Biomarker Panels for Guiding Dysregulated Host Response Therapy

Description

BACKGROUND

Host response is a complex pathophysiologic process arising from an insult such as infection, trauma, burns, and other injuries. Diverse host responses can manifest clinically, including immune response, inflammatory response, coagulopathic response, and any other type of response to bodily insult. In some cases, host response to bodily insult can go awry, causing acute, life-threatening syndromes. As referred to herein, “dysregulated host response” refers to such cases in which host response to bodily insult goes awry, and thereby causes acute, life-threatening syndromes. For example, dysregulated immune response to infection can manifest clinically as sepsis. As another example, dysregulated immune response to a non-infection insult, such as, for example, burns, can manifest clinically as Systemic Inflammatory Response Syndrome (SIRS)⁵².

Sepsis is an acute, life-threatening syndrome caused by a dysregulated immune response to infection.^1,2Approximately 1.7 million patients are diagnosed with sepsis each year.¹⁵According to a recent study based on electronic medical record data from more than 7 million hospitalizations across 409 US hospitals, sepsis has an estimated 6% hospital admission rate.¹⁵The average length of stay for septic patients is 75% greater than most other conditions, and its mortality accounts for more than 50% of hospital deaths in hospitals.¹⁶Sepsis ranks as one of the highest costs among all hospital admissions, representing approximately 13% of total US hospital costs, or more than $24 billion in hospital expenses.¹⁶Sepsis costs increase based on sepsis severity level and timing of clinical presentation (e.g., at the hospital admission or during the hospital stay). Sepsis cases that were not present at hospital admission spend almost twice the amount of time in the hospital, in the intensive care unit, and on mechanical ventilation, compared to patients in which sepsis was presented at the hospital admission.¹⁷

Beyond early recognition, according to the Surviving Sepsis Campaign guidelines, the cornerstone for initial sepsis management is currently based on five main actions known as the “1-hour bundle”. The “1-hour bundle” includes: (1) lactate level measurement; (2) blood cultures collection; (3) broad-spectrum antibiotics administration; (4) rapid fluid administration of 30 ml/kg crystalloid for hypotension or lactate ≥4 mmol/L and (5) vasopressors for patients that remain hypotensive during or after resuscitation to maintain mean arterial pressure ≥65 mmHg.¹⁸

After applying this initial approach, patients are frequently assessed over the following hours according to their clinical response. For those patients with poor clinical response, further adjustments, in terms of the amount of fluids given and/or in terms of the choice of antibiotic therapy and measurements for source control (e.g. device removal, surgical procedures, or additional investigation), can be made.

Despite the appropriate application of these actions, close to 30% of septic patients remain hypotensive, requiring vasopressors to maintain a mean arterial pressure ≥65 mmHg, and then are characterized as having septic shock,¹⁹a subtype of sepsis and a condition that has an expected hospital mortality in excess of 40%.¹Of septic shock patients, close to 40% continue to show no clinical improvement (refractory septic shock), defined as a systolic blood pressure <90 mmHg for more than one hour following both adequate fluid resuscitation and vasopressor therapy. In this set of refractory septic shock patients, glucocorticoid therapy may provide improvement.¹

Corticosteroids remain a controversial therapy for sepsis patients. Specifically, current guidelines provide a weak recommendation for corticosteroids sepsis patients by stating that either steroids and no steroids are reasonable management options.²⁰

Despite often-promising preclinical studies, more than 100 interventional trials have failed to demonstrate significantly improved survival among sepsis patients, leaving clinicians with limited interventions, and patients with mortality rates as high as 40% among those who develop septic shock.^4-12

Similar trends have also been observed for other manifestations of dysregulated host response not caused by infection, such as, for examples SIRS, which can be caused by severe burns.

SUMMARY

Embodiments disclosed herein relate to methods, non-transitory computer-readable mediums, systems, and kits for determining patient subtypes, determining therapy recommendations for patients, and generating therapeutic hypotheses for patient subtypes. In various embodiments described herein, the methods involve analyzing quantitative data of one or more biomarker sets derived from a sample obtained from a patient using a patient subtype classifier. The patient subtype classifier outputs a classification for the patient that guides the determination of a therapy recommendation.

Disclosed herein is a method for determining a patient subtype, the method comprising: obtaining or having obtained quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one or more of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8, wherein biomarker 2 is one or more of SERPINB1 or GSPT1, and wherein biomarker 3 is one or more of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A, wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one or more of ZNF831, MME, CD3G, or STOM, wherein biomarker 5 is one or more of ECSIT, LAT, or NCOA4, and wherein biomarker 6 is one or more of SLC1A5, IGF2BP2, or ANXA3, wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one or more of C14orf159 or PUM2, wherein biomarker 8 is one or more of EPB42 or RPS6KA5, and wherein biomarker 9 is one or more of EPB42 or GBP2; and wherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one or more of MSH2, DCTD, or MMP8, wherein biomarker 11 is one or more of HK3, UCP2, or NUP88, and wherein biomarker 12 is one or more of GABARAPL2 or CASP4; and wherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one or more of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G, wherein biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; and determining a classification of a subject based on the quantitative data using a patient subtype classifier.

In various embodiments, the at least one biomarker set is group 5, and wherein biomarker 13 is one or more of STOM, MME, BNT3A2, or HLA-DPA1. In various embodiments, wherein the at least one biomarker set is group 5, and wherein biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1. In various embodiments, the at least one biomarker set is group 5, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, or ANXA3.

Additionally disclosed herein is a method for determining a therapy recommendation for a patient, the method comprising: obtaining or having obtained quantitative data for two or more biomarkers selected from the group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, TOMM70A, ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, ANXA3, C14orf159, PUM2, EPB42, RPS6KA5, GBP2, MSH2, DCTD, HK3, UCP2, NUP88, GABARAPL2, and CASP4; and determining a classification of a subject based on the quantitative data using a patient subtype classifier.

Additionally disclosed herein is a method for determining a therapy recommendation for a patient, the method comprising: obtaining or having obtained quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises two or more biomarkers selected from a group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1 GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A, wherein group 2 comprises two or more biomarkers selected from a group consisting of ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, and ANXA3, wherein group 3 comprises two or more biomarkers selected from a group consisting of C14orf159, PUM2, EPB42, RPS6KA5, EPB42, and GBP2; and wherein group 4 comprises two or more biomarkers selected from a group consisting of MSH2, DCTD, MMP8, HK3, UCP2, NUP88, GABARAPL2, and CASP4; and wherein group 5 comprises two or more biomarkers selected from a group consisting of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, CD3G, EPB42, GSPT1, LAT, HK3, SERPINB1, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, and TNFRSF1A; and determining a classification of a subject based on the quantitative data using a patient subtype classifier. In various embodiments, methods described herein further comprise identifying a therapy recommendation for the subject based at least in part on the classification.

Additionally disclosed herein is a method for determining a therapy recommendation for a patient, the method comprising: obtaining a classification of a subject exhibiting a dysregulated host response, the classification having been determined by: obtaining or having obtained quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one or more of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8, wherein biomarker 2 is one or more of SERPINB1 or GSPT1, and wherein biomarker 3 is one or more of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A, wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one or more of ZNF831, MME, CD3G, or STOM, wherein biomarker 5 is one or more of ECSIT, LAT, or NCOA4, and wherein biomarker 6 is one or more of SLC1A5, IGF2BP2, or ANXA3, wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one or more of C14orf159 or PUM2, wherein biomarker 8 is one or more of EPB42 or RPS6KA5, and wherein biomarker 9 is one or more of EPB42 or GBP2; and wherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one or more of MSH2, DCTD, or MMP8, wherein biomarker 11 is one or more of HK3, UCP2, or NUP88, and wherein biomarker 12 is one or more of GABARAPL2 or CASP4; and wherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one or more of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G, wherein biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; and determining the classification based on the quantitative data using a patient subtype classifier; and identifying a therapy recommendation for the subject based at least in part on the classification.

In various embodiments, the dysregulated host response of the subject comprises one of sepsis and dysregulated host response not caused by infection. In various embodiments, the classification of the subject comprises one of subtype A or subtype B. In various embodiments, the classification of the subject comprises one of subtype A, subtype B, or subtype C. In various embodiments, responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject comprises at least no immunosuppressive therapy. In various embodiments, responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject further comprises at least no corticosteroid therapy. In various embodiments, the therapy recommendation identified for the subject further comprises no hydrocortisone.

In various embodiments, responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy. In various embodiments, responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor, a blocker of complement components, a blocker of complement component receptors, and a blocker of a pro-inflammatory cytokine. In various embodiments, the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, anti-C5a, anti-C3a, anti-C5aR, anti-C3aR, anti-TNF-alpha, and anti-IL-6, Anti-HMGB1, ST2 antibody, IL-33 antibody.

In various embodiments, responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, modulators of coagulation therapy, and modulators of vascular permeability therapy. In various embodiments, responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor and an anticoagulant. In various embodiments, the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, activated protein C, antithrombin, and thrombomodulin. In various embodiments, methods further comprise administering or having administered therapy to the subject based on the therapy recommendation.

In various embodiments, obtaining or having obtained quantitative data comprises: obtaining a sample from a subject exhibiting dysregulated host response, wherein the sample comprises a plurality of biomarkers; and determining the quantitative data from the obtained sample. In various embodiments, the obtained sample comprises a blood sample from the subject. In various embodiments, the subject exhibiting dysregulated host response does not exhibit shock, and wherein the at least one biomarker set is one of group 1, group 3, or group 4. In various embodiments, the subject exhibiting dysregulated host response is further exhibiting shock, and wherein the at least one biomarker set is one of group 1, group 2, group 4, group 5, group 6, group 7, or group 8. In various embodiments, the subject exhibiting dysregulated host response is an adult subject, and wherein the at least one biomarker set is one of group 1, group 2, group 3, group 5, group 6, group 7, or group 8. In various embodiments, the subject exhibiting dysregulated host response is a pediatric subject, and wherein the at least one biomarker set is one of group 1, group 4, group 5, group 6, group 7, or group 8.

In various embodiments, the quantitative data is determined by one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction.

In various embodiments, the quantitative data is determined by: contacting a sample with a reagent; generating a plurality of complexes between the reagent and the plurality of biomarkers in the sample; and detecting the plurality of complexes to obtain a dataset associated with the sample, wherein the dataset comprises the quantitative data. In various embodiments, the classification of the subject is determined by: determining, for at least one candidate classification of the subject, a classification-specific score for the subject; determining, by the patient subtype classifier, based on the classification-specific score, the classification of the subject.

In various embodiments, determining the classification-specific score comprises: determining a first subscore of the quantitative data for the subject for one or more biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more biomarkers of the candidate classification are increased relative to the quantitative data for the one or more biomarkers for one or more control subjects; determining a second subscore of the quantitative expression for the subject for one or more additional biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more additional biomarkers of the candidate classification are decreased relative to the quantitative data for the one or more additional biomarkers for the one or more control subjects; and determining a difference between the first subscore and the second subscore, the first and second geometric subscore optionally subject to scaling, and the difference comprising the classification-specific score for the subject. In various embodiments, one or both of the first subscore and the second subscore are geometric means.

In various embodiments, the patient subtype classifier is a machine-learned model. In various embodiments, the machine-learned model is a support vector machine (SVM). In various embodiments, the support vector machine receives, as input, one or more classification-specific scores and outputs the classification of the subject. In various embodiments, the patient subtype classifier determines the classification of the subject by: comparing the classification-specific scores to one or more threshold values; and determining the classification of the subject based on the comparisons. In various embodiments, at least one of the one or more threshold values is a fixed value. In various embodiments, at least one of the one or more threshold values is determined using training samples, the at least one threshold value representing a value on a ROC curve nearest to maximum sensitivity or maximum specificity.

In various embodiments, methods disclosed herein further comprise, prior to determining a classification of the subject using a patient subtype classifier, normalizing the quantitative data based on quantitative data for one or more housekeeping genes. In various embodiments, the candidate classifications of the subject comprise subtype A, subtype B, and subtype C. In various embodiments, the at least one biomarker set is group 1, and wherein the patient subtype classifier has an average accuracy of at least 82.93%. In various embodiments, the at least one biomarker set is group 2, and wherein the patient subtype classifier has an average accuracy of at least 89.6%. In various embodiments, the at least one biomarker set is group 3, and wherein the patient subtype classifier has an average accuracy of at least 86.3%. In various embodiments, the at least one biomarker set is group 4, and wherein the patient subtype classifier has an average accuracy of at least 98.3%.

In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation. In various embodiments, the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response not provided corticosteroid therapy is greater than or equal to a threshold statistical significance. In various embodiments, the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A or subtype C. In various embodiments, the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is greater than or equal to a threshold statistical significance.

In various embodiments, the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be favorably responsive to corticosteroid therapy. In various embodiments, the subtype is subtype B.

In various embodiments, the therapy recommendation identified for the subject comprises a no therapy recommendation, wherein the no therapy recommendation is identified at least by: determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and not provided corticosteroid therapy is less than a threshold statistical significance; and determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is less than a threshold statistical significance. In various embodiments, a statistical significance comprises a p-value, and wherein the threshold statistical significance comprises at least 0.1.

In various embodiments, the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 1 or group 4. In various embodiments, the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A or subtype C.

In various embodiments, the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A. In various embodiments, the therapy recommendation identified for the subject further comprises a no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises a subtype likely to be non-responsive to corticosteroid therapy. In various embodiments, the subtype is subtype B or subtype C.

In various embodiments, the therapy recommendation identified for the subject further comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 2. In various embodiments, the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be adversely responsive to corticosteroid therapy. In various embodiments, the subtype is subtype C.

In various embodiments, the therapy recommendation identified for the subject further comprises a no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A or subtype B. In various embodiments, the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 3. In various embodiments, the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A or subtype C. In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be responsive to corticosteroid therapy. In various embodiments, the subtype is subtype B.

Additionally disclosed herein is a method for identifying a candidate therapeutic, the method comprising: accessing a differentially expressed gene database comprising gene level fold changes between patients of different subtypes; determining at least a threshold number of genes are differentially expressed in patients of a first subtype in comparison to patients of a second subtype, wherein each of the differentially expressed genes is involved in a common biological pathway; and determining a candidate therapeutic likely to be effective for patients of the first subtype, wherein the candidate therapeutic is effective in modulating expression of at least one of the genes that are differentially expressed in patients of the first subtype. In various embodiments, the differentially expressed gene database is generated by: obtaining labeled patient data, wherein labels of the labeled patient data identify patients that are classified into one of two or more subtypes; generating the differentially expressed gene database for at least one or more genes by at least determining gene-level fold changes between patient data with a label indicating a first subtype and patient data with a label indicating a second subtype. In various embodiments, the labels of the labeled patient data are generated by applying a clustering analysis or by applying a patient subtype classifier. In various embodiments, at least the threshold number of genes is at least three genes, at least four genes, at least five genes, at least six genes, at least seven genes, at least eight genes, at least nine genes, or at least ten genes. In various embodiments, determining a candidate therapeutic for patients of the first subtype further comprises: analyzing one or both of: therapeutic pharmacology data comprising data for the candidate therapeutic; and host response pathobiology comprising data for patients of the first subtype.

Additionally disclosed herein a non-transitory computer readable medium for determining a patient subtype, the non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to: obtain quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one or more of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8, wherein biomarker 2 is one or more of SERPINB1 or GSPT1, and wherein biomarker 3 is one or more of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A, wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one or more of ZNF831, MME, CD3G, or STOM, wherein biomarker 5 is one or more of ECSIT, LAT, or NCOA4, and wherein biomarker 6 is one or more of SLC1A5, IGF2BP2, or ANXA3, wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one or more of C14orf159 or PUM2, wherein biomarker 8 is one or more of EPB42 or RPS6KA5, and wherein biomarker 9 is one or more of EPB42 or GBP2; and wherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one or more of MSH2, DCTD, or MMP8, wherein biomarker 11 is one or more of HK3, UCP2, or NUP88, and wherein biomarker 12 is one or more of GABARAPL2 or CASP4; and wherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one or more of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G, wherein biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; and determine a classification of a subject based on the quantitative data using a patient subtype classifier.

In various embodiments, the at least one biomarker set is group 5, and wherein biomarker 13 is one or more of STOM, MME, BNT3A2, or HLA-DPA1. In various embodiments, the at least one biomarker set is group 5, and wherein biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1. In various embodiments, the at least one biomarker set is group 5, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, or ANXA3.

Additionally disclosed herein is a non-transitory computer readable medium for determining a therapy recommendation for a patient, the non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to: obtain quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises two or more biomarkers selected from a group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1 GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A, wherein group 2 comprises two or more biomarkers selected from a group consisting of ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, and ANXA3, wherein group 3 comprises two or more biomarkers selected from a group consisting of C14orf159, PUM2, EPB42, RPS6KA5, EPB42, and GBP2; and wherein group 4 comprises two or more biomarkers selected from a group consisting of MSH2, DCTD, MMP8, HK3, UCP2, NUP88, GABARAPL2, and CASP4; and wherein group 5 comprises two or more biomarkers selected from a group consisting of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, CD3G, EPB42, GSPT1, LAT, HK3, SERPINB1, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, and TNFRSF1A; and determine a classification of a subject based on the quantitative data using a patient subtype classifier. In various embodiments, the instructions further comprise instructions that, when executed by the processor, cause the processor to identify a therapy recommendation for the subject based at least in part on the classification.

Additionally disclosed herein is a non-transitory computer readable medium for determining a therapy recommendation for a subject, the non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to: obtain a classification of the subject exhibiting a dysregulated host response, the classification having been determined by: obtaining or having obtained quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one or more of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8, wherein biomarker 2 is one or more of SERPINB1 or GSPT1, and wherein biomarker 3 is one or more of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A, wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one or more of ZNF831, MME, CD3G, or STOM, wherein biomarker 5 is one or more of ECSIT, LAT, or NCOA4, and wherein biomarker 6 is one or more of SLC1A5, IGF2BP2, or ANXA3, wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one or more of C14orf159 or PUM2, wherein biomarker 8 is one or more of EPB42 or RPS6KA5, and wherein biomarker 9 is one or more of EPB42 or GBP2; and wherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one or more of MSH2, DCTD, or MMP8, wherein biomarker 11 is one or more of HK3, UCP2, or NUP88, and wherein biomarker 12 is one or more of GABARAPL2 or CASP4; and wherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one or more of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G, wherein biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; and determining the classification based on the quantitative data using a patient subtype classifier; and identify a therapy recommendation for the subject based at least in part on the classification.

In various embodiments, responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy. In various embodiments, responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor, a blocker of complement components, a blocker of complement component receptors, and a blocker of a pro-inflammatory cytokine. In various embodiments, the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, anti-C5a, anti-C3a, anti-C5aR, anti-C3aR, anti-TNF-alpha, and anti-IL-6, Anti-HMGB1, ST2 antibody, IL-33 antibody. In various embodiments, responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, modulators of coagulation therapy, and modulators of vascular permeability therapy.

In various embodiments, responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor and an anticoagulant. In various embodiments, the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, activated protein C, antithrombin, and thrombomodulin.

In various embodiments, the instructions that cause the processor to obtain quantitative data further comprises instructions that, when executed by the processor, cause the processor to: obtain a sample from a subject exhibiting dysregulated host response, wherein the sample comprises a plurality of biomarkers; and determine the quantitative data from the obtained sample. In various embodiments, the obtained sample comprises a blood sample from the subject.

In various embodiments, the subject exhibiting dysregulated host response does not exhibit shock, and wherein the at least one biomarker set is one of group 1, group 3, or group 4. In various embodiments, the subject exhibiting dysregulated host response is further exhibiting shock, and wherein the at least one biomarker set is one of group 1, group 2, group 4, group 5, group 6, group 7, or group 8. In various embodiments, the subject exhibiting dysregulated host response is an adult subject, and wherein the at least one biomarker set is one of group 1, group 2, group 3, group 5, group 6, group 7, or group 8. In various embodiments, the subject exhibiting dysregulated host response is a pediatric subject, and wherein the at least one biomarker set is one of group 1, group 4, group 5, group 6, group 7, or group 8. In various embodiments, the quantitative data is determined by one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction.

In various embodiments, the classification of the subject is determined by: determining, for at least one candidate classification of the subject, a classification-specific score for the subject; determining, by the patient subtype classifier, based on the classification-specific score, the classification of the subject. In various embodiments, the instructions that cause the processor to determine the classification-specific score further comprises instructions that, when executed by the processor, cause the processor to: determine a first subscore of the quantitative data for the subject for one or more biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more biomarkers of the candidate classification are increased relative to the quantitative data for the one or more biomarkers for one or more control subjects; determine a second subscore of the quantitative expression for the subject for one or more additional biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more additional biomarkers of the candidate classification are decreased relative to the quantitative data for the one or more additional biomarkers for the one or more control subjects; and determine a difference between the first subscore and the second subscore, the first and second geometric subscore optionally subject to scaling, and the difference comprising the classification-specific score for the subject. In various embodiments, one or both of the first subscore and the second subscore are geometric means. In various embodiments, the patient subtype classifier is a machine-learned model. In various embodiments, the machine-learned model is a support vector machine (SVM). In various embodiments, the support vector machine receives, as input, one or more classification-specific scores and outputs the classification of the subject.

In various embodiments, the patient subtype classifier determines the classification of the subject by: comparing the classification-specific scores to one or more threshold values; and determining the classification of the subject based on the comparisons. In various embodiments, at least one of the one or more threshold values is a fixed value. In various embodiments, at least one of the one or more threshold values is determined using training samples, the at least one threshold value representing a value on a ROC curve nearest to maximum sensitivity or maximum specificity. In various embodiments, further comprising, prior to determining a classification of the subject using a patient subtype classifier, normalizing the quantitative data based on quantitative data for one or more housekeeping genes.

In various embodiments, wherein the candidate classifications of the subject comprise subtype A, subtype B, and subtype C. In various embodiments, the at least one biomarker set is group 1, and wherein the patient subtype classifier has an average accuracy of at least 82.93%. In various embodiments, the patient subtype classifier has an average accuracy of at least 89.6%. In various embodiments, the patient subtype classifier has an average accuracy of at least 86.3%. In various embodiments, the at least one biomarker set is group 4, and wherein the patient subtype classifier has an average accuracy of at least 98.3%.

In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation. In various embodiments, the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response not provided corticosteroid therapy is greater than or equal to a threshold statistical significance. In various embodiments, the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A or subtype C.

In various embodiments, the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is greater than or equal to a threshold statistical significance. In various embodiments, the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be favorably responsive to corticosteroid therapy. In various embodiments, the subtype is subtype B.

In various embodiments, the therapy recommendation identified for the subject further comprises no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises subtype B. In various embodiments, the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises sepsis, wherein the at least one biomarker set is one of group 2, group 3, or group 4. In various embodiments, the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A.

In various embodiments, the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 3. In various embodiments, the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A or subtype C.

In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be responsive to corticosteroid therapy. In various embodiments, the subtype is subtype B.

Additionally disclosed herein is a non-transitory computer readable medium for identifying a candidate therapeutic, the non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to: access a differentially expressed gene database comprising gene level fold changes between patients of different subtypes; determine at least a threshold number of genes are differentially expressed in patients of a first subtype in comparison to patients of a second subtype, wherein each of the differentially expressed genes is involved in a common biological pathway; and determine a candidate therapeutic likely to be effective for patients of the first subtype, wherein the candidate therapeutic is effective in modulating expression of at least one of the genes that are differentially expressed in patients of the first subtype. In various embodiments, the differentially expressed gene database is generated by: obtaining labeled patient data, wherein labels of the labeled patient data identify patients that are classified into one of two or more subtypes; generating the differentially expressed gene database for at least one or more genes by at least determining gene-level fold changes between patient data with a label indicating a first subtype and patient data with a label indicating a second subtype. In various embodiments, the labels of the labeled patient data are generated by applying a clustering analysis or by applying a patient subtype classifier. In various embodiments, at least the threshold number of genes is at least three genes, at least four genes, at least five genes, at least six genes, at least seven genes, at least eight genes, at least nine genes, or at least ten genes. In various embodiments, determining a candidate therapeutic for patients of the first subtype further comprises: analyzing one or both of: therapeutic pharmacology data comprising data for the candidate therapeutic; and host response pathobiology comprising data for patients of the first subtype.

Additionally disclosed herein is a system for determining a patient subtype, the system comprising: a set of reagents used for determining quantitative data for at least one biomarker set from a test sample from a subject, the at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one or more of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8, wherein biomarker 2 is one or more of SERPINB1 or GSPT1, and wherein biomarker 3 is one or more of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A, wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one or more of ZNF831, MME, CD3G, or STOM, wherein biomarker 5 is one or more of ECSIT, LAT, or NCOA4, and wherein biomarker 6 is one or more of SLC1A5, IGF2BP2, or ANXA3, wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one or more of C14orf159 or PUM2, wherein biomarker 8 is one or more of EPB42 or RPS6KA5, and wherein biomarker 9 is one or more of EPB42 or GBP2; and wherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one or more of MSH2, DCTD, or MMP8, wherein biomarker 11 is one or more of HK3, UCP2, or NUP88, and wherein biomarker 12 is one or more of GABARAPL2 or CASP4; and wherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one or more of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G, wherein biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; and an apparatus configured to receive a mixture of one or more reagents in the set and the test sample and to measure the quantitative data for the at least one biomarker set from the test sample; and a computer system communicatively coupled to the apparatus to obtain the quantitative data for the at least one biomarker set and to determine a classification of the subject based on the quantitative data using a patient subtype classifier.

In various embodiments, the at least one biomarker set is group 5, and wherein biomarker 13 is one or more of STOM, MME, BNT3A2, or HLA-DPA1. In various embodiments, biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1. In various embodiments, the at least one biomarker set is group 5, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, or ANXA3.

Additionally disclosed herein is a system for determining a patient subtype, the system comprising: a set of reagents used for determining quantitative data for two or more biomarkers selected from the group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, TOMM70A, ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, ANXA3, C14orf159, PUM2, EPB42, RPS6KA5, GBP2, MSH2, DCTD, HK3, UCP2, NUP88, GABARAPL2, and CASP4; and an apparatus configured to receive a mixture of one or more reagents in the set and the test sample and to measure the quantitative data for the at least one biomarker set from the test sample; and a computer system communicatively coupled to the apparatus to obtain the quantitative data for the at least one biomarker set and to determine a classification of the subject based on the quantitative data using a patient subtype classifier.

Additionally disclosed herein is a system for determining a patient subtype, the system comprising: a set of reagents used for determining quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises two or more biomarkers selected from a group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1 GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A, wherein group 2 comprises two or more biomarkers selected from a group consisting of ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, and ANXA3, wherein group 3 comprises two or more biomarkers selected from a group consisting of C14orf159, PUM2, EPB42, RPS6KA5, EPB42, and GBP2; and wherein group 4 comprises two or more biomarkers selected from a group consisting of MSH2, DCTD, MMP8, HK3, UCP2, NUP88, GABARAPL2, and CASP4; and wherein group 5 comprises two or more biomarkers selected from a group consisting of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, CD3G, EPB42, GSPT1, LAT, HK3, SERPINB1, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, and TNFRSF1A; and an apparatus configured to receive a mixture of one or more reagents in the set and the test sample and to measure the quantitative data for the at least one biomarker set from the test sample; and a computer system communicatively coupled to the apparatus to obtain the quantitative data for the at least one biomarker set and to determine a classification of the subject based on the quantitative data using a patient subtype classifier. In various embodiments, the computer system is configured to identify a therapy recommendation for the subject based at least in part on the classification.

Additionally disclosed herein is a system for determining a therapy recommendation for a subject, the system comprising: a computer system configured to: obtain a classification of the subject exhibiting a dysregulated host response, the classification having been determined by: obtaining or having obtained quantitative data for at least one biomarker set obtained from the subject, the at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one or more of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8, wherein biomarker 2 is one or more of SERPINB1 or GSPT1, and wherein biomarker 3 is one or more of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A, wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one or more of ZNF831, MME, CD3G, or STOM, wherein biomarker 5 is one or more of ECSIT, LAT, or NCOA4, and wherein biomarker 6 is one or more of SLC1A5, IGF2BP2, or ANXA3, wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one or more of C14orf159 or PUM2, wherein biomarker 8 is one or more of EPB42 or RPS6KA5, and wherein biomarker 9 is one or more of EPB42 or GBP2; and wherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one or more of MSH2, DCTD, or MMP8, wherein biomarker 11 is one or more of HK3, UCP2, or NUP88, and wherein biomarker 12 is one or more of GABARAPL2 or CASP4; and wherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one or more of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G, wherein biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; and determine the classification based on the quantitative data using a patient subtype classifier; and identify a therapy recommendation for the subject based at least in part on the classification.

In various embodiments, responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject further comprises at least no corticosteroid therapy. In various embodiments, the therapy recommendation identified for the subject further comprises no hydrocortisone. In various embodiments, responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy. In various embodiments, responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor, a blocker of complement components, a blocker of complement component receptors, and a blocker of a pro-inflammatory cytokine. In various embodiments, the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, anti-C5a, anti-C3a, anti-C5aR, anti-C3aR, anti-TNF-alpha, and anti-IL-6, Anti-HMGB1, ST2 antibody, IL-33 antibody.

In various embodiments, the sample comprises a blood sample from the subject. In various embodiments, the subject exhibiting dysregulated host response does not exhibit shock, and wherein the at least one biomarker set is one of group 1, group 3, or group 4. In various embodiments, the subject exhibiting dysregulated host response is further exhibiting shock, and wherein the at least one biomarker set is one of group 1, group 2, group 4, group 5, group 6, group 7, or group 8. In various embodiments, the subject exhibiting dysregulated host response is an adult subject, and wherein the at least one biomarker set is one of group 1, group 2, group 3, group 5, group 6, group 7, or group 8. In various embodiments, the subject exhibiting dysregulated host response is a pediatric subject, and wherein the at least one biomarker set is one of group 1, group 4, group 5, group 6, group 7, or group 8.

In various embodiments, the classification of the subject is determined by: determining, for at least one candidate classification of the subject, a classification-specific score for the subject; determining, by the patient subtype classifier, based on the classification-specific score, the classification of the subject. In various embodiments, determine the classification-specific score further comprises: determine a first subscore of the quantitative data for the subject for one or more biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more biomarkers of the candidate classification are increased relative to the quantitative data for the one or more biomarkers for one or more control subjects; determine a second subscore of the quantitative expression for the subject for one or more additional biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more additional biomarkers of the candidate classification are decreased relative to the quantitative data for the one or more additional biomarkers for the one or more control subjects; and determine a difference between the first subscore and the second subscore, the first and second geometric subscore optionally subject to scaling, and the difference comprising the classification-specific score for the subject. In various embodiments, one or both of the first subscore and the second subscore are geometric means.

In various embodiments, at least one of the one or more threshold values is a fixed value. In various embodiments, at least one of the one or more threshold values is determined using training samples, the at least one threshold value representing a value on a ROC curve nearest to maximum sensitivity or maximum specificity. In various embodiments, prior to determining a classification of the subject using a patient subtype classifier, normalizing the quantitative data based on quantitative data for one or more housekeeping genes.

In various embodiments, the candidate classifications of the subject comprise subtype A, subtype B, and subtype C. In various embodiments, the at least one biomarker set is group 1, and wherein the patient subtype classifier has an average accuracy of at least 82.93%. In various embodiments, the patient subtype classifier has an average accuracy of at least 89.6%. In various embodiments, the patient subtype classifier has an average accuracy of at least 86.3%. In various embodiments, the at least one biomarker set is group 4, and wherein the patient subtype classifier has an average accuracy of at least 98.3%.

In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation. In various embodiments, the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response not provided corticosteroid therapy is greater than or equal to a threshold statistical significance. In various embodiments, the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A or subtype C. In various embodiments, the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is greater than or equal to a threshold statistical significance.

In various embodiments, the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises sepsis, wherein the at least one biomarker set is one of group 2, group 3, or group 4. In various embodiments, the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A.

In various embodiments, the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 3. In various embodiments, the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy. In various embodiments, the subtype is subtype A or subtype C.

Additionally disclosed herein is a a system for identifying a candidate therapeutic, the system comprising: a storage device storing a differentially expressed gene database comprising gene level fold changes between patients of different subtypes; a computational device configured to: access one or more gene level fold changes corresponding to differentially expressed genes in the differentially expressed gene database; determine at least a threshold number of genes are differentially expressed in patients of a first subtype in comparison to patients of a second subtype, wherein each of the differentially expressed genes is involved in a common biological pathway; and determine a candidate therapeutic likely to be effective for patients of the first subtype, wherein the candidate therapeutic is effective in modulating expression of at least one of the genes that are differentially expressed in patients of the first subtype. In various embodiments, the differentially expressed gene database is generated by: obtaining labeled patient data, wherein labels of the labeled patient data identify patients that are classified into one of two or more subtypes; generating the differentially expressed gene database for at least one or more genes by at least determining gene-level fold changes between patient data with a label indicating a first subtype and patient data with a label indicating a second subtype. In various embodiments, the labels of the labeled patient data are generated by applying a clustering analysis or by applying a patient subtype classifier. In various embodiments, at least the threshold number of genes is at least three genes, at least four genes, at least five genes, at least six genes, at least seven genes, at least eight genes, at least nine genes, or at least ten genes. In various embodiments, determining a candidate therapeutic for patients of the first subtype further comprises: analyzing one or both of: therapeutic pharmacology data comprising data for the candidate therapeutic; and host response pathobiology comprising data for patients of the first subtype.

Additionally disclosed herein is a kit for determining a patient subtype, the kit comprising: a set of reagents for determining quantitative data for at least one biomarker set from a test sample from a subject, the at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one or more of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8, wherein biomarker 2 is one or more of SERPINB1 or GSPT1, and wherein biomarker 3 is one or more of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A, wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one or more of ZNF831, MME, CD3G, or STOM, wherein biomarker 5 is one or more of ECSIT, LAT, or NCOA4, and wherein biomarker 6 is one or more of SLC1A5, IGF2BP2, or ANXA3, wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one or more of C14orf159 or PUM2, wherein biomarker 8 is one or more of EPB42 or RPS6KA5, and wherein biomarker 9 is one or more of EPB42 or GBP2; and wherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one or more of MSH2, DCTD, or MMP8, wherein biomarker 11 is one or more of HK3, UCP2, or NUP88, and wherein biomarker 12 is one or more of GABARAPL2 or CASP4; and wherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one or more of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G, wherein biomarker 14 is one or more of EPB42, GSPT1, LAT, HK3, or SERPINB1, and wherein biomarker 15 is one or more of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; and instructions for using the set of reagents to determine the quantitative data for the at least one biomarker set.

Additionally disclosed herein is a kit for determining a patient subtype, the kit comprising: a set of reagents for determining quantitative data for two or more biomarkers selected from the group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, TOMM70A, ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, ANXA3, C14orf159, PUM2, EPB42, RPS6KA5, GBP2, MSH2, DCTD, HK3, UCP2, NUP88, GABARAPL2, and CASP4; and instructions for using the set of reagents to determine the quantitative data for the at least one biomarker set.

Additionally disclosed herein is a kit for determining a patient subtype, the kit comprising: a set of reagents for determining quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises two or more biomarkers selected from a group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1 GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A, wherein group 2 comprises two or more biomarkers selected from a group consisting of ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, and ANXA3, wherein group 3 comprises two or more biomarkers selected from a group consisting of C14orf159, PUM2, EPB42, RPS6KA5, EPB42, and GBP2; and wherein group 4 comprises two or more biomarkers selected from a group consisting of MSH2, DCTD, MMP8, HK3, UCP2, NUP88, GABARAPL2, and CASP4; and wherein group 5 comprises two or more biomarkers selected from a group consisting of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, CD3G, EPB42, GSPT1, LAT, HK3, SERPINB1, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, and TNFRSF1A; and instructions for using the set of reagents to determine the quantitative data for the at least one biomarker set.

In various embodiments, the instructions comprise instructions for determining the quantitative data by performing one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction.

In various embodiments, the set of reagents comprises at least three primer sets for amplifying at least three biomarkers, wherein the at least three primer sets comprise pairs of single-stranded DNA primers for amplifying the at least three biomarkers, and wherein at least one of the at least three biomarkers is selected from the group consisting of the biomarkers EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, ZNF831, MME, CD3G, STOM, C14orf159, PUM2, MSH2, DCTD, BNT3A2, or HLA-DPA1, at least one biomarker of the at least three biomarkers is selected from the group consisting of the biomarkers SERPINB1, GSPT1, ECSIT, LAT, NCOA4, EPB42, RPS6KA5, HK3, UCP2, or NUP88, and at least one biomarker of the at least three biomarkers is selected from the group consisting of the biomarkers MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, PRPF3, TOMM70A, EPB42, GABARAPL2, CASP4, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, OR TNFRSF1A.

In various embodiments, the at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 7 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 8, a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 9 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 10, a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 11 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 12, and a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 13 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 14, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 15 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 16, a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 17 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 18, and a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 19 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 20, and wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 1 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 2; a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 3 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 4, and a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 5 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 6.

In various embodiments, at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 7 and a reverse primer comprising SEQ ID NO. 8, a forward primer comprising SEQ ID NO. 9 and a reverse primer comprising SEQ ID NO. 10, a forward primer comprising SEQ ID NO. 11 and a reverse primer comprising SEQ ID NO. 12, and a forward primer comprising SEQ ID NO. 13 and a reverse primer comprising SEQ ID NO. 14, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 15 and a reverse primer comprising SEQ ID NO. 16, a forward primer comprising SEQ ID NO. 17 and a reverse primer comprising SEQ ID NO. 18, and a forward primer comprising SEQ ID NO. 19 and a reverse primer comprising SEQ ID NO. 20, and wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 1 and a reverse primer comprising SEQ ID NO. 2; a forward primer comprising SEQ ID NO. 3 and a reverse primer comprising SEQ ID NO. 4, and a forward primer comprising SEQ ID NO. 5 and a reverse primer comprising SEQ ID NO. 6.

In various embodiments, at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 21 and a reverse primer comprising SEQ ID NO. 22, and a forward primer comprising SEQ ID NO. 23 and a reverse primer comprising SEQ ID NO. 24, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 25 and a reverse primer comprising SEQ ID NO. 26, and a forward primer comprising SEQ ID NO. 29 and a reverse primer comprising SEQ ID NO. 30, and wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 25 and a reverse primer comprising SEQ ID NO. 26, and a forward primer comprising SEQ ID NO. 27 and a reverse primer comprising SEQ ID NO. 28.

In various embodiments, the set of reagents comprises at least three primer sets for amplifying at least three biomarkers, wherein each primer set of the at least three primer sets comprises a forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer for amplifying one of the at least three biomarkers, and wherein at least one of the at least three biomarkers is selected from the group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, ZNF831, MME, CD3G, STOM, C14orf159, PUM2, MSH2, DCTD, BNT3A2, or HLA-DPA1, at least one biomarker of the at least three biomarkers is selected from the group consisting of SERPINB1, GSPT1, ECSIT, LAT, NCOA4, EPB42, RPS6KA5, HK3, UCP2, or NUP88, and at least one biomarker of the at least three biomarkers is selected from the group consisting of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, PRPF3, TOMM70A, EPB42, GABARAPL2, CASP4, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, OR TNFRSF1A.

In various embodiments, at least one of the at least three primer sets is selected from the group consisting of: a forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer, each of which is configured to enable amplification of at least one biomarker selected from the group consisting of: EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, ZNF831, MME, CD3G, STOM, C14orf159, PUM2, MSH2, DCTD, BNT3A2, or HLA-DPA1, a forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer, each of which is configured to enable amplification of at least one biomarker selected from the group consisting of: SERPINB1, GSPT1, ECSIT, LAT, NCOA4, EPB42, RPS6KA5, HK3, UCP2, or NUP88, and a forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer, each of which is configured to enable amplification of at least one biomarker selected from the group consisting of: MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, PRPF3, TOMM70A, EPB42, GABARAPL2, CASP4, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, OR TNFRSF1A.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other features, aspects, and advantages of the present disclosure will become better understood with regard to the following description, and accompanying drawings, where:

FIG. 1A is a block diagram of a process for identifying subtypes of dysregulated host response patients, building a patient subtype classifier, and evaluating efficacy of corticosteroid therapy for dysregulated host response patients based on subtype classifications identified using the patient subtype classifier, in accordance with an embodiment.

FIG. 1B is a system environment overview for determining a therapy recommendation for a patient, in accordance with an embodiment.

FIG. 2 is a graph of the individual accuracies determined for each combination of three biomarkers, with one biomarker from each subtype, for the Full Model.

FIG. 3 is a graph of the individual accuracies determined for each combination of three biomarkers, with one biomarker from each subtype, for the SS Model.

