Patent Application
20220082554
The present invention relates to markers for predicting and/or judging the efficacy of therapies with PD-1 signal inhibitor-containing drugs.
The results of recent clinical trials have revealed that the anti-PD-1 antibody therapy is more effective than conventional standard therapies in various cancers (Non-Patent Documents Nos. 1-3). The response rate of PD-1 antibody therapy was 20-30% when used alone and 60-70% when combined with other therapies, showing a dramatic improvement compared to conventional immunotherapies. However, about one half of the patients were non-responsive. Little is known about why those patients are non-responsive to the PD-1 antibody therapy.
At present, two markers are approved by the Food and Drug Administration (FDA), USA for anti-PD-1 antibody therapy.
One is PD-L1 expression on at least 50% of tumor cells. Pembrolizumab is applicable to those patients with 50% or greater expression in first-line treatment (Non-Patent Document No. 4).
The other relates to the number of mutations in tumor cells. When more than a specific number of mutations are found in solid tumor cells, Nivolumab may be applied to such tumors (Non-Patent Document No. 5).
While the markers approved by FDA focus on tumors, the present inventors have found markers focusing on hosts' blood metabolites and immunostimulation (Patent Document No. 1).
It is an object of the present invention to provide markers for predicting and/or judging the efficacy prior to or at an early stage of disease therapy with a PD-1 signal inhibitor-containing drug.
The present inventors have investigated into plasma metabolites and cellular markers (including mitochondria-related markers) derived from the blood of non-small cell lung cancer (NSCLC) patients before and after treatment with an anti-PD-1 antibody Nivolumab. The results revealed that metabolites related to microbiome, energy metabolism and redox are correlated with responsiveness to Nivolumab. The present inventors have also revealed that cellular markers are correlated with plasma metabolites. The present invention has been achieved based on these findings.
A summary of the present invention is described as below.
Since the therapy using anti-PD-1 antibody, a representative of PD-1 signal inhibitors, is very expensive, judging the efficacy prior to or at an early stage of such therapy will lead to reduction of the cost of the treatment.
The present specification encompasses the contents disclosed in the specification and/or drawings of Japanese Patent Application No. 2019-000181 based on which the present patent application claims priority.
a) A schematic diagram of this study. b) Comparison of 247 metabolites between non-responders and responders at each time point was summarized in volcano plots. Metabolites with log 2|fold change|>1.0 and −log 10 (p-value)>1.3 were considered significant. Metabolites showing significant differences between responders and non-responders are listed in table. Modality indicates the analytical platform (GC-MS or LC-MS) used for metabolite measurement. The area under the curve (AUC) of each metabolite in relation to responsiveness was calculated from the univariate logistic regression analysis and is shown in the rightmost column of the table. c) The peak areas measured by GC- or LC-MS of each gut microbiota-related metabolite in non-responders (NR) and responders (R). d) The peak areas of redox/energy metabolism-related metabolites. Each dot represents one patient. Error bars show median and interquartile range. *p<0.05, **p<0.01 by Kruskal-Wallis test followed by Dunn's multiple comparisons test (c and d).
a) The stepwise Akaike's information criterion (AIC) regression procedure selected the best predictive combinations of metabolites I (in the 1st sample alone), II (in the 1st and 2nd samples) and III (in the 1st, 2nd and 3rd samples). Detailed data on each metabolite are shown in
a) A stepwise AIC regression procedure selected the best predictive combinations of cellular markers I (in the 1st sample alone), II (in the 1st and 2nd samples) and III (in the 1st, 2nd and 3rd samples). b) Two representative NSCLC samples (non-responder and responder) after gating on CD8+ PBMC (left). X axis represents CCR7 and Y axis PD-1. Frequency of PD-1high CD8+ T cells in non-responders and responders in 1st samples (right). c) Representative histograms of Mito SOX on gated CD4+ and CD8+ T cells (left). Ratio of Mito SOX levels in CD8+ and CD4+ T cells (Mito SOX CD8/CD4) for non-responders and responders in the 1st samples (right). d) PGC-1αβ of the 1st, 2nd and 3rd samples among CD8+ PBMC (upper left). Change in MFI of PGC-1αβ between the 1st, 2nd and 3rd samples (upper right). The solid line and the dotted line represent responders and non-responders, respectively. Fold change of PGC-1αβ expression between non-responders and responders in the 2nd relative to 1st samples (lower left) and in the 3rd relative to 2nd samples (lower right). e) Frequency of CD4+ T cells among PBMC in the 1st and 2nd samples (left). The solid line and the dotted line represent responders and non-responders, respectively. Fold change of CD4+ T cell frequency in the 2nd relative to 1st samples between non-responders and responders (right). Each dot represents one patient. Error bars show median and interquartile range. *p<0.05, **p<0.01, ****p<0.0001 by Wilcoxon rank sum test.
a) LDA evaluated the accuracy of cellular marker combination I. Canonical plot of LDA for determination of LDA-R and LDA-NR is shown. Each dot represents one patient. The vertical dotted line indicates the cut-off value. b) Kaplan-Meier plots of PFS and OS of LDA-R and LDA-NR determined by cellular marker combination I. c) Canonical plot for LDA based on cellular marker combination II. d) Kaplan-Meier plots of PFS and OS of LDA-R and LDA-NR determined by the cellular marker combination II. e) Canonical plot for LDA based on cellular marker combination III. f) Kaplan-Meier plots of PFS and OS of LDA-R and LDA-NR determined by cellular marker combination III. *p<0.05, **p<0.01, ***p<0.001 by Gehan-Breslow-Wilcoxon test (b, d, and f). g) The ROC curve of five-fold cross validation for cellular marker combinations I, II and III.
a) Scatter plots between cellular markers (X-axis) and metabolite markers (Y-axis). Filled circles represent responders and open circles non-responders. r: Spearman correlation coefficients. b) A clustered heatmap of absolute correlation coefficients over all marker pairs detected in a) above (using Spearman correlation distance, and complete linkage). Dark denotes higher correlation (|r| close to 1) and light lower correlation (|r| close to 0). The markers clustered into 3 groups, which were designated as metabolic categories 1, 2 and 3. c) Summary of the Spearman correlation analysis. Metabolic category 1 (gut microbiota-related metabolites) is correlated with PGC-1αβ of CD8+ T cells; metabolic category 2 (FAO-related metabolites) is correlated with the frequency of PD-1high CD8+ T cells; and metabolic category 3 (redox-related metabolites) is correlated with the T cell Mito SOX marker.
a) Kaplan-Meier plots of PFS and OS of patients sorted by the criteria of PFS>3 months (solid line) and PFS≤3 months (dotted line). b) Kaplan-Meier plots of PFS and OS of patients sorted by the criteria of PFS>6 months (solid line) and PFS≤6 months (dotted line). **p<0.01, ***p<0.001, ****p<0.0001 by Gehan-Breslow-Wilcoxon test. c) Kaplan-Meier plots of PFS and OS of patients sorted by frequency of PD-L1 expression on tumors. The dashed line, solid line, and dotted line show patients with high PD-L1 expression (>50%), low PD-L1 expression (1-50%), and rare PD-L1 expression (<1%), respectively. d) The solid line and dotted line show patients with positive expression (>1%) and negative expression (<1%) of PD-L1, respectively.
a) The peak area measured by GC- or LC-MS of each microbiota related metabolite in patients without pre-antibiotics treatment (ATB(−)) and with pre-antibiotics treatment (ATB(+)) are shown. These graphs display the data of the 1st+2nd+3rd samples. **p<0.01, ****p<0.0001 by Wilcoxon rank sum test. b) The peak areas measured by GC-MS of hippuric acid and indoxyl sulfate in non-responders (NR) and responders (R) are shown. Each dot represents one patient. Error bars show median and interquartile range. *p<0.05, **p<0.01, ***p<0.001 by Kruskal-Wallis test followed by Dunn's multiple comparisons test. c) The peak areas of acylcarnitine species between 1st, 2nd and 3rd samples are shown. The solid line and dotted line represent responders and non-responders, respectively. *p<0.05, **p<0.01 by Wilcoxon rank sum test.
a-c) Graphs show comparison of peak areas of metabolite markers selected by stepwise regression analysis between non-responders and responders. Each dot represents one patient. Error bars indicate the median and interquartile range. *p<0.05, **p<0.01 by Wilcoxon rank sum test.
a) Graph shows comparison of PD-1 positive frequency among CD8+ T cells in non-responders and responders in the 1st samples. b) FACS data show the expression levels of PD-1, Ki-67, Granzyme B, IFN-γ, T-bet and EOMES among CD8+ T cells in PBMC. Representative NSCLC samples are depicted in left panels. Right panels show the frequencies of Ki-67+, Granzyme B+, T-bet+ and EOMES+ cells in PD-1hi, PD-1low, and PD-1(−) CD8+ T cells. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 by Kruskal Wallis test followed by Dunn's multiple comparisons test. c) CD4+ T cells were sorted into naive, Tcm, Tem and Temra subsets according to the expression of CD45RO and CCR7 (left panel). The frequencies of CD4+ Tcm and CD4+ Temra in non-responders and responders are shown (right panels). **p<0.01 by Wilcoxon rank sum test.
Hereinbelow, the present invention will be described in detail.
