HK1152751A - Methods for detecting major adverse cardiovascular and cerebrovascular events - Google Patents

Description

Method for detecting serious adverse cardiovascular and cerebrovascular events

RELATED APPLICATIONS

This application claims priority and benefit from U.S. provisional patent application serial No. 60/998,563 filed on 10/2007 and U.S. provisional patent application serial No. 60/998,756 filed on 10/11/2007, each of which is hereby incorporated by reference in its entirety, under section 35U.S. C.119 (e).

Technical Field

The present application relates to methods for predicting whether a human will suffer from a severely adverse cardiovascular or cerebrovascular event (or MACCE). More particularly, the present application relates to methods of screening individuals at risk of having or developing a severely adverse cardiovascular or cerebrovascular event by using one or more analytes.

Background

Heart attack is the single leading cause of death (seewww.americanheart.org). One of every 5 deaths in the united states is caused by a heart attack. In 2004, 452,327 deaths in the united states were due to heart attacks resulting from approximately 1,200,000 new and recurrent cardiovascular events.

Stroke is the third leading cause of death in the United states (seewww.americanheart.org). In 2004, stroke caused by approximately 700,000 newly-occurring and recurrent cerebrovascular events killed 150,147 individuals. Stroke is the leading cause of severe long-term disability in the united states. About 5,700,000 survivors of stroke still live today in the United states. 2,400,000 are male, and 3,300,000 are female.

Heart attacks and certain types of strokes can be caused by the rupture of a vulnerable atherosclerotic plaque (Naghavi et al Circulation 108: 1664-72& 108: 1772-8, 2003). Currently, the risk of having a heart attack or stroke is assessed in the general population by considering certain clinical and biochemical risk factors (Wilson et al, Circulation 97: 1837-47, 1998; ATP III, JAMA 285: 2486-97, 2001), but these features do not fully account for cardiovascular risk (Khot et al, JAMA 290: 898-.

If the ability to predict future heart attacks or stroke could be improved, it would be possible to take preventative measures to individuals at risk that the overall incidence of the leading cause of these deaths would be reduced.

Measurement of various proteins and metabolites in the blood of an individual provides a "window" for viewing the biochemical status of the individual, and may provide a better indication of the status of his or her cardiovascular system and the likelihood that the subject will experience a future heart attack or stroke (Vasan, Circulation 113: 2335-62, 2006). Based on this principle, we have conducted the following study with the following objectives: finding a molecular biomarker profile (e.g., a set of proteins, metabolites, a set of proteins and metabolites, or a set of other analytes that may include proteins and/or metabolites) in blood or plasma; related algorithms were found to predict recent serious adverse cardiovascular or cerebrovascular events (MACCE).

Thus, there remains a need for diagnostic methods for predicting severe acute cardiac events. In particular, there is a need for reliable and cost-effective methods and compositions to provide diagnosis and/or prediction of severely adverse cardiovascular and cerebrovascular events.

Brief description of the invention

The present invention relates to methods for predicting that an individual will suffer from a major adverse cardiovascular or cerebrovascular event (or MACCE). The method involves measuring the levels of one or more of certain analytes (e.g., proteins and metabolites) in a plasma or serum sample derived from a blood sample of the individual, and then employing a decision algorithm to predict whether the individual will likely experience MACCE.

Markers have been identified in the blood that can be used to predict the likelihood of a major adverse cardiovascular or cerebrovascular event in a subject. When the levels of these markers differ from the norm, a severe adverse cardiovascular or cerebrovascular event is predicted. The methods of the invention utilize one or more analytes to predict a severely adverse cardiovascular or cerebrovascular event. In particular, subjects at risk of developing a severely adverse cardiovascular or cerebrovascular event can be screened by using one or more of these analytes and determining whether the levels of these analytes differ from the standard.

The analytes of the present invention are not only useful for predicting the likelihood of a serious adverse cardiovascular or cerebrovascular event. The analytes may be used to screen for candidate drugs that prevent serious adverse cardiovascular or cerebrovascular events. The analyte may also be used to identify a subject in need of monitoring his health. Likewise, the analyte may be used to validate animal models for coronary artery disease or cerebrovascular disease.

In one aspect, the invention provides a method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human. The method includes measuring a concentration of a set of analytes in a blood-based sample of a person. The set of analytes may include alpha-fetoprotein, cancer antigen 125, glutathione S-transferase, and tissue factor. The method further includes determining a MACCE index for the set of analytes, identifying the human as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event if the MACCE index is greater than zero, or identifying the human as having a decreased likelihood of a major adverse cardiovascular or cerebrovascular event if the MACCE index is less than or equal to zero.

In another aspect, the present invention provides a method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, wherein the method comprises determining a MACCE index having a value indicative of the likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentration of a set of analytes in a blood-based sample of a human. The set of analytes may include alpha-fetoprotein, cancer antigen 125, glutathione S-transferase, and tissue factor. The method further includes transmitting, displaying, storing or outputting at least one of the MACCE index, the likelihood of a major adverse cardiovascular or cerebrovascular event, or an equivalent thereof, to a user's interface device, computer-readable storage medium, or a local or remote computer system.

In yet another aspect, the invention provides a method of treating a human. The method includes determining a MACCE index having a value indicative of the likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentration of a set of analytes in a blood-based sample of a human. The set of analytes may include alpha-fetoprotein, cancer antigen 125, glutathione S-transferase, and tissue factor. The method may further comprise recommending, approving or administering treatment if the human is identified as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event.

According to various embodiments, the collection of analytes can also include CD40, fibrinogen, IL-3, IL-8, SGOT, and von Willebrand factor (von Willebrand factor). According to certain embodiments, determining a MACCE index for a set of analytes may include normalizing the measured concentration of each analyte to obtain a normalized concentration, multiplying the normalized concentration of each analyte by an analyte constant to obtain a value for the analyte, and then adding the analyte values for each analyte to obtain the MACCE index. According to certain embodiments, normalizing the measured concentration may comprise subtracting the population mean from the measured concentration to obtain a result, dividing the result by the standard deviation of the population mean.

In certain embodiments, the method comprises measuring the concentration of a set of analytes in a blood-based sample of a human, wherein the set of analytes can consist of: alpha-fetoprotein, cancer antigen 125, CD40, fibrinogen, glutathione S-transferase, IL-3, IL-8, SGOT, tissue factor, and von Willebrand factor. The method further includes determining a MACCE index for the set of analytes, identifying the human as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event if the MACCE index is greater than zero, or identifying the human as having a decreased likelihood of a major adverse cardiovascular or cerebrovascular event if the MACCE index is less than or equal to zero.

In yet another aspect, the invention provides a method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, wherein the method comprises determining a MACCE index having a value indicative of the likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentration of a set of analytes in a blood-based sample of the human. The collection of analytes may be comprised of: alpha-fetoprotein, cancer antigen 125, CD40, fibrinogen, glutathione S-transferase, IL-3, IL-8, SGOT, tissue factor, and von Willebrand factor. The method further includes transmitting, displaying, storing or outputting at least one of the MACCE index, the likelihood of a major adverse cardiovascular or cerebrovascular event, or an equivalent thereof, to a user's interface device, computer-readable storage medium, or a local or remote computer system.

Another aspect of the invention provides a method of treating a human, wherein the method comprises determining a MACCE index having a value indicative of the likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentration of a set of analytes in a blood-based sample of the human. The collection of analytes may be comprised of: alpha-fetoprotein, cancer antigen 125, CD40, fibrinogen, glutathione S-transferase, IL-3, IL-8, SGOT, tissue factor, and von Willebrand factor. The method may further comprise recommending, approving or administering treatment if the person is identified as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event.

According to various embodiments of the present invention, determining a MACCE index for a set of analytes may include normalizing the measured concentration of each analyte to obtain a normalized concentration, multiplying the normalized concentration of each analyte by a constant for the analyte to obtain a value for the analyte, and then adding the analyte values for each analyte to obtain the MACCE index. According to certain embodiments, normalizing the measured concentration comprises subtracting the population mean from the measured concentration to obtain a result, dividing the result by the standard deviation of the population mean.

In other aspects, the invention provides methods of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, wherein the methods comprise measuring the concentration of at least one analyte in a collection of analytes in a blood-based sample of the human. The analyte collection consisted of: cysteine, von Willebrand factor, IL-8, 16:0/18:1 phosphatidylcholine, N-carboxy-alanine, fibrinogen, MMP-2, 18:0/20:4 phosphatidylethanolamine, apolipoprotein A1, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, alpha-1 antitrypsin, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 20:4 lysophosphatidylcholine, 16:0 sphingomyelin, SHBG, 18:1/17:1, 16:0 triacylglycerol, arabinose, and 18:1/18:1/17:0 triacylglycerol. The method further includes identifying the person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on comparing the measured concentration to a predetermined threshold.

In another aspect, the invention provides a method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, wherein the method comprises comparing the measured concentration of at least one analyte of a set of analytes in a blood-based sample of the human with a predetermined threshold value to identify the likelihood of a major adverse cardiovascular or cerebrovascular event. The set of analytes is selected from the following: cysteine, von Willebrand factor, IL-8, 16:0/18:1 phosphatidylcholine, N-carboxy-alanine, fibrinogen, MMP-2, 18:0/20:4 phosphatidylethanolamine, apolipoprotein A1, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, alpha-1 antitrypsin, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 20:4 lysophosphatidylcholine, 16:0 sphingomyelin, SHBG, 18:1/17:1, 16:0 triacylglycerol, arabinose, and 18:1/18:1/17:0 triacylglycerol. The method may further include transmitting, displaying, storing, or outputting at least one of the measured concentration, the predetermined threshold, the likelihood of a major adverse cardiovascular or cerebrovascular event to a user interface device, a computer readable storage medium, or a local or remote computer system.

In a further aspect, the present invention provides a method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, wherein the method comprises measuring the concentration of at least one analyte selected from the group consisting of: 16:0/18:1 phosphatidylcholine, 18:0/20:4 phosphatidylethanolamine, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 20:4 lysophosphatidylcholine, 16:0 sphingomyelin, 18:1/17:1/16:0 triacylglycerol, and 18:1/18:1/17:0 triacylglycerol. The method may further comprise identifying the human as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on comparing the measured concentration to a predetermined threshold.

In yet another aspect, the invention provides a method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, wherein the method comprises comparing the measured concentration of at least one analyte of a set of analytes in a blood-based sample of the human with a predetermined threshold value to identify the likelihood of a major adverse cardiovascular or cerebrovascular event. The analyte is selected from: 16:0/18:1 phosphatidylcholine, 18:0/20:4 phosphatidylethanolamine, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 20:4 lysophosphatidylcholine, 16:0 sphingomyelin, 18:1/17:1/16:0 triacylglycerol, and 18:1/18:1/17:0 triacylglycerol. The method may further include transmitting, displaying, storing, or outputting at least one of the measured concentration, the predetermined threshold, the likelihood of a major adverse cardiovascular or cerebrovascular event to a user interface device, a computer readable storage medium, or a local or remote computer system.

According to various embodiments, the predetermined threshold value for each of the following analytes may be the lower limit of the fourth quartile of each corresponding analyte in table 4: cysteine, von Willebrand factor, IL-8, 16:0/18:1 phosphatidylcholine, N-carboxy-alanine, fibrinogen, MMP-2, 18:0/20:4 phosphatidylethanolamine, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, alpha-1 antitrypsin, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 16:0 sphingomyelin, SHBG, 18:1/17:1, 16:0 triacylglycerol, and 18:1/18:1/17:0 triacylglycerol, wherein the measured concentration falling within the fourth quartile increases the likelihood of a serious adverse cardiovascular or cerebrovascular event. According to certain embodiments, the predetermined threshold value for each of the following analytes may be the lower limit of the third and fourth quartile for each respective analyte in table 4: cysteine, von Willebrand factor, IL-8, 16:0/18:1 phosphatidylcholine, N-carboxy-alanine, fibrinogen, MMP-2, 18:0/20:4 phosphatidylethanolamine, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, alpha-1 antitrypsin, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 16:0 sphingomyelin, SHBG, 18:1/17:1, 16:0 triacylglycerol, and 18:1/18:1/17:0 triacylglycerol, wherein the measured concentrations falling within the third and fourth quartiles increase the likelihood of a major adverse cardiovascular or cerebrovascular event.

According to various embodiments, the predetermined threshold for each of the analytes apolipoprotein a1, 20:4 lysophosphatidylcholine, and arabinose may be the upper limit of the first quartile of each corresponding analyte in table 4, wherein the measured concentration falling within the first quartile increases the likelihood of a serious adverse cardiovascular or cerebrovascular event. According to certain embodiments, the predetermined threshold for each of the analytes apolipoprotein a1, 20:4 lysophosphatidylcholine, and arabinose may be an upper limit of the first and second quartile of each respective analyte in table 4, wherein the measured concentration falling within the first and second quartile increases the likelihood of a major adverse cardiovascular or cerebrovascular event.

In various embodiments of the invention, the blood-based sample may be serum or plasma.

In yet another aspect, the invention provides a method for predicting a major adverse cardiovascular or cerebrovascular event in a human, wherein the method can comprise obtaining at least one blood-based sample from the human, measuring the absolute concentration of one or more analytes identified in appendix 1 in the sample from the human, and identifying the human as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the absolute concentration of the one or more analytes identified in appendix 1.

In yet another aspect, the invention provides a method for predicting a major adverse cardiovascular or cerebrovascular event in a human, wherein the method comprises measuring in a sample from the human an absolute concentration of one or more analytes identified in appendix 1, comparing the absolute concentration to a predetermined threshold value, and identifying the human as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the absolute concentration of the one or more analytes identified in appendix 1, and transmitting, displaying, storing or outputting at least one of the likelihood of a major adverse cardiovascular or cerebrovascular event, the absolute concentration, the predetermined threshold value, or an equivalent thereof, to an interface device, computer readable storage medium, or a local or remote computer system of a user.

In certain aspects, the invention provides a method for predicting a major adverse cardiovascular or cerebrovascular event in a human, wherein the method comprises obtaining at least one blood-based sample from the human, measuring the relative concentrations of one or more analytes identified in appendix 1 in said sample from the human, and identifying the human as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the relative concentrations of the one or more analytes identified in appendix 1.

In yet another aspect, the invention provides a method for predicting a major adverse cardiovascular or cerebrovascular event in a human, wherein the method comprises measuring in a sample from the human the relative concentration of one or more analytes identified in appendix 1, comparing the relative concentration to a predetermined threshold, identifying the human as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the relative concentration of the one or more analytes identified in appendix 1, and transmitting, displaying, storing or outputting at least one of the likelihood of a major adverse cardiovascular or cerebrovascular event, the relative concentration, the predetermined threshold, or an equivalent thereof, to an interface device, computer readable storage medium, or a local or remote computer system of a user.

