Patent Application
20070218519
The present invention relates generally to the identification of biological markers associated with an increased risk of developing Diabetes, as well as methods of using such biological markers in diagnosis and prognosis of Diabetes.
Diabetes Mellitus describes a metabolic disorder characterized by chronic hyperglycemia with disturbances of carbohydrate, fat and protein metabolism that result from defects in insulin secretion, insulin action, or both. The effects of Diabetes Mellitus include long-term damage, dysfunction and failure of various organs. Diabetes may be present with characteristic symptoms such as thirst, polyuria, blurring of vision, chronic infections, slow wound healing, and weight loss. In its most severe forms, ketoacidosis or a non-ketotic hyperosmolar state may develop and lead to stupor, coma and, in the absence of effective treatment, death. Often symptoms are not severe, not recognized, or may be absent. Consequently, hyperglycemia sufficient to cause pathological and functional changes may be present for a long time, occasionally up to ten years, before a diagnosis is made, usually by the detection of high levels of glucose in urine after overnight fasting during a routine medical work-up. The long-term effects of Diabetes Mellitus include progressive development of complications such as retinopathy with potential blindness, nephropathy that may lead to renal failure, neuropathy, microvascular changes, and autonomic dysfunction. People with Diabetes are also at increased risk of cardiovascular, peripheral vascular, and cerebrovascular disease (together, “arteriovascular” disease). There is also an increased risk of cancer. Several pathogenetic processes are involved in the development of Diabetes. These include processes which destroy the insulin-secreting beta cells of the pancreas with consequent insulin deficiency, and changes in liver and smooth muscle cells that result in the resistance to insulin uptake. The abnormalities of carbohydrate, fat and protein metabolism are due to deficient action of insulin on target tissues resulting from insensitivity to insulin or lack of insulin.
Regardless of the underlying cause, Diabetes Mellitus is subdivided into Type 1 Diabetes and Type 2 Diabetes. Type 1 Diabetes results from autoimmune mediated destruction of the beta cells of the pancreas. The rate of destruction is variable, and the rapidly progressive form is commonly observed in children, but may also occur in adults. The slowly progressive form of Type 1 Diabetes generally occurs in adults and is sometimes referred to as latent autoimmune Diabetes in adults (LADA). Some patients, particularly children and adolescents, may exhibit ketoacidosis as the first manifestation of the disease. Others have modest fasting hyperglycemia that can rapidly change to severe hyperglycemia and/or ketoacidosis in the presence of infection or other stress. Still others, particularly adults, may retain residual beta cell function sufficient to prevent ketoacidosis for many years. Individuals with this form of Type 1 Diabetes often become dependent on insulin for survival and are at risk for ketoacidosis. Patients with Type 1 Diabetes exhibit little or no insulin secretion as manifested by low or undetectable levels of plasma C-peptide. However, there are some forms of Type 1 Diabetes which have no known etiology, and some of these patients have permanent insulinopenia and are prone to ketoacidosis, but have no evidence of autoimmunity. These patients are referred to as “Type 1 idiopathic.”
Type 2 Diabetes is the most common form of Diabetes and is characterized by disorders of insulin action and insulin secretion, either of which may be the predominant feature. Both are usually present at the time that this form of Diabetes is clinically manifested. Type 2 Diabetes patients are characterized with a relative, rather than absolute, insulin deficiency and are resistant to the action of insulin. At least initially, and often throughout their lifetime, these individuals do not need insulin treatment to survive. Type 2 Diabetes accounts for 90-95% of all cases of Diabetes. This form of Diabetes can go undiagnosed for many years because the hyperglycemia is often not severe enough to provoke noticeable symptoms of Diabetes or symptoms are simply not recognized. The majority of patients with Type 2 Diabetes are obese, and obesity itself may cause or aggravate insulin resistance. Many of those who are not obese by traditional weight criteria may have an increased percentage of body fat distributed predominantly in the abdominal region (visceral fat). Ketoacidosis is infrequent in this type of Diabetes and usually arises in association with the stress of another illness. Whereas patients with this form of Diabetes may have insulin levels that appear normal or elevated, the high blood glucose levels in these diabetic patients would be expected to result in even higher insulin values had their beta cell function been normal. Thus, insulin secretion is often defective and insufficient to compensate for the insulin resistance. On the other hand, some hyperglycemic individuals have essentially normal insulin action, but markedly impaired insulin secretion.
Diabetic hyperglycemia may be decreased by weight reduction, increased physical activity, and/or pharmacological treatment. There are several biological mechanisms that are associated with hyperglycemia such as insulin resistance, insulin secretion, and gluconeogenesis, and there are orally active drugs available that act on one or more of these mechanisms. With lifestyle and/or drug intervention, glucose levels can return to near-normal levels, but this is usually temporary. With time, additional second-tier drugs are often required additions to the treatment approach. Often with time, even these multi-drug approaches fail, at which point insulin injections are instituted.
Over 18 million people in the United States have Type 2 Diabetes, and of these, about 5 million do not know they have the disease. These persons who do not know they have the disease and who do not exhibit the classic symptoms of Diabetes present a major diagnostic and therapeutic challenge.
There is a large group in the United States, nearly 41 million persons, who are at significant risk of developing Type 2 Diabetes. They are broadly referred to in the literature as “pre-diabetics.” A “pre-diabetic” or a subject with pre-Diabetes represents any person or population with a greater risk than the broad population for conversion to Type 2 Diabetes in a given period of time. The risk of developing Type 2 Diabetes increases with age, obesity, and lack of physical activity. It occurs more frequently in women with prior gestational Diabetes, and in individuals with hypertension and/or dyslipidemia. Its frequency varies in different ethnic subgroups. Type 2 Diabetes is often associated with strong familial, likely genetic, predisposition, however the genetics of this form of Diabetes are complex and not clearly defined.
Pre-diabetics often have fasting glucose levels between normal and frank diabetic levels. Occasionally in research, these persons are tested for their tolerance to glucose. Abnormal glucose tolerance, or “impaired glucose tolerance” can be an indication that an individual is on the path toward Diabetes; it requires the use of a 2-hour oral glucose tolerance test for its detection. However, it has been shown that impaired glucose tolerance is by itself entirely asymptomatic and unassociated with any functional disability. Indeed, insulin secretion is typically greater in response to a mixed meal than in response to a pure glucose load; as a result, most persons with impaired glucose tolerance are rarely, if ever, hyperglycemic in their daily lives, except when they undergo diagnostic glucose tolerance tests. Thus, the importance of impaired glucose tolerance resides exclusively in its ability to identify persons at increased risk of future disease (Stem et al, 2002). In studies conducted by Stem and others, the sensitivity and false-positive rates of impaired glucose tolerance as a predictor of future conversion to Type 2 Diabetes was 50.9% and 10.2%, respectively, representing an area under the Receiver-Operating Characteristic Curve of 77.5% and a p-value of 0.20. Because of its cost, reliability, and inconvenience, the oral glucose tolerance test is seldom used in routine clinical practice. Moreover, patients whose Diabetes is diagnosed solely on the basis of an oral glucose tolerance test have a high rate of reversion to normal on follow-up and may in fact represent false-positive diagnoses. Stem and others reported that such cases were almost 5 times more likely to revert to non-diabetic status after 7 to 8 years of follow-up compared with persons meeting conventional fasting or clinical diagnostic criteria. Clearly, there is a need for improved methods of assessing the risk of future Diabetes.
Often a person with impaired glucose tolerance will be found to have at least one or more of the common arteriovascular disease risk factors. This clustering has been termed “Syndrome X,” or “Metabolic Syndrome” by some researchers and can be indicative of a pre-diabetic state. Alone, each component of the cluster conveys increased arteriovascular and diabetic disease risk, but together as a combination they become much more significant. This means that the management of persons with hyperglycemia and other features of Metabolic Syndrome should focus not only on blood glucose control but also include strategies for reduction of other arteriovascular disease risk factors. Furthermore, such risk factors are non-specific for Diabetes or pre-Diabetes and are not in themselves a basis for a diagnosis of Diabetes, or of diabetic status.
It should furthermore be noted that an increased risk of conversion to Diabetes implies an increased risk of converting to arteriovascular disease and events. Diabetes itself is one of the most significant single risk factors for arteriovascular disease, and is in fact often termed a “coronary heart disease equivalent” by itself, indicating a greater than 20 percent ten-year risk of an arteriovascular event, in a similar range with stable angina and just below the most significant independent risk factors, such as survivorship of a previous arteriovascular event. The same is true of other arteriovascular disease, such as peripheral artery disease or cerebrovascular disease.
It is well documented that pre-Diabetes can be present for ten or more years before the detection of glycemic disorders like Diabetes. Treatment of pre-diabetics with drugs such as acarbose, metformin, troglitazone and rosiglitazone can postpone or prevent Diabetes; yet few pre-diabetics are treated. A major reason, as indicated above, is that no simple laboratory test exists to determine the actual risk of an individual to develop Diabetes. Thus, there remains a need in the art for methods of identifying and diagnosing these individuals who are not yet diabetics, but who are at significant risk of developing Diabetes.
The present invention relates in part to the discovery that certain biological markers, such as proteins, nucleic acids, polymorphisms, metabolites, and other analytes, as well as certain physiological conditions and states, are present in subjects with an increased risk of developing Diabetes Mellitus or a pre-diabetic condition such as, but not limited to, Metabolic Syndrome (Syndrome X), conditions characterized by impaired glucose regulation and/or insulin resistance, such as Impaired Glucose Tolerance (IGT) and Impaired Fasting Glycemia (IFG), but where such subjects do not exhibit some or all of the conventional risk factors of these conditions, or subjects who are asymptomatic for these conditions.
Accordingly, the invention provides biological markers of Diabetes or pre-diabetic conditions that can be used to monitor or assess the risk of subjects experiencing such diabetic or pre-diabetic conditions, to diagnose or identify subjects with a diabetic or pre-diabetic condition, to monitor the risk for development of a diabetic or pre-diabetic condition, to monitor subjects that are undergoing therapies for Diabetes or a pre-diabetic condition, to differentially diagnose disease states associated with Diabetes or a pre-diabetic condition from other diseases, or within sub-classifications of Diabetes or pre-diabetic conditions, to evaluate changes in the risk of Diabetes or pre-diabetic conditions, and to select therapies for use in treating subjects with Diabetes or a pre-diabetic condition, or for use in treating subjects who are at risk for developing Diabetes or a pre-diabetic condition. Preferably, the present invention provides use of biological markers, some of which are unrelated to Diabetes or have not heretofore been identified as related to Diabetes, but are related to early biological changes that can lead to the development of Diabetes or a pre-diabetic condition, to detect and identify subjects who exhibit none of the symptoms for Diabetes, i.e., who are asymptomatic for Diabetes or pre-diabetic conditions or have only non-specific indivators of potential pre-diabetic conditions, such as arteriovascular risk factors, or who exhibit none or few of the conventional risk factor of Diabetes. Significantly, many of the biomarkers disclosed herein have shown little individual significance in the diagnosis of Diabetes, but when used in combination (in “panels”) with other disclosed markers and combined with the herein disclosed mathematical classification algorithms, becomes significant discriminates of the pre-Diabetes patient or population from one who is not pre-diabetic.
Accordingly, in one aspect, the present invention provides a method with a predetermined level of predictability for assessing a risk of development of Diabetes Mellitus or a pre-diabetic condition in a subject comprising: measuring the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a sample from the subject, and measuring a clinically significant alteration in the level of the one or more, preferably two or more DBRISKMARKERS in the sample, wherein the alteration indicates an increased risk of developing Diabetes Mellitus or a pre-diabetic condition in the subject.
In one embodiment, the Diabetes Mellitus comprises Type 1 Diabetes, Type 2 Diabetes, or gestational Diabetes. In other embodiments, the pre-diabetic condition comprises IFG, IGT, Metabolic Syndrome, or Syndrome X.
The level of DBRISKMARKERS can be measured electrophoretically or immunochemically. Where the detection is immunochemical, the detection can be by radioimmunoassay, immunofluorescence assay or by an enzyme-linked immunosorbent assay. The detection can also be achieved by specific oligonucleotide hybridization.
In some embodiments, the subject has not been previously diagnosed or identified as having the Diabetes Mellitus or the pre-diabetic condition. In other embodiments, the subject is asymptomatic for the Diabetes Mellitus or the pre-diabetic condition.
The sample as defined by the present invention can be serum, blood plasma, blood cells, endothelial cells, tissue biopsies, ascites fluid, bone marrow, interstitial fluid, sputum, or urine.
In one embodiment of the present invention, the level of expression of five or more DBRISKMARKERS is measured, but can also encompass measurement of ten or more, twenty-five or more, or fifty or more DBRISKMARKERS.
In another aspect, a method with a predetermined level of predictability for diagnosing or identifying a subject having Diabetes Mellitus or a pre-diabetic condition is provided, comprising measuring the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a sample from the subject, and comparing the level of the effective amount of the one or more (or two or more) DBRISKMARKERS to a reference value.
In one embodiment, the reference value is an index value. The reference value can also be derived from one or more risk prediction algorithms or computed indices for the Diabetes or pre-diabetic condition.
Another aspect of the present invention provides a method with a predetermined level of predictability for assessing a risk of impaired glucose tolerance in a subject comprising measuring the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a sample from the subject, and measuring a clinically significant alteration in the level of the one or more (or two or more) DBRISKMARKERS in the sample, wherein the alteration indicates an increased risk of impaired glucose tolerance in the subject.
In one embodiment, the subject has not been previously diagnosed as having impaired glucose tolerance. In another embodiment, the subject is asymptomatic for the impaired glucose tolerance.
In another aspect, a method with a predetermined level of predictability for diagnosing or identifying a subject having impaired glucose tolerance is provided, comprising measuring the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a sample from the subject, and comparing the level of the effective amount of the one or more (preferably two or more) DBRISKMARKERS to a reference value. The reference value can be an index value.