The present invention provides a test method comprising predicting and/or judging the efficacy of therapy with a PD-1 signal inhibitor-containing drug using the following (i) and/or (ii) as indicator(s):
(i) at least one metabolite in serum and/or plasma selected from the group consisting of alanine, 4-cresol, cysteine, hippuric acid, oleic acid, indoxyl sulfate, ribose, indoleacetate, uric acid, trans-urocanic acid, pipecolic acid, N-acetylglucosamine, indolelactic acid, arabinose, arabitol, cystine, indoxyl sulfate, gluconic acid, citrulline, creatinine, N-acetylaspartic acid, pyroglutamic acid, trimethyllysine, asy-dimethylarginine, sym-dimethylarginine, methylhistidine, acylcarnitine, 3-aminoisobutyric acid, acetylcarnosine, arginine, N-acetylornitine, 3-hydroxyisovaleric acid, pyruvic acid, α-ketoglutaric acid, GSSG, 2-hydrobutyric acid, 1,5-anhydro-D-sorbitol, glutamine, glycine, lysine, taurine, AMP, acetylcarnosine, 3-hydroxybutyric acid, 2-hydroxyisovaleric acid, acetoacetic acid, tryptophan, 2-hydroxyglutaric acid, malic acid, quinolinic acid, caproic acid, isoleucine, GSH and 3-OH-kynurenine
(ii) at least one cellular marker in peripheral blood selected from the group consisting of the frequency of CD4+ T cells among peripheral blood mononuclear cells (% of CD4+ T cells among PBMC), the frequency of CD8+ T cells among peripheral blood mononuclear cells (% of CD8+ T cells among PBMC), the frequency of nave T cells among CD8+ T cells (% of Tnaive among CD8+ T cells), the frequency of central memory T cells among CD4+ T cells (% of Tcm among CD4+ T cells), the frequency of central memory T cells among CD8+ T cells (% of Tcm among CD8+ T cells), the frequency of effector memory T cells among CD8+ T cells (% of Tem among CD8+ T cells), the frequency of terminally differentiated effector memory T cells among CD4+ T cells (% of Temra among CD4+ T cells), the frequency of terminally differentiated effector memory T cells among CD8+ T cells (% of Temra among CD8+ T cells), the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells [e.g., the ratio of mitochondrial reactive oxygen species production in CD8+ T cells to mitochondrial reactive oxygen species production in CD4+ T cells (Mito SOX CD8/CD4), the ratio of mitochondrial mass in CD8+ T cells to mitochondrial mass in CD4+ T cells (Mito mass CD8/CD4)], PGC-1α and PGC-1β expression in CD8+ T cells (PGC-1αβ (MF1) of CD8+ T cells), the frequency of PD-1 highly expressing CD8+ T cell population (% of PD-1high among CD8+ T cells), the frequency of FoxP3 lowly expressing CD45RA+ T cell population among CD4+ T cells (% of FoxP3low CD45RA+ among CD4+ T cells), the frequency of T-bet highly expressing T cell population among CD4+ T cells (% of T-bethigh among CD4+ T cells), the frequency of T-bet lowly expressing T cell population among CD4+ T cells (% of T-betlow among CD4+ T cells), the frequency of T-bet expressing T cell population among CD8+ T cells (% of T-bet+ among CD8+ T cells) and the frequency of T-bet and EOMES expressing T cell population among CD8+ T cells (% of T-bet+ EOMES+ among CD8+ T cells).
In one embodiment of the present invention, the therapeutic efficacy in a subject of a PD-1 signal inhibitor-containing drug may be predicted and/or judged from the measured values of the patient based on a criterion whether the level(s) of metabolite and/or cellular marker at the time point or the ratio of two time points as indicated in Tables 4 and 5 (Example 1 described later) is(are) high or low (Changes in R relative to NR).
Examples of such criteria are described below. Each of these criteria may be used alone or in combination.
Table 4 (Metabolites Detected in Serum and/or Plasma)
(Before Treatment (Drug Administration) (Time Point: 1st))
In one preferred embodiment of the present invention, the metabolite of (i) may be at least one metabolite selected from the group consisting of hippuric acid, arabinose and acylcarnitine, and the cellular marker of (ii) may be at least one cellular marker selected from the group consisting of the frequency of PD-1 highly expressing CD8+ T cell population, the ratio of mitochondrial reactive oxygen species production in CD8+ T cells to mitochondrial reactive oxygen species production in CD4+ T cells, and PGC-1α and PGC-1β expression in CD8+ T cells.
As used herein, the term “PD-1 signal” refers to the signal transduction mechanism which PD-1 bears. As one aspect of this mechanism, PD-1 inhibits T cell activation in collaboration with its ligands PD-L1 and PD-L2. PD-1 (Programmed cell death-1) is a membrane protein expressed in activated T cells and B cells. Its ligands PD-L1 and PD-L2 are expressed in various cells such as antigen-presenting cells (monocytes, dendritic cells, etc.) and cancer cells. PD-1, PD-L1 and PD-L2 work as inhibitory factors which inhibit T cell activation. Certain types of cancer cells and virus-infected cells escape from host immune surveillance by expressing the ligands of PD-1 to thereby inhibit T cell activation.
PD-1 signal inhibitors are drugs which block PD-1 signals. By blocking PD-1 signals, PD-1 signal inhibitors cancel inhibition of T cell activation and enhance immune surveillance so as to exert therapeutic effect on cancers and virus infections (PD-1 blockade therapy). As PD-1 signal inhibitors, substances which specifically bind to PD-1, PD-L1 or PD-L2 may be given. Such substances include, but are not limited to, proteins, polypeptides, oligopeptides, nucleic acids (including natural-type and artificial nucleic acids), low molecular weight organic compounds, inorganic compounds, cell extracts, and extracts from animals, plants, soils or the like. These substances may be either natural or synthetic products. Preferable PD-1 signal inhibitors are antibodies. More preferably, antibodies such as anti-PD-1 antibody, anti-PD-L1 antibody and anti-PD-L2 antibody may be given. Any type of antibody may be used as long as it is capable of inhibiting PD-1 signal. The antibody may be any of polyclonal antibody, monoclonal antibody, chimeric antibody, single chain antibody, humanized antibody or human-type antibody. Methods for preparing such antibodies are known. The antibody may be derived from any organisms such as human, mouse, rat, rabbit, goat or guinea pig. As used herein, the term “antibody” is a concept encompassing antibodies of smaller molecular sizes such as Fab, F(ab)′2, ScFv, Diabody, VH, VL, Sc(Fv)2, Bispecific sc(Fv)2, Minibody, scFv-Fc monomer or scFv-Fc dimer.
In the present invention, the PD-1 signal inhibitor-containing drug may be a PD-1 signal inhibitor as a single drug that does not contain other medical ingredients; alternatively, it may be a combined drug containing a PD-1 signal inhibitor and other medical ingredients.
Other medical ingredients may be ones that enhance or reinforce the drug efficacy of PD-1 signal inhibitors, or ones that alleviate the adverse effect of PD-1 signal inhibitors. The drug efficacy of other medical ingredients is not particularly limited. Currently approved concomitant drugs include, but are not limited to, ipilimumab, cisplatin, carboplatin, paclitaxel and pemetrexed. A large number of concomitant drugs are under development, including those which are currently used as standard of care, immune regulation-related drugs (such as those relating to energy metabolism), and angiogenic inhibitors (1-3).
The PD-1 signal inhibitor-containing drug may be used as an active ingredient of an anticancer agent, an anti-infective agent or a combination thereof.
Subjects to be tested may be either persons who have already received administration of other drugs (such as anti-cancer agent) or persons who have not received such administration. Preferably, subjects have already received administration of other drugs.
In the present invention, prediction and/or judgement of the efficacy of therapy with a PD-1 signal inhibitor-containing drug may be made at any stage of the following: before start of treatment, after start of treatment, between a plurality of times of drug administration, etc.
In one preferred embodiment of the present invention, at least one member selected from the group consisting of hippuric acid, arabinose and acylcarnitine in the serum and/or plasma of a subject may be measured.
Hippuric acid is one of microbiota-derived metabolites and known to be produced preferentially by Clostridium (4, 5).
Arabinose is converted into xylulose-5-phosphate (an intermediate of pentose phosphate cycle) and enters into the metabolic pathway for synthesizing NADPH and nucleic acid. Enteric bacteria, in particular, are highly expressing enzymes for the above conversion, which potentially affects gut microbiota (6).
Acylcarnitine is converted from short-chain acyl-CoA and carnitine by carnitine-dependent enzymes, carnitine acetyltransferase and carnitine palmitoyltransferase 1 (CPT1) which are localized in the mitochondrial matrix. Acylcarnitine may be 4- to 6-carbon acylcarnitine; specifically, butyrylcarnitine, isovalerylcarnitine and hexanoylcarnitine. Butyrylcarnitine, the 4-carbone acylcarnitine, serves as a fatty acid transporter into mitochondria to generate ATP. Acylcarnitine species with various carbon numbers are released from cells once the function of fatty acid oxidation (FAO) is attenuated.
In the present invention, at least one member selected from the group consisting of cysteine, cystine, arginine and glutathione disulfide in serum and/or plasma may be used as the indicator of (i).
In the biosynthesis and metabolism of glutathione (GSH), cysteine binds to L-glutamate to form L-γ-glutamylcysteine, which then binds to glycine to yield GSH. GSH is oxidized into its oxidized form (GSSG) after reaction with reactive oxygen species (ROS) (
Cystine is a component of glutathione (
Arginine is important for composing cytoskeleton and as an energy source. It is believed that immunosuppression resulting from arginine depletion is occurring under tumor local environment where macrophage produces arginase highly.
Alanine, 4-cresol, cysteine, hippuric acid, oleic acid, indoxyl sulfate, ribose, indoleacetate, uric acid, trans-urocanic acid, pipecolic acid, N-acetylglucosamine, indolelactic acid, arabinose, arabitol, cystine, indoxyl sulfate, gluconic acid, citrulline, creatinine, N-acetylaspartic acid, pyroglutamic acid, trimethyllysine, asy-dimethylarginine, sym-dimethylarginine, methylhistidine, acylcarnitine, 3-aminoisobutyric acid, acetylcarnosine, arginine, N-acetylornitine, 3-hydroxyisovaleric acid, pyruvic acid, α-ketoglutaric acid, GSSG, 2-hydrobutyric acid, 1,5-anhydro-D-sorbitol, glutamine, glycine, lysine, taurine, AMP, acetylcarnosine, 3-hydroxybutyric acid, 2-hydroxyisovaleric acid, acetoacetic acid, tryptophan, 2-hydroxyglutaric acid, malic acid, quinolinic acid, caproic acid, isoleucine, GSH and 3-OH-kynurenine may be measured by mass spectrometry (e.g., GC-MS or LC-MS)
In the present invention, alanine, 4-cresol, cysteine, hippuric acid, oleic acid, indoxyl sulfate, ribose, indoleacetate, uric acid, trans-urocanic acid, pipecolic acid, N-acetylglucosamine, indolelactic acid, arabinose, arabitol, cystine, indoxyl sulfate, gluconic acid, citrulline, creatinine, N-acetylaspartic acid, pyroglutamic acid, trimethyllysine, asy-dimethylarginine, sym-dimethylarginine, methylhistidine, acylcarnitine, 3-aminoisobutyric acid, acetylcarnosine, arginine, N-acetylornitine, 3-hydroxyisovaleric acid, pyruvic acid, α-ketoglutaric acid, GSSG, 2-hydrobutyric acid, 1,5-anhydro-D-sorbitol, glutamine, glycine, lysine, taurine, AMP, acetylcarnosine, 3-hydroxybutyric acid, 2-hydroxyisovaleric acid, acetoacetic acid, tryptophan, 2-hydroxyglutaric acid, malic acid, quinolinic acid, caproic acid, isoleucine, GSH and 3-OH-kynurenine in serum and/or plasma may be used as biomarkers for predicting and/or judging the efficacy of therapy with a PD-1 signal inhibitor-containing drug.