FIG. 4 is a graph of the individual accuracies determined for each combination of three biomarkers, with one biomarker from each subtype, for the S Model.

FIG. 5 is a graph of the individual accuracies determined for each combination of three biomarkers, with one biomarker from each subtype, for the P Model.

FIGS. 6A-6D are graphs of individual accuracies determined for each combination of three biomarkers, with one biomarker from each subtype, for the SS.B1, SS.B2, SS.B3, and SS.B4 models, respectively.

FIG. 7 is an example flow process for determining therapeutic hypotheses for patient subtypes, in accordance with an embodiment.

FIG. 8 depicts the conclusions of the further analysis of Tables 6 and 7, in accordance with an embodiment.

FIG. 9 depicts a heat map depicting differential expression of genes from Table 6 for dysregulated host response patients having subtypes A, B, and C, and for healthy subjects without dysregulated host response, in accordance with an embodiment.

FIG. 10 depicts risk of morality for dysregulated host response patients having subtypes A, B, and C, in accordance with an embodiment.

FIG. 11 depicts differential expression of the genes of Table 7 that are associated with pharmacology of hydrocortisone therapy (e.g., regulation of the glucocorticoid receptor signaling pathway) for the subtypes A, B, and C, in accordance with an embodiment.

FIG. 12 provides support for a hypothesis of differential response to checkpoint inhibition therapy between the subtypes A, B, and C, by depicting differential expression of genes of Table 7 that are associated with pharmacology of checkpoint inhibition therapy (e.g., regulation of immune checkpoints and related immune functions mediated by cytokines) for subtypes A, B, and C, in accordance with an embodiment.

FIG. 13 depicts an example of a precision platform clinical trial design, in accordance with an embodiment.

FIG. 16 illustrates an example computer for implementing the methods described herein, in accordance with an embodiment.

The figures depict various embodiments of the present disclosure for purposes of illustration only. One skilled in the art will readily recognize from the following discussion that alternative embodiments of the structures and methods illustrated herein can be employed without departing from the principles of the disclosure described herein.

DETAILED DESCRIPTION
I. Definitions

In general, terms used in the claims and the specification are intended to be construed as having the plain meaning understood by a person of ordinary skill in the art. Certain terms are defined below to provide additional clarity. In case of conflict between the plain meaning and the provided definitions, the provided definitions are to be used.

The term “patient” or “subject” encompasses or organism, mammals including humans or non-humans (e.g., non-human primates, canines, felines, murines, bovines, equines, and porcines), whether in vivo, ex vivo, or in vitro, male or female.

The term “sample” can include a single cell or multiple cells or fragments of cells or an aliquot of body fluid, such as a blood sample, taken from a subject, by means including venipuncture, excretion, ejaculation, massage, biopsy, needle aspirate, lavage sample, scraping, surgical incision, or intervention or other means known in the art. Examples of an aliquot of body fluid include amniotic fluid, aqueous humor, bile, lymph, breast milk, interstitial fluid, blood, blood plasma, cerumen (earwax), Cowper's fluid (pre-ejaculatory fluid), chyle, chyme, female ejaculate, menses, mucus, saliva, urine, vomit, tears, vaginal lubrication, sweat, serum, semen, sebum, pus, pleural fluid, cerebrospinal fluid, synovial fluid, intracellular fluid, and vitreous humour.

The terms “marker,” “markers,” “biomarker,” and “biomarkers” encompass, without limitation, lipids, lipoproteins, proteins, cytokines, chemokines, growth factors, peptides, nucleic acids, genes, and oligonucleotides, together with their related complexes, metabolites, mutations, variants, polymorphisms, modifications, fragments, subunits, degradation products, elements, and other analytes or sample-derived measures. In particular embodiments discussed herein, the biomarkers are genes. However, in alternative embodiments, the biomarkers can include any other measurable substance in a sample from a subject. A marker can also include mutated proteins, mutated nucleic acids, variations in copy numbers, and/or transcript variants, in circumstances in which such mutations, variations in copy number and/or transcript variants are useful for generating a predictive model, or are useful in predictive models developed using related markers (e.g., non-mutated versions of the proteins or nucleic acids, alternative transcripts, etc.). In some embodiments, the biomarkers discussed throughout this disclosure can include a nucleic acid, including DNA, modified (e.g., methylated) DNA, cDNA, and RNA, including coding (e.g., mRNAs) and non-coding RNA (e.g., sncRNAs), a protein, including a post-transcriptionally modified protein (e.g., phosphorylated, glycosylated, myristilated, etc. proteins), a nucleotide (e.g., adenosine triphosphate (ATP), adenosine diphosphate (ADP), and adenosine monophosphate (AMP)), including cyclic nucleotides such as cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP), a biologic, an ADC, a small molecule, such as oxidized and reduced forms of nicotinamide adenine dinucleotide (NADP/NADPH), a volatile compound, and any combination thereof.

The term “antibody” is used in the broadest sense and specifically covers monoclonal antibodies (including full length monoclonal antibodies), polyclonal antibodies, multi specific antibodies (e.g., bispecific antibodies), and antibody fragments that are antigen-binding so long as they exhibit the desired biological activity, e.g., an antibody or an antigen-binding fragment thereof.

“Antibody fragment,” and all grammatical variants thereof, as used herein are defined as a portion of an intact antibody comprising the antigen binding site or variable region of the intact antibody, wherein the portion is free of the constant heavy chain domains (i.e. CH2, CH3, and CH4, depending on antibody isotype) of the Fc region of the intact antibody. Examples of antibody fragments include Fab, Fab′, Fab′-SH, F(ab′)₂, and Fv fragments; diabodies; any antibody fragment that is a polypeptide having a primary structure consisting of one uninterrupted sequence of contiguous amino acid residues (referred to herein as a “single-chain antibody fragment” or “single chain polypeptide”).

The term “obtaining or having obtained quantitative data” encompasses obtaining a set of data determined from at least one sample. Obtaining a dataset encompasses obtaining a sample and processing the sample to experimentally determine the data. The phrase also encompasses receiving a set of data, e.g., from a third party that has processed the sample to experimentally determine the dataset. Additionally, the phrase encompasses mining data from at least one database or at least one publication or a combination of databases and publications. A dataset can be obtained by one of skill in the art via a variety of known ways including stored on a storage memory.

Any terms not directly defined herein shall be understood to have the meanings commonly associated with them as understood within the art of the disclosure. Certain terms are discussed herein to provide additional guidance to the practitioner in describing the compositions, devices, methods and the like of aspects of the disclosure, and how to make or use them. It will be appreciated that the same thing can be said in more than one way. Consequently, alternative language and synonyms can be used for any one or more of the terms discussed herein. No significance is to be placed upon whether or not a term is elaborated or discussed herein. Some synonyms or substitutable methods, materials and the like are provided. Recital of one or a few synonyms or equivalents does not exclude use of other synonyms or equivalents, unless it is explicitly stated. Use of examples, including examples of terms, is for illustrative purposes only and does not limit the scope and meaning of the aspects of the disclosure herein.

Additionally, as used in the specification, the singular forms “a,” “an” and “the” include plural referents unless the context clearly dictates otherwise.

II. Overview: Biomarker Panels for Guiding Dysregulated Host Response Therapy

FIG. 1A is a block diagram of a process for identifying subtypes of dysregulated host response patients (row 1), building patient subtype classifiers (row 2), and evaluating efficacy of therapies for dysregulated host response patients based on subtype classifications identified using the patient subtype classifiers (row 3), in accordance with an embodiment. To identify subtypes of dysregulated host response patients, working datasets compiled from historical transcriptomic data from sepsis patients were created as described in further detail below. Then, clustering analysis was performed on the working dataset to identify subtypes of dysregulated host response patients based on differential biomarker expression. These clusters are labeled (e.g., subtype A, subtype B, subtype C, etc.) such that the data can be used for training and building a model (second row).

In various embodiments, the process of building a model that predicts patient subtypes, hereafter referred to as a patient subtype classifier, involves using the labeled data. The labeled data is analyzed to select biomarkers (e.g., “gene selection” as shown in FIG. 1A) that are informative for predicting certain patient subtypes. In various embodiments, patient subtype classifiers were trained using the labeled training data using. As depicted in the embodiment in FIG. 1A, the patient subtype classifier (depicted as a triangle) can be trained to classify a patient into one of three subtypes (e.g., subtype A, subtype B, and subtype C). In some embodiments, fewer (e.g., two subtypes) or additional (e.g., more than three) subtypes can be predicted by the patient subtype classifier. The patient subtype classifier can undergo validation using a test dataset (e.g., dataset other than the labeled training data) to ensure sufficient classifier performance

The trained patient subtype classifiers can be deployed to classify specific patients. In one embodiment, the patient subtype classifier analyzes data derived from randomized controlled trial (RCT) data pertaining to one or more patients and outputs predictions for the patients. For example, the patient subtype classifier analyzes quantitative biomarker expression data for patients that have been involved in a randomized controlled trial and classifies the patients in one of the different subtypes.

IIA. System Environment Overview

FIG. 1B depicts an overview of a system environment for determining a therapy recommendation 140 for a patient 110, in accordance with an embodiment. The system environment 100 provides context in order to introduce a marker quantification assay 120 and a patient classification system 130.

In various embodiments, a test sample is obtained from the subject 110. The test sample is analyzed to determine quantitative values of one or more biomarkers by performing the marker quantification assay 120. The marker quantification assay 120 may be a quantitative reverse transcription polymerase chain reaction (RT-PCR) assay, a microarray, a sequencing assay, or an immunoassay, examples of which are described in further detail below. The quantitative values of biomarkers can be quantified RT-PCR data, transcriptomics data, and/or RNA-seq data. The quantified expression values of the biomarkers are provided to the patient classification system 130.

Generally, the patient classification system 130 includes one or more computers, such as example computer 1600 as discussed below with respect to FIG. 16. Therefore, in various embodiments, the steps described in reference to the patient classification system 130 are performed in silico. The patient classification system 130 analyzes the received biomarker expression values from the marker quantification assay 120. In various embodiments, the patient classification system 130 determines a classification for the patient 110. For example, a classification for the patient 110 can be one of multiple subtypes characterized by the quantitative biomarkers of the patient 110. In various embodiments, the patient classification system 130 determines a therapy recommendation 140 for the patient 110. In such embodiments, the patient classification system 130 determines a therapy recommendation 140 for the patient 110 based on a classification of the patient 110.

In various embodiments, the patient classification system 130 applies a patient subtype classifier to predict a classification for patient 110. In various embodiments, a patient subtype classifier can be a machine-learned model. In such embodiments, the patient classification system 130 can train the patient subtype classifier using training data and/or deploy the patient subtype classifier to analyze the quantitative expression values of biomarkers of the patient 110.

In various embodiments, the marker quantification assay 120 and the patient classification system 130 can be employed by different parties. For example, a first party performs the marker quantification assay 120 which then provides the results to a second party which implements the patient classification system 130. For example, the first party may be a clinical laboratory that obtains test samples from subjects 110 and performs the assay 120 on the test samples. The second party receives the expression values of biomarkers resulting from the performed assay 120 analyzes the expression values using the patient classification system 130.

In various embodiments, the patient classification system 130 can be a distributed computing system implemented in a cloud computing environment. For example, steps performed by the patient classification system 130 can be performed using systems in geographically different locations. In particular embodiments, the patient classification system 130 receives quantitative biomarker data from the marker quantification assay 120 at a first location. The patient classification system 130 transmits the quantitative biomarker data and analyzes the quantitative biomarker data to predict a classification using a patient subtype classifier at a second location (e.g., cloud computing). The patient classification system 130 can further transmit the classification back to the first location for subsequent use.

Cloud computing can be employed to offer on-demand access to the shared set of configurable computing resources. The shared set of configurable computing resources can be rapidly provisioned via virtualization and released with low management effort or service provider interaction, and then scaled accordingly. A cloud-computing model can be composed of various characteristics such as, for example, on-demand self-service, broad network access, resource pooling, rapid elasticity, measured service, and so forth. A cloud-computing model can also expose various service models, such as, for example, Software as a Service (“SaaS”), Platform as a Service (“PaaS”), and Infrastructure as a Service (“IaaS”). A cloud-computing model can also be deployed using different deployment models such as private cloud, community cloud, public cloud, hybrid cloud, and so forth. In this description and in the claims, a “cloud-computing environment” is an environment in which cloud computing is employed.

In various embodiments, the marker quantification assay 120 and patient classification system 130 are implemented in a critical care setting such that a therapy recommendation is to be generated for a patient 110 within a maximum amount of time. In various embodiments, the maximum amount of time is 30 minutes. In various embodiments, the maximum amount of time is 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 11 hours, or 12 hours. In other embodiments, the marker quantification assay 120 and patient classification system 130 are not implemented in a critical care setting.

IIB. Methods for Determining a Therapy Recommendation

In various embodiments, the patient classification system 130 (as described above in reference to FIG. 1B) analyzes quantitative data for a biomarker set, the quantitative data derived from a patient (e.g., patient 110 in FIG. 1B), and determines a therapy recommendation for the patient. Generally, the patient classification system 130 applies a patient subtype classifier that analyzes the quantitative data for the biomarker set and classifies the patient in a classification. The patient classification system 130 can determine a therapy recommendation for the patient based on the classification of the patient.

The patient classification system 130 receives quantitative data from the marker quantification assay 120. Here, the quantitative data from the marker quantification assay 120 can include quantitative levels of one or more biomarkers that were determined from a sample obtained from a patient. In various embodiments, the patient classification system 130 normalizes the quantitative data. For example the patient classification system 130 can normalize the quantitative data based on study-specific parameters (such that data is normalized for a study) and/or based on parameters specific for a particular assay or platform used to generate the quantitative data. In various embodiments, the patient classification system 130 can normalize the quantitative data according to normalization parameters derived the healthy samples. In such embodiments, the resulting quantitative data are normalized across patients and studies at the end of the normalization process. Such embodiments that involve normalizing quantitative data can be implemented during research settings (non-critical care settings). In some embodiments, the patient classification system 130 need not normalize the quantitative data prior to analysis by the patient subtype classifier. Such embodiments that do not involve normalizing the quantitative data can be implemented in critical care settings where a rapid analysis and classification is needed for a patient 110. The patient classification system 130 analyzes the quantitative data, which hereafter also encompasses normalized quantitative data.

As one example, the patient classification system 130 analyzes quantitative data for a biomarker set derived from a microarray analysis. The patient classification system 130 applies a patient subtype classifier that analyzes the quantitative microarray data and classifies the patient, which can later be used to determine a therapy recommendation. As another example, the patient classification system 130 analyzes qPCR data, which measures the relative or absolute expression level of biomarkers. In various embodiments, normalization or calibration processes are implemented. The quantitative data of the biomarker set are used to calculate the scores for different classifications (e.g., subtypes), which then will be used for subtype assignment by a patient subtype classifier. As another example, the patient classification system 130 analyzes RNA sequencing data, which includes relative expression levels of model genes and their transcripts. Using sequencing reads alignment methods (e.g. Hisat2, and Bowtie2), expression estimation methods (e.g. StringTie, Salmon) and normalization processes (e.g. quantile normalization), the estimated expression of model genes can be used to calculate classification-specific scores for downstream classification by a patient subtype classifier. In various embodiments, the patient classification system 130 can convert quantitative data derived from a first type of assay to quantitative data of a second type of assay using normalization factors. For example, the patient classification system 130 can convert quantitative data derived from microarray data to either qPCR data or RNA sequencing data. The conversion can entail one or more normalization factors involving normalization or calibration processes for qPCR data or normalization processes (e.g., quantile normalization) for RNA sequencing data. Thus, the patient classification system 130 can apply different patient subtype classifiers to analyze different types of quantitative data.

The patient classification system 130 implements the patient subtype classifier to analyze quantitative data for biomarkers. In one embodiment, the patient subtype classifier is a trained machine-learned model. Thus, the patient subtype classifier can be trained to receive, as input, quantitative data of a biomarker set, and analyze the input to output a classification for the patient. In some embodiments, the patient subtype classifier is not a machine-learned model. In various embodiments, patient subtype classifier outputs a prediction of one classification for the patient out of X possible classifications. For example, the patient subtype classifier can output a prediction of a patient subtype for the patient out of a possible X patient subtypes. In various embodiments, X may be two possible classifications. In various embodiments, X may be more than two possible classifications. In various embodiments, X may be three, four, five, six, seven, eight, nine, or ten possible classifications. In various embodiments, X may be more than ten possible classifications.

In some embodiments, the patient classification system 130 calculates scores from the quantitative data and then provides the calculated scores as input to the patient subtype classifier. Thus, the patient subtype classifier determines a classification for the patient based on the calculated scores.

In various embodiments, the patient classification system 130 calculates multiple scores, each score corresponding to a patient subtype (e.g., classification). For example, if the goal is to classify the patient in a classification out of X possible classifications, the patient classification system 130 calculates X scores. The X scores are then provided as input to the patient subtype classifier to predict the classification. These scores are hereafter referred to as classification-specific scores.

In various embodiments, to calculate a classification-specific score, the patient classification system 130 determines subscores derived from quantitative data of one or more biomarkers in the biomarker set and uses the subscores to determine the classification-specific score. In one embodiment, a subscore is calculated from one or more biomarkers that are differentially expressed in the patient in comparison to a control value. In various embodiments, the control value may be a value derived from a different set of patients, such as healthy patients. In various embodiments, the control value may be a baseline value derived from the same patient (e.g., a baseline value corresponding to when the same patient was previously healthy).

In various embodiments, the patient classification system 130 determines a subscore determined from quantitative data of one or more biomarkers that are upregulated in the patient in comparison to the control value. In various embodiments, the patient classification system 130 determines a subscore determined from quantitative data of one or more biomarkers that are downregulated in the patient in comparison to the control value. In various embodiments, the patient classification system 130 determines a first subscore determined from quantitative data of one or more biomarkers that are upregulated in the patient in comparison to the control value and further determines a second subscore determined from quantitative data of one or more biomarkers that are downregulated in the patient in comparison to a control value. In various embodiments, a subscore can be an aggregation of the quantitative data of the one or more biomarkers. For example, a subscore can be a mean, a median, or a geometric mean of quantitative data of the one or more biomarkers. In various embodiments, the patient classification system 130 can further scale the sub scores.

In various embodiments, the quantitative data of one or more biomarkers that are analyzed refer to biomarkers that have been previously categorized as influencing the particular subtype that the classification-specific score is being calculated for. For example, if the patient classification system 130 is determining a classification-specific for subtype A, the patient classification system 130 determines subscores using quantitative data of biomarkers that are categorized as influencing the subtype A. Examples of biomarkers that are categorized with certain subtypes are shown below in Tables 1, 2A-2B, 3, and 4A-4D. Specifically, row number 1 in each of Tables 1, 2A-2B, 3, and 4A-4D show biomarkers that are categorized with subtype A, row number 2 in each of Tables 1, 2A-2B, 3, and 4A-4D show biomarkers that are categorized with subtype B, and row number 3 in each of Tables 1, 2A-2B, 3, and 4A-4D show biomarkers that are categorized with subtype C.

In various embodiments, the patient classification system 130 combines one or more subscores to determine the classification-specific score. For example, the patient classification system 130 can determine a difference between a first subscore and a second subscore. The difference can represent the classification-specific score.

As a specific example, the patient classification system 130 can determine a classification-specific score using the following steps: the patient classification system 130 determines a first geometric mean of the quantitative expression data for the subject for one or more biomarkers of the candidate classification, wherein the quantitative expression data for the subject for the one or more biomarkers of the candidate classification are increased relative to the quantitative expression data for the one or more biomarkers for one or more control subjects. The patient classification system 130 determines a second geometric mean of the quantitative expression for the subject for one or more additional biomarkers of the candidate classification, wherein the quantitative expression data for the subject for the one or more additional biomarkers of the candidate classification are decreased relative to the quantitative expression data for the one or more additional biomarkers for the one or more control subjects. The patient classification system 130 determines a difference between the first geometric mean and the second geometric mean, the first and second geometric means optionally subject to scaling. Here, the difference can represent the classification-specific score.

In various embodiments, the patient classification system 130 determines multiple classification-specific scores and provides them as input to the patient subtype classifier. The patient subtype classifier analyzes the classification-specific scores and outputs a classification for the patient. Embodiments of the patient subtype classifier are described in further detail below.

In various embodiments, based on the classification-specific scores, the patient subtype classifier outputs a classification. For example, the patient subtype classifier may analyze X classification-specific scores and outputs a prediction for one class out of Xpossible classifications. As another, the patient subtype classifier may analyze X classification-specific scores and outputs a prediction for one class out of two possible classifications. As a specific example, the patient subtype classifier may analyze 3 classification-specific scores (e.g., specific for subtype A, subtype B, and subtype C), and outputs a prediction for a class out of two possible classifications (e.g., subtype A v. not subtype A, subtype B v. not subtype B, or subtype C v. not subtype C).

Generally, the classification determined by the patient subtype classifier guides the selection of a therapy recommendation. In various embodiments, the therapy recommendation refers to whether a therapy is likely to be beneficial to a patient. In particular embodiments, the disease of interest is sepsis and therefore, the therapy recommendation pertain to whether a corticosteroid therapy, such as hydrocortisone, is likely to be of benefit to a patient. In one embodiment, the therapy recommendation can indicate whether the patient is likely to be “favorably responsive” or “non-responsive” to a therapy. In one embodiment, the therapy recommendation can indicate whether the patient is likely to be “favorably responsive”, “adversely responsive”, or “non-responsive” to a therapy.

Examples of a therapy include: immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, anti-inflammatory therapy, a checkpoint inhibitor, a blocker of complement components, a blocker of complement component receptors, a blocker of a pro-inflammatory cytokine, modulators of coagulation therapy, and modulators of vascular permeability therapy. Additional examples of a therapy include: GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, anti-C5a, anti-C3a, anti-C5aR, anti-C3aR, anti-TNF-alpha, and anti-IL-6, Anti-HMGB1, ST2 antibody, IL-33 antibody, activated protein C, antithrombin, and thrombomodulin.

Additional examples of a therapy and corresponding therapy recommendations for different patient subtypes (e.g., subtype A, subtype B, and subtype C) are shown below in Table 8. Specifically, the therapy recommendations are shown in the column titled “Subtype Hypothesis” and support for that hypothesis is found in the column titled “Evidence.” Altogether the therapy recommendation determined by the patient classification system 130 can be provided to guide therapy for the patient.

The impact of a particular therapy and a patient subtype, such as those hypothesized in Table 8, may have been previously determined by analyzing patient cohorts who have received the particular therapy. For example, such patient cohorts may have been involved in a clinical trial. Thus, the patients may be exhibiting dysregulated host responses and therefore, were enrolled in the trial. Therefore, patients in the clinical trial are classified with a patient subtype (e.g., using the methods described above) and their responses to the therapy (e.g., favorably responsive, adversely responsive, non-responsive) are tracked and recorded. For each subtype, the responses of patients receiving the therapy are compared to control patients. If the comparison yields a statistically significant difference patients of the subtype are labeled as favorably responsive or adversely responsive to the therapy. If the comparison does not yield a statistically significant difference (e.g., p-value not greater than a threshold value), patients of the subtype are labeled as non-responsive to the therapy. In various embodiments, the statistical significance threshold is a p-value, where the p-value is any one of 0.01, 0.0.5, or 0.1.

In particular embodiments, the compared measurable on which statistical significance is determined is patient mortality. Therefore, the mortality of patients who receive a therapy is compared to mortality of control patients to determine whether there is statistical significance indicating an effect due to the therapy. For example, if the patients of a subtype who receive a therapy exhibit a statistically significantly increased survival time in comparison to control patients who did not receive the therapy, then patients of this subtype can be identified as favorably responsive to the therapy. As another example, if patients of a subtype who receive a therapy exhibit a statistically significantly decreased survival time in comparison to control patients who did not receive the therapy, then patients of this subtype can be identified as adversely responsive to the therapy. As yet another example, if patients of a subtype who receive a therapy do not exhibit a statistically significantly increased or a statistically significantly decreased survival time in comparison to control patients who did not receive the therapy, then patients of this subtype can be identified as not responsive to the therapy.

IIB. Methods for Determining a Therapy Hypothesis

In various embodiments, methods disclosed herein involve the identification of therapeutic hypotheses for different patient subtypes. In various embodiments, the process of identifying a therapeutic hypothesis is performed by the patient classification system 130. In some embodiments, the process of identifying a therapeutic hypothesis is performed by third party system which provides a therapeutic hypothesis to the patient classification system 130. In various embodiments, a therapy hypothesis is specific for a patient subtype. Therefore, a therapy hypothesis is useful for identifying a therapy recommendation, as discussed above in reference to FIG. 1B.

Generally, a therapeutic hypothesis involves analyzing genes that are differentially expressed across different subtypes. By identifying patterns of differentially expressed genes that are implicated in certain known biological pathways, certain patient subtypes can be associated with particular dysregulated pathways. A therapeutic hypothesis comprising a candidate therapeutic can be selected. Here, a candidate therapeutic can modulate parts of the dysregulated pathways, thereby representing a possible avenue of therapy for treating particular patient subtypes.

Reference is now made to FIG. 7, which depicts an example flow process for determining therapeutic hypotheses for patient subtypes, in accordance with an embodiment. Generally, FIG. 7 depicts the use of labeled data 610 to generate differentially expressed gene data 620. The differentially expressed gene data 620 can be used to identify a therapeutic hypothesis 650. In some embodiments, the differentially expressed gene data 620 is analyzed together with therapeutic pharmacology data 630 and response pathobiology data 640 to determine the therapeutic hypothesis. In some embodiments, the differentially expressed gene data 620 is analyzed with one of therapeutic pharmacology data 630 or respond pathobiology data 640 to determine the therapeutic hypothesis 650. In some embodiments, only the differentially expressed gene data 620 is analyzed to determine the therapeutic hypothesis 650.

The labeled data 610 represents patient data that have been labeled with one or more classifications. For example, the labeled data 610 can be labeled with patient subtypes (e.g., subtype A, subtype B, subtype C, etc.). In various embodiments, the patient data comprises quantitative data of one or more biomarkers of patients. In various embodiments, the patient data is clinical trial data and therefore, the quantitative data of one or more biomarkers can be data obtained from patients enrolled in the clinical trial.

The labels of the labeled data can be previously generated through various means. In various embodiments, the labels of the data can be generated using a model, such as a patient subtype classifier described herein. For example, the quantitative data of biomarkers from patients are analyzed using the patient subtype classifier to predict a classification for patients. Thus, the predicted classification for each patient can serve as a label for the labeled data. In various embodiments, the labels of the data can be generated through a clustering analysis. For example, the quantitative data of biomarkers can be analyzed through unsupervised clustering, thereby generating clusters of patients that have similar expression of various biomarkers. Each cluster of patients can be labeled. In various embodiments, a cluster can be labeled based on outcomes of patients in the clinical trials. For example, if a majority of patients in a cluster exhibited prolonged survival time in response to a therapy, the cluster can be labeled as a subtype that is responsive to the therapy.

The differentially expressed gene data 620 comprises gene level fold changes of biomarker expression between patients of different subtypes. Using the labeled data 610, gene expression from patients of individual subtypes are aggregated and compared across subtypes. For example, a statistical measure of gene expression for patients of a subtype can be determined (e.g., a mean, a median, a mode, a geometric mean). The statistical measure of gene expression for patients of a first subtype are compared to a statistical measure of gene expression for patients of a second subtype. This can be performed across the different patient subtypes and across various genes. Thus, the differentially expressed gene data 620 includes gene level fold changes of different biomarkers across different patient subtypes.

In various embodiments, the differentially expressed gene data 620 includes gene level fold changes for at least twenty biomarkers. In various embodiments, the differentially expressed gene data 620 includes gene level fold changes for at least fifty biomarkers. In various embodiments, the differentially expressed gene data 620 includes gene level fold changes for at least 100 biomarkers, at least 200 biomarkers, at least 300 biomarkers, at least 400 biomarkers, at least 500 biomarkers, at least 1000 biomarkers, at least 2000 biomarkers, at least 3000 biomarkers, at least 4000 biomarkers, at least 5000 biomarkers, at least 10,000 biomarkers, at least 50,000 biomarkers, or at least 100,000 biomarkers.

In various embodiments, the differentially expressed gene data 620 can be represented as a database or a table that documents gene level fold changes between patients of different subtypes. An example of such a gene level fold changes between patient subtypes is shown below in Table 7. Specifically, for each gene, a gene level fold change (e.g., ratio) between different subtypes (e.g., subtype A/subtype B denoted as “A/B”) is shown.

To determine a therapeutic hypothesis 650, patterns of gene level fold changes are identified across the differentially expressed gene data 620. In various embodiments, patterns of gene level fold changes refer to at least a threshold number of genes that are differentially expressed in a first patient subtype in comparison to a second patient subtype. In various embodiments, patterns of gene level fold changes refer to at least a threshold number of genes that are overexpressed in a first patient subtype in comparison to a second patient subtype. In various embodiments, patterns of gene level fold changes refer to at least a threshold number of genes that are underexpressed in a first patient subtype in comparison to a second patient subtype.

In various embodiments, the threshold number of genes include genes that are involved in a common biological pathway. Example biological pathways include, but are not limited to: innate immune pathways, chronic inflammation pathways, acute inflammation pathways, coagulation pathways, complement pathways, signaling pathways (e.g., TLR signaling pathway or glucocorticoid receptor signaling pathway), and the like. In various embodiments, the involvement of genes in certain biological pathways is curated from publicly available databases such as the Reactome Pathway Database or the KEGG Pathway database.

In various embodiments, the threshold number of genes involved in a common biological pathway is at least 2 genes. In various embodiments, the threshold number of genes is at least 3 genes, at least 4 genes, at least 5 genes, at least 6 genes, at least 7 genes, at least 8 genes, at least 9 genes, at least 10 genes, at least 15 genes, at least 20 genes, at least 25 genes, at least 50 genes, at least 75 genes, at least 100 genes, at least 200 genes, at least 300 genes, at least 400 genes, at least 500 genes, or at least 1000 genes. In various embodiments, the threshold number of genes involved in a common biological pathway is 2 genes. In various embodiments, the threshold number of genes involved in a common biological pathway is 3 genes, 4 genes, 5 genes, 6 genes, 7 genes, 8 genes, 9 genes, 10 genes, 11 genes, 12 genes, 13 genes, 14 genes, 15 genes, 16 genes, 17 genes, 18 genes, 19 genes, 20 genes, 25 genes, 50 genes, 75 genes, 100 genes, 200 genes, 300 genes, 400 genes, 500 genes, 600 genes, 700 genes, 800 genes, 900 genes, or 1000 genes.

Altogether, patterns of gene level fold changes, as indicated by a threshold number of genes involved in a common biological pathway, are useful for understanding the underlying biology that may be involved in a patient subtype. For example, genes involved in inflammation may be differentially expressed in subtype A in comparison to those genes in subtype B. Thus, subtype A can be associated or characterized by inflammation based processes.

The patterns of gene level fold changes between subtypes is analyzed to determine a therapeutic hypothesis 650 which, in some scenarios, includes a class of a candidate therapeutic of a candidate therapeutic itself (e.g., including but not limited to a drug therapy or a gene therapy). For example, given the characterization that a particular patient subtype is associated with an underlying biological pathway or process, a target involved in the biological pathway or process can serve as a druggable target. Thus, a class of a candidate therapeutic or a candidate therapeutic that modulates the target involved in the biological pathway can be promising as a therapeutic hypothesis 650. Examples of a class of a therapy include, but are not limited to: immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, anti-inflammatory therapy, a checkpoint inhibitor, a blocker of complement components, a blocker of complement component receptors, a blocker of a pro-inflammatory cytokine, modulators of coagulation therapy, and modulators of vascular permeability therapy. Examples of a candidate therapy include but are not limited to: GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, anti-C5a, anti-C3a, anti-C5aR, anti-C3aR, anti-TNF-alpha, and anti-IL-6, Anti-HMGB1, ST2 antibody, IL-33 antibody, activated protein C, antithrombin, and thrombomodulin.

The therapeutic pharmacology data 630 is useful for developing a therapeutic hypothesis for a particular class of therapy or for a particular candidate therapy. Generally, therapeutic pharmacology data 630 is useful for understanding what therapeutic effects, if any, can be imparted by a class of therapy or candidate therapy. For example, therapeutic pharmacology data 630 can include molecular data of therapeutics, clinical pharmacology data of therapeutics (e.g., pharmacokinetics and pharmacodynamics data), and/or data identifying therapeutics that are useful for modulating activity in particular biological pathways. For example, for a given candidate therapeutic (e.g., an anti-PD-1 inhibitor), the therapeutic pharmacology data 630 is useful for understanding how different patients respond to the anti-PD-1 inhibitor.

Examples of therapeutic pharmacology data 630 is shown in FIG. 12. For example, PD-1 blockade is expected to up-regulate IL-7 and CTLA-4 blockade is expected to up-regulate INF-gamma and to stimulate immune activity more broadly. In patients with down-regulated immune activity, PD-L1 and CTLA-4 is up-regulated, while IL-7 and INF-gamma are down-regulated. Therefore, blockade of PD-1/PD-L1 will likely result in up-regulation of IL-7 and blockade of CTLA-4 upregulation of INF-gamma, and stimulation of immune activity more broadly.

The response pathobiology data 640 is useful for developing a hypothesis as to therapeutic effects, independent of a particular candidate therapeutic, that may benefit a particular patient subtype. In various embodiments, response pathobiology data 640 can include patient data corresponding to patients that responded favorably. In various embodiments, response pathobiology data 640 includes patient data of patient subtypes that indicate differential expression of biomarkers associated with certain biological activity. The differentially expressed biomarkers can be promising targets for modulation. For example, dysregulated host response patients of subtype A exhibit up-regulation of biomarkers associated with innate immune activity involved in pathogen recognition (e.g., via recognition of pathogen-associated molecular patterns (PAMPs)), up-regulation of biomarkers associated with innate immune regulation, and up-regulation of biomarkers associated with adaptive immune activity. As another example, dysregulated host response patients of subtype B exhibit up-regulation of biomarkers associated with innate immune activity involved in recognition of damage-associated molecular patterns (DAMPs), up-regulation of biomarkers associated with DAMPs, up-regulation of biomarkers associated with inflammation (e.g. TNF-alpha), up-regulation of biomarkers associated with complement activity, down-regulation of biomarkers associated with adaptive immune activity, up-regulation of biomarkers associated with adaptive immune suppression, and up-regulation of markers associated with increased risk of acute kidney injury. As another example, subtype C patients exhibit down-regulation of biomarkers associated with innate and adaptive immune activity, up-regulation of biomarkers associated with DAMPs, up-regulation of biomarkers associated with cellular recruitment (e.g. G-CSF and GM-CSF), up-regulation of biomarkers associated with increased risk of thrombosis, and up-regulation of biomarkers associated with coagulation.

The therapeutic hypothesis 650 for a patient subtype can be subsequently tested and validated. For example, the therapeutic hypothesis 650 can be tested in pre-clinical or clinical studies and trials (e.g., a randomized controlled trial) by providing subjects or patients of the subtype a candidate therapeutic and monitoring their response.

IIC. Patient Subtype Classifier

In various embodiments, the patient subtype classifier is a machine-learned model that analyzes quantitative data of biomarkers or classification-specific scores derived from quantitative data of biomarkers and predicts a classification. In various embodiments, the patient subtype classifier is any one of a regression model (e.g., linear regression, logistic regression, or polynomial regression), decision tree, random forest, support vector machine, Naïve Bayes model, k-means cluster, or neural network (e.g., feed-forward networks, convolutional neural networks (CNN), deep neural networks (DNN), autoencoder neural networks, generative adversarial networks, or recurrent networks (e.g., long short-term memory networks (LSTM), bi-directional recurrent networks, deep bi-directional recurrent networks), or any combination thereof.

The patient subtype classifier can be trained using a machine learning implemented method, such as any one of a linear regression algorithm, logistic regression algorithm, decision tree algorithm, support vector machine classification, Naïve Bayes classification, K-Nearest Neighbor classification, random forest algorithm, deep learning algorithm, gradient boosting algorithm, and dimensionality reduction techniques such as manifold learning, principal component analysis, factor analysis, autoencoder regularization, and independent component analysis, or combinations thereof. In various embodiments, the patient subtype classifier is trained using supervised learning algorithms, unsupervised learning algorithms, semi-supervised learning algorithms (e.g., partial supervision), weak supervision, transfer, multi-task learning, or any combination thereof.

In various embodiments, the patient subtype classifier has one or more parameters, such as hyperparameters or model parameters. Hyperparameters are generally established prior to training. Examples of hyperparameters include the learning rate, depth or leaves of a decision tree, number of hidden layers in a deep neural network, number of clusters in a k-means cluster, penalty in a regression model, and a regularization parameter associated with a cost function. Model parameters are generally adjusted during training. Examples of model parameters include weights associated with nodes in layers of neural network, support vectors in a support vector machine, and coefficients in a regression model. The model parameters of the patient subtype classifier are trained (e.g., adjusted) using the training data to improve the predictive capacity of the patient subtype classifier.

In some embodiments, the patient subtype classifier is a regression, such as a logistic regression. Parameters of the logistic regression are trained using the training data such that when the logistic regression is applied, it outputs a classification based on the different classification-specific scores. The parameters of the logistic regression can be trained to maximize the differences between the different classifications (e.g., subtype A, subtype B, and subtype C).

In some embodiments, the patient subtype classifier is a support vector machine. The support vector machine is trained with a single or a set of hyperplanes that maximizes the differences among the X different classifications. In one embodiment, the support vector machine is trained with single or a set of hyperplanes that maximizes the differences among 3 different classifications (e.g., subtype A, subtype B, and subtype C). As a specific example, the support vector machine is trained with a set of hyperplanes that maximizes the differences among the 3 different classification-specific scores (e.g., scores for each of subtype A, subtype B, and subtype C). Therefore, the trained support vector machine can use the hyperplanes to output a prediction of a classification when provided quantitative data of biomarkers or classification-specific scores derived from quantitative data of biomarkers.

In some embodiments, the patient subtype classifier may be a non-machine learned model. The patient subtype classifier may employ one or more threshold values for comparison against the classification-specific scores. Depending on the comparison between the threshold values and the classification-specific scores, the patient subtype classifier outputs a predicted classification. In various embodiments, a threshold value is specific for a classification. Therefore, there may be X threshold values to be compared against X classification-specific scores.

In some embodiments, a threshold value may be a fixed value (e.g., fixed value=0). Here, the classification-specific scores are compared to the fixed threshold value and patient subtype classifier determines the classification based on the comparison. For example, assuming there are two classification-specific scores, the patient subtype classifier may compare each of the first classification-specific score and the second classification-specific score to the fixed threshold. In one embodiment, if the first classification-specific score is greater than the fixed threshold value and the second classification-specific score is less than a fixed threshold value, then the patient subtype classifier can output a particular classification. Similar logic can be applied for determining classifications using more than two classification-specific scores.

In some embodiments, a threshold value may be determined from training samples including data from patients who have been classified (e.g., classified as subtype A, subtype B, and/or subtype C). Such a threshold value may derived from a receiver operating curve (ROC) demonstrating the sensitivity/specificity of a model that classified the patients of the training samples. For example, for patients in the training sample classified as subtype A, a receiver operating curve is generated that demonstrates the sensitivity and specificity of the classifier. The threshold value can be the top-left part of the plot, representing the closest point in the ROC to perfect sensitivity or specificity.

The classification-specific scores are compared to corresponding threshold values, and based on the comparison, the patient subtype classifier determines the classification. For example, assuming there are two classification-specific scores for subtype A and subtype B, the patient subtype classifier may compare the subtype A classification-specific score to a subtype A threshold value and may further compare the subtype B classification-specific score to a subtype B threshold value. Thus, depending on the two comparisons, the patient subtype classifier determines the classification. In one embodiment, if the first classification-specific score is greater than the first threshold value and the second classification-specific score is less than a second threshold value, then the patient subtype classifier can output a particular classification. Similar logic can be applied for determining classifications using more than two classification-specific scores and/or more than two threshold values. Examples of subtype specific threshold values that are derived from training samples are described below in Table 18.

IID. Biomarker Panel

Embodiments described herein involve the analysis of biomarkers. As described herein, a biomarker panel, also referred to as a biomarker set, can be implemented to analyze values of biomarkers for a patient. In various embodiments, a biomarker panel can be a multivariate biomarker panel. In such embodiments, the multivariate biomarker panel includes more than one biomarker. In various embodiments, the multivariate biomarker panel includes two biomarkers. In various embodiments, the multivariate biomarker panel includes 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50 biomarkers. In particular embodiments, the multivariate biomarker panel includes 3 biomarkers. In particular embodiments, the multivariate biomarker panel includes 4 biomarkers. In particular embodiments, the multivariate biomarker panel includes 5 biomarkers. In particular embodiments, the multivariate biomarker panel includes 6 biomarkers. In particular embodiments, the multivariate biomarker panel includes 8 biomarkers. In particular embodiments, the multivariate biomarker panel includes 10 biomarkers. In particular embodiments, the multivariate biomarker panel includes 15 biomarkers. In particular embodiments, the multivariate biomarker panel includes 16 biomarkers. In particular embodiments, the multivariate biomarker panel includes 24 biomarkers.