Alternatively, the reference value can be derived from one or more risk prediction algorithms or computed indices for impaired glucose tolerance.
Another aspect of the invention provides a method with a predetermined level of predictability for assessing the progression of Diabetes Mellitus or a pre-diabetic condition in a subject, comprising detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a first sample from the subject at a first period of time; detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS in a second sample from the subject at a second period of time; and comparing the level of the effective amount of the one or more (or two or more) DBRISKMARKERS detected in step (a) to the amount detected in step (b), or to a reference value.
In one embodiment, the subject has previously been diagnosed or identified as suffering from the Diabetes Mellitus or the pre-diabetic condition. In another embodiment, the subject has previously been treated for the Diabetes Mellitus or the pre-diabetic condition. In yet another embodiment, the subject has not been previously diagnosed or identified as suffering from the Diabetes Mellitus or the pre-diabetic condition. In other embodiments, the subject is asymptomatic for the Diabetes Mellitus or the pre-diabetic condition.
In the context of the invention, the first sample can be taken from the subject prior to being treated for the Diabetes Mellitus or the pre-diabetic condition. The second sample can taken from the subject after being treated for the Diabetes Mellitus or the pre-diabetic condition. The reference value can be derived from one or more subjects who have suffered from Diabetes Mellitus or a pre-diabetic condition.
In another aspect of the present invention, a method with a predetermined level of predictability for assessing the progression of impaired glucose tolerance associated with Diabetes Mellitus or a pre-diabetic condition in a subject is provided, comprising detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a first sample from the subject at a first period of time; detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS in a second sample from the subject at a second period of time; and comparing the level of the effective amount of the one or more (or two or more) DBRISKMARKERS detected in step (a) to the amount detected in step (b), or to a reference value.
The subject can be one who has previously been treated for the Diabetes Mellitus or the pre-diabetic condition. The subject can also be one who has not been previously diagnosed or identified as having impaired glucose tolerance or suffering from the Diabetes Mellitus or the pre-diabetic condition. Alternatively, the subject can be asymptomatic for the impaired glucose tolerance, or is asymptomatic for the Diabetes Mellitus or the pre-diabetic condition.
In yet another aspect, a method with a predetermined level of predictability for monitoring the effectiveness of treatment for Diabetes Mellitus or a pre-diabetic condition is provided, comprising detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a first sample from the subject at a first period of time; detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS in a second sample from the subject at a second period of time; and comparing the level of the effective amount of the one or more (or two or more) DBRISKMARKERS detected in step (a) to the amount detected in step (b), or to a reference value, wherein the effectiveness of treatment is monitored by a change in the level of the effective amount of one or more, preferably two or more DBRISKMARKERS from the subject.
In one embodiment, the treatment for the Diabetes Mellitus or the pre-diabetic condition comprises exercise regimens, dietary supplements, therapeutic agents, surgical intervention, and prophylactic agents. In another embodiment, the reference value is derived from one or more subjects who show an improvement in Diabetes risk factors as a result of one or more treatments for the Diabetes Mellitus or the pre-diabetic condition. The effectiveness of treatment can be additionally monitored by detecting changes in body mass index (BMI), insulin levels, blood glucose levels, HDL levels, systolic and/or diastolic blood pressure, or combinations thereof. Changes in blood glucose levels can be detected by an oral glucose tolerance test.
Another aspect of the present invention provides a method with a predetermined level of predictability for selecting a treatment regimen for a subject diagnosed with or at risk for Diabetes Mellitus or a pre-diabetic condition comprising detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a first sample from the subject at a first period of time; optionally detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS in a second sample from the subject at a second period of time; and comparing the level of the effective amount of the one or more (or two or more) DBRISKMARKERS detected in step (a) to a reference value, or optionally, to the amount detected in step (b).
The present invention also provides a Diabetes Mellitus reference expression profile, comprising a pattern of marker levels of an effective amount of one or more, preferably two or more markers selected from the group consisting of DBRISKMARKERS1-260, taken from one or more subjects who do not have the Diabetes Mellitus.
An impaired glucose tolerance reference expression profile is also provided by the invention, comprising a pattern of marker levels of an effective amount of one or more, preferably two or more markers selected from the group consisting of DBRISKMARKERS1-260, taken from one or more subjects who do not have impaired glucose tolerance.
In another aspect, a Diabetes Mellitus subject expression profile is provided, comprising a pattern of marker levels of an effective amount of one or more, preferably two or more markers selected from the group consisting of DBRISKMARKERS1-260 taken from one or more subjects who have the Diabetes Mellitus, are at risk for developing the Diabetes Mellitus, or are being treated for the Diabetes Mellitus.
In another aspect, an impaired glucose tolerance subject expression profile is provided, comprising a pattern of marker levels of an effective amount of one or more, preferably two or more markers selected from the group consisting of DBRISKMARKERS1-260 taken from one or more subjects who have impaired glucose tolerance, are at risk for developing impaired glucose tolerance, or are being treated for impaired glucose tolerance.
The present invention also provides a kit comprising a plurality of DBRISKMARKER detection reagents that detect the corresponding DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260, sufficient to generate the profiles of the invention. The detection reagent can comprise one or more antibodies or fragments thereof. Alternatively, or additionally, the detection reagent can comprise one or more oligonucleotides or one or more aptamers.
The present invention also provides, in another aspect, a machine readable media containing one or more Diabetes Mellitus reference expression profiles according to the invention, or one or more Diabetes Mellitus subject expression profiles according to the invention, and optionally, additional test results and subject information.
A machine readable media containing one or more impaired glucose tolerance reference expression profiles according to invention is also contemplated, or one or more impaired glucose tolerance subject expression profiles according to the invention, and optionally, additional test results and subject information.
In another aspect, a DBRISKMARKER panel comprising one or more DBRISKMARKERS that are indicative of a physiological and/or biochemical pathway associated with Diabetes Mellitus or a pre-diabetic condition is provided. In one embodiment, the physiological and biochemical pathways comprise autoimmune regulation, inflammation and endothelial function (including cytokine-cytokine receptor interactions, cell adhesion molecules (CAMs), focal adhesions, leukocyte transendothelial migration, natural killer cell mediated cytotoxicity, regulation of the actin cytoskeleton, adherens/tight/gap junctions, and extracellular matrix (ECM)-receptor interaction), adipocyte development and maintenance (including adipocytokines, cell cycle, apoptosis, and neuroactive ligand-receptor interaction) as well as hematopoietic cell lineage, complement and coagulation cascades, intra- and extracellular cell signaling pathways (including the mTOR, TGF-β, MAPK, insulin, GnRH, Toll-like receptor, Jak-STAT, PPAR, T-cell receptor, B-cell receptor, FcεRI, calcium, Wnt, and VEGF signaling pathways and other cell communication mechanisms), in addition to those pathways that are commonly associated with Type 1 and Type 2 Diabetes Mellitus.
A DBRISKMARKER panel comprising one or more DBRISKMARKERS that are indicative of a site associated with Diabetes Mellitus or a pre-diabetic condition is also provided, wherein the site can comprise beta cells, endothelial cells, skeletal and smooth muscle, or peripheral, cardiovascular, or cerebrovascular arteries.
In other aspects, a DBRISKMARKER panel comprising one or more DBRISKMARKERS that are indicative of the progression of Diabetes Mellitus or a pre-diabetic condition is provided.
The present invention further provides a DBRISKMARKER panel comprising one or more DBRISKMARKERS that are indicative of the speed of progression of Diabetes Mellitus or a pre-diabetic condition. The invention also concerns a DBRISKMARKER panel comprising one or more DBRISKMARKERS that are specific to one or more types of Diabetes Mellitus and a DBRISKMARKER panel comprising one or more DBRISKMARKERS that are specific to a pre-diabetic condition.
A DBRISKMARKER panel comprising one or more DBRISKMARKERS selected from mathematical classification algorithms and factor analysis approach is provided, utilizing a relevant past cohort of subjects, or calculated indices which were developed in such past cohorts. In particular, a DBRISKMARKER panel of one or more, preferably two or more DBRISKMARKERS selected from a subset of the disclosed DBRISKMARKERS comprising Leptin (LEP), Haptoglobin (HP), Insulin-like growth factor binding protein 3 (ILGFBP3), Resistin (RETN), Matrix Metallopeptidase 2 (MMP-2), Angiotensin I converting enzyme (peptidyl dipeptidase A)-1 (ACE), complement component 4A (Rogers blood group)(C4A), CD14 molecule (CD14), selectin E (endothelial adhesion molecule)(SELE), colony stimulating factor 1 (macrophage) (CSF1), and vascular endothelial growth factor (VEGF), c-reactive protein, pentraxin-related (CRP), Tumor Necrosis Factor Receptor Superfamily Member 1A (TNFRSF1A), RAGE (Advanced Glycosylation End Product-specific Receptor [AGER]), CD26 (dipeptidyl peptidase 4; DPP4), and their statistical and/or functional equivalents within mathematical classification algorithms using one or more of these DBRISKMARKERS.
A method for treating one or more subjects at risk for developing Diabetes Mellitus or a pre-diabetic condition is also contemplated by the present invention, comprising detecting the presence of increased levels of at least one, preferably two different DBRISKMARKERS present in a sample from the one or more subjects; and treating the one or more subjects with one or more Diabetes-modulating drugs until altered levels of the at least one, preferably two different DBRISKMARKERS return to a baseline value measured in one or more subjects at low risk for developing the Diabetes Mellitus or the pre-diabetic condition, or a baseline value measured in one or more subjects who show improvements in Diabetes risk markers as a result of treatment with one or more Diabetes-modulating drugs.
The Diabetes-modulating drugs can comprise sulfonylureas; biguanides; insulin, insulin analogs; peroximsome proliferator-activated receptor-γ (PPAR-γ) agonists; dual-acting PPAR agonists; insulin secretagogues; analogs of glucagon-like peptide-1 (GLP-1); inhibitors of dipeptidyl peptidase IV (DPP4); pancreatic lipase inhibitors; α-glucosidase inhibitors; and combinations thereof. In one embodiment, the improvements in Diabetes risk markers as a result of treatment with one or more Diabetes-modulating drugs comprise a reduction in body mass index (BMI), a reduction in blood glucose levels, an increase in insulin levels, an increase in HDL levels, a reduction in systolic and/or diastolic blood pressure, or combinations thereof.
In another aspect, a method of evaluating changes in the risk of impaired glucose tolerance in a subject diagnosed with or at risk for developing a pre-diabetic condition is provided, comprising detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a first sample from the subject at a first period of time; optionally detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS in a second sample from the subject at a second period of time; and comparing the level of the effective amount of the one or more (or two or more) DBRISKMARKERS detected in step (a) to a reference value, or optionally, the amount in step (b).
The present invention further provides a method of differentially diagnosing disease states associated with Diabetes Mellitus or a pre-diabetic condition in a subject comprising detecting the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a sample from the subject; and comparing the level of the effective amount of the one or more (or two or more) DBRISKMARKERS detected in step (a) to the Diabetes Mellitus disease subject expression profile of the invention, to the impaired glucose tolerance subject expression profile of the invention, or to a reference value.
Further, in a method of diagnosing or identifying a subject at risk for developing Diabetes or a pre-diabetic condition by analyzing Diabetes risk factors, the present invention provides an improvement comprising measuring the level of an effective amount of one or more, preferably two or more DBRISKMARKERS selected from the group consisting of DBRISKMARKERS1-260 in a sample from the subject, and measuring a clinically significant alteration in the level of the one or more (or two or more) DBRISKMARKERS in the sample, wherein the alteration indicates an increased risk of developing Diabetes Mellitus or a pre-diabetic condition in the subject.
In yet another aspect of the present invention, in a method of diagnosing or identifying a subject at risk for developing Diabetes or a pre-diabetic condition by analyzing Diabetes risk factors, the present invention provides an improvement comprising: measuring the level of an effective amount of one or more DBRISKMARKERS selected from the group consisting of: Leptin (LEP), Haptoglobin (HP), Insulin-like growth factor binding protein 3 (ILGFBP3), Resistin (RETN), Matrix Metallopeptidase 2 (MMP-2), Angiotensin I converting enzyme (peptidyl dipeptidase A)-1 (ACE), complement component 4A (C4A), CD14 molecule (CD14), selectin E (SELE), colony stimulating factor 1 (macrophage; CSF1), and vascular endothelial growth factor (VEGF), c-reactive protein, pentraxin-related (CRP), Tumor Necrosis Factor Receptor Superfamily Member 1A (TNFRSF1A), RAGE (Advanced Glycosylation End Product-specific Receptor [AGER]), and CD26 (dipeptidyl peptidase 4; DPP4), and measuring a clinically significant alteration in the level of the one or more DBRISKMARKERS in the sample, wherein the alteration indicates an increased risk of developing Diabetes Mellitus or a pre-diabetic condition in the subject.
In a method of diagnosing or identifying a subject at risk for developing Diabetes or a pre-diabetic condition by analyzing Diabetes risk factors, the present invention provides an improvement comprising: measuring the level of an effective amount of two or more DBRISKMARKERS selected from the group consisting of: Leptin (LEP), Haptoglobin (HP), Insulin-like growth factor binding protein 3 (ILGFBP3), Resistin (RETN), Matrix Metallopeptidase 2 (MMP-2), Angiotensin I converting enzyme (peptidyl dipeptidase A)-1 (ACE), complement component 4A (C4A), CD14 molecule (CD14), selectin E (SELE), colony stimulating factor 1 (macrophage; CSF1), and vascular endothelial growth factor (VEGF), c-reactive protein, pentraxin-related (CRP), Tumor Necrosis Factor Receptor Superfamily Member 1A (TNFRSF1A), RAGE (Advanced Glycosylation End Product-specific Receptor [AGER]), and CD26 (dipeptidyl peptidase 4; DPP4), and measuring a clinically significant alteration in the level of the two or more DBRISKMARKERS in the sample, wherein the alteration indicates an increased risk of developing Diabetes Mellitus or a pre-diabetic condition in the subject.
Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice of the present invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are expressly incorporated by reference in their entirety. In cases of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples described herein are illustrative only and are not intended to be limiting.
Other features and advantages of the invention will be apparent from and are encompassed by the following detailed description and claims.
The following Detailed Description, given by way of example, but not intended to limit the invention to specific embodiments described, may be understood in conjunction with the accompanying Figures, incorporated herein by reference, in which:
FIGS. 3A-3RR are graphic illustrations of the KEGG pathways highlighting three or more DBRISKMARKERS in each disclosed pathway.
The present invention relates to the identification of biomarkers associated with subjects having Diabetes or a pre-diabetic condition, or who are pre-disposed to developing Diabetes or a pre-diabetic condition. Accordingly, the present invention features methods for identifying subjects who are pre-disposed to developing Diabetes or a pre-diabetic condition, including those subjects who are asymptomatic for Diabetes or a pre-diabetic condition by detection of the biomarkers disclosed herein. These biomarkers are also useful for monitoring subjects undergoing treatments and therapies for Diabetes or pre-diabetic conditions, and for selecting therapies and treatments that would be efficacious in subjects having Diabetes or a pre-diabetic condition, wherein selection and use of such treatments and therapies slow the progression of Diabetes or pre-diabetic conditions, or substantially delay or prevent its onset.
“Diabetes Mellitus” in the context of the present invention encompasses Type 1 Diabetes, both autoimmune and idiopathic and Type 2 Diabetes (together, “Diabetes”). The World Health Organization defines the diagnostic value of fasting plasma glucose concentration to 7.0 mmol/l (126 mg/dl) and above for Diabetes Mellitus (whole blood 6.1 mmol/l or 110 mg/dl), or 2-hour glucose level ≧11.1 mmol/L (≧200 mg/dL). Other values suggestive of or indicating high risk for Diabetes Mellitus include elevated arterial pressure ≧140/90 mm Hg; elevated plasma triglycerides (≧1.7 mmol/L; 150 mg/dL) and/or low HDL-cholesterol (<0.9 mmol/L, 35 mg/dl for men; <1.0 mmol/L, 39 mg/dL women); central obesity (males:waist to hip ratio>0.90; females:waist to hip ratio>0.85) and/or body mass index exceeding 30 kg/m2; microalbuminuria, where the urinary albumin excretion rate ≧20 μg/min or albumin:creatinine ratio ≧30 mg/g).
“Pre-diabetic condition” refers to a metabolic state that is intermediate between normal glucose homeostasis and metabolism and states seen in frank Diabetes Mellitus. Pre-diabetic conditions include, without limitation, Metabolic Syndrome (“Syndrome X”), Impaired Glucose Tolerance (IGT), and Impaired Fasting Glycemia (IFG). IGT refers to post-prandial abnormalities of glucose regulation, while IFG refers to abnormalities that are measured in a fasting state. The World Health Organization defines values for IFG as a fasting plasma glucose concentration of 6.1 mmol/L (100 mg/dL) or greater (whole blood 5.6 mmol/L; 100 mg/dL), but less than 7.0 mmol/L (126 mg/dL)(whole blood 6.1 mmol/L; 110 mg/dL). Metabolic syndrome according to the National Cholesterol Education Program (NCEP) criteria are defined as having at least three of the following: blood pressure ≧130/85 mm Hg; fasting plasma glucose ≧6.1 mmol/L; waist circumference >102 cm (men) or >88 cm (women); triglycerides ≧1.7 mmol/L; and HDL cholesterol <1.0 mmol/L (men) or 1.3 mmol/L (women).
“Pre-Diabetes” in the context of the present invention indicates the physiological state, in an individual or in a population, of having a higher than normal expected rate of disease conversion to frank Type 2 diabetes mellitus. Such absolute expected rate of conversion to frank Type 2 diabetes in Pre-Diabetes populations may be up to 1 percent or more per annum, and preferably 2 percent per annum or more. It may also be stated in terms of a relative risk from normal between quartiles of risk or as a likelihood ratio between differing biomarker and index scores, including those coming from the invention. Unless otherwise noted, and without limitation, when a categorical positive diagnosis of Pre-Diabetes is stated here, it is defined experimentally by the group of patients with an expected conversion rate to Type 2 Diabetes of two percent (2%) per annum over the coming 7.5 years, or fifteen percent (15%) of those testing at a given threshold value (the selected Pre-Diabetes clinical cutoff). When a continuous measure of Pre-Diabetes conversion risk is produced, having a “pre-diabetic condition” encompasses any expected annual rate of conversion above that seen in a normal reference or general unselected normal prevalence population.
“Impaired glucose tolerance” (IGT) is defined as having a blood glucose level that is higher than normal, but not high enough to be classified as Diabetes Mellitus. A subject with IGT will have two-hour glucose levels of 140 to 199 mg/dL (7.8 to 11.0 mmol) on the 75-g oral glucose tolerance test. These glucose levels are above normal but below the level that is diagnostic for Diabetes. Subjects with impaired glucose tolerance or impaired fasting glucose have a significant risk of developing Diabetes and thus are an important target group for primary prevention.
“Insulin resistance” refers to a condition in which the cells of the body become resistant to the effects of insulin, that is, the normal response to a given amount of insulin is reduced. As a result, higher levels of insulin are needed in order for insulin to exert its effects.
“Normal glucose levels” is used interchangeably with the term “normoglycemic” and refers to a fasting venous plasma glucose concentration of less than 6.1 mmol/L (110 mg/dL). Although this amount is arbitrary, such values have been observed in subjects with proven normal glucose tolerance, although some may have IGT as measured by oral glucose tolerance test (OGTT).
Two hundred and sixty biomarkers have been identified as being found to have altered or modified presence or concentration levels in subjects who have Diabetes, or who exhibit symptoms characteristic of a pre-diabetic condition, or have Pre-Diabetes (as defined herein) such as those subjects who are insulin resistant, have altered beta cell function or at risk of developing Diabetes based upon known clinical parameters or risk factors, such as family history of Diabetes, low activity level, poor diet, excess body weight (especially around the waist), age greater than 45 years, high blood pressure, high levels of triglycerides, HDL cholesterol of less than 35, previously identified impaired glucose tolerance, previous Diabetes during pregnancy (“gestational Diabetes Mellitus”) or giving birth to a baby weighing more than nine pounds, and ethnicity.
The biomarkers and methods of the present invention allow one of skill in the art to identify, diagnose, or otherwise assess those subjects who do not exhibit any symptoms of Diabetes or a pre-diabetic condition, but who nonetheless may be at risk for developing Diabetes or experiencing symptoms characteristic of a pre-diabetic condition.
The term “biomarker” in the context of the present invention encompasses, without limitation, proteins, nucleic acids, polymorphisms of proteins and nucleic acids, elements, metabolites, and other analytes. Biomarkers can also include mutated proteins or mutated nucleic acids. The term “analyte” as used herein can mean any substance to be measured and can encompass electrolytes and elements, such as calcium. Finally, biomarkers can also refer to non-analyte physiological markers of health status encompassing other clinical characteristics such as, without limitation, age, ethnicity, diastolic and systolic blood pressure, body-mass index, and resting heart rate.
Proteins, nucleic acids, polymorphisms, and metabolites whose levels are changed in subjects who have Diabetes or a pre-diabetic condition, or are predisposed to developing Diabetes or a pre-diabetic condition are summarized in Table 1 and are collectively referred to herein as, inter alia, “Diabetes risk-associated proteins”, “DBRISKMARKER polypeptides”, or “DBRISKMARKER proteins”. The corresponding nucleic acids encoding the polypeptides are referred to as “Diabetes risk-associated nucleic acids”, “Diabetes risk-associated genes”, “DBRISKMARKER nucleic acids”, or “DBRISKMARKER genes”. Unless indicated otherwise, “DBRISKMARKER”, “Diabetes risk-associated proteins”, “Diabetes risk-associated nucleic acids” are meant to refer to any of the sequences disclosed herein. The corresponding metabolites of the DBRISKMARKER proteins or nucleic acids can also be measured, as well as any of the aforementioned conventional risk marker metabolites previously disclosed, including, without limitation, such metabolites as dehydroepiandrosterone sulfate (DHEAS); c-peptide; cortisol; vitamin D3; 5-hydroxytryptamine (5-HT; serotonin); oxyntomodulin; estrogen; estradiol; and digitalis-like factor, herein referred to as “DBRISKMARKER metabolites”. Non-analyte physiological markers of health status (e.g., such as age, ethnicity, diastolic or systolic blood pressure, body-mass index, and other non-analyte measurements commonly used as conventional risk factors) are referred to as “DBRISKMARKER physiology”. Calculated indices created from mathematically combining measurements of one or more, preferably two or more of the aforementioned classes of DBRISKMARKERS are referred to as “DBRISKMARKER indices”. Proteins, nucleic acids, polymorphisms, mutated proteins and mutated nucleic acids, metabolites, and other analytes are, as well as common physiological measurements and indices constructed from any of the preceding entities, are included in the broad category of “DBRISKMARKERS”.
A “subject” in the context of the present invention is preferably a mammal. The mammal can be a human, non-human primate, mouse, rat, dog, cat, horse, or cow, but are not limited to these examples. Mammals other than humans can be advantageously used as subjects that represent animal models of Diabetes Mellitus or pre-Diabetes conditions. A subject can be male or female. A subject can be one who has been previously diagnosed or identified as having Diabetes or a pre-diabetic condition, and optionally has already undergone treatment for the Diabetes or pre-diabetic condition. Alternatively, a subject can also be one who has not been previously diagnosed as having Diabetes or a pre-diabetic condition. For example, a subject can be one who exhibits one or more risk factors for Diabetes or a pre-diabetic condition, or a subject who does not exhibit Diabetes risk factors, or a subject who is asymptomatic for Diabetes or pre-Diabetes. A subject can also be one who is suffering from or at risk of developing Diabetes or a pre-diabetic condition.
A “sample” in the context of the present invention is a biological sample isolated from a subject and can include, for example, serum, blood plasma, blood cells, endothelial cells, tissue biopsies, lymphatic fluid, ascites fluid, interstitital fluid (also known as “extracellular fluid” and encompasses the fluid found in spaces between cells, including, inter alia, gingival crevicular fluid), bone marrow, sputum, or urine.
One or more, preferably two or more DBRISKMARKERS can be detected in the practice of the present invention. For example, two (2), five (5), ten (10), fifteen (15), twenty (20), forty (40), fifty (50), seventy-five (75), one hundred (100), one hundred and twenty five (125), one hundred and fifty (150), one hundred and seventy-five (175), two hundred (200), two hundred and ten (210), two hundred and twenty (220), two hundred and thirty (230), two hundred and forty (240), two hundred and fifty (250) or more DBRISKMARKERS can be detected. In some aspects, all 260 DBRISKMARKERS disclosed herein can be detected. Preferred ranges from which the number of DBRISKMARKERS can be detected include ranges bounded by any minimum selected from between one and 260, particularly two, five, ten, twenty, fifty, seventy-five, one hundred, one hundred and twenty five, one hundred and fifty, one hundred and seventy-five, two hundred, two hundred and ten, two hundred and twenty, two hundred and thirty, two hundred and forty, two hundred and fifty, paired with any maximum up to the total known DBRISKMARKERS, particularly five, ten, twenty, fifty, and seventy-five. Particularly preferred ranges include two to five (2-5), two to ten (2-10), two to fifty (2-50), two to seventy-five (2-75), two to one hundred (2-100), five to ten (5-10), five to twenty (5-20), five to fifty (5-50), five to seventy-five (5-75), five to one hundred (5-100), ten to twenty (10-20), ten to fifty (10-50), ten to seventy-five (10-75), ten to one hundred (10-100), twenty to fifty (20-50), twenty to seventy-five (20-75), twenty to one hundred (20-100), fifty to seventy-five (50-75), fifty to one hundred (50-100), one hundred to one hundred and twenty-five (100-125), one hundred and twenty-five to one hundred and fifty (125-150), one hundred and fifty to one hundred and seventy five (150-175), one hundred and seventy-five to two hundred (175-200), two hundred to two hundred and ten (200-210), two hundred and ten to two hundred and twenty (210-220), two hundred and twenty to two hundred and thirty (220-230), two hundred and thirty to two hundred and forty (230-240), two hundred and forty to two hundred and fifty (240-250), and two hundred and fifty to more than two hundred and fifty (250+).
Diagnostic and Prognostic Methods
The risk of developing Diabetes or Pre-Diabetes can be detected with a “pre-determined level of predictability” by examining an “effective amount” of DBRISKMARKER proteins, nucleic acids, polymorphisms, metabolites, and other analytes in a test sample (e.g., a subject derived sample) and comparing the effective amounts to reference or index values, often utilizing mathematical algorithms in order to combine information from results of multiple individual DBRISKMARKERS into a single measurement or index. Subjects identified as having an increased risk of Diabetes or a pre-diabetic condition can optionally be selected to receive treatment regimens, such as administration of prophylactic or therapeutic compounds such as “diabetes-modulating drugs” as defined herein, or implementation of exercise regimens or dietary supplements to prevent or delay the onset of Diabetes or Pre-Diabetes. A sample isolated from the subject can comprise, for example, blood, plasma, blood cells, endothelial cells, tissue biopsies, lymphatic fluid, serum, bone marrow, ascites fluid, interstitial fluid (including, for example, gingival crevicular fluid), urine, sputum, or other bodily fluids.