In one embodiment of the present invention, it is possible to predict that therapy with a PD-1 signal inhibitor-containing drug is effective when cysteine level in serum and/or plasma before administration of the drug is high and hippuric acid level in serum and/or plasma before administration of the drug is high.
In another embodiment of the present invention, it is possible to judge that therapy with a PD-1 signal inhibitor-containing drug is effective when arabinose level in serum and/or plasma after 1st administration of the drug is high; arginine level in serum and/or plasma after 1st administration of the drug is high; and butyrylcarnitine level in serum and/or plasma after 1st administration of the drug is low.
In still another embodiment of the present invention, it is possible to judge that therapy with a PD-1 signal inhibitor-containing drug is effective when hippuric acid level in serum and/or plasma before administration of the drug is high; cystine level in serum and/or plasma after 1st administration of the drug is high; glutathione disulfide level in serum and/or plasma after 2nd administration of the drug is high; and butyrylcarnitine level in serum and/or plasma after 2nd administration of the drug is low.
Further, in one preferred embodiment of the present invention, at least one member selected from the group consisting of the frequency of PD-1 highly expressing CD8+ T cell population, the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells, and PGC-1α and PGC-1β expression in CD8+ T cells in peripheral blood may be measured.
PD-1 highly expressing CD8+ T cell population is called exhausted cells, and is led to unresponsiveness after undergoing multiple times of division. High frequency of this cell population means that the ratio of exhausted, unresponsive cells is high, which is believed to result in weakened immune response to cancers.
As used herein, the expression “PD-1 highly expressing” means that the fluorescence intensity is higher than 103 when PD-1 expression is detected by flow cytometry. However, this value is variable. The reference value may be determined using patient samples stained with isotype control, patient samples with no (or a few) cases of high positive, and patient samples with many cases of high positive.
The ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells is likely to reflect the intensity of immunity that injures tumors. The mitochondrial activity of CD8+ T cells (killer T cells) is deeply involved in the status of intracellular energy metabolism, and is capable of reflecting the cytotoxic activity of killer T cells that injure cancer cells. The mitochondrial activity is lowered when CD8+ T cells are exhausted to become nonresponsive (10).
PGC-1 is a major regulator for mitochondrial biogenesis and mitochondrial metabolic pathways (oxidative phosphorylation (OXPHOS), fatty acid oxidation (FAO), and others) (10, 11). PGC-1α and PGC-1β, called transcriptional coactivators, bind to multiple transcription factors to thereby promote mitochondrial biosynthesis, mitochondria-involving energy metabolism and mitochondrial activity-related transcription (11).
PGC-1α and PGC-1β expression in CD8+ T cells is reflecting the mitochondrial activity of CD8+ T cells. When CD8+ T cells have been led to the state of exhaustion and unresponsiveness, PGC-1 expression decreases. Conversely, unresponsiveness in CD8+ T cells can be recovered by allowing high expression of PGC-1 (10).
Peripheral blood CD8+ T cells may be collected as described below. Briefly, blood samples taken from patients are placed on Ficoll and centrifuged at 2000 rpm. The buffy coat between erythrocyte phase and plasma phase is collected and washed with a cell culture medium. From the resultant lymphocytes, CD8+ T cells are isolated with a magnetic cell sorter (Miltenyi Biotec).
Peripheral blood CD4+ T cells can be collected by isolating them with a magnetic cell sorter in the same manner as described for CD8+ T cells.
The frequency of PD-1 highly expressing CD8+ T cell population may be measured by staining leucocytes from patient's peripheral blood with fluorescently labeled antibodies against PD-1 and CD8, and conducting imaging, flow cytometry or microplate analysis (for details, see the method in Example described later).
The mitochondrial activation status of CD4+ T cells or CD8+ T cells may be detected using ROS (reactive oxygen species), OCR (oxygen consumption rate), membrane potential, mitochondrial mass, or the like as an indicator.
ROS may be detected by staining CD4+ T cells or CD8+ T cells with a dye Mito SOX, and conducting imaging, flow cytometry or microplate analysis.
OCR may be measured with XF96 Extracellular Flux Analyzer (Seahorse Biosciences). Briefly, isolated CD8+ T cells are plated on a dedicated cell culture plate. A sensor cartridge is placed on the plate. Oligomycin, FCCP, Antimycine A and Rotenone are poured into the injection port of the sensor cartridge. Then, the plate with the sensor cartridge is set in XF96 Extracellular Flux Analyzer. Oxygen concentration and hydrogen ion concentration in the semi-closed microenvironment between cells and the sensor are measured.
Membrane potential may be detected by staining CD4+ T cells or CD8+ T cells with a dye MitoTracker Deep Red and conducting imaging, flow cytometry or microplate analysis.
Mitochondrial mass may be detected by staining CD4+ T cells or CD8+ T cells with a dye MitoTracker Green and conducting imaging, flow cytometry or microplate analysis.
To calculate the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells, a measured value which can be an indicator for mitochondrial activation status in CD8+ T cells may be divided by a measured value which can be an indicator for mitochondrial activation status in CD4+ T cells. As ratios of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells, the ratio of mitochondrial reactive oxygen species production (Mito SOX level) between CD8+ T cells and CD4+ T cells (Mito SOX CD8/CD4), the ratio of mitochondrial mass in CD8+ T cells to mitochondrial mass in CD4+ T cells (Mito mass CD8/CD4), and the like may be enumerated.
PGC-1α and PGC-1β expression may be detected using a monoclonal antibody which recognizes both PGC-1α and PGC-1β (hereinafter, sometimes expressed as “PGC-1αβ”).
In the present invention, the frequency of CD4+ T cells in peripheral blood mononuclear cells (PBMC) may be used as an indicator of (ii).
The frequency of CD4+ T cells in PBMC may be measured by staining leucocytes from patient's peripheral blood with a fluorescently labeled antibody against CD4, and conducting imaging, flow cytometry or microplate analysis (for details, see the method in Example described later).
The frequency of CD4+ T cells among peripheral blood mononuclear cells (PBMC) (% of CD4+ T cells among PBMC), the frequency of CD8+ T cells among PBMC (% of CD8+ T cells among PBMC), the frequency of nave T cells among CD8+ T cells (% of Tnaive among CD8+ T cells), the frequency of central memory T cells among CD4+ T cells (% of Tcm among CD4+ T cells), the frequency of central memory T cells among CD8+ T cells (% of Tcm among CD8+ T cells), the frequency of effector memory T cells among CD8+ T cells (% of Tem among CD8+ T cells), the frequency of terminally differentiated effector memory T cells among CD4+ T cells (% of Temra among CD4+ T cells), the frequency of terminally differentiated effector memory T cells among CD8+ T cells (% of Temra among CD8+ T cells), the ratio of mitochondrial mass in CD8+ T cells to mitochondrial mass in CD4+ T cells (Mito mass CD8/CD4), the frequency of FoxP3 lowly expressing CD45RA+ T cell population among CD4+ T cells (% of FoxP3low CD45RA+ among CD4+ T cells), the frequency of T-bet highly expressing T cell population among CD4+ T cells (% of T-bethigh among CD4+ T cells), the frequency of T-bet lowly expressing T cell population among CD4+ T cells (% of T-betlow among CD4+ T cells), the frequency of T-bet expressing T cell population among CD8+ T cells (% of T-bet+ among CD8+ T cells) and the frequency of T-bet and EOMES expressing T cell population among CD8+ T cells (% of T-bet+ EOMES+ among CD8+ T cells) may be measured by known methods.
The frequency of CD4+ T cells among PBMC (% of CD4+ T cells among PBMC) means the frequency of helper T cells. Helper T cells abundantly produce cytokines that help activation of CD8+ killer T cells.
The frequency of CD8+ T cells among PBMC (% of CD8+ T cells among PBMC) means the proportion of killer T cells that kill cancer cells.
The frequency of nave T cells among CD8+ T cells (% of Tnaive among CD8+ T cells) means the frequency of initial cells that generate effector CD8+ T cells which have a high capacity to kill cancer cells. When this frequency is high, it is highly possible that the number of effector CD8+ T cells will increase.
The frequency of central memory T cells among CD4+ T cells (% of Tcm among CD4+ T cells) means the proportion of initial cells that generate effector CD4+ T cells which have a high cytokine production capacity. When this frequency is high, it is highly possible that the number of effector CD4+ T cells will increase.
The frequency of central memory T cells among CD8+ T cells (% of Tcm among CD8+ T cells) means the proportion of initial cells that generate effector CD8+ T cells which have a high capacity to kill cancer cells. When this frequency is high, it is highly possible that the number of effector CD8+ T cells will increase.
The frequency of effector memory T cells among CD8+ T cells (% of Tem among CD8+ T cells) means the frequency of a cell population containing a large number of CD8+ T cells which have a high capacity to kill cancer cells.
The frequency of terminally differentiated effector memory T cells among CD4+ T cells (% of Temra among CD4+ T cells) means the frequency of a cell population containing those exhausted cells that have reached the final stage of differentiation and have a low cytokine production capacity.
The ratio of mitochondrial mass in CD8+ T cells to mitochondrial mass in CD4+ T cells (Mito mass CD8/CD4) is suggesting that mitochondria in CD8+ T cells are activated and showing that intracellular energy metabolism therein is activated.
The frequency of FoxP3 lowly expressing CD45RA+ T cell population among CD4+ T cells (% of FoxP3low CD45RA+ among CD4+ T cells) shows the frequency of activated CD4+ T cells before antigen sensitization.
The frequency of T-bet highly expressing T cell population among CD4+ T cells (% of T-bethigh among CD4+ T cells) shows the frequency of Th1 helper T cells which help CD8+ killer T cells efficiently.