According to various embodiments of the method, the MACCE index, the likelihood of a major adverse cardiovascular or cerebrovascular event, the measured concentration, the predetermined threshold, or an equivalent thereof is displayed on a screen or tangible medium, or the MACCE index is transmitted to a medical industry personnel, a person providing medical insurance, or a physician.

In another aspect, the invention provides a method of treatment, wherein the method can comprise identifying a person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the concentration of the one or more analytes identified in appendix 1 measured in a blood-based sample, and recommending, approving or administering treatment if the person is identified as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event.

In yet another aspect, the invention provides a method of identifying a person who should not receive treatment, wherein the method can comprise identifying a person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the concentration of the one or more analytes identified in appendix 1 measured in a blood-based sample, and rejecting the recommendation, approval, or administration of treatment unless the person is identified as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event.

The foregoing and other features and advantages of the invention will be more fully understood from the following drawings, description, examples and claims.

Brief Description of Drawings

The foregoing and other objects, features and advantages of the invention will be more fully understood from the following description of various illustrative embodiments, when read together with the accompanying drawings. In the drawings, for example, reference characters generally refer to the same parts throughout the different views. It should be understood that the drawings described below are for illustration purposes only. The drawings are not intended to limit the scope of the present invention in any way.

Fig. 1 is a graph depicting the Receiver Operating Characteristic (ROC) curve for predicting the occurrence of MACCE over two years using the levels of the first 20 analytes of table 5 in plasma samples according to exemplary embodiments of the present invention.

Figure 2 is a graph depicting the Receiver Operating Characteristic (ROC) curve for predicting the occurrence of MACCE over two years using the levels of the top 10 protein analytes of table 6 in plasma samples in an exemplary embodiment of the invention.

Detailed description of the invention

Analytes predictive of a severely adverse cardiovascular or cerebrovascular event have been identified. When one or more of these analytes are present in a body fluid sample from an individual in an amount different from the standard amount, they may indicate that the individual is at risk of having or developing a severely adverse cardiovascular or cerebrovascular event. The present application incorporates herein by reference in its entirety each of the following patent applications: U.S. provisional patent application serial No. 60/998,563 filed on 10/2007 and U.S. provisional patent application serial No. 60/998,756 filed on 11/10/2007.

Throughout the specification, when a composition is described as having, including, or comprising specific components, or when a method is described as having, including, or comprising specific method steps, it is intended that the composition of the invention also consists essentially of, or consists of, the recited components, and that the method of the invention also consists essentially of, or consists of, the recited method steps.

In the present application, when an element or component is considered to be included in and/or selected from a list of such elements or components, it is to be understood that the element or component can be any one of the elements or components and can be selected from two or more of the elements or components. In addition, it will be understood that elements and/or features of the compositions, apparatus or methods described herein may be combined in various ways without departing from the spirit and scope of the invention, whether explicitly described or implied herein.

The use of the terms "comprising," "including," "having," "containing," and "containing" are generally to be construed as open-ended and not limiting unless expressly stated otherwise.

Unless specifically stated otherwise, use of the singular herein includes the plural (and vice versa). In addition, the invention also includes the particular numerical values themselves when the term "about" is used before a numerical value, unless specifically stated otherwise. The term "about" as used herein refers to a deviation of ± 10% of the nominal value.

It will be understood that the order of steps or order of performing certain actions is immaterial so long as the invention can still be practiced. In addition, two or more steps or acts may be performed simultaneously.

Seven hundred twenty-three (723) analytes were identified. Certain methods of the invention utilize two or more of these analytes to predict a severely adverse cardiovascular or cerebrovascular event. In particular, in these methods, a subject sample can be screened to determine whether the level of each of two or more of the seven hundred twenty-three (723) analytes in the sample differs from a standard sample. A screen is considered a "positive screen" (i.e., an individual is at risk of developing a major adverse cardiovascular or cerebrovascular event) if the sample contains an amount of each of two or more of these analytes that is different from the standard amount of these analytes. Greater sensitivity and specificity can be obtained in classifying samples of severely adverse cardiovascular or cerebrovascular events, typically by using more analyte species. Samples that may contain these analytes may be derived from a variety of biological samples (e.g., bodily fluids, tissues, cells) obtained from a variety of sources (e.g., whole blood, plasma, serum, urine, cerebrospinal fluid, epithelial cells, and endothelial cells). It should be understood that all possible combinations of the seven hundred twenty three (723) analytes disclosed herein (not just the fifty (50) analytes identified in table 5) may be used in the methods of the invention.

Seven hundred twenty-three (723) analytes were identified using the method detailed in example 1. Briefly, a collection of seven hundred twenty-three (723) analytes were analyzed by a specific classification analysis protocol to obtain spectral peaks. Appendices 1 to 4 provide specific molecules containing the seven hundred twenty three (723) analytes analyzed. These peaks characterize a particular molecule. Fifty peaks (and other preferred peaks described below) were identified using these peaks (which included all 723 analytes). With respect to peak characterizing molecules, 50 peaks are identified from a much larger number of peaks, meaning that the analyte is selected from a group of molecules comprising more than 50 molecules corresponding to the 50 analytes. The analyte (i.e., molecule) can be any type of molecule. Analytes (molecules) include, but are not limited to: proteins, peptides, amino acids, lipids, steroids, nucleic acids, metabolites, and elements. Table 5 provides specific molecules comprising the 50 peaks (i.e., identifying the selected analyte). Column 2 of table 5 ranks the peaks by weight. Thus, the highest weighted peak is ranked at1 st and the lowest weighted peak is ranked at 50 th.

Now that the analytes are known (i.e., any of the seven hundred twenty three analytes identified, or preferably any of the 50 analytes identified), they can be used to screen individuals to determine whether the amount or absolute concentration of each of two or more of these analytes in an individual's sample differs from the amount of each of the two or more standard analytes to determine the relative concentration of each of the two or more analytes in an individual's sample as compared to the standard, defining the individual as having or at risk of developing a severely adverse cardiovascular or cerebrovascular event with a certain specificity and sensitivity. Of course, the measured amount or concentration of analyte may be normalized prior to the comparison. The sensitivity and specificity required for the assay can be selected based on the amount of analyte examined. The criteria may be actual samples or previously generated empirical data. The standard may be from a person known to be normal. It is known that a normal person may be a healthy person and may have taken a predetermined diet at a predetermined time before sampling. In addition, the sample may be obtained from a known normal human of the same sex as the subject. Alternatively, the analyte may be compared to an analyte from a subject with a known major adverse cardiovascular or cerebrovascular event, in which case the similarity between the two samples will be examined, or the relative concentration of the analyte compared to a standard. Various techniques and/or kits may be used by a medical practitioner to screen a subject sample in order to determine the level and/or amount of a particular analyte in the subject sample. Examples of such assays are set forth below, including but not limited to immunoassays, mass spectrometry, chromatography, chemical analysis, colorimetric analysis, spectrophotometric analysis, electrochemical analysis, and nuclear magnetic resonance. In addition, the assay can be performed on any biological sample, including whole blood, plasma, serum, cerebrospinal fluid, saliva, urine, semen, nipple secretions, pancreatic juice, and combinations thereof. These assays are selected based on their best suitability for detecting a particular analyte and their best suitability for use with a particular biological sample. Thus, a variety of assays may be used to detect a desired analyte, and samples from one or more sources may be analyzed.

Analytes can be detected and/or quantified by using one or more separation methods. For example, suitable separation methods can include mass spectrometry methods, such as ionization spray mass spectrometry (ESI-MS), ESI-MS/MS, ESI-MS/(MS)ⁿ(n is an integer greater than zero), matrix assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS), surface enhanced laser desorption ionization time-of-flight mass spectrometry (SELDI-TOF-MS), silicon surface desorption/ionization (DIOS), Secondary Ion Mass Spectrometry (SIMS), quadrupole time-of-flight (Q-TOF), atmospheric pressure chemical ionization mass spectrometry (APCI-MS), APCI-MS/MS, APCI- (MS)ⁿAtmospheric pressure photoionization mass spectrometry (APPI-MS), APPI-MS/MS and APPI- (MS)ⁿ. Other mass analysis methods may include, among others, quadrupoles, Fourier Transform Mass Spectrometry (FTMS), and ion traps. Spectroscopic techniques that may also be used include magnetic resonance spectroscopy and optical spectroscopy.

Other suitable separation methods include chemical extraction partitioning, column chromatography, ion exchange chromatography, hydrophobic (reverse phase) liquid chromatography, isoelectric focusing, one-way polyacrylamide gel electrophoresis (PAGE), two-way polyacrylamide gel electrophoresis (2D-PAGE), or other chromatographic techniques (e.g., thin layer, gas or liquid chromatography), or any combination thereof. In one embodiment, the biological sample to be assayed may be fractionated prior to applying the separation method.

Chromatography (e.g., liquid chromatography ("LC")) and mass spectrometry ("MS") tandem can be used to detect and quantify one or more analytes. LC can be used to separate molecules that may include an analyte in a sample from an individual. A small amount of sample dissolved in a solvent may be injected into an injection port of an LC device that may be maintained at an elevated temperature. The LC column of the device contains a solid matrix which may be polar or non-polar. The molecules will elute at different times because of their different polarity in the sample and their different affinity for the solid matrix in the column. The stronger the affinity of the molecule for the matrix, the longer the retention time of the molecule in the column. As the molecules exit the column, they enter the mass spectrometer. The mass spectrometer ionizes the molecules. In tandem mass spectrometry mode, each compound entering the mass spectrometer breaks into ions of various masses and abundances, if the system can be properly calibrated, forming a signature (signature pattern) unique to that substance. By comparing each peak of the tandem mass spectrometer to a computerized database, computers are generally able to identify the molecules with high confidence. Alternatively or additionally, such comparison may be performed by manual detection. Once the identity is determined, the computer integrates the area under each peak, thereby determining the relative amount of each molecule in the mixture. When any of the molecules can be identified as an analyte, the amount of the analyte can be compared to the amount of analyte from the standard to determine if there is a difference.

Analytes can also be detected and/or quantified by methods that do not require physical separation of the analyte itself. For example, Nuclear Magnetic Resonance (NMR) spectroscopy can be used to resolve the analyte profile (profile) from a mixture of complex molecules. For example, similar applications of NMR to define tumors are disclosed in Hagberg, NMR biomed.11: 148-56 (1998). Additional methods include nucleic acid amplification techniques that can be used to determine analyte profiles without physical separation of individual molecules. (e.g. inSee alsoStordreur et al, j.immunol.methods 259: 55-64(2002) and Tan et al, Proc. nat' l Acad. Sci. USA 99: 11387-11392(2002)).

For example, an analyte in a sample can also be detected and/or quantified, e.g., by binding the analyte to a binding moiety capable of specifically binding the analyte. Binding moieties can include, for example, members of a ligand-receptor pair (i.e., a pair of molecules capable of specific binding). Binding moieties may also include, for example, members of specific binding pairs, such as antibody-antigen, enzyme-substrate, nucleic acid-nucleic acid, protein-protein, or other specific binding pairs known in the art. Binding proteins with increased affinity for the target can be designed. Optionally, the binding moiety may be linked to a detectable label, such as an enzymatic label, a fluorescent label, a radioactive label, a phosphorescent label, or a colored particle label. The labeled complex can be detected, for example, visually or with the aid of a spectrophotometer or other detector, and/or the labeled complex can be quantified.

Analytes can also be detected and/or quantified using gel electrophoresis techniques available in the art. In two-dimensional gel electrophoresis, molecules are first separated according to their isoelectric points in a pH gradient gel. The resulting gel is then placed on a second polyacrylamide gel and the molecules are separated according to molecular weight (c)Ginseng radix (Panax ginseng C.A. Meyer) SeeFor example, O' Farrell j. biol. chem.250: 4007-4021(1975)). Analytes of a major adverse cardiovascular or cerebrovascular event can be detected by first isolating molecules from a sample obtained from an individual suspected of being at risk for a major adverse cardiovascular or cerebrovascular event, and then separating the molecules by two-dimensional gel electrophoresis to generate a characteristic two-dimensional gel electrophoresis pattern. This profile is then compared to a standard gel profile generated by isolating the molecules isolated from a standard (e.g., a healthy or severe acute cardiac event subject) under the same or similar conditions. Standard gel profiles can be stored and retrieved from the electropherogram electronic database. Thus, it can be determined whether the amount of the marker in the subject is different from the amount in the standard. The presence of multiple (e.g., 2-50) severely adverse cardiovascular or cerebrovascular event analytes on a bi-directional gel in amounts different from those in known normal standards indicates an individual is severely adverse cardiovascular or brain eventsPositive screening for vascular events. Thus, the detection allows prediction and treatment of severely adverse cardiovascular or cerebrovascular events.

The analyte may be detected and/or quantified using any of a variety of immunoassay techniques useful in the art. For example, a sandwich immunoassay format can be used to detect and/or quantify an analyte in a subject sample. Alternatively, conventional immunohistochemistry procedures may be used with one or more labeled binding proteins to detect and/or quantify the presence of an analyte in a sample.

In sandwich immunoassays, two antibodies capable of binding the analyte are typically used, e.g., one immobilized on a solid support and the other free in solution and labeled with a detectable chemical compound. Examples of chemical labels that may be used for the second antibody include radioisotopes, fluorescent compounds and enzymes, or other molecules that produce a colored or electrochemically active product when contacted with a reactant or enzyme substrate. When a sample containing the analyte is placed in the system, the analyte binds both the immobilized and the labeled antibody, forming a "sandwich" immune complex on the support surface. The analyte forming the complex is detected by washing away unbound sample components and excess labeled antibody and measuring the amount of labeled antibody that forms a complex with the analyte on the surface of the support. Alternatively, an antibody free in solution that is labeled with a chemical moiety (e.g., a hapten) can be detected by a third antibody labeled with a detectable moiety that binds to the free antibody or, for example, a hapten conjugated thereto.

Both sandwich immunoassays and tissue immunohistochemistry methods are highly specific and extremely sensitive, provided that labels with good detection limits are used. A detailed overview of immunoassay design, theory, and protocol can be found in a number of textbooks in the art, including Butt, w.r.,Practical Immunology，edited by Marcel Dekker, New York (1984) and Harlow et alAntibodies，A LaboratoryApproach，Edited by Cold Spring harbor laboratory (1988).