In various embodiments, the multivariate biomarker panel includes biomarkers selected from the following markers: EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, TOMM70A, ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, ANXA3, C14orf159, PUM2, EPB42, RPS6KA5, GBP2, MSH2, DCTD, HK3, UCP2, NUP88, GABARAPL2, and CASP4.

In various embodiments, the multivariate biomarker panel includes at least two biomarkers selected from the following markers: EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, TOMM70A, ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, ANXA3, C14orf159, PUM2, EPB42, RPS6KA5, GBP2, MSH2, DCTD, HK3, UCP2, NUP88, GABARAPL2, and CASP4.

In various embodiments, the multivariate biomarker panel include X number of biomarkers, where X is the number of possible classifications that the patient subtype classifier can predict. For example, for a patient subtype classifier that predicts three different subtypes (e.g., subtype A, subtype B, and subtype C), the multivariate biomarker panel can include three different biomarkers.

In various embodiments, the multivariate biomarker panel includes a first biomarker selected from EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, a second biomarker selected from SERPINB1 and GSPT1, and a third biomarker selected from MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A. Example accuracies of a multivariate biomarker panel implementing combinations of three biomarkers described above is shown in FIG. 2.

In various embodiments, the multivariate biomarker panel includes one or more biomarkers selected from EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, one or more biomarkers selected from SERPINB1 and GSPT1, and one or more biomarkers selected from MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A.

In one embodiment, the multivariate biomarker panel includes a first biomarker selected from ZNF831, MME, CD3G, and STOM, a second biomarker selected from ECSIT, LAT, and NCOA4, and a third biomarker selected from SLC1A5, IGF2BP2, and ANXA3. Example accuracies of a multivariate biomarker panel implementing combinations of three biomarkers described above is shown in FIG. 3.

In one embodiment, the multivariate biomarker panel includes a first biomarker selected from C14orf159 and PUM2, a second biomarker selected from EPB42 and RPS6KA5, and a third biomarker selected from GBP2. Example accuracies of a multivariate biomarker panel implementing combinations of three biomarkers described above is shown in FIG. 4.

In one embodiment, the multivariate biomarker panel includes a first biomarker selected from MSH2, DCTD, and MMP8, a second biomarker selected from HK3, UCP2, and NUP88, and a third biomarker selected from GABARAPL2 and CASP4. Example accuracies of a multivariate biomarker panel implementing combinations of three biomarkers described above is shown in FIG. 5.

In one embodiment, the multivariate biomarker panel includes a first biomarker selected from STOM, ZNF831, CD3G, MME, BTN3A2, and HLA-DPA1, a second biomarker selected from EPB42, GSPT1, LAT, HK3, and SERPINB1, and a third biomarker selected from GBP2, TNFRSF1A, SLC1A5, IGF2BP2, and ANXA3. Example accuracies of a multivariate biomarker panel implementing combinations of three biomarkers described above is shown in FIG. 6A.

In one embodiment, the multivariate biomarker panel includes a first biomarker selected from STOM, ZNF831, CD3G, MME, BTN3A2, and HLA-DPA1, a second biomarker selected from EPB42, GSPT1, LAT, HK3, and SERPINB1, and a third biomarker selected from GBP2, SLC1A5, IGF2BP2, and ANXA3. Example accuracies of a multivariate biomarker panel implementing combinations of three biomarkers described above is shown in FIG. 6B.

In one embodiment, the multivariate biomarker panel includes a first biomarker selected from STOM, MME, BTN3A2, HLA-DPA1, and EVL, a second biomarker selected from EPB42, GSPT1, LAT, HK3, and SERPINB1, and a third biomarker selected from BTN3A2, TNFRSF1A, SLC1A5, IGF2BP2, and ANXA3. Example accuracies of a multivariate biomarker panel implementing combinations of three biomarkers described above is shown in FIG. 6C.

In one embodiment, the multivariate biomarker panel includes a first biomarker selected from STOM, MME, BTN3A2, HLA-DPA1, and EVL, a second biomarker selected from EPB42, GSPT1, LAT, HK3, and SERPINB1, and a third biomarker selected from GBP2, SLC1A5, IGF2BP2, and ANXA3. Example accuracies of a multivariate biomarker panel implementing combinations of three biomarkers described above is shown in FIG. 6D.

Although the embodiments described above may refer to a “first biomarker,” “second biomarker,” and/or “third biomarker,” the terms “first biomarker,” “second biomarker,” and/or “third biomarker,” each encompass one or more biomarkers. For example, a “first biomarker” can refer to one or more biomarkers selected from EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8. A “second biomarker” can refer to one or more biomarkers selected from SERPINB1 and GSPT1. A “third biomarker” can refer to one or more biomarkers selected from MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A.

In one embodiment, the multivariate biomarker panel includes four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty, twenty one, twenty two, twenty three, or twenty four biomarkers selected from EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A.

In one embodiment, the multivariate biomarker panel includes four, five, six, seven, eight, nine, or ten biomarkers selected from ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, and ANXA3.

In one embodiment, the multivariate biomarker panel includes four or five biomarkers selected from C14orf159, PUM2, EPB42, RPS6KA5, and GBP2.

In one embodiment, the multivariate biomarker panel includes four, five, six, seven, or eight biomarkers selected from MSH2, DCTD, MMP8, HK3, UCP2, NUP88, GABARAPL2 and CASP4.

In one embodiment, the multivariate biomarker panel includes four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen or sixteen biomarkers selected from STOM, ZNF831, CD3G, MME, BTN3A2, HLA-DPA1, EPB42, GSPT1, LAT, HK3, SERPINB1, GBP2, TNFRSF1A, SLC1A5, IGF2BP2, and ANXA3.

In one embodiment, the multivariate biomarker panel includes four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, or fifteen biomarkers selected from STOM, ZNF831, CD3G, MME, BTN3A2, HLA-DPA1, EPB42, GSPT1, LAT, HK3, SERPINB1, GBP2, SLC1A5, IGF2BP2, and ANXA3.

In one embodiment, the multivariate biomarker panel includes four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, or fifteen biomarkers selected from STOM, MME, BTN3A2, HLA-DPA1, EVL, EPB42, GSPT1, LAT, HK3, SERPINB1, BTN3A2, TNFRSF1A, SLC1A5, IGF2BP2, and ANXA3.

In one embodiment, the multivariate biomarker panel includes four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, or fifteen biomarkers selected from STOM, MME, BTN3A2, HLA-DPA1, EVL, EPB42, GSPT1, LAT, HK3, SERPINB1, GBP2, SLC1A5, IGF2BP2, and ANXA3.

IIE. Assays

As shown in FIG. 1B, the system environment 100 involves implementing a marker quantification assay 120 for determining quantitative data for one or more biomarkers. Examples of an assay (e.g., marker quantification assay 120) for one or more markers include DNA assays, microarrays, polymerase chain reaction (PCR), RT-PCR, Southern blots, Northern blots, antibody-binding assays, enzyme-linked immunosorbent assays (ELISAs), flow cytometry, protein assays, Western blots, nephelometry, turbidimetry, chromatography, mass spectrometry, immunoassays, including, by way of example, but not limitation, RIA, immunofluorescence, immunochemiluminescence, immunoelectrochemiluminescence, or competitive immunoassays, immunoprecipitation. The information from the assay can be quantitative and sent to a computer system as described in further detail in reference to FIG. 16. The information can also be qualitative, such as observing patterns or fluorescence, which can be translated into a quantitative measure by a user or automatically by a reader or computer system.

In various embodiments, the assay can be any one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction. In particular embodiments, the assay is a RT-qPCR assay or a LAMP assay. For example, in a critical care setting where a classification and therapy recommendation is to be rapidly developed for a patient (e.g., within 30 minutes or within 2 hours), assay can be RT-qPCR or a LAMP assay that enables rapid quantification of the biomarkers in a sample obtained from the patient.

In various embodiments, the marker quantification assay 120 involves performing sequencing to obtain sequence reads (e.g., sequence reads for generating a sequencing library). The sequence reads can be quantified to determine quantitative data of biomarkers. Sequence reads can be achieved with commercially available next generation sequencing (NGS) platforms, including platforms that perform any of sequencing by synthesis, sequencing by ligation, pyrosequencing, using reversible terminator chemistry, using phospholinked fluorescent nucleotides, or real-time sequencing. As an example, amplified nucleic acids may be sequenced on an Illumina MiSeq platform.

When pyrosequencing, libraries of NGS fragments are cloned in-situ amplified by capture of one matrix molecule using granules coated with oligonucleotides complementary to adapters. Each granule containing a matrix of the same type is placed in a microbubble of the “water in oil” type and the matrix is cloned amplified using a method called emulsion PCR. After amplification, the emulsion is destroyed and the granules are stacked in separate wells of a titration picoplate acting as a flow cell during sequencing reactions. The ordered multiple administration of each of the four dNTP reagents into the flow cell occurs in the presence of sequencing enzymes and a luminescent reporter, such as luciferase. In the case where a suitable dNTP is added to the 3′ end of the sequencing primer, the resulting ATP produces a flash of luminescence within the well, which is recorded using a CCD camera. It is possible to achieve a read length of more than or equal to 400 bases, and it is possible to obtain 10⁶readings of the sequence, resulting in up to 500 million base pairs (megabytes) of the sequence. Additional details for pyrosequencing is described in Voelkerding et al., Clinical Chem., 55: 641-658, 2009; MacLean et al., Nature Rev. Microbiol., 7: 287-296; U.S. Pat. Nos. 6,210,891; 6,258,568; each of which is hereby incorporated by reference in its entirety.

On the Solexa/Illumina platform, sequencing data is produced in the form of short readings. In this method, fragments of a library of NGS fragments are captured on the surface of a flow cell that is coated with oligonucleotide anchor molecules. An anchor molecule is used as a PCR primer, but due to the length of the matrix and its proximity to other nearby anchor oligonucleotides, elongation by PCR leads to the formation of a “vault” of the molecule with its hybridization with the neighboring anchor oligonucleotide and the formation of a bridging structure on the surface of the flow cell. These DNA loops are denatured and cleaved. Straight chains are then sequenced using reversibly stained terminators. The nucleotides included in the sequence are determined by detecting fluorescence after inclusion, where each fluorescent and blocking agent is removed prior to the next dNTP addition cycle. Additional details for sequencing using the Illumina platform is found in Voelkerding et al., Clinical Chem., 55: 641-658, 2009; MacLean et al., Nature Rev. Microbiol., 7: 287-296; U.S. Pat. Nos. 6,833,246; 7,115,400; 6,969,488; each of which is hereby incorporated by reference in its entirety.

Sequencing of nucleic acid molecules using SOLiD technology includes clonal amplification of the library of NGS fragments using emulsion PCR. After that, the granules containing the matrix are immobilized on the derivatized surface of the glass flow cell and annealed with a primer complementary to the adapter oligonucleotide. However, instead of using the indicated primer for 3′ extension, it is used to obtain a 5′ phosphate group for ligation for test probes containing two probe-specific bases followed by 6 degenerate bases and one of four fluorescent labels. In the SOLiD system, test probes have 16 possible combinations of two bases at the 3′ end of each probe and one of four fluorescent dyes at the 5′ end. The color of the fluorescent dye and, thus, the identity of each probe, corresponds to a certain color space coding scheme. After many cycles of alignment of the probe, ligation of the probe and detection of a fluorescent signal, denaturation followed by a second sequencing cycle using a primer that is shifted by one base compared to the original primer. In this way, the sequence of the matrix can be reconstructed by calculation; matrix bases are checked twice, which leads to increased accuracy. Additional details for sequencing using SOLiD technology is found in Voelkerding et al., Clinical Chem., 55: 641-658, 2009; MacLean et al., Nature Rev. Microbiol., 7: 287-296; U.S. Pat. Nos. 5,912,148; 6,130,073; each of which is incorporated by reference in its entirety.

In particular embodiments, HeliScope from Helicos BioSciences is used. Sequencing is achieved by the addition of polymerase and serial additions of fluorescently-labeled dNTP reagents. Switching on leads to the appearance of a fluorescent signal corresponding to dNTP, and the specified signal is captured by the CCD camera before each dNTP addition cycle. The reading length of the sequence varies from 25-50 nucleotides with a total yield exceeding 1 billion nucleotide pairs per analytical work cycle. Additional details for performing sequencing using HeliScope is found in Voelkerding et al., Clinical Chem., 55: 641-658, 2009; MacLean et al., Nature Rev. Microbiol., 7: 287-296; U.S. Pat. Nos. 7,169,560; 7,282,337; 7,482,120; 7,501,245; 6,818,395; 6,911,345; 7,501,245; each of which is incorporated by reference in its entirety.

In some embodiments, a Roche sequencing system 454 is used. Sequencing 454 involves two steps. In the first step, DNA is cut into fragments of approximately 300-800 base pairs, and these fragments have blunt ends. Oligonucleotide adapters are then ligated to the ends of the fragments. The adapter serve as primers for amplification and sequencing of fragments. Fragments can be attached to DNA-capture beads, for example, streptavidin-coated beads, using, for example, an adapter that contains a 5′-biotin tag. Fragments attached to the granules are amplified by PCR within the droplets of an oil-water emulsion. The result is multiple copies of cloned amplified DNA fragments on each bead. At the second stage, the granules are captured in wells (several picoliters in volume). Pyrosequencing is carried out on each DNA fragment in parallel. Adding one or more nucleotides leads to the generation of a light signal, which is recorded on the CCD camera of the sequencing instrument. The signal intensity is proportional to the number of nucleotides included. Pyrosequencing uses pyrophosphate (PPi), which is released upon the addition of a nucleotide. PPi is converted to ATP using ATP sulfurylase in the presence of adenosine 5′phosphosulfate. Luciferase uses ATP to convert luciferin to oxyluciferin, and as a result of this reaction, light is generated that is detected and analyzed. Additional details for performing sequencing 454 is found in Margulies et al. (2005) Nature 437: 376-380, which is hereby incorporated by reference in its entirety.

Ion Torrent technology is a DNA sequencing method based on the detection of hydrogen ions that are released during DNA polymerization. The microwell contains a fragment of a library of NGS fragments to be sequenced. Under the microwell layer is the hypersensitive ion sensor ISFET. All layers are contained within a semiconductor CMOS chip, similar to the chip used in the electronics industry. When dNTP is incorporated into a growing complementary chain, a hydrogen ion is released that excites a hypersensitive ion sensor. If homopolymer repeats are present in the sequence of the template, multiple dNTP molecules will be included in one cycle. This results in a corresponding amount of hydrogen atoms being released and in proportion to a higher electrical signal. This technology is different from other sequencing technologies that do not use modified nucleotides or optical devices. Additional details for Ion Torrent Technology is found in Science 327 (5970): 1190 (2010); US Patent Application Publication Nos. 20090026082, 20090127589, 20100301398, 20100197507, 20100188073, and 20100137143, each of which is incorporated by reference in its entirety.

In various embodiments, immunoassays designed to quantitate markers can be used in screening including multiplex assays. Measuring the concentration of a target marker in a sample or fraction thereof can be accomplished by a variety of specific assays. For example, a conventional sandwich type assay can be used in an array, ELISA, RIA, etc. format. Other immunoassays include Ouchterlony plates that provide a simple determination of antibody binding. Additionally, Western blots can be performed on protein gels or protein spots on filters, using a detection system specific for the markers as desired, conveniently using a labeling method.

Protein based analysis, using an antibody that specifically binds to a polypeptide (e.g. marker), can be used to quantify the marker level in a test sample obtained from a subject. In various embodiments, an antibody that binds to a marker can be a monoclonal antibody. In various embodiments, an antibody that binds to a marker can be a polyclonal antibody. For multiplex analysis of markers, arrays containing one or more marker affinity reagents, e.g. antibodies can be generated. Such an array can be constructed comprising antibodies against markers. Detection can utilize one or a panel of marker affinity reagents, e.g. a panel or cocktail of affinity reagents specific for one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty, twenty one, twenty two, twenty three, twenty four or more markers.

In various embodiments, determining the quantitative expression data for each of the at least three biomarkers comprises: contacting the sample with a reagent; generating a plurality of complexes between the reagent and the plurality of biomarkers in the sample; and detecting the plurality of complexes to obtain a dataset associated with the sample, wherein the dataset comprises the quantitative expression data for the biomarker.

EXAMPLES
III. Dsyregulated Host Response Patient Subtypes

Custom processing of 14 datasets from sepsis studies from the literature was performed to identify dysregulated host response subtypes.⁴⁶For each study, patients were classified as either adult or pediatric. To distinguish between pediatric and adult patients, manual literature review was performed. Then, adult patients were classified as either sepsis (S) or septic shock (SS), septic shock being a subset of sepsis. To distinguish between adult sepsis and adult septic shock patients, the rate of patient vasopressor use reported in the literature (normally at the first day) was used. If the rate of patient vasopressor use was more than 50%, the whole study cohort was classified as septic shock. In contrast, if the rate of patient vasopressor use was less than 50%, the whole study cohort was classified as sepsis. Based on these classifications of adult or pediatric and sepsis or septic shock, patient samples were classified as Full samples (including adult, pediatric, sepsis, and septic shock patient samples), SS samples (including only adult septic shock patient samples), S samples (including only adult sepsis patient samples), and P samples (including only pediatric sepsis and septic shock patient samples).

Following classification of the patient samples from each literature study, for each study, biomarker expression data were normalized within the study and curated with methodologies specific to the study's array platform technology and to the study's available data format. Healthy control samples and patient samples were processed by the COCONUT framework,⁴⁷which normalized the samples with the same array platform and transformed patient expression data according to normalization parameters derived from the healthy samples. The resulting expression data were quantile normalized across patients and studies at the end of the normalization process.

The COINCIDE algorithm was then used to rank the genes based on the expression data.⁴⁷Then, for each set of classified patient samples (e.g., Full samples, SS samples, S samples, and P samples), for each subset of genes ranked (i.e. 100, 250, 500, 1000, 1500 genes, so on and so forth), the COMMUNAL clustering algorithm was used to identify the optimal number of clusters,^{47, 49}as well as to label each patient sample.⁴⁶COMMUNAL maps of cluster optimality were generated for each set of classified patient samples, in which the X-axis is the number of clusters, the Y-axis is the number of included genes, and the Z-axis is the mean validity score.

The COMMUNAL map of cluster optimality of a Full Model (including adult, pediatric, sepsis, and septic shock patient samples), exhibited three clusters for 574 out of 700 training samples. The remaining training samples were reported as inconclusive.

The COMMUNAL map of cluster optimality for a SS Model (including only adult septic shock patient samples), exhibited three clusters for 115 out of 165 training samples.

The COMMUNAL map of cluster optimality for a S Model (including only adult septic patient samples), exhibited four clusters for 153 out of 308 training samples. However, the fourth cluster did not reveal consistent results among different clustering algorithms.

The COMMUNAL map of cluster optimality for a P Model (including only pediatric sepsis and septic shock patient samples), exhibited three clusters for 180 out of 227 training samples.

Stable optima were consistently observed at K=3 clusters. Using the patient expression data and cluster labels, Gene Ontology (GO) analysis was performed to characterize the nature and functionality of each cluster, hereinafter referred to as a “subtype”. Subtypes were named as A (lower mortality, adaptive immune activation), B (higher mortality, innate immune activation), and C (higher mortality, older, and with clinical and molecular evidence of coagulopathy).⁴⁶The biological functions indicated in the GO analysis demonstrated distinct characteristics among the different subtypes, indicating high potential for guided treatment.

IV. Dysregulated Host Response Patient Subtype Classifiers

Eight classification Models, including the Full Model based on the Full samples, the SS Model based on the SS samples, the S Model based on the S samples, the P Model based on the P samples, as well as the SS.B1, SS.B2, SS.B3, and SS.B4 models were developed. As detailed below, to train the classification Models based on the associated samples, training labels for each training sample were determined using unsupervised clustering procedures including, normalization, the COCONUT method, the COINCIDE method, and the COMMUNAL method.

The methodology of building the classifiers was guided by a number of considerations, particularly in data transformation and normalization, which impact classifier performance the most. Specifically, the classifiers were built based on the following considerations. First, the classification is time-sensitive because dysregulated host response progression is dynamic (e.g., patients can transition from one subtype to another over time). Therefore, time matched data were analyzed. The time matched data analyzed included data from blood collected from patients within 24 hours of sepsis diagnosis. In cases in which time series data existed, data from the first time point was used. Second, the final classification was envisioned to be a measure of a few biomarkers selected from tens of thousands of biomarkers, so down-selection of the most important biomarkers was implemented. Third, the training sets for the subtype classifiers did not have any outcome labels based on a randomized placebo-controlled trial design, so the trial datasets were selected exclusively as a test set for classifier performance evaluation. Fourth, the VANISH trial raw expression data were measured with the Illumina platform and reported in a different format than the format of expression data of the training set, so the normalization used in the clustering process required special consideration. Fifth, the classification process applied similar data transformation to the training set and the test set to achieve the best performance. Finally, the transformation and normalization strategies that worked best for the clustering process and even the training process may not necessarily perform well in classifying subtypes to differentiate corticosteroid response because the training set did not involve outcome data.

Based on these six considerations, the classifiers were built with a normalization scheme for both the training and test expression data. A platform normalization matrix was built out of all genes of all healthy and sepsis samples. As the number of samples in the matrix was large, individual samples' expression data were quantile normalized against the matrix as a perturbation. To train the classifiers, the expression data from the training set was batch normalized and curated, and then normalized by the platform normalization matrix, as described in detail below.

Sets of potentially significant biomarkers were identified by Significance Analysis of Microarrays (SAM).⁴⁸As another example, sets of potentially significant biomarkers are identifiable using qPCR or RNA sequencing data. qPCR measures the relative or absolute expression level of biomarkers. Normalization or calibration processes are implemented. RNA sequencing data measures relative expression levels of model genes and their transcripts. Using sequencing reads alignment methods (e.g. Hisat2, and Bowtie2), expression estimation methods (e.g. StringTie, Salmon) and normalization processes (e.g. quantile normalization), the estimated expression of model genes are quantified.

These sets of potentially significant biomarker sets were down-selected by at least 2-fold change, and forward-search methodology was used to identify a small set of biomarkers for feature calculation.⁵¹The calculated features (e.g., summarized differential gene expression)⁵¹and clustering label of each sample were finally used to train the multi-class classifiers, implemented as e1071::svm with radial kernel, 0.1 gamma, and 10 cost. Tables 1, 2A-2B, and 3 below depict the genes identified for each subtype (e.g., A, B, and C) for each classifier (e.g., the Full Model, the SS Model, the S Model, and the P Model). Specifically, Table 1 depicts the genes for each subtype (e.g., A, B, and C) for the Full Model, Table 2A depicts the genes for each subtype (e.g., A, B, and C) for the SS Model, Table 2B depicts the genes for each subtype (e.g., A, B, and C) for the S Model, and Table 3 depicts the genes for each subtype (e.g., A, B, and C) for the P Model. Note that in certain embodiments, the entire set of genes for a given Model is used to train and/or test the Model. However, in alternative embodiments, only a subset of the set of genes for a given Model is used to train and/or test the Model. For example, in some embodiments, at least one gene from each subtype A, B, and C (e.g., at least one gene from each row 1, 2, and 3 in one of the below Tables 1, 2A, 2B, and 3) may be used to train and/or test a model.

TABLE 1

Full Model Biomarkers

Row Number
Subtype
Role
Biomarkers

1
A
up
EVL, BTN3A2, HLA-DPA1, IDH3A,

ACBD3, EXOSC10, SNRK

down
MMP8

2
B
up
SERPINB1

down
GSPT1

3
C
up
MPP1, HMBS, TAL1, C9orf78,

POLR2L

down
SLC27A3, BTN3A2, DDX50,

FCHSD2, GSTK1, UBE2E1,

TNFRSF1A, PRPF3, TOMM70A

TABLE 2A

SS Model Biomarkers

Row Number
Subtype
Role
Biomarkers

1
A
up
ZNF831, MME, CD3G

down
STOM

2
B
up

down
ECSIT, LAT, NCOA4

3
C
up
SLC1A5, IGF2BP2, ANXA3

down

TABLE 2B

S Model Biomarkers

Row Number
Subtype
Role
Biomarkers

1
A
up
C14orf159, PUM2

down

2
B
up

down
EPB42, RPS6KA5

3
C
up
EPB42

down
GBP2

TABLE 3

P Model Biomarkers

Row Number
Subtype
Role
Biomarkers

1
A
up
MSH2, DCTD

down
MMP8

2
B
up
HK3

down
UCP2, NUP88

3
C
up
GABARAPL2

down
CASP4

Additional models were created in order to include at least one up- and one down-gene in the model to enable the calculation of scores in an assay based on relative gene expression. Two methods were applied based on forward selection and backward elimination. Forward selection is an iterative method that starts with no genes in the model. In each iteration, features are added that improves the model until the addition of a new variable does not improve the performance of the model. In backward elimination, all the genes are included and then the least significant feature is removed at each iteration if there is improvement in the performance of the model. This is repeated until no improvement is observed from the removal of features. As an example, the SS model was taken as a starting point for the creation of an alternative model. The metric used for evaluating model performance was leave-one-out accuracy and the model's similarity in labeling patients when compared to the Full model.

In this exercise, the backward elimination method produced superior results. Tables 4A-4D depicts four additional models generated by this method named SS.B1, SS.B2, SS.B3, and SS.B4.

TABLE 4A

SS.B1

Row Number
Subtype
Role
Biomarkers

1
A
down
STOM

up
ZNF831, CD3G, MME,

BTN3A2, HLA-DPA1

2
B
down
EPB42, GSPT1, LAT

up
HK3, SERPINB1

3
C
down
GBP2, TNFRSF1A

up
SLC1A5, IGF2BP2, ANXA3

TABLE 4B

SS.B2

Row Number
Subtype
Role
Biomarkers

1
A
down
STOM

up
ZNF831, CD3G, MME,

BTN3A2, HLA-DPA1

2
B
down
EPB42, GSPT1, LAT

up
HK3, SERPINB1

3
C
down
GBP2

up
SLC1A5, IGF2BP2, ANXA3

TABLE 4C

SS.B3

Row Number
Subtype
Role
Biomarkers

1
A
down
STOM

up
MME, BTN3A2, HLA-DPA1, EVL

2
B
down
EPB42, GSPT1, LAT

up
HK3, SERPINB1

3
C
down
BTN3A2, TNFRSF1A

up
SLC1A5, IGF2BP2, ANXA3

TABLE 4D

SS.B4

Row Number
Subtype
Role
Biomarkers

1
A
down
STOM

up
MME, BTN3A2, HLA-DPA1, EVL

2
B
down
EPB42, GSPT1, LAT

up
HK3, SERPINB1

3
C
down
GBP2

up
SLC1A5, IGF2BP2, ANXA3

TABLE 4E

Identification of biomarkers included in each of

the Full model, SS model, S model, P model, SS.B1

model, SS.B2 model, SS.B3 model, and SS.B4 model.

Gene
Alias
Uniprot ID

EVL
Enah/Vasp-like
Q9UI08

BTN3A2
Butyrophilin Subfamily 3 Member
P78410

A2

HLA-DPA1
Major Histocompatibility Complex,
P20036

Class II, DP Alpha 1

IDH3A
Isocitrate Dehydrogenase (NAD(+))
P50213

3 Catalytic Subunit Alpha

ACBD3
Acyl-CoA Binding Domain
Q9H3P7

Containing 3

EXOSC10
Exosome Component 10
Q01780

SNRK
SNF Related Kinase
Q9NRH2

MMP8
Matrix Metallopeptidase 8
P22894

SERPINB1
Serpin Family B Member 1
P30740

GSPT1
G1 To S Phase Transition 1
P15170

MPP1
Membrane Palmitoylated Protein 1
Q00013

HMBS
Hydroxymethylbilane Synthase
P08397

TAL1
TAL BHLH Transcription Factor 1,
P17542

Erythroid Differentiation Factor

C9orf78
Chromosome 9 Open Reading Frame
Q9NZ63

78

POLR2L
RNA Polymerase II, I And III
P62875

Subunit L

SLC27A3
Solute Carrier Family 27 Member 3
Q5K4L6

DDX50
DExD-Box Helicase 50
Q9BQ39

FCHSD2
FCH And Double SH3 Domains 2
O94868

GSTK1
Glutathione S-Transferase Kappa 1
Q9Y2Q3

UBE2E1
Ubiquitin Conjugating Enzyme E2
P51965

E1

TNFRSF1A
TNF Receptor Superfamily Member
P19438

1A

PRPF3
Pre-MRNA Processing Factor 3
O43395

TOMM70A
Translocase Of Outer Mitochondrial
O94826

Membrane 70

ZNF831
Zinc Finger Protein 831
Q5JPB2

MME
Membrane Metalloendopeptidase
P08473

CD3G
CD3g Molecule
P09693

STOM
Stomatin
P27105

ECSIT
ECSIT Signaling Integrator
Q9BQ95

LAT
Linker For Activation Of T Cells
O43561

NCOA4
Nuclear Receptor Coactivator 4
Q13772

SLC1A5
Solute Carrier Family 1 Member 5
Q15758

IGF2BP2
Insulin lake Growth Factor 2
Q9Y6M1

MRNA Binding Protein 2

ANXA3
Annexin A3
P12429

C14orf159
D-Glutamate Cyclase
Q7Z3D6

PUM2
Pumilio RNA Binding Family
Q8TB72

Member 2

EPB42
Erythrocyte Membrane Protein Band
P16452

4.2

RPS6KA5
Ribosomal Protein S6 Kinase A5
O75582

GBP2
Guanylate Binding Protein 2
P32456

MSH2
MutS Homolog 2
P43246

DCTD
DCMP Deaminase
P32321

HK3
Hexokinase 3
P52790

UCP2
Uncoupling Protein 2
P55851

NUP88
Nucleoporin 88
Q99567

GABARAPL2
GABA Type A Receptor Associated
P60520

Protein Like 2

CASP4
Caspase 4
P49662

Table 5 depicts primer sets for amplifying genes identified by the SS Model and depicted above in Table 2A, primer sets for amplifying genes identified by the S Model and depicted above in Table 3B, and primer sets for amplifying genes identified by the SS.B2 Model and depicted above in Table 4B. Each primer set includes a pair of single-stranded DNA primers (i.e., a forward primer and a reverse primer) for amplifying one gene by, for example, RT-qPCR. In some embodiments the entire sequence of a primer may be used in amplification of the associated gene. In alternative embodiments, at least 15 contiguous nucleotides of a primer sequence may be used in amplification of the associated gene. In certain embodiments, primer sequences other than those mentioned provided in Table 5 can be used to amplify one or more of the genes from Tables 1, 2A, 2B, 3, and 4A-4D.

TABLE 5

RT-qPCR Primer Sequences

Forward
Reverse

Primer
Primer

Forward
Reverse
SEQ ID
SEQ ID

Gene
Model
Subtype
Primer
Primer
NO.
NO.

SLC1A5
SS/SS.B2
C
ATCACCATC
CCACAGCCA
1
2

CTGGTCACG
GGATCAAGG

G
AG

IGF2BP2
SS
C
AAGACCGTG
TTTCCCTGAT
3
4

AACGAACT
CTTGCGCTG

GCA
T

ANXA3
SS/SS.B2
C
CGAGCCTTG
TGTTCGAAT
5
6

AAGGGTATT
GTCCAAAAG

GG
GTCA

ZNF831
SS/SS.B2
A
ACCTGGGTG
GGTGATTCT
7
8

CGAAGAAG
GAGGTGGCA

AAG
CA

MME
SS/SS.B2
A
AACTTTGCA
GCAGAGTTC
9
10

CAGGTGTGG
TGCAAAGTC

TG
CC

CD3G
SS/SS.B2
A
GCCCCTCAA
AGGAGGAGA
11
12

GGATCGAG
ACACCTGGA

AAG
CT

STOM
SS/SS.B2
A
AAAGGTGG
AAGGGCTGC
13
14

AGCGTGTGG
AGGAGATTC

AAA
AG

ECSIT
SS
B
CCGGAGGA
CATGCACAT
15
16

GTGGAACCT
GGCGAAGAC

CTA
AG

LAT
SS/SS.B2
B
TGTGTCCCA
CAGCTCCTG
17
18

GGAACTGC
CAGATTCTC

ATC
GT

NCOA4
SS
B
GGGCAACCT
CAAACTGCA
19
20

CAGCCAGTT
GGGAGGCCA

AT
TA

C14orf159
S
A
CCCTCCCGT
TTCTGGATC
21
22

CGGTCATTA
ATCTCGGCG

AG
TG

PUM2
S
A
TGCACAAGA
GGTGGTCCT
23
24

TTCGACCTC
CCAATAGGT

ACA
CC

EPB42
S/SS.B2
B, C
TGCCATCAA
CTCTCTGTGA
25
26

GATGCCAG
ATGAGCCCC

AGAA
C

GBP2
S/SS.B2
C
CAGGGCCCA
GGCTCCAAT
27
28

GTTAATGGC
GATTTGCTTC

A
TCA

RPS6KA5
S
B
AGCAACCTT
ACTCTCACT
29
30

CCACGCCTT
GGAACTGCT

TA
GC

GSPT1
SS.B2

GACTTCCCT
TCACAGTAT
31
32

CAGATGGGT
TGTGCAGGG

CG
TCA

IGFBP2
SS.B2

AAGACCGTG
TTTCCCTGAT
33
34

AACGAACT
CTTGCGCTG

GCA
T

HK3
SS.B2

GAACGCTCT
CTCTGACTG
35
36

ACAAGCTGC
CAGGAACGT

AC
GA

SERPINB1
SS.B2

TCCTGCTGC
GTCCACTCA
37
38

CGGATGAC
TGCAACTTTT

ATT
CCA

BTN3A2
SS.B2

GCTGACTTA
CAGAGCGGG
39
40

TTGGTATCG
AAATAAGCC

GACG
TAAGA

HLA-DPA1
SS.B2

CCAGGGGA
AGAGCTTGA
41
42

CCCTGTGAA
AGGGTCAGC

ATA
AAT

In certain embodiments, genes may be amplified by methods other than RT-qPCR. For example, in some embodiments, genes may be amplified via LAMP (loop-mediated isothermal amplification). In such embodiments in which a gene is amplified via LAMP, a primer set for amplifying the gene includes a forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer.

Sensitivity analysis was performed for each classifier, for each combination of three genes, with one gene selected from each subtype. Specifically, the accuracies of the classifiers were measured to demonstrate that the accuracy of each classifier in identifying a subject's subtype using any combination of three genes, with one gene from each subtype, is greater than 50% (e.g., greater than random chance).

To calculate accuracy for a given classifier for a given combination of three genes, leave-one-out accuracy of the training samples of the training dataset on which the classifier was trained was implemented. The training dataset included N training samples, each training sample including a label y and features x. The leave-one-out accuracy for the classifier for the combination of three genes was calculated based on N calculations. The Ni calculation leaves out the training sample i during training of the classifier. Then, the trained classifier is used to make a prediction z_ifor the features x_ithat corresponds to the training sample i that was left out of the training data set. The prediction z_iis then compared to the label y_ito determine the accuracy of the prediction. The leave-one-out accuracy for the classifier for the combination of three genes was calculated as the number of correct predictions z, divided by N.

FIGS. 2-5 depict the individual accuracies determined for each combination of three genes, with one gene from each subtype, for the Full, SS, S, and P Models, respectively. Specifically, FIG. 2 is a graph of the individual accuracies determined for each combination of three genes, with one gene from each subtype, for the Full Model. FIG. 3 is a graph of the individual accuracies determined for each combination of three genes, with one gene from each subtype, for the SS Model. FIG. 4 is a graph of the individual accuracies determined for each combination of three genes, with one gene from each subtype, for the S Model. FIG. 5 is a graph of the individual accuracies determined for each combination of three genes, with one gene from each subtype, for the P Model. As shown in FIGS. 2-5, each classifier, for each combination of three genes, with one gene from each subtype, demonstrated an accuracy of greater than 50% (e.g., greater than random chance). Furthermore, the average accuracies of the Full, SS, S, and P Models were 82.93%, 89.6%, 86.3%, and 98.3%, respectively. Therefore, each classifier demonstrated an average accuracy of greater than 50% (e.g., greater than random chance). Included in each of FIGS. 2-5 is the accuracy of a model that incorporates all of the genes for a particular model (denoted as “full” in each respective figure). For example, for the Full model, incorporating all of the genes refers to a Full model that analyzes all the biomarkers shown in Table 1. For the SS model, incorporating all of the genes refers to the SS model that analyzes all the biomarkers shown in Table 2A. For the S model, incorporating all of the genes refers to the S model that analyzes all the biomarkers shown in Table 2B. For the P model, incorporating all of the genes refers to the SS model that analyzes all the biomarkers shown in Table 3.

FIGS. 6A-6D are graphs of individual accuracies determined for each combination of three biomarkers, with one biomarker from each subtype, for the SS.B1, SS.B2, SS.B3, and SS.B4 models, respectively. These models exhibit accuracies of 89.57%. Included in each of FIGS. 6A-6D is the accuracy of each model that incorporates all of the genes for a particular model (denoted as “full” in each respective figure). For example, for the SS.B1 model, incorporating all of the genes refers to a SS.B1 model that analyzes all the biomarkers shown in Table 4A. For example, for the SS.B2 model, incorporating all of the genes refers to a SS.B2 model that analyzes all the biomarkers shown in Table 4B. For example, for the SS.B3 model, incorporating all of the genes refers to a SS.B3 model that analyzes all the biomarkers shown in Table 4C. For example, for the SS.B4 model, incorporating all of the genes refers to a SS.B4 model that analyzes all the biomarkers shown in Table D.

V. Identification of Therapeutics for Treatment of Dysregulated Host Response Patient Subtypes

Based on the differential biomarker expression determined for each dysregulated host response subtype, an immune state was determined for each subtype. Specifically, subtype A was determined to be associated with the adaptive immune state, subtype B was determined to be associated with the innate immune state and the complement immune state, and subtype C was determined to be associated with the coagulopathic immune state. Then, biomarkers indicated as related to dysregulated host response, immune state, and the pharmacology of existing therapeutics were identified in the literature. Table 6 below depicts a representative list of genes associated with dysregulated host response, immune state, and the pharmacology of existing therapeutics that were identified from the literature.

TABLE 6

Representative Examples of Genes Associated with Dysregulated host

response, Immune State, and Pharmacology of Existing Therapeutics

Gene
Uniprot
Immune state
Protein/Function
Effect

TREM1
Q9NP99
Innate
PAMP
Triggering receptor
Pro-

expressed by myeloid
inflammatory

cells-1

CD180
Q99467
Innate
PAMP
Controls B cell
Pro-

recognition and signaling
inflammatory

of LPS via TLR4

MIF
P14174
Innate
PAMP
Stimulated by bacterial
Pro-

antigens (and
inflammatory

glucocorticoids thus

counteracting it's effects)

CD14
P08571
Innate
PAMP
PAMP recognition
Pro-

inflammatory

IL15
P40933
Innate
PAMP
IL-15. Secreted following
Pro-

viral infection. Induces
inflammatory

the proliferation of natural

killer cells.