The amount of the DBRISKMARKER protein, nucleic acid, polymorphism, metabolite, or other analyte can be measured in a test sample and compared to the normal control level. The term “normal control level”, means the level of a DBRISKMARKER protein, nucleic acid, polymorphism, metabolite, or other analyte, or DBRISKMARKER physiology or indices, typically found in a subject not suffering from Diabetes or a pre-diabetic condition and not likely to have Diabetes or a pre-diabetic condition, e.g., relative to samples collected from young subjects who were monitored until advanced age and were found not to develop Diabetes or a pre-diabetic condition. Alternatively, the normal control level can mean the level of a DBRISKMARKER protein, nucleic acid, polymorphism, metabolite, or other analyte typically found in a subject suffering from Diabetes or a pre-diabetic condition. The normal control level can be a range or an index. Alternatively, the normal control level can be a database of patterns from previously tested subjects. A change in the level in the subject-derived sample of a DBRISKMARKER protein, nucleic acid, polymorphism, metabolite, or other analyte compared to the normal control level can indicate that the subject is suffering from or is at risk of developing Diabetes or a pre-diabetic condition. In contrast, when the methods are applied prophylactically, a similar level compared to the normal control level in the subject-derived sample of a DBRISKMARKER protein, nucleic acid, polymorphism, metabolite, or other analyte can indicate that the subject is not suffering from, is not at risk or is at low risk of developing Diabetes or a pre-diabetic condition.
The difference in the level of DBRISKMARKERS is preferably statistically significant. By “statistically significant”, it is meant that the alteration is greater than what might be expected to happen by chance alone. Statistical significance can be determined by any method known in the art. For example, statistical significance can be determined by p-value. The p-value is a measure of probability that a difference between groups during an experiment happened by chance. (P(z>zobserved)). For example, a p-value of 0.01 means that there is a 1 in 100 chance the result occurred by chance. The lower the p-value, the more likely it is that the difference between groups was caused by treatment. An alteration is statistically significant if the p-value is at least 0.05. Preferably, the p-value is 0.04, 0.03, 0.02, 0.01, 0.005, 0.001 or less. As noted below, and without any limitation of the invention, achieving statistical significance generally but not always requires that combinations of several DBRISKMARKERS be used together in panels and combined with mathematical algorithms in order to achieve a statistically significant DBRISKMARKER index.
The “diagnostic accuracy” of a test, assay, or method concerns the ability of the test, assay, or method to distinguish between subjects having Diabetes or a pre-diabetic condition, or at risk for Diabetes or a pre-diabetic condition is based on whether the subjects have a “clinically significant presence” or a “clinically significant alteration” in the levels of a DBRISKMARKER. By “clinically significant presence” or “clinically significant alteration”, it is meant that the presence of the DBRISKMARKER (e.g., mass, such as milligrams, nanograms, or mass per volume, such as milligrams per deciliter or copy number of a transcript per unit volume) or an alteration in the presence of the DBRISKMARKER in the subject (typically in a sample from the subject) is higher than the predetermined cut-off point (or threshold value) for that DBRISKMARKER and therefore indicates that the subject has Diabetes or a pre-diabetic condition for which the sufficiently high presence of that protein, nucleic acid, polymorphism, metabolite or analyte is a marker.
The present invention may be used to make categorical or continuous measurements of the risk of conversion to Type 2 Diabetes, thus diagnosing a category of subjects defined as Pre-Diabetic.
In the categorical scenario, the methods of the present invention can be used to discriminate between Normal and Pre-Diabetes subject cohorts. In this categorical use of the invention, the terms “high degree of diagnostic accuracy” and “very high degree of diagnostic accuracy” refer to the test or assay for that DBRISKMARKER (or DBRISKMARKER index; wherein DBRISKMARKER value encompasses any individual measurement whether from a single DBRISKMARKER or derived from an index of DBRISKMARKERS) with the predetermined cut-off point correctly (accurately) indicating the presence or absence of Pre-Diabetes. A perfect test would have perfect accuracy. Thus, for subjects who have Pre-Diabetes, the test would indicate only positive test results and would not report any of those subjects as being “negative” (there would be no “false negatives”). In other words, the “sensitivity” of the test (the true positive rate) would be 100%. On the other hand, for subjects who did not have Pre-Diabetes, the test would indicate only negative test results and would not report any of those subjects as being “positive” (there would be no “false positives”). In other words, the “specificity” (the true negative rate) would be 100%. See, e.g., O'Marcaigh A S, Jacobson R M, “Estimating The Predictive Value Of A Diagnostic Test, How To Prevent Misleading Or Confusing Results,” Clin. Ped. 1993, 32(8): 485-491, which discusses specificity, sensitivity, and positive and negative predictive values of a test, e.g., a clinical diagnostic test. In other embodiments, the present invention may be used so as to discriminate Pre-Diabetes from Diabetes, or Diabetes from Normal. Such use may require a different DBRISKMARKER panel, mathematical algorithm, and/or cut-off point, but be subject to the same aforementioned measurements of diagnostic accuracy for the intended use.
In the categorical diagnosis of a disease, changing the cut point or threshold value of a test (or assay) usually changes the sensitivity and specificity, but in a qualitatively inverse relationship. For example, if the cut point is lowered, more subjects in the population tested will typically have test results over the cut point or threshold value. Because subjects who have test results above the cut point are reported as having the disease, condition, or syndrome for which the test is conducted, lowering the cut point will cause more subjects to be reported as having positive results (e.g., that they have Diabetes, Pre-Diabetes, or a pre-diabetic condition). Thus, a higher proportion of those who have Diabetes or Pre-Diabetes will be indicated by the test to have it. Accordingly, the sensitivity (true positive rate) of the test will be increased. However, at the same time, there will be more false positives because more people who do not have the disease, condition, or syndrome (e.g., people who are truly “negative”) will be indicated by the test to have DBRISKMARKER values above the cut point and therefore to be reported as positive (e.g., to have the disease, condition, or syndrome) rather than being correctly indicated by the test to be negative. Accordingly, the specificity (true negative rate) of the test will be decreased. Similarly, raising the cut point will tend to decrease the sensitivity and increase the specificity. Therefore, in assessing the accuracy and usefulness of a proposed medical test, assay, or method for assessing a subject's condition, one should always take both sensitivity and specificity into account and be mindful of what the cut point is at which the sensitivity and specificity are being reported because sensitivity and specificity may vary significantly over the range of cut points.
There is, however, an indicator that allows representation of the sensitivity and specificity of a test, assay, or method over the entire range of test (or assay) cut points with just a single value. That indicator is derived from a Receiver Operating Characteristics (“ROC”) curve for the test, assay, or method in question. See, e.g., Shultz, “Clinical Interpretation Of Laboratory Procedures,” chapter 14 in Teitz, Fundamentals of Clinical Chemistry, Burtis and Ashwood (eds.), 4th edition 1996, W.B. Saunders Company, pages 192-199; and Zweig et al., “ROC Curve Analysis: An Example Showing The Relationships Among Serum Lipid And Apolipoprotein Concentrations In Identifying Subjects With Coronory Artery Disease,” Clin. Chem., 1992, 38(8): 1425-1428.
An ROC curve is an x-y plot of sensitivity on the y-axis, on a scale of zero to one (e.g., 100%), against a value equal to one minus specificity on the x-axis, on a scale of zero to one (e.g., 100%). In other words, it is a plot of the true positive rate against the false positive rate for that test, assay, or method. To construct the ROC curve for the test, assay, or method in question, subjects can be assessed using a perfectly accurate or “gold standard” method that is independent of the test, assay, or method in question to determine whether the subjects are truly positive or negative for the disease, condition, or syndrome (for example, coronary angiography is a gold standard test for the presence of coronary atherosclerosis). The subjects can also be tested using the test, assay, or method in question, and for varying cut points, the subjects are reported as being positive or negative according to the test, assay, or method. The sensitivity (true positive rate) and the value equal to one minus the specificity (which value equals the false positive rate) are determined for each cut point, and each pair of x-y values is plotted as a single point on the x-y diagram. The “curve” connecting those points is the ROC curve.
The ROC curve is often used in order to determine the optimal single clinical cut-off or treatment threshold value where sensitivity and specificity are maximized; such a situation represents the point on the ROC curve which describes the upper left corner of the single largest rectangle which can be drawn under the curve.
The total area under the curve (“AUC”) is the indicator that allows representation of the sensitivity and specificity of a test, assay, or method over the entire range of cut points with just a single value. The maximum AUC is one (a perfect test) and the minimum area is one half (e.g. the area where there is no discrimination of normal versus disease). The closer the AUC is to one, the better is the accuracy of the test. It should be noted that implicit in all ROC and AUC is the definition of the disease and the post-test time horizon of interest.
By a “high degree of diagnostic accuracy”, it is meant a test or assay (such as the test of the invention for determining the clinically significant presence of DBRISKMARKERS, which thereby indicates the presence of Diabetes or a pre-diabetic condition) in which the AUC (area under the ROC curve for the test or assay) is at least 0.70, desirably at least 0.75, more desirably at least 0.80, preferably at least 0.85, more preferably at least 0.90, and most preferably at least 0.95.
By a “very high degree of diagnostic accuracy”, it is meant a test or assay in which the AUC (area under the ROC curve for the test or assay) is at least 0.80, desirably at least 0.85, more desirably at least 0.875, preferably at least 0.90, more preferably at least 0.925, and most preferably at least 0.95.
Alternatively, in low disease prevalence tested populations (defined as those with less than 1% rate of occurrences per annum), ROC and AUC can be misleading as to the clinical utility of a test, and absolute and relative risk ratios as defined elsewhere in this disclosure can be employed to determine the degree of diagnostic accuracy. Populations of subjects to be tested can also be categorized into quartiles, where the top quartile (25% of the population) comprises the group of subjects with the highest relative risk for developing or suffering from Diabetes or a pre-diabetic condition and the bottom quartile comprising the group of subjects having the lowest relative risk for developing Diabetes or a pre-diabetic condition. Generally, values derived from tests or assays having over 2.5 times the relative risk from top to bottom quartile in a low prevalence population are considered to have a “high degree of diagnostic accuracy,” and those with five to seven times the relative risk for each quartile are considered to have a very high degree of diagnostic accuracy. Nonetheless, values derived from tests or assays having only 1.2 to 2.5 times the relative risk for each quartile remain clinically useful are widely used as risk factors for a disease; such is the case with total cholesterol and for many inflammatory markers with respect to their prediction of future cardiovascular events.
The predictive value of any test depends on the sensitivity and specificity of the test, and on the prevalence of the condition in the population being tested. This notion, based on Bayes' theorem, provides that the greater the likelihood that the condition being screened for is present in a subject or in the population (pre-test probability), the greater the validity of a positive test and the greater the likelihood that the result is a true positive. Thus, the problem with using a test in any population where there is a low likelihood of the condition being present is that a positive result has limited value (i.e., more likely to be a false positive). Similarly, in populations at very high risk, a negative test result is more likely to be a false negative. By defining the degree of diagnostic accuracy, i.e., cut points on a ROC curve, defining an acceptable AUC value, and determining the acceptable ranges in relative concentration of what constitutes an effective amount of the DBRISKMARKERS of the invention allows for one of skill in the art to use the DBRISKMARKERS to diagnose or identify subjects with a pre-determined level of predictability.
Alternative methods of determining diagnostic accuracy must be used with continuous measurements of risk, which are commonly used when a disease category or risk category (such as Pre-Diabetes) has not yet been clearly defined by the relevant medical societies and practice of medicine.
“Risk” in the context of the present invention can mean “absolute” risk, which refers to that percentage probability that an event will occur over a specific time period. Absolute risk can be measured with reference to either actual observation post-measurement for the relevant time cohort, or with reference to index values developed from statistically valid historical cohorts that have been followed for the relevant time period. “Relative” risk refers to the ratio of absolute risks of a subject's risk compared either to low risk cohorts or average population risk, which can vary by how clinical risk factors are assessed. Odds ratios, the proportion of positive events to negative events for a given test result, are also commonly used (odds are according to the formula p/(1−p) where p is the probability of event and (1−p) is the probability of no event) to no-conversion. Alternative continuous measures which may be assessed in the context of the present invention include time to Diabetes conversion and therapeutic Diabetes conversion risk reduction ratios.
For such continuous measures, measures of diagnostic accuracy for a calculated index are typically based on linear regression curve fits between the predicted continuous value and the actual observed values (or historical index calculated value) and utilize measures such as R squared, p values and confidence intervals. It is not unusual for predicted values using such algorithms to be reported including a confidence interval (usually 90% or 95% CI) based on a historical observed cohort's predictions, as in the test for risk of future breast cancer recurrence commercialized by Genomic Health (Redwood City, Calif.).
The ultimate determinant and gold standard of true risk conversion to Diabetes is a actual conversions within a sufficiently large population and observed over the length of time claimed. However, this is problematic, as it is necessarily a retrospective point of view, coming after any opportunity for preventive interventions. As a result, subjects suffering from or at risk of developing Diabetes or a pre-diabetic condition are commonly diagnosed or identified by methods known in the art, and future risk is estimated based on historical experience and registry studies. Such methods include, but are not limited to, measurement of systolic and diastolic blood pressure, measurements of body mass index, in vitro determination of total cholesterol, LDL, HDL, insulin, and glucose levels from blood samples, oral glucose tolerance tests, stress tests, measurement of human serum C-reactive protein (hsCRP), electrocardiogram (ECG), c-peptide levels, anti-insulin antibodies, anti-beta cell-antibodies, and glycosylated hemoglobin (HbA1c). Additionally, any of the aforementioned methods can be used separately or in combination to assess if a subject has shown an “improvement in Diabetes risk factors.” Such improvements include, without limitation, a reduction in body mass index (BMI), a reduction in blood glucose levels, an increase in HDL levels, a reduction in systolic and/or diastolic blood pressure, an increase in insulin levels, or combinations thereof.