The frequency of T-bet lowly expressing T cell population among CD4+ T cells (% of T-betlow among CD4+ T cells) shows the frequency of helper T cells whose T1 helper capacity is weak.
The frequency of T-bet expressing T cell population among CD8+ T cells (% of T-bet+ among CD8+ T cells) shows the frequency of CD8+ killer T cells which have a high capacity to kill cancers.
The frequency of T-bet and EOMES expressing T cell population among CD8+ T cells (% of T-bet+ EOMES+ among CD8+ T cells) shows the frequency of memory or exhausted CD8+ T cells.
In the present invention, the frequency of CD4+ T cells among peripheral blood mononuclear cells (% of CD4+ T cells among PBMC), the frequency of CD8+ T cells among peripheral blood mononuclear cells (% of CD8+ T cells among PBMC), the frequency of nave T cells among CD8+ T cells (% of Tnaive among CD8+ T cells), the frequency of central memory T cells among CD4+ T cells (% of Tcm among CD4+ T cells), the frequency of central memory T cells among CD8+ T cells (% of Tcm among CD8+ T cells), the frequency of effector memory T cells among CD8+ T cells (% of Tem among CD8+ T cells), the frequency of terminally differentiated effector memory T cells among CD4+ T cells (% of Temra among CD4+ T cells), the frequency of terminally differentiated effector memory T cells among CD8+ T cells (% of Temra among CD8+ T cells), the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells [e.g., the ratio of mitochondrial reactive oxygen species production in CD8+ T cells to mitochondrial reactive oxygen species production in CD4+ T cells (Mito SOX CD8/CD4), the ratio of mitochondrial mass in CD8+ T cells to mitochondrial mass in CD4+ T cells (Mito mass CD8/CD4)], PGC-1α and PGC-16 expression in CD8+ T cells (PGC-1αβ (MF1) of CD8+ T cells), the frequency of PD-1 highly expressing CD8+ T cell population (% of PD-1high among CD8+ T cells), the frequency of FoxP3 lowly expressing CD45RA+ T cell population among CD4+ T cells (% of FoxP3low CD45RA+ among CD4+ T cells), the frequency of T-bet highly expressing T cell population among CD4+ T cells (% of T-bethigh among CD4+ T cells), the frequency of T-bet lowly expressing T cell population among CD4+ T cells (% of T-betlow among CD4+ T cells), the frequency of T-bet expressing T cell population among CD8+ T cells (% of T-bet+ among CD8+ T cells) and the frequency of T-bet and EOMES expressing T cell population among CD8+ T cells (% of T-bet+ EOMES+ among CD8+ T cells) in peripheral blood may be used as cellular markers for predicting and/or judging the efficacy of therapy with a PD-1 signal inhibitor-containing drug.
In one embodiment of the present invention, therapy with a PD-1 signal inhibitor-containing drug can be predicted effective when the frequency of PD-1 highly expressing CD8+ T cell population in peripheral blood before administration of the drug is low and the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells (e.g., Mito SOX CD8/CD4) in peripheral blood before administration of the drug is high.
In another embodiment of the present invention, therapy with a PD-1 signal inhibitor-containing drug can be judged effective when the frequency of PD-1 highly expressing CD8+ T cell population in peripheral blood before administration of the drug is low; the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells (e.g., Mito SOX CD8/CD4) in peripheral blood before administration of the drug is high; the ratio of PGC-1α and PGC-16 expression in CD8+ T cells in peripheral blood after 1st administration of the drug to the corresponding expression before administration of the drug is low; and the ratio of the frequency of CD4+ T cells in PBMC after 1st administration of the drug to the corresponding frequency before administration of the drug is high.
In still another embodiment of the present invention, therapy with a PD-1 signal inhibitor-containing drug can be judged effective when the frequency of PD-1 highly expressing CD8+ T cell population in peripheral blood before administration of the drug is low; the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells (e.g., Mito SOX CD8/CD4) in peripheral blood before administration of the drug is high; the ratio of PGC-1α and PGC-1β expression in CD8+ T cells in peripheral blood after 2nd administration of the drug to the corresponding expression after 1st administration of the drug is high; and the ratio of the frequency of CD4+ T cells in PBMC after 1st administration of the drug to the corresponding frequency before administration of the drug is high.
As used herein, with respect to the value (level) of biomarker or cellular marker, “high” or “low” means that the value of responder is “high” or “low” relative to the value of non-responder. As regards criteria for judging “high” or “low”, 95% confidence interval of quantitative values of responders or non-responders may be used as a reference value. Alternatively, a cut-off value may be set from ROC curves or statistical analysis (such as LDA analysis) and used as a reference value. Methods for determining these values are known.
In Example described later, the present inventors investigated into the concentrations of plasma metabolites and cellular markers in the blood from 60 non-small-cell lung cancer patients before and after administration of Nivolumab (anti-PD-1 antibody). As a result, the following calculation formulas were obtained by LDA analysis. These formulas may vary when the population to be analyzed has changed. Mark * appearing in formulas 1 to 6 means multiplication.
−0.3023987485021*[PD-1 high 1st]+2.94519844381265*[Mito SOX CD8/CD4 1st]−1.87379126258675 (Formula 1)
−0.2522084377427*[PD-1 high 1st]+3.58270233554892*[Mito SOX CD8/CD4 1 st]+1.92950430842982*[CD4(%) 2nd/1 st]−1.2170886788589*[PGC1ab of CD8 2nd/1st]−3.29839771768711 (Formula 2)
−0.2808914943715*[PD-1 high 1st]+3.29512762046372*[Mito SOX CD8/CD4 1 st]+1.55202212350758*[CD4(%) 2nd/1 st]+2.00254807664124*[PGClab of CD8 3rd/2nd]−5.97495334069991 (Formula 3)
Metabolite combination I (Cut-off value 0)
9.92415868963082*[Cysteine 1st]+0.00000054083199*[Hippuric acid (LC-MS) 1st]−44.661430966258*[Unk8 1st]−1.46007680395094 (Formula 4)
Metabolite combination II (Cut-off value 0)
207.062010223744*[Arabinose 2nd]+0.00000031741819*[Arginine 2nd]−0.0000003768213*[Butyrylcarnitine 2nd]−1.95925442024969 (Formula 5)
Metabolite combination III (Cut-off value 0)
0.00000044275294*[Hippuric acid (LC-MS) 1st]+12.0744069146569*[Cystine 2nd]+0.00003552481124*[GSSG 3rd]−0.00000008812702408*[Butyrylcarnitine 3rd]−2.67048929974173 (Formula 6)
PD-1high 1 st: the frequency of PD-1 highly expressing CD8+ T cell population in peripheral blood before administration of a PD-1 signal inhibitor-containing drug.
Mito SOX CD8/CD4 1st: the ratio of mitochondrial reactive oxygen species production in CD8+ T cells to the corresponding production in CD4+ T cells in peripheral blood before administration of a PD-1 signal inhibitor-containing drug.
CD4(%) 2nd/1st: the ratio of the frequency of CD4+ T cells in PBMC after 1st administration of a PD-1 signal inhibitor-containing drug to the frequency of CD4+ T cells in PBMC before administration of the drug. (When this ratio is larger than 1, it can be said that the frequency of CD4+ T cells in PBMC has increased after 1st administration of the PD-1 signal inhibitor-containing drug, compared to the frequency before administration of the drug.)
PGC1ab of CD8 2nd/1st: the ratio of PGC-1α and PGC-1β expression in CD8+ T cells in peripheral blood after 1″ administration of a PD-1 signal inhibitor-containing drug to PGC-1α and PGC-1β expression in CD8+ T cells in peripheral blood before administration of the PD-1 signal inhibitor-containing drug. (When this ratio is smaller than 1, it can be said that PGC-1α and PGC-1β expression in CD8+ T cells in peripheral blood has decreased after 1st administration of the PD-1 signal inhibitor-containing drug, compared to the expression before administration of the drug.)
PGC1ab of CD8 3rd/2nd: the ratio of PGC-1α and PGC-1β expression in CD8+ T cells in peripheral blood after 2nd administration of a PD-1 signal inhibitor-containing drug to PGC-1α and PGC-1β expression in CD8+ T cells in peripheral blood after 1st administration of the PD-1 signal inhibitor-containing drug. (When this ratio is larger than 1, it can be said that PGC-1α and PGC-1β expression in CD8+ T cells in peripheral blood has increased after 2nd administration of the PD-1 signal inhibitor-containing drug, compared to the expression after 1st administration of the drug.)
Cysteine 1st: plasma cysteine level before administration of a PD-1 signal inhibitor-containing drug (as measured by GC-MS).
Hippuric acid (LC-MS) 1st: plasma hippuric acid level before administration of a PD-1 signal inhibitor-containing drug (as measured by LC-MS).
Unk8 1st: plasma Unk8 (substance unidentified) level before administration of a PD-1 signal inhibitor-containing drug (as measured by GC-MS).
Arabinose 2nd: plasma arabinose level after 1st administration of a PD-1 signal inhibitor-containing drug (as measured by GC-MS).
Arginine 2nd: plasma arginine level after 1st administration of a PD-1 signal inhibitor-containing drug (as measured by GC-MS).
Butyrylcarnitine 2nd: plasma butyrylcarnitine level after 1st administration of a PD-1 signal inhibitor-containing drug (as measured by GC-MS).
Cystine 2nd: plasma cystine level after 1st administration of a PD-1 signal inhibitor-containing drug (as measured by GC-MS).
GSSG 3rd: plasma glutathione disulfide level after 2nd administration of a PD-1 signal inhibitor-containing drug (as measured by GC-MS).
Butyrylcarnitine 3rd: plasma butyrylcarnitine level after 2nd administration of a PD-1 signal inhibitor-containing drug (as measured by GC-MS).
When therapy with a PD-1 signal inhibitor-containing drug has been predicted effective according to the method of the present invention, therapy with the drug may be started.
When therapy with a PD-1 signal inhibitor-containing drug has been judged effective according to the method of the present invention, therapy with the drug may be continued.