IL6
P05231
Innate
PAMP
IL-6
Pro and Anti-

inflammatory

TLR2
O60603
Innate
PAMP/DAMP
Recognizes bacterial,
Pro-

fungal, viral, and certain
inflammatory

endogenous substances

TLR6
Q9Y2C9
Innate
PAMP
Recognizes lipopeptides
Pro-

derived from gram-
inflammatory

positive bacteria and

mycoplasma and several

fungal cell wall

saccharides

NLRP1
Q9C000
Innate
PAMP
Notch-like receptor.
Pro-

activates an antibacterial
inflammatory

immune response

CASP1
P29466
Innate
PAMP/DAMP
Interleukin-1 converting
Pro-

enzyme
inflammatory

IL1B
P01584
Innate
PAMP/DAMP
IL-1β
Pro-

inflammatory

IL18
Q14116
Innate
PAMP/DAMP
IL-18
Pro-

inflammatory

PYCARD
Q9ULZ3
Innate
PAMP/DAMP
ASC (PRR), activates
Pro-

caspsase 1 and pro-
inflammatory

inflammatory cytokines

TLR4
O00206
Innate
PAMP/DAMP
PRR that activates innate
Pro-

immunity via NF-kB
inflammatory

TNF
P01375
Innate
PAMP/DAMP
TNF-α expressed by
Pro-

macrophages
inflammatory

EBI3
Q14213
Innate
PAMP/DAMP
IL-27B, Expressed by
Adaptive

APC via TLR4 activation,
function

activates Th1, Tr1, inhibits

Th2, Th17, Treg

IL27
Q8NEV9
Innate
PAMP/DAMP
IL-27, Expressed by APC
Adaptive

via TLR4 activation,
function

activates Th1, Tr1, inhibits

Th2, Th17, Treg

IL1RN
P18510
Innate
PAMP/DAMP
IL-1 receptor antagonist,
Anti-

prevents IL-1A and B
inflammatory

from binding

HSPD1
P10809
Innate
DAMP
HSP60
Pro-

inflammatory

IL1RL1
Q01638
Innate
DAMP
ST2 (receptor of IL-33
Pro-

which is upregulated by
inflammatory

DAMPs)

S100A9
P06702
Innate
DAMP
Heat shock protein
Pro-

(DAMP trigger)
inflammatory

HSPA1B
P0DMV8
Innate
DAMP
Heat shock 70 kDa protein
Pro-

1B (DAMP trigger)
inflammatory

HSPA1A
P0DMV9
Innate
DAMP
Heat shock 70 kDa protein
Pro-

1A (DAMP trigger)
inflammatory

MPO
P05164
Innate
DAMP
Myeloperoxidase (DAMP
Pro-

trigger)
inflammatory

ELANE
P08246
Innate
DAMP
Neutrophil elastase
Pro-

(DAMP trigger)
inflammatory

CTSG
P08311
Innate
DAMP
Cathepsin G (DAMP
Pro-

trigger)
inflammatory

HMGB1
P09429
Innate
DAMP
d HMG-1 (DAMP trigger)
Pro-

inflammatory

CD24
P25063
Innate
DAMP
In association with
Pro-

SIGLEC10 may be
inflammatory

involved in the selective

suppression of the immune

response to danger-

associated molecular

patterns (DAMPs) such as

HMGB1, HSP70 and

HSP90.

SIGIRR
Q6IA17
Innate

Single Ig IL-1-related
Pro-

receptor, attenuates TLR4
inflammatory

activity

CSF3
P09919
Innate
Cell
G-CSF (pro and anti-
Pro-

recruitment
inflammatory), expression
inflammatory

triggered by IL-17

CSF2
P04141
Innate
Cell
GM-CSF, encoded in Th2,
Pro-

recruitment,
stimulates stem cells to
inflammatory

Innate
produce granulocytes

immune
(neutrophils, eosinophils,

stimlation
and basophils) and

monocytes via STAT5

C3AR1
Q16581
Complement
Complement
C3a receptor
Complement

activity

C5AR1
P21730
Complement
Complement
C5a receptor
Complement

activity

C5AR2
Q9P296
Complement
Complement
C5a receptor
Complement

activity

STAT1
P42224
Adaptive

Activated by IFNa, IFNg,
Pro-

EGF, PDGF, IL-6.
inflammatory

Activates Th1, inhibits

Th17, Treg

IFNG
P01579
Adaptive

INF-γ (innate and
Pro-

adaptive)
inflammatory

LTA
P01374
Adaptive

TNF-β, Lymphotoxin-
Pro-

alpha (LT-α), expressed
inflammatory

by lymphocytes activates

innate immunity via NF-

kB

STAT4
Q14765
Adaptive

IFN-γ production triggered
Pro-

by IL-12
inflammatory

CD28
P10747
Adaptive

T cell co-stimulation, IL-6
Pro-

stimulation, IL-10
inflammatory

stimulation

CD3D
P04234
Adaptive

T-cell surface glycoprotein
Pro-

CD3 gamma chain
inflammatory

CD3G
P07766
Adaptive

T-cell surface glycoprotein
Pro-

CD3 gamma chain
inflammatory

CD3E
P09693
Adaptive

T-cell surface glycoprotein
Pro-

CD3 gamma chain
inflammatory

PTMA
P06454
Adaptive

Thymosin al (increases
Pro-

HLA-DR)
inflammatory

IL7R
P16871
Adaptive

IL-7 receptor (IL-7
Pro-

decreases Treg)
inflammatory

IL7
P13232
Adaptive

IL-7 (IL-7 decreases Treg)
Pro-

inflammatory

TNFRSF18
Q9Y5U5
Adaptive

Glucocorticoid-induced
Pro-

tumor necrosis factor
inflammatory

receptor family-related

gene (GITR) involved in

inhibiting the suppressive

activity of T-regulatory

cells and extending the

survival of T-effector

cells, decreases IL-10

IL13
P35225
Adaptive
Pro and anti-
IL-13 (Th2 cytokine)
Anti-

inflammatory
mediator of allergic
inflammatory

inflammatory response

GZMB
P10144
Adaptive

Granzyme-B, expressed
Anti-

by Treg to lyse T cells
inflammatory

TNFRSF14
Q92956
Adaptive

a APC HVEM (suppresses
Anti-

adaptive and activates
inflammatory

innate)

BTLA
Q7Z6A9
Adaptive

a Tcell BTLA (Inhibitory
Anti-

cell surface receptor)
inflammatory

PDCD1
Q15116
Adaptive

b APC PD-1 (Inhibitory
Anti-

cell surface receptor)
inflammatory

CD274
Q9NZQ7
Adaptive

b Tcell PD-L1 (Inhibitory
Anti-

cell surface receptor)
inflammatory

HLA-DRA
P01903
Adaptive

c APC Peptide
Anti-

presentation
inflammatory

LAG3
P18627
Adaptive

c Tcell LAG-3 (Inhibitory
Anti-

cell surface receptor)
inflammatory

CEACAM1
P13688
Adaptive

d APC ligand for TIM-3
Anti-

inflammatory

HAVCR2
Q8TDQ0
Adaptive

d Tcell TIM-3 (Inhibitory
Anti-

cell surface receptor)
inflammatory

CD86
P42081
Adaptive

e APC Ligand for CTLA-4
Anti-

inflammatory

CTLA4
P16410
Adaptive

e Tcell CTLA-4
Anti-

(Inhibitory cell surface
inflammatory

receptor)

IL10
P22301
Adaptive

IL-10, expressed by Th2
Anti-

inflammatory

TGFB1
P01137
Adaptive

TGF-β (inhinits Th and
Anti-

cytokines)
inflammatory

IL2RA
P01589
Adaptive

Treg activity
Anti-

inflammatory

FOXP3
Q9BZS1
Adaptive

FoxP3 (Treg)
Anti-

inflammatory

SERPINE1
P05121
Coagulation

Plasminogen activator
Pro-

inhibitor-1 (PAI-1),
coagulant

elevated risk of

thrombosis

F2
P00734
Coagulation

Thrombin
Pro-

coagulant

F3
P13726
Coagulation

Tissue factor
Pro-

coagulant

F5
P12259
Coagulation

Factor V
Pro-

coagulant

F7
P08709
Coagulation

Factor VII
Pro-

coagulant

F8
P00451
Coagulation

Factor VIII
Pro-

coagulant

F10
P00742
Coagulation

Factor X
Pro-

coagulant

F12
P00748
Coagulation

Factor XII
Pro-

coagulant

F13A1
P00488
Coagulation

Factor XIII, A1
Pro-

polypeptide
coagulant

ITGA2B
P08514
Coagulation

Integrin alpha 2b.
Pro-

Following activation
coagulant

integrin alpha-IIb/beta-3

brings about

platelet/platelet interaction

through binding of soluble

fibrinogen. This step leads

to rapid platelet

aggregation which

physically plugs ruptured

endothelial cell surface.

ITGB3
P05106
Coagulation

Integrin beta 3. The
Pro-

ITGB3 protein product is
coagulant

the integrin beta chain beta

3. Integrins are integral

cell-surface proteins

composed of an alpha

chain and a beta chain. A

given chain may combine

with multiple partners

resulting in different

integrins. Integrin beta 3 is

found along with the alpha

IIb chain in platelets.

Integrins are known to

participate in cell adhesion

as well as cell-surface-

mediated signaling.

FGA
P02671
Coagulation

Fibrinogen alpha chain
Pro-

coagulant

FGB

Coagulation

Fibrinogen beta chain
Pro-

coagulant

FIBCD1
Q8N539
Coagulation

Fibrinogen c domain
Pro-

containing 1
coagulant

PTAFR
P25105
Coagulation
Platelet
Platelet-activating factor
Pro-

Activation
receptor
coagulant

THBD
P07204
Coagulation

Thrombomodulin
Anti-

coagulant

TFPI
P10646
Coagulation

Tissue factor pathway
Anti-

inhibitor
coagulant

SERPINC1
P01008
Coagulation

Antithrombin
Anti-

coagulant

PROS1
P07225
Coagulation

Protein S
Anti-

coagulant

PROC

Coagulation

Protein C
Anti-

coagulant

S1PR3
Q99500
Vascular

Maintaining vascular
Decreased

Permeability

integrity
permeability

S1PR1
P21453
Vascular

Sphingosine-1-phosphate
Decreased

Permeability

receptor 1, T cell
permeability

suppression

ANGPT1
Q15389
Vascular

Angiopoietin 1
Decreased

Permeability

permeability

ANGPT2
O15123
Vascular

angiopoietin 2
Increased

Permeability

permeability

For each gene in Table 6, the fold-change in gene expression was calculated between subtypes. Specifically, for each subtyping Model (Full/S/SS/P), linear regression was used to compare each gene expression among A/B/C subtypes. In order to adjust batch effects of microarray dataset from different studies, study IDs were included in the linear regression model. From the linear regression model, the coefficients of subtypes were used to calculate gene expression fold changes and Benjamini-Hochberg (BH)⁵³adjusted p-values of subtypes were used to indicate if expression differences were statistically significant. Table 7 below depicts a representative dataset for subtype fold-changes in expression of the genes in Table 6. The fold-changes in gene expression between subtypes (e.g. fold change “A/B”=2{circumflex over ( )}(A−B) where A and B are the log 2 mean expression for the listed gene for the given subtype A and B) are listed as the numerical values in the table. Bold or underlined indicates a statistically significant fold-change as determined by BH. Bold indicates up-regulation and underlined indicates down-regulation. This dataset was then used to identify therapeutic candidates for the treatment of dysregulated host response taking into account whether the gene is expected to be appreciably expressed in blood.

TABLE 7

Representative Examples of Fold-Changes in Gene Expression Between A/B/C Subtypes

Gene
A/B
A/C
B/A
B/C
C/A
C/B
Examples of Related Therapeutics

TREM1

1.382

1.786

0.724

1.2

0.56
0.833
nangibotide (MOTREM), TREM-1 inhibitor

CD180

1.612

1.635

0.62
0.935

0.612

1.069

MIF

1.348

1.281

0.742

0.988

0.781

1.012

CD14
0.934

1.698

1.071

1.724

0.589

0.58

IL15
1.07

1.758

0.934

1.547

0.569

0.646

IL-15, NIZ985

IL6
1.019
0.963
0.981

0.949

1.038

1.054

Tocilizumab/anti-IL-6R

NLRP1

1.56

1.706

0.641

1.025

0.586

0.975

CASP1
0.916

1.692

1.091

1.81

0.591

0.552

Emricasan (Novartis), pan-caspsase inhibitor

IL1B
0.965

2.017

1.036

1.903

0.496

0.525

IL1R1 (Amgen)

IL18

0.824

1.059

1.214

1.35

0.944

0.741

CXCL8
1.03
1.016
0.97
0.966
0.984
1.035

PYCARD

0.832

1.496

1.202

1.722

0.669

0.581

Emricasan, pan-caspase inhibitor

TLR4

0.577

1.1

1.733

1.87

0.909

0.535

Resatorvid (Takeda), Eritoran (Eisai), HU-003

(Huons), NI-0101 (NovImmune)

TNF

0.785

1.091

1.274

1.433

0.917

0.698

CytoFab (anti-TNF-α AstraZeneca),

Adalimumab/Humira, Infliximab/Remicade,

Nerelimomab, Humicade, Afelimomab,

rhTNFbp (TNF binding protein)

EBI3

0.555

0.711

1.801

1.198

1.406

0.834

IL27

0.794

0.985

1.26

1.18

1.015

0.848

IL1RL1

0.902

0.906

1.109

1.029

1.104

0.972

MPO

0.547

0.407

1.828

0.669

2.459

1.494

S100A9

0.803

0.938

1.245

1.165

1.066

0.858

HSPA1B

0.594

0.632

1.683

1.224

1.582

0.817

HSPA1A

0.686

0.723

1.457

1.113

1.383

0.899

ELANE

0.553

0.356

1.807

0.529

2.811

1.891

CTSG
0.738

0.507

1.355

0.588

1.973

1.702

HMGB1
0.92
1.069
1.088

1.222

0.935

0.818

IL33
0.998
0.988
1.002
0.983
1.012
1.017

HSP90B1
1.112
1.108
0.899
1.033
0.902
0.968

HSPD1

1.19

1.161

0.84
1.04

0.861

0.962

S100A8
1.099
0.925
0.91
0.909
1.081
1.1

IL1A
0.929
1.053
1.076
1.128
0.949
0.886
IL1R1 (Amgen)

C3AR1

0.48
1.034

2.084

2.021

0.967

0.495

C5AR1

0.728

1.262

1.373

1.611

0.792

0.621

IFX-1 (anti-C5a InflaRx), Soliris, Ultomiris

(anti-C5a Alexion), Avacopan (anti-C5aR

C5AR2

0.835

1.046

1.198

1.234

0.956

0.81
ChemoCentryx), C5a inhibitor/CaCP 29

(InflaRx)

C2
1.071
1.134
0.934
1.1
0.882
0.909

C4B
0.718
0.494
1.393
0.544
2.024
1.838

SIGIRR

1.682

1.668

0.594

0.99

0.599

1.01

CPB2
1.011
0.978
0.989
0.966
1.022
1.036

IL12A
1.039
1.007
0.962
0.989
0.993
1.011

IL12B
1.006
0.987
0.994
0.992
1.013
1.008

IL4
1.035
1.004
0.966
0.975
0.996
1.026

IL5
1.012
0.966
0.988
0.96
1.036
1.042

IL13
1.032

0.936

0.969

0.936

1.069

1.068

STAT1
1.093

2.243

0.915

1.924

0.446

0.52

IFNG

1.216

1.051

0.823

0.953
0.952
1.049
INF-gama, Actimmune, Recombinant protein,

Genentech

LTA

1.235

0.951

0.81

0.668

1.052

1.497

STAT4

1.902

2.03

0.526

1.039

0.493

0.963

HLA-DRA

2.62

2.47

0.382

0.906

0.405

1.104

CD28

1.338

1.41

0.748

1.006

0.709

0.994
AB103, Atox Bio (peptide CD28 Antagonist)

(contraindicated)

CD3D

2.984

2.611

0.335

0.809

0.383

1.236

CD3G

2.755

2.492

0.363

0.918

0.401

1.089

CD3E

2.799

2.132

0.357

0.746

0.469

1.34

PTMA

1.608

1.352

0.622

0.883

0.74
1.132
Thymosin alpha I (Roche), Thymalfasin

peptide, T-lymphocyte subset modulators; Th1

cell stimulants; Th2-cell-inhibitors

(immunostimulant) (SciClone Pharmaceuticals)

IL7R

3.286

2.409

0.304

0.74

0.415

1.351

IL7

1.147

1.199

0.872

1.009

0.834

0.991
CYT-107 (IL-7 Revnimmune)

IL17A
1
0.986
1
0.968
1.014
1.033

IL3
1.01
0.991
0.99
1.006
1.009
0.994

GZMB

2.204

2.453

0.454

1.152

0.408

0.868

BTLA

1.93

1.708

0.518

0.913

0.585

1.095

TNFRSF14
1.119

1.752

0.894

1.472

0.571

0.679

LAG3

1.465

1.272

0.683

0.972

0.786

1.029

PDCD1

1.104

1.006

0.906

0.902

0.994

1.109

Nivolumab, anti-PD-1 monoclonal antibody,

pembrolizumab/Keytruda

CD274

0.613

1.46

1.632

2.274

0.685

0.44
Anti-PD-L1 (BMS-936559)

CD86

2.177

1.939

0.459

0.89

0.516

1.124

HAVCR2

0.819

1.201

1.221

1.45

0.832

0.69

CTLA4

1.158

1.044

0.863

0.975
0.958
1.026
anti-CTLA-4 monoclonal antibody,

Ipilimumab, Medarex

IL10

0.639

0.783

1.566

1.233

1.277

0.811

FOXP3
1.036

0.921

0.965

0.903

1.085

1.107

IL2
1.017
0.985
0.983
0.975
1.016
1.025
IL-2 Roncoleukin

IL2RA
0.964
1.024
1.038

1.114

0.977

0.897

TNFRSF18

1.057

0.954

0.946

0.938

1.048

1.066

TGFB1

0.875

0.996

1.143

1.156

1.004

0.865

SERPINE1
1.007

0.904

0.993

0.887

1.106

1.127

defibrotide

F2
1.005

0.934

0.995

0.928

1.07

1.077

Antithrombin (CSL Behring), tanogitran

(antagonizes Factors Xa and IIa), Boehringer

Ingelheim

F3
0.981

0.89
1.02

0.933

1.124

1.072

F5

0.603

1.077

1.658

1.603

0.929

0.624

F7
1.017

0.887

0.983

0.893

1.127

1.12

F8

0.599

0.959

1.67

1.516

1.042

0.66

F9
0.997
0.978
1.003
0.979
1.022
1.021
TNX-832, Sunol cH36, mAb, Factor IX

inhibitors; Factor X inhibitors

F10
1.017

0.939

0.984

0.937

1.065

1.068

TNX-832, Sunol cH36, mAb, Factor IX

inhibitors; Factor X inhibitors, tanogitran

(antagonizes Factors Xa and IIa), Boehringer

Ingelheim

F11
1.008
0.976
0.992
0.98
1.024
1.021

F12

0.656

0.762

1.526

1.163

1.313

0.86

F13A1

1.67
0.763

0.599

0.463

1.311

2.158

ITGA2B
0.977

0.513

1.023

0.528

1.948

1.892

ITGB3
0.918

0.607

1.09

0.623

1.647

1.605

FGA
1.011

0.937

0.989

0.947

1.067

1.055

FGB
0.997

0.944

1.003
0.955

1.059

1.047

FGG
0.995
0.982
1.005
0.975
1.018
1.026

FIBCD1
1.046

0.864

0.956

0.751

1.157

1.332

PTAFR

0.722

1.21

1.384

1.403

0.827

0.713

Minopafant (PAF antagonist), Pafase

(inactivates PAF), TCV-309 (PAF antagonist),

YM-264 (PAF antagonist), SM-12502 (PAF

antagonist), UK-74505 (PAF receptor

antagonist), Ginkgolide B (PAF inhinbitor),

Epafipase (Recombinant Human Platelet-

Activating Factor Acetylhydrolase)

CSF3
1.04

0.895

0.962

0.875

1.117

1.142

G-CSF, Filgrastim, Recombinant protein,

Immunostimulants, Amgen (contraindicsted)

CSF2
1.015

0.879

0.985

0.882

1.138

1.133

Sargramostim (Genzyme, etc.)

PROS1

0.784

0.447

1.275

0.581

2.238

1.722

PROC

1.051

0.973

0.952

0.927

1.028

1.079

THBD

0.666

1.231

1.501

1.769

0.812

0.565

Thrombomodulin, ART-123 (Asahi), Protein C

Stimulant

TFPI

0.908

0.645

1.101

0.74

1.55

1.351

tifacogin (recombinant TFPI)

SERPINC1
1.019

0.95
0.981
0.958

1.052

1.044

PROCR
1.023
1.009
0.977
1.009
0.991
0.991

S1PR3

1.58

1.639

0.633

1.013

0.61
0.988

S1PR1

0.579

1.035

1.727

1.687

0.966

0.593

ANGPT1
1.026

0.929

0.975

0.906

1.076

1.104

ANGPT2
1.021
0.964
0.98

0.948

1.037

1.055

TEK
1.013
0.966
0.987
0.971
1.035
1.029

AGT
1.036

0.935

0.966

0.909

1.07

1.1

GIAPREZA (La Jolla)

ACE
1.008
0.956
0.992

0.892

1.046

1.122

GIAPREZA (La Jolla)

REN
1.02
0.947
0.98

0.931

1.056

1.074

GIAPREZA (La Jolla)

ACE2
1.012
0.981
0.988
0.974
1.019
1.027
GIAPREZA (La Jolla)

AGTR1
1.022
0.989
0.979
0.981
1.011
1.019
GIAPREZA (La Jolla)

GLP1R
1.021

0.911

0.979

0.917

1.098

1.09

Exenatide, Byetta, Bydureon, GLP-1 receptor

agonist (Amylin Pharmaceuticals)

TNFRSF1B
1.044

1.719

0.958

1.628

0.582

0.614

p75 TNF receptor, Recombinant protein, TNF

blocker, Amgen

IL11
1.036
0.932
0.965

0.913

1.073

1.096

Oprelvekin, Thrombocytopenia

TNFRSF1A

0.773

1.435

1.293

1.75

0.697

0.571

Lenercept (Roche)

LTF

0.13

0.129

7.68
1.141

7.752

0.876
Talactoferrin alfa, Apolactoferrin, recombinant

lactoferrin (Agennix)

IL3RA

0.796

0.776

1.256

1.111

1.289

0.9
Interleukin-3-receptor-alpha-subunit-

antagonists, Talacotuzumab

FLT3
0.971
1.041
1.03
1.062
0.961
0.942
Flt3 ligand, Mobista, Fms-like tyrosine kinase

3 stimulants, increases Treg proliferation,

Amgen

TLR3
1.033
1.017
0.968
0.997
0.984
1.003
Poly-ICLC, TLR3 agonist, Janssen, Peptide P7

MS4A1

2.023

1.954

0.494

0.862

0.512

1.16
Rituximab, destroys B cells expressing CD20

CASP2
0.975

1.101

1.026

1.119

0.908

0.894

Emricasan

CASP3

0.774

1.177

1.292

1.442

0.85

0.694

Emricasan

CASP4
0.9

1.642

1.111

1.791

0.609

0.558

Emricasan

CASP5

0.744

2.056

1.343

2.593

0.486

0.386

Emricasan

CASP6

1.205

1.362

0.83

1.118

0.734

0.894

Emricasan

CASP7
0.976

1.233

1.025

1.221

0.811

0.819

Emricasan

CASP8
0.992

1.307

1.008

1.303

0.765

0.767

Emricasan

CASP9

0.81
0.965

1.235

1.142

1.037

0.876

Emricasan

CASP10
0.975
1.047
1.026

1.1

0.955

0.909

Emricasan

CASP12
1.003
0.985
0.997
0.985
1.015
1.015
Emricasan

CASP14
0.996
0.985
1.004

0.929

1.015

1.077

Emricasan

BDKRB2
1.035
0.973
0.966

0.942

1.028

1.062

deltibant (bradykinin-2 (BK-2) receptor

antagonist), NPC-17761 (bradykinin-2 (BK-2)

receptor antagonist)

KNG1
1.01
0.985
0.99
0.979
1.015
1.021
deltibant (bradykinin-2 (BK-2) receptor

antagonist), NPC-17761 (bradykinin-2 (BK-2)

receptor antagonist)

PLA2G3
0.987
0.966
1.013
0.964
1.036
1.037
varespladib (sPLA2 inhibitor), IPP-201007

(sPLA2 inhibitor)

PLAT
1.019
0.991
0.982
0.979
1.009
1.022
defibrotide

PDYN
1.002
0.972
0.998
0.976
1.029
1.025
naloxone

PENK
0.985
0.983
1.015
0.993
1.018
1.007
naloxone

OPRM1
1.008
0.99
0.992
0.99
1.01
1.01
naloxone

POMC

1.093

0.971

0.915

0.971
0.971
0.971
naloxone

PNOC

1.697

1.367

0.589

0.826

0.732

1.21

naloxone

NOS2
0.971

0.961

0.971
0.971

1.041

0.971
GW-274150 (NOS inhibitor), hemoximer (NO

scavenger), nebacumab (NO scavenger), ONO-

1714 (iNOS inhibitor), Tilarginine (NO

synthase inhibitor), Norathiol (NO-inhibitor),

targinine (NOS inhibitor), aSeptiMab (anti-

NOS)

ADM

0.565

1.263

1.771

2.185

0.792

0.458

adrecizumab (stabilizes/increases

adrenomedullin and reverses vascular

permeability)

RAMP2
0.971

0.9
0.971

0.903

1.111

1.108

adrecizumab (stabilizes/increases

adrenomedullin and reverses vascular

permeability)

RAMP3
0.971
0.971
0.971

0.952

0.971

1.05

adrecizumab (stabilizes/increases

adrenomedullin and reverses vascular

permeability)

CALCRL
1.016
1.006
0.984
0.996
0.994
1.004
adrecizumab (stabilizes/increases

adrenomedullin and reverses vascular

permeability)

FAS

0.668

0.971

1.497

1.718

0.971

0.582

asunercept (blocks CD95 ligand)

FASLG

1.3

1.159

0.769

0.952

0.863

0.971
asunercept (blocks CD95 ligand)

ADRA2A
0.971
0.971
0.971

0.901

0.971

1.11

centhaquin, Alpha-2A adrenergic receptor

agonist and Alpha-1 adrenergic receptor

antagonist: reduces blood lactate and increase

blood pressure

ADRA1A
0.971
0.971
0.971

0.911

0.971

1.098

centhaquin, Alpha-2A adrenergic receptor

agonist and Alpha-1 adrenergic receptor

antagonist: reduces blood lactate and increase

blood pressure

TMSB4X
0.971
0.971
0.971

1.106

0.971

0.905

timbetasin (synthetic TB4)

ACTA1
0.971

0.937

0.971
0.971

1.067

0.971
timbetasin

ACTA2
1.01
1.017
0.99
1.056
0.983
0.947
timbetasin

ACTC1
1.005
0.966
0.995
0.971
1.035
1.03
timbetasin

ACTB

0.819

0.971

1.22

1.132

0.971

0.883

timbetasin

ACTG1

0.878

0.971

1.139

1.193

0.971

0.838

timbetasin

ACTG2
0.971

0.941

0.971

0.94

1.062

1.064

timbetasin

GPX1

1.618

0.741

0.618

0.45

1.35

2.22

Rexis (enhances Glutathione peroxidase)

GPX2
0.971
0.971
0.971

0.848

0.971

1.18

Rexis (enhances Glutathione peroxidase)

GPX3
0.971

0.848

0.971

0.847

1.18

1.181

Rexis (enhances Glutathione peroxidase)

GPX4

1.776

0.858

0.563

0.487

1.166

2.055

Rexis (enhances Glutathione peroxidase)

GPX5
0.971
0.971
0.971

0.894

0.971

1.119

Rexis (enhances Glutathione peroxidase)

GPX6
1.037
0.999
0.965
0.962
1.001
1.039
Rexis (enhances Glutathione peroxidase)

GPX7
0.886
1.057
1.128
1.192
0.947
0.839
Rexis (enhances Glutathione peroxidase)

GPX8
0.967
0.965
1.034
0.981
1.036
1.019
Rexis (enhances Glutathione peroxidase)

Cxcl9
1.019
1.022
0.981
1.035
0.979
0.966
ISU201

Cxcl10
1.167

1.467

0.857

1.301

0.681

0.768

ISU201

Icam1

0.789

1.148

1.268

1.455

0.871

0.687

ISU201

Vcam1
1.035
0.965
0.966

0.943

1.037

1.06

ISU201

IL12B
1.006
0.987
0.994
0.992
1.013
1.008
ISU201

Csf1
1.017
0.955
0.983

0.894

1.047

1.118

ISU201

PCSK9

0.76

0.838

1.316

1.212

1.194

0.825

LGT-209, anti-PCSK9 antibody

TLR2

0.563

1.181

1.775

2.055

0.847

0.487

Peptide P13

TLR9
1.067
1.013
0.937
0.958
0.987
1.044
Peptide P13, Peptide P16

TLR6

0.796

1.548

1.256

1.779

0.646

0.562

Tinospora cordifolia derivative

ALOX5AP

0.528

0.701

1.893

1.422

1.427

0.703

AKI: montelukast

PLA2G4A

0.721

1.068

1.387

1.467

0.936

0.682

AKI: montelukast

MGST2
0.945

1.209

1.059

1.292

0.827

0.774

AKI: montelukast

CYSLTR1
1.058

1.847

0.945

1.682

0.541

0.595

AKI: montelukast

CYSLTR2

1.098

0.866

0.911

0.831

1.155

1.204

AKI: montelukast

LTB4R2
0.988
0.942
1.012

0.917

1.062

1.09

AKI: montelukast

LCN2

0.073

0.09

13.764
1.458

11.128

0.686
AKI: montelukast

Bdnf
1.007
0.983
0.993
0.989
1.018
1.011
Hydrocortisone

Ncoa2
0.943

1.107

1.061

1.162

0.903

0.861

Hydrocortisone

Nr3c1

0.712

1.121

1.404

1.564

0.892

0.64
Hydrocortisone

Ntrk2
1.015
0.979
0.986

0.962

1.021

1.04

Hydrocortisone

Ppp5c
1.031
1.007
0.97
0.984
0.993
1.016
Hydrocortisone

Arntl

0.857

1.446

1.166

1.634

0.691

0.612

Hydrocortisone

Clock

1.091

1.317

0.916

1.141

0.759

0.876

Hydrocortisone

Cry1

1.372

1.206

0.729

0.866

0.829

1.155
Hydrocortisone

Cry2

1.198

1.127

0.835

0.99

0.887

1.01
Hydrocortisone

Phb

1.517

1.357

0.659

0.926

0.737

1.08
Hydrocortisone

Per1
0.891

0.796

1.122
0.904

1.257

1.106
Hydrocortisone

Arid1a
0.97

1.24

1.031

1.226

0.807

0.815

Hydrocortisone

Ptges3

1.189

1.044

0.841

0.875
0.958
1.143
Hydrocortisone

Ywhah

0.778

0.784

1.285

0.991

1.276

1.01
Hydrocortisone

FIG. 8 depicts the conclusions of this further analysis of Tables 6 and 7, in accordance with an embodiment. Dysregulated host response patients of subtype A exhibit up-regulation of biomarkers associated with innate immune activity involved in pathogen recognition (e.g., via recognition of pathogen-associated molecular patterns (PAMPs)), up-regulation of biomarkers associated with innate immune regulation, and up-regulation of biomarkers associated with adaptive immune activity. Dysregulated host response patients of subtype B exhibit up-regulation of biomarkers associated with innate immune activity involved in recognition of damage-associated molecular patterns (DAMPs), up-regulation of biomarkers associated with DAMPs, up-regulation of biomarkers associated with inflammation (e.g. TNF-alpha), up-regulation of biomarkers associated with complement activity, down-regulation of biomarkers associated with adaptive immune activity, up-regulation of biomarkers associated with adaptive immune suppression, and up-regulation of markers associated with increased risk of acute kidney injury. Subtype C patients exhibit down-regulation of biomarkers associated with innate and adaptive immune activity, up-regulation of biomarkers associated with DAMPs, up-regulation of biomarkers associated with cellular recruitment (e.g. G-CSF and GM-CSF), up-regulation of biomarkers associated with increased risk of thrombosis, and up-regulation of biomarkers associated with coagulation.

These findings of differential biomarker expression between subtypes A, B, and C inform general therapeutic strategies. FIG. 9 depicts a heat map depicting differential expression of genes from Table 6 for dysregulated host response patients having subtypes A, B, and C, and for healthy subjects without dysregulated host response, in accordance with an embodiment. As discussed below with regard to FIG. 10, subtype A patients exhibit relatively low mortality, which may be attributable to relatively beneficial host response. In fact, as shown in FIG. 9, differential expression of genes for dysregulated host response patients having subtype A most closely resembles differential expression of genes for healthy subjects without dysregulated host response. Thus, in subtype A patients, it may be beneficial to avoid immunomodulatory agents exhibiting immunosuppressive effects that suppress the beneficial host response. In subtype B patients, it may be beneficial to stimulate adaptive immune activity, attenuate innate immune stimulants (e.g. TNF-α), attenuate complement immune activity, attenuate DAMPs and/or block DAMP receptors, and activate PAMP receptors. In subtype C patients, it may be beneficial to simulate adaptive immune activity, administer anticoagulants or agents that indirectly attenuate pro-coagulation factors, decrease vascular permeability, attenuate DAMPs and/or block DAMP receptors, and activate PAMP receptors.

FIG. 10 depicts risk of mortality for dysregulated host response patients having subtypes A, B, and C, in accordance with an embodiment. As mentioned above, subtype A patients exhibit a low risk of morality, relative to subtype B and C patients. Furthermore, subtype C patients exhibit a high risk of morality, relative to subtype A and B patients. Therefore, the subtyping Models may be used as a prognostic to assess the risk of mortality of a dysregulated host response patient.

VI. Evaluation of Therapeutics for Dysregulated Host Response Patient Subtypes

As discussed above, the genes of Tables 6 and 7 are associated with pharmacology of existing therapeutics. For instance, examples of existing therapeutics that are associated with certain genes are indicated in Table 7. Analysis of these genes of Tables 6 and 7 according to subtype, informs the use of the existing therapeutics associated with these genes for treating dysregulated host response patients of the subtype. Specifically, Table 8 depicts therapeutic hypotheses for systemic immune patients having subtypes A, B, and C, determined based on the analysis of differential gene expression of Table 7, in accordance with an embodiment.

As a specific example, while anti-TNF-alpha has failed to show benefit in past sepsis clinical trials, analysis of differential gene expression according to subtype can inform which specific subtype of patients may respond to anti-TNF-alpha. In this example, the TNF gene is seen to be up-regulated in patients having subtype B and thus, subtype B patients may specifically respond to anti-TNF-alpha therapy.

Table 8 below summarizes an analysis of existing therapeutics that are anticipated to provide the desired therapeutic effects for subtypes A, B, and C mentioned above.

TABLE 8

Representative Examples of Therapeutic Hypotheses for Dysregulated host response Patient Subtypes

Genetic
Trade

Sepsis
Anticipated
Subtype

Name
Name
Description
Hypothesis
Effect
Hypothesis
Evidence

Anti-
BMS-
Anti-
Blocks upregulation
Increase
May benefit
Type B and C

PD-L1
936559
PD-L1
of PD-1/PD-L1 to
adaptive
subtype B
patients have a

restore immune cell
immune
and C
suppressed

function
activity

adaptive

immune

response and

Type B up-

regulated PD-

L1 and down-

regulated

INF-g

PD-L1
BMS-
Peptide that
Blocks upregulation
Increase
May benefit
Type B and C

blocker
986189
blocks
of PD-1/PD-L1 to
adaptive
subtype B
patients have a

PD/PD-L1
restore immune cell
immune
and C
suppressed

function
activity

adaptive

immune

response and

Type B up-

regulated PD-

L1 and down-

regulated

INF-g

Anti-
CM-24
anti-

Increase
May benefit
Type B and C

CEACAM1

CEACAM1

adaptive
subtype B
patients have a

immune
and C
suppressed

activity

adaptive

immune

response and

up-regulated

CEACAM1

and TIM-3

Anti-
MK-1966
anti-IL-10

Increase
May benefit
Type B patients

IL-10R

receptor

adaptive
subtype B
have a

immune

suppressed

activity

adaptive

immune

response and

up-regulated

IL-10

TNF
JTE 607
Reduces
These results suggest
Decrease
May benefit
Type B

inhibitor

TNF-α,
that JTE-607 can
inflammation,
subtype B
exhibits

IL-1β, IL-6,
inhibit the production
increase

relative high

IL-8, IL-10
of inflammatory
adaptive

gene

cytokines such as

expression of

tumor necrosis factor-

TNF-a (pro-

alpha, interleukin-6

inflammatory

and cytokine-induced

cytokine) and

neutrophil

IL-10 (adaptive

chemoattractant and

immune

attenuate acid-

suppressant)

induced lung injury in

rats. This agent might

be therapeutically

useful for lung injury

IL-7
CYT-107
IL-7,
A defining
Increase
May benefit
IL-7 gene

immune
pathophysiologic
adaptive
subtype B
expression is

stimulant
feature of sepsis is
immune
and C
relatively low

profound apoptosis-
activity

in subtype B

induced death and

and C and these

depletion of CD4+

Types exhibit

and CD8+ T cells.

down-

Interleukin-7 (IL-7) is

regulation of

an antiapoptotic

genes

common γ-chain

associated with

cytokine that is

immune

essential for

activity.

lymphocyte

proliferation and

survival. Clinical

trials of IL-7 in over

390 oncologic and

lymphopenic patients

showed that IL-7 was

safe, invariably

increased CD4+ and

CD8+ lymphocyte

counts, and improved

immunity.

tanogitran
Antithrombin:

anti-
May benefit
Type C patients

antagonizes

coagulant
subtype C
have

Factors Xa

upregulated

and IIa

genes related to

coagulation

TNX-832,
mAb Factor IX
Tissue factor (TF) is a
anti-
May benefit
Type C patients

Sunol
inhibitors;
transmembrane
coagulant
subtype C
have

cH36
Factor X
glycoprotein that acts

upregulated

inhibitors
as the principal

genes related to

initiator of the

coagulation

extrinsic coagulation

pathway. TF is a key

mediator between the

immune system and

coagulation and is the

principal activator of

coagulation. Vessel

injury or pathological

conditions leading to

the exposure TF in the

vascular adventitia

layer or induction of

TF expression on

endothelial cells and

monocytes permits

interactions between

TF and coagulation

factor VIIa (FVIIa)

resulting in the

formation of the high

affinity TF-FVIIa

complex. TNX-832

(formerly known as

Sunol-cH36), directed

against human TF,

which can block the

pathological

complications of TF-

dependent thrombus

formation. The

blockage by TNX-832

of initiating events in

the extrinsic

coagulation pathway

may attenuate the

effects on pro-

inflammatory events

tifacogin
TFPI: anti-
Systemic activation of
anti-
May benefit
Type C patients

coaggulant
coagulation and
coagulant
subtype C
have

thrombus formation in

upregulated

the microvasculature

genes related to

accompanies organ

coagulation

dysfunction and

excess mortality in

severe sepsis. Tissue

factor

(thromboplastin) is a

major initiator of the

blood coagulation

process. Endothelial

damage is common in

severe sepsis, as

shown by elevations

in endothelial derived

factors, such as von

Willebrand factor,

and by the presence of

coagulation

abnormalities,

including

prolongation of

prothrombin time, in

more than 90% of

patients who are

severely ill and

infected. It is

hypothesized that in

patients with severe

sepsis, TFPI may

protect the

microvasculature

endothelium from

coagulation and

sepsis-induced injury.

This hypothesis is

supported by several

preclinical studies in

which exogenous

TFPI expressed in

mammalian cells

and/or Escherichia

coli improved

outcome in septic

animals

iloprost
Anti-
combination therapy
anti-
May benefit
Coagulation in

trometamol +
coagulant
in septic shock
coagulant
subtype C
subtype C

eptifibatide

patients is expected to

deactivate the

endothelium and

restore vascular

integrity, reduce

formation of

microvascular

thrombosis and

dissolve existing clots

in the

microcirculation and

maintain platelet

counts, thereby

improving platelet-

mediated immune

function and reducing

the risk of bleeding.

Together this is

expected to translate

into reduced organ

failure and improved

outcome in patients

with septic shock.

Pafase,
PAF
BN 52021 is an
anti-
May benefit
PAF receptor

Ginkgolide B
inhinbitor
effective and specific
coagulant
subtype B
upregulated in

PAF receptor

subtype B

antagonist (PAFra)

with proven inhibiting

effects on PAF-

induced events, i.e. in

vitro on platelet Study

Design aggregation,

and in animals on

shock events induced

by endotoxin

(hypotension,

gastrointestinal

disorders and

bronchial spasm). [7]

It has also been

demonstrated that

preventive

administration of

BN 52021 in rats

attenuated the

reactions to injected

endotoxin: the

mortality rate was

decreased and the

release of

thromboxane and

prostaglandin factor

1-α

(PGF1-α) was

reduced.

Epafipase or
Recombinant
The therapeutic
anti-
May benefit
PAF receptor

Pafase
Human
rationale for the
coagulant
subtype B
upregulated in

Platelet-
administration of

subtype B

Activating
rPAF-AH in

Factor
severe sepsis is to

Acetylhydrolase
increase PAF-AH

activity in the

presence of

generalized

inflammation and

coagulation. The

therapeutic

potential for this

strategy was

supported

by the results from a

phase II trial of

rPAF-AH in 127

patients with severe

sepsis (36). A phase

III trial was

undertaken

to confirm these

results in patients at

risk

for ARDS and

mortality from severe

sepsis.