The oral glucose tolerance test (OGTT) is principally used for diagnosis of Diabetes Mellitus or pre-diabetic conditions when blood glucose levels are equivocal, during pregnancy, or in epidemiological studies (Definition, Diagnosis and Classification of Diabetes Mellitus and its Complications, Part 1, World Health Organization, 1999). The OGTT should be administered in the morning after at least 3 days of unrestricted diet (greater than 150 g of carbohydrate daily) and usual physical activity. A reasonable (30-50 g) carbohydrate-containing meal should be consumed on the evening before the test. The test should be preceded by an overnight fast of 8-14 hours, during which water may be consumed. After collection of the fasting blood sample, the subject should drink 75 g of anhydrous glucose or 82.5 g of glucose monohydrate in 250-300 ml of water over the course of 5 minutes. For children, the test load should be 1.75 g of glucose per kg body weight up to a total of 75 g of glucose. Timing of the test is from the beginning of the drink. Blood samples must be collected 2 hours after the test load. As previously noted, a diagnosis of impaired glucose tolerance (IGT) has been noted as being only 50% sensitive, with a >10% false positive rate, for a 7.5 year conversion to diabetes when used at the WHO cut-off points. This is a significant problem for the clinical utility of the test, as even relatively high risk ethnic groups have only a 10% rate of conversion to Diabetes over such a period unless otherwise enriched by other risk factors; in an unselected general population, the rate of conversion over such periods is typically estimated at 5-6%, or less than 1% per annum.
Other methods of measuring glucose in blood include reductiometric methods known in the art such as, but not limited to, the Somogyi-Nelson method, methods using hexokinase and glucose dehydrogenase, immobilized glucose oxidase electrodes, the o-toluidine method, the ferricyanide method and the neocuprine autoanalyzer method. Whole blood glucose values are usually about 15% lower than corresponding plasma values in patients with a normal hematocrit reading, and arterial values are generally about 7% higher than corresponding venous values. Subjects taking insulin are frequently requested to build up a “glycemic profile” by self-measurement of blood glucose at specific times of the day. A “7-point profile” is useful, with samples taken before and 90 minutes after each meal, and just before going to bed.
A subject suffering from or at risk of developing Diabetes or a pre-diabetic condition may also be suffering from or at risk of developing arteriovascular disease, hypertension or obesity. Type 2 Diabetes in particular and arteriovascular disease have many risk factors in common, and many of these risk factors are highly correlated with one another. The relationships among these risk factors may be attributable to a small number of physiological phenomena, perhaps even a single phenomenon. Subjects suffering from or at risk of developing Diabetes, arteriovascular disease, hypertension or obesity are identified by methods known in the art. For example, Diabetes is frequently diagnosed by measuring fasting blood glucose levels or insulin. Normal adult glucose levels are 60-126 mg/dl. Normal insulin levels are 7 mU/mL±3 mU. Hypertension is diagnosed by a blood pressure consistently at or above 140/90. Risk of arteriovascular disease can also be diagnosed by measuring cholesterol levels. For example, LDL cholesterol above 137 or total cholesterol above 200 is indicative of a heightened risk of arteriovascular disease. Obesity is diagnosed for example, by body mass index. Body mass index (BMI) is measured (kg/m2 (or lb/in2×704.5)). Alternatively, waist circumference (estimates fat distribution), waist-to-hip ratio (estimates fat distribution), skinfold thickness (if measured at several sites, estimates fat distribution), or bioimpedance (based on principle that lean mass conducts current better than fat mass (i.e. fat mass impedes current), estimates % fat) is measured. The parameters for normal, overweight, or obese individuals is as follows: Underweight: BMI <18.5; Normal: BMI 18.5 to 24.9; Overweight: BMI=25 to 29.9. Overweight individuals are characterized as having a waist circumference of >94 cm for men or >80 cm for women and waist to hip ratios of ≧0.95 in men and ≧0.80 in women. Obese individuals are characterized as having a BMI of 30 to 34.9, being greater than 20% above “normal” weight for height, having a body fat percentage >30% for women and 25% for men, and having a waist circumference >102 cm (40 inches) for men or 88 cm (35 inches) for women. Individuals with severe or morbid obesity are characterized as having a BMI of ≧35. Because of the interrelationship between Diabetes and arteriovascular disease, some or all of the individual DBRISKMARKERS and DBRISKMARKER panels of the present invention may overlap or be encompassed by biomarkers of arteriovascular disease, and indeed may be useful in the diagnosis of the risk of arteriovascular disease.
Risk prediction for Diabetes Mellitus or a pre-diabetic condition can also encompass risk prediction algorithms and computed indices that assess and estimate a subject's absolute risk for developing Diabetes or a pre-diabetic condition with reference to a historical cohort. Risk assessment using such predictive mathematical algorithms and computed indices has increasingly been incorporated into guidelines for diagnostic testing and treatment, and encompass indices obtained from and validated with, inter alia, multi-stage, stratified samples from a representative population. A plurality of conventional Diabetes risk factors are incorporated into predictive models. A notable example of such algorithms include the Framingham Heart Study (Kannel, W. B., et al., (1976) Am. J. Cardiol. 38: 46-51) and modifications of the Framingham Study, such as the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III), also know as NCEP/ATP III, which incorporates a patient's age, total cholesterol concentration, HDL cholesterol concentration, smoking status, and systolic blood pressure to estimate a person's 10-year risk of developing arteriovascular disease, which is commonly found in subjects suffering from or at risk for developing Diabetes Mellitus, or a pre-diabetic condition. The same Framingham algorithm has been found to be modestly predictive of the risk for developing Diabetes Mellitus, or a pre-diabetic condition.
Other Diabetes risk prediction algorithms include, without limitation, the San Antonio Heart Study (Stem, M. P. et al, (1984) Am. J. Epidemiol. 120: 834-851; Stem, M. P. et al, (1993) Diabetes 42: 706-714; Burke, J. P. et al, (1999) Arch. Intern. Med. 159: 1450-1456), Archimedes (Eddy, D. M. and Schlessinger, L. (2003) Diabetes Care 26(11): 3093-3101; Eddy, D. M. and Schlessinger, L. (2003) Diabetes Care 26(11): 3102-3110), the Finnish-based Diabetes Risk Score (Lindström, J. and Tuomilehto, J. (2003) Diabetes Care 26(3): 725-731), and the Ely Study (Griffin, S. J. et al, (2000) Diabetes Metab. Res. Rev. 16: 164-171), the contents of which are expressly incorporated herein by reference.
Archimedes is a mathematical model of Diabetes that simulates the disease state person-by-person, object-by-object and comprises biological details that are continuous in reality, such as the pertinent organ systems, more than 50 continuously interacting biological variables, and the major symptoms, tests, treatments, and outcomes commonly associated with Diabetes.
Archimedes includes many diseases simultaneously and interactively in a single integrated physiology, enabling it to address features such as co-morbidities, syndromes, treatments and other multiple effects. The Archimedes model includes Diabetes and its complications, such as coronary artery disease, congestive heart failure, and asthma. The model is written in differential equations, using object-oriented programming and a construct called “features”. The model comprises the anatomy of a subject (all simulated subjects have organs, such as hearts, livers, pancreases, gastrointestinal tracts, fat, muscles, kidneys, eyes, limbs, circulatory systems, brains, skin, and peripheral nervous systems), the “features” that determine the course of the disease and representing real physical phenomena (e.g., the number of milligrams of glucose in a deciliter of plasma, behavioral phenomena, or conceptual phenomena (e.g., the “progression” of disease), risk factors, incidence, and progression of the disease, glucose metabolism, signs and tests, diagnosis, symptoms, health outcomes of glucose metabolism, treatments, complications, deaths from Diabetes and its complications, deaths from other causes, care processes, and medical system resources. For a typical application of the model, there are thousands of simulated subjects, each with a simulated anatomy and physiology, who will get simulated diseases, can seek care at simulated health care facilities, will be seen by simulated health care personnel in simulated facilities, will be given simulated tests and treatments, and will have simulated outcomes. As in reality, each of the simulated patients is different, with different characteristics, physiologies, behaviors, and responses to treatments, all designed to match the individual variations seen in reality.
The model is built by development of a non-quantitative or conceptual description of the pertinent biology and pathology—the variables and relationships—as best they are understood with current information. Studies are then identified that pertain to the variables and relationships, and typically comprise basic research, epidemiological, and clinical studies that experts in the field identify as the foundations of their own understanding of the disease. That information is used to develop differential equations that relate the variables. The development of any particular equation in the Archimedes model involves finding the form and coefficients that best fit the available information about the variables, after which the equations are programmed into an object-oriented language. This is followed by a series of exercises in which the parts of the model are tested and debugged, first one at a time, and then in appropriate combinations, using inputs that have known outputs. The entire model can then be used to simulate a complex trial, which demonstrates not only the individual parts of the model, but also the connections between all the parts. The Archimedes calculations are performed using distributed computing techniques. Archimedes has been validated as a realistic representation of the anatomy, pathophysiology, treatments and outcomes pertinent to Diabetes and its complications (Eddy, D. M. and Schlessinger, L. (2003) Diabetes Care 26(11) 3102-3110).
The Finland-based Diabetes Risk Score is designed as a screening tool for identifying high-risk subjects in the population and for increasing awareness of the modifiable risk factors and healthy lifestyle. The Diabetes Risk Score was determined from a random population sample of 35- to 64-year old Finnish men and women with no anti-diabetic drug treatment at baseline, and followed for 10 years. Multivariate logistic regression model coefficients were used to assign each variable category a score. The Diabetes Risk Score comprises the sum of these individual scores and validated in an independent population survey performed in 1992 with a prospective follow-up for 5 years. Age, BMI, waist circumference, history of anti-hypertensive drug treatment and high blood glucose, physical activity, and daily consumption of fruits, berries, or vegetables were selected as categorical variables.
The Finland-based Diabetes Risk Score values are derived from the coefficients of the logistic model by classifying them into five categories. The estimated probability (p) of drug-treated Diabetes over a 10-year span of time for any combination of risk factors can be calculated from the following coefficients:
where β0 is the intercept and β1, β2, and so on represent the regression coefficients of the various categories of the risk factors x1, x2, and so on.
The sensitivity relates to the probability that the test is positive for subjects who will get drug-treated Diabetes in the future and the specificity reflects the probability that the test is negative for subjects without drug-treated Diabetes. The sensitivity and the specificity with 95% confidence interval (CI) were calculated for each Diabetes Risk Score level in differentiating the subjects who developed drug-treated Diabetes from those who did not. ROC curves were plotted for the Diabetes Risk score, the sensitivity was plotted on the y-axis and the false-positive rate (1-specificity) was plotted on the x-axis. The more accurately discriminatory the test, the steeper the upward portion of the ROC curve, and the higher the AUC, the optimal cut point being the peak of the curve.
Statistically significant independent predictors of future drug-treated Diabetes in the Diabetes Risk Score are age, BMI, waist circumference, antihypertensive drug therapy, and history of high blood glucose levels. The Diabetes Risk Score model comprises a concise model that includes only these statistically significant variables and a full model, which includes physical activity and fruit and vegetable consumption.
The San Antonio Heart Study is a long-term, community-based prospective observational study of Diabetes and arteriovascular disease in Mexican Americans and non-Hispanic Caucasians. The study initially enrolled 3,301 Mexican-American and 1,857 non-Hispanic Caucasian men and non-pregnant women in two phases between 1979 and 1988. Participants were 25-64 years of age at enrollment and were randomly selected from low, middle, and high-income neighborhoods in San Antonio, Tex. A 7-8 year follow-up exam followed approximately 73% of the surviving individuals initially enrolled in the study. Baseline characteristics such as medical history of Diabetes, age, sex, ethnicity, BMI, systolic and diastolic blood pressure, fasting and 2-hour plasma glucose levels, fasting serum total cholesterol, LDL, and HDL cholesterol levels, as well as triglyceride levels, were compiled and assessed. A multiple logistic regression model with incident Diabetes as the dependent variable and the aforementioned baseline characteristics were applied as independent variables. Using this model, univariate odds ratios can be computed for each potential risk factor for men and women separately and for both sexes combined. For continuous risk factors, the odds ratios can be presented for a 1-SD increment. A multivariate predicting model with both sexes combined can be developed using a stepwise logistic regression procedure in which the variables that had shown statistically significant odds ratios when examined individually were allowed to enter the model. This multivariable model is then analyzed by ROC curves and 95% CIs of the areas under the ROC curves estimated by non-parametric algorithms such as those described by DeLong (DeLong E. R. et al, (1988) Biometrics 44: 837-45). The results of the San Antonio Heart Study indicate that pre-diabetic subjects have an atherogenic pattern of risk factors (possibly caused by obesity, hyperglycemia, and especially hyperinsulinemia), which may be present for many years and may contribute to the risk of macrovascular disease as much as the duration of clinical Diabetes itself.
Despite the numerous studies and algorithms that have been used to assess the risk of Diabetes or a pre-diabetic condition, the evidence-based, multiple risk factor assessment approach is only moderately accurate for the prediction of short- and long-term risk of manifesting Diabetes or a pre-diabetic condition in individual asymptomatic or otherwise healthy subjects. Such risk prediction algorithms can be advantageously used in combination with the DBRISKMARKERS of the present invention to distinguish between subjects in a population of interest to determine the risk stratification of developing Diabetes or a pre-diabetic condition. The DBRISKMARKERS and methods of use disclosed herein provide tools that can be used in combination with such risk prediction algorithms to assess, identify, or diagnose subjects who are asymptomatic and do not exhibit the conventional risk factors.
The data derived from risk prediction algorithms and from the methods of the present invention can be compared by linear regression. Linear regression analysis models the relationship between two variables by fitting a linear equation to observed data. One variable is considered to be an explanatory variable, and the other is considered to be a dependent variable. For example, values obtained from the Archimedes or San Antonio Heart analysis can be used as a dependent variable and analyzed against levels of one or more DBRISKMARKERS as the explanatory variables in an effort to more fully define the underlying biology implicit in the calculated algorithm score (see Examples). Alternatively, such risk prediction algorithms, or their individual inputs, which are generally DBRISKMARKERS themselves, can be directly incorporated into the practice of the present invention, with the combined algorithm compared against actual observed results in a historical cohort.