The present invention provides a method of diagnosis and therapy for a disease, comprising predicting and/or judging the efficacy of therapy with a PD-1 signal inhibitor-containing drug in a subject using the following (i) and/or (ii) as indicator(s), and administering to the subject in a therapeutically effective amount when the therapy with the PD-1 signal inhibitor-containing drug has been predicted and/or judged effective:
(i) at least one metabolite in serum and/or plasma selected from the group consisting of alanine, 4-cresol, cysteine, hippuric acid, oleic acid, indoxyl sulfate, ribose, indoleacetate, uric acid, trans-urocanic acid, pipecolic acid, N-acetylglucosamine, indolelactic acid, arabinose, arabitol, cystine, indoxyl sulfate, gluconic acid, citrulline, creatinine, N-acetylaspartic acid, pyroglutamic acid, trimethyllysine, asy-dimethylarginine, sym-dimethylarginine, methylhistidine, acylcarnitine, 3-aminoisobutyric acid, acetylcarnosine, arginine, N-acetylornitine, 3-hydroxyisovaleric acid, pyruvic acid, α-ketoglutaric acid, GSSG, 2-hydrobutyric acid, 1,5-anhydro-D-sorbitol, glutamine, glycine, lysine, taurine, AMP, acetylcarnosine, 3-hydroxybutyric acid, 2-hydroxyisovaleric acid, acetoacetic acid, tryptophan, 2-hydroxyglutaric acid, malic acid, quinolinic acid, caproic acid, isoleucine, GSH and 3-OH-kynurenine
(ii) at least one cellular marker in peripheral blood selected from the group consisting of the frequency of CD4+ T cells among peripheral blood mononuclear cells (% of CD4+ T cells among PBMC), the frequency of CD8+ T cells among peripheral blood mononuclear cells (% of CD8+ T cells among PBMC), the frequency of nave T cells among CD8+ T cells (% of Tnaive among CD8+ T cells), the frequency of central memory T cells among CD4+ T cells (% of Tcm among CD4+ T cells), the frequency of central memory T cells among CD8+ T cells (% of Tcm among CD8+ T cells), the frequency of effector memory T cells among CD8+ T cells (% of Tem among CD8+ T cells), the frequency of terminally differentiated effector memory T cells among CD4+ T cells (% of Temra among CD4+ T cells), the frequency of terminally differentiated effector memory T cells among CD8+ T cells (% of Temra among CD8+ T cells), the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells [e.g., the ratio of mitochondrial reactive oxygen species production in CD8+ T cells to mitochondrial reactive oxygen species production in CD4+ T cells (Mito SOX CD8/CD4), the ratio of mitochondrial mass in CD8+ T cells to mitochondrial mass in CD4+ T cells (Mito mass CD8/CD4)], PGC-1α and PGC-1β expression in CD8+ T cells (PGC-1αβ (MF1) of CD8+ T cells), the frequency of PD-1 highly expressing CD8+ T cell population (% of PD-1high among CD8+ T cells), the frequency of FoxP3 lowly expressing CD45RA+ T cell population among CD4+ T cells (% of FoxP3low CD45RA+ among CD4+ T cells), the frequency of T-bet highly expressing T cell population among CD4+ T cells (% of T-bethigh among CD4+ T cells), the frequency of T-bet lowly expressing T cell population among CD4+ T cells (% of T-betlow among CD4+ T cells), the frequency of T-bet expressing T cell population among CD8+ T cells (% of T-bet+ among CD8+ T cells) and the frequency of T-bet and EOMES expressing T cell population among CD8+ T cells (% of T-bet+ EOMES+ among CD8+ T cells).
The present invention also provides a pharmaceutical composition comprising a PD-1 signal inhibitor-containing drug as an active ingredient, which is for use in a method of diagnosis and therapy for a disease, comprising predicting and/or judging the efficacy of therapy with the PD-1 signal inhibitor-containing drug in a subject using the following (i) and/or (ii) as indicator(s), and administering to the subject in a therapeutically effective amount when the therapy with the PD-1 signal inhibitor-containing drug has been predicted and/or judged effective:
(i) at least one metabolite in serum and/or plasma selected from the group consisting of alanine, 4-cresol, cysteine, hippuric acid, oleic acid, indoxyl sulfate, ribose, indoleacetate, uric acid, trans-urocanic acid, pipecolic acid, N-acetylglucosamine, indolelactic acid, arabinose, arabitol, cystine, indoxyl sulfate, gluconic acid, citrulline, creatinine, N-acetylaspartic acid, pyroglutamic acid, trimethyllysine, asy-dimethylarginine, sym-dimethylarginine, methylhistidine, acylcarnitine, 3-aminoisobutyric acid, acetylcarnosine, arginine, N-acetylornitine, 3-hydroxyisovaleric acid, pyruvic acid, α-ketoglutaric acid, GSSG, 2-hydrobutyric acid, 1,5-anhydro-D-sorbitol, glutamine, glycine, lysine, taurine, AMP, acetylcarnosine, 3-hydroxybutyric acid, 2-hydroxyisovaleric acid, acetoacetic acid, tryptophan, 2-hydroxyglutaric acid, malic acid, quinolinic acid, caproic acid, isoleucine, GSH and 3-OH-kynurenine
(ii) at least one cellular marker in peripheral blood selected from the group consisting of the frequency of CD4+ T cells among peripheral blood mononuclear cells (% of CD4+ T cells among PBMC), the frequency of CD8+ T cells among peripheral blood mononuclear cells (% of CD8+ T cells among PBMC), the frequency of nave T cells among CD8+ T cells (% of Tnaive among CD8+ T cells), the frequency of central memory T cells among CD4+ T cells (% of Tcm among CD4+ T cells), the frequency of central memory T cells among CD8+ T cells (% of Tcm among CD8+ T cells), the frequency of effector memory T cells among CD8+ T cells (% of Tem among CD8+ T cells), the frequency of terminally differentiated effector memory T cells among CD4+ T cells (% of Temra among CD4+ T cells), the frequency of terminally differentiated effector memory T cells among CD8+ T cells (% of Temra among CD8+ T cells), the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells [e.g., the ratio of mitochondrial reactive oxygen species production in CD8+ T cells to mitochondrial reactive oxygen species production in CD4+ T cells (Mito SOX CD8/CD4), the ratio of mitochondrial mass in CD8+ T cells to mitochondrial mass in CD4+ T cells (Mito mass CD8/CD4)], PGC-1α and PGC-1β expression in CD8+ T cells (PGC-1αβ (MF1) of CD8+ T cells), the frequency of PD-1 highly expressing CD8+ T cell population (% of PD-1high among CD8+ T cells), the frequency of FoxP3 lowly expressing CD45RA+ T cell population among CD4+ T cells (% of FoxP3low CD45RA+ among CD4+ T cells), the frequency of T-bet highly expressing T cell population among CD4+ T cells (% of T-bethigh among CD4+ T cells), the frequency of T-bet lowly expressing T cell population among CD4+ T cells (% of T-betlow among CD4+ T cells), the frequency of T-bet expressing T cell population among CD8+ T cells (% of T-bet+ among CD8+ T cells) and the frequency of T-bet and EOMES expressing T cell population among CD8+ T cells (% of T-bet+ EOMES+ among CD8+ T cells).
The pharmaceutical composition of the present invention may be used as an anticancer agent, an anti-infective agent or a combination thereof.
When the pharmaceutical composition of the present invention is administered as an anticancer agent, target cancers or tumors includes, but are not limited to, leukemia, lymphoma (e.g., Hodgkin's disease, non-Hodgkin's lymphoma), multiple myeloma, brain tumors, breast cancer, endometrial cancer, cervical cancer, ovarian cancer, esophageal cancer, stomach cancer, appendix cancer, colon cancer, liver cancer, gallbladder cancer, bile duct cancer, pancreatic cancer, adrenal cancer, gastrointestinal stromal tumor, mesothelioma, head and neck cancer (such as laryngeal cancer), oral cancer (such as floor of mouth cancer), gingival cancer, tongue cancer, buccal mucosa cancer, salivary gland cancer, nasal sinus cancer (e.g., maxillary sinus cancer, frontal sinus cancer, ethmoid sinus cancer, sphenoid sinus cancer), thyroid cancer, renal cancer, lung cancer, osteosarcoma, prostate cancer, testicular tumor (testicular cancer), renal cell carcinoma, bladder cancer, rhabdomyosarcoma, skin cancer (e.g., basal cell cancer, squamous cell carcinoma, malignant melanoma, actinic keratosis, Bowen's disease, Paget's disease) and anal cancer.
When the pharmaceutical composition of the present invention is administered as an anti-infective agent, target infections include, but are not limited to, bacterial infections [various infections caused by Streptococcus (e.g., group Aβ hemolytic streptococcus, pneumococcus), Staphylococcus aureus (e.g., MSSA, MRSA), Staphylococcus epidermidis, Enterococcus, Listeria, Neisseria meningitis aureus, Neisseria gonorrhoeae, pathogenic Escherichia coli (e.g., 0157:H7), Klebsiella (Klebsiella pneumoniae), Proteus, Bordetella pertussis, Pseudomonas aeruginosa, Serratia, Citrobacter, Acinetobacter, Enterobacter, mycoplasma, Clostridium or the like]; tuberculosis, cholera, plague, diphtheria, dysentery, scarlet fever, anthrax, syphilis, tetanus, leprosy, Legionella pneumonia (legionellosis), leptospirosis, Lyme disease, tularemia, Q fever, and the like], rickettsial infections (e.g., epidemic typhus, scrub typhus, Japanese spotted fever), chlamydial infections (e.g., trachoma, genital chlamydial infection, psittacosis), fungal infections (e.g., aspergillosis, candidiasis, cryptococcosis, trichophytosis, histoplasmosis, Pneumocystis pneumonia), parasitic protozoan infections (e.g., amoebic dysentery, malaria, toxoplasmosis, leishmaniasis, cryptosporidiosis), parasitic helminthic infections (e.g., echinococcosis, schistosomiasis japonica, filariasis, ascariasis, diphyllobothriasis latum), and viral infections [e.g., influenza, viral hepatitis, viral meningitis, acquired immune deficiency syndrome (AIDS), adult T-cell leukemia, Ebola hemorrhagic fever, yellow fever, cold syndrome, rabies, cytomegalovirus infection, severe acute respiratory syndrome (SARS), progressive multifocal leukoencephalopathy, chickenpox, herpes zoster, hand-foot-and-mouth disease, dengue, erythema infectiosum, infectious mononucleosis, smallpox, rubella, acute anterior poliomyelitis (polio), measles, pharyngoconjunctival fever (pool fever), Marburg hemorrhagic fever, hantavirus renal hemorrhagic fever, Lassa fever, mumps, West Nile fever, herpangina and chikungunya fever].