Minopafant,
PAF

anti-
May benefit
PAF receptor

TCV-309,
antagonist

coagulant
subtype B
upregulated in

YM-264,

subtype B

SM-12502,

UK-74505

NI-0101
Blocks
Toll-Like Receptor 4
Decrease
May benefit
TLR4 gene

TLR4
(TLR4) signal
inflammation
subtype B
upregulated in

pathway plays an

subtype B vs.

important role in

subtype A

initiating the innate

immune response and

its activation by

bacterial endotoxin is

responsible for

chronic and acute

inflammatory

disorders that are

becoming more and

more frequent in

developed countries.

Modulation of the

TLR4 pathway is a

potential strategy to

specifically target

these pathologies.

HU-003
Blocks
Toll-Like Receptor 4
Decrease
May benefit
TLR4 gene

TLR4
(TLR4) signal
inflammation
subtype B
upregulated in

pathway plays an

subtype B vs.

important role in

subtype A

initiating the innate

immune response and

its activation by

bacterial endotoxin is

responsible for

chronic and acute

inflammatory

disorders that are

becoming more and

more frequent in

developed countries.

Modulation of the

TLR4 pathway is a

potential strategy to

specifically target

these pathologies.

Eritoran
Blocks
Toll-Like Receptor 4
Decrease
May benefit
TLR4 gene

TLR4
(TLR4) signal
inflammation
subtype B
upregulated in

pathway plays an

subtype B vs.

important role in

subtype A

initiating the innate

immune response and

its activation by

bacterial endotoxin is

responsible for

chronic and acute

inflammatory

disorders that are

becoming more and

more frequent in

developed countries.

Modulation of the

TLR4 pathway is a

potential strategy to

specifically target

these pathologies.

Resatorvid
Blocks
Toll-Like Receptor 4
Decrease
May benefit
TLR4 gene

TLR4
(TLR4) signal
inflammation
subtype B
upregulated in

pathway plays an

subtype B vs.

important role in

subtype A

initiating the innate

immune response and

its activation by

bacterial endotoxin is

responsible for

chronic and acute

inflammatory

disorders that are

becoming more and

more frequent in

developed countries.

Modulation of the

TLR4 pathway is a

potential strategy to

specifically target

these pathologies.

CytoFab
anti-
CytoFab is a
Decrease
May benefit
TNF-α gene

TNF-α
polyclonal antibody
inflammation
subtype B
upregulated in

against tumor necrosis

subtype B vs.

factor alpha, which is

subtype A

produced in vast

quantities in sepsis

patients and

contributes to the

symptoms and organ

dysfunctions that

eventually kill the

patient. Phase IIb

results showed that

CytoFab significantly

reduced TNF-alpha in

the blood and lung

tissues of sepsis

patients, and patients

required five days'

less mechanical

ventilation than when

treated with placebo.

There was also a trend

towards improved

survival;

approximately one

third of patients with

severe sepsis die from

major organ failure at

present.

Nerelimomab
Nerelimomab
anti-

Decrease
May benefit
TNF-α gene

TNF-α

inflammation
subtype B
upregulated in

subtype B vs.

subtype A

Humicade
anti-

Decrease
May benefit
TNF-α gene

TNF-α

inflammation
subtype B
upregulated in

subtype B vs.

subtype A

rhTNFbP
TNF binding

Decrease
May benefit
TNF-α gene

protein

inflammation
subtype B
upregulated in

subtype B vs.

subtype A

p55 TNF
Lenercept
recombinant
Lenercept is a
Decrease
May benefit
TNF-α gene

receptor

TNF receptor
recombinant protein
inflammation
subtype B
upregulated

p55, binds
that is constructed by

and p55 TNF

TNF-a
fusing human soluble

receptor up-

p55 TNF receptors

regulated in

(extracellular domain)

subtype B vs.

to an immunoglobulin

subtype A

G1 heavy chain

fragment and is

expressed as a

dimeric molecule in

Chinese hamster

ovary cells.

Preclinical studies

demonstrated that

lenercept binds to and

neutralizes TNF and

prevents death in a

variety of animal

models of sepsis and

septic shock

Bicizar
Complement
C1-esterase inhibitor
Decrease
May benefit
Complement

C1 inhibitor
(C1 INH) is an alpha-
inflammation
subtype B
system in

globulin controlling

highly

the first part of the

activated in

classic complement

subtype B vs.

pathway and is a

subtype A

natural inhibitor of

complement and

contact system

proteases and a major

downregulator of

inflammatory

processes in blood.

During sepsis, an

overactive

complement system

may compromise the

eff ectiveness of

innate immunity.

anti-C5a
IFX-1/
anti-C5a
Given the strong pro-
Decrease
May benefit
Complement

CaCP29

inflammatory and
inflammation
subtype B
system in

modulatory activities

highly

of C5a signaling,

activated in

therapeutic

subtype B vs.

intervention at the

subtype A

level of C5a or the

C5a receptor (C5aR;

CD88) remains a

focal area.

Neutralizing

antibodies against

C5a have

demonstrated

protective effects in

experimental sepsis.

anti-C5aR
Avacopan
anti-C5aR

Decrease
May benefit
Complement

(will be

inflammation
subtype B
system in

approved in

highly

2020)

activated in

subtype B vs.

subtype A

ISU201
suppressed

Decrease
May benefit
TNF-a and

accumulation of

inflammation
subtype B
Icam1 are up-

pulmonary

and C
regulated in

neutrophils

subtype B and

and

and vcam1 and

eosinophils,

Csf1 genes are

while

up-regulated in

accelerating

subtype C

the decline

in CXCL1,

TNF-α, and

IL-6 in

lavage fluid

and lung

tissue.

ISU201

significantly

reduced

peak

expression

of mRNA

for the

chemokines

Cxcl9 and

Cxcl10, the

adhesion

molecules

Icam1 and

Vcam1, and the

proinflammatory

cytokines

Il1b,

Il12p40,

and Csf1.

PGX-100
Reduces

Decrease
May benefit
TNF-a and

(modified
IL-6, IL-8,

inflammation
subtype B
complement

heparin)
TNF-a,

system up-

CRP. CRP

regulated in

activates the

subtype B vs.

complement

subtype A

system

LGT-209
anti-PCSK9
anti-PCSK9 antibody:
Decrease
May benefit
PCSK9 gene is

antibody
LGT-209 as a novel
inflammation
subtype B
up-regulated in

means to clear

subtype B

endotoxin and other

bacterial toxins out of

a patient's system

centhaquin
Alpha-2A

Increase
May benefit
Both receptors

adrenergic

blood
subtype C
are up-

receptor

pressure

regulated in

agonist and

subtype C

Alpha-1

adrenergic

receptor

antagonist:

reduces

blood lactate

and increase

blood

pressure

Thrombomodulin
ART-123/
Protein C
Thrombomodulin is
anti-
May benefit
Type C patients

REMODULIN/
Stimulant,
an endothelial cell
coagulant
subtype C
have

treprostinil
thrombomodulin
surface

upregulated

transmembrane

genes related to

protein critical to the

coagulation.

regulation of

intravascular

coagulation. rhTM

was approved and

now is being used

clinically for the

treatment of

disseminated

intravascular

coagulation (DIC) in

Japan. As its

mechanism of action,

thrombin-rhTM

complex catalyzes the

activation of protein

C. Activated protein

C proteolytically

inactivates

coagulation co-factors

Va and VIIIa, thereby

inhibiting

amplification of the

coagulation system

asunercept
blocks CD95

Reduces
May benefit
CD95 is

ligand

tissue
subtype B
uptregulated in

receptor

damage

subtype B

antagonist

Rexis
enhances

Reduces
May benefit
Glutathione

Glutathione

tissue
subtype C
peroxidase

peroxidase

damage

genes up-

regulated in

subtype C

IL-15,
Pro-

Immune
May benefit
IL-15 gene

NIZ985
inflammatory

stimulant
subtype C
expression is

cytokine

low in subtype

C relative to

Typa A

anti-
Keytruda/
anti-
Blocks upregulation
Increase
May benefit
Type B and C

PD-1
pembrolizumab
PD-1
of PD-1/PD-L1 to
adaptive
subtype B
patients have a

restore immune cell
immune
and C
suppressed

function
activity

adaptive

immune

response

anti-
Nivolumab
anti-
Blocks upregulation
Increase
May benefit
Type B and C

PD-1

PD-1
of PD-1/PD-L1 to
adaptive
subtype B
patients have a

restore immune cell
immune
and C
suppressed

function
activity

adaptive

immune

response

anti-
Ipilimumab/
anti-
CTLA-4 is a negative
Increase
May benefit
Type B and C

CTLA-4
YERVOY
CTLA-4
co-stimulatory
adaptive
subtype B
patients have a

molecule that acts in a
immune
and C
suppressed

fashion similar to PD-
activity

adaptive

1 to induce

immune

suppression of T cell

response, IL-2

function.

is also down-

regulated in

subtype B

patients

Ulinastatin
Ulinastatin
inactivates
The exact mechanism
anti-
May benefit
Type C patients

many serine
of action of
coagulant
subtype C
have

proteases,
ulinastatin in sepsis is
immune

upregulated

including
not clear, it is likely
activity

genes related to

trypsin,
that it may attenuate

coagulation

chymotrypsin,
the inflammatory

kallikrein,
response by acting at

plasmin,
multiple sites. Many

granulocyte
of the intermediaries

elastase,
in the systemic

cathepsin,
inflammatory

thrombin,
processes are serine

and factors
proteases. These

IXa, Xa,
include trypsin,

XIa, and
thrombin,

XlIa
chymotrypsin,

kallikrein, plasmin,

neutrophil elastase,

cathepsin, neutrophil

protease-3, and

coagulation factors

IXa, Xa, XIa, and

XlIa. It is now being

recognized that

besides their

proteolytic activity,

these proteases have

an important role in

regulation of

inflammation through

inter- and intracellular

signaling pathways.

To counter-regulate

the effect of these

proteases, several

protease inhibitors are

produced by the liver

in the presence of

inflammation; these

include acute phase

reactants such as α1-

antitrypsin and

proteins of the inter-

α-inhibitor family.

Urinary trypsin

inhibitor is one such

important protease

inhibitor found in

human blood and

urine; it has been also

referred to in the

literature as

ulinastatin or bikunin

Adalimumab
Humira
anti-

Decrease
May benefit
TNF-α gene

TNF-α

inflammation
subtype B
upregulated in

subtype B vs.

subtype A

Infliximab
Remicade
anti-

Decrease
May benefit
TNF-α gene

TNF-α

inflammation
subtype B
upregulated in

subtype B vs.

subtype A

p75 TNF
Adalimumab
p75 TNF

Decrease
May benefit
TNF-α gene

receptor

receptor,

inflammation
subtype B
upregulated

Binds TNF-a

subtype B vs.

subtype A

anti-C5a
Ultomiris
anti-C5a

Decrease
May benefit
Complement

inflammation
subtype B
system in

highly

activated in

subtype B vs.

subtype A

anti-C5a
Soliris
anti-C5a

Decrease
May benefit
Complement

inflammation
subtype B
system in

highly

activated in

subtype B vs.

subtype A

IL1R1
Kineret/
INF-gama,
TNF-α and IL-1 (a
Immune
May benefit
INF-gamma

anakinra
immune
term used for a family
stimulant
subtype B
gene is less

stimulant
of proteins, including

expressed in

IL-1α and IL-1β) are

subtype B vs.

powerful

subtype A

proinflammatory

cytokines that have

been implicated in a

large number of

infectious and

noninfectious

inflammatory

diseases. The release

of TNF-α from

macrophages begins

within 30 minutes

after the inciting

event, following gene

transcription and

RNA translation,

which established this

mediator to be an

early regulator of the

immune response.

TNF-α acts via

specific

transmembrane

receptors, TNF

receptor (TNFR)1,

and TNFR2, leading

to the activation of

immune cells and the

release of an array of

downstream

immunoregulatory

mediators. Likewise,

IL-1 is released

primarily from

activated

macrophages in a

timely manner similar

to TNF-α, signals

through two distinct

receptors, termed IL-1

receptor type I (IL-

1R1) and IL-1R2, and

has comparable

downstream effects

on immune cells

progesterone
reduces IL-6

Decrease
May benefit
Type B

and TNF-a

inflammation
subtype B
exhibits

relative high

gene

expression of

TNF-a

Thymosin
Thymalfasin
Thymosin alpha 1
Immune
May benefit
Thymosin

alpha I
peptide,
(Ta1) is a naturally
stimulant
subtype B
alpha I gene

(SciClone
T-lymphocyte
occurring thymic

highly

Pharmaceuticals,
subset
peptide. It acts as an

expressed in

Roche)
modulators;
endogenous regulator

subtype A vs.

Th1 cell
of both the innate and

subtype B and

stimulants;
adaptive immune

drug could

Th2-cell-
systems. It is used

increase

inhibitors
worldwide for treating

adaptive

diseases associated

immune

with immune

activity

dysfunction including

viral infections such

as hepatitis B and C,

certain cancers, and

for vaccine

enhancement

Actimmune
INF-gama,
TNF-α and IL-1 (a
Immune
May benefit
INF-gamma

immune
term used for a family
stimulant
subtype B
gene is less

stimulant
of proteins, including

expressed in

IL-1α and IL-1β) are

subtype B vs.

powerful

subtype A

proinflammatory

cytokines that have

been implicated in a

large number of

infectious and

noninfectious

inflammatory

diseases. The release

of TNF-α from

macrophages begins

within 30 minutes

after the inciting

event, following gene

transcription and

RNA translation,

which established this

mediator to be an

early regulator of the

immune response.

TNF-α acts via

specific

transmembrane

receptors, TNF

receptor (TNFR)1,

and TNFR2, leading

to the activation of

immune cells and the

release of an array of

downstream

immunoregulatory

mediators. Likewise,

IL-1 is released

primarily from

activated

macrophages in a

timely manner similar

to TNF-α, signals

through two distinct

receptors, termed IL-1

receptor type I

(IL-1R1) and IL-1R2,

and has comparable

downstream effects

on immune cells

defibrotide
protects the

anti-
May benefit
Type C has

cells lining

coagulant
subtype C
Plasminogen

bloods

activator

vessels and

inhibitor-1

preventing

upregulated

blood

clotting.

mixture of

single-

stranded

oligonucleotides

that is

purified

from the

intestinal

mucosa of

pigs

nangibotide
Anti-TREM-1,

Anti-
May benefit
TREM-1 gene

(MOTREM)
blocks

inflammatory
subtype B
is highly

TREM-1

expressed in

which is a

subtype A

trigger of

patients but

pathogen-

these patients

induced

already exhibit

inflammation

relatively low

mortality

EA-230
Reduces

Anti-
May benefit
Reducing IL-10

IL-6, IL-10,

inflammatory
subtype B
and TNF-a

INF-g, TNF-a,

whose genes

E-Selectin

are highly

expressed in B

may be

beneficial but

reducing INF-g

whose genes

are highly

expressed in

subtype A may

not be

beneficial

curcumin
NF-kB

Anti-
May benefit
Type A may

inhibitor

inflammatory
subtype B
benefit from

pathogen-

mediated

inflammation

that required

NF-kB

Emricasan
pan-caspsase

Anti-
May benefit
Up-regulated in

inhibitor

inflammatory
subtype B
subtype B vs. C

and A

IL1R1
Inhinits

Anti-
May benefit
May benefit

IL-1A, IL-1B,

inflammatory
subtype B
subtype B since

and IL-1

IL-1 receptor

receptor

antagonist gene

antagonist

is highly

expressed in

subtype B vs.

subtype A

AB103
peptide

Immune
May be
CD28 gene is

CD28

suppressant
contraindicated
highly

Antagonist

expressed in

subtype A

patients and

these patients

already exhibit

relatively low

mortality.

AB103 would

suppress

adaptive

immune

activity.

Filgrastim
G-CSF,

Immune
May benefit
G-CSF is

immune

stimulant
subtype C
highly

stimulant

expressed in

subtype C

which has the

worst outcomes

Sagramostim
GM-CSF,

Immune
May benefit
GM-CSF may

immune

stimulant
subtype B
increase innate

stimulant

and C
activity

associated with

pathogen

recognition and

subtype B and

C exhibit

down-

regulation of

immune

activity

associated with

pathogen

clearance.

Roncoleukin
IL-2,

Immune
May be
Would

promotes

suppressant
contraindicated
suppress

T-reg

immune

activity

adrecizumab
stabilizes/increases

Decrease
May benefit
Stabilizes a

adrenomedullin and

vascular
subtype C
vasodilator that

reverses

permeability

is already

vascular

down-regulated

permeability

in subtype C

while it's

receptors are

up-regulated in

subtype C

Talacotuzumab
Interleukin-

Immune
Type B and
IL-3 receptor

3-receptor-

stimulant
C may
up-regulated in

alpha-

benefit
subtype B and C

subunit-

antagonists

Mobista
Flt3 ligand,

Immune
May be
Would

Fms-like

suppressant
contraindicated
suppress

tyrosine

adaptive

kinase 3

immune

stimulants,

activity

increases

Treg

proliferation

Rituximab
Destroys B

Immune
May be
CD20 gene

cells

suppressant
contraindicated
expression is

expressing

up-regulated in

CD20

subtype A

GW-274150,
NOS

May benefit
INOS up-

Tilarginine,
inhibitor

subtype C
regulated in

Norathiol,

subtype C

targinine

timbetasin
synthetic

May benefit
TB4 up-

TB4

subtype C
regulated in

subtype B vs C

Peptide
TLR2

Anti-
May benefit
TLR2 is up-

P13
inhibitor

inflammatory
subtype B
regulated in

subtype B

Tinospora
TRL6

Anti-
May benefit
TLR6 is up-

cordifolia
inhibitor

inflammatory
subtype B
regulated in

derivative

subtype B

Tocilizumab
ACTE
Anti-

Anti-
May benefit
Type B patients

MRA
IL-6

inflammatory
subtype B
are inflammed

abatacept
Fc region
Suppression of
Immune
May be
Patients

of the
adaptive immune
suppression
contraindicated
exhibiting

immunoglobulin
activity. Abatacept

adaptive

IgG1
binds to the CD80 and

immune activity

fused to the
CD86 molecules, and

exhibit lower

extracellular
prevents co-

mortality

domain of
stimulation for T cell

CTLA-4
activation.

Abetimus
Made of four

Anti-
Type B and
Type B and C

double-

inflammatory
C patients
patients exhibit

stranded

may benefit
up-regulation of

oligodeoxyri

DAMP-mediated innate

bonucleotides

immune activity

that are

relative to

attached to a

subtype A

carrier

patients

platform and

are designed

to block

specific B-

cell anti

double

stranded

DNA

antibodies

Abrilumab
Anti-α4β7
α4β7 integrin is a
Anti-
Type B
Type B patients

antibody
validated target in
inflammatory
patients may
exhibit up-

inflammatory bowel

benefit
regulated

disease. Gut-specific

expression of

homing is the

TNF-alpha gene

mechanism by which

activated T cells and

antibody-secreting

cells (ASCs) are

targeted to both

inflamed and non-

inflamed regions of

the gut in order to

provide an effective

immune response.

This process relies on

the key interaction

between the integrin

α4β7 and the

addressin MadCAM-1

on the surfaces of the

appropriate cells.

Additionally, this

interaction is

strengthened by the

presence of CCR9, a

chemokine receptor,

which interacts with

TECK.

adalimumab
Anti-TNF-
Attenuation of pro-
Anti-
Type B
Type B patients

alpha
inflammatory
inflammatory
patients may
exhibit up-

cytokines TNF-alpha

benefit
regulated

and IL-6

expression of

TNF-alpha gene

Afelimomab
Anti-TNF-
Attenuation of pro-
Anti-
Type B
Type B patients

alpha
inflammatory
inflammatory
patients may
exhibit up-

cytokines TNF-alpha

benefit
regulated

and IL-6

expression of

TNF-alpha gene

Alefacept
Fusion
Suppression of
Immune
May be
Patients

protein
adaptive immune
suppressant
contraindicated
exhibiting

combining
activity. Inhibits the

adaptive

part of an
activation of CD4+

immune activity

antibody with
and CD8+ T cells by

exhibit lower

a protein that
interfering with CD2

mortality

blocks the
on the T cell

growth of
membrane thereby

some types of
blocking the

T cells
costimulatory

molecule LFA-3/CD2

interaction and

induces apoptosis of

memory-effector T

lymphocytes.

anakinra
Recombinant

Immune
Type B and
INF-gama gene

human

stimulant
C patients
is less expressed

interleukin-1

may benefit
in subtype B

receptor

and C vs.

antagonist

subtype A

Andecaliximab
Recombinant

Anti-
Type B and
Type B and C

chimeric

inflammatory
C patients
patients exhibit

IgG4

may benefit
up-regulation of

monoclonal

MMP9 and

antibody

DAMP-

against

mediated innate

metalloproteinase-9

immune activity

(MMP9)

relative to

subtype A

patients

Anrukinzumab
Anti-
IL-13 is a mediator of
Anti-
Type C
Type C patients

interleukin 13
allergic inflammatory
inflammatory
patients may
have IL-13 up-

monoclonal
response

benefit
regulated

antibody

relative to

subtype A

Anti-
Infusion of

Immune
May be
Patients

lymphocyte
animal-

suppressant
contraindicated
exhibiting

globulin
antibodies

adaptive

against

immune activity

human T

exhibit lower

cells

mortality

Anti
Infusion of

Immune
May be
Patients

thymocyte
horse or

suppressant
contraindicated
exhibiting

globulin
rabbit-

adaptive

derived

immune activity

antibodies

exhibit lower

against

mortality

human T

cells

antifolate
Class of
Interferes with cell-
Immune
May be
Patients

antimetabolite
mediated immune
suppressant
contraindicated
exhibiting

medications
response. Antifolates

pathogen-

that
act specifically during

specific innate

antagonise
DNA and RNA

and adaptive

the actions of
synthesis, and thus are

immune activity

folic acid
cytotoxic during the

exhibit lower

(vitamin B9),
S-phase of the cell

mortality

typically via
cycle, exhibiting a

inhibiting
greater toxic effect on

dihydrofolate
rapidly dividing cells

reductase
such as malignant

(DHFR)
cells, myeloid cells,

as well

gastrointestinal and

oral mucosa.

Apolizumab
Humanized

Immune
May be
Patients

monoclonal

suppressant
contraindicated
exhibiting

antibody

pathogen-

against HLA-

specific innate

DR beta

and adaptive

immune activity

exhibit lower

mortality

Apremilast
Small
Down-regulation of
Anti-
May benefit
Type B patients

molecule
pro-inflammatory
inflammatory
subtype B but
exhibit up-

inhibitor of
cytokines (e.g. TNF-

risk of
regulated

the enzyme
alpha) and up-

contraindication
expression of

phosphodiest
regulation of adaptive

TNF-alpha gene

erase 4
immune suppression

and up-

(PDE4)
(IL-10)

regulation of IL-

(enzyme that

10 which may

breaks down

suppress

cyclic

beneficial

adenosine

adaptive

monophosph

immune activity

ate (cAMP))

resulting in

down-

regulation if

TNF-alpha,

IL-17, and

IL-23, and

up-regulation

of IL-10

Aselizumab
Humanized
Interferes with
Immune
May be
Patients

monoclonal
leukocyte function
suppressant
contraindicated
exhibiting

antibody

adaptive

against

immune activity

CD62L

exhibit lower

mortality

Atezolizumab
Humanized,
Interferes with
Immune
Type B and
Type B and C

engineered
adaptive immune
stimulant
C patients
patients exhibit

monoclonal
suppression

may benefit
adaptive

antibody of

immune

IgG1 isotype

suppression,

against the

subtype B

protein

patients exhibit

programmed

up-regulation of

cell death-

PD-L1 gene

ligand 1

relative to other

types, subtype C

patients exhibit

up-regulation of

PD-1 gene, and

patients

exhibiting

adaptive

immune activity

exhibit lower

mortality

Avelumab
Whole
Interruption of
Immune
Type B and
Type B and C

human
adaptive immune
stimulant
C patients
patients exhibit

monoclonal
suppression to

may benefit
adaptive

antibody of
increase adaptive

immune

isotype IgG1
immune activity.

suppression,

that binds
Formation of a PD-

subtype B

to the
1/PD-L1

patients exhibit

programmed
receptor/ligand

up-regulation of

death-ligand
complex leads to

PD-L1 gene

1 (PD-L1)
inhibition of CD8+

relative to other

T cells, and therefore

types, subtype C

inhibition of an

patients exhibit

immune reaction.

up-regulation of

Avelumab blocks the

PD-1 gene, and

formation of PD-

patients

1/PDL1 ligand pairs

exhibiting

is blocked and CD8+

adaptive

T cell immune

immune activity

response should be

exhibit lower

increased.

mortality

azathioprine
Azathioprine
By inhibiting purine
Immune
May be
Patients

inhibits
synthesis, less DNA
suppressant
contraindicated
exhibiting

purine
and RNA are

adaptive

synthesis.
produced for the

immune activity

Purines are
synthesis of white

exhibit lower

needed to
blood cells, thus

mortality

produce
causing

DNA and
immunosuppression.

RNA.

Basiliximab
Chimeric
Prevents T cells from
Immune
May be
Patients

mouse-
replicating and from
suppressant
contraindicated
exhibiting

human
activating B cells and

adaptive

monoclonal
thus production of

immune activity

antibody to
antibodies

exhibit lower

the α chain

mortality

(CD25) of

the IL-2

receptor of T

cells

Belatacept
Fusion
Suppression of
Immune
May be
Patients

protein
adaptive immune
suppressant
contraindicated
exhibiting

composed of
activity. Prevents co-

adaptive

the Fc
stimulation for T cell

immune activity

fragment of a
activation.

exhibit lower

human IgG1

mortality

immunoglobulin

linked to the

extracellular

domain of

CTLA-4

Belimumab
Human
Belimumab reduces
Immune
May be
Patients

monoclonal
the number of
suppressant
contraindicated
exhibiting

antibody that
circulating B cells

adaptive

inhibits B-

immune activity

cell

exhibit lower

activating

mortality

factor

(BAFF)

Benralizumab
Murine
Binds to IL-5R via its
Immune
May be
Patients

humanized
Fab domain, blocking
suppressant
contraindicated
exhibiting

monocolonal
the binding of IL-5 to

adaptive

antibody
its receptor and

immune activity

against the
resulting in inhibition

exhibit lower

alpha-chain
of eosinophil

mortality

of the
differentiation and

interleukin-5
maturation in bone

receptor
marrow. In addition,

(CD125)
this antibody is able

to bind through its

afucosylated Fc

domain to the RIIIa

region of the Fcy

receptor on NK cells,

macrophages, and

neutrophils, thus

strongly inducing

antibody-dependent,

cell-mediated

cytotoxicity in both

circulating and tissue-

resident eosinophils.

Bertilimumab
Human
CCL11 selectively
Immune
May be
Patients

monoclonal
recruits eosinophils
suppressant
contraindicated
exhibiting

antibody that
by inducing their

adaptive

binds to
chemotaxis, and

immune activity

eotaxin-1
therefore, is

exhibit lower

implicated in allergic

mortality

responses.

Besilesomab
Mouse
Diagnostic use only
Immune
May be
Diagnostic use

monoclonal

suppressant
contraindicated
only

antibody

labelled

with the

radioactive

isotope

technetium-

99m. It is

used to detect

inflammatory

lesions and

metastases. It

binds to an

immunoglobulin,

IgG1

isotype.

Bleselumab
Anti-CD40
CD40 is a
Immune
May be
Patients

monoclonal
costimulatory protein
suppressant
contraindicated
exhibiting

antibody
found on antigen-

adaptive

presenting cells and is

immune activity

required for their

exhibit lower

activation

mortality

Blisibimod
Tetrameric
Antagonist of B -cell
Immune
May be
Patients

BAFF
activating factor
suppressant
contraindicated
exhibiting

binding
(BAFF)

adaptive

domain fused

immune activity

to a human

exhibit lower

IgG1 Fc

mortality

region

Brazikumab
Monoclonal
Inhibits Th17 function
Immune
May be
Patients

antibody that

suppressant
contraindicated
exhibiting

binds to the

adaptive

IL23 receptor

immune activity

exhibit lower

mortality

Briakinumab
Human
IL-12 is involved in
Immune
May be
Patients

monoclonal
the differentiation of
suppressant
contraindicated
exhibiting

antibody
naive T cells into Th1

adaptive

targetting
cells

immune activity

IL-12 and

exhibit lower

IL-23

mortality

Brodalumab
Human
Blocks recruitment of
Immune
May be
Patients

monoclonal
immune cells, such as
suppressant
contraindicated
exhibiting

antibody
monocytes and

adaptive

targetting
neutrophils to the site

immune activity

interleukin
of inflammation.

exhibit lower

17 receptor A

mortality

Canakinumab
Human
Attenuates IL-1 beta
Anti-
Type B
Type B patients

monoclonal

inflammatory
patients may
exhibit up-

antibody

benefit
regulation of

targeted at

inflammatory

interleukin-1

cytokines

beta

Carlumab
Human
CCL2 recruits
Immune
May be
Patients

recombinant
monocytes, memory
suppressant
contraindicated
exhibiting

monoclonal
T cells, and dendritic

adaptive

antibody
cells to the sites of

immune activity

(type IgG1
inflammation

exhibit lower

kappa) that
produced by either

mortality

targets
tissue injury or

human CC
infection

chemokine

ligand 2

(CCL2)

Cedelizumab
Murine
CD4+ T helper cells
Immune
May be
Patients

humanized
are white blood cells
suppressant
contraindicated
exhibiting

monocolonal
that are an essential

adaptive

antibody
part of the human

immune activity

against CD4
immune system.

exhibit lower

Depletion impairs

mortality

immune activity.

Certolizumab
Fragment of
Attenuates TNF-alpha
Anti-
Type B
Type B patients

pegol
a monoclonal

inflammatory
patients may
have up-

antibody

benefit
regulated TNF-

specific to

alpha gene

tumor

expression

necrosis

relastive to

factor alpha

subtype A

patients

chloroquine
Antimalarial
Against rheumatoid
Immune
May be
Patients

drug
arthritis, it operates by
suppressant
contraindicated
exhibiting

inhibiting lymphocyte

adaptive

proliferation,

immune activity

phospholipase A2,

exhibit lower

antigen presentation

mortality thus

in dendritic cells,

inhibition of

release of enzymes

lymphocyte

from lysosomes,

proliferation and

release of reactive

antigen

oxygen species from

presentation

macrophages, and

could be

production of IL-1.

detrimental,

subtype B

patients have

up-regulation of

pro-

inflammatory

cytokines

including

phospholipase

A2 activity thus

inhibition of

phospholipase

A2, release of

enzymes from

lysosomes,

release of

reactive oxygen

species from

macrophages,

and production

of IL-1 could be

beneficial, and

subtype C

patients

similarly exhibit

inflammation

from cell and

tissue damage

and thus

inhibition of

enzyme release

and reactive

oxygen species

may be

beneficial in

these patients.

Clazakizumab
Aglycosylated,
Attenuation of pro-
Anti-
Type B
Type B patients

humanized
inflammatory
inflammatory
patients may
have up-

rabbit
cytokine IL-6

benefit
regulated pro-

monoclonal

inflammatory

antibody

cytokines

against

interleukin-6

Clenoliximab
Chimeric
CD4+ T helper cells
Immune
May be
Patients

Macacairus/
are white blood cells
suppressant
contraindicated
exhibiting

Homo

that are an essential

adaptive

sapiens

part of the human

immune activity

monoclonal
immune system.

exhibit lower

antibody
Depletion impairs

mortality

against CD4
immune activity.

corticosteroids
Class of
Anti-inflammatory,
Anti-
Type B and
Immunosupressive

steroid
immunosuppressive,
inflammatory
C patients
effects may

hormones
anti-proliferative, and

may benefit
harm subtype A

that are
vasoconstrictive

patients, anti-

produced in
effects

inflammatory

the adrenal

effects may

cortex of

benefit subtype

vertebrates,

B patients,

as well as the

vasoconstrictive

synthetic

effects may

analogues of

benefit subtype

these

C patients.

hormones

cyclosporine
Immunosuppressant
Lower the activity of
Immune
May be
Patients

medication
T-cells
suppressant
contraindicated
exhibiting

and natural

adaptive

product

immune activity

exhibit lower

mortality

Daclizumab
Humanized
Reduction of T-cell
Immune
May be
Patients

monoclonal
responses and
suppressant
contraindicated
exhibiting

antibody that
expansion of CD56

adaptive

binds to
bright natural killer

immune activity

CD25, the
cells

exhibit lower

alpha subunit

mortality

of the IL-2

receptor of

T-cells

Hydroxychloroquine
Antimalarial
Against rheumatoid
Immune
May be
Type A

amyloquilone
arthritis, it operates by
suppressant
contraindicated
patients exhibit

drug
inhibiting lymphocyte

lower mortality

proliferation,

and thus

phospholipase A2,

inhibition of

antigen presentation

lymphocyte

in dendritic cells,

proliferation

release of enzymes

and antigen

from lysosomes,

presentation

release of reactive

could prolong

oxygen species from

viral clearance.

macrophages, and

subtype B

production of IL-1

patients exhibit

up-regulation

of pro-

inflammatory

cytokines and

thus the anti-

inflammatory

properties of

hydroxychloro

quine may be

beneficial to

these patients.

Azithromycin
Macrolide
Exhibit anti-
Anti-
Type B
Type B patients

antibiotic
inflammatory
inflammatory
patients may
exhibit up-

properties via

benefit
regulation of

suppression of pro-

pro-

inflammatory host

inflammatory

response that may

cytokines and

contribute to

thus the anti-

inflammation of the

inflammatory

airways

properties of

azithromycin

may be

beneficial to

these patients.

Anti-GM-

Immune
May be
GM-CSF may

CSF

suppresant
contraindicated
increase innate

activity

associated with

pathogen

recognition and

subtype B and

C exhibit

down-

regulation of

immune

activity

associated with

pathogen

clearance.

CD24Fc
DAMP

Anti-
Type B and
Type B and C

receptor

inflammtory
C patients
patients exhibit

blocker

may benefit
up-regulation

of DAMPs

which may

contribute to

inflammation

VI.A. Dysregulated Host Response Patient Subtype A

VI.A.1. Corticosteroids

As discussed in detail above, septic patients that remain hypotensive and require vasopressors to maintain a mean arterial pressure ≥65 mmHg are characterized as having septic shock—a condition that exhibits a hospital mortality in excess of 40%. Septic shock patients that show no clinical improvement (defined as having a systolic blood pressure <90 mmHg for more than one hour following both adequate fluid resuscitation and vasopressor therapy) are deemed refractory to vasopressor therapy and are thus characterized as refractory septic shock patients. In many cases, refractory septic shock patients are given corticosteroid therapy, such as hydrocortisone, based on rationale that the therapy may enable vasopressor responsiveness.

To evaluate the efficacy of hydrocortisone therapy in sepsis patients having subtypes A, B, and C, differential expression of the genes of Table 7 that are associated with pharmacology of hydrocortisone therapy were evaluated for the subtypes A, B, and C. Specifically, FIG. 11 depicts differential expression of the genes of Table 7 that are associated with pharmacology of hydrocortisone therapy (e.g., regulation of the glucocorticoid receptor signaling pathway) for the subtypes A, B, and C, in accordance with an embodiment. As shown in FIG. 11, subtype A patients exhibit differential expression of genes associated with glucocorticoid receptor signaling than subtype B patients. Specifically, relative to subtype B patients, subtype A patients exhibit down-regulation of genes associated with positive regulation of the glucocorticoid receptor signaling pathway, but up-regulation of genes associated with negative regulation of the glucocorticoid receptor signaling pathway. In other words, relative to subtype A patients, subtype B patients exhibit up-regulation of genes associated with positive regulation of the glucocorticoid receptor signaling pathway, but down-regulation of genes associated with negative regulation of the glucocorticoid receptor signaling pathway.

Due to this differential expression of genes associated with glucocorticoid receptor signaling between patients of subtypes A, B, and C, it was hypothesized that hydrocortisone therapy is differentially effective for the different subtypes. To test this hypothesis, multiple cohort datasets were analyzed for differential expressions and survival rates to evaluate the effect of hydrocortisone among different dysregulated host response subtypes. Specifically, the constructed classifiers discussed above. were applied to two placebo-controlled trials: the VANISH trial and a Burn-Induced SIRS trial to evaluate the survival rate of the patients that received hydrocortisone therapy.^{13, 50}

To evaluate hydrocortisone therapy response in dysregulated host response patients of the identified dysregulated host response patient subtypes, as discussed in detail below³¹the patient subtype classifiers were applied to a transcriptomic dataset from a placebo-controlled hydrocortisone clinical trials in sepsis patients and burn-induced SIRS patients that failed to show a difference in mortality between the treatment and placebo arms of the trial. Differential responses to hydrocortisone therapy were identified for the different patient subtypes. Specifically, one patient subtype is shown to benefit from hydrocortisone, and one or both of the other patient subtypes are shown to worsen with hydrocortisone.

The test expression data from each trial were normalized by the platform normalization matrix described above¹³so that the test data were more consistent with the training data. The classifiers (e.g., the Full Model, the SS Model, the S Model, and the P Model) were then applied to the normalized data such that the patients were classified into A, B, and C subtypes. In contrast to the COCONUT method, the normalization approach described herein is simpler because it does not use controls and instead employs a platform normalization matrix, and then selects all of the samples from the matrix used by the target platform of the target sample and then co-normalizes them together. Therefore, each sample in the target samples was normalized independently with the normalization matrix of the sample array platform.

Survival and mortality rates were calculated at day 28 because survival and mortality labels at other time points were not available. A single-time-point survival analysis was performed to observe the difference of survival rate between the hydrocortisone therapy group and the placebo group in each subtype. Binomial and Chi-squared with continuity correction tests were used to test for the significance of these differences. Mortality reduction when ruling out hydrocortisone was calculated as: 1−(Placebo's mortality rate/Hydrocortisone's mortality rate). Conversely, mortality reduction when ruling in hydrocortisone was calculated as: 1−(Hydrocortisone's mortality rate/Placebo's mortality rate), whichever denominator was larger.

Tables 9-16 below depict the survival analyses for each subtype (e.g., A, B, and C) for each classifier (e.g., the Full Model, the SS Model, the S Model, and the P Model) for each trial. Specifically, Tables 9 and 13 depicts survival analysis for each subtype (e.g., A, B, and C) for the Full Model, Tables 10 and 14 depicts survival analysis for each subtype (e.g., A, B, and C) for the SS Model, Tables 11 and 15 depicts survival analysis for each subtype (e.g., A, B, and C) for the S Model, and Tables 12 and 16 depicts survival analysis for each subtype (e.g., A, B, and C) for the P Model.