Generally, a matter of design consideration for immunoassays includes the preparation of antibodies (e.g., monoclonal or polyclonal antibodies) that have sufficiently high binding specificity for a target to form a complex that can be reliably distinguished from non-specific interacting products. The term "antibody" as used herein is understood to mean a binding protein, such as an antibody or other protein comprising an immunoglobulin variable region-like binding domain with appropriate binding affinity and specificity for a target. The higher the antibody binding specificity, the lower the concentration of target that can be detected. The term "specific binding" as used herein is understood to mean that the binding affinity of a binding moiety (e.g., a binding protein) for a target is greater than about 10⁵M^-1More preferably greater than about 10⁷M^-1。

Antibodies to isolated target analytes for assays to predict an individual's serious adverse cardiovascular or cerebrovascular events can be generated using standard immunological procedures well known and described in the art. See, e.g.Practical ImmunologySee above. Briefly, the isolated analyte may be used to produce antibodies in a xenogeneic host, such as a mouse, goat or other suitable mammal. The analyte may be combined with a suitable adjuvant capable of increasing the yield of antibody in the host and may be injected into the host, for example, by intraperitoneal administration. Any adjuvant suitable for stimulating an immune response in a host may be used. Commonly used adjuvants are complete Freund's adjuvant (emulsions containing inactivated and dried microbial cells, available from, for example, Calbiochem Corp., San Diego or Gibco, Grand Island, NY). Where multiple antigens need to be injected, subsequent injections may contain the antigen in combination with an incomplete adjuvant (e.g., a cell-free emulsion). Polyclonal antibodies can be isolated from a host producing the antibodies by extracting serum containing antibodies to the protein of interest. Monoclonal antibodies can be generated as follows: host cells producing the desired antibody are isolated, fused with myeloma cells using standard procedures known in the immunization art, and selected for hybrid cells (hybridomas) that specifically react with the target and have the desired binding affinity.

Antibody binding domains can also be biosynthetically generated, and the amino acid sequence of the binding domain manipulated to enhance binding affinity to a target-preferred epitope. Specific antibody methods are well understood and are described in the literature. More detailed descriptions about their preparation can be found, for example, inPractical Immunology(see above).

Alternatively, the sample can be assayed for the presence of an analyte using genetically engineered biosynthetic antibody binding sites known in the art, such as the binding site for BABS or sFv. Methods of making and using BABS comprising: (i) v non-covalently associated or linked via disulfide bonds_HAnd V_LA dimer; (ii) covalently linked V_H-V_LA single-stranded binding site; (iii) v alone_HOr V_LA domain; or (iv) a single chain antibody binding site. In addition, BABS can be obtained with the necessary specificity for the analyte by cloning phage antibodies from combinatorial gene libraries (see, e.g., Clackson et alNature352: 624-628(1991)). Briefly, phages, each expressing BABS on its capsid surface with immunoglobulin variable regions encoded by variable region gene sequences derived from mice pre-immunized with an isolated analyte or fragment thereof, are screened for binding activity to the immobilized analyte. Phage that bind the immobilized analyte are harvested and the BABS-encoding gene can be sequenced. The resulting nucleic acid sequence encoding a BABS of interest can be expressed using conventional expression systems to produce a BABS protein.

The isolated analytes can also be used to develop diagnostic and other tissue evaluation kits and assays to monitor the level of the analyte in a tissue or fluid sample. For example, a kit may include an antibody or other specific binding protein that specifically binds to one or more analytes and allows for the detection and/or quantification of the presence and/or amount of the one or more analytes in a tissue or fluid sample.

Suitable kit protocols for detecting one or more analytes include, but are not limited to: containers or other means for capturing a sample to be assessed (means) and means for detecting the presence and/or amount of one or more analytes described herein in a sample. Means for detection in one embodiment include, but are not limited to, one or more antibodies specific for these analytes and means for detecting binding of the antibodies to these analytes by, for example, a standard sandwich immunoassay as described herein. When the presence of an analyte located within a cell (e.g., from a tissue sample) is to be detected, the kit may further include means for disrupting cellular structure so as to expose intracellular components.

The analyte of the present invention may comprise a nucleic acid of a specific sequence. One or more analytes can be detected and/or quantified by determining (using, e.g., real-time quantitative PCR (RT-PCR)) the amount or absolute concentration of analyte nucleic acid in a sample, and comparing the measured amount to a standard to determine the relative concentration of analyte nucleic acid in the sample. RT-PCR effectively measures the amount of analyte nucleic acid generated by PCR. A positive result represents that the amount of analyte nucleic acid measured is different from the amount of analyte from the standard, or represents a relative concentration whose value is above or below zero.

Primers can be developed that are complementary to the nucleic acid sequence of a particular nucleic acid analyte. These primers direct the polymerase to replicate and amplify the particular nucleic acid. RT-PCR detects the cumulative amount of nucleic acid analyte amplified during the reaction. During the exponential phase of the PCR reaction, the accumulated nucleic acid analyte can be measured. Calibration standards with known nucleic acid concentrations can be used to prepare a standard curve from which the amount of nucleic acid analyte in a test sample can be inferred.

Once the amount or absolute concentration of the nucleic acid analyte in the sample is known, it can be compared to the amount of analyte from the standard to determine the relative concentration of the nucleic acid analyte in the sample. Criteria for defining subjects with major adverse cardiovascular or cerebrovascular events can be determined empirically. For example, the amount can be determined by amplifying the nucleic acid analyte in a sample from a population of one or more individuals known to be normal and quantitatively analyzing the amount of nucleic acid analyte in the population.

Likewise, additional forms of sample chemical analysis may be performed. For example, a quantitative test may be performed that indicates the amount or absolute concentration of each analyte in the sample. Colorimetry is a quantitative chemical analysis that measures the intensity of color produced by the reaction of a sample with a reagent as an indication of the amount of the substance to be measured in the sample. A reagent may be provided that produces a color in the test sample upon reaction with the analyte. The color intensity may depend on the amount of analyte in the sample. By comparing the intensity to a calibrated color card and/or standard, the amount of analyte in the sample can be determined. This amount can then be compared to the amount of analyte from a standard (e.g., from a known normal human) to determine the relative concentration of the analyte in the sample.

In addition, urinalysis can be used to determine the amount or absolute concentration of an analyte in a urine sample. Urine samples are tested using a variety of different instruments and techniques. Some test strips (dipsticks) are thin plastic strips that change color in the presence of some particular substance. The test strip may be used to measure the amount of analyte.

Comparing the absolute level or concentration of each of at least two analytes with the level of each of the two analytes from a standard to determine the relative concentration of each of the analytes not only diagnoses as having or at risk of having a major adverse cardiovascular or cerebrovascular event, but the same comparison method is also applicable to other applications. For example, the analyte may be used to screen candidate drugs for the treatment of a severely adverse cardiovascular or cerebrovascular event. In such cases, the drug candidate treatment may be monitored by monitoring the analyte level. Effectiveness is determined when the absolute concentration of the analyte returns from the diseased level to the standard level, whereby the relative concentration reaches zero. Furthermore, for any drug that has been found to be effective in treating a severely adverse cardiovascular or cerebrovascular event, some subjects may be responders and some may not. Thus, the analyte may be monitored during the treatment to determine whether the drug is effective by determining whether the absolute level or concentration of the analyte returns to a standard level, whereby the relative concentration approaches zero. Of course, there may not be any currently known responder and non-responder populations, and thus the efficacy of drug treatment in subjects with any serious adverse cardiovascular or cerebrovascular events can be monitored over time. When it is ineffective, its use may be terminated and another drug provided instead.

Furthermore, the relative concentration may be determined by comparing the absolute level or concentration of each of the at least two analytes with the level of each of the two analytes from the standard, as a prophylactic screening measure, and not only when a detrimental cardiovascular or cerebrovascular event is observed (i.e., after the disease may have developed). For example, to obtain a diagnosis of or at risk of having a major adverse cardiovascular or cerebrovascular event at the earliest possible time, the subject may be monitored after a certain age and at predetermined intervals, assuming that the subject has no evidence of an adverse cardiovascular or cerebrovascular event. Upon positive screening, the medical professional may recommend further monitoring of disease progression, and/or the medical professional may begin treatment with a drug or other therapy.

The results of the analysis, including, for example, the amount or absolute concentration of each analyte, the relative concentration of each analyte to a standard, and/or the likelihood of having or being at risk of having a major adverse cardiovascular or cerebrovascular event, may be displayed or output to a user's interface device, computer readable storage medium, or local or remote computer system. Displaying or outputting the results or diagnosis, meaning that the results of any of the foregoing analyses are communicated to the user using any medium (e.g., oral, written, visual display, etc.), computer readable medium, or computer system. Those skilled in the art will appreciate that the output results are not limited solely to output to a user or to a connected external component (e.g., a computer system or computer memory), but may alternatively or additionally be output to an internal component (e.g., any computer readable medium). The computer readable medium may include, but is not limited to, a hard disk, a floppy disk, a CD-ROM, a DVD, and a DAT. The computer-readable medium does not include a carrier wave or other form of wave for data transmission. It will be apparent to those skilled in the art that the various sample assessment and diagnostic methods disclosed and claimed herein can be, but need not be, computer-implemented and can, for example, implement display or output steps, such as by oral or written (e.g., written in text) communication with the individual.

In addition, the analytes can be used to validate animal models of major adverse cardiovascular or cerebrovascular events. For example, in any particular model, a sample can be analyzed to determine whether the level of an analyte in an animal is the same as the level of an analyte in a subject with a known major adverse cardiovascular or cerebrovascular event. This will validate the model, for example, to test drug candidates in the manner described above.

Example 1

This example illustrates a method for identifying analytes of a major adverse cardiovascular or cerebrovascular event. Briefly, the study included one hundred thirty-six (136) subjects. The definition of a subject as a disease or control is achieved by identifying a subject who has experienced a severe adverse cardiovascular or cerebrovascular event within 2 years of index cardiac catheterization (index cardiac catheterisation), defined as: myocardial infarction, percutaneous coronary intervention, or death.

Frozen plasma samples were from collected samples obtained in long-term studies of individuals undergoing cardiac catheterization at the university of duke medical center (CATHGEN study). Plasma samples were from 136 study subjects, with 68 cases and 68 matched controls (defined below). Cases and controls were matched according to their Coronary Artery Disease (CAD) index, age, gender and race. Some study subjects in both the case and control groups showed no signs of coronary artery disease during both cardiac catheterization and angiography. Other subjects had different severity of coronary stenosis as indicated by a higher CAD index score.

The clinical parameters recorded for all samples are those shown in table 1 below. These parameters (other than those at the time of case and control matching) are included as additional clinical factors (described below) in some statistical analyses.

Table 1: study of clinical parameters of subjects

Characteristics of	Class/unit	Description of the invention
			Age (age)	Year of year	Age at intubation time (integer)
Race of a ethnic group	Caucasian, African-American and others
			CAD index		Coronary artery disease severity index
BMI	·＜25·25-29.9·＞＝30	Body mass index
			Dyslipidemia	Yes/no	History of dyslipidemia
Family history	Yes/no	Family history of coronary heart disease
			Hypertension (hypertension)	Yes/no	Is a history of hypertension or systolic BP > 140 or diastolic BP > 90
Diabetes mellitus	Yes/no	Is type I or II diabetes
			Smoking status	Yes/no	History of smoking
Creatinine clearance rate	Continuous measurement	Cockcroft-Gault approximation based on age, mass, plasma creatinine, and gender
			Daily administration of aspirin	Yes/no	Obtained from professional therapeutic drug tables

Use of statins

Yes/no

Obtained from professional therapeutic drug tables

Cases were identified as having the following characteristics: prior to index cannulation (index catheterization), subjects had no history of Coronary Artery Bypass Graft (CABG), Percutaneous Coronary Intervention (PCI) due to cardiac events, or Myocardial Infarction (MI); the subject subsequently did not have CABG; the ejection fraction of the subject is more than or equal to 40% (when the index is inserted) and is not reduced; MI, PCI or death occurred within 2 years after index cannulation and blood sampling. Those subjects who apparently were non-coronary artery disease-related deaths (i.e., pulmonary hypertension, cancer, etc.) were excluded. Those subjects with PCI-related events within 7 days of index intubation were excluded.

The control example was identified as having the following characteristics: prior to index cannulation, the subject had no history of Coronary Artery Bypass Graft (CABG), Percutaneous Coronary Intervention (PCI) due to cardiac events, or Myocardial Infarction (MI); the subject subsequently did not have CABG; the ejection fraction of the subject is more than or equal to 40% (when the index is inserted) and is not reduced; no MI, PCI or death occurred for at least 2 years after index intubation and blood sampling.

The samples were subjected to comprehensive bioanalysis using the platform provided in table 2 below and described in detail below.

Table 2: biological analysis program

Proteomics	Analyte vinegar for use in statistical analysis
		LC-MALDI-MS/MS discovery proteomics	217
Directed proteomics (multiplex immunoassay)	66
		Metabolomics
Lipid LC/MS metabolomics	125
		Polar LC/MS metabolomics	164
GC/MS metabolomics	130
		Free fatty acids	21

A total of 723 analytes were successfully measured in the bioanalytical portion of the study. The contribution of each bioanalytical platform to this total number is shown in table 2 above. The bioanalytical platform results for each platform were analyzed and also combined into a single integrated data set for subsequent statistical analysis (see discussion below).

Lipid liquid chromatography ("LC") tandem mass spectrometry ("LC/MS") metabolomics

The liquid chromatography and mass spectrometry conditions used in the present method are conditions optimized for the resolution and detection of lipid molecules.

The following materials were used according to this method. Solvents and reagents including water, methanol and isopropanol and dichloromethane (HPLC grade) were purchased from vwr (usa). Formic acid (99%), ammonium acetate, dichloromethane and reserpine were purchased from Sigma-Aldrich (Milwaukee, Wis.). HPLC protective column (Symmetry 300C 4)3.5 μ 2.1 × 10mm) and analytical column (C4)3.5 μ 2.1 × 150mm) from Waters (Milford, Ma). Autosampler vials, 300 μ L total recovery vials and screw caps were purchased from Waters (Milford, Ma). 500 μ L Eppendorf tubes for storage were purchased from VWR. Lipid internal standards 17:0LPC, 24:0PC and 40:0PC were obtained from Avanti Polarlipids (Alabaster, AL), while 30:0PC and 51:0TG were obtained from Sigma-Aldrich (Milwaukee, Wis.).

Samples were extracted according to the following protocol. To normalize the data 5 internal standards were added to the extraction solution. Internal standard stock solutions were prepared by dissolving appropriate weighed amounts of the various internal standard compounds in 1:9DCM/IPA solution to obtain final concentrations ranging from 1-3mg/ml depending on the specific lipid. A300. mu.L aliquot of each of the 5 stock solutions was added to 300mL of a 1:9DCM/IPA solution to prepare an extract solution (reference SOP: BG-QC27R 01). This solution was used as a lipid extraction solution containing 5 internal standards, 17:0LPC, 24:0PC, 30:0PC, 40:0PC, 51:0TG added at concentration ratios of 2: 1: 2: 3: 2(μ g/mL), respectively.

Samples for LC/MS analysis were prepared according to the following protocol. Each batch of samples was removed from the freezer along with the quality control samples and thawed on ice. An appropriate volume of extraction solution was added to each sample vial containing 10 μ L of plasma. The vial was gently vortexed and centrifuged at 14,000RPM for 10 minutes at 24 ℃. The supernatant was transferred to an intermediate vial, mixed by vortexer for several seconds, and 2 aliquots of 60 μ Ι _ each were pipetted as replicate samples into two barcode autosampler vials for LC/MS analysis. The remaining supernatant was stored at-40 ℃ for LC/MS/MS analysis. The remaining extract containing the precipitate was discarded.