A linear regression line has an equation of the form Y=a+bX, where X is the explanatory variable and Y is the dependent variable. The slope of the line is b, and a is the intercept (the value of y when x=0). A numerical measure of association between two variables is the “correlation coefficient,” or R, which is a value between −1 and 1 indicating the strength of the association of the observed data for the two variables. This is also often reported as the square of the correlation coefficient, as the “coefficient of determination” or R2; in this form it is the proportion of the total variation in Y explained by fitting the line. The most common method for fitting a regression line is the method of least-squares. This method calculates the best-fitting line for the observed data by minimizing the sum of the squares of the vertical deviations from each data point to the line (if a point lies on the fitted line exactly, then its vertical deviation is 0). Because the deviations are first squared, then summed, there are no cancellations between positive and negative values.
After a regression line has been computed for a group of data, a point which lies far from the line (and thus has a large residual value) is known as an outlier. Such points may represent erroneous data, or may indicate a poorly fitting regression line. If a point lies far from the other data in the horizontal direction, it is known as an influential observation. The reason for this distinction is that these points have may have a significant impact on the slope of the regression line. Once a regression model has been fit to a group of data, examination of the residuals (the deviations from the fitted line to the observed values) allows one of skill in the art to investigate the validity of the assumption that a linear relationship exists. Plotting the residuals on the y-axis against the explanatory variable on the x-axis reveals any possible non-linear relationship among the variables, or might alert the skilled artisan to investigate “lurking variables.” A “lurking variable” exists when the relationship between two variables is significantly affected by the presence of a third variable which has not been included in the modeling effort.
Linear regression analyses can be used, inter alia, to predict the risk of developing Diabetes or a pre-diabetic condition based upon correlating the levels of DBRISKMARKERS in a sample from a subject to that subjects' actual observed clinical outcomes, or in combination with, for example, calculated Archimedes risk scores, San Antonio Heart risk scores, or other known methods of diagnosing or predicting the prevalence of Diabetes or a pre-diabetic condition. Of particular use, however, are non-linear equations and analyses to determine the relationship between known predictive models of Diabetes and levels of DBRISKMARKERS detected in a subject sample. Of particular interest are structural and synactic classification algorithms, and methods of risk index construction, utilizing pattern recognition features, including established techniques such as the Kth-Nearest Neighbor, Boosting, Decision Trees, Neural Networks, Bayesian Networks, Support Vector Machines, and Hidden Markov Models. Most commonly used are classification algorithms using logistic regression, which are the basis for the Framingham, Finnish, and San Antonio Heart risk scores. Furthermore, the application of such techniques to panels of multiple DBRISKMARKERS is encompassed by or within the ambit of the present invention, as is the use of such combination to create single numerical “risk indices” or “risk scores” encompassing information from multiple DBRISKMARKER inputs. An example using logistic regression is described herein in the Examples.
Factor analysis is a mathematical technique by which a large number of correlated variables (such as Diabetes risk factors) can be reduced to fewer “factors” that represent distinct attributes that account for a large proportion of the variance in the original variables (Hanson, R. L. et al, (2002) Diabetes 51: 3120-3127). Thus, factor analysis is well suited for identifying components of Diabetes Mellitus and pre-diabetic conditions such as IGT, IFG, and Metabolic Syndrome. Epidemiological studies of factor “scores” from these anlyses can further determine relations between components of the metabolic syndrome and incidence of Diabetes. The premise underlying factor analysis is that correlations observed among a set of variables can be explained by a small number of unique unmeasured variables, or “factors”. Factor analysis involves two procedures: 1) factor extraction to estimate the number of factors, and 2) factor rotation to determine constituents of each factor in terms of the original variables.
Factor extraction can be conducted by the method of principal components. These components are linear combinations of the original variables that are constructed so that each component has a correlation of zero with each of the other components. Each principal component is associated with an “eigen-value,” which represents the variance in the original variables explained by that component (with each original variable standardized to have a variance of 1). The number of principal components that can be constructed is equal to the number of original variables. In factor analysis, the number of factors is customarily determined by retention of only those components that account for more of the total variance than any single original variable (i.e., those components with eigen-values of >1).
Once the number of factors has been established, then factor rotation is conducted to determine the composition of factors that has the most parsimonious interpretation in terms of the original variables. In factor rotation, “factor loadings,” which represent correlations of each factor with the original variables, are changed so that these factor loadings are made as close to 0 or 1 as possible (with the constraint that the total amount of variance explained by the factors remains unchanged). A number of methods for factor rotation have been developed and can be distinguished by whether they require the final set of factors to remain uncorrelated with one another (also known as “orthogonal methods”) or by whether they allow factors to be correlated (“oblique methods”). In interpretation of factor analysis, the pattern of factor loadings is examined to determine which original variables represent primary constituents of each factor. Conventionally, variables that have a factor loading of >0.4 (or less than −0.4) with a particular factor are considered to be its major constituents. Factor analysis can be very useful in constructing DBRISKMARKER panels from their constituent components, and in grouping substitutable groups of markers.
Levels of an effective amount of DBRISKMARKER proteins, nucleic acids, polymorphisms, metabolites, or other analytes also allows for the course of treatment of Diabetes or a pre-diabetic condition to be monitored. In this method, a biological sample can be provided from a subject undergoing treatment regimens, e.g., drug treatments, for Diabetes. Such treatment regimens can include, but are not limited to, exercise regimens, dietary supplementation, surgical intervention, and treatment with therapeutics or prophylactics used in subjects diagnosed or identified with Diabetes or a pre-diabetic condition. If desired, biological samples are obtained from the subject at various time points before, during, or after treatment. Levels of an effective amount of DBRISKMARKER proteins, nucleic acids, polymorphisms, metabolites, or other analytes can then be determined and compared to a reference value, e.g. a control subject or population whose diabetic state is known or an index value or baseline value. The reference sample or index value or baseline value may be taken or derived from one or more subjects who have been exposed to the treatment, or may be taken or derived from one or more subjects who are at low risk of developing Diabetes or a pre-diabetic condition, or may be taken or derived from subjects who have shown improvements in Diabetes risk factors as a result of exposure to treatment. Alternatively, the reference sample or index value or baseline value may be taken or derived from one or more subjects who have not been exposed to the treatment. For example, samples may be collected from subjects who have received initial treatment for Diabetes or a pre-diabetic condition and subsequent treatment for Diabetes or a pre-diabetic condition to monitor the progress of the treatment. A reference value can also comprise a value derived from risk prediction algorithms or computed indices from population studies such as those disclosed herein.
The DBRISKMARKERS of the present invention can thus be used to generate a “reference expression profile” of those subjects who do not have Diabetes or a pre-diabetic condition such as impaired glucose tolerance, and would not be expected to develop Diabetes or a pre-diabetic condition. The DBRISKMARKERS disclosed herein can also be used to generate a “subject expression profile” taken from subjects who have Diabetes or a pre-diabetic condition like impaired glucose tolerance. The subject expression profiles can be compared to a reference expression profile to diagnose or identify subjects at risk for developing Diabetes or a pre-diabetic condition, to monitor the progression of disease, as well as the rate of progression of disease, and to monitor the effectiveness of Diabetes or pre-Diabetes treatment modalities. The reference and subject expression profiles of the present invention can be contained in a machine-readable medium, such as but not limited to, analog tapes like those readable by a VCR, CD-ROM, DVD-ROM, USB flash media, among others. Such machine-readable media can also contain additional test results, such as, without limitation, measurements of conventional Diabetes risk factors like systolic and diastolic blood pressure, blood glucose levels, insulin levels, BMI indices, and cholesterol (LDL and HDL) levels. Alternatively or additionally, the machine-readable media can also comprise subject information such as medical history and any relevant family history. The machine-readable media can also contain information relating to other Diabetes-risk algorithms and computed indices such as those described herein.
Differences in the genetic makeup of subjects can result in differences in their relative abilities to metabolize various drugs, which may modulate the symptoms or risk factors of Diabetes or a pre-diabetic condition. Subjects that have Diabetes or a pre-diabetic condition, or at risk for developing Diabetes or a pre-diabetic condition can vary in age, ethnicity, body mass index (BMI), total cholesterol levels, blood glucose levels, blood pressure, LDL and HDL levels, and other parameters. Accordingly, use of the DBRISKMARKERS disclosed herein allow for a pre-determined level of predictability that a putative therapeutic or prophylactic to be tested in a selected subject will be suitable for treating or preventing Diabetes or a pre-diabetic condition in the subject.
To identify therapeutics or drugs that are appropriate for a specific subject, a test sample from the subject can be exposed to a therapeutic agent or a drug, and the level of one or more of DBRISKMARKER proteins, nucleic acids, polymorphisms, metabolites or other analytes can be determined. The level of one or more DBRISKMARKERS can be compared to sample derived from the subject before and after treatment or exposure to a therapeutic agent or a drug, or can be compared to samples derived from one or more subjects who have shown improvements in Diabetes or pre-Diabetes risk factors as a result of such treatment or exposure. Examples of such therapeutics or drugs frequently used in Diabetes treatments, and may modulate the symptoms or risk factors of Diabetes include, but are not limited to, sulfonylureas like glimepiride, glyburide (also known in the art as glibenclamide), glipizide, gliclazide; biguanides such as metformin; insulin (including inhaled formulations such as Exubera), and insulin analogs such as insulin lispro (Humalog), insulin glargine (Lantus), insulin detemir, and insulin glulisine; peroxisome proliferator-activated receptor-γ (PPAR-γ) agonists such as the thiazolidinediones including troglitazone (Rezulin), pioglitazone (Actos), rosiglitazone (Avandia), and isaglitzone (also known as netoglitazone); dual-acting PPAR agonists such as BMS-298585 and tesaglitazar; insulin secretagogues including metglitinides such as repaglinide and nateglinide; analogs of glucagon-like peptide-1 (GLP-1) such as exenatide (AC-2993) and liraglutide (insulinotropin); inhibitors of dipeptidyl peptidase IV like LAF-237; pancreatic lipase inhibitors such as orlistat; α-glucosidase inhibitors such as acarbose, migitol, and voglibose; and combinations thereof, particularly metformin and glyburide (Glucovance), metformin and rosiglitazone (Avandamet), and metformin and glipizide (Metaglip). Such therapeutics or drugs have been prescribed for subjects diagnosed with Diabetes or a pre-diabetic condition, and may modulate the symptoms or risk factors of Diabetes or a pre-diabetic condition.
A subject sample can be incubated in the presence of a candidate agent and the pattern of DBRISKMARKER expression in the test sample is measured and compared to a reference profile, e.g., a Diabetes reference expression profile or a non-Diabetes reference expression profile or an index value or baseline value. The test agent can be any compound or composition or combination thereof. For example, the test agents are agents frequently used in Diabetes treatment regimens and are described herein.
Table 1 comprises the two-hundred and sixty (260) DBRISKMARKERS of the present invention. One skilled in the art will recognize that the DBRISKMARKERS presented herein encompasses all forms and variants, including but not limited to, polymorphisms, isoforms, mutants, derivatives, precursors including nucleic acids, receptors (including soluble and transmembrane receptors), ligands, and post-translationally modified variants, as well as any multi-unit nucleic acid, protein, and glycoprotein structures comprised of any of the DBRISKMARKERS as constituent subunits of the fully assembled structure.
One skilled in the art will note that the above listed DBRISKMARKERS come from a diverse set of physiological and biological pathways, including many which are not commonly accepted to be related to diabetes. For convenience and ease of analysis, a representative subset of approximately fifty of the disclosed DBRISKMARKERS was studied in depth in order to elucidate the more important pathways.
In
The strongest signal comes from inflammatory markers concentrated on the cytokine-cytokine receptor and adipocytokine signaling pathways, and significantly the Jak-STAT signalling pathway, which is concentrated in a group of markers including LEP (Leptin) and HP (Haptoglobin). Another overlapping signal also covers the MAPK and insulin signaling pathways and, interestingly, the mTOR signaling pathway, coming from DBRISKMARKERS including ILGFBP3 (Insulin-like growth factor binding protein 3) and such DBRISKMARKERS as VEGF. This group also has the overlapping involvement of ECM-receptor interaction and cell adhesion molecule (CAMs) pathways, together with complement and coagulation cascades and hematopoietic cell lineages and toll-like receptor pathways, perhaps indicating endothelial and vascular changes, and is further represented by CD14 and CSF1 (M-CSF). A final signal, involving the DBRISKMARKERS such as VEGF and SELE (E-Selectin), is concentrated on focal adhesion, ECM and other pathways related to vascular and endothelial remodeling. The kinetics of these expression relative to status of pre-diabetic risk remains to be ascertained and validated, but it is believed that such distinct patterns may allow a more biologically detailed and clinically useful signal from the DBRISKMARKERS as well as opportunities for pattern recognition within the DBRISKMARKER panel algorithms combining the biomarker signals.
The above discussion for convenience focuses on a subset of the DBRISKMARKERS; other DBRISKMARKERS and even biomarkers which are not listed in the above table but related to these physiological and biological pathways may prove to be useful given the signal and information provided from these studies. To the extent that other participants within the total list of DBRISKMARKERS are also relevant pathway participants in Pre-Diabetes they may be functional equivalents to the biomarkers thus far disclosed. DBRISKMARKERS provided they additionally share certain defined characteristics of a good biomarker, which would include both this biological process involvement and also analytically important characterisitics such as the bioavailability of said markers at a useful signal to noise ration, and in a useful sample matrix such as blood serum. Such requirements typically limit the usefulness of many members of a biological KEGG pathway, as this is unlikely to be generally the case, and frequently occurs only in pathway members that constitute secretory substances, those accessible on the plasma membranes of cells, as well as those that are released into the serum upon cell death, due to apotosis or for other reasons such as endothelial remodeling or other cell turnover or cell necrotic processes, whether or not said is related to the disease progression of Pre-Diabetes and Diabetes. However, the remaining and future biomarkers that meet this high standard for DBRISKMARKERS are likely to be quite valuable. Our invention encompasses such functional and statistical equivalents to the aforelisted DBRISKMARKERS. Furthermore, the statistical utility of such additional DBRISKMARKERS is substantially dependent on the cross-correlation between markers and new markers will often be required to operate within a panel in order to elaborate the meaning of the underlying biology.