TABLE 9

Full Model VANISH Trial Survival Analysis

Hydro
B
C
A
CA
Total

Alive
16
11
9
20
36

Dead
6
8
8
16
22

Total
22
19
17
36
58

Survival rate
72.7%
57.9%
52.9%
55.6%
62.1%

Mortality rate
27.3%
42.1%
47.1%
44.4%
37.9%

Placebo

Alive
11
18
15
33
44

Dead
8
4
3
7
15

Total
19
22
18
40
59

Survival rate
57.9%
81.8%
83.3%
82.5%
74.6%

Mortality rate
42.1%
18.2%
16.7%
17.5%
25.4%

Grand total
41
41
35
76
117

Mortality Reduction
Hydro
Placebo
Placebo
Placebo
Placebo

Group

Binomial P-value
1E−01
1E−02
3E−03
2E−04
2E−02

Chi-squared P-value
5E−01
2E−01
1E−01
2E−02
2E−01

Mortality reduction
35.2%
56.8%
64.6%
60.6%
33.0%

Overall Mortality
34.1%
29.3%
31.4%
30.3%
31.6%

TABLE 10

SS Model VANISH Trial Survival Analysis

Hydro
B
C
A
CA
Total

Alive
5
16
15
31
36

Dead
5
8
9
17
22

Total
10
24
24
48
58

Survival rate
50.0%
66.7%
62.5%
64.6%
62.1%

Mortality rate
50.0%
33.3%
37.5%
35.4%
37.9%

Placebo

Alive
9
17
18
35
44

Dead
6
8
1
9
15

Total
15
25
19
44
59

Survival rate
60.0%
68.0%
94.7%
79.5%
74.6%

Mortality rate
40.0%
32.0%
5.3%
20.5%
25.4%

Grand total
25
49
43
92
117

Mortality Reduction
Placebo
Placebo
Placebo
Placebo
Placebo

Group

Binomial P-value
4E−01
5E−01
2E−06
1E−02
2E−02

Chi-squared P-value
9E−01
8E−01
3E−02
2E−01
2E−01

Mortality reduction
20.0%
4.0%
86.0%
42.2%
33.0%

Overall Mortality
44.0%
32.7%
23.3%
28.3%
31.6%

TABLE 11

S Model VANISH Trial Survival Analysis

Hydro
B
C
A
CA
Total

Alive
12
8
16
8
20

Dead
9
4
9
4
13

Total
21
12
25
12
33

Survival rate
57.1%
66.7%
64.0%
66.7%
60.6%

Mortality rate
42.9%
33.3%
36.0%
33.3%
39.4%

Placebo

Alive
17
8
19
8
25

Dead
9
5
1
5
14

Total
26
13
20
13
39

Survival rate
65.4%
61.5%
95.0%
61.5%
64.1%

Mortality rate
34.6%
38.5%
5.0%
38.5%
35.9%

Grand total
47
25
45
25
72

Mortality Reduction
29
16
35
16
45

Group

Binomial P-value
18
9
10
9
27

Chi-squared P-value
0.0762
0.0225
4.5145
0.0225
0.0037

Mortality reduction
Placebo
Hydro
Placebo
Hydro
Placebo

Overall Mortality
0.28
0.48
0.000002
5E−01
4E−01

TABLE 12

P Model VANISH Trial Survival Analysis

Hydro
B
C
A
CA
Total

Alive
23
4
9
13
36

Dead
8
4
10
14
22

Total
31
8
19
27
58

Survival rate
74.2%
50.0%
47.4%
48.1%
62.1%

Mortality rate
25.8%
50.0%
52.6%
51.9%
37.9%

Placebo

Alive
20
9
15
24
44

Dead
8
5
2
7
15

Total
28
14
17
31
59

Survival rate
71.4%
64.3%
88.2%
77.4%
74.6%

Mortality rate
28.6%
35.7%
11.8%
22.6%
25.4%

Grand total
59
22
36
58
117

Mortality Reduction
Hydro
Placebo
Placebo
Placebo
Placebo

Group

Binomial P-value
5E−01
3E−01
2E−05
9E−04
2E−02

Chi-squared P-value
1E+00
8E−01
2E−02
4E−02
2E−01

Mortality reduction
9.7%
28.6%
77.6%
56.5%
33.0%

Overall Mortality
27.1%
40.9%
33.3%
36.2%
31.6%

TABLE 13

Full Model Burn-Induced SIRS Trial Survival Analysis

Hydro
B
C
A
CA
Total

Alive
6
1
2
3
9

Dead
1
4
1
5
6

Total
7
5
3
8
15

Survival rate
85.7%
20.0%
66.7%
37.5%
60.0%

Mortality rate
14.3%
80.0%
33.3%
62.5%
40.0%

Placebo

Alive
3
5
5
10
13

Dead
2
0
0
0
2

Total
5
5
5
10
15

Survival rate
60.0%
100.0%
100.0%
100.0%
86.7%

Mortality rate
40.0%
0.0%
0.0%
0.0%
13.3%

Grand total
12
10
8
18
30

Mortality
Hydro
Placebo
Placebo
Placebo
Placebo

Reduction

Group

Binomial
2E−01
0E+00
0E+00
0E+00
1E−02

P-value

Chi-squared
7E−01
5E−02
8E−01
2E−02
2E−01

P-value

Mortality
64.3%
100.0%
100.0%
100.0%
66.7%

reduction

Overall
25.0%
40.0%
12.5%
27.8%
26.7%

Mortality

TABLE 14

SS Model Burn-Induced SIRS Trial Survival Analysis

Hydro
B
C
A
CA
Total

Alive
2
3
4
7
9

Dead
0
4
2
6
6

Total
2
7
6
13
15

Survival rate
100.0%
42.9%
66.7%
53.8%
60.0%

Mortality rate
0.0%
57.1%
33.3%
46.2%
40.0%

Placebo

Alive
5
4
4
8
13

Dead
0
1
1
2
2

Total
5
5
5
10
15

Survival rate
100.0%
80.0%
80.0%
80.0%
86.7%

Mortality rate
0.0%
20.0%
20.0%
20.0%
13.3%

Grand total
7
12
11
23
30

Mortality Reduction
Placebo
Placebo
Placebo
Placebo
Placebo

Group

Binomial P-value
1E+00
3E−02
3E−01
3E−02
1E−02

Chi-squared P-value
—
5E−01
9E−01
4E−01
2E−01

Mortality reduction
—
65.0%
40.0%
56.7%
66.7%

Overall Mortality
0.0%
41.7%
27.3%
34.8%
26.7%

TABLE 15

S Model Burn-Induced SIRS Trial Survival Analysis

Hydro
B
C
A
CA
Total

Alive
6
0
3
3
9

Dead
1
3
2
5
6

Total
7
3
5
8
15

Survival rate
85.7%
0.0%
60.0%
37.5%
60.0%

Mortality rate
14.3%
100.0%
40.0%
62.5%
40.0%

Placebo

Alive
4
4
5
9
13

Dead
1
0
1
1
2

Total
5
4
6
10
15

Survival rate
80.0%
100.0%
83.3%
90.0%
86.7%

Mortality rate
20.0%
0.0%
16.7%
10.0%
13.3%

Grand total
12
7
11
18
30

Mortality Reduction
Hydro
Placebo
Placebo
Placebo
Placebo

Group

Binomial P-value
6E−01
0E+00
2E−01
4E−04
1E−02

Chi-squared P-value
6E−01
6E−02
9E−01
7E−02
2E−01

Mortality reduction
28.6%
100.0%
58.3%
84.0%
66.7%

Overall Mortality
16.7%
42.9%
27.3%
33.3%
26.7%

TABLE 16

P Model Burn-Induced SIRS Trial Survival Analysis

Hydro
B
C
A
CA
Total

Alive
5
0
4
4
9

Dead
1
4
1
5
6

Total
6
4
5
9
15

Survival rate
83.3%
0.0%
80.0%
44.4%
60.0%

Mortality rate
16.7%
100.0%
20.0%
55.6%
40.0%

Placebo

Alive
4
3
6
9
13

Dead
2
0
0
0
2

Total
6
3
6
9
15

Survival rate
66.7%
100.0%
100.0%
100.0%
86.7%

Mortality rate
33.3%
0.0%
0.0%
0.0%
13.3%

Grand total
12
7
11
18
30

Mortality
Hydro
Placebo
Placebo
Placebo
Placebo

Reduction

Group

Binomial
4E−01
0E+00
0E+00
0E+00
1E−02

P-value

Chi-squared
1E+00
6E−02
9E−01
4E−02
2E−01

P-value

Mortality
50.0%
100.0%
100.0%
100.0%
66.7%

reduction

Overall
25.0%
57.1%
9.1%
27.8%
26.7%

Mortality

An alternative method for identifying patients that may be harmed by immunosuppressive effects of hydrocortisone is based on employing A and B scores to identify patients expected to exhibit increased immune activity and lower inflammation. In simple terms, this method is based on a classifying patients with a high A score and low B score.

In one example, previously identified subtypes of sepsis patients were used to tune the model to identify these type A and B patients. Two distinct sepsis response signatures (SRS1 and SRS2) were identified in five public studies (E-MTAB-4421, E-MTAB-4451, E-MTAB-5273, E-MTAB-5274, and E-MTAB-7581), where HumanHT-12 v4 BeadChip were used to generate the gene expression profiles of the patient samples. The processed data of those five studies were downloaded and processed using R programming language and software environment for statistical analysis (version 3.6.3). The Bioconductor annotation package, illuminaHumanv4.db (version 1.26.0), was used to annotate microarray probes and expression levels of genes were determined by each individual probe or mean of probes belonging to the same gene. In order to remove cohort biases, the Bioconductor package, limma (version 3.42.2), was used to remove batch effects. Using ss.b2 panel genes, subtype A, B, and C scores were calculated by geometric mean of up/down genes.

To build the classifier, we defined E-MTAB-4421, E-MTAB-4451, E-MTAB-5273, and E-MTAB-5274 as the training dataset and VANISH (E-MTAB-7581) as the testing dataset. We used features (subtype A, B, and C scores) and class labels (SRS1 vs SRS2) in the training dataset to build a machine-learned classifier based on support-vector machine (SVM) method. SVM is a supervised machine learning method for classification analysis. The algorithm finds a single or a set of hyperplanes that maximize the margin among subtype A, B, and C scores. In order to capture non-linear data, the kernel function was used. R package e1071 was used to build the SVM classifier with following parameters: method=“C-classification”, kernal=“radial”, gamma=0.1, and cost=10.

The accuracy of the classifiers was evaluated by Leave-One-Out (LOO) cross-validation over the training dataset. Also the classifier was applied to 117 controlled samples from VANISH trial. The patients predicted as Type-A (SRS2-like) exhibited significant 28-day mortality reduction when hydrocortisone was applied in comparison to placebo. These Type-A exhibited 75.5% mortality reduction in the placebo group in comparison to the hydrocortisone group (Fisher exact test p-value 0.0093). The Type-A (SRS2-like) and Type-B (SRS1-like) classifier exhibited an accuracy of 88.6%. Table 17 below depict the survival analyses for each subtype for the SS.B2 model.

TABLE 17

SS.B2 Model VANISH Trial Survival Analysis

Hydro
A
B
Total

Alive
22
14
36

Dead
14
8
22

Total
36
22
58

Survival rate
61.1%
63.6%
62.1%

Mortality rate
38.9%
36.4%
37.9%

Placebo

Alive
31
13
44

Dead
3
12
15

Total
34
25
59

Survival rate
91.2%
52.0%
74.6%

Mortality rate
8.8%
48.0%
25.4%

Grand total
70
47
117

Mortality Reduction
Placebo
Hydro
Placebo

Group

Fisher exact test
5E−03
6E−01
2E−01

Binomial P-value
1E−06
2E−01
2E−02

Chi-squared P-value
8E−03
6E−01
2E−01

Mortality reduction
77.3%
24.2%
33.0%

Overall Mortality
24.3%
42.6%
31.6%

In addition to the SVM-method, thresholds can be employed to scores in order to define A vs. B labels. We discovered that subtype A and B scores play an important role in subtype SRS1 and SRS2 classification. Therefore we applied a heuristic threshold (threshold=0) to subtype A and B scores to classify SRS1-like and SRS2-like in VANISH: SRS2-like label was assigned to samples with subtype A score >0 and subtype B score <0 and SRS1-like label was assigned to the rest of samples. With the simple heuristic threshold (threshold=0), the patients predicted as SRS2-like exhibited 85.2% 28-day mortality reduction in the placebo group in comparison to the hydrocortisone group (Fisher exact test p-value 0.0159).

Besides the heuristic threshold, we also derived the thresholds for subtype A and B scores using the training dataset. Same as the SVM method, we defined E-MTAB-4421, E-MTAB-4451, E-MTAB-5273, and E-MTAB-5274 as the training dataset and VANISH (E-MTAB-7581) as the testing dataset. To identify the best threshold of subtype A score to classify subtype SRS1 and SRS2 in training dataset, we fitted subtype A scores and SRS subtype labels into ROC curve (receiver operating characteristic curve) and identified the threshold for subtype A score (A threshold=−0.2664) which is the closest point to the top-left part of the plot with perfect sensitivity or specificity. With the same method, the optimal B score threshold (B threshold=0.3179) was selected to classify subtype SRS1 and SRS2 in the training set. We applied the defined optimal thresholds for subtype A and B scores to the VANISH trail: the samples whose subtype A scores are above the A threshold and subtype B scores are below the B threshold were labeled as SRS2-like and the rest VANISH trail samples were labeled as SRS1-like. With such classification, the patients with the SRS2-like label showed 81.7% 28-day mortality reduction in the placebo group in comparison to the hydrocortisone group (Fisher exact test p-value 0.0065).

Various thresholds can be employed in order to optimize for mortality reduction (mr) and for the number of patients who may benefit (percentage of patients that are B). Table 18 below depict the survival analyses for each subtype for the SS.B2 model.

TABLE 18

SS.B2 Model VANISH Trial Survival Analysis

hydro_alive,

Accuracy
% of

Fisher
hydro_dead,

A score
B score
in training
patients

exact
placebo_alive,

class
cutoff
cutoff
set
that are B
mr
test
placebo_dead

B
0.0000
0.0000
0.74830
23.08%
83.8%
0.01831837
7, 9, 10, 1

A
0.0000
0.0000
0.74830
76.92%
5.8%
1
29, 13, 34, 14

B
−0.2664
0.3179
0.82483
41.88%
82.5%
0.00361372
13, 11, 23, 2

A
−0.2664
0.3179
0.82483
58.12%
15.4%
0.80002799
23, 11, 21, 13

B
0.0000
0.0000
0.75283
23.08%
83.8%
0.01831837
7, 9, 10, 1

A
0.0000
0.0000
0.75283
76.92%
5.8%
1
29, 13, 34, 14

B
−0.1277
0.3630
0.83673
39.32%
83.3%
0.00267971
11, 11, 22, 2

A
−0.1277
0.3630
0.83673
60.68%
17.7%
0.62103399
25, 11, 22, 13

B
0.0000
0.0000
0.76871
23.08%
83.8%
0.01831837
7, 9, 10, 1

A
0.0000
0.0000
0.76871
76.92%
5.8%
1
29, 13, 34, 14

B
−0.0406
0.1067
0.79592
25.64%
87.3%
0.00669665
7, 9, 13, 1

A
−0.0406
0.1067
0.79592
74.36%
0.5%
1
29, 13, 31, 14

B
0.0000
0.0000
0.74830
23.93%
85.2%
0.01587078
7, 9, 11, 1

A
0.0000
0.0000
0.74830
76.07%
3.8%
1
29, 13, 33, 14

B
−0.2664
0.3179
0.82483
39.32%
81.7%
0.00651932
12, 10, 22, 2

A
−0.2664
0.3179
0.82483
60.68%
10.3%
0.80648544
24, 12, 22, 13

B
0.0000
0.0000
0.75283
17.09%
82.5%
0.02810193
4, 7, 8, 1

A
0.0000
0.0000
0.75283
82.91%
12.3%
0.82470462
32, 15, 36, 14

B
−0.1277
0.3630
0.83673
29.91%
79.0%
0.01164257
8, 9, 16, 2

A
−0.1277
0.3630
0.83673
70.09%
0.0%
1
28, 13, 28, 13

B
0.0000
0.0000
0.76871
37.61%
86.8%
0.01260373
15, 10, 18, 1

A
0.0000
0.0000
0.76871
62.39%
3.8%
1
21, 12, 26, 14

B
−0.0406
0.1067
0.79592
41.88%
79.2%
0.01807688
15, 10, 22, 2

A
−0.0406
0.1067
0.79592
58.12%
2.1%
1
21, 12, 22, 13

Response to hydrocortisone therapy for each subtype of patients identified by each Model for each of the sepsis and burn-induced SIRS patient studies, was evaluated based on a p-values and mortality reduction for the subtype. To evaluate possible adverse response to hydrocortisone therapy for a subtype, a binomial p-value was calculated as the probability of achieving a random survival rate of less than or equal to the survival rate P(X<=x) observed in the hydrocortisone therapy group for the subtype, assuming that the survival rate observed in the placebo therapy group for the subtype was an expected survival rate for patients receiving no hydrocortisone therapy. To evaluate possible favorable response to hydrocortisone therapy for a subtype, a chi-squared p-value was calculated for the survival rate P(X>=x) observed in the hydrocortisone therapy group for the subtype. The chi-squared p-value was calculated with continuity correction when computed for 2-by-2 tables.

As an example, assuming a chosen, statistically significant p-value of at least 0.1, hydrocortisone therapy response was evaluated based on mortality reduction, as well as binomial and chi-squared p-values, for sepsis and SIRS patients, for each subtype, and for each Model, as follows.

First, assuming the chosen statistically significant p-value of at least 0.1, hydrocortisone therapy response was evaluated for sepsis patients. As shown in Tables 9-12, sepsis patients assigned to the A subtype by the Full, SS, S, and P Models exhibited statistically significant 28-day mortality reduction when placebo was applied in comparison to hydrocortisone therapy. Specifically, the Full, SS, S, and P Models identified a subtype, A, exhibiting 64.6%, 86.0%, 86.1%, and 77.6%, respectively, lower mortality in the placebo group when compared to the hydrocortisone therapy group. As shown in Table 9, sepsis patients assigned to the B subtype by the Full Model exhibited statistically significant 28-day mortality reduction when hydrocortisone was applied in comparison to placebo. Specifically, the Full Model identified a subtype, B, exhibiting 35.2%, lower mortality in the hydrocortisone group when compared to the placebo group. As shown in Table 9, sepsis patients assigned to the C subtype by the Full Model exhibited statistically significant 28-day mortality reduction when placebo was applied in comparison to hydrocortisone therapy. Specifically, the Full Model identified a subtype, C, exhibiting 56.8% lower mortality in the placebo group when compared to the hydrocortisone therapy group.

Additionally, assuming the chosen statistically significant p-value of at least 0.1, hydrocortisone therapy response was evaluated for SIRS patients. As shown in Tables 13, 15, and 16, SIRS patients assigned to the A subtype by the Full, S, and P Models exhibited statistically significant 28-day mortality reduction when placebo was applied in comparison to hydrocortisone therapy. Specifically, the Full, S, and P Models identified a subtype, A, exhibiting 100% lower mortality in the placebo group when compared to the hydrocortisone therapy group. As shown in Table 15, SIRS patients assigned to the B subtype by the S Model exhibited statistically significant 28-day mortality reduction when hydrocortisone was applied in comparison to placebo. Specifically, the S Model identified a subtype, B, exhibiting 28.6%, lower mortality in the hydrocortisone group when compared to the placebo group. As shown in Tables 13-16, SIRS patients assigned to the C subtype by the Full, SS, S, and P Models exhibited statistically significant 28-day mortality reduction when placebo was applied in comparison to hydrocortisone therapy. Specifically, the Full, SS, S, and P Models identified a subtype, A, exhibiting 100%, 65%, 100%, and 100%, respectively, lower mortality in the placebo group when compared to the hydrocortisone therapy group.

Based on these observations of differential mortality reduction as a result of hydrocortisone therapy or placebo therapy between the subtypes A, B, and C identified for both sepsis and SIRS patients by the Full, SS, S, and P Models, the subtypes can be assigned more descriptive titles such as “favorably responsive”, “adversely responsive”, and “non-responsive” to corticosteroid therapy. For example, assuming the chosen statistically significant p-value of at least 0.1, subtypes can be assigned titles as follows.

First, assuming the chosen statistically significant p-value of at least 0.1, subtypes A, B, and C identified for sepsis patients by the Full, SS, S, and P Models can be assigned titles as follows. Because mortality reduction was statistically significant in the placebo group compared to the hydrocortisone therapy group for sepsis patients assigned to subtype A by the Full, SS, S, and P Models, sepsis patients assigned to subtype A by at least one of the Full, SS, S, and P Models can be colloquially referred to as “adversely responsive” to corticosteroid therapy. Similarly, because mortality reduction was statistically significant in the placebo group compared to the hydrocortisone therapy group for sepsis patients assigned to subtype C by the Full Model, sepsis patients assigned to subtype C by the Full Model can be colloquially referred to as “adversely responsive” to corticosteroid therapy. Conversely, because mortality reduction was statistically significant in the hydrocortisone therapy group compared to the placebo group for sepsis patients assigned to subtype B by the Full Model, sepsis patients assigned to subtype B by the Full Model can be colloquially referred to as “favorably responsive” to corticosteroid therapy. Finally, because correlation between therapy group and mortality reduction was not statistically significant for sepsis patients assigned to subtypes B and C by the SS, S, and P Models, sepsis patients assigned to subtype B or C by at least one of the SS, S, and P Models can colloquially referred to as “non-responsive” to corticosteroid therapy.

Additionally, assuming the chosen statistically significant p-value of at least 0.1, subtypes A, B, and C identified for SIRS patients by the Full, SS, S, and P Models can be assigned titles as follows. Because mortality reduction was statistically significant in the placebo group compared to the hydrocortisone therapy group for SIRS patients assigned to subtype A by the Full, S, and P Models, SIRS patients assigned to subtype A by at least one of the Full, S, and P Models can be colloquially referred to as “adversely responsive” to corticosteroid therapy. Similarly, because mortality reduction was statistically significant in the placebo group compared to the hydrocortisone therapy group for SIRS patients assigned to subtype C by the Full, SS, S, and P Models, SIRS patients assigned to subtype C by at least one of the Full, SS, S, and P Models can be colloquially referred to as “adversely responsive” to corticosteroid therapy. Conversely, because mortality reduction was statistically significant in the hydrocortisone therapy group compared to the placebo group for SIRS patients assigned to subtype B by the S Model, SIRS patients assigned to subtype B by the S Model one can be colloquially referred to as “favorably responsive” to corticosteroid therapy. Because correlation between therapy group and mortality reduction was not statistically significant for SIRS patients assigned to subtype B by the Full, SS, and P Models, SIRS patients assigned to subtype B by at least one of the Full, SS, and P Models can be colloquially referred to as “non-responsive” to corticosteroid therapy. Finally, because correlation between therapy group and mortality reduction was not statistically significant for SIRS patients assigned to subtype A by the SS Model, SIRS patients assigned to subtype A by the SS Model can be colloquially referred to as “non-responsive” to corticosteroid therapy.

Further based on these observations of mortality reduction in sepsis and SIRS patients assigned to subtypes A, B, and C by the Full, SS, S, and P Models, subtyped sepsis and SIRS patients may be provided treatment recommendations accordingly. For instance, in one embodiment, patients subtyped as “favorably responsive” to corticosteroid therapy can be recommended treatment with corticosteroids, while patients subtyped as “adversely responsive” to corticosteroid therapy can be recommended no corticosteroid therapy, and while patients subtyped as “non-responsive” to corticosteroid therapy can be provided with no therapy recommendation.

Discrepancies between treatment recommendations for a given subtype (e.g., subtype A, B, or C) across models (e.g., the Full, SS, S, and P Model) and types of dysregulated host response (e.g., sepsis or SIRS) are due to the fact that statistical significance of mortality reduction for a subtype varies according to the model used to assign the subtype, as well as the type of dysregulated host response. For example, as discussed above, for a statistically significant p-value of at least 0.1, sepsis patients that are determined to be of subtype C by the Full Model may be subtyped as “adversely responsive” to corticosteroid therapy, and thus recommended no corticosteroid therapy. Conversely, for a statistically significant p-value of at least 0.1, sepsis patients that are determined to be of subtype C by the SS Model may be subtyped as “non-responsive” to corticosteroid therapy, and thus may not be provided with a therapy recommendation, while SIRS patients that are determined to be of subtype C by the SS Model may be subtyped as “adversely responsive” to corticosteroid therapy, and thus recommended no corticosteroid therapy. Therefore, the titles assigned to subtypes A, B, and C for each model and for each type of dysregulated host response, and thus the therapy recommendations, are dependent upon the chosen statistically significant p-value. In alternative embodiments, the statistically significant p-value may be adjusted, and thus the titles assigned to subtypes A, B, and C, as well as the therapy recommendations, may be adjusted. For example, in some embodiments, the statistically significant p-value may be less than 0.1.

Furthermore, as described above, survival analyses were performed independently from classifier training, which prevented the training from overfitting issues. Thus, the observations of differential response to corticosteroid therapy among the three different subtypes can likely be attributed to the fundamental link between therapy and the biological nature of each subtype. For instance, the most significant molecular functions from the GO analysis of the A subtype were antigen binding, MEW protein complex binding, and cytokine binding, which are strong indicators for adaptive immune response. In the survival analysis results for the A subtype, significant mortality reduction was observed in the placebo group compared to the corticosteroid therapy group, inferring that the corticosteroid therapy might be potentially disturbing the already working adaptive immune response of the A subtype patients. On the contrary, according to the GO analysis of the B subtype, the B subtype was significantly enriched with interleukin (IL)-1 receptor and complement component Cl, indicating a more likely innate immune response. Indeed, for the B subtype, instead of a mortality reduction in the placebo group, a mortality reduction was observed with corticosteroid therapy.

VI.B. Dysregulated Host Response Patient Subtypes B and C

VI.B.1. Immune Stimulants: Checkpoint Inhibitors, Interleukins, and Mediators of T-Cell Regulation Attenuation

As discussed in detail above, subtype B and C patients may benefit from immune stimulants. Examples of therapies for stimulating the immune system include checkpoint inhibitors, interleukins such as IL-7, and therapies that attenuate the regulation and suppression of T-cell function such as blockers of IL-10, and TGF-β.

As shown in FIG. 12, subtypes B and C exhibit down-regulation of immune markers including IL-7 and INF-γ. Conversely, subtype A exhibits up-regulation of immune markers including IL-7 and INF-γ. PD-1 and PD-L1 are receptor/ligand immune inhibitory cell surface markers. Checkpoint inhibition of PD-1/PD-L1 interaction results in upregulation of IL-7. As shown in FIG. 12, subtype B patients exhibit up-regulation of PD-L1 and down-regulation of IL-7. Thus subtype B patients may benefit from anti-PD-1 and anti-PD-L1 therapy.

CD28 interacts with CD86 and CD80 to mediate stimulation of T-cell function. CTLA-4 interacts with CD86 and CD80 to mediate inhibition of T-cell function. Checkpoint inhibition of CTLA-4 causes upregulation of INF-γ. As shown in FIG. 12, subtype B and C patients exhibit an increased ratio of CTLA-4/CD28 and decreased expression of INF-γ. Therefore, subtype B and C patients may benefit from anti-CTLA-4 therapy.

TIM-3 interacts with CEACAM-1 to mediate inhibition of T cell function. As shown in FIG. 12, subtype B and C patients exhibit up-regulation of CEACAM-1 and TIM-3. Therefore, subtype B and C patients may benefit from anti-CEACAM-1 and anti-TIM-3 therapy.

VI.C. Dysregulated Host Response Patient Subtype C

VI.C.1. Modulators of Coagulation and Modulators of Vascular Permeability

As discussed in detail above, subtype C patients exhibit coagulopathy and may benefit from modulators of coagulation such as anticoagulants and modulators of vascular permeability. Specifically, therapies that indirectly modulate coagulation factors, such as activated protein C and antithrombin, may be of particular benefit to subtype C patients due to the complexity of the coagulation system and difficulty of managing coagulation by targeting specific coagulation factors directly.

VII. Benefits Conferred by Systemic Immume Response Patient Subtype Classifiers

VII.A. Improvement of Acute Care

Syndromes caused by dysregulated host response, such as sepsis, are not single diseases, but rather are heterogeneous processes. As a result, evaluation of effective therapies has been hampered by limitations in the ability to classify patients into homogeneous subtypes based on pathogenesis. The improved ability to subtype patients exhibiting dysregulated host response can therefore enable identification and evaluation of effective new therapies for treating dysregulated host response syndromes such as sepsis.

VII.B. Precision Clinical Trials

The improved ability to subtype patients exhibiting dysregulated host response also enables the design and execution of precision clinical trials and the ability to test effectiveness potential new therapies by targeting the therapies to specific subtypes of patients. The improved ability to subtype patients exhibiting dysregulated host response also allows for predictive therapy enrichment in positively-responsive patients and avoiding the use of therapies in non-responsive or adversely-responsive patients.

FIG. 13 depicts an example of a precision clinical trial design, in accordance with an embodiment. FIG. 13 depicts an example of a precision platform clinical trial design, in accordance with an embodiment.

VII.C. Precision Care

The improved ability to subtype patients exhibiting dysregulated host response also enables the delivery of precision care. The patient subtype classifiers discussed throughout this disclosure allow for the development of tests for guiding dysregulated host response therapy, and in particular for guiding dysregulated host response therapy in acute care. Specifically, the patient subtype classifiers discussed throughout this disclosure can serve as a companion diagnostic to enable the safe and effective use of dysregulated host response therapy.

FIG. 14 depicts an example workflow for the use of the patient subtype classifiers discussed throughout this disclosure, in targeting therapies for septic shock patients, in accordance with an embodiment. The same approach can similarly be used to target therapies for patients exhibiting sepsis other than septic shock, as well as other dysregulated host response syndromes resulting from insults other than infection, such as burns, acute respiratory distress syndrome, acute kidney injury, and/or any other insults.

VII.D. Patient Subtyping Test

FIG. 15 depicts an example dysregulated host response patient subtyping test that employs an FDA-cleared patient sample collection system (e.g., PAXgene Blood RNA System), and an FDA-cleared Real Time PCR system (e.g. the Thermo Fisher Quantstudio Dx System), in accordance with an embodiment. An RT-qPCR test that quantifies the absolute and/or relative expression levels of genes that enable patient subtyping may be run using a testing system such as the one depicted in FIG. 15. This test can then be used in precision trials and in precision care as discussed above.

In some embodiments, the subtyping test can be differently configured. For example, the subtyping test need not employ the manual RNA extraction and assay preparation step shown in FIG. 15. In such embodiments, the sample can be directly added to a system for performing RT-qPCR and the extraction and PCR analysis can be performed all in one.

VIII. Example Computer

FIG. 16 illustrates an example computer 1600 for implementing the methods described herein, in accordance with an embodiment. The computer 1600 includes at least one processor 1601 coupled to a chipset 1602. The chipset 1602 includes a memory controller hub 1610 and an input/output (I/O) controller hub 1611. A memory 1603 and a graphics adapter 1606 are coupled to the memory controller hub 1610, and a display 1609 is coupled to the graphics adapter 1606. A storage device 1604, an input device 1607, and network adapter 1608 are coupled to the I/O controller hub 1611. Other embodiments of the computer 1600 have different architectures.

The storage device 1604 is a non-transitory computer-readable storage medium such as a hard drive, compact disk read-only memory (CD-ROM), DVD, or a solid-state memory device. The memory 1603 holds instructions and data used by the processor 1601. The input interface 1607 is a touch-screen interface, a mouse, track ball, or other type of pointing device, a keyboard, or some combination thereof, and is used to input data into the computer 1600. In some embodiments, the computer 1600 can be configured to receive input (e.g., commands) from the input interface 1607 via gestures from the user. The graphics adapter 1606 displays images and other information on the display 1609. The network adapter 1608 couples the computer 1600 to one or more computer networks.

The computer 1600 is adapted to execute computer program modules for providing functionality described herein. As used herein, the term “module” refers to computer program logic used to provide the specified functionality. Thus, a module can be implemented in hardware, firmware, and/or software. In one embodiment, program modules are stored on the storage device 1604, loaded into the memory 1603, and executed by the processor 1601.

The types of computers 1600 used to implement the methods described herein can vary depending upon the embodiment and the processing power required by the entity. For example, the diagnostic/treatment system can run in a single computer 1600 or multiple computers 1600 communicating with each other through a network such as in a server farm. The computers 1600 can lack some of the components described above, such as graphics adapters 1606, and displays 1609.

IX. Example Kit Implementation

Also disclosed herein are kits for determining a therapy recommendation for an individual. Such kits can include reagents for detecting expression levels of one or biomarkers and instructions for classifying based on the detected expression levels and selecting a therapy recommendation based on the classification.

The detection reagents can be provided as part of a kit. Thus, the invention further provides kits for detecting the presence of a panel of biomarkers of interest in a biological test sample. A kit can comprise a set of reagents for generating a dataset via at least one protein detection assay (e.g., immunoassay or RT-PCR assay) that analyzes the test sample from the subject. In various embodiments, the set of reagents enable detection of quantitative expression levels of biomarkers described in any of Tables 1, 2A-2B, 3, and 4A-4D. In certain aspects, the reagents include one or more antibodies that bind to one or more of the markers. The antibodies may be monoclonal antibodies or polyclonal antibodies. In some aspects, the reagents can include reagents for performing ELISA including buffers and detection agents. In some aspects, the reagents include primers that are designed to hybridize with nucleic acids transcribed from genes identified in any of Tables 1, 2A-2B, 3, and 4A-4D.

A kit can include instructions for use of a set of reagents. For example, a kit can include instructions for performing at least one biomarker detection assay such as an immunoassay, a protein-binding assay, an antibody-based assay, an antigen-binding protein-based assay, a protein-based array, an enzyme-linked immunosorbent assay (ELISA), flow cytometry, a protein array, a blot, a Western blot, nephelometry, turbidimetry, chromatography, mass spectrometry, enzymatic activity, proximity extension assay, and an immunoassay selected from RIA, immunofluorescence, immunochemiluminescence, immunoelectrochemiluminescence, immunoelectrophoretic, a competitive immunoassay, and immunoprecipitation.

In various embodiments, a kit can include instructions for performing at least one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction.

In various embodiments, the kit includes instructions for determining quantitative expression data for three biomarkers, the instructions including: contacting the sample with a reagent; generating a plurality of complexes between the reagent and the plurality of biomarkers in the sample; and detecting the plurality of complexes to obtain a dataset associated with the sample, wherein the dataset comprises the quantitative expression data for the biomarker.

In various embodiments, the kits include instructions for practicing the methods disclosed herein (e.g., methods for training and/or implementing a patient subtype classifier). These instructions can be present in the subject kits in a variety of forms, one or more of which can be present in the kit. One form in which these instructions can be present is as printed information on a suitable medium or substrate, e.g., a piece or pieces of paper on which the information is printed, in the packaging of the kit, in a package insert, etc. Yet another means would be a computer readable medium, e.g., diskette, CD, hard-drive, network data storage, etc., on which the information has been recorded. Yet another means that can be present is a website address which can be used via the internet to access the information at a removed site. Any convenient means can be present in the kits.

X. Additional Considerations

All references, issued patents and patent applications cited within the body of the specification are hereby incorporated by reference in their entirety, for all purposes.

The foregoing description of the embodiments of the disclosure has been presented for the purpose of illustration; it is not intended to be exhaustive or to limit the disclosure to the precise forms disclosed. Persons skilled in the relevant art can appreciate that many modifications and variations are possible in light of the above disclosure.

Some portions of this description describe the embodiments of the disclosure in terms of algorithms and symbolic representations of operations on information. These algorithmic descriptions and representations are commonly used by those skilled in the data processing arts to convey the substance of their work effectively to others skilled in the art. These operations, while described functionally, computationally, or logically, are understood to be implemented by computer programs or equivalent electrical circuits, microcode, or the like.

Any of the steps, operations, or processes described herein can be performed or implemented with one or more hardware or software modules, alone or in combination with other devices. In one embodiment, a software module is implemented with a computer program product including a computer-readable non-transitory medium containing computer program code, which can be executed by a computer processor for performing any or all of the steps, operations, or processes described.

Embodiments may also relate to an apparatus for performing the operations herein. This apparatus may be specially constructed for the required purposes, and/or it may comprise a general-purpose computing device selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a non-transitory, tangible computer readable storage medium, or any type of media suitable for storing electronic instructions, which may be coupled to a computer system bus. Furthermore, any computing systems referred to in the specification may include a single processor or may be architectures employing multiple processor designs for increased computing capability.

Embodiments of the disclosure may also relate to a product that is produced by a computing process described herein. Such a product may include information resulting from a computing process, where the information is stored on a non-transitory, tangible computer-readable storage medium and may include any embodiment of a computer program product or other data combination described herein.

Finally, the language used in the specification has been principally selected for readability and instructional purposes, and it may not have been selected to delineate or circumscribe the inventive subject matter. It is therefore intended that the scope of the disclosure be limited not by this detailed description, but rather by any claims that issue on an application based hereon. Accordingly, the disclosure of the embodiments of the disclosure is intended to be illustrative, but not limiting, of the scope of the disclosure.

XI. Additional Embodiments

Disclosed herein is a method comprising: obtaining a sample from a subject exhibiting dysregulated host response, wherein the sample comprises a plurality of biomarkers; determining quantitative expression data for at least one biomarker selected from the group consisting of the biomarkers listed in Row 1 of one of Tables 1, 2A, 2B, and 3, at least one biomarker selected from the group consisting of the biomarkers listed in Row 2 of the one of Tables 1, 2A, 2B, and 3, and at least one biomarker selected from the group consisting of the biomarkers listed in Row 3 of the one of Tables 1, 2A, 2B, and 3; and determining a classification of the subject based on the quantitative expression data using a patient subtype classifier, the classification of the subject comprising one of subtype A, subtype B, or subtype C.

Additionally disclosed herein is a method comprising: obtaining a classification of a subject exhibiting dysregulated host response, the classification of the subject comprising one of subtype A, subtype B, or subtype C; and identifying a therapy recommendation for the subject based at least in part on the classification, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject comprises at least no immunosuppressive therapy, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy, and wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, modulators of coagulation therapy, and modulators of vascular permeability therapy.

Additionally disclosed herein is a computer-implemented method comprising: obtaining quantitative expression data from a sample from a subject exhibiting dysregulated host response for at least one biomarker selected from the group consisting of the biomarkers listed in Row 1 of one of Tables 1, 2A, 2B, and 3, at least one biomarker selected from the group consisting of the biomarkers listed in Row 2 of the one of Tables 1, 2A, 2B, and 3, and at least one biomarker selected from the group consisting of the biomarkers listed in Row 3 of the one of Tables 1, 2A, 2B, and 3; and determining, by a computer processor, a classification of the subject based on the quantitative expression data using a patient subtype classifier, the classification of the subject comprising one of subtype A, subtype B, or subtype C.

Additionally disclosed herein is a computer-implemented method comprising: obtaining, by a computer processor, a classification of a subject exhibiting dysregulated host response, the classification of the subject comprising one of subtype A, subtype B, or subtype C; and identifying a therapy recommendation for the subject based at least in part on the classification, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject comprises at least no immunosuppressive therapy, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy, and wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, modulators of coagulation therapy, and modulators of vascular permeability therapy.

Additionally disclosed herein is a non-transitory computer-readable storage medium storing computer program instructions that when executed by a computer processor, cause the computer processor to: store quantitative expression data from a sample from a subject exhibiting dysregulated host response, the quantitative expression data for at least one biomarker selected from the group consisting of the biomarkers listed in Row 1 of one of Tables 1, 2A, 2B, and 3, at least one biomarker selected from the group consisting of the biomarkers listed in Row 2 of the one of Tables 1, 2A, 2B, and 3, and at least one biomarker selected from the group consisting of the biomarkers listed in Row 3 of the one of Tables 1, 2A, 2B, and 3; and determine a classification of the subject based on the quantitative expression data using a patient subtype classifier, the classification of the subject comprising one of subtype A, subtype B, or subtype C.

Additionally disclosed herein is a non-transitory computer-readable storage medium storing computer program instructions that when executed by a computer processor, cause the computer processor to: obtain a classification of a subject exhibiting dysregulated host response, the classification of the subject comprising one of subtype A, subtype B, or subtype C; and identify a therapy recommendation for the subject based at least in part on the classification, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject comprises at least no immunosuppressive therapy, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy, and wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, modulators of coagulation therapy, and modulators of vascular permeability therapy.

Additionally disclosed herein is a system comprising: a storage memory for storing quantitative expression data from a sample from a subject exhibiting dysregulated host response, the quantitative expression data for at least one biomarker selected from the group consisting of the biomarkers listed in Row 1 of one of Tables 1, 2A, 2B, and 3, at least one biomarker selected from the group consisting of the biomarkers listed in Row 2 of the one of Tables 1, 2A, 2B, and 3, and at least one biomarker selected from the group consisting of the biomarkers listed in Row 3 of the one of Tables 1, 2A, 2B, and 3; and a processor communicatively coupled to the storage memory for determining a classification of the subject based on the quantitative expression data using a patient subtype classifier, the classification of the subject comprising one of subtype A, subtype B, or subtype C.

Additionally disclosed herein is a system comprising: a processor for: obtaining a classification of a subject exhibiting dysregulated host response, the classification of the subject comprising one of subtype A, subtype B, or subtype C; and identifying a therapy recommendation for the subject based at least in part on the classification, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject comprises at least no immunosuppressive therapy, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy, and wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, modulators of coagulation therapy, and modulators of vascular permeability therapy.

Additionally disclosed herein is a kit comprising: a plurality of reagents for determining, from a sample obtained from a subject exhibiting dysregulated host response, quantitative expression data for at least one biomarker selected from the group consisting of the biomarkers listed in Row 1 of one of Tables 1, 2A, 2B, and 3, at least one biomarker selected from the group consisting of the biomarkers listed in Row 2 of the one of Tables 1, 2A, 2B, and 3, and at least one biomarker selected from the group consisting of the biomarkers listed in Row 3 of the one of Tables 1, 2A, 2B, and 3; and instructions for using the plurality of reagents to determine the quantitative expression data from the sample from the subject.

Additionally disclosed herein is a composition comprising at least three primer sets for amplifying at least three biomarkers, wherein each primer set of the at least three primer sets comprises a pair of single-stranded DNA primers for amplifying one of the at least three biomarkers, and wherein at least one of the at least three biomarkers is selected from the group consisting of the biomarkers listed in Row 1 of one of Tables 1, 2A, 2B, and 3, at least one biomarker of the at least three biomarkers is selected from the group consisting of the biomarkers listed in Row 2 of the one of Tables 1, 2A, 2B, and 3, and at least one biomarker of the at least three biomarkers is selected from the group consisting of the biomarkers listed in Row 3 of the one of Tables 1, 2A, 2B, and 3.