High Performance Liquid Chromatography (HPLC) analysis was performed according to the following protocol. The chromatographic instrument consists of the following parts: a Waters 2795Alliance HPLC system with quaternary gradient system, autosampler, external column heater set at 45 ℃ and HPLC analytical column with built-in pre-column. The analyte was separated and delivered to the mass spectrometer at a flow rate of 350 μ L/min. From 95% H₂Solvent A consisting of O and 5% MeOH and solvent A consisting of 99% MeOH and 1% H₂Solvent B consisting of O was subjected to analyte separation using a binary gradient (0 min-20% B, 2 min-20% B, 4 min-80% B, 20 min-100% B, 25 min-100% B, 25.1 min-20% B, 35 min-20% B). Both solvent streams contained 10mM ammonium acetate and 0.1% formic acid. 5 μ L of sample was injected into a 20 μ L sample loop and loaded onto a pre-column prior to separation by the analytical column. Between each sample, the injection needle, sample ring and syringe were washed with isopropanol/MeOH solution.

Waters/Micromass quadrupole time-of-flight instrument (Waters, Q-ToF) equipped with LockSpray ESI source^TM) LC/MS analysis was performed. The ESI source block temperature was maintained at120 ℃ and the desolventizing gas temperature was set at 320 ℃. MS signal sensitivity was optimized with a reserpine peak at 609.2814. Mass calibration was performed in MS mode using protonated polyalanine peaks. An average instrument resolution of about 8500 was oriented to 5 IS peaks with a LockMass calibration accuracy of 3-5 ppm. MS data were obtained at 7-23 minutes during LC elution on the basis of the previous study. All LC/MS data were obtained in centroid mode (centroid mode) by scanning m/z 300-1000 in 0.3 seconds.

MS/MS analysis was then performed on the same LC/MS system with the same LC conditions to identify the analyte. MS/MS calibration was performed at m/z 1084 using the peak fragment of the parent polyalanine (precorsor). All MS/MS data were obtained in centroid mode by scanning m/z 100-1000 within 1 or 2 seconds. LC retention time correction is performed on the target list (target list) as necessary. MS/MS data were collected by directional analysis.

Peak detection was performed according to the following protocol. The accepted running raw data file is subjected to peak picking and integration. Only peaks that reach a defined threshold enter the alignment process. Each peak is characterized by an m/z value and retention time, e.g., the peak detected at m/z 510 and eluting at 7.88 minutes is labeled 510-0788. Each sample was performed independently.

After the peaks are selected, the MLL documents generated by the peak selection algorithm are "aligned" to each other. The program sets a retention time window using internal standard peaks from all injections, and then logically "aligns" each analyte within the batch based on the average retention time of the reference peak. At the end of the process, the physical retention time may drift from 510_0785 to 510_0788, and an additional threshold may be applied for peak area.

Quality Control (QC) of each lot was evaluated by a variety of measurements including calculating the percent relative standard deviation (% RSD) of internal standard to all samples, plotting the extracted area of the internal standard peak (intensity) as a function of time to identify trends and check the change in retention time between before and after alignment. A visual QC check was also performed for each injection during the analysis. These checks focused on how well the coverage of repeated injections, the behavior of QC samples over time, and overall instrument trends. Small changes in retention time and peak area that may occur during the study may be corrected in the data normalization process (e.g., a small temporal trend), but more significant changes (e.g., local injection) may require re-analysis.

During LC/MS measurements, each molecular species of interest is typically present as multiple chemical entities. This may be due to several factors: first, there are a number of stable isotopes of the constituent elements; it is composed ofNext, the analyte was analyzed by reaction with sodium (Na), potassium (K), ammonium (NH)₄) Adduct formation to form pseudo molecular ions; and other analytes, produce multiple types of ions. In addition, mass spectrometry ionization processes can often be accompanied by the breakdown of the original molecule. Thus, the data set obtained is an order of magnitude larger than the number of peaks representing the analyte. These additional peaks are informative, promote noise ingress, reduce statistical efficacy, mask the effects of bioinformatics analysis, and may make it more difficult to reach a clear scientific conclusion because of the large number of peaks. After peak selection and comparison of spectra, and prior to or concurrent with normalization and statistical analysis of the data, an automated computer platform was used that brings the measured entities together according to their putative starting compounds. The core algorithm scans the data by a Retention Time (RT) with a narrow moving time window. Candidate peaks with similar retention times are grouped together according to the specific chemical characteristics of the platform according to a known mass difference pattern.

After peak detection and alignment, several additional data processing steps are required before the data can be used for univariate and multivariate statistical analysis. These steps include removal of peaks designated as sample/study blanks and other background peaks, batch calibration and normalization. The end of these steps represents an important phase called "Locked Data Set".

Mass spectrometry profiling (profiling) and subsequent data processing yields a list of analytes for subsequent statistical analysis. The same or similar LC/MS settings as the initial profiling experiments were used to target MS/MS for the major analytes (usually the most abundant peaks of each analyte family). The directional LC/MS/MS spectrum is obtained in centroid mode. The MS/MS spectra obtained by the directed MS/MS analysis of the analyte were interpreted by combining retention time, mass, nitrogen regularity, Fatty Acid (FA) side chain specific fragment ions and comparing with reference spectra from a lipid MS/MS library.

Gas chromatography tandem mass spectrometry (GC/MS) metabolomics

GC has the advantages of high separation efficiency and constant retention time, and is sensitive in combination and selective (electron impact) in mass detection. However, many compounds contain polar functional groups that are either thermally unstable at the temperatures required for their separation or are not volatile at all. In addition, the peak shape of compounds with polar functional groups may be unsatisfactory due to unwanted column interactions (e.g., irreversible adsorption). To address these problems, the compounds must be derivatized prior to GC analysis.

The following materials were used according to this method. Solvents and reagents include methanol, pyridine, ethoxyamine hydrochloride, and N-methyl-N-trimethylsilyltrifluoroacetamide (MSTFA). The internal standard comprises: leucine-D3; phenylalanine-D5; glutamic acid-D3; glucose-D7; cholic acid-D4; alanine-D4; trifluoroacetylanthracene (TFAA); difluorobiphenyl (DFBP); and dicyclohexyl phthalate (DCHP).

Samples were thawed at room temperature and 10. mu.L of 250 ng/. mu.L standard aqueous solutions of leucine-D3, phenylalanine-D5, glutamic acid-D3, and glucose-D7 were added, followed by 400. mu.L of methanol (for protein precipitation). The sample was vortexed, centrifuged at 10000rpm for 10 minutes and the supernatant was transferred to an autosampler vial. After the nitrogen drying step, 10 μ L of 250 ng/. mu.L cholic acid D4 and alanine-D4 standard pyridine solution was added. Before capping the autosampler vial cap, 30 μ L of the ethoxyamine hydrochloride pyridine solution was added and the sample was incubated at 40 ℃ for 90 minutes.

The remaining internal standards were added after oximation: mu.L of 250 ng/. mu.L of difluorobiphenyl, dicyclohexylphthalate and trifluoroacetylanthracene standard pyridine solution. The sample was silylated by the addition of 100 μ L of MSTFA and heating at 40 ℃ for 50 minutes. The samples were centrifuged at 3500rpm for 10 minutes before being injected into the GC.

GC/MS analysis was performed using an Agilent 6890N gas chromatograph equipped with a PTV (temperature programmed vaporizer) syringe and a CTCAnalytics Combi-Pal autosampler. Detection was performed using an Agilent 5973Mass Selective detector. The system is controlled by Enhanced chemstation g1701CA Version d.01.02 software.

All compounds eluted from the GC/MS column were detected in a full scan manner. Each peak is characterized by its retention time and various fragments (m/z values). The amount of data (i.e., the number of variables) is reduced by using a directional processing procedure.

The orientation method reduces the number of variables to at most 1/20. Most of the program products can be removed from the data by not including these peaks in the target table. A study specific target table was compiled for this study. Information from a previous human plasma GC/MS project was used in the process. Representative chromatograms were selected from study samples (including extreme samples) to include as many peaks as possible observed in this study in the target table. At the initial stage of the project, the identity of these peaks was not known, they were labeled 'unknown 1' to 'unknown x', each peak being characterized by its retention time and a quantitative ion. Peaks found in the program blank were removed from the target table. In cases where the source of the peak is uncertain, the peak is left in the target list to avoid possible loss of the analyte of interest.

In the study samples, individual analytes/compounds found in representative chromatograms that differ from the internal standard were matched to a database of reference compounds (retention time and mass spectra). After this matching step, the tentative IDs of the plurality of compounds are specified based on the reference database.

When using the directional processing method, no alignment peak retention time is required. In the case where retention time drift is observed (internal standard is used as 'flag'), the target table can be adjusted. All compounds present in the target table were first integrated using a standard integration setting. These settings are not applicable to all compounds and samples and may lead to erroneous results for integration of certain compounds. Procedures have been developed to find integration errors. For all targets, the peak area can be calculated as the deviation from the mean (of all study samples and QC samples). The results were plotted for the detection of problematic peaks (false integration of those peaks across multiple samples) and problematic samples (integration of multiple peaks within a sample). For the problem samples, manual integration was performed. For the problematic peak, automatic integration is performed after the integration setting is adjusted.

In each step of sample preparation and analysis, one or more internal standards are added in order to monitor the quality of the data after analysis. After each batch, the internal standard response after correction for DCHP was evaluated in all samples. The batch can be approved if the corrected peak area of each internal standard deviates from the average value of the batch by less than 20%. If one or more samples deviate by more than 20%, these samples are re-analyzed in the next batch. Finally, a locked data set for GC/MS analysis can be generated.

The first identification step may be to match the list of target compounds to a database of reference standards. A variety of compounds from the target list can be identified and validated by comparison to a reference database and analysis of reference standards. After univariate and multivariate analysis, the priority unknowns were first evaluated by checking the raw data and QC results. They are divided into two categories: i) compounds with very low intensity and/or high RSD in QC standards; and ii) compounds with good signal in one or more samples and low RSD in QC standards. Authentication is initiated with the second class. For both classes, additional checks of the program product are carried out.

Following this analysis, additional identification methods are selected depending on the individual (compound to be identified). For some compounds, it was found that it could be searched in commercial spectral libraries without additional ID experiments. For other compounds, chemical ionization, accurate mass or other derivatization experiments were performed.

Polar LC/MS metabolomics

The following materials were used according to the method: methanol, Biosolve, LC-MS grade, cat # 13687801 or equivalent; HCl, Merck, 37%, cat # 1.00317 or equivalent; n-butanol, Baker, cat # 8017 or equivalent; Milli-Q water, water purified using ELGA system or equivalent; dithiothreitol DTT, Sigma 99%, cat # D9779 or equivalent; formic acid, Merck, cat # 1.00264 or equivalent; acetonitrile, Biosolve HPLC-S gradient grade, cat # 12000701 or equivalent; phenylalanine-D5, CDN isotope 99 at% D, cat No. D-1597 or equivalent; glutamine-D3, CDN isotope 99 at% D, cat No. D-1196 or equivalent; leucine-D3, CDN isotope 99 atomic% D, cat # D-1973 or equivalent; alanine-D3, CDN isotope 99 at% D, cat # D-1462 or equivalent; sarcosine-D3, CDN isotope 99 at% D, cat No. D-1462 or equivalent; methionine-D4, CDN isotope 99 at% D, cat No. D-1462 or equivalent; methyl- (D3) -histidine, CDN isotope 99 at% D, cat # D-1462 or equivalent; tyrosine-D7, CDN isotope 99 at% D, cat # D-1462 or equivalent.

The following instruments were used according to the method: a vortex oscillator; a centrifuge; an oven; an evaporator; ThermoFinnigan Surveyor or Surveyor Plus HPLC; ThermoFinnigan LTQ ion trap mass spectrometer equipped with an E SI source.

To a 10. mu.l plasma sample from a small Eppendorf vial was added 10. mu. lIS-working solution and the sample briefly vortexed. After addition of the DTT solution and removal of the proteins from the sample, the supernatant was lyophilized. The sample was then derivatized with HCl-butanol at 65 ℃. Excess reagent was removed by lyophilization. The samples were reconstituted with an aqueous solution of DTT containing underivatized tyrosine D7 as an internal standard.

High performance liquid chromatography was performed as follows:

column: Varian/Chrompack Inertsil 5. mu. mODS-3100 x 3mm

Protection of the column: Varian/Chrompack R210X 2mm internal diameter

Mobile phase A: 0.1% formic acid

1ml of formic acid was added to 1000ml of water, mixed and degassed by sonication for 5 minutes.

Mobile phase B: 80% acetonitrile/0.1% formic acid

800ml of acetonitrile are taken up in 1000ml of water, 1ml of formic acid are added, mixed and degassed by ultrasound for 5 minutes.

Column temperature: 25-30 deg.C

Autosampler temperature: 10 deg.C

Sample introduction volume: 10 μ l

Mass spectrometry was performed with a separation ratio of about 4:1, i.e., about 75. mu.l was flowed to the ESI source. ESI and MS were set as:

ESI spray voltage: 4kV

Heating the capillary: 250 ℃ to 300 DEG C

Gas covering: 50-60

Auxiliary gas: 5

Polarity: is just

Scanning range: 125-1250

Number of micro-scans: 5-6

Maximum injection time: 200ms

Source CID: 0-5V

Source distance: position C

The exact mass of the selected sample is analyzed. For this purpose, two QC samples and various "extreme" samples from each derivative lot were selected. Accurate mass experiments were performed on a thermo finnigan LTQ-FTMS instrument with the same setup as described above. Ion detection was performed in FTMS. The resolution at m/z400 is about 100,000.

All compounds eluted from the LC/MS column were detected in full scan mode. Each peak is characterized by its retention time and various ions (m/z values). The amount of data (i.e. the number of variables) is reduced by applying a targeted treatment program where each compound in the chromatogram is represented in most cases by a unique entry in the target table.

The orientation method reduces the number of variables to at most 1/20. Most of the program products can be removed from the data by not including these peaks in the target table.

A study-specific target table for this study was compiled. This was done by picking all QCY peaks and from the first few batches of study samples. The resulting peak table was classified based on intensity, and 1000 highest peaks were selected. From this partial table, the isotopic peaks and the known adducts were removed. The identity of the peaks is not known at this stage and they are labelled as m/z with retention time (minutes).

When using the directional processing method, no alignment peak retention time is required. In the case where retention time drift is observed (internal standard is used as 'flag'), the target table can be adjusted.

After all rounds of analysis were completed, all compounds present in the target table were first integrated using the standard integration setting. These settings are not applicable to all compounds and samples and may lead to erroneous results for integration of certain compounds. Procedures have been developed to find integration errors. For all targets, the peak area can be calculated as the deviation from the mean (of all study samples and QC samples).