As is shown in
It was previously noted that many of the individual markers listed, when used alone and not as a member of a multi-marker panel of DBRISKMARKERS, have little or no statistically significant differences in their concentration levels between Normal, Pre-Diabetes, and Diabetes populations, and thus cannot reliably be used alone in classifying any patient between those three states (Normal, Pre-Diabetes, or Diabetes). As also previously mentioned, a common measure of statistical significance is the p value, which indicates the probability that an observation has arisen by chance rather than correlation or causation; preferably, such p values are 0.05 or less, representing a 95% chance that the observation of interest arose by other than chance.
Despite this individual marker performance, it is the subject matter of our invention that certain specific combinations of two or more DBRISKMARKERS of the present invention can also be used as multi-marker panels comprising combinations of DBRISKMARKERS that are known to be involved in one or more physiological or biological pathways, and that such information can be combined and made clinically useful through the use of various statistical classification algorithms, including those commonly used such as logistic regression. In fact, it is the further detailed subject matter of the invention, that such algorithms, when optimized for their best clinical classification performance (as measured by line fitting statistics such as R2) across a reasonably large group of potentially contributing DBRISKMARKERS as continuous measurements of the risk of conversion to Type 2 Diabetes, will commonly share one of a discrete number of multimarker components motifs and combinations. These include, solely within the representative group of DBRISKMARKERS previously assayed, strong significance around groupings aound the marker LEPTIN (LEP), and in particular the various permutations and component combinations of LEPTIN, HAPTOGLOBIN (HP), INSULIN-LIKE GROWTH FACTOR BINDING PROTEIN 3 (IGFBP3), and RESISTIN (RETN) including, without limitation, the various subsets of two or more of each of the foregoing markers, and the combination of those sets with additional markers. An alternative general strategy to that of using LEPTIN and its supporting cluster of partner markers involves the use of TNFR1 and CD26, typically together as a cluster, but either alone or with other markers (including with the use of LEPTIN and any of the other individually mentioned family of markers in panels of three or more DBRISKMARKERS). A third, generally lower performing strategy than that of LEP is to use more generalized markers of inflammation, such as C-REACTIVE PROTEIN (CRP), RECEPTOR FOR ADVANCED GLYCOSYLATION ENDPRODUCTS (RAGE, now AGER), and general cytokines, adipocytokines, and complement and coagulation cascade members such as IL-18, ADIPONECTIN (ADIPOQ), ADIPISIN (aka COMPLEMENT FACTOR D or CFD), and PAI-1 (SERPINE1), among the others disclosed, in larger numbers or in combination with more specific DBRISKMARKERS.
The general concept of how two less specific or lower performing biomarkers are combined into novel and more useful combinations for the purpose of diagnosing PRE-DIABETES, is a subject and key aspect of the invention. An illustrative example,
As the lower left panel indicates, while the test has maintained its sensitivity (a larger patient cohort might show an improvement, but LEPTIN had excellent starting performance, and only one false negative remains). However, specificity has declined dramatically, to a level worse than either marker alone, due to the higher number of false positives called (58 together versus 29 for LEPTIN alone or 45 for HAPTOGLOBIN alone). More typically, an improvement in sensitivity at the cost of a drop in specificity is expected when two markers are used in this way together.
In contrast, in the lower right panel, the same two markers are tested together when combined using a standard logistic regression algorithm. In this scenario, sensitivity remains maintained, but specificity has increased to a higher level than either marker is capable of alone. The logistic algorithm scenario is shown across all cut-offs in the following ROC curve, and has the a higher AUC than either marker alone (unfortunately, again due to the small sample size of the disease cohort, this AUC difference does not quite make statistical significance; however, it is clear from the preceeding categorical analysis that the combination is a superior test, with a lower false positive rate and false negative rate)).
This example illustrated several concepts. The first is that multiple markers can often yield better performance than the individual components when proper mathematical and clinical algorithms are used; this is often evident in both sensitivity and specificity, and results in a greater AUC. The second key concept is that there is often novel unperceived information in existing markers, as was necessary in order to achieve the new algorithm combined level of specificity. The final concept is that this hidden information may hold true even for markers which are generally regarded to have suboptimal clinical performance on their own, as did the HAPTOGLOBIN in the example, at only 62.5% sensitivity and 41.5% specificity, a conclusion which would not be obvious prior to testing the two markers together with an algorithm. In fact, the suboptimal performance in terms of high false positive rates on the individual test in may very well be the indicator that some important additional information is contained within the tests results—information which would not be elucidated absent the combination with a second marker and a mathematical algorithm. The example in
The disadvantages of such forward selection techniques is the possibility of non-step wise solutions, where synergistic information can be gained by also testing a “step backwards” in order to reassess each existing markers remaining contribution (as noted, the beta coefficients do change) and to test for such synergies that might be cloaked by the legacy steps taken to get to the current panel size. This forward and backwards technique can be combined with a balancing factor providing input as to when the additional complexity of more markers outweighs the incremental gain to further marker additions, a searching technique commonly called a “stepwise.” searching algorithm. It is clear from the R2 graph in
Several techniques can be used to generate such best marker addition algorithms, building the optimal DBRISKMARKER additions at each step.
Other statistical tools such as factor and cross-marker correlation/covariance analysis allow more rationale approaches to panel construction.
A DBRISKMARKER panel is comprised of a series of individual DBRISKMARKER components. Within our study using 50 representative DBRISKMARKERS, there are three core marker approaches which can be used independently or, when larger panels are desired, in combination in order to achieve high performance in a DBRISKMARKER panel: the first, which we term the Key Individual Marker approach in
An second alternative approach is to begin building a DBRISKMARKER panel using what we have defined in
Key Combination Markers do not have a set order of hierarchy or order beyond the common upfront pairing of TNFR1 and CD26, and of several of the other members with Key Individual Markers, notably E-Selctin, MCSF, and VEGF. Often the Key Combination Markers are added late in a DBRISKMARKER panel construction approach, of when factor and information redundancy makes multiple statistically similar high performance solutions to the optimal DBRISKMARKER panel possible.
A final, third approach is to work within the group of more generalized inflammation cytokine, adipokine and coagulation markers, including CRP, RAGE, IL-18, ADIPONECTIN, ACTIVIN_A, and ADIPISIN, This is a common fill-in strategy for approaches begun with Key Individual or Key Combination Markers, as the more generalized and broad information content of some of these multi-potent markers (such as CRP and RAGE in particular) makes them amenable to being added to many different panel combinations without creating information redundancy.
Examples of specific DBRISKMARKER panel construction using the above general techniques are also disclosed herein, without limitation of the foregoing, our techniques of marker panel construction, or the applicability of alternative DBRISKMARKERS or biomarkers from functionally equivalent classes which are also involved in the same constituent physiological and biological pathways.
Levels of the DBRISKMARKERS can be determined at the protein or nucleic acid level using any method known in the art. For example, at the nucleic acid level, Northern and Southern hybridization analysis, as well as ribonuclease protection assays using probes which specifically recognize one or more of these sequences can be used to determine gene expression. Alternatively, expression can be measured using reverse-transcription-based PCR assays (RT-PCR), e.g., using primers specific for the differentially expressed sequence of genes. Expression can also be determined at the protein level, e.g., by measuring the levels of peptides encoded by the gene products described herein, or activities thereof. Such methods are well known in the art and include, e.g., immunoassays based on antibodies to proteins encoded by the genes, aptamers or molecular imprints. Any biological material can be used for the detection/quantification of the protein or its activity. Alternatively, a suitable method can be selected to determine the activity of proteins encoded by the marker genes according to the activity of each protein analyzed.
The DBRISKMARKER proteins, polypeptides, mutations, and polymorphisms thereof can be detected in any suitable manner, but is typically detected by contacting a sample from the subject with an antibody which binds the DBRISKMARKER protein, polypeptide, mutation, or polymorphism and then detecting the presence or absence of a reaction product. The antibody may be monoclonal, polyclonal, chimeric, or a fragment of the foregoing, as discussed in detail above, and the step of detecting the reaction product may be carried out with any suitable immunoassay. The sample from the subject is typically a biological fluid as described above, and may be the same sample of biological fluid used to conduct the method described above.
Immunoassays carried out in accordance with the present invention may be homogeneous assays or heterogeneous assays. In a homogeneous assay the immunological reaction usually involves the specific antibody (e.g., anti-DBRISKMARKER protein antibody), a labeled analyte, and the sample of interest. The signal arising from the label is modified, directly or indirectly, upon the binding of the antibody to the labeled analyte. Both the immunological reaction and detection of the extent thereof can be carried out in a homogeneous solution. Immunochemical labels which may be employed include free radicals, radioisotopes, fluorescent dyes, enzymes, bacteriophages, or coenzymes.
In a heterogeneous assay approach, the reagents are usually the sample, the antibody, and means for producing a detectable signal. Samples as described above may be used. The antibody can be immobilized on a support, such as a bead (such as protein A and protein G agarose beads), plate or slide, and contacted with the specimen suspected of containing the antigen in a liquid phase. The support is then separated from the liquid phase and either the support phase or the liquid phase is examined for a detectable signal employing means for producing such signal. The signal is related to the presence of the analyte in the sample. Means for producing a detectable signal include the use of radioactive labels, fluorescent labels, or enzyme labels. For example, if the antigen to be detected contains a second binding site, an antibody which binds to that site can be conjugated to a detectable group and added to the liquid phase reaction solution before the separation step. The presence of the detectable group on the solid support indicates the presence of the antigen in the test sample. Examples of suitable immunoassays are oligonucleotides, immunoblotting, immunofluorescence methods, chemiluminescence methods, electrochemiluminescence or enzyme-linked immunoassays.
Those skilled in the art will be familiar with numerous specific immunoassay formats and variations thereof which may be useful for carrying out the method disclosed herein. See generally E. Maggio, Enzyme-Immunoassay, (1980) (CRC Press, Inc., Boca Raton, Fla.); see also U.S. Pat. No. 4,727,022 to Skold et al. titled “Methods for Modulating Ligand-Receptor Interactions and their Application,” U.S. Pat. No. 4,659,678 to Forrest et al. titled “Immunoassay of Antigens,” U.S. Pat. No. 4,376,110 to David et al., titled “Immunometric Assays Using Monoclonal Antibodies,” U.S. Pat. No. 4,275,149 to Litman et al., titled “Macromolecular Environment Control in Specific Receptor Assays,” U.S. Pat. No. 4,233,402 to Maggio et al., titled “Reagents and Method Employing Channeling,” and U.S. Pat. No. 4,230,767 to Boguslaski et al., titled “Heterogenous Specific Binding Assay Employing a Coenzyme as Label.”
Antibodies can be conjugated to a solid support suitable for a diagnostic assay (e.g., beads such as protein A or protein G agarose, microspheres, plates, slides or wells formed from materials such as latex or polystyrene) in accordance with known techniques, such as passive binding. Antibodies as described herein may likewise be conjugated to detectable labels or groups such as radiolabels (e.g., 35S, 125I, 131I), enzyme labels (e.g., horseradish peroxidase, alkaline phosphatase), and fluorescent labels (e.g., fluorescein, Alexa, green fluorescent protein) in accordance with known techniques.
Antibodies can also be useful for detecting post-translational modifications of DBRISKMARKER proteins, polypeptides, mutations, and polymorphisms, such as tyrosine phosphorylation, threonine phosphorylation, serine phosphorylation, glycosylation (e.g., O-GlcNAc). Such antibodies specifically detect the phosphorylated amino acids in a protein or proteins of interest, and can be used in immunoblotting, immunofluorescence, and ELISA assays described herein. These antibodies are well-known to those skilled in the art, and commercially available. Post-translational modifications can also be determined using metastable ions in reflector matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF) (Wirth, U. et al. (2002) Proteomics 2(10): 1445-51).
For DBRISKMARKER proteins, polypeptides, mutations, and polymorphisms known to have enzymatic activity, the activities can be determined in vitro using enzyme assays known in the art. Such assays include, without limitation, kinase assays, phosphatase assays, reductase assays, among many others. Modulation of the kinetics of enzyme activities can be determined by measuring the rate constant KM using known algorithms, such as the Hill plot, Michaelis-Menten equation, linear regression plots such as Lineweaver-Burk analysis, and Scatchard plot.
Using sequence information provided by the database entries for the DBRISKMARKER sequences, expression of the DBRISKMARKER sequences can be detected (if present) and measured using techniques well known to one of ordinary skill in the art. For example, sequences within the sequence database entries corresponding to DBRISKMARKER sequences, or within the sequences disclosed herein, can be used to construct probes for detecting DBRISKMARKER RNA sequences in, e.g., Northern blot hybridization analyses or methods which specifically, and, preferably, quantitatively amplify specific nucleic acid sequences. As another example, the sequences can be used to construct primers for specifically amplifying the DBRISKMARKER sequences in, e.g., amplification-based detection methods such as reverse-transcription based polymerase chain reaction (RT-PCR). When alterations in gene expression are associated with gene amplification, deletion, polymorphisms, and mutations, sequence comparisons in test and reference populations can be made by comparing relative amounts of the examined DNA sequences in the test and reference cell populations.
Expression of the genes disclosed herein can be measured at the RNA level using any method known in the art. For example, Northern hybridization analysis using probes which specifically recognize one or more of these sequences can be used to determine gene expression. Alternatively, expression can be measured using reverse-transcription-based PCR assays (RT-PCR), e.g., using primers specific for the differentially expressed sequences.