In various embodiments, the one of Tables 1, 2A, 2B, and 3 comprises Table 1, wherein the at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 7 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 8, a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 9 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 10, a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 11 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 12, and a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 13 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 14, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 15 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 16, a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 17 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 18, and a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 19 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 20, and wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 1 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 2; a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 3 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 4, and a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 5 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 6.

In various embodiments, the one of Tables 1, 2A, 2B, and 3 comprises Table 1, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 7 and a reverse primer comprising SEQ ID NO. 8, a forward primer comprising SEQ ID NO. 9 and a reverse primer comprising SEQ ID NO. 10, a forward primer comprising SEQ ID NO. 11 and a reverse primer comprising SEQ ID NO. 12, and a forward primer comprising SEQ ID NO. 13 and a reverse primer comprising SEQ ID NO. 14, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 15 and a reverse primer comprising SEQ ID NO. 16, a forward primer comprising SEQ ID NO. 17 and a reverse primer comprising SEQ ID NO. 18, and a forward primer comprising SEQ ID NO. 19 and a reverse primer comprising SEQ ID NO. 20, and wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 1 and a reverse primer comprising SEQ ID NO. 2; a forward primer comprising SEQ ID NO. 3 and a reverse primer comprising SEQ ID NO. 4, and a forward primer comprising SEQ ID NO. 5 and a reverse primer comprising SEQ ID NO. 6.

In various embodiments, the one of Tables 1, 2A, 2B, and 3 comprises Table 2B, wherein the at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 21 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 22, and a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 23 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 24, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 25 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 26, and a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 29 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 30, and wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 25 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 26, and a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 27 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 28.

In various embodiments, the one of Tables 1, 2A, 2B, and 3 comprises Table 2B, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 21 and a reverse primer comprising SEQ ID NO. 22, and a forward primer comprising SEQ ID NO. 23 and a reverse primer comprising SEQ ID NO. 24, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 25 and a reverse primer comprising SEQ ID NO. 26, and a forward primer comprising SEQ ID NO. 29 and a reverse primer comprising SEQ ID NO. 30, and wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 25 and a reverse primer comprising SEQ ID NO. 26, and a forward primer comprising SEQ ID NO. 27 and a reverse primer comprising SEQ ID NO. 28.

Additionally disclosed herein is a composition comprising at least three primer sets for amplifying at least three biomarkers, wherein each primer set of the at least three primer sets comprises a forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer for amplifying one of the at least three biomarkers, and wherein at least one of the at least three biomarkers is selected from the group consisting of the biomarkers listed in Row 1 of one of Tables 1, 2A, 2B, and 3, at least one biomarker of the at least three biomarkers is selected from the group consisting of the biomarkers listed in Row 2 of the one of Tables 1, 2A, 2B, and 3, and at least one biomarker of the at least three biomarkers is selected from the group consisting of the biomarkers listed in Row 3 of the one of Tables 1, 2A, 2B, and 3.

In various embodiments, at least one of the at least three primer sets is selected from the group consisting of: a forward outer primer configured to enable amplification of the at least one biomarker listed in Row 1 of Tables 1, 2A, 2B, and 3, a backward outer primer configured to enable amplification of the at least one biomarker listed in Row 1 of Tables 1, 2A, 2B, and 3, a forward inner primer configured to enable amplification of the at least one biomarker listed in Row 1 of Tables 1, 2A, 2B, and 3, a backward inner primer configured to enable amplification of the at least one biomarker listed in Row 1 of Tables 1, 2A, 2B, and 3, a forward loop primer configured to enable amplification of the at least one biomarker listed in Row 1 of Tables 1, 2A, 2B, and 3, and a backward loop primer configured to enable amplification of the at least one biomarker listed in Row 1 of Tables 1, 2A, 2B, and 3, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward outer primer configured to enable amplification of the at least one biomarker listed in Row 2 of Tables 1, 2A, 2B, and 3, a backward outer primer configured to enable amplification of the at least one biomarker listed in Row 2 of Tables 1, 2A, 2B, and 3, a forward inner primer configured to enable amplification of the at least one biomarker listed in Row 2 of Tables 1, 2A, 2B, and 3, a backward inner primer configured to enable amplification of the at least one biomarker listed in Row 2 of Tables 1, 2A, 2B, and 3, a forward loop primer configured to enable amplification of the at least one biomarker listed in Row 2 of Tables 1, 2A, 2B, and 3, and a backward loop primer configured to enable amplification of the at least one biomarker listed in Row 2 of Tables 1, 2A, 2B, and 3, and wherein at least one of the at least three primer sets is selected from the group consisting of: a forward outer primer configured to enable amplification of the at least one biomarker listed in Row 3 of Tables 1, 2A, 2B, and 3, a backward outer primer configured to enable amplification of the at least one biomarker listed in Row 3 of Tables 1, 2A, 2B, and 3, a forward inner primer configured to enable amplification of the at least one biomarker listed in Row 3 of Tables 1, 2A, 2B, and 3, a backward inner primer configured to enable amplification of the at least one biomarker listed in Row 3 of Tables 1, 2A, 2B, and 3, a forward loop primer configured to enable amplification of the at least one biomarker listed in Row 3 of Tables 1, 2A, 2B, and 3, and a backward loop primer configured to enable amplification of the at least one biomarker listed in Row 3 of Tables 1, 2A, 2B, and 3.

In various embodiments, the dysregulated host response comprises one of sepsis and dysregulated host response not caused by infection. In various embodiments, methods described above further comprise administering or having administered therapy to the subject based on the therapy recommendation. In various embodiments, responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject further comprises at least no corticosteroid therapy. In various embodiments, responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor, a blocker of complement components, a blocker of complement component receptors, and a blocker of a pro-inflammatory cytokine. In various embodiments, responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor and an anticoagulant. In various embodiments, responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject further comprises no hydrocortisone.

In various embodiments, responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject further comprises at least one of anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, IL-7, anti-C5a, anti-C3a, anti-C5aR, anti-C3aR, anti-TNF-alpha, and anti-IL-6. In various embodiments, responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject further comprises at least one of anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, IL-7, activated protein C, antithrombin, and thrombomodulin. In various embodiments, the classification is pre-determined. In various embodiments, the method further comprises determining the classification, and wherein determining the classification comprises: obtaining a sample from the subject exhibiting dysregulated host response, wherein the sample comprises a plurality of biomarkers; determining quantitative expression data for at least three biomarkers; and determining the classification of the subject based on the quantitative expression data using a patient subtype classifier.

In various embodiments, the at least three biomarkers comprise at least one biomarker selected from the group consisting of the biomarkers listed in Row 1 of one of Tables 1, 2A, 2B, and 3, at least one biomarker selected from the group consisting of the biomarkers listed in Row 2 of the one of Tables 1, 2A, 2B, and 3, and at least one biomarker selected from the group consisting of the biomarkers listed in Row 3 of the one of Tables 1, 2A, 2B, and 3. In various embodiments, the obtained sample comprises a blood sample from the subject. In various embodiments, the method further comprises determining that the subject exhibiting dysregulated host response does not exhibit shock, and wherein the one of Tables 1, 2A, 2B, and 3 comprises one of Tables 1, 2B, and 4. In various embodiments, the method further comprises determining that the subject exhibiting dysregulated host response is further exhibiting shock, and wherein the one of Tables 1, 2A, 2B, and 3 comprises one of Tables 1, 2A, and 3. In various embodiments, the method further comprises determining that the subject exhibiting dysregulated host response is an adult subject, and wherein the one of Tables 1, 2A, 2B, and 3 comprises one of Tables 1, 2A, and 2B. In various embodiments, the method further comprises determining that the subject exhibiting dysregulated host response is a pediatric subject, and wherein the one of Tables 1, 2A, 2B, and 3 comprises one of Tables 1 and 3.

In various embodiments, the quantitative expression data for at least one of the at least three biomarkers is determined by one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction.

In various embodiments, determining a classification of the subject based on the quantitative expression data using a patient subtype classifier comprises: determining, by the patient subtype classifier, for each candidate classification of the subject, a classification-specific score for the subject by: determining a first geometric mean of the quantitative expression data for the subject for one or more biomarkers of the candidate classification, wherein the quantitative expression data for the subject for the one or more biomarkers of the candidate classification are increased relative to the quantitative expression data for the one or more biomarkers for one or more control subjects; determining a second geometric mean of the quantitative expression for the subject for one or more additional biomarkers of the candidate classification, wherein the quantitative expression data for the subject for the one or more additional biomarkers of the candidate classification are decreased relative to the quantitative expression data for the one or more additional biomarkers for the one or more control subjects; and determining a difference between the first geometric mean and the second geometric mean, the first and second geometric means optionally subject to scaling, and the difference comprising the classification-specific score for the subject; and determining, by the patient subtype classifier, using a multi-class regression model, based on the classification-specific score for each candidate classification of the subject, the classification of the subject, wherein the candidate classifications of the subject comprise subtype A, subtype B, and subtype C.

In various embodiments, the method further comprises prior to determining a classification of the subject based on the quantitative expression data using a patient subtype classifier, normalizing the quantitative expression data based on quantitative expression data for one or more housekeeping genes.

In various embodiments, the patient subtype classifier is a machine-learned model. In various embodiments, the one of Tables 1, 2A, 2B, and 3 comprises Table 1, and wherein the patient subtype classifier has an average accuracy of at least 82.93%. In various embodiments, the one of Tables 1, 2A, 2B, and 3 comprises Table 2A, and wherein the patient subtype classifier has an average accuracy of at least 89.6%. In various embodiments, the one of Tables 1, 2A, 2B, and 3 comprises Table 2B, and wherein the patient subtype classifier has an average accuracy of at least 86.3%. In various embodiments, the one of Tables 1, 2A, 2B, and 3 comprises Table 3, and wherein the patient subtype classifier has an average accuracy of at least 98.3%.

In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation, identifying that the therapy recommendation for the subject comprises at least no corticosteroid therapy comprises determining that a statistical significance of an average day 28 reduction in mortality of subjects exhibiting dysregulated host response, determined based on the one of Tables 1, 2A, 2B, and 3 to be of the determined classification of the subject, and not provided corticosteroid therapy, is greater than or equal to a threshold statistical significance, identifying that the therapy recommendation for the subject comprises corticosteroid therapy comprises determining that a statistical significance of an average day 28 reduction in mortality of subjects exhibiting dysregulated host response, determined based on the one of Tables 1, 2A, 2B, and 3 to be of the determined classification of the subject, and provided corticosteroid therapy, is greater than or equal to a threshold statistical significance, and identifying that the therapy recommendation for the subject comprises no therapy recommendation comprises: determining that a statistical significance of an average day 28 reduction in mortality of subjects exhibiting dysregulated host response, determined based on the one of Tables 1, 2A, 2B, and 3 to be of the determined classification of the subject, and not provided corticosteroid therapy, is less than a threshold statistical significance; and determining that a statistical significance of an average day 28 reduction in mortality of subjects exhibiting dysregulated host response, determined based on the one of Tables 1, 2A, 2B, and 3 to be of the determined classification of the subject, and provided corticosteroid therapy, is less than a threshold statistical significance.

In various embodiments, a statistical significance comprises a p-value, and wherein the threshold statistical significance comprises at least 0.1. In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation, wherein the dysregulated host response comprises sepsis, wherein the one of Tables 1, 2A, 2B, and 3 comprises Table 1, wherein identifying that the therapy recommendation for the subject comprises no corticosteroid therapy comprises determining that the classification of the subject comprises subtype A or subtype C, and wherein identifying that the therapy recommendation for the subject comprises corticosteroid therapy comprises determining that the classification of the subject comprises subtype B.

In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation, wherein the dysregulated host response comprises dysregulated host response not caused by infection, wherein the one of Tables 1, 2A, 2B, and 3 comprises Table 1 or Table 3, wherein identifying that the therapy recommendation for the subject comprises no corticosteroid therapy comprises determining that the classification of the subject comprises subtype A or subtype C, and wherein identifying that the therapy recommendation for the subject comprises no therapy recommendation comprises determining that the classification of the subject comprises subtype B.

In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation, wherein the dysregulated host response comprises sepsis, wherein the one of Tables 1, 2A, 2B, and 3 comprises Table 2A, Table 2B or Table 3, wherein identifying that the therapy recommendation for the subject comprises no corticosteroid therapy comprises determining that the classification of the subject comprises subtype A, and wherein identifying that the therapy recommendation for the subject comprises no therapy recommendation comprises determining that the classification of the subject comprises subtype B or subtype C.

In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation, wherein the dysregulated host response comprises dysregulated host response not caused by infection, wherein the one of Tables 1, 2A, 2B, and 3 comprises Table 2A, wherein identifying that the therapy recommendation for the subject comprises no corticosteroid therapy comprises determining that the classification of the subject comprises subtype C, and wherein identifying that the therapy recommendation for the subject comprises no therapy recommendation comprises determining that the classification of the subject comprises subtype A or subtype B.

In various embodiments, the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation, wherein the dysregulated host response comprises dysregulated host response not caused by infection, wherein the one of Tables 1, 2A, 2B, and 3 comprises Table 2B, wherein identifying that the therapy recommendation for the subject comprises no corticosteroid therapy comprises determining that the classification of the subject comprises subtype A or subtype C, and wherein identifying that the therapy recommendation for the subject comprises corticosteroid therapy comprises determining that the classification of the subject comprises subtype B.

In various embodiments, an average day 28 reduction in mortality of subjects exhibiting dysregulated host response that comprises sepsis, determined to be of subtype A based on Table 3, and not provided corticosteroid therapy, is between 5.0%-77.6%, compared to subjects exhibiting dysregulated host response that comprises sepsis, determined to be of subtype A based on Table 3, and provided corticosteroid therapy. In various embodiments, an average day 28 reduction in mortality of subjects exhibiting dysregulated host response that comprises sepsis, determined to be of subtype B based on Table 1, and provided corticosteroid therapy, is between 5.0%-35.2%, compared to subjects exhibiting dysregulated host response that comprises sepsis, determined to be of subtype B based on Table 1, and not provided corticosteroid therapy. In various embodiments, an average day 28 reduction in mortality of subjects exhibiting dysregulated host response that comprises dysregulated host response not caused by infection, determined to be of subtype A or subtype C based on Table 1, and not provided corticosteroid therapy, is between 5.0%-100.0%, compared to subjects exhibiting dysregulated host response that comprises dysregulated host response not caused by infection, determined to be of subtype A or subtype C based on Table 1, and provided corticosteroid therapy. In various embodiments, an average day 28 reduction in mortality of subjects exhibiting dysregulated host response that comprises dysregulated host response not caused by infection, determined to be of subtype A or subtype C based on Table 2A, and not provided corticosteroid therapy, is between 5.0%-56.7%, compared to subjects exhibiting dysregulated host response that comprises dysregulated host response not caused by infection, determined to be of subtype A or subtype C based on Table 2A, and provided corticosteroid therapy.

In various embodiments, an average day 28 reduction in mortality of subjects exhibiting dysregulated host response that comprises dysregulated host response not caused by infection, determined to be of subtype A or subtype C based on Table 2B, and not provided corticosteroid therapy, is between 5.0%-100.0%, compared to subjects exhibiting dysregulated host response that comprises dysregulated host response not caused by infection, determined to be of subtype A or subtype C based on Table 2B, and provided corticosteroid therapy. In various embodiments, an average day 28 reduction in mortality of subjects exhibiting dysregulated host response that comprises dysregulated host response not caused by infection, determined to be of subtype A or C based on Table 3, and not provided corticosteroid therapy, is between 5.0%-100.0%, compared to subjects exhibiting dysregulated host response that comprises dysregulated host response not caused by infection, determined to be of subtype A or C based on Table 3, and provided corticosteroid therapy.

XII. References

1. Rather A R, Kasana B. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) [Internet]. JMS SKIMS. 2015. pp. 162-164. doi:10.33883/jms.v18i2.269

2. Bellani G, Laffey J G, Pham T, Fan E, Brochard L, Esteban A, et al. Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA. 2016; 315: 788-800.

3. Liu V, Escobar G J, Greene J D, Soule J, Whippy A, Angus D C, et al. Hospital deaths in patients with sepsis from 2 independent cohorts. JAMA. 2014; 312: 90-92.

4. Marshall J C. Why have clinical trials in sepsis failed? Trends Mol Med. 2014; 20: 195-203.

5. Vincent J-L, Mongkolpun W. Non-antibiotic therapies for sepsis: an update. Expert Rev Anti Infect Ther. 2019; 17: 169-175.

6. Young D, Lamb S E, Shah S, MacKenzie I, Tunnicliffe W, Lall R, et al. High-Frequency Oscillation for Acute Respiratory Distress Syndrome [Internet]. New England Journal of Medicine. 2013. pp. 806-813. doi:10.1056/nejmoa1215716

7. Ferguson N D, Cook D J, Guyatt G H, Mehta S, Hand L, Austin P, et al. High-Frequency Oscillation in Early Acute Respiratory Distress Syndrome [Internet]. New England Journal of Medicine. 2013. pp. 795-805. doi:10.1056/nejmoa1215554

8. Meade M O, Young D, Hanna S, Zhou Q, Bachman T E, Bollen C, et al. Severity of Hypoxemia and Effect of High-Frequency Oscillatory Ventilation in Acute Respiratory Distress Syndrome [Internet]. American Journal of Respiratory and Critical Care Medicine. 2017. pp. 727-733. doi:10.1164/rccm.201609-1938oc

9. Meade M O, Cook D J, Guyatt G H, Slutsky A S, Arabi Y M, James Cooper D, et al. Ventilation Strategy Using Low Tidal Volumes, Recruitment Maneuvers, and High Positive End-Expiratory Pressure for Acute Lung Injury and Acute Respiratory Distress Syndrome [Internet]. JAMA. 2008. p. 637. doi:10.1001/jama.299.6.637

10. Hodgson C, Goligher E C, Young M E, Keating J L, Holland A E, Romero L, et al. Recruitment manoeuvres for adults with acute respiratory distress syndrome receiving mechanical ventilation. Cochrane Database Syst Rev. 2016; 11: CD006667.

11. Walkey A J, Del Sorbo L, Hodgson C L, Adhikari N K J, Wunsch H, Meade M O, et al. Higher PEEP versus Lower PEEP Strategies for Patients with Acute Respiratory Distress Syndrome. A Systematic Review and Meta-Analysis [Internet]. Annals of the American Thoracic Society. 2017. pp. S297-S303. doi:10.1513/annalsats.201704-338ot

12. Investigators WGFTARFARDST (art), Writing Group for the Alveolar Recruitment for Acute Respiratory Distress Syndrome Trial (ART) Investigators, Cavalcanti A B, Suzumura É A, Laranjeira L N, de Moraes Paisani D, et al. Effect of Lung Recruitment and Titrated Positive End-Expiratory Pressure (PEEP) vs Low PEEP on Mortality in Patients With Acute Respiratory Distress Syndrome [Internet]. JAMA. 2017. p. 1335. doi:10.1001/jama.2017.14171

13. Antcliffe D B, Burnham K L, Al-Beidh F, Santhakumaran S, Brett S J, Hinds C J, et al. Transcriptomic Signatures in Sepsis and a Differential Response to Steroids. From the VANISH Randomized Trial [Internet]. American Journal of Respiratory and Critical Care Medicine. 2019. pp. 980-986. doi:10.1164/rccm.201807-1419oc

14. Calfee C S, Delucchi K L, Sinha P, Matthay M A, Hackett J, Shankar-Hari M, et al. Acute respiratory distress syndrome subphenotypes and differential response to simvastatin: secondary analysis of a randomised controlled trial. Lancet Respir Med. 2018; 6: 691-698.

15. Rhee C, Dantes R, Epstein L, Murphy D J, Seymour C W, Iwashyna T J, et al. Incidence and Trends of Sepsis in U S Hospitals Using Clinical vs Claims Data, 2009-2014. JAMA. 2017; 318: 1241-1249.

16. Paoli C J, Reynolds M A, Sinha M, Gitlin M, Crouser E. Epidemiology and Costs of Sepsis in the United States—An Analysis Based on Timing of Diagnosis and Severity Level. Crit Care Med. 2018; 46: 1889-1897.

17. Rhee C, Jones T M, Hamad Y, Pande A, Varon J, O'Brien C, et al. Prevalence, Underlying Causes, and Preventability of Sepsis-Associated Mortality in US Acute Care Hospitals. JAMA Netw Open. 2019; 2: e187571.

18. Levy M M, Evans L E, Rhodes A. The Surviving Sepsis Campaign Bundle: 2018 update [Internet]. Intensive Care Medicine. 2018. pp. 925-928. doi:10.1007/s00134-018-5085-0

19. Jones S L, Ashton C M, Kiehne L B, Nicolas J C, Rose A L, Shirkey B A, et al. Outcomes and Resource Use of Sepsis-associated Stays by Presence on Admission, Severity, and Hospital Type. Med Care. 2016; 54: 303-310.

20. Lamontagne F, Rochwerg B, Lytvyn L, Guyatt G H, Møller M H, Annane D, et al. Corticosteroid therapy for sepsis: a clinical practice guideline. BMJ. 2018; 362: k3284.

21. Hahn E O, Houser H B, Rammelkamp C H Jr, Denny F W, Wannamaker L W. Effect of cortisone on acute streptococcal infections and poststreptococcal complications. J Clin Invest. 1951; 30: 274-281.

22. Schumer W. Steroids in the Treatment of Clinical Septic Shock [Internet]. Annals of Surgery. 1976. pp. 333-341. doi:10.1097/00000658-197609000-00011

23. Sprung C L, Caralis P V, Marcial E H, Pierce M, Celbard M A, Long W N, et al. The Effects of High-Dose Corticosteroids in Patients with Septic Shock, a Prospective, Controlled Study [Internet]. Survey of Anesthesiology. 1985. p. 238. doi:10.1097/00132586-198508000-00028

24. Bone R C, Fisher C J, Clemmer T P, Slotman G J, Metz C A, Balk R A, et al. A Controlled Clinical Trial of High-Dose Methylprednisolone in the Treatment of Severe Sepsis and Septic Shock [Internet]. New England Journal of Medicine. 1987. pp. 653-658. doi:10.1056/nejm198709103171101

25. Bollaert P E, Charpentier C, Levy B, Debouverie M, Audibert G, Larcan A. Reversal of late septic shock with supraphysiologic doses of hydrocortisone. Crit Care Med. 1998; 26: 645-650.

26. Annane, Annane, Bellissant, Sebille, Lesieur, Mathieu, et al. Impaired pressor sensitivity to noradrenaline in septic shock patients with and without impaired adrenal function reserve [Internet]. British Journal of Clinical Pharmacology. 2002. pp. 589-597. doi:10.1046/j.1365-2125.1998.00833.x

27. Rothwell P. Cortisol response to corticotropin and survival in septic shock [Internet]. The Lancet. 1991. pp. 582-583. doi:10.1016/0140-6736(91)91641-7

28. Moran J L, Chapman M J, O'Fathartaigh M S, Peisach A R, Pannall P R, Leppard P. Hypocortisolaemia and adrenocortical responsiveness at onset of septic shock [Internet]. Intensive Care Medicine. 1994. pp. 489-495. doi:10.1007/bf01711901

29. Annane D. A 3-Level Prognostic Classification in Septic Shock Based on Cortisol Levels and Cortisol Response to Corticotropin [Internet]. JAMA. 2000. p. 1038. doi:10.1001/jama.283.8.1038

30. Annane D. Effect of Treatment With Low Doses of Hydrocortisone and Fludrocortisone on Mortality in Patients With Septic Shock [Internet]. JAMA. 2002. p. 862. doi:10.1001/jama.288.7.862

31. Gordon A C, Mason A J, Thirunavukkarasu N, Perkins G D, Cecconi M, Cepkova M, et al. Effect of Early Vasopressin vs Norepinephrine on Kidney Failure in Patients With Septic Shock: The VANISH Randomized Clinical Trial. JAMA. 2016; 316: 509-518.

32. Dellinger R P, The Surviving Sepsis Campaign Guidelines Committee including The Pediatric Subgroup*, Levy M M, Rhodes A, Annane D, Gerlach H, et al. Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock, 2012 [Internet]. Intensive Care Medicine. 2013. pp. 165-228. doi:10.1007/s00134-012-2769-8

33. Rhodes A, Evans L E, Alhazzani W, Levy M M, Antonelli M, Ferrer R, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016. Intensive Care Med. 2017; 43: 304-377.

34. Venkatesh B, Finfer S, Cohen J, Rajbhandari D, Arabi Y, Bellomo R, et al. Adjunctive Glucocorticoid Therapy in Patients with Septic Shock. N Engl J Med. 2018; 378: 797-808.

35. Annane D, Renault A, Brun-Buisson C, Megarbane B, Quenot J-P, Siami S, et al. Hydrocortisone plus Fludrocortisone for Adults with Septic Shock [Internet]. New England Journal of Medicine. 2018. pp. 809-818. doi:10.1056/nejmoa1705716

36. Identification numbers for sepsis-related clinical trials currently registered on clinicaltrials.gov as of Jul. 1, 2019: NCT03509350, NCT03401398, NCT02473263, NCT03872011, NCT03592693, NCT02602210, NCT03710187, NCT03333278, NCT03380507, NCT03389555, NCT03422159, NCT03649633, NCT03913468, NCT03649009, NCT01817153, NCT03540628, NCT03821714, NCT03258619, NCT03336814, NCT03748537. In: ClinicalTrials.gov [Internet]. [cited Jul. 1, 2019]. Available: www.clinicaltrials.gov

37. Stress Hydrocortisone In Pediatric Septic Shock—Full Text View—ClinicalTrials.gov [Internet]. [cited 16 Jul. 2019]. Available: https://clinicaltrials.gov/ct2/show/NCT03401398

38. Wong H R, Atkinson S J, Cvijanovich N Z, Anas N, Allen G L, Thomas N J, et al. Combining Prognostic and Predictive Enrichment Strategies to Identify Children With Septic Shock Responsive to Corticosteroids. Crit Care Med. 2016; 44: e1000-3.

39. Hager D N, Hooper M H, Bernard G R, Busse L W, Ely E W, Fowler A A, et al. The Vitamin C, Thiamine and Steroids in Sepsis (VICTAS) Protocol: a prospective, multi-center, double-blind, adaptive sample size, randomized, placebo-controlled, clinical trial. Trials. 2019; 20: 197.

40. Marshall J C, Vincent J-L, Fink M P, Cook D J, Rubenfeld G, Foster D, et al. Measures, markers, and mediators: Toward a staging system for clinical sepsis. A Report of the Fifth Toronto Sepsis Roundtable, Toronto, Ontario, Canada, Oct. 25-26, 2000 [Internet]. Critical Care Medicine. 2003. pp. 1560-1567. doi:10.1097/01.ccm.0000065186.67848.3a

41. Pachot A, Lepape A, Vey S, Bienvenu J, Mougin B, Monneret G. Systemic transcriptional analysis in survivor and non-survivor septic shock patients: a preliminary study. Immunol Lett. 2006; 106: 63-71.

42. Wong H R, Cvijanovich N, Lin R, Allen G L, Thomas N J, Willson D F, et al. Identification of pediatric septic shock subclasses based on genome-wide expression profiling. BMC Med. 2009; 7: 34.

43. Maslove D M, Tang B M, McLean A S. Identification of sepsis subtypes in critically ill adults using gene expression profiling. Crit Care. 2012; 16: R183.

44. Davenport E E, Burnham K L, Radhakrishnan J, Humburg P, Hutton P, Mills T C, et al. Genomic landscape of the individual host response and outcomes in sepsis: a prospective cohort study. Lancet Respir Med. 2016; 4: 259-271.

45. Scicluna B P, van Vught L A, Zwinderman A H, Wiewel M A, Davenport E E, Burnham K L, et al. Classification of patients with sepsis according to blood genomic endotype: a prospective cohort study. Lancet Respir Med. 2017; 5: 816-826.

46. Sweeney T E, Azad T D, Donato M, Haynes W A, Perumal T M, Henao R, et al. Unsupervised Analysis of Transcriptomics in Bacterial Sepsis Across Multiple Datasets Reveals Three Robust Clusters. Crit Care Med. 2018; 46: 915-925.

47. Sweeney T E, Wong H R, Khatri P. Robust classification of bacterial and viral infections via integrated host gene expression diagnostics. Sci Transl Med. 2016; 8: 346ra91.

48. Tusher V G, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA. 2001; 98: 5116-5121.

49. Sweeney T E, Chen A C, Gevaert O. Combined Mapping of Multiple clUsteriNg ALgorithms (COMMUNAL): A Robust Method for Selection of Cluster Number, K. Sci Rep. 2015; 5: 16971.

50. Plassais, Jonathan, et al. “Transcriptome modulation by hydrocortisone in severe burn shock: ancillary analysis of a prospective randomized trial.” Critical Care, vol. 21, no. 158, 16 Jun. 2017, doi:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5473974/.

51. Timothy, Sweeney E., et al. “A comprehensive time-course-based multicohort analysis of sepsis and sterile inflammation reveals a robust diagnostic gene set.” Science Translational Medicine, vol. 7, no. 287, 13 May 2015, doi:https://stm.sciencemag.org/content/7/287/287ra71.

52. Balk, Robert A. “Systemic inflammatory response syndrome (SIRS) Where did it come from and is it still relevant today?” Virulence, vol. 5, no. 1, 13 Nov. 2013, pp. 20-26, doi:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3916374/.

53. Benjamini, Y., and Hochberg, Y. (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B, 57, 289-300. http://www.jstor.org/stable/2346101.