The pharmaceutical composition of the present invention is administered to human or animal subjects systemically or locally by an oral or parenteral route.
PD-1 signal inhibitors (e.g., anti-PD-1 antibody, anti-PD-L1 antibody or anti-PD-L2 antibody) may be dissolved in buffers such as PBS, physiological saline or sterile water, optionally filter- or otherwise sterilized before being administered to human or animal subjects by injection or infusion. To the solution of PD-1 signal inhibitors, additives such as coloring agents, emulsifiers, suspending agents, surfactants, solubilizers, stabilizers, preservatives, antioxidants, buffers, isotonizing agents and the like may be added. As routes of administration, intravenous, intramuscular, intraperitoneal, subcutaneous or intradermal administration and the like may be selected.
The content of the PD-1 signal inhibitor (e.g., anti-PD-1 antibody, anti-PD-L1 antibody or anti-PD-L2 antibody) in a preparation varies with the type of the preparation and is usually 1-100% by weight, preferably 50-100% by weight. Such a preparation may be formulated into a unit dosage form.
The dose, frequency and number of administration of PD-1 signal inhibitor (e.g., anti-PD-1 antibody, anti-PD-L1 antibody or anti-PD-L2 antibody) may vary with the symptoms, age and body weight of the human or animal subject, the method of administration, dosage form and more. For example, in terms of the amount of the active ingredient usually ranges from 0.1 to 100 mg/kg body weight, preferably 1-10 mg/kg body weight, and may be administered once per adult at a frequency that enables confirmation of efficacy. Commencement, continuation or cancellation of administration of PD-1 signal inhibitor may be determined based on test results using the above-described bio-markers and/or cellular markers.
The present invention also provides use of reagents for measuring the biomarkers of the following (i) and/or (ii) in manufacturing a product for diagnosis for determining administration of a PD-1 signal inhibitor-containing drug:
(i) at least one metabolite in serum and/or plasma selected from the group consisting of alanine, 4-cresol, cysteine, hippuric acid, oleic acid, indoxyl sulfate, ribose, indoleacetate, uric acid, trans-urocanic acid, pipecolic acid, N-acetylglucosamine, indolelactic acid, arabinose, arabitol, cystine, indoxyl sulfate, gluconic acid, citrulline, creatinine, N-acetylaspartic acid, pyroglutamic acid, trimethyllysine, asy-dimethylarginine, sym-dimethylarginine, methylhistidine, acylcarnitine, 3-aminoisobutyric acid, acetylcarnosine, arginine, N-acetylornitine, 3-hydroxyisovaleric acid, pyruvic acid, α-ketoglutaric acid, GSSG, 2-hydrobutyric acid, 1,5-anhydro-D-sorbitol, glutamine, glycine, lysine, taurine, AMP, acetylcarnosine, 3-hydroxybutyric acid, 2-hydroxyisovaleric acid, acetoacetic acid, tryptophan, 2-hydroxyglutaric acid, malic acid, quinolinic acid, caproic acid, isoleucine, GSH and 3-OH-kynurenine
(ii) at least one cellular marker in peripheral blood selected from the group consisting of the frequency of CD4+ T cells among peripheral blood mononuclear cells (% of CD4+ T cells among PBMC), the frequency of CD8+ T cells among peripheral blood mononuclear cells (% of CD8+ T cells among PBMC), the frequency of nave T cells among CD8+ T cells (% of Tnaive among CD8+ T cells), the frequency of central memory T cells among CD4+ T cells (% of Tcm among CD4+ T cells), the frequency of central memory T cells among CD8+ T cells (% of Tcm among CD8+ T cells), the frequency of effector memory T cells among CD8+ T cells (% of Tem among CD8+ T cells), the frequency of terminally differentiated effector memory T cells among CD4+ T cells (% of Temra among CD4+ T cells), the frequency of terminally differentiated effector memory T cells among CD8+ T cells (% of Temra among CD8+ T cells), the ratio of mitochondrial activation status in CD8+ T cells to mitochondrial activation status in CD4+ T cells [e.g., the ratio of mitochondrial reactive oxygen species production in CD8+ T cells to mitochondrial reactive oxygen species production in CD4+ T cells (Mito SOX CD8/CD4), the ratio of mitochondrial mass in CD8+ T cells to mitochondrial mass in CD4+ T cells (Mito mass CD8/CD4)], PGC-1α and PGC-1β expression in CD8+ T cells (PGC-1αβ (MF1) of CD8+ T cells), the frequency of PD-1 highly expressing CD8+ T cell population (% of PD-1high among CD8+ T cells), the frequency of FoxP3 lowly expressing CD45RA+ T cell population among CD4+ T cells (% of FoxP3low CD45RA+ among CD4+ T cells), the frequency of T-bet highly expressing T cell population among CD4+ T cells (% of T-bethigh among CD4+ T cells), the frequency of T-bet lowly expressing T cell population among CD4+ T cells (% of T-betlow among CD4+ T cells), the frequency of T-bet expressing T cell population among CD8+ T cells (% of T-bet+ among CD8+ T cells) and the frequency of T-bet and EOMES expressing T cell population among CD8+ T cells (% of T-bet+ EOMES+ among CD8+ T cells).
The “product for diagnosis” is a concept encompassing kits, equipment (mass spectrometer, imaging system, flow cytometry, microplate and other analytical equipment), etc. Products for diagnosis may suitably contain programs for analysis, operation manuals, and the like.
As regards reagents for measuring biomarkers, internal standards, reagents for derivatization (e.g., derivatizing reagent for GC/MS), organic solvents for extraction, and combinations thereof may be enumerated when the biomarker is a metabolite. When the biomarker is a cellular marker, reagents for isolating cells (e.g., beads coated with antibodies to cellular marker (CD4, CD8, CD45RA) proteins), antibodies to the protein as a target of detection (PGC-1α, PGC-1β, FoxP3, T-bet, or EOMES), antibodies for detecting the target protein (preferably, labeled with an enzyme or the like), substrates for the enzyme used as a label (those which generate color, fluorescence or luminescence upon reaction with the enzyme), standard samples, and combinations thereof may be enumerated.
Hereinbelow, the present invention will be described more specifically with reference to the following Examples.
To identify reliable biomarkers for the host immunity, the present inventors investigated the levels of plasma metabolites and cellular markers in immune cells in the blood of patients with non-small cell lung cancer before and after administration of Nivolumab. A validation study using machine learning for the statistically selected biomarkers showed that a combination of four specific metabolites derived from gut microbiome, fatty acid oxidation and redox provided high probability (AUC=0.91). A combination of four T cell markers, including those related to mitochondrial activation and the frequencies of CD8+ PD-1high and CD4+ T cells, demonstrated even higher prediction value (AUC=0.96). When the present inventors statistically selected the highest predictive combination from the pool of all markers, the above-mentioned four T cell markers were selected; and no metabolite was included therein. It was believed that this result is due to strong linkage between metabolites and T cell markers. These findings indicate that combination of biomarkers reflecting host immune activity is quite valuable for the responder prediction.
PD-1 and CTLA-4 are key players in immune suppression upon tumor growth (12-15). Blocking these molecules individually or together rejuvenates immune surveillance and can induce strong anti-tumor activity in mice and humans (12-14, 16, 17). Based on the promising outcomes in clinical trials using these immune checkpoint blockade mAbs, the FDA has approved antibodies against CTLA-4, PD-1, or its ligand, PD-L1, to treat various human cancers (3, 15). Regardless of the impressive clinical efficacy of the PD-1 blockade immunotherapy, a significant portion of cancer patients remain non-responsive (3, 15-17). Therefore, therapeutic efficacy prediction biomarkers (hereinafter, “responder biomarkers”) to distinguish responders and non-responders are desperately required to save cost and time for those patients.
PD-L1 high expression on tumor tissue detectable by immunostaining has been used as a responder biomarker for NSCLC (18, 19). The FDA has recently approved microsatellite instability-high (MSI-H) or DNA mismatch repair deficiency (dMMR) as common responder biomarkers for various solid tumors (20, 21). These markers, however, cannot cover all of the responsive patients because the responses of tumor-reactive killer T cells are affected not only by tumor properties but also by host immune activity. Several groups have identified candidates for responder biomarkers by analysis of cell components at tumor sites or in peripheral blood. The proposed markers are the frequencies of CD8+ T cells, CD4+ T cells, eosinophils, neutrophils, subsets of suppressive macrophages, and subsets of T cells (22). Immune regulators such as specific cytokines or chemokines are also listed as candidates for biomarkers (22). However, the prediction value of each of these single markers is not high enough for the clinical use.
It has been known that gut microbiota and immune activity mutually affect as the present inventors have demonstrated using IgA-deficient or PD-1-deficient mouse models (23, 24). As regards tumor immunity, recent reports also suggested that gut microbiota and their metabolites are related to the efficacy of immune check point inhibitors (22, 25, and 26). Especially, a specific enteric bacterium, Akkermansia muciniphila, and diversity of gut flora are reported to correlate with the responsiveness to PD-1 blockade therapy (26). Although the detailed mechanism remains largely unknown, gut flora is believed to activate innate immunity and upregulate the baseline of anti-tumor immunity in the periphery (26-28). However, it is still elusive whether the above-mentioned microbiota or its associated factors could be responder biomarkers for PD-1 blockade therapy.
Recently, tight linkage between T cell energy metabolism and immune activity has been disclosed (29, 30). The present inventors have examined how energy metabolism in T cells regulates anti-tumor immunity (31, 32). From the results of measurement of oxygen consumption rate and dye staining of mitochondrial mass, reactive oxygen species (ROS) and membrane potential (these items are potential indicators for mitochondrial activity), the present inventors have found that during PD-1 blockade therapy, mitochondria in tumor-reactive T cells are activated. Furthermore, the present inventors showed that increase in mitochondria-derived ROS augments the efficacy of PD-1 blockade therapy and reported that this was mediated by peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) activation (31). It had previously been known that the level of ROS signal was tightly linked to T cell activation and cytokine production (33). The present inventors therefore examined whether these mitochondria-related factors could be predictive biomarkers for host immune activity.