Quality control of the entire data set was performed based on IS responses of all samples except blank and examination of the relative standard deviation of all target compounds in QCY and CTRL samples. These results were compared to those observed in previous studies. After these steps are completed, a "lock data set" for the polar LC/MS platform may be generated. These data sets form the basis for all statistical analyses that follow.

The first identification step is to match the list of target compounds to a reference standard database containing a variety of amino acids and related compounds. The retention time and the exact mass are compared and, if necessary, the MS/MS spectra are compared.

The remaining target compounds were only identified after being prioritized (univariate/multivariate). First, by checking the raw data and QC results, the priority unknowns are evaluated. They are divided into two categories: i) compounds with very low intensity and/or high RSD in QC standards; and ii) compounds with good signal in one or more samples and low RSD in QC standards. Authentication is initiated with the second class. For both types, additional checks of the program product are carried out; the blank was examined for the presence of these compounds.

For further characterization, retention time, accurate mass and MS/MS data were used. Based on the exact mass and knowledge about the derivatives used, the KEGG, Merck and Chemfinder databases were searched for possible elemental compositions. Possible hits were evaluated, for which purpose retention times and MS/MS spectra were used. In several cases, standards were purchased and analyzed in solution after adding the standards to the study samples.

LC/MS profiling of free fatty acids

The plasma was freed of proteins and extracted with IPA. Free fatty acids were separated on an analytical column, ionized by electrospray in negative ion mode, and detected in full scan mode. Various free fatty acids were analyzed in external standards and calculated (approximated) using calibration standards if it was desired to determine the true concentration in plasma.

Samples were prepared and analyzed with the following materials. The instrument included an eppendorf tube centrifuge, Thermo Electron LTQ, Thermo Electron surveyor HPLC with autosampler and column oven, and a vortex shaker. Chemicals include demineralized water (ELGA System 4 or any other system preparation, such as Millipore); isopropanol p.a. (Baker 6068 or equivalent); methanol (HPLC grade, Chromasolve 34966); dichloromethane p.a. (Baker 7053 or equivalent); ammonium acetate p.a., (Fluka 09690 or equivalent); and formic acid p.a. (merck1.00264.1000 or equivalent).

The material comprises Alltech Prosphere C4HPLC column (150 × 3.0mm, 5 μm inner diameter), part No. (part No.) 35548; symmAn etry 300C4 guard column (10x 2.1mm inner diameter 3.5 μm), part No. 186000278; autosampler vials (Alltech part number AV055201, 32 x 11.6 mm); an Eppendorf tube; and pipettes (e.g., Eppendorf).

The following criteria were used: heptadecanoic acid C17:0FA (Sigma-Aldrich, H3500); c14:0FA myristic acid (Sigma M3128); c16:0FA palmitic acid (Sigma P5585); c16:1FA palmitoleic acid (Sigma P9417); c18:0FA stearic acid (Sigma S4751); c18:1FA oleic acid (Sigma O1008); c18:2FA linoleic acid (Sigma L1376); and C20:4FA arachidonic acid (Sigma A9673).

The internal standard solution was prepared as follows. A heptadecanoic acid C17:0FA (1mg/mL) stock solution was prepared by weighing 5mg of heptadecanoic acid C17:0FA in a 5mL volumetric flask, adding 500. mu.L of DCM, dissolving with ultrasound, and filling to 5mL with IPA. IS working solution (for protein precipitation and lipid extraction) was prepared at a concentration of 1. mu.g/mL in IPA containing C17:0 FA.

Samples were prepared as follows. mu.L of IS working solution was added to 10. mu.L of thawed plasma sample in an eppendorf cup and vortexed. The solution was subjected to eppendorf centrifugation at10,000 rpm for 3-5 minutes, transferring 50. mu.L of clear supernatant in 2 portions each into two autosampler vials with internal cannula (insert) and storing the remaining extract at < -18 ℃.

With Alltech Prosphere C4HPLC columns (150X 3.2mm inner diameter 5 μm) and Phenomenex wire trap C4 guard column (4X 3mm inner diameter 5 μm) were used for liquid chromatography and mass spectrometry. Mobile phase a consisted of 5% methanol in buffer; mobile phase B consisting of 2mM NH₄Methanol composition of Ac; the needle wash consisted of 60% IPA/MeOH. The gradient is as follows:

time (minutes)	Flow rate (mL/min)	％A	％B
				0	0.4	70	30
2	0.4	70	30
				6	0.4	30	70
10	0.4	0	100
				15	0.4	0	100

15.1	0.4	70	30
				20	0.4	70	30

The column temperature was 40 ℃; the autosampler temperature was 20 ℃; the injection volume was 20. mu.L.

For the mass spectrometry procedure, the LC/MS instrument was adjusted with the following calibration standards:

instrument for measuring the position of a moving object	ThermoFinnigan LTQ
		Mode(s)	Negative pole
Heater capillary tube	250℃
		Spray voltage	3.5kV
Cover gas	35arb
		Auxiliary gas	15arb
Spray location	C
		Scanning range	180-400
Scanning speed	Automatic
		Multiplier	Automatic

Selected target analytes were integrated with Therno XCalibur LCQuan V2.0. Data were processed with a target table containing at least the following fatty acids (based on previous studies): c12:0FA, C12:1FA, C14:0FA, C14:1FA, C16:0FA, C16:1FA, C16:2FA, C18:0FA, C18:1FA, C18:2FA, C18:3FA, C20:0FA, C20:1FA, C20:2FA, C20:3FA, C20:4FA, C20: 5FA, C22:5FA, C22:6FA and C22:7 FA.

Integrated data for all compounds from the target table were obtained. The results are an Excel table containing the integrated values for each target compound.

LC-MALDI-MS/MS discovery proteomics

Proteomics workflow is based on multidimensional liquid chromatography-MS/MS analysis of peptides produced from plasma samples. The advantages of this method include insensitivity to multiple heterogeneity of plasma proteins that leads to difficulties in the two-way gel-based methods, the automation of liquid chromatography techniques for preparation and the possibility of using stable isotopically labeled peptides.

The recently proposed iTRAQTM method (iTRAQ is a trademark of Applied BioSystems) is particularly suitable for this study because the labeling reaction is easy to perform to completion, the reaction is universal, almost all peptides acquire iTRAQ labels, and each experiment allows multiplex quantitation of up to 4 samples. The latter is important in this study where consistent quantification (and identification) of the same protein among samples is a key requirement. Another important advantage of stable isotopic labeling is the robustness of the method in terms of liquid chromatographic behavior variations of peptides between samples (once the library of differentially labeled is pooled).

In the iTRAQ method, an enzymatic digest from a protein sample is treated with a reagent (N-hydroxysuccinimide (NHS) ester) that derivatizes the free primary amino groups (N-terminus (if unblocked) and lysine residues) of the peptide. The four different classes of reagents are characterized by the same mass but different stable isotope labeling positions. Thus, in MS mode, all four differently labeled peptides appear as a single component when the ionic signal reflects the molecular weight of the peptide. In the MS/MS model, subtle differences in the marker structure become visible, while the peptide backbone segments (so-called a, b and y) remain isomeric, yielding four different reporter segments corresponding to m/z 114, 115, 116 and 117, respectively. By determining the relative intensities of these reporter ions, a relative quantitative determination of peptides from four different samples can be accomplished.

In this study, a MALDI TOF/TOF mass spectrometer (ABI 4800) was used, in which the sample was placed in a MALDI dish by an automated partial collection procedure. A particular advantage of MALDI LC-MS/MS is the off-line feature of LC-MS coupling. It is possible to first analyze the entire 2D LC isolate in MS mode before generating peptide identification and quantitative measurement data in MS/MS mode. This feature may be very convenient when a constant series of peptides must be measured in a large number of samples.

One unique aspect of analyzing plasma samples is the huge background represented by the abundant plasma proteins (albumin, immunoglobulins, etc.). To address this challenge, a number of cancellation techniques (delete techniques) are used. In this study, chicken IgY antibody columns were used to eliminate twelve abundant proteins from the samples. Proteins not retained on the antibody column were recovered on a reverse phase column. The recovered protein was reduced, alkylated and digested with trypsin. The resulting peptide mixture was labeled with iTRAQ reagents and combined with three other samples designated for the same iTRAQ mixture. The combined quadruple (four-plex) mixture was first fractionated on strong cation exchange chromatography. 10 fractions were collected from the SCX elution. These fractions were further analyzed by HPLC MS/MS after combining some fractions.

iTRAQ analysis allows the relative quantification of four different samples to be measured in a single 2D LC-MS/MS experiment (quaternary). However, assigning four initial samples into an iTRAQ quaternary mixture does not allow for a comparison of two initial samples assigned to different iTRAQ mixtures. This is solved as follows: making a reference pool (reference pool) of aliquots of the quantity of iTRAQ mixtures analyzed according to the project; constant composition of each iTRAQ mixture; sample preparation of this reference aliquot was performed along with the other 3 members of the same iTRAQ mixture; and labeled with 117 reagent.

A reference library was prepared by combining approximately 25% of each initial study sample analyzed from the study. Aliquots from the reference library were referred to as QCR samples.

MS/MS data acquisition may be guided by a list of MS parents for both inclusion and exclusion. The initial inclusion list was derived from analysis performed during study enrollment. At this stage, several samples from the sample QC library were processed via a complete proteomics workflow. After identifying the peptides and matching the proteins from these mixtures, an inclusion list and an exclusion list are compiled.

The inclusion list content includes: fully alkylated and iTRAQ-labeled trypsin treatment of the peptides; fully alkylated and iTRAQ-labeled hemitrypsin treated peptides and excluded those peptides produced due to chymotrypsin activity (cleavage at F, Y and W); and peptides manually identified as having post-translational modifications but not routinely sought by Mascot.

Exclusion list content includes: 2,000 of the most abundant unidentified components, including peptides with incomplete iTRAQ labeling or alkylation; a peptide of molecular weight of more than 3,000Da containing multiple missed trypsin cleavage sites; peptides produced by the chymotrypsin side activity of trypsin; in the Inclusion List as M + H⁺Probable M + K of peptide abundant in form⁺Forms thereof; and autolytic peptides of trypsin.

A custom software application was used to take into account inclusion/exclusion criteria and to generate MS/MS data collection on AB4800 based on a separate text file listing the selected parent list for inclusion or exclusion. The software also includes automatic adjustment of possible HPLC retention time drifts. Additional MS/MS spectra were obtained in an opportunistic manner as allowed by the post-maternal acquisition time involved in the first analysis.

Samples were thawed on ice from 12 batches per day, 12 batches consisting of 9 initial samples and 3 reference library samples, and samples were processed consecutively with one reference library sample followed by 3 initial samples (i.e., samples 1,2, 3, then R, 4, 5, 6, R, etc.). Once the sample was completely melted, it was degreased. Samples were diluted with PBS and extracted with trichlorotrifluoroethane (Freon 13) followed by a centrifugation step. The top aqueous phase was transferred to a new tube and stored briefly on ice, then placed in an LC system autosampler that had been frozen for rich protein clearance.

The 12 most abundant plasma protein proteins in the defatted sample were removed using an abundant protein clearance (APR). The program uses a two-column approach (2DLC) in which 12 proteins including albumin, IgG, IgA, transferrin, alpha-1 antitrypsin, haptoglobin family, fibrinogen, alpha-1 acid glycoprotein, IgM, alpha-2 macroglobulin, and HDL (apolipoprotein a-I and AII) are targeted by affinity IgY columns, while the cleared material flows through and is captured on a Reverse Phase (RP) column. The RP column was washed separately and eluted stepwise. The elution peak consisting of the washed out cleared material was collected in a fraction collector and subsequently dried by vacuum centrifugation concentration (speed vacuum) for further processing in the workflow. The RP column was then washed with an anti-extraction solution that eliminated the residue remaining to the next round. The IgY columns were then eluted separately and neutralized before the next round. 12 defatted samples (single batch) were processed within 24 hours, followed by overnight vacuum centrifugation evaporative concentration of the collected fractions.

Cleared plasma samples were reduced, alkylated, digested and labeled with iTRAQ. The dried cleared sample was resuspended in 1M TEAB buffer pH 8.5. TCEP and SDS were added to denature and reduce the samples. After incubation for 1 hour at 71 ℃, the samples were alkylated with iodoacetamide for 30 minutes at room temperature in the dark. Trypsin suspended in N-acetyl-L-cysteine was added and the sample was allowed to digest overnight at 48 ℃. The following day, iTRAQ reagent was thawed and resuspended in 100% ethanol. The sample is labeled with a specific iTRAQ reagent. After incubation of the tags at room temperature for 1 hour, the acetylation reaction was stopped by adding 4. mu.l of 1M ammonium bicarbonate. The iTRAQ-labeled samples in the same mixture were combined and dried in Speedvac before being resuspended for cation exchange chromatography.

The pooled iTRAQ mixtures were subjected to strong cation exchange fractionation on a Vision Chromatography Station (Applied Biosystems) using a 4.6x50mm polysulfonoethyl strong cation exchange column. The separation uses a two-column process in which a peptide library is first loaded onto a PorosR2 reverse phase column in order to remove a significant amount of iTRAQ hydrolysate.

Peptides were eluted to SCX columns by loading buffer SCX with 35% ACN content. The more hydrophobic material (i.e. residual trypsin) stuck to the RP column was removed by high organic washing. Peptides bound to the SCX column were eluted with a gradient. Before injecting the sample, the sample pH was adjusted to 3.5 with 1N HCl. 0.8mL of 10 fractions (A1 to A10) were collected from SCX separation. These fractions were dried in Speedvac and resuspended in HPLC loading buffer.

LC-MALDI-MS/MS discovery proteomics

HPLC and MALDI coupling was performed off-line by partial collection onto MALDI targets. The apparatus is an UltMate chromatography system from Dionex-LC Packings (Hercules, Calif.) equipped with a ProbotMALDI spotting device (spotting device). An integrated syringe pump on the fraction collector allows co-mixing of the HPLC eluate and MALDI matrix before placement on a MALDI dish (plate). To ensure good mass calibration of the MALDI MS spectra, the internal standard-ACTH (18-39) peptide; MH⁺2465.199-was added to the matrix solution.

Spotting on MALDI discs was configured in such a way that three rounds of HPLC were placed on a single 4800MALDI disc. Each round of HPLC contained 6x35 array fractions (210 fractions/HPLC). Six rounds of HPLC, representing iTRAQ mixtures, were spotted into two MALDI discs.

The HPLC-MALDI operating parameters are listed below. HPLC grade water and analytical grade acetonitrile were from Burdick & Jackson, ammonium acetate, ammonium citrate, TFA and α -cyano-4-hydroxycinnamic acid were from Sigma.