Claims

1. A method for determining a patient subtype, the method comprising: obtaining or having obtained quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8,wherein biomarker 2 is one of SERPINB1 or GSPT1, andwherein biomarker 3 is one of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A,wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one of ZNF831, MME, CD3G, or STOM,wherein biomarker 5 is one of ECSIT, LAT, or NCOA4, andwherein biomarker 6 is one of SLC1A5, IGF2BP2, or ANXA3,wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one of C14orf159 or PUM2,wherein biomarker 8 is one of EPB42 or RPS6KA5, andwherein biomarker 9 is one of EPB42 or GBP2; andwherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one of MSH2, DCTD, or MMP8,wherein biomarker 11 is one of HK3, UCP2, or NUP88, andwherein biomarker 12 is one of GABARAPL2 or CASP4; andwherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G,wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1, andwherein biomarker 15 is one of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; anddetermining a classification of a subject based on the quantitative data using a patient subtype classifier.
2. The method of claim 1, wherein the at least one biomarker set is group 5, and wherein biomarker 13 is one of STOM, MME, BNT3A2, or HLA-DPA1.
3. The method of claim 1 or 2, wherein the at least one biomarker set is group 5, and wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1.
4. The method of any one of claims 1-3, wherein the at least one biomarker set is group 5, and wherein biomarker 15 is one of SLC1A5, IGF2BP2, or ANXA3.
5. A method for determining a therapy recommendation for a patient, the method comprising: obtaining or having obtained quantitative data for two or more biomarkers selected from the group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, TOMM70A, ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, ANXA3, C14orf159, PUM2, EPB42, RPS6KA5, GBP2, MSH2, DCTD, HK3, UCP2, NUP88, GABARAPL2, and CASP4; anddetermining a classification of a subject based on the quantitative data using a patient subtype classifier.
6. A method for determining a therapy recommendation for a patient, the method comprising: obtaining or having obtained quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises two or more biomarkers selected from a group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1 GSPT1, MPP1, HMBS, TALL C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A,wherein group 2 comprises two or more biomarkers selected from a group consisting of ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, and ANXA3,wherein group 3 comprises two or more biomarkers selected from a group consisting of C14orf159, PUM2, EPB42, RPS6KA5, EPB42, and GBP2; andwherein group 4 comprises two or more biomarkers selected from a group consisting of MSH2, DCTD, MMP8, HK3, UCP2, NUP88, GABARAPL2, and CASP4; andwherein group 5 comprises two or more biomarkers selected from a group consisting of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, CD3G, EPB42, GSPT1, LAT, HK3, SERPINB1, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, and TNFRSF1A; anddetermining a classification of a subject based on the quantitative data using a patient subtype classifier.
7. The method of any one of claims 1-6, further comprising identifying a therapy recommendation for the subject based at least in part on the classification.
8. A method for determining a therapy recommendation for a patient, the method comprising: obtaining a classification of a subject exhibiting a dysregulated host response, the classification having been determined by: obtaining or having obtained quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8,wherein biomarker 2 is one of SERPINB1 or GSPT1, andwherein biomarker 3 is one of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A,wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one of ZNF831, MME, CD3G, or STOM,wherein biomarker 5 is one of ECSIT, LAT, or NCOA4, andwherein biomarker 6 is one of SLC1A5, IGF2BP2, or ANXA3,wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one of C14orf159 or PUM2,wherein biomarker 8 is one of EPB42 or RPS6KA5, andwherein biomarker 9 is one of EPB42 or GBP2; andwherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one of MSH2, DCTD, or MMP8,wherein biomarker 11 is one of HK3, UCP2, or NUP88, andwherein biomarker 12 is one of GABARAPL2 or CASP4; andwherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G,wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1, andwherein biomarker 15 is one of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; anddetermining the classification based on the quantitative data using a patient subtype classifier; andidentifying a therapy recommendation for the subject based at least in part on the classification.
9. The method of claim 8, wherein the dysregulated host response of the subject comprises one of sepsis and dysregulated host response not caused by infection.
10. The method of any one of claims 1-9, wherein the classification of the subject comprises one of subtype A or subtype B.
11. The method of any one of claims 1-9, wherein the classification of the subject comprises one of subtype A, subtype B, or subtype C.
12. The method of claim 10 or 11, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject comprises at least no immunosuppressive therapy.
13. The method of claim 10 or 11, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject further comprises at least no corticosteroid therapy.
14. The method of claim 13, wherein the therapy recommendation identified for the subject further comprises at least one of no hydrocortisone.
15. The method of claim 10 or 11, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy.
16. The method of claim 10 or 11, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor, a blocker of complement components, a blocker of complement component receptors, and a blocker of a pro-inflammatory cytokine.
17. The method of claim 16, wherein the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, anti-C5a, anti-C3a, anti-C5aR, anti-C3aR, anti-TNF-alpha, and anti-IL-6, Anti-HMGB1, ST2 antibody, IL-33 antibody.
18. The method of claim 11, wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, modulators of coagulation therapy, and modulators of vascular permeability therapy.
19. The method of claim 11, wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor and an anticoagulant.
20. The method of claim 19, wherein the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, activated protein C, antithrombin, and thrombomodulin.
21. The method of any one of claims 7-20, further comprising administering or having administered therapy to the subject based on the therapy recommendation.
22. The method of any one of claims 1-21, wherein obtaining or having obtained quantitative data comprises: obtaining a sample from a subject exhibiting dysregulated host response, wherein the sample comprises a plurality of biomarkers; anddetermining the quantitative data from the obtained sample.
23. The method of claim 22, wherein the obtained sample comprises a blood sample from the subject.
24. The method of any one of claim 1 or 7-23, wherein the subject exhibiting dysregulated host response does not exhibit shock, and wherein the at least one biomarker set is one of group 1, group 3, or group 4.
25. The method of any one of claim 1 or 7-23, wherein the subject exhibiting dysregulated host response is further exhibiting shock, and wherein the at least one biomarker set is one of group 1, group 2, group 4, group 5, group 6, group 7, or group 8.
26. The method of any one of claim 1 or 7-23, wherein the subject exhibiting dysregulated host response is an adult subject, and wherein the at least one biomarker set is one of group 1, group 2, group 3, group 5, group 6, group 7, or group 8.
27. The method of any one of claim 1 or 7-23, wherein the subject exhibiting dysregulated host response is a pediatric subject, and wherein the at least one biomarker set is one of group 1, group 4, group 5, group 6, group 7, or group 8.
28. The method of any one of claims 1-27, wherein the quantitative data is determined by one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction.
29. The method of any one of claims 1-28, wherein the quantitative data is determined by: contacting a sample with a reagent;generating a plurality of complexes between the reagent and the plurality of biomarkers in the sample; anddetecting the plurality of complexes to obtain a dataset associated with the sample, wherein the dataset comprises the quantitative data.
30. The method of any one of claims 1-29, wherein the classification of the subject is determined by: determining, for at least one candidate classification of the subject, a classification-specific score for the subject;determining, by the patient subtype classifier, based on the classification-specific score, the classification of the subject.
31. The method of claim 30, wherein determining the classification-specific score comprises: determining a first subscore of the quantitative data for the subject for one or more biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more biomarkers of the candidate classification are increased relative to the quantitative data for the one or more biomarkers for one or more control subjects;determining a second subscore of the quantitative expression for the subject for one or more additional biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more additional biomarkers of the candidate classification are decreased relative to the quantitative data for the one or more additional biomarkers for the one or more control subjects; anddetermining a difference between the first subscore and the second subscore, the first and second geometric subscore optionally subject to scaling, and the difference comprising the classification-specific score for the subject; and
32. The method of claim 31, wherein one or both of the first subscore and the second subscore are geometric means.
33. The method of any one of claims 1-32, wherein the patient subtype classifier is a machine-learned model.
34. The method of claim 33, wherein the machine-learned model is a support vector machine (SVM).
35. The method of claim 34, where the support vector machine receives, as input, one or more classification-specific scores and outputs the classification of the subject.
36. The method of claim 30 or 31, wherein the patient subtype classifier determines the classification of the subject by: comparing the classification-specific scores to one or more threshold values; anddetermining the classification of the subject based on the comparisons.
37. The method of claim 36, wherein at least one of the one or more threshold values is a fixed value.
38. The method of claim 36, wherein at least one of the one or more threshold values is determined using training samples, the at least one threshold value representing a value on a ROC curve nearest to maximum sensitivity or maximum specificity.
39. The method of any one of claims 1-38, further comprising, prior to determining a classification of the subject using a patient subtype classifier, normalizing the quantitative data based on quantitative data for one or more housekeeping genes.
40. The method of any one of claims 30-39, wherein the candidate classifications of the subject comprise subtype A, subtype B, and subtype C.
41. The method of any one of claim 1 or 7-40, wherein the at least one biomarker set is group 1, and wherein the patient subtype classifier has an average accuracy of at least 82.93%.
42. The method of any one of claim 1 or 7-40, wherein the at least one biomarker set is group 2, and wherein the patient subtype classifier has an average accuracy of at least 89.6%.
43. The method of any one of claim 1 or 7-40, wherein the at least one biomarker set is group 3, and wherein the patient subtype classifier has an average accuracy of at least 86.3%.
44. The method of any one of claim 1 or 7-40, wherein the at least one biomarker set is group 4, and wherein the patient subtype classifier has an average accuracy of at least 98.3%.
45. The method of claim 7 or 8, wherein: the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation.
46. The method of claim 45, wherein the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response not provided corticosteroid therapy is greater than or equal to a threshold statistical significance.
47. The method of claim 45, wherein the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy.
48. The method of claim 47, wherein the subtype is subtype A or subtype C.
49. The method of claim 45, wherein the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is greater than or equal to a threshold statistical significance.
50. The method of claim 45, wherein the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be favorably responsive to corticosteroid therapy.
51. The method of claim 50, wherein the subtype is subtype B.
52. The method of claim 45, wherein the therapy recommendation identified for the subject comprises a no therapy recommendation, wherein the no therapy recommendation is identified at least by: determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and not provided corticosteroid therapy is less than a threshold statistical significance; anddetermining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is less than a threshold statistical significance.
53. The method of any one of claims 46-52, wherein a statistical significance comprises a p-value, and wherein the threshold statistical significance comprises at least 0.1.
54. The method of claim 45, wherein the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 1 or group 4.
55. The method of claim 54, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy.
56. The method of claim 55, wherein the subtype is subtype A or subtype C.
57. The method of claim 45, wherein the therapy recommendation identified for the subject further comprises no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises subtype B.
58. The method of claim 45, wherein the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises sepsis, wherein the at least one biomarker set is one of group 2, group 3, or group 4.
59. The method of claim 58, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy.
60. The method of claim 59, wherein the subtype is subtype A
61. The method of claim 45, wherein the therapy recommendation identified for the subject further comprises a no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises a subtype likely to be non-responsive to corticosteroid therapy.
62. The method of claim 61, wherein the subtype is subtype B or subtype C.
63. The method of claim 45, wherein the therapy recommendation identified for the subject further comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 2.
64. The method of claim 63, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be adversely responsive to corticosteroid therapy.
65. The method of claim 64, wherein the subtype is subtype C.
66. The method of claim 45, wherein the therapy recommendation identified for the subject further comprises a no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy.
67. The method of claim 66, wherein the subtype is subtype A or subtype B.
68. The method of claim 45, wherein the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 3.
69. The method of claim 68, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy.
70. The method of claim 69, wherein the subtype is subtype A or subtype C.
71. The method of claim 45, wherein the therapy recommendation identified for the subject further comprises corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be responsive to corticosteroid therapy.
72. The method of claim 71, wherein the subtype is subtype B.
73. A method for identifying a candidate therapeutic, the method comprising: accessing a differentially expressed gene database comprising gene level fold changes between patients of different subtypes;determining at least a threshold number of genes are differentially expressed in patients of a first subtype in comparison to patients of a second subtype, wherein each of the differentially expressed genes is involved in a common biological pathway; anddetermining a candidate therapeutic likely to be effective for patients of the first subtype, wherein the candidate therapeutic is effective in modulating expression of at least one of the genes that are differentially expressed in patients of the first subtype.
74. The method of claim 73, wherein the differentially expressed gene database is generated by: obtaining labeled patient data, wherein labels of the labeled patient data identify patients that are classified into one of two or more subtypes;generating the differentially expressed gene database for at least one or more genes by at least determining gene-level fold changes between patient data with a label indicating a first subtype and patient data with a label indicating a second subtype.
75. The method of claim 73 or 74, wherein the labels of the labeled patient data are generated by applying a clustering analysis or by applying a patient subtype classifier.
76. The method of any one of claims 73-75, wherein at least the threshold number of genes is at least three genes, at least four genes, at least five genes, at least six genes, at least seven genes, at least eight genes, at least nine genes, or at least ten genes.
77. The method of any one of claims 73-76, wherein determining a candidate therapeutic for patients of the first subtype further comprises: analyzing one or both of: therapeutic pharmacology data comprising data for the candidate therapeutic; andhost response pathobiology comprising data for patients of the first subtype.
78. A non-transitory computer readable medium for determining a patient subtype, the non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to: obtain quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8,wherein biomarker 2 is one of SERPINB1 or GSPT1, andwherein biomarker 3 is one of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A,wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one of ZNF831, MME, CD3G, or STOM,wherein biomarker 5 is one of ECSIT, LAT, or NCOA4, andwherein biomarker 6 is one of SLC1A5, IGF2BP2, or ANXA3,wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one of C14orf159 or PUM2,wherein biomarker 8 is one of EPB42 or RPS6KA5, andwherein biomarker 9 is one of EPB42 or GBP2; andwherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one of MSH2, DCTD, or MMP8,wherein biomarker 11 is one of HK3, UCP2, or NUP88, andwherein biomarker 12 is one of GABARAPL2 or CASP4; andwherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G,wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1, andwherein biomarker 15 is one of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; anddetermine a classification of a subject based on the quantitative data using a patient subtype classifier.
79. The non-transitory computer readable medium of claim 78, wherein the at least one biomarker set is group 5, and wherein biomarker 13 is one of STOM, MME, BNT3A2, or HLA-DPA1.
80. The non-transitory computer readable medium of claim 78 or 79, wherein the at least one biomarker set is group 5, and wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1.
81. The non-transitory computer readable medium of any one of claims 78-80, wherein the at least one biomarker set is group 5, and wherein biomarker 15 is one of SLC1A5, IGF2BP2, or ANXA3.
82. A non-transitory computer readable medium for determining a therapy recommendation for a patient, the non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to: obtain quantitative data for two or more biomarkers selected from the group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, TOMM70A, ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, ANXA3, C14orf159, PUM2, EPB42, RPS6KA5, GBP2, MSH2, DCTD, HK3, UCP2, NUP88, GABARAPL2, and CASP4; anddetermine a classification of a subject based on the quantitative data using a patient subtype classifier.
83. A non-transitory computer readable medium for determining a therapy recommendation for a patient, the non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to: obtain quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises two or more biomarkers selected from a group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1 GSPT1, MPP1, HMBS, TALL C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A,wherein group 2 comprises two or more biomarkers selected from a group consisting of ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, and ANXA3,wherein group 3 comprises two or more biomarkers selected from a group consisting of C14orf159, PUM2, EPB42, RPS6KA5, EPB42, and GBP2; andwherein group 4 comprises two or more biomarkers selected from a group consisting of MSH2, DCTD, MMP8, HK3, UCP2, NUP88, GABARAPL2, and CASP4; andwherein group 5 comprises two or more biomarkers selected from a group consisting of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, CD3G, EPB42, GSPT1, LAT, HK3, SERPINB1, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, and TNFRSF1A; anddetermine a classification of a subject based on the quantitative data using a patient subtype classifier.
84. The non-transitory computer readable medium of any one of claims 78-83, further comprising instructions that, when executed by the processor, cause the processor to identify a therapy recommendation for the subject based at least in part on the classification.
85. A non-transitory computer readable medium for determining a therapy recommendation for a subject, the non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to: obtain a classification of the subject exhibiting a dysregulated host response, the classification having been determined by: obtaining or having obtained quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8,wherein biomarker 2 is one of SERPINB1 or GSPT1, andwherein biomarker 3 is one of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A,wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one of ZNF831, MME, CD3G, or STOM,wherein biomarker 5 is one of ECSIT, LAT, or NCOA4, andwherein biomarker 6 is one of SLC1A5, IGF2BP2, or ANXA3,wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one of C14orf159 or PUM2,wherein biomarker 8 is one of EPB42 or RPS6KA5, andwherein biomarker 9 is one of EPB42 or GBP2; andwherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one of MSH2, DCTD, or MMP8,wherein biomarker 11 is one of HK3, UCP2, or NUP88, andwherein biomarker 12 is one of GABARAPL2 or CASP4; andwherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G,wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1, andwherein biomarker 15 is one of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; anddetermining the classification based on the quantitative data using a patient subtype classifier; andidentify a therapy recommendation for the subject based at least in part on the classification.
86. The non-transitory computer readable medium of claim 85, wherein the dysregulated host response of the subject comprises one of sepsis and dysregulated host response not caused by infection.
87. The non-transitory computer readable medium of any one of claims 78-86, wherein the classification of the subject comprises one of subtype A or subtype B.
88. The non-transitory computer readable medium of any one of claims 78-86, wherein the classification of the subject comprises one of subtype A, subtype B, or subtype C.
89. The non-transitory computer readable medium of claim 87 or 88, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject comprises at least no immunosuppressive therapy.
90. The non-transitory computer readable medium of claim 87 or 88, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject further comprises at least no corticosteroid therapy.
91. The non-transitory computer readable medium of claim 90, wherein the therapy recommendation identified for the subject further comprises at least one of no hydrocortisone.
92. The non-transitory computer readable medium of claim 87 or 88, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy.
93. The non-transitory computer readable medium of claim 87 or 88, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor, a blocker of complement components, a blocker of complement component receptors, and a blocker of a pro-inflammatory cytokine.
94. The non-transitory computer readable medium of claim 93, wherein the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, anti-C5a, anti-C3a, anti-C5aR, anti-C3aR, anti-TNF-alpha, and anti-IL-6, Anti-HMGB1, ST2 antibody, IL-33 antibody.
95. The non-transitory computer readable medium of claim 88, wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, modulators of coagulation therapy, and modulators of vascular permeability therapy.
96. The non-transitory computer readable medium of claim 88, wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor and an anticoagulant.
97. The non-transitory computer readable medium of claim 96, wherein the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, activated protein C, antithrombin, and thrombomodulin.
98. The non-transitory computer readable medium of any one of claims 78-97, wherein the instructions that cause the processor to obtain quantitative data further comprises instructions that, when executed by the processor, cause the processor to: obtain a sample from a subject exhibiting dysregulated host response, wherein the sample comprises a plurality of biomarkers; anddetermine the quantitative data from the obtained sample.
99. The non-transitory computer readable medium of claim 98, wherein the obtained sample comprises a blood sample from the subject.
100. The non-transitory computer readable medium of any one of claim 78 or 84-99, wherein the subject exhibiting dysregulated host response does not exhibit shock, and wherein the at least one biomarker set is one of group 1, group 3, or group 4.
101. The non-transitory computer readable medium of any one of claim 78 or 84-99, wherein the subject exhibiting dysregulated host response is further exhibiting shock, and wherein the at least one biomarker set is one of group 1, group 2, group 4, group 5, group 6, group 7, or group 8.
102. The non-transitory computer readable medium of any one of claim 78 or 84-99, wherein the subject exhibiting dysregulated host response is an adult subject, and wherein the at least one biomarker set is one of group 1, group 2, group 3, group 5, group 6, group 7, or group 8.
103. The non-transitory computer readable medium of any one of claim 78 or 84-99, wherein the subject exhibiting dysregulated host response is a pediatric subject, and wherein the at least one biomarker set is one of group 1, group 4, group 5, group 6, group 7, or group 8.
104. The non-transitory computer readable medium of any one of claims 78-103, wherein the quantitative data is determined by one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction.
105. The non-transitory computer readable medium of any one of claims 78-104, wherein the quantitative data is determined by: contacting a sample with a reagent;generating a plurality of complexes between the reagent and the plurality of biomarkers in the sample; anddetecting the plurality of complexes to obtain a dataset associated with the sample, wherein the dataset comprises the quantitative data.
106. The non-transitory computer readable medium of any one of claims 78-105, wherein the classification of the subject is determined by: determining, for at least one candidate classification of the subject, a classification-specific score for the subject;determining, by the patient subtype classifier, based on the classification-specific score, the classification of the subject.
107. The non-transitory computer readable medium of claim 106, wherein the instructions that cause the processor to determine the classification-specific score further comprises instructions that, when executed by the processor, cause the processor to: determine a first subscore of the quantitative data for the subject for one or more biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more biomarkers of the candidate classification are increased relative to the quantitative data for the one or more biomarkers for one or more control subjects;determine a second subscore of the quantitative expression for the subject for one or more additional biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more additional biomarkers of the candidate classification are decreased relative to the quantitative data for the one or more additional biomarkers for the one or more control subjects; anddetermine a difference between the first subscore and the second subscore, the first and second geometric subscore optionally subject to scaling, and the difference comprising the classification-specific score for the subject; and
108. The non-transitory computer readable medium of claim 107, wherein one or both of the first subscore and the second subscore are geometric means.
109. The non-transitory computer readable medium of any one of claims 78-108, wherein the patient subtype classifier is a machine-learned model.
110. The non-transitory computer readable medium of claim 109, wherein the machine-learned model is a support vector machine (SVM).
111. The non-transitory computer readable medium of claim 110, where the support vector machine receives, as input, one or more classification-specific scores and outputs the classification of the subject.
112. The non-transitory computer readable medium of claim 110 or 111, wherein the patient subtype classifier determines the classification of the subject by: comparing the classification-specific scores to one or more threshold values; anddetermining the classification of the subject based on the comparisons.
113. The non-transitory computer readable medium of claim 112, wherein at least one of the one or more threshold values is a fixed value.
114. The non-transitory computer readable medium of claim 112, wherein at least one of the one or more threshold values is determined using training samples, the at least one threshold value representing a value on a ROC curve nearest to maximum sensitivity or maximum specificity.
115. The non-transitory computer readable medium of any one of claims 78-114, further comprising, prior to determining a classification of the subject using a patient subtype classifier, normalizing the quantitative data based on quantitative data for one or more housekeeping genes.
116. The non-transitory computer readable medium of any one of claims 106-115, wherein the candidate classifications of the subject comprise subtype A, subtype B, and subtype C.
117. The non-transitory computer readable medium of any one of claim 78 or 85-116, wherein the at least one biomarker set is group 1, and wherein the patient subtype classifier has an average accuracy of at least 82.93%.
118. The non-transitory computer readable medium of any one of claim 78 or 85-116, and wherein the patient subtype classifier has an average accuracy of at least 89.6%.
119. The non-transitory computer readable medium of any one of claim 78 or 85-116, and wherein the patient subtype classifier has an average accuracy of at least 86.3%.
120. The non-transitory computer readable medium of any one of claim 78 or 85-116, wherein the at least one biomarker set is group 4, and wherein the patient subtype classifier has an average accuracy of at least 98.3%.
121. The non-transitory computer readable medium of claim 84 or 85, wherein: the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation.
122. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response not provided corticosteroid therapy is greater than or equal to a threshold statistical significance.
123. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy.
124. The non-transitory computer readable medium of claim 123, wherein the subtype is subtype A or subtype C.
125. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is greater than or equal to a threshold statistical significance.
126. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be favorably responsive to corticosteroid therapy.
127. The non-transitory computer readable medium of claim 126, wherein the subtype is subtype B.
128. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation identified for the subject comprises a no therapy recommendation, wherein the no therapy recommendation is identified at least by: determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and not provided corticosteroid therapy is less than a threshold statistical significance; anddetermining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is less than a threshold statistical significance.
129. The non-transitory computer readable medium of claim 122-128, wherein a statistical significance comprises a p-value, and wherein the threshold statistical significance comprises at least 0.1.
130. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 1 or group 4.
131. The non-transitory computer readable medium of claim 130, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy.
132. The non-transitory computer readable medium of claim 131, wherein the subtype is subtype A or subtype C.
133. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation identified for the subject further comprises no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises subtype B.
134. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises sepsis, wherein the at least one biomarker set is one of group 2, group 3, or group 4.
135. The non-transitory computer readable medium of claim 134, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy.
136. The non-transitory computer readable medium of claim 135, wherein the subtype is subtype A.
137. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation identified for the subject further comprises a no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises a subtype likely to be non-responsive to corticosteroid therapy.
138. The non-transitory computer readable medium of claim 137, wherein the subtype is subtype B or subtype C.
139. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation identified for the subject further comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 2.
140. The non-transitory computer readable medium of claim 139, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be adversely responsive to corticosteroid therapy.
141. The non-transitory computer readable medium of claim 140, wherein the subtype is subtype C.
142. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation identified for the subject further comprises a no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy.
143. The non-transitory computer readable medium of claim 142, wherein the subtype is subtype A or subtype B.
144. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 3.
145. The non-transitory computer readable medium of claim 144, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy.
146. The non-transitory computer readable medium of claim 145, wherein the subtype is subtype A or subtype C.
147. The non-transitory computer readable medium of claim 121, wherein the therapy recommendation identified for the subject further comprises corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be responsive to corticosteroid therapy.
148. The non-transitory computer readable medium of claim 147, wherein the subtype is subtype B.
149. A non-transitory computer readable medium for identifying a candidate therapeutic, the non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to: access a differentially expressed gene database comprising gene level fold changes between patients of different subtypes;determine at least a threshold number of genes are differentially expressed in patients of a first subtype in comparison to patients of a second subtype, wherein each of the differentially expressed genes is involved in a common biological pathway; anddetermine a candidate therapeutic likely to be effective for patients of the first subtype, wherein the candidate therapeutic is effective in modulating expression of at least one of the genes that are differentially expressed in patients of the first subtype.
150. The non-transitory computer readable medium of claim 149, wherein the differentially expressed gene database is generated by: obtaining labeled patient data, wherein labels of the labeled patient data identify patients that are classified into one of two or more subtypes;generating the differentially expressed gene database for at least one or more genes by at least determining gene-level fold changes between patient data with a label indicating a first subtype and patient data with a label indicating a second subtype.
151. The non-transitory computer readable medium of claim 149 or 150, wherein the labels of the labeled patient data are generated by applying a clustering analysis or by applying a patient subtype classifier.
152. The non-transitory computer readable medium of any one of claims 149-151, wherein at least the threshold number of genes is at least three genes, at least four genes, at least five genes, at least six genes, at least seven genes, at least eight genes, at least nine genes, or at least ten genes.
153. The non-transitory computer readable medium of any one of claims 149-152, wherein determining a candidate therapeutic for patients of the first subtype further comprises: analyzing one or both of:therapeutic pharmacology data comprising data for the candidate therapeutic; andhost response pathobiology comprising data for patients of the first subtype.
154. A system for determining a patient subtype, the system comprising: a set of reagents used for determining quantitative data for at least one biomarker set from a test sample from a subject, the at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8,wherein biomarker 2 is one of SERPINB1 or GSPT1, andwherein biomarker 3 is one of MPP1, HMBS, TALL C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A,wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one of ZNF831, MME, CD3G, or STOM,wherein biomarker 5 is one of ECSIT, LAT, or NCOA4, andwherein biomarker 6 is one of SLC1A5, IGF2BP2, or ANXA3,wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one of C14orf159 or PUM2,wherein biomarker 8 is one of EPB42 or RPS6KA5, andwherein biomarker 9 is one of EPB42 or GBP2; andwherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one of MSH2, DCTD, or MMP8,wherein biomarker 11 is one of HK3, UCP2, or NUP88, andwherein biomarker 12 is one of GABARAPL2 or CASP4; andwherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G,wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1, andwherein biomarker 15 is one of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; andan apparatus configured to receive a mixture of one or more reagents in the set and the test sample and to measure the quantitative data for the at least one biomarker set from the test sample; anda computer system communicatively coupled to the apparatus to obtain the quantitative data for the at least one biomarker set and to determine a classification of the subject based on the quantitative data using a patient subtype classifier.
155. The system of claim 154, wherein the at least one biomarker set is group 5, and wherein biomarker 13 is one of STOM, MME, BNT3A2, or HLA-DPA1.
156. The system of claim 154 or 155, wherein the at least one biomarker set is group 5, and wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1.
157. The system of any one of claims 154-156, wherein the at least one biomarker set is group 5, and wherein biomarker 15 is one of SLC1A5, IGF2BP2, or ANXA3.
158. A system for determining a patient subtype, the system comprising: a set of reagents used for determining quantitative data for two or more biomarkers selected from the group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, TOMM70A, ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, ANXA3, C14orf159, PUM2, EPB42, RPS6KA5, GBP2, MSH2, DCTD, HK3, UCP2, NUP88, GABARAPL2, and CASP4; andan apparatus configured to receive a mixture of one or more reagents in the set and the test sample and to measure the quantitative data for the at least one biomarker set from the test sample; anda computer system communicatively coupled to the apparatus to obtain the quantitative data for the at least one biomarker set and to determine a classification of the subject based on the quantitative data using a patient subtype classifier.
159. A system for determining a patient subtype, the system comprising: a set of reagents used for determining quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises two or more biomarkers selected from a group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1 GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A,wherein group 2 comprises two or more biomarkers selected from a group consisting of ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, and ANXA3,wherein group 3 comprises two or more biomarkers selected from a group consisting of C14orf159, PUM2, EPB42, RPS6KA5, EPB42, and GBP2; andwherein group 4 comprises two or more biomarkers selected from a group consisting of MSH2, DCTD, MMP8, HK3, UCP2, NUP88, GABARAPL2, and CASP4; andwherein group 5 comprises two or more biomarkers selected from a group consisting of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, CD3G, EPB42, GSPT1, LAT, HK3, SERPINB1, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, and TNFRSF1A; andan apparatus configured to receive a mixture of one or more reagents in the set and the test sample and to measure the quantitative data for the at least one biomarker set from the test sample; anda computer system communicatively coupled to the apparatus to obtain the quantitative data for the at least one biomarker set and to determine a classification of the subject based on the quantitative data using a patient subtype classifier.
160. The system of any one of claims 154-159, wherein the computer system is configured to identify a therapy recommendation for the subject based at least in part on the classification.
161. A system for determining a therapy recommendation for a subject, the system comprising: a computer system configured to: obtain a classification of the subject exhibiting a dysregulated host response, the classification having been determined by: obtaining or having obtained quantitative data for at least one biomarker set obtained from the subject, the at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5,wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8,wherein biomarker 2 is one of SERPINB1 or GSPT1, andwherein biomarker 3 is one of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A,wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one of ZNF831, MME, CD3G, or STOM,wherein biomarker 5 is one of ECSIT, LAT, or NCOA4, andwherein biomarker 6 is one of SLC1A5, IGF2BP2, or ANXA3,wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one of C14orf159 or PUM2,wherein biomarker 8 is one of EPB42 or RPS6KA5, andwherein biomarker 9 is one of EPB42 or GBP2; andwherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one of MSH2, DCTD, or MMP8,wherein biomarker 11 is one of HK3, UCP2, or NUP88, andwherein biomarker 12 is one of GABARAPL2 or CASP4; andwherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G,wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1, andwherein biomarker 15 is one of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; anddetermine the classification based on the quantitative data using a patient subtype classifier; andidentify a therapy recommendation for the subject based at least in part on the classification.
162. The system of claim 161, wherein the dysregulated host response of the subject comprises one of sepsis and dysregulated host response not caused by infection.
163. The system of any one of claims 154-162, wherein the classification of the subject comprises one of subtype A or subtype B.
164. The system of any one of claims 154-162, wherein the classification of the subject comprises one of subtype A, subtype B, or subtype C.
165. The system of claim 163 or 164, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject comprises at least no immunosuppressive therapy.
166. The system of claim 163 or 164, wherein responsive to the classification of the subject comprising subtype A, the therapy recommendation identified for the subject further comprises at least no corticosteroid therapy.
167. The system of claim 166, wherein the therapy recommendation identified for the subject further comprises at least one of no hydrocortisone.
168. The system of claim 163 or 164, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, blocking of complement activity therapy, and anti-inflammatory therapy.
169. The system of claim 163 or 164, wherein responsive to the classification of the subject comprising subtype B, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor, a blocker of complement components, a blocker of complement component receptors, and a blocker of a pro-inflammatory cytokine.
170. The system of claim 163, wherein the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, anti-C5a, anti-C3a, anti-C5aR, anti-C3aR, anti-TNF-alpha, and anti-IL-6, Anti-HMGB1, ST2 antibody, IL-33 antibody.
171. The system of claim 164, wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject comprises at least one of no therapy recommendation, immune stimulation therapy, suppression of immune regulation therapy, blocking of immune suppression therapy, modulators of coagulation therapy, and modulators of vascular permeability therapy.
172. The system of claim 164, wherein responsive to the classification of the subject comprising subtype C, the therapy recommendation identified for the subject further comprises at least one of a checkpoint inhibitor and an anticoagulant.
173. The system of claim 172, wherein the therapy recommendation identified for the subject further comprises at least one of GM-CSF, anti-PD-1, anti-PD-L1, anti-CLTA-4, anti-CEACAM-1, anti-TIM-3, anti-BTLA, IL-7, INF-gamma, IFN-beta 1a regulator, IL-22 agonist, IFN-alpha regulator, IFN-lambda regulator, IFN-alpha 2b stimulant, activated protein C, antithrombin, and thrombomodulin.
174. The system of any one of claims 154-163, wherein the sample comprises a blood sample from the subject.
175. The system of any one of claim 154 or 161-174, wherein the subject exhibiting dysregulated host response does not exhibit shock, and wherein the at least one biomarker set is one of group 1, group 3, or group 4.
176. The system of any one of claim 154 or 161-174, wherein the subject exhibiting dysregulated host response is further exhibiting shock, and wherein the at least one biomarker set is one of group 1, group 2, group 4, group 5, group 6, group 7, or group 8.
177. The system of any one of claim 154 or 161-174, wherein the subject exhibiting dysregulated host response is an adult subject, and wherein the at least one biomarker set is one of group 1, group 2, group 3, group 5, group 6, group 7, or group 8.
178. The system of any one of claim 154 or 161-174, wherein the subject exhibiting dysregulated host response is a pediatric subject, and wherein the at least one biomarker set is one of group 1, group 4, group 5, group 6, group 7, or group 8.
179. The system of any one of claims 154-178, wherein the quantitative data is determined by one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction.
180. The system of any one of claims 154-179, wherein the classification of the subject is determined by: determining, for at least one candidate classification of the subject, a classification-specific score for the subject;determining, by the patient subtype classifier, based on the classification-specific score, the classification of the subject.
181. The system of claim 180, wherein determine the classification-specific score further comprises: determine a first subscore of the quantitative data for the subject for one or more biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more biomarkers of the candidate classification are increased relative to the quantitative data for the one or more biomarkers for one or more control subjects;determine a second subscore of the quantitative expression for the subject for one or more additional biomarkers of the candidate classification, wherein the quantitative data for the subject for the one or more additional biomarkers of the candidate classification are decreased relative to the quantitative data for the one or more additional biomarkers for the one or more control subjects; anddetermine a difference between the first subscore and the second subscore, the first and second geometric subscore optionally subject to scaling, and the difference comprising the classification-specific score for the subject; and
182. The system of claim 181, wherein one or both of the first subscore and the second subscore are geometric means.
183. The system of any one of claims 154-182, wherein the patient subtype classifier is a machine-learned model.
184. The system of claim 183, wherein the machine-learned model is a support vector machine (SVM).
185. The system of claim 184, where the support vector machine receives, as input, one or more classification-specific scores and outputs the classification of the subject.
186. The system of claim 180 or 181, wherein the patient subtype classifier determines the classification of the subject by: comparing the classification-specific scores to one or more threshold values; anddetermining the classification of the subject based on the comparisons.
187. The system of claim 186, wherein at least one of the one or more threshold values is a fixed value.
188. The system of claim 186, wherein at least one of the one or more threshold values is determined using training samples, the at least one threshold value representing a value on a ROC curve nearest to maximum sensitivity or maximum specificity.
189. The system of any one of claims 154-188, further comprising, prior to determining a classification of the subject using a patient subtype classifier, normalizing the quantitative data based on quantitative data for one or more housekeeping genes.
190. The system of any one of claims 180-189, wherein the candidate classifications of the subject comprise subtype A, subtype B, and subtype C.
191. The system of any one of claim 154 or 161-190, wherein the at least one biomarker set is group 1, and wherein the patient subtype classifier has an average accuracy of at least 82.93%.
192. The system of any one of claim 154 or 161-190, and wherein the patient subtype classifier has an average accuracy of at least 89.6%.
193. The system of any one of claim 154 or 161-190, and wherein the patient subtype classifier has an average accuracy of at least 86.3%.
194. The system of any one of claim 154 or 161-190, wherein the at least one biomarker set is group 4, and wherein the patient subtype classifier has an average accuracy of at least 98.3%.
195. The system of claim 158 or 161, wherein the therapy recommendation identified for the subject further comprises corticosteroid therapy, no corticosteroid therapy, or no therapy recommendation.
196. The system of claim 195, wherein the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response not provided corticosteroid therapy is greater than or equal to a threshold statistical significance.
197. The system of claim 195, wherein the therapy recommendation comprises a no corticosteroid therapy, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy.
198. The system of claim 197, wherein the subtype is subtype A or subtype C.
199. The system of claim 195, wherein the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is greater than or equal to a threshold statistical significance.
200. The system of claim 195, wherein the therapy recommendation comprises a corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be favorably responsive to corticosteroid therapy.
201. The system of claim 200, wherein the subtype is subtype B.
202. The system of claim 195, wherein the therapy recommendation identified for the subject comprises a no therapy recommendation, wherein the no therapy recommendation is identified at least by: determining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and not provided corticosteroid therapy is less than a threshold statistical significance; anddetermining that a statistical significance of a reduction in mortality of subjects exhibiting dysregulated host response and provided corticosteroid therapy is less than a threshold statistical significance.
203. The system of any one of claims 196-202, wherein a statistical significance comprises a p-value, and wherein the threshold statistical significance comprises at least 0.1.
204. The system of claim 195, wherein the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 1 or group 4.
205. The system of claim 204, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy.
206. The system of claim 205, wherein the subtype is subtype A or subtype C.
207. The system of claim 195, wherein the therapy recommendation identified for the subject further comprises no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises subtype B.
208. The system of claim 195, wherein the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises sepsis, wherein the at least one biomarker set is one of group 2, group 3, or group 4.
209. The system of claim 208, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be adversely responsive to corticosteroid therapy.
210. The system of claim 209, wherein the subtype is subtype A.
211. The system of claim 195, wherein the therapy recommendation identified for the subject further comprises a no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises a subtype likely to be non-responsive to corticosteroid therapy.
212. The system of claim 211, wherein the subtype is subtype B or subtype C.
213. The system of claim 195, wherein the therapy recommendation identified for the subject further comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 2.
214. The system of claim 213, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be adversely responsive to corticosteroid therapy.
215. The system of claim 214, wherein the subtype is subtype C.
216. The system of claim 195, wherein the therapy recommendation identified for the subject further comprises a no therapy recommendation, wherein the no therapy recommendation is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy.
217. The system of claim 216, wherein the subtype is subtype A or subtype B.
218. The system of claim 195, wherein the therapy recommendation identified for the subject comprises a no corticosteroid therapy, wherein the dysregulated host response comprises dysregulated host response not caused by infection, and wherein the at least one biomarker set is group 3.
219. The system of claim 218, wherein the no corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype like to be non-responsive to corticosteroid therapy.
220. The system of claim 219, wherein the subtype is subtype A or subtype C.
221. The system of claim 195, wherein the therapy recommendation identified for the subject further comprises corticosteroid therapy, wherein the corticosteroid therapy is identified by determining that the classification of the subject comprises a subtype likely to be responsive to corticosteroid therapy.
222. The system of claim 221, wherein the subtype is subtype B.
223. A system for identifying a candidate therapeutic, the system comprising: a storage device storing a differentially expressed gene database comprising gene level fold changes between patients of different subtypes;a computational device configured to: access one or more gene level fold changes corresponding to differentially expressed genes in the differentially expressed gene database;determine at least a threshold number of genes are differentially expressed in patients of a first subtype in comparison to patients of a second subtype, wherein each of the differentially expressed genes is involved in a common biological pathway; anddetermine a candidate therapeutic likely to be effective for patients of the first subtype, wherein the candidate therapeutic is effective in modulating expression of at least one of the genes that are differentially expressed in patients of the first subtype.
224. The system of claim 223, wherein the differentially expressed gene database is generated by: obtaining labeled patient data, wherein labels of the labeled patient data identify patients that are classified into one of two or more subtypes;generating the differentially expressed gene database for at least one or more genes by at least determining gene-level fold changes between patient data with a label indicating a first subtype and patient data with a label indicating a second subtype.
225. The system of claim 223 or 224, wherein the labels of the labeled patient data are generated by applying a clustering analysis or by applying a patient subtype classifier.
226. The system of any one of claims 223-225, wherein at least the threshold number of genes is at least three genes, at least four genes, at least five genes, at least six genes, at least seven genes, at least eight genes, at least nine genes, or at least ten genes.
227. The system of any one of claims 223-226, wherein determining a candidate therapeutic for patients of the first subtype further comprises: analyzing one or both of: therapeutic pharmacology data comprising data for the candidate therapeutic; andhost response pathobiology comprising data for patients of the first subtype.
228. A kit for determining a patient subtype, the kit comprising: a set of reagents for determining quantitative data for at least one biomarker set from a test sample from a subject, the at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises biomarker 1, biomarker 2, and biomarker 3, wherein biomarker 1 is one of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, or MMP8,wherein biomarker 2 is one of SERPINB1 or GSPT1, andwherein biomarker 3 is one of MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, or TOMM70A,wherein group 2 comprises biomarker 4, biomarker 5, and biomarker 6, wherein biomarker 4 is one of ZNF831, MME, CD3G, or STOM,wherein biomarker 5 is one of ECSIT, LAT, or NCOA4, andwherein biomarker 6 is one of SLC1A5, IGF2BP2, or ANXA3,wherein group 3 comprises biomarker 7, biomarker 8, and biomarker 9, wherein biomarker 7 is one of C14orf159 or PUM2,wherein biomarker 8 is one of EPB42 or RPS6KA5, andwherein biomarker 9 is one of EPB42 or GBP2; andwherein group 4 comprises biomarker 10, biomarker 11, and biomarker 12, wherein biomarker 10 is one of MSH2, DCTD, or MMP8,wherein biomarker 11 is one of HK3, UCP2, or NUP88, andwherein biomarker 12 is one of GABARAPL2 or CASP4; andwherein group 5 comprises biomarker 13, biomarker 14, and biomarker 15, wherein biomarker 13 is one of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, or CD3G,wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1, andwherein biomarker 15 is one of SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, OR TNFRSF1A; andinstructions for using the set of reagents to determine the quantitative data for the at least one biomarker set.
229. The kit of claim 228, wherein the at least one biomarker set is group 5, and wherein biomarker 13 is one of STOM, MME, BNT3A2, or HLA-DPA1.
230. The kit of claim 228 or 229, wherein the at least one biomarker set is group 5, and wherein biomarker 14 is one of EPB42, GSPT1, LAT, HK3, or SERPINB1.
231. The kit of any one of claims 228-230, wherein the at least one biomarker set is group 5, and wherein biomarker 15 is one of SLC1A5, IGF2BP2, or ANXA3.
232. A kit for determining a patient subtype, the kit comprising: a set of reagents for determining quantitative data for two or more biomarkers selected from the group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1, GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, TOMM70A, ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, ANXA3, C14orf159, PUM2, EPB42, RPS6KA5, GBP2, MSH2, DCTD, HK3, UCP2, NUP88, GABARAPL2, and CASP4; andinstructions for using the set of reagents to determine the quantitative data for the at least one biomarker set.
233. A kit for determining a patient subtype, the kit comprising: a set of reagents for determining quantitative data for at least one biomarker set selected from the group consisting of the biomarker sets of group 1, group 2, group 3, group 4, or group 5, wherein group 1 comprises two or more biomarkers selected from a group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, SERPINB1 GSPT1, MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, TNFRSF1A, PRPF3, and TOMM70A,wherein group 2 comprises two or more biomarkers selected from a group consisting of ZNF831, MME, CD3G, STOM, ECSIT, LAT, NCOA4, SLC1A5, IGF2BP2, and ANXA3,wherein group 3 comprises two or more biomarkers selected from a group consisting of C14orf159, PUM2, EPB42, RPS6KA5, EPB42, and GBP2; andwherein group 4 comprises two or more biomarkers selected from a group consisting of MSH2, DCTD, MMP8, HK3, UCP2, NUP88, GABARAPL2, and CASP4; andwherein group 5 comprises two or more biomarkers selected from a group consisting of STOM, MME, BNT3A2, HLA-DPA1, ZNF831, CD3G, EPB42, GSPT1, LAT, HK3, SERPINB1, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, BTN3A2, and TNFRSF1A; andinstructions for using the set of reagents to determine the quantitative data for the at least one biomarker set.
234. The kit of any one of claims 228-233, wherein the instructions comprise instructions for determining the quantitative data by performing one of RT-qPCR (quantitative reverse transcription polymerase chain reaction), qPCR (quantitative polymerase chain reaction), PCR (polymerase chain reaction), RT-PCR (reverse transcription polymerase chain reaction), SDA (strand displacement amplification), RPA (recombinase polymerase amplification), MDA (multiple displacement amplification), HDA (helicase dependent amplification), LAMP (loop-mediated isothermal amplification), RCA (rolling circle amplification), NASBA (nucleic acid-sequence-based amplification), and any other isothermal or thermocycled amplification reaction.
235. The kit of any one of claims 228-234, wherein the set of reagents comprises at least three primer sets for amplifying at least three biomarkers, wherein the at least three primer sets comprise pairs of single-stranded DNA primers for amplifying the at least three biomarkers, andwherein at least one of the at least three biomarkers is selected from the group consisting of the biomarkers EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, ZNF831, MME, CD3G, STOM, C14orf159, PUM2, MSH2, DCTD, BNT3A2, or HLA-DPA1,at least one biomarker of the at least three biomarkers is selected from the group consisting of the biomarkers SERPINB1, GSPT1, ECSIT, LAT, NCOA4, EPB42, RPS6KA5, HK3, UCP2, or NUP88, andat least one biomarker of the at least three biomarkers is selected from the group consisting of the biomarkers MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, PRPF3, TOMM70A, EPB42, GABARAPL2, CASP4, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, OR TNFRSF1A.
236. The kit of claim 235, wherein the at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 7 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 8,a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 9 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 10,a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 11 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 12, anda forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 13 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 14,wherein at least one of the at least three primer sets is selected from the group consisting of:a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 15 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 16,a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 17 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 18, anda forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 19 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 20, andwherein at least one of the at least three primer sets is selected from the group consisting of:a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 1 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 2;a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 3 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 4, anda forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 5 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 6.
237. The kit of claim 235, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 7 and a reverse primer comprising SEQ ID NO. 8,a forward primer comprising SEQ ID NO. 9 and a reverse primer comprising SEQ ID NO. 10,a forward primer comprising SEQ ID NO. 11 and a reverse primer comprising SEQ ID NO. 12, anda forward primer comprising SEQ ID NO. 13 and a reverse primer comprising SEQ ID NO. 14,wherein at least one of the at least three primer sets is selected from the group consisting of:a forward primer comprising SEQ ID NO. 15 and a reverse primer comprising SEQ ID NO. 16,a forward primer comprising SEQ ID NO. 17 and a reverse primer comprising SEQ ID NO. 18, anda forward primer comprising SEQ ID NO. 19 and a reverse primer comprising SEQ ID NO. 20, andwherein at least one of the at least three primer sets is selected from the group consisting of:a forward primer comprising SEQ ID NO. 1 and a reverse primer comprising SEQ ID NO. 2;a forward primer comprising SEQ ID NO. 3 and a reverse primer comprising SEQ ID NO. 4, anda forward primer comprising SEQ ID NO. 5 and a reverse primer comprising SEQ ID NO. 6.
238. The kit of claim 235, wherein the at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 21 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 22, anda forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 23 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 24,wherein at least one of the at least three primer sets is selected from the group consisting of:a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 25 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 26, anda forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 29 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 30, andwherein at least one of the at least three primer sets is selected from the group consisting of:a forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 25 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 26, anda forward primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 27 and a reverse primer comprising at least 15 contiguous nucleotides of SEQ ID NO. 28.
239. The kit of claim 235, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward primer comprising SEQ ID NO. 21 and a reverse primer comprising SEQ ID NO. 22, anda forward primer comprising SEQ ID NO. 23 and a reverse primer comprising SEQ ID NO. 24,wherein at least one of the at least three primer sets is selected from the group consisting of:a forward primer comprising SEQ ID NO. 25 and a reverse primer comprising SEQ ID NO. 26, anda forward primer comprising SEQ ID NO. 29 and a reverse primer comprising SEQ ID NO. 30, andwherein at least one of the at least three primer sets is selected from the group consisting of:a forward primer comprising SEQ ID NO. 25 and a reverse primer comprising SEQ ID NO. 26, anda forward primer comprising SEQ ID NO. 27 and a reverse primer comprising SEQ ID NO. 28.
240. The kit of any one of claims 228-234, wherein the set of reagents comprises at least three primer sets for amplifying at least three biomarkers, wherein each primer set of the at least three primer sets comprises a forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer for amplifying one of the at least three biomarkers, andwherein at least one of the at least three biomarkers is selected from the group consisting of EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, ZNF831, MME, CD3G, STOM, C14orf159, PUM2, MSH2, DCTD, BNT3A2, or HLA-DPA1,at least one biomarker of the at least three biomarkers is selected from the group consisting of SERPINB1, GSPT1, ECSIT, LAT, NCOA4, EPB42, RPS6KA5, HK3, UCP2, or NUP88, andat least one biomarker of the at least three biomarkers is selected from the group consisting of MPP1, HMBS, TALL C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, PRPF3, TOMM70A, EPB42, GABARAPL2, CASP4, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, OR TNFRSF1A.
241. The kit of claim 240, wherein at least one of the at least three primer sets is selected from the group consisting of: a forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer, each of which is configured to enable amplification of at least one biomarker selected from the group consisting of: EVL, BTN3A2, HLA-DPA1, IDH3A, ACBD3, EXOSC10, SNRK, MMP8, ZNF831, MME, CD3G, STOM, C14orf159, PUM2, MSH2, DCTD, BNT3A2, or HLA-DPA1,a forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer, each of which is configured to enable amplification of at least one biomarker selected from the group consisting of: SERPINB1, GSPT1, ECSIT, LAT, NCOA4, EPB42, RPS6KA5, HK3, UCP2, or NUP88, anda forward outer primer, a backward outer primer, a forward inner primer, a backward inner primer, a forward loop primer, and a backward loop primer, each of which is configured to enable amplification of at least one biomarker selected from the group consisting of: MPP1, HMBS, TAL1, C9orf78, POLR2L, SLC27A3, BTN3A2, DDX50, FCHSD2, GSTK1, UBE2E1, PRPF3, TOMM70A, EPB42, GABARAPL2, CASP4, SLC1A5, IGF2BP2, ANXA3, GBP2, TNFRSF1, OR TNFRSF1A.

CROSS REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of and priority to U.S. Provisional Patent Application No. 62/909,530 filed Oct. 2, 2019 and U.S. Provisional Patent Application No. 63/009,331 filed Apr. 13, 2020, the entire disclosures of which are each hereby incorporated by reference in its entirety for all purposes.

PCT Information

Filing Document	Filing Date	Country	Kind
PCT/US2020/054033	10/2/2020	WO

Provisional Applications (2)

	Number	Date	Country
	62909530	Oct 2019	US
	63009331	Apr 2020	US

Biomarker Panels for Guiding Dysregulated Host Response Therapy

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

International Classifications

Abstract

Description

Claims

CROSS REFERENCE TO RELATED APPLICATIONS

PCT Information

Provisional Applications (2)