In the present study using blood samples from 60 NSCLC patients, the present inventors demonstrated that a combination of several plasma metabolites could serve as a good responder biomarker (AUC=0.91 by validation study using machine learning). These metabolites include those related to gut microbiota (hippuric acid), fatty acid oxidation (butyrylcarnitine), and redox (cystine and glutathione disulfide). Further, the present inventors demonstrated that a combination of several T cell markers could serve as a better biomarker (AUC=0.96 by validation study using machine learning). These T cell markers were those related to suppressive function (the frequency of PD-1high population) and mitochondrial activities (PGC-1α and ROS expression) in CD8+ killer T cells. These T cell markers strongly linked with the above-mentioned metabolite markers. These data indicate that a combination of gut microbiota and T cell energy metabolism could be a highly predictive biomarker for PD-1 blockade therapy.
Gut Microbiota- and Energy Metabolism-Related Metabolites are Correlating with Responsiveness to PD-1 Blockade Cancer Immunotherapy.
Accumulated evidence indicates that there is a considerable association between immune responses and gut microbiota (30). However, it remains unknown whether specific metabolites can serve as predictive biomarkers for PD-1 blockade therapy in humans. In order to investigate how metabolites are associated with anti-tumor immunity, the present inventors analyzed metabolites and T cell functional markers in plasma and PBMC, respectively, in 60 NSLC patients before and after administration of Nivolumab (
The present inventors examined the probability for each of the predictive biomarker candidates listed in
Considering that patients are administered Nivolumab every two weeks in the current clinical protocol for NSCLC, metabolite combinations I and II are useful as predictive biomarkers, while metabolite combination III may be less valuable for clinical use despite the highest reliability.
The present inventors' previous reports have shown that mitochondrial activation and energy metabolism in T cells are strongly associated with responses to PD-1 blockade therapy (31, 32). Therefore, the present inventors investigated cellular markers for effector function, energy metabolism, mitochondrial status, and immune activation in CD8+ T cells in patients' PBMC. Among the total 52 markers shown in Table 3 and their ratio (fold change) between each time point, the present inventors extracted 26 items with significant differences between responders and non-responders (Table 5). Further, from these 26 items, the best combinations of predictive biomarkers were selected using the stepwise regression method as described for metabolites. As shown in
The present inventors assessed the error rates of cellular marker combinations I, II, and III by the method described above. LDA demonstrated clear separation between responders and non-responders at an error rate of 19.1% by cellular marker combination I (
Subsequently, the present inventors conducted the stepwise regression method to select the best combination of markers among the total metabolic and cellular markers showing significant differences between responders and non-responders (Table 4 and Table 5). Surprisingly, all selected markers were cellular markers, resulting in the same combinations as shown in
As a result of cluster analysis, cellular markers and metabolites were roughly divided into three groups (
Anti-tumor immunity is regulated by both tumor and immune activity. As shown in this study, the functional activity of killer T cells is related to a complex network of different high-order function systems such as gut microbiome and energy metabolism. Therefore, a single biomarker is not sufficient, and a combination of multi-markers is required to improve the accuracy of prediction. Current clinical biomarkers approved by the FDA are PD-L1 high expression in NSCLC, cervical cancer, esophageal cancer, etc. and MSI-H and/or dMMR for various solid tumors. However, a significant portion of NSCLC patients with PD-L1 low expression is responsive and the number of patients with MSI-H and/or dMMR on their tumor is very limited (around 1% of NSCLC patients) (35, 42). Considering that the expression frequency of PD-L1 in tumor could not significantly discriminate responders from non-responders in this study, currently approved biomarkers based solely on tumor properties are incomplete as predictive biomarkers for PD-1 blockade therapy. For investigating tumor factors, at least biopsy specimens are needed, which usually puts a huge burden on patients. On the other hand, blood-based tests to examine immune cell properties and metabolites would be much simpler and more patient-friendly. Although it has been reported that several markers associated with host immunity (such as the frequency of eosinophils, neutrophils, tumor-associated macrophages, CD4+ T cells in PBMC, and the frequency of infiltrated CD8+ T cells or PD-1+ CD8+ T cells in tumor sites) are related to responsiveness, these single immune-related markers are insufficient in terms of prediction accuracy (22, 43). In this study, the present inventors demonstrated for the first time that the combination of several cellular markers for T cell activation, including mitochondrial activation status in T cells, can discriminate responders from non-responders with high accuracy; this has led to the finding that the combination has a potential to serve as a good biomarker. Correlation analyses suggested that metabolites associated with gut microbiota, energy metabolism and redox are reflecting the anti-tumor activation status of peripheral killer T cells.
It has been known that the host immune activity and gut microbiota affect each other (23). Further, gut microbiota is known to be associated with various diseases, and the relationship between gut microbiota-derived blood metabolites and disease etiology has been studied (5, 44, 45). In the present study, it is noticeable that gut microbiota-derived metabolites are correlated with the behavior of PGC-1αβ, a master molecule for mitochondrial regulation in the peripheral CD8+ T cells (10, 11). It has been reported that hippuric acid, one of the microbiota-derived metabolites, can be an indicator of gut microbiota diversity and is produced bundantly by Clostridiales bacteria (4, 5). Recent studies have reported that the diversity of gut microbiota and the abundance of Clostridiales bacteria are correlated with responsiveness to PD-1 blockade therapy (25). The present study demonstrated that hippuric acid and other gut microbiota-derived metabolites show higher levels in responders. This may reflect increase of specific enterobacterial species which activate the immune system and the diversity of gut microbiota. Although the mechanism by which gut microbiota regulates anti-tumor immunity still remains largely unknown (26), the results of the present study suggest that gut microbiota and microbiota-derived metabolites are deeply correlated with the mitochondrial activation status of peripheral CD8+ T cells.
Changes in acylcarnitine levels showed correlation with the frequency of PD-1high CD8+ T cells. Acylcarnitines are converted from short-chain acyl-CoAs and carnitine by carnitine-dependent enzymes, carnitine acetyltransferase and carnitine palmitory transferase I (CPT1), which are localized in the mitochondrial matrix. In other words, carnitine serves as a shuttle to transfer acyl-CoAs from the cytoplasm to the inside of mitochondria (7-9). When FAO is promoted, acylcarnitines are substantially transported into the mitochondrial matrix, resulting in decrease in plasma acylcarnitine levels (7-9, 46, 47). Therefore, it is held that plasma levels of acylcarnitine species could be an indicator for mitochondrial activity or FAO under inflammatory reaction (48-50). In the present study, acylcarnitines (butyrylcarnitine, isovalerylcarnitine, and hexanoylcarnitine) were elevated in non-responders after treatment with PD-1 antibody, suggesting decrease in FAO in killer T cells (47). Higher acylcarnitine levels and higher frequency of PD-1high CD8+ T cells were observed in non-responders. These two markers were correlated with each other. This indicates the presence of a large number of exhausted killer T cells with reduced FAO in non-responders. These results are consistent with a previous report that FAO pathway in effector killer T cells is important for robust anti-tumor activity by PD-1 blockade (32).
Mitochondrial ROS upregulation is critical for T cell activation and cytokine production (31, 51). However, since an extremely high concentration of ROS is toxic to cells, they control ROS levels appropriately using the GSH/GSSG system (37). Oxidized GSH (GSSG) is extracellularly released into the plasma via a complex efflux system (52, 53). Therefore, it is known that plasma GSSG level is an indicator of cellular oxidation status and various immune-related diseases (52, 53). In the present study, responders saw an increase in mitochondrial ROS in killer T cells before treatment, and GSSG and GSH components (cystine and pyroglutamic acid) in responder plasma after treatment showed high levels. The correlation between these markers can be explained by ROS and the mechanism of the GSH/GSSG system (37).
To date, it has been believed that PD-1high CD8+ T cells in PBMC contain tumor-specific T cells because their Ki-67 frequency is extremely high (54, 55). In the present study, production of granzyme B and IFN-γ was decreased in PD-1high CD8+ T cells. This suggests that they are exhausted as a result of vigorous division. Indeed, PD-1high CD8+ T cells expressed less T-bet, but expressed the exhaustion marker EOMES highly. Although it has been reported that PD-1high CD8+ T cells are abundant in responder tumor tissues (56), the results of the present study show that PD-1high CD8+ T cells are not abundant in the peripheral blood of responders. This discrepancy may be attributed to 1) the number of tumor-reactive CD8+ T cells in the periphery being less in responders because such T cells preferentially move to tumor sites, and 2) the function and the extent of exhaustion of PD-1high CD8+ T cells being different between tumor sites and peripheral blood. Further analysis is necessary to clarify the reasons behind this discrepancy.
The results of the present study showed that among all metabolites and cellular makers tested, a combination of several cellular markers can provide the highest prediction accuracy. Metabolites which are capable of predicting therapeutic efficacy are strongly correlated with the cellular markers. However, considering the convenience of measuring specific metabolites and the difficulties involved in achieving stable measurements of cellular markers between different facilities, a combination of particular metabolites might be more suitable for clinical application. According to the present study, practical use of combinatorial biomarkers for cancer immunotherapy has been suggested. Through establishment of such biomarkers, those patients who are non-responsive to anti-PD-1 antibody monotherapy would be provided an opportunity to select different therapeutic options. As a consequence, enhancement of therapeutic effect can be expected.
Among NSCLC patients receiving Nivolumab (anti-PD-1 antibody) at Kyoto University Hospital, those who consented to the collection and storage of blood samples during treatment, and allowed review of their medical records for past medical history, cancer tumor type, toxicity assessments, clinical response, survival, and laboratory data were selected as subjects. The present study was approved by the Ethics Committee of Kyoto University. The present inventors enrolled 60 patients with NSCLC, all of which had previously received other chemotherapy. Patients received Nivolumab (3 mg/kg) every 2 weeks until disease progression or the emergence of an unacceptable side effect. Blood samples were collected immediately before the 1st, 2nd and 3rd Nivolumab administration. Among the 60 patients enrolled, 6 patients dropped out and 7 patients had to cancel the therapy due to severe side effect, leaving the data of 47 patients for analysis. Tumor size was measured by CT and evaluated for response using Response Evaluation Criteria in Solid Tumors 1.1 (RECIST 1.1).
For functional analysis of PD-1high CD8+ T cell population, the present inventors enrolled 12 other patients with NSCLC receiving Nivolumab or Pembrolizumab.