Alternatively, DBRISKMARKER protein and nucleic acid metabolites can be measured. The term “metabolite” includes any chemical or biochemical product of a metabolic process, such as any compound produced by the processing, cleavage or consumption of a biological molecule (e.g., a protein, nucleic acid, carbohydrate, or lipid). Metabolites can be detected in a variety of ways known to one of skill in the art, including the refractive index spectroscopy (RI), ultra-violet spectroscopy (UV), fluorescence analysis, radiochemical analysis, near-infrared spectroscopy (near-IR), nuclear magnetic resonance spectroscopy (NMR), light scattering analysis (LS), mass spectrometry, pyrolysis mass spectrometry, nephelometry, dispersive Raman spectroscopy, gas chromatography combined with mass spectrometry, liquid chromatography combined with mass spectrometry, matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) combined with mass spectrometry, ion spray spectroscopy combined with mass spectrometry, capillary electrophoresis, NMR and IR detection. (See, WO 04/056456 and WO 04/088309, each of which are hereby incorporated by reference in their entireties) In this regard, other DBRISKMARKER analytes can be measured using the above-mentioned detection methods, or other methods known to the skilled artisan.
Kits
The invention also includes a DBRISKMARKER-detection reagent, e.g., nucleic acids that specifically identify one or more DBRISKMARKER nucleic acids by having homologous nucleic acid sequences, such as oligonucleotide sequences, complementary to a portion of the DBRISKMARKER nucleic acids or antibodies to proteins encoded by the DBRISKMARKER nucleic acids packaged together in the form of a kit. The oligonucleotides can be fragments of the DBRISKMARKER genes. For example the oligonucleotides can be 200, 150, 100, 50, 25, 10 or less nucleotides in length. The kit may contain in separate containers a nucleic acid or antibody (either already bound to a solid matrix or packaged separately with reagents for binding them to the matrix), control formulations (positive and/or negative), and/or a detectable label. Instructions (e.g., written, tape, VCR, CD-ROM, etc.) for carrying out the assay may be included in the kit. The assay may for example be in the form of a Northern hybridization or a sandwich ELISA as known in the art.
For example, DBRISKMARKER detection reagents can be immobilized on a solid matrix such as a porous strip to form at least one DBRISKMARKER detection site. The measurement or detection region of the porous strip may include a plurality of sites containing a nucleic acid. A test strip may also contain sites for negative and/or positive controls. Alternatively, control sites can be located on a separate strip from the test strip. Optionally, the different detection sites may contain different amounts of immobilized nucleic acids, e.g., a higher amount in the first detection site and lesser amounts in subsequent sites. Upon the addition of test sample, the number of sites displaying a detectable signal provides a quantitative indication of the amount of DBRISKMARKERS present in the sample. The detection sites may be configured in any suitably detectable shape and are typically in the shape of a bar or dot spanning the width of a test strip.
Alternatively, the kit contains a nucleic acid substrate array comprising one or more nucleic acid sequences. The nucleic acids on the array specifically identify one or more nucleic acid sequences represented by DBRISKMARKERS1-260. In various embodiments, the expression of 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 40, 50, 100, 125, 150, 175, 200, 210, 220, 230, 240 or more of the sequences represented by DBRISKMARKERS1-260 can be identified by virtue of binding to the array. The substrate array can be on, e.g., a solid substrate, e.g., a “chip” as described in U.S. Pat. No. 5,744,305. Alternatively, the substrate array can be a solution array, e.g., xMAP (Luminex, Austin, Tex.), Cyvera (Illumina, San Diego, Calif.), CellCard (Vitra Bioscience, Mountain View, Calif.) and Quantum Dots' Mosaic (Invitrogen, Carlsbad, Calif.).
The skilled artisan can routinely make antibodies, nucleic acid probes, e.g., oligonucleotides, aptamers, siRNAs, antisense oligonucleotides, against any of the DBRISKMARKERS in Table 1.
The protein biomarker panels were determined by analyzing 64 proteins in human serum samples derived from a group of 96 normal, pre-diabetic, and diabetic persons.
Source Reagents: A large and diverse array of vendors that were used to source immunoreagents as a starting point for assay development, such as, but not limited to, Abazyme, Abnova, Affinity Biologicals, AntibodyShop, Biogenesis, Biosense Laboratories, Calbiochem, Cell Sciences, Chemicon International, Chemokine, Clontech, Cytolab, DAKO, Diagnostic BioSystems, eBioscience, Endocrine Technologies, Enzo Biochem, Eurogentec, Fusion Antibodies, Genesis Biotech, GloboZymes, Haematologic Technologies, Immunodetect, Immunodiagnostik, Immunometrics, Immunostar, Immunovision, Biogenex, Invitrogen, Jackson ImmunoResearch Laboratory, KMI Diagnostics, Koma Biotech, LabFrontier Life Science Institute, Lee Laboratories, Lifescreen, Maine Biotechnology Services, Mediclone, MicroPharm Ltd., ModiQuest, Molecular Innovations, Molecular Probes, Neoclone, Neuromics, New England Biolabs, Novocastra, Novus Biologicals, Oncogene Research Products, Orbigen, Oxford Biotechnology, Panvera, PerkinElmer Life Sciences, Pharmingen, Phoenix Pharmaceuticals, Pierce Chemical Company, Polymun Scientific, Polysiences, Inc., Promega Corporation, Proteogenix, Protos Immunoresearch, QED Biosciences, Inc., R&D Systems, Repligen, Research Diagnostics, Roboscreen, Santa Cruz Biotechnology, Seikagaku America, Serological Corporation, Serotec, SigmaAldrich, StemCell Technologies, Synaptic Systems GmbH, Technopharm, Terra Nova Biotechnology, TiterMax, Trillium Diagnostics, Upstate Biotechnology, US Biological, Vector Laboratories, Wako Pure Chemical Industries, and Zeptometrix. A search for capture antibodies, detection antibodies, and analytes was performed to configure a working sandwich immunoassay. The reagents were ordered and received into inventory.
Immunoassays were developed in three steps: Prototyping, Validation, and Kit Release. Prototyping was conducted using standard ELISA formats when the two antibodies used in the assay were from different host species. Using standard conditions, anti-host secondary antibodies conjugated with horse radish peroxidase were evaluated in a standard curve. If a good standard curve was detected, the assay proceeded to the next step. Assays that had the same host antibodies went directly to the next step (e.g., mouse monoclonal sandwich assays).
Validation of a working assay was performed using the Zeptosense detection platform from Singulex, Inc. (St. Louis, Mo.). The detection antibody was first conjugated to the fluorescent dye Alexa 647. The conjugations used standard NHS ester chemistry, for example, according to the manufacturer. Once the antibody was labeled, the assay was tested in a sandwich assay format using standard conditions. Each assay well was solubilized in a denaturing buffer, and the material was read on the Zeptosense platform.
Each component of the kit including manufacturer, catalog numbers, lot numbers, stock and working concentrations, standard curve, and serum requirements were compiled into a standard operating procedures for each biomarker assay. This kit was then released for use to test clinical samples.
Samples were collected from several sources. In all cases, sufficient clinical annotations were available to calculate risk factors using the model developed by Stem et al. (2002). Typically, a minimum of the following clinical annotations were available from each study: Date of collection, age, sex, height, weight, waist, BMI, ethnicity, family history, diastolic and systolic blood pressure, fasting glucose levels, cholesterol. The samples were collected using standard protocols, and were stored at −80 C from the time of collection.
Clinical samples arrived frozen on dry ice, and each sample was stored at −80 C. Each sample typically had many clinical annotations associated with it. The clinical annotations associated with each sample set were brought into a standardized nomenclature prior to import. All of the clinical annotations associated with each sample were then imported into a relational database.
The frozen aliquots of clinical samples were thawed and aliquotted for use in the laboratory. Each clinical sample was thawed on ice, and aliquots were dispensed into barcoded tubes (daughter tubes). Each daughter tube was stored at −80 C until it was needed for the immunoassays. The daughter tubes were then arrayed into sample plates. Each barcoded daughter tube to be assayed was arrayed into barcoded 96 or 384 well plates (sample plates). The daughter tube to sample plate well mapping was tracked by the relational database.
Each sample plate was prepared for immunoassay analysis. The 384 well barcoded assay plates were dedicated to one biomarker per plate. Typically, 4-12 assay plates were derived from each sample plate depending upon the amount of serum required for each assay. The sample plate went through a series of dilutions to ensure that the clinical samples were at an appropriate dilution for each immunoassay. The clinical samples were then deposited into the assay plate wells in triplicate for each marker. Again, tracking of each sample plate well to assay plate well was tracked in the relational database. The assays were then be processed using standard immunoassay procedures, and the assay plate was read on the Zeptosense instrument. Each run contained data for a single biomarker across about 384 clinical samples. The resulting data files were then imported back into the relational database, where standard curves were calculated and the concentration values for each biomarker for each sample were calculated.
The biomarker values assigned to each clinical sample were reassociated with the original clinical annotations. The quantitative biomarker data were correlated to the clinical annotations associated with each sample. Diabetes risk over 7.5 years was calculated using the model developed by Stern et al. (2002). The clinical model is of the form of a logistic equation
p=1/(1+e−x),
where
x=−13.415+0.028(age)+0.661(sex)+0.412(MA)+0.079(FG)+0.018(SBP)−0.039(HDL)+0.070(BMI)+0.481(family history).
In this equation, p=the probability of developing diabetes over the 7.5 year follow-up period; age is in years; sex=1 if female, 0 if male; MA=1 if Mexican American, 0 if non-Hispanic white; FG=fasting glucose in mg/dL; SBP=systolic blood pressure in mm Hg; HDL=high-density lipoprotein cholesterol level in mg/dL; BMI=body mass index in kg/m2; and family history=1 if at least one parent or sibling has diabetes or 0 if not (Stem et al. 2002).
In order to estimate risk for the cohort patient samples, the following modifications were made to these parameters. First, African Americans and Hispanics were included in the high risk group with Mexican Americans and patients with a diagnosis of hypertension were assumed to have a SBP=150 and patients without an SBP=125. The rest of the data were available in the clinical record. Raw concentration data for each marker were log10 transformed and used as the inputs for several linear regression models on the logit transfom of risk (x in the above equation).
Linear regression of x on the log10 biomarker concentration on each univariate, bivariate, and tri-variate basis by marker sets was performed via a complete search of all combinations. The quality of models was judged on the basis of the coefficient of determination, R2.
Models larger than three markers were developed using forward, backward, and stepwise selection based on Akaike Information Criterion (AIC). Alternatives to these marker sets were identified by eliminating each marker and searching the remaining set for the best replacement, where ‘best’ is the marker with the highest R2 value.
A full model was also created by adding a single variable to the null model one by one until all markers were used. Each marker was selected based on the coefficient of determination of the complete marker set being used up to that point. Selected fits of these models were used to calculate sensitivity and specificity of any individual model.
The uniqueness/similarity of biomarker concentrations was investigated using principle components analysis (PCA), Hierarchical clustering, and simple correlation. The results of the PCA were evaluated graphically using scree plots, bi-plots, and sample projections to quantify how much independent variation existed among these markers. Hierarchical clustering, using the standardized (mean=0, sd=1) concentrations, was based on euclidian distance as a distance metric and Ward's method as the means of agglomeration. Clusters were used to identify markers behaving similarly.
The following is an illustrative example of a method that was used in developing protein biomarker tests in accordance with the invention.
Assay Analyte: C-Reactive Protein
Each individual well on a NUNC Maxisorp 384-well plate was coated with 20 μl of capture antibody diluted in coating buffer (0.05 M carbonate, pH 9.6; diluted to 1 μg/mL and prepared immediately before use) and incubated overnight at room temperature. The plate was then washed three times in 100 μl of Wash buffer A (PBS with 0.1% Tween 20), and blocked in 30 μl PBS buffer containing 1% BSA, 5% sucrose, 0.05% NaN3 for analyte capture for at least two hours at room temperature. After incubation, blocking buffer was removed and the blocked plates air-dried for at least 5 hours at room temperature and prepared for storage at 4° C. or for Zeptosense assay.
Samples were diluted 1:400 in Assay Buffer (BS buffer containing 1% BSA, 0.1% Triton X-100. To the blocked and dried plate, 20 μl/well of standards and diluted unknown samples were added and allowed to incubate overnight at room temperature. After incubation, the plate washed five times in wash buffer B (BS buffer with 0.02% Triton X-100 and 0.0001% BSA), detection antibody A647 was diluted to 50 ng/ml in assay buffer and was added to the wells in an amount of 20 μl/well. The detection antibody was allowed to bind for 2 hours at room temperature, after which the plate washed five times in 100 μl of wash buffer B. A standard curve was generated using a control diluted to 100 ng/ml in a calibrator. Serial dilutions from 100 ng/ml to 0.01 pg/ml in calibrator diluent (assay buffer+additional 5% BSA) were prepared.
It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.
This application claims priority from U.S. Provisional Application Ser. No. 60/725,462, filed on Oct. 11, 2005. Each of the applications and patents cited in this text, as well as each document or reference cited in each of the applications and patents (including during the prosecution of each issued patent; “application cited documents”), and each of the U.S. and foreign applications or patents corresponding to and/or claiming priority from any of these applications and patents, and each of the documents cited or referenced in each of the application cited documents, are hereby expressly incorporated herein by reference. More generally, documents or references are cited in this text, either in a Reference List before the claims, or in the text itself; and, each of these documents or references (“herein-cited references”), as well as each document or reference cited in each of the herein-cited references (including any manufacturer's specifications, instructions, etc.), is hereby expressly incorporated herein by reference. Documents incorporated by reference into this text may be employed in the practice of the invention.
| Number | Date | Country | |
|---|---|---|---|
| 60725462 | Oct 2005 | US |