HPLC MALDI parameter

HPLC buffer a: 95% water, 5% acetonitrile, 0.1% TFA

HPLC buffer B: 10% Water, 90% acetonitrile, 0.1% TFA

HPLC loading buffer: 95% Water, 5% acetonitrile, 0.1% TFA, 2mM ammonium acetate

Matrix solution: 10g/L alpha-cyano-4-hydroxycinnamic acid in 15% water, 85% acetonitrile

0.1g/L diammonium citrate

Internal mass calibration standard: 200 FIELD/L ACTH (18-39) peptide (in matrix solution)

HPLC analysis column: 0.180x150mm C18(Dionex-LC Packings)

HPLC trap column: 0.3X5mm C18(Dionex-LC Packings)

HPLC flow rate: 2 microliter/minute

Flow rate of the substrate: 2 microliter/minute

Fraction collection interval: 9 seconds

HPLC gradient program as follows:

0-5% B within 8 min

5-10% B in 2 min

10-37.5% B in 30 min

Within 37.5-100% B2 min

UV recordings were recorded by monitoring the 280nm wavelength with 45-nL flow cytometry.

The injected amount of each SCX fraction was determined to maximize sample loading but avoid overloading the HPLC column. These amounts were determined in preliminary registration experiments.

Peptide standard mixtures (from Michrom) were injected to monitor the consistency of HPLC retention time prior to each set of 6 injections representing MALDI samples from iTRAQ mixtures. Note that HPLC elution times were treated without interruption for moderation by the parent selection program, which was performed by LC-MS runs automatically against adjacent SCX fractions. After injection of the standard QC mix, 6 SCX fractions were injected, yielding 2 MALDI discs.

The discs were set in sets of two MALDI discs and placed in the autosampler of an AB4800 instrument. Each disc runs in MS mode through the following sequence:

disc models were generated with 4800 standard mixes (Applied Biosystems) spotted on the disc at 8 "calibration" positions and the default calibration was updated;

run LC fractions with internal mass calibration in MS mode using the m/z 2465.199 (peptide standard) peak and m/z 568.136 (matrix trimer) peak.

The main objective of the precursor selection is to minimize acquisition time (exclude excess precursors), eliminate interference from neighboring peaks in the MS spectrum, and consistently select the same precursor for all iTRAQ mixtures. MS/MS access and processing parameters were optimized to yield the best possible measurement of iTRAQ report peaks at m/z 114, 115, 116, 117. Unless the iTRAQ signal exceeds the signal-to-noise criteria after only 1,050 transmissions, care is taken to set the MS/MS data acquisition to accumulate 2,100 laser transmissions. In this way the acquisition time can be reduced without risking quality for iTRAQ peak area measurements.

The AB4800MS/MS spectra were obtained from an Oracle database located on the data acquisition control computer with a custom Perl script driven by pipeline control software. MS/MS peaks were processed into two different documents to separately record quantitative iTRAQ peak data and peptide cleavage peaks. Each individual MS/MS spectrum is identified by a unique database peak table ID and a unique MALDI disc database ID from which the data was collected. These two values allow further integration with other information recorded in both the collected data and the internal Laboratory Information Management System (LIMS). The iTRAQ quantitative information is transferred to a database from the first document called PeptidedB. The second output is an mgf (Mascot Generic format) document, which contains peptide cleavage peaks and additional relevant annotation lines that allow tracking between Mascot results and input MS/MS spectra.

The peaks included in the MGF profile were filtered so as to: removing high quality peaks (within 64Da or parent MH)⁺) (ii) a Low quality peaks (< 146Da) are removed; and limiting peak "density" to 15 fragment masses per 200Da window.

A separate MGF archive was made for each MS/MS job run operating on AB4800 (an MS/MS job could consist of 1 to thousands of separate MS/MS data acquisitions acquired from the same MALDI disk). Separate MGF files corresponding to the same iTRAQ mixture were juxtaposed before beginning Mascot search.

Each MS/MS spectrum is correlated in PeptideDB with the ion intensities of the m/z 114, 115, 116 and 117 ion signals. As described above, the iTRAQ ratio was determined relative to the m/z 117 intensity (since the label was used for the reference library). The ratio is calculated as: peak area ratios for 4800 data (cluster areas not used by Applied BioSystems software) and peak count ratios for QToF data.

After database search, peptide sequences that match the MS/MS spectrum inherit the identification number (identification number) assigned to the MS/MS spectrum. Protein ratios are derived from the ratios of peptides assigned to the protein as average ratios (or average ratios in Applied BioSystems software). However, the final peptide-protein assignment in the BGM workflow will not be performed until the peptide identification for each iTRAQ mixture is completed. At this point, not only is the protein assignment organized, but individual proteins can be separated into nodes based on quantitative information throughout the study (i.e., resolving protein isoforms, peptide sequences that provide polymorphic sites, allelic variants, etc.). Protein nodes-which are ultimately used for statistical and associative network analysis-are then represented as the average peptide ratio matched to them.

When propagating peptide ratios as protein ratios or protein node ratios, only the following are considered: (a) those peptides that are fully modified to the desired extent; i.e. completely labelled with iTRAQ and fully alkylated if they contain cysteine; or (b) peptides that do not match the proteins targeted by the APR, or do not match trypsin, hemoglobin (these proteins are considered contaminants).

Peptide/protein identification was performed using a commercially available Mascot database search program (matrix science ltd., UK), expert confirmation by scientists, and MS/MS spectrum matching program, three-fold.

The spectral matching procedure was performed immediately after data acquisition of the entire proteomic component of the study was completed and Mascot search results were uploaded into PeptideDB. The aim was to find MS/MS spectra that matched the peptide sequence according to the automated validation criteria in any iTRAQ mixture analyzed in the proteomics project. Based on a close match: parent mass; SCX fraction number and HPLC retention time; and the mass of the most abundant fragment ion to determine the identity of the peptide. This makes it possible to rescue a large number of MS/MS spectra that do not meet the automatic validation criteria of Mascot, but have automatic validation or expert validation in other iTRAQ mixtures. The importance of the procedure is to increase the performance of peptides and proteins measured in as many samples as possible in proteomic studies.

After peptide identification was completed throughout the study (with the most complete set of peptides possible), the protein confirmation tool was run to identify the smallest set of proteins that accounted for all confirmed peptide sequences. In protein validation tool terminology, peptide sequences are assigned to: (a) a class of proteins representing unique genes in the genome; or (b) a protein prototype (protein exemplar) that represents a preferred (usually the one with the most annotated information) sequence instance within a certain protein class (among splice variants, polymorphic forms, etc.).

A number of data processing steps were applied to the entire data set, which steps affected the number of proteins/peptides reported in this study and the identification metrics. These data processing steps include:

finishing peptide measurements: the iTRAQ signals from multiple MS/MS measurements of the same peptide (same sequence and same modification) in the same iTRAQ mixture were added and the results were measured with a single ratio from this point forward.

Eliminating outliers: standard statistical procedure to remove 5% of abnormal peptide measurements. A peptide measurement is considered an outlier if it falls completely outside the distribution (greater than two standard deviations) of all peptide measurements that match the same protein.

Final proteinPrime/node assignment: immediately after the protein validation tool analysis was completed, the protein lists for univariate and multivariate statistical analyses were finalized with the quantitative analysis results for the entire project. The final protein defines the node of the relevant network analysis. This process is necessary to correctly follow the post-processing forms of the protein (i.e., C3a and C3b from complement C3, fibrinopeptide a from the alpha chain of fibrinogen, alpha-microglobulin, bikunin from AMBP _ HUMAN, etc.) and polymorphic variants of the protein.

Quantitative performance was assessed by the consistency of the ratio of peptides matched to the same protein. The relative standard deviation of these peptide measurements indicates the degree of accuracy of the protein measurements (as the average of the matching peptide measurements). These procedures reduce the peptide relative standard deviation by a few percent.

Directed proteomics: multiplex immunoassay

The concentrations of the 89 target proteins shown in table 3 below were measured in a multiplex immunoassay system (Rules Based Medicine, Austin, TX, USA) [ McDade & Fulton, DeviceDiagn Ind 19: 75-82, 1997; fulton et al, Clin Chem 43: 1749-56, 1997]. The system performs a large number of simultaneous reactions by using a flow cytometer and a digital signal processor to perform real-time analysis of a variety of microsphere-based assays. The 3 main components of the system are a desktop flow cytometer, microspheres, and computer hardware and software. The plasma volume used in these measurements was 50 microliters. The fluorescence signal recorded by the flow cytometer for each target protein can be compared to a standard curve to calculate the concentration of each target protein. Quality control samples were included in this analysis.

Table 3: 89 proteins measured in multiplex immunoassay

1. Adiponectin	30.GST	60. Lymphotactin
			2. Alpha-1 antitrypsin	31.G-CSF	61.MDC
3. Alpha-fetoprotein	32.GM-CSF	62.MIP-1α
			4. Alpha-2 macroglobulin	33. Growth hormone	63.MIP-1β
5. Apolipoprotein A-1	34. Haptoglobin protein	64.MMP-2
			6. Apolipoprotein C-III	35. Immunoglobulin A	65.MMP-3
7. Apolipoprotein H	36. Immunoglobulin E	66.MMP-9
			8. Beta-2 microglobulin	37. Immunoglobulin M	67.MCP-1

9.BDNF	38. Insulin	68. Myeloperoxidase
			C-reactive protein	39.IGF-1	69. Myoglobin
11. Calcitonin	40.ICAM-1	70.PAI-1
			12. Cancer antigen 19-9	41. Interferon-gamma	71.PAPP-A
13. Cancer antigen 125	42. Interleukin-1 alpha	72. Free PSA
			14. Carcinoembryonic antigen	43. Interleukin-1 beta	73. Prostatic acid phosphatase
15.CD40	44. Interleukin-1 ra	74.RANTES
			CD40 ligand	45. Interleukin-2	75. Serum amyloid P
17. Complement 3	46. Interleukin-3	76.SGOT
			18.CK-MB	47. Interleukin-4	77. Sex hormone binding globulin
19. Endothelin-1	48. Interleukin-5	78. Stem cell factor
			20. Eotaxin	49. Interleukin-6	79. Thrombopoietin
21. Epidermal growth factor	50. Interleukin-7	80. Thyroid-binding globulin
			22.ENA-78	51. Interleukin-8	81. Thyroid stimulating hormone
23. ErythropoiesisVegetable extract	52. Interleukin-10	82. Tissue factor
			24.ENRAGE	53. Interleukin-12 p40	83.TIMP-1
Factor VII	54. Interleukin-12 p70	84. Tumor necrosis factor-alpha
			26. Fatty acid binding proteins	55. Interleukin-13	85. Tumor necrosis factor-beta
27. Ferritin	56. Interleukin-15	86. Tumor necrosis factor RII
			28. Fibrinogen	57. Interleukin-16	87.VCAM-1
29. Basic FGF	58. Leptin protein	88.VEGF
				59. Lipoprotein (a)	89. Von willebrand factor

The targeted protein analysis yields useful information about 66 protein concentrations in the sample sets from cases and controls. These 66 protein datasets were used for univariate and multivariate analysis of the individual bioanalytical platform and combined with other bioanalytical platform datasets to produce a comprehensive dataset containing 723 analytes.

Statistical method

Statistical analysis was performed to address the main goals of the study, namely the discovery of molecular analytes in blood or plasma (pools), and the associated algorithms to predict recent MACCE. Subjects who developed MACCE during the 2-year follow-up period are referred to as cases, while subjects who did not develop MACCE during the 2-year follow-up period are referred to as controls.

The statistical analysis included the following components:

1. study subject baseline characteristics;

2. principal Component Analysis (PCA) to visualize the separation, if any, between cases and controls from the data summarized by each platform;

3. a univariate model for detecting individual analytes statistically significantly correlated with recent MACCE probabilities after adjustment according to conventional risk factors;

4. multivariate model: obtaining a classifier of an optimal analyte subset, wherein the optimal analyte subset is statistically significantly correlated with recent MACCE probabilities with or without consideration of conventional risk factors. Receiver Operating Characteristic (ROC) curves based on the results produced by the classifier with or without consideration of conventional clinical factors are compared.

Base line specialAnd (3) carrying out mark: the baseline characteristic mean and ratio of cases and controls were calculated. Statistical significance of baseline characteristics between cases and controlsThe writability is based on Wilcoxon rank sum test (continuous variable) and chi-square test (for scale values and RxC table).

Principal component analysis: PCA scores were generated for the data obtained from each metabolomics and proteomics platform in this study. The functions (motivations) of these analytes include visualizing the separation between cases and controls, identifying outlier samples, and assessing the quality of the data set.

Single variable method: the basic univariate analysis is based on conditional logistic regression, taking into account 1-to-1 matching of cases to controls. Univariate analysis consisted of a regression model that evaluated the statistical significance (case versus control) of each plasma analyte alone in association with the results after adjustment to the relevant clinical covariates. Since cases matched controls in age, race, gender, and CAD index, these variables did not appear in the regression model. The conditional logistic regression model is adjusted according to confounding factors not included in the matching strategy. The selection of conventional risk factors for inclusion into the model is based on a review of recent papers in the medical literature that report on findings from investigations investigating the effect of analytes on cardiovascular outcomes. Each plasma analyte was assigned a quartile based on the distribution of the analyte in the control subjects. The statistical significance of each analyte in the predicted outcome was assessed using a likelihood ratio test. The Storey-based method (J.R.Statist.Soc.B 64: 479-. Special cases with more than 50% missing values in cases or controls were excluded from this univariate analysis.

A second set of univariate analyses was performed in which the analyte values were converted to binary variables, where 0 represents a missing value and 1 represents no missing measurement (otherwise). As with previous univariate analysis, the statistical significance of each analyte in the predicted results was assessed using the likelihood ratio test. The Storey-based method (J.R.Statist.Soc.B 64: 479-.

A graphical representation of a statistically significant box-and-wisker plot (box-and-wisker plot) for each analyte was also made, depicting the case and control, respectively.

A graph was prepared comparing FDR adjusted p-values to the original p-values. The p-value distribution observed in the plot was used as the basis for selecting the p-value threshold used to select putative analytes in each platform. When data from all platforms is available, FDR adjusted p values are generated by integrating data from all platforms. A final list of univariate analytes is generated from the integrated analysis. Selected results are shown in table 4 below.