Peripheral blood samples were collected in 7 ml vacutainers containing EDTA (Venoject II, VP-NA070K, Terumo, Tokyo, Japan), immediately stored in a CubeCooler (Forte Grow Medical Co. Ltd., Tochigi, Japan) and kept at 4° C. until centrifugation at 4° C. at 3000 rpm for 15 min. All the harvested plasma samples were then stored at 80° C. until analysis. For GCMS analysis, 50 μl of plasma was mixed with 256 μl of a solvent mixture (methanol:water:chloroform=2.5:1:1) containing 2.34 μg/ml of 2-isopropylmalic acid (Sigma-Aldrich), which was utilized as an internal standard. The obtained mixture was shaken at 1200 rpm for 30 min at 37° C. (Maximizer MBR-022UP, Taitec). After centrifugation at 16000×g for 5 min at 25° C., 150 μl of supernatant was collected and mixed with 140 μl of purified water followed by vortex mixing for 5 s. After centrifugation at 16,000×g for 5 min at 25° C., 180 μl of supernatant was dried in a centrifugal evaporator (CVE-3100, Tokyo Rikakikai Co. Ltd.). The dried sample was dissolved in 80 μl of methoxyamine solution (20 mg/ml in pyridine, Sigma-Aldrich) and shaken at 1,200 rpm for 30 min at 37° C. Forty microliters of N-methyl-N-trimethylsilyltrifluoroacetamide solution (GL science) were added for trimethylsilyl derivatization, followed by agitation at 1,200 rpm for 30 min at 37° C. After centrifugation, 50 μl of supernatant was transferred to a glass vial and subjected to GCMS measurement. For LCMS analysis, the metabolite extraction protocol was slightly changed. Fifty microliters of plasma was mixed with 256 μl of methanol and shaken at 1200 rpm for 10 min at 37° C. After centrifugation at 16,000×g for 30 min at 25° C., 150 μl of supernatant was mixed with 90 μl of 1% acetic acid in water and 120 μl of chloroform, followed by a vortex mixing for 15 s. After centrifugation at 2,000×g for 10 min at 25° C., 150 μl of the upper layer was dried and solubilized in 50 μl of 0.1% formic acid in water, and then subjected to LCMS analysis.
GCMS analysis was performed with a GCMS-QP2010 Ultra (Shimadzu). The derivatized metabolites were separated on a DB-5 column (30 m×0.25 mm id, film thickness 1.0 μm, Agilent Technologies). The helium carrier gas was set at a flow rate of 39 cm/sec. The inlet temperature was 280° C. and the column temperature was first held at 80° C. for 2 min, then raised at a rate of 15° C./min to 330° C., and held for 6 min.
GC-MS measurement was performed in scan mode. Metabolite peaks were identified by agreement with spectrum library and semi-quantitated using internal standards. One microliter of the sample was injected into the GCMS in the split mode (split ratio 1:3). Mass spectra were obtained under the following conditions: electron ionization (ionization voltage 70 eV), ion source temperature 200° C., full scan mode in the range of m/z 85-500, scan rate 0.3 sec/scan. Identification of chromatographic peaks was performed using the NIST library or Shimadzu GC/MS database, and further confirmed with authentic commercial standards. For semi-quantitative analysis, the area of each metabolite peak was calculated and divided by the area of the internal standard peak. LC separation was conducted on a Shim-pack GIST C18-AQ column (3 μm, 150 mm×2.1 mm id, Shimadzu GLC) with a Nexera UHPLC system (Shimadzu). The mobile phase consisted of 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B). The gradient program was as follows: 0-3 min, 0% B; 3-15 min, linear gradient to 60% B; 15-17.5 min, 95% B; 17.5-20 min, linear gradient to 0% B; hold for 4 min; flow rate, 0.2 ml/min. The column oven temperature was maintained at 40° C. The LC system was coupled with a triple-quadruple mass spectrometer LCMS-8060 (Shimadzu). LCMS-8060 was operated with the electrospray ionization and multiple reaction monitoring mode. All ion transitions and collision energies were optimized experimentally by using authentic standards of each metabolite. Three microliters of the sample were injected into the CLAMS system. Quality control (QC, a pooled plasma) samples were subjected to the same preparation protocol, and every 10 and 5 samples were injected for GEMS and CLAMS analysis, respectively. Each metabolite's signals were normalized with a QC-based correction method using the smooth-spline algorithm (57-59). Information on all the measured metabolites, including retention time, m/z, and ion transitions was summarized in Tables 1 and 2.
PBMCs were isolated from blood by Ficoll density gradient centrifugation. PBMCs were immediately stained using the following antibodies: anti-CD8a (RPA-T8), -CD8 (SK1), -CD4 (RPA-T4, SK3), -CD45RA (HI100), -CD45RO (UCHL1), -CCR7 (3D12), -PD-1 (EH12.2H7), -Tim3 (F38-2E2), -KLRG1 (13F12F2), -CD25 (BC96), -CXCR3 (G025H7), -CCR6 (G034E3), -T-bet (4B10), -EOMES (WD1928), -Ki-67(SolA15), -CTLA-4 (BNI3), -p-mTOR (MRRBY), -p-Akt1(Ser473) (SDRNR), -granzyme B (GB11), -IFN-γ (4S.B3), and -FOXP3 (236A/E7). PGC-1 expression was detected with anti-PGC-1α6 (rabbit polyclonal, Abcum, ab72230) which recognizes both PGC-1α and PGC-16, followed by secondary staining with goat anti-rabbit IgG (Santa Cruz Biotech, sc-3739). Dead cell discrimination was performed using 7-AAD staining solution (TONBO, 13-6993). Intracellular staining was performed using a FOXP3 Fixation Kit (eBioscience). For staining intracellular phosphoproteins, cells were permeabilized with 0.5% Triton-X and fixed with 1.5% paraformaldehyde before staining. Samples were measured on BD Canto II flow cytometer (BD Bioscience). Data were collected using BD FACS Diva Software version 6.1.3 and further analyzed with FlowJo 10.4 (Tree Star Inc.). For data analysis, data were gated on live (7AAD negative) and single cells. Measurement of mitochondrial mass, membrane potential, mitochondrial superoxide, and cellular ROS were performed using MitoTracker Green, MitoTracker Deep Red, MitoSOX Red, and CellROX Green reagents, respectively (all from Life Technologies). These dyes were added to cells, which were then incubated at 37° C. in a 5% CO2 humidified incubator for 30 min, followed by surface staining. As regards samples after Nivolumab administration, anti-PD-1 (EH12.2H7, APC-conjugated) antibody was added to cells, which were then incubated at 37° C. in a 5% CO2 humidified incubator for 60 min, followed by other surface staining.
Data are shown as the median and interquartile range. For comparison of two groups, a Wilcoxon rank sum test was conducted. For comparison between multiple groups, a Kruskal-Wallis test followed by Dunn's test for multiple comparison was conducted. The stepwise AIC regression procedure was performed to select the best marker combination. Then, LDA was performed by using estimated biomarker combination to predict the reliability. The prediction model was evaluated with five-fold cross validation. The survival rates of different groups of patients were calculated with the Kaplan-Meier method and presented graphically as survival curves. Comparison of survival curves between two groups was analyzed by Gehan-Breslow-Wilcoxon test. The Spearman correlation coefficient was used to calculate the association between cellular markers and metabolite markers. IMP software (Version 12.0.0; SAS Institute Inc.; Cary, N.C., USA), R software (Version 3.4.4), DataRobot (Version 4.3.0) and Prism software (Version 6.0h; GraphPad Software) were used for data management and statistical analyses. Significance level was set at 0.05 for all tests.
cells among CD4
T cells
cells among CD4
T cells
T cells among PBMC
T cells among PBMC
cells among: CD4
T cells
cells among CD4
T cells
cells among CD4* T cells
T cells
T cells
CD25
cells among CD4
T cells
cells among CD4
T cells
CTLA4
cells among CD4
T cells
cells among CD4
T cells
cells among CD8
T cells
cells among CD4
T cells
CCR
cells among CD8
T cells
cells among CD8
T cells
cells among CD4
T cells
CD45
cells among CD4
T cells
FoxP3
cells among CD4
T cells
cells among CD8
T cells
cells
cells among CD4
T cells
T cells
KLRG1
cells among CD4
T cells
cells among CD8
T cells
T cells
cells among CD8
T cells
EOMES
cells among CD8
T cells
EOMES
cells among CD8
T cells
EOMES
cells among CD8
T cells
T cells
T cells
T cells
T cells
T cells
T cells
cells
T cells
cells among CD4
T cells
cells among CD8
T cells
T cells
T cells
T cells
T cells
T cells
T cells
T cells
T cells
T cells
T cells
T cells
indicates data missing or illegible when filed
T cells among PBMC
T cells among PBMC
T cells among PBMC
T cells among PBMC
T cells
T cells
T cells
T cells
T cells
T cells
T cells
T cells
T cells
T cells
T cells
among CD4
T cells
T cells
T cells
among CD8
T cells
EOMES
among CD8
T cells
indicates data missing or illegible when filed
CD8+ T cells from age matched 30 healthy donors were gated and PD-1 staining data were overlaid (
It is a known fact that PD-1 antibody therapy does not work well when lung cancer cells have an EGFR mutation. LDA analysis was performed to determine whether the cellular marker combination II of the present invention can discriminate the non-responsiveness resulting from the presence of EGFR mutation (8 patients having an EGFR mutation were selected as subjects from the patients of Example 1). As a result, the error rate was 0%, making it clear that the cellular marker combination II is capable of discriminating the non-responsiveness due to EGFR mutation (
In order to examine whether the cellular marker combination II can also judge therapeutic efficacy in other cancer species, samples from 11 head & neck cancer patients (after or before administration of Nivolumab at Kyoto University Hospital in the second-line treatment or thereafter in the same manner as in treatment of lung cancer) were stained in the same manner and subjected to LDA analysis. As a result, efficacy could be judged at an error rate of 9.1% (
All publications, patents and patent applications cited herein are incorporated herein by reference in their entirety.
By using the biomarkers and/or cellular markers of the present invention, the efficacy of therapy with a PD-1 signal inhibitor-containing drug can be predicted and/or judged before or at an early stage of the therapy. As a result, the efficiency of therapy can be improved while reducing its cost.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2019-000181 | Jan 2019 | JP | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2019/046582 | 11/28/2019 | WO | 00 |