Table 4: single variable analysis results (part 1)

Platform	ID	Unit of	P value	Odds Ratio (OR)4)	Ratio of mean concentration in case to mean concentration in control
						Polar LC/MS	Cysteine	Arbitrary Unit (AU)	＜0.01	15.70	1.17
Directed proteomics	Von willebrand factor	Microgram/ml	＜0.01	17.67	1.44
						Directed proteomics	IL-8	pg/ml	＜0.01	15.57	1.25
Lipid LC/MS	16:0/18:1PC	AU	＜0.01	17.67	1.12
						GCMS	N-carboxy-alanine	AU	＜0.01	14.31	1.16
Directed proteomics	Fibrinogen	mg/ml	＜0.01	7.55	1.28
						Directed proteomics	MMP-2	ng/ml	＜0.01	17.85	1.47
Lipid LC/MS	18:0/20:4PE	AU	＜0.01	27.53	1.15
						Directed proteomics	Apolipoprotein A1	mg/ml	＜0.01	0.09	0.86
Lipid LC/MS	16:0/22:6PE	AU	0.01	14.87	1.08
						Lipid LC/MS	18:1/18:0/18:0TG	AU	0.01	11.43	1.20
Directed proteomics	Alpha-1 antitrypsin	mg/ml	0.01	5.59	1.12

Lipid LC/MS	18:2/18:1/17:0TG	AU	0.01	69.36	1.11
						Lipid LC/MS	20:1/18:1/18:1TG	AU	0.02	12.64	1.27
Lipid LC/MS	16:0/16:0PC	AU	0.02	6.93	1.12
						Lipid LC/MS	20:4LPC	AU	0.02	0.22	0.87
Lipid LC/MS	16:0SM	AU	0.02	7.91	1.08
						Directed proteomics	SHBG	nmol/ml	0.03	6.81	1.25
Lipid LC/MS	18:1/17:1/16:0TG	AU	0.03	11.84	1.11
						GCMS	Arabinose	AU	0.03	0.14	0.87
Lipid LC/MS	18:1/18:1/17:0TG	AU	0.03	10.68	1.16

Table 4B: single variable analysis results (part 2)

Platform	ID	First quartile	Second quartile	Second quartile	Fourth quartile
						Polar LC/MS	Cysteine	＜2041.1	2041.2-2699.2	2699.3-3675.9	＞3675.9
Directed proteomics	Von willebrand factor	＜35.2	35.2-47.7	47.8-69.0	＞69.0
						Directed proteomics	IL-8	＜14.1	14.1-17.4	17.5-21.4	＞21.4
Lipid LC/MS	16:0/18:1PC	＜30075.8	30075.9-35320.5	35320.5-61392.4	＞61392.4
						GCMS	N-carboxy-alanine	＜3431.4	3431.4-3853.5	3853.6-4324.6	＞4324.6

Directed proteomics	Fibrinogen	＜3.5	3.5-4.3	4.3-5.2	＞5.2
						Directed proteomics	MMP-2	＜1037.5	1037.5-1520.0	1520.1-2157.5	＞2157.5
Lipid LC/MS	18:0/20:4PE	＜1861.8	1861.8-2138.4	2138.5-2429.0	＞2429.0
						Directed proteomics	Apolipoprotein A1	＜0.30	0.30-0.36	0.37-0.46	＞0.46
Lipid LC/MS	16:0/22:6PE	＜1811.8	1811.8-2066.1	2066.2-2476.7	＞2476.7
						Lipid LC/MS	18:1/18:0/18:0TG	＜15038.4	15038.4-17619.3	17619.4-22441.1	＞22441.1
Directed proteomics	Alpha-1 antitrypsin	＜1.50	1.50-1.69	1.70-1.96	＞1.96
						Lipid LC/MS	18:2/18:1/17:0TG	＜1963.7	1963.7-4369.5	4369.6-12667.6	＞12667.7
Lipid LC/MS	20:1/18:1/18:1TG	＜44210.6	44210.6-54354.2	54354.3-68063.3	＞68063.3
						Lipid LC/MS	16:0/16:0PC	＜4859.1	4859.1-6019.0	6019.1-8158.0	＞8158.0
Lipid LC/MS	20:4LPC	＜4203.5	4203.5-5082.6	5082.7-6311.4	＞6311.4
						Lipid LC/MS	16:0SM	＜4289.8	4298.8-5239.6	5239.7-6554.6	＞6554.6
Directed proteomics	SHBG	＜25.8	25.8-38.8	38.9-62.9	＞62.9
						Lipid LC/MS	18:1/17:1/16:0TG	＜1649.8	1649.8-1861.9	1862.0-2189.9	＞2189.9
GCMS	Arabinose	＜26944.7	26944.7-33111.6	33111.7-43695.4	＞43695.4
						Lipid LC/MS	18:1/18:1/17:0TG	＜1827.5	1827.5-2473.7	2473.8-3245.7	＞3245.7

In the above table 4, PC means phosphatidylcholine, PE means phosphatidylethanolamine, TG means triacylglycerol, SM means sphingomyelin, LPC means lyso-phosphatidylcholine, IL means interleukin, and x: y means a fatty acid chain containing x carbon atoms and y double bonds. The entry in the column labeled 'odds ratio'(s) is the odds ratio corresponding to a comparison of subjects whose analyte concentrations fall within the fourth quartile, relative to subjects whose analyte concentrations fall within the first quartile, where the quartile is the quartile of the overall distribution of analyte concentrations in all subjects of the study. For analytes measured by mass spectrometry, the units of measurement are denoted as 'arbitrary units' or 'AU'. The concentration measurements from the mass spectrometer represent the processed ion counts detected by the detector of the mass spectrometry instrument. For analytes measured by mass spectrometry, no absolute quantification is performed.

The following fatty acid analytes are correlated with MACCE risk, with the first analyte having the strongest MACCE risk correlation and the last analyte having the weakest MACCE risk correlation within the list of analytes identified: 18:2/18:1/17:0TG, 18:0/20:4PE, 16:0/18:1PC, 16:0/22:6PE, 20:4LPC, 20:1/18:1/18:1TG, 18:1/17:1/16:0TG, 18:1/18:0/18:0TG, 18:1/18:1/17:0TG, 16:0SM, and 16:0/16:0 PC.

For each of the fatty acid analytes identified in table 4, the higher the fatty acid concentration, except for 20:4LPC, the stronger the correlation with MACCE risk likely (e.g., subjects falling within the fourth quartile of the population distribution of 18:0/20:4PE are more likely to suffer from MACCE than subjects falling within the lowest, i.e., first quartile). For 20:4LPC, the opposite is likely to be true: subjects falling within the lowest or first quartile of the 20:4LPC population distribution are more likely to suffer from MACCE than subjects falling within the highest, i.e., fourth quartile.

According to certain embodiments, a clinician may identify a subject with an analyte measurement falling within the fourth quartile range in table 4 above as an individual who is likely to be at high risk for MACCE within 2 years (a odds ratio of less than 1 means that a subject with the first quartile value is at higher risk, except for apolipoprotein a1, 20:4LPC, or arabinose with a odds ratio of less than 1). According to another embodiment, a clinician may identify subjects with analyte measurements falling within the second, third, or fourth quartile ranges in table 4 above as likely to be at high risk for MACCE within 2 years. According to yet another embodiment, a clinician may identify subjects whose analyte measurements fall within the third or fourth quartile ranges in table 4 above as being likely to be at high risk for MACCE within 2 years.

Multivariate classification method

Supervised multivariate prediction model: a multivariate predictive model was constructed in which the outcome to be predicted was MACCE occurring within 2 years of index intubation. The model input variables are: (i) bioanalytical platform-specific plasma analytes; and (ii) profiling the analyzed plasma analytes by all bioanalytical platforms together. Each of the various multivariate models was constructed with or without adjustment according to conventional clinical factors.

A multivariate classifier Random Forest (Breiman, Machine Learning 45: 5-32, 2001) was applied to the data to obtain a multivariate fingerprint predicting MACCE. To assess the statistical significance of the resulting model, an external cross-validation and marker displacement assay was performed (Ambroise & McLachlan, Proc. Natl. Acad. Sci.99: 6562-2, 2002). The following outputs are available from each classifier:

1. predictive multivariate plasma analyte candidates for MACE prediction in the 'case' and 'control' groups were reported, the analytes ranked by classification accuracy importance;

2. the sensitivity versus specificity ROC curve for each classifier (see, e.g., fig. 1 and 2).

Special cases with more than 50% of missed-findings in case or control were excluded from this multivariate analysis. Particular cases with less than 50% of missed measures in cases and controls have missed measures entered using the method described by Troyanskaya et al (Bioinformatics 17(6)520-525, 2001).

Analyte identified by the method

By using multivariate statistical classifier analysis methods (Random forms-Breiman, Machine Learning 45: 5-32, 2001) andcombination of 723 analytes Analyte analysis data setWe have discovered a collection of analytes and algorithms that can be used to predict the occurrence of a severely adverse cardiovascular or cerebrovascular event in an individual within 2 years from the time point at which a blood sample was obtained. Table 5 below gives the 50 most important scored lines in the list of 723 total analytes in their order or importance in the Random forms classifierAnalytes, wherein the order or importance is based on one starting seed (assay # 1). A partial analyte importance list from Random forms analysis and recursive feature elimination to obtain a classifier containing only 50 analyte components with the same starting seed as analysis #1 is shown in the column labeled "order of importance for analysis # 2". A list of the importance of the partial analytes is shown in analysis #3, #4 and #5 columns from another 3 Random forms analyses and recursive feature elimination methods to obtain classifiers containing only 50 analyte components from different starting seeds. Regardless of what starting seed is used for the Random forms analysis or whether recursive feature elimination is used to obtain a fixed 50 analyte component classifier, the most important 20 analytes are almost always within the top 50 analytes ranked according to importance from assay # 1.

Table 5: the 50 most important analytes for predicting MACCE

Analyte of interest	Importance order from analysis #1	Importance order from analysis #2	Importance order from analysis #3	Importance order from analysis #4	Importance order from analysis #5
						Tissue factor	1	3	3	2	3
Cancer antigen 125	2	2	1	3	2
						Glutathione S-transferase	3	1	2	1	1
Alpha-fetoprotein	4	4	4	4	4
						IPI00028413(ITIH3)	5	6	12	9	11
IL-3	6	9	8	7	8
						103_0at233 (arabinose-related fragment (M))	7	7	7	10	10
Von willebrand factor	8	5	6	5	7
						IPI00011264(CFHR1)	9	12	15	12	12
IL-8	10	*	10	11	14

178_1at107 (cysteine M + H (M))	11	11	*	6	5
						114_ Oat120 (acetoacetate-related fragment)	12	13	11	18	17
(M))
						Factor VII	13		27	23	*
IPI00019755(GSTO1)	14	25	17	35	21
						292_0at200 (erythronic acid related fragment (M))	15	16	25	*	20
IPI00022417(LRG1)	16	30	37	15	15
						MMP-2	17	14	9	14	6
Fibrinogen	18	8	5	8	9
						TNF RII	19	19	13	*	*
TNF-β	20	10	16	21	16
						IPI00550991(SERPINA3)	21	20	29	19	18
IPI00298971(VTN)	22	21	23	27	*
						IPI00641737(HP)	23	*	34	20	27
IPI00296176(F9)	24	46	*	*	32
						202_0at311 (L-Tryptophan M + H (M))	25	*	30	36	*
ICAM-1	26	28		25	19
						217_0at294 (inositol-related fragment (M))	27	*	31	*	*

103_0at114 (acetoacetate-related fragment (M))	28	*	36	28	33
						259_ Oat375 (unknown P7478_ uk15 related fragment (M))	29	*	*	*	43
IPI00167093(CFHR1)	30	48	*	*	*
						IPI00021885(FGA)	31	*	*	*	*
IPI00022429(ORM1)	32	*	*	*	*
						IPI00019943(AFM)	33	17	35	34	30
226_2at018 (methylhistidine M + H (M))	34	44	39	44	31
						IPI00647915(TAGLN2)	35
191_0at110 (3-hydroxybutyric acid-related fragment (M))	36	*	*	*	*
						361_0at372 (sucrose-related fragment (M))	37	*	*	*	*
230_0at194(L-4-hydroxyproline	38	*	*	*	*
						Related fragment (M)
1880at097 (pyruvate-related fragment (M))	39	*	*	*	*
						IPI00290283(MASP 1)	40	*	*	*	*
IPI00654888(KLKB 1)	41	35	*	*	*
						IPI00020996(IGFALS)	42	*	*	*	*
IPI00029658(EFEMP 1)	43	24	20	30	28
						764_1213(16:0/22:6PEM+H(M))	44	33	*	37	*

IPI00745872(ALB)	45	*	18	17	26
						SGOT	46	15	19	13	13
278_2at143	47	*	*	*	*
						IPI00745933	48	*	*	*	*

^＊The analyte is not a member of the first 50 analytes in a particular assay

A list of all analytes from the Random forms analysis with 4 starting seeds is shown in appendix 1-4, and in addition the first 50 analytes from the Random forms plus recursive feature elimination method with the same four starting seeds is shown in appendices 5-8.

By using the levels of the first 20 analytes from assay #1 in a plasma sample of an individual in conjunction with an algorithm, it would be possible to predict the occurrence of MACCE within 2 years for that individual with a sensitivity of 87% and a specificity of 87%. This is illustrated by the Receiver Operating Characteristic (ROC) curve shown in fig. 1.

By using multivariate statistical classifier analysis methods (Random forms-Breiman, Machine Learning 45: 5-32, 2001) andof 66 analytesIndividualsOrientation Proteomic bioanalytical data setWe have discovered a set of analytes and algorithms that can be used to predict the occurrence of a heart attack within 2 years of the time point at which an individual obtained a blood sample. Table 6 below gives the 10 most important analytes in one such subset of analytes and their order or importance in the classifier derived from multivariate analysis.

Table 6: prediction of 10 of the most important analytes of MACCE

Protein analytes	Importance of variables in classifiers
		Glutathione S-transferase	0.048
Cancer antigen 125	0.045
		Tissue factor	0.04
Alpha-fetoprotein	0.028
		Von willebrand factor	0.023
Fibrinogen	0.02
		SGOT	0.015

IL-3	0.014
		IL-8	0.013
CD40	0.008

By using the levels of these first 10 analytes in plasma samples from individuals in conjunction with an algorithm, it would be possible to predict the occurrence of MACCE within 2 years for an individual with a sensitivity of 82% and a specificity of 87%. This is illustrated by the Receiver Operating Characteristic (ROC) curve shown in fig. 2.

According to certain embodiments, the MACCE index may be defined by the following linear equation for the first ten (10) analytes:

y＝c₁x₁+c₂x₂+c₃x₃+c₄x₄+c₅x₅+c₆x₆+c₇x₇+c₈x₈+c₉x₉+c₁₀x₁₀

[ equation 1]

Wherein c is₁To c₁₀The values correspond to the values listed in Table 9 below, variable x₁To x₁₀Representative of the analyte measurements listed in table 7 below. Analyte such as aboveDirected proteomics: multiplex immunoassayThe method described in section a measurements. The units of measurement for each analyte are shown in the table below. The values must be normalized before the measured values are substituted into equation 1. That is, each measurement value is subtracted by the average of all values derived from the entire study population (i.e., all cases and control subjects), and the results are then divided by the standard deviation of analyte measurements derived from the entire study population (i.e., all cases and control subjects). The corresponding values (i.e., population mean and standard deviation) for the 10 analytes of equation 1 are listed in table 8.

Using an equation such as equation 1, a subject with a measurement of these 10 analytes, each normalized to the appropriate value in table 8 as described above, from which the resulting value y, i.e., MACCE index, is greater than zero, will be defined as being at risk for MACCE within 2 years, otherwise will be defined as not being at risk for MACCE within 2 years.

Table 7: equation 1, variable x

Variables in equation 1	Analyte of interest	Measurement Unit
			x1	Alpha-fetoprotein	Nanogram/ml (ng/ml)
x2	Cancer antigen 125	Unit/ml (U/ml)
			x3	CD40	ng/ml
x4	Fibrinogen	mg/ml
			x5	Glutathione S-transferase	ng/ml
x6	IL-3	ng/ml
			x7	IL-8	Picogram per milliliter (pg/ml)

x8	SGOT	Microgram/ml
			x9	Tissue factor	ng/ml
x10	Von willebrand factor	Microgram/ml

Table 8: standard value

The analytes set forth in equation 1	Population mean (in units of measurement in Table 7)	Standard deviation (in units of measurement in Table 7)
			Alpha-fetoprotein	1.729	1.138
Cancer antigen 125	20.180	45.201
			CD40	1.195	2.916
Fibrinogen	4.515	1.603
			Glutathione S-transferase	1.877	0.971
IL-3	0.555	0.139
			IL-8	19.795	12.619
SGOT	26.108	7.069
			Tissue factor	3.392	2.135
Von willebrand factor	54.738	29.320

Table 9: equation 1, c variables

c1	-0.26243
		c2	0.0947013
c3	0.00672228
		c4	-0.102292
c5	0.138434
		c6	-0.0191226

c7	-0.121236
		c8	0.0659886
c9	0.181278
		c10	-0.0900255

Another MACCE index can be defined for the first four (4) analytes by the following linear equation:

y＝c₁x₁+c₂x₂+c₃x₃+c₄x₄[ equation 2]

Wherein c is₁To c₄The values are presented as the values listed in Table 11 below, variable x₁To x₁₀Representative of the analyte measurements of table 10 below. Analyte as aboveDirected proteomics: multiple purpose Meta-immunoassayThe measurements described in section. The units of measurement for each analyte are shown in the table below, and the values must be normalized before the measurements can be substituted into equation 2. That is, each measurement value is subtracted by the average of all values derived from the entire study population (i.e., all cases and control subjects), and the results are then divided by the standard deviation of analyte measurements derived from the entire study population (i.e., all cases and control subjects). The corresponding values (i.e., population mean and standard deviation) for the 4 analytes of equation 2 are listed in table 8 above.

Table 10: equation 2, variable x

Variables in equation 2	Analyte of interest	Measurement Unit
			x1	Alpha-fetoprotein	Nanogram/ml (ng/ml)
x2	Cancer antigen 125	Unit/ml (U/ml)
			x3	Glutathione S-transferase	ng/ml
x4	Tissue factor	ng/ml

Table 11: equation 2, c variables

c1	-0.240085
		c2	0.0331361
c3	0.17824
		c4	0.207721

Using an equation such as equation 2, subjects with measurements of these four (4) analytes, each normalized to the appropriate value in table 8 as described above, with the resulting value y of these 4 analytes in the above equation, i.e., a MACCE index, greater than zero, would be defined as being at risk for MACCE within 2 years, and otherwise would be defined as not being at risk for MACCE within 2 years.

An additional multivariate classification analysis method, microarray predictive analysis (PAM, Tibshirani et al, PNAS 99: 6567-. This analysis yields the best set of minimal analytes for defining cases or controls. The analyte collection of four (4) PAM analyses performed with the same four seeds as analyses #2, #3, #4 and #5 in the Random forms case above is shown in the table below:

table 12: seed #1

Seed #1

Sensitivity is 83.8%; specificity is 83.8%

Table 13: seed #2

Seed of corn#2

Sensitivity is 82.4%; specificity 80.9%

Table 14: seed #3

Seed of corn#3

Sensitivity is 80.9%; specificity 80.9%

Table 15: seed #4

Seed of corn#4

Sensitivity is 80.9%; specificity 82.4%

This alternative multivariate classifier analysis of PAM produced essentially the same set of analytes for each different starting seed, with the ranking order of importance of the analytes being consistent, although different from the Random forms analysis.

The present invention includes the identity of a set of analytes that can be measured in a blood sample from an individual and used in conjunction with an algorithm to calculate a MACCE index and predict the occurrence of MACCE in the individual within 2 years after the blood sample.

The invention also includes algorithms that can be used with blood, plasma or serum level information for a collection of analytes to calculate a MACCE index and predict the occurrence of MACCE in the individual within 2 years from the time of blood sampling, as described for a collection of analytes (e.g., the Breiman collection) (Machine Learning 45: 5-32, 2001).

The present invention also includes an instrument that can measure analyte pool levels from a blood, plasma or serum sample of an individual, which can be used with an algorithm to calculate a MACCE index and predict the occurrence of MACCE in the individual within 2 years after the blood sample is taken.

The present invention also includes reagents (antibodies and other types of affinity reagents) that can be used in assays that measure the level of a pool of analytes in blood, plasma or serum that can be used with algorithms to calculate MACCE indices and predict the occurrence of MACCE in an individual within 2 years after blood sampling.

The present invention is intended to cover embodiments in other specific forms without departing from its spirit or essential characteristics. The foregoing embodiments are therefore to be considered in all respects illustrative rather than limiting of the inventive teachings described herein. The scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are intended to be embraced therein.

Claims (29)

1. A method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, the method comprising:

measuring the concentration of a set of analytes in a human blood-based sample, the set of analytes comprising alpha-fetoprotein, cancer antigen 125, glutathione S-transferase, and tissue factor;

determining a MACCE index for the set of analytes; and

identifying the human as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event if the MACCE index is greater than zero; or the MACCE index is less than or equal to zero, identifying the human as having a reduced likelihood of a major adverse cardiovascular or cerebrovascular event.

2. A method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, the method comprising:

determining a MACCE index having a value indicative of the likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentrations of a set of analytes in a blood-based sample of a human, wherein the set of analytes comprises alpha-fetoprotein, a cancer antigen 125, glutathione S-transferase, and tissue factor; and

transmitting, displaying, storing or outputting at least one of the MACCE index, the likelihood of a major adverse cardiovascular or cerebrovascular event, or an equivalent thereof, to a user's interface device, computer-readable storage medium, or a local or remote computer system.

3. A method of treating a human, the method comprising:

determining a MACCE index having a value indicative of the likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentrations of a set of analytes in a blood-based sample of a human, wherein the set of analytes comprises alpha-fetoprotein, a cancer antigen 125, glutathione S-transferase, and tissue factor; and

recommending, approving or administering treatment if the human is identified as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event.

4. The method of any one of claims 1-3, wherein the collection of analytes further comprises CD40, fibrinogen, IL-3, IL-8, SGOT, and von willebrand factor.

5. The method of any one of claims 1 to 4, wherein determining the MACCE index for the set of analytes comprises:

normalizing the measured concentration of each analyte to obtain a normalized concentration;

multiplying the normalized concentration of each analyte by a constant of the analyte to obtain a value of the analyte; and

and adding the analyte values of the analytes to obtain the MACCE index.

6. The method of claim 5, wherein normalizing the measured concentration comprises subtracting a population mean from the measured concentration to obtain a result, and then dividing the result by a standard deviation of the population mean.

7. A method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, the method comprising:

measuring in a blood-based sample of a human the concentration of a set of analytes consisting of alpha-fetoprotein, cancer antigen 125, CD40, fibrinogen, glutathione S-transferase, IL-3, IL-8, SGOT, tissue factor, and von willebrand factor;

determining a MACCE index for the set of analytes; and

identifying the human as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event if the MACCE index is greater than zero; or the MACCE index is less than or equal to zero, identifying the human as having a reduced likelihood of a major adverse cardiovascular or cerebrovascular event.

8. A method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, the method comprising:

determining a MACCE index having a value indicative of the likelihood of a serious adverse cardiovascular or cerebrovascular event based on the measured concentrations of a set of analytes in a blood-based sample of a human, wherein the set of analytes consists of alpha-fetoprotein, cancer antigen 125, CD40, fibrinogen, glutathione S-transferase, IL-3, IL-8, SGOT, tissue factor, and von willebrand factor; and

transmitting, displaying, storing or outputting at least one of the MACCE index, the likelihood of a major adverse cardiovascular or cerebrovascular event, or an equivalent thereof, to a user's interface device, computer-readable storage medium, or a local or remote computer system.

9. A method of treating a human, the method comprising:

determining a MACCE index having a value indicative of the likelihood of a serious adverse cardiovascular or cerebrovascular event based on the measured concentrations of a set of analytes in a blood-based sample of a human, wherein the set of analytes consists of alpha-fetoprotein, cancer antigen 125, CD40, fibrinogen, glutathione S-transferase, IL-3, IL-8, SGOT, tissue factor, and von willebrand factor; and

recommending, approving or administering treatment if the human is identified as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event.

10. The method of any one of claims 7-9, wherein determining the MACCE index for the set of analytes comprises:

normalizing the measured concentration of each analyte to obtain a normalized concentration;

multiplying the normalized concentration of each analyte by a constant of the analyte to obtain a value of the analyte; and

and adding the analyte values of the analytes to obtain the MACCE index.

11. The method of claim 10, wherein normalizing the measured concentration comprises subtracting a population mean from the measured concentration to obtain a result, and then dividing the result by a standard deviation of the population mean.

12. A method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, the method comprising:

measuring in a human blood-based sample the concentration of at least one analyte selected from the group consisting of: cysteine, von willebrand factor, IL-8, 16:0/18:1 phosphatidylcholine, N-carboxy-alanine, fibrinogen, MMP-2, 18:0/20:4 phosphatidylethanolamine, apolipoprotein A1, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, alpha-1 antitrypsin, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 20:4 lysophosphatidylcholine, 16:0 sphingomyelin, SHBG, 18:1/17:1, 16:0 triacylglycerol, arabinose, and 18:1/18:1/17:0 triacylglycerol; and

identifying the person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on a comparison of the measured concentration to a predetermined threshold.

13. A method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, the method comprising:

comparing the measured concentration of at least one analyte in a set of analytes in a blood-based sample of a person to a predetermined threshold value to identify a likelihood of a major adverse cardiovascular or cerebrovascular event, wherein the set of analytes is selected from the group consisting of: cysteine, von willebrand factor, IL-8, 16:0/18:1 phosphatidylcholine, N-carboxy-alanine, fibrinogen, MMP-2, 18:0/20:4 phosphatidylethanolamine, apolipoprotein A1, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, alpha-1 antitrypsin, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 20:4 lysophosphatidylcholine, 16:0 sphingomyelin, SHBG, 18:1/17:1, 16:0 triacylglycerol, arabinose, and 18:1/18:1/17:0 triacylglycerol; and

transmitting, displaying, storing or outputting at least one of the measured concentration, the predetermined threshold, and the serious adverse cardiovascular or cerebrovascular event to a user interface device, a computer readable storage medium, or a local or remote computer system.

14. A method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, the method comprising:

measuring in a human blood-based sample the concentration of at least one analyte selected from the group consisting of: 16:0/18:1 phosphatidylcholine, 18:0/20:4 phosphatidylethanolamine, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 20:4 lysophosphatidylcholine, 16:0 sphingomyelin, 18:1/17:1/16:0 triacylglycerol, and 18:1/18:1/17:0 triacylglycerol; and

identifying the person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on a comparison of the measured concentration to a predetermined threshold.

15. A method of assessing the probability of a major adverse cardiovascular or cerebrovascular event in a human, the method comprising:

comparing the measured concentration of at least one analyte of a set of analytes in a blood-based sample of a person to a predetermined threshold value to identify a likelihood of a major adverse cardiovascular or cerebrovascular event, wherein the analyte is selected from the group consisting of: 16:0/18:1 phosphatidylcholine, 18:0/20:4 phosphatidylethanolamine, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 20:4 lysophosphatidylcholine, 16:0 sphingomyelin, 18:1/17:1/16:0 triacylglycerol, and 18:1/18:1/17:0 triacylglycerol; and

transmitting, displaying, storing or outputting at least one of the measured concentration, the predetermined threshold, and the serious adverse cardiovascular or cerebrovascular event to a user interface device, a computer readable storage medium, or a local or remote computer system.

16. A method according to any one of claims 12 to 15, wherein the predetermined threshold value for each of the following analytes is the lower limit of the fourth quartile of the respective analyte in table 4: cysteine, von Willebrand factor, IL-8, 16:0/18:1 phosphatidylcholine, N-carboxy-alanine, fibrinogen, MMP-2, 18:0/20:4 phosphatidylethanolamine, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, alpha-1 antitrypsin, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 16:0 sphingomyelin, SHBG, 18:1/17:1, 16:0 triacylglycerol, and 18:1/18:1/17:0 triacylglycerol, wherein the measured concentration falling within the fourth quartile increases the likelihood of a serious adverse cardiovascular or cerebrovascular event.

17. A method according to any one of claims 12 to 15, wherein the predetermined threshold value for each of the following analytes is the lower limit of the third and fourth quartile of the respective analyte in table 4: cysteine, von Willebrand factor, IL-8, 16:0/18:1 phosphatidylcholine, N-carboxy-alanine, fibrinogen, MMP-2, 18:0/20:4 phosphatidylethanolamine, 16:0/22:6 phosphatidylethanolamine, 18:1/18:0/18:0 triacylglycerol, alpha-1 antitrypsin, 18:2/18:1/17:0 triacylglycerol, 20:1/18:1/18:1 triacylglycerol, 16:0/16:0 phosphatidylcholine, 16:0 sphingomyelin, SHBG, 18:1/17:1, 16:0 triacylglycerol, and 18:1/18:1/17:0 triacylglycerol, wherein the measured concentrations falling within the third and fourth quartiles increase the likelihood of a major adverse cardiovascular or cerebrovascular event.

18. The method of any one of claims 12 to 15, wherein the predetermined threshold value for each analyte of apolipoprotein a1, 20:4 lysophosphatidylcholine and arabinose is the upper limit of the first quartile of the respective corresponding analyte in table 4, wherein the measured concentration falling within the first quartile increases the likelihood of a major adverse cardiovascular or cerebrovascular event.

19. The method of any one of claims 12 to 15, wherein the predetermined threshold for each analyte of apolipoprotein a1, 20:4 lysophosphatidylcholine and arabinose is the upper limit of the first and second quartile of the respective corresponding analyte in table 4, wherein the measured concentration falling within the first and second quartile increases the likelihood of a major adverse cardiovascular or cerebrovascular event.

20. The method of any one of the preceding claims, wherein the blood-based sample comprises serum or plasma.

21. A method for predicting a major adverse cardiovascular or cerebrovascular event in a human, said method comprising:

obtaining at least one blood-based sample from a human;

measuring in a sample from the person the absolute concentration of one or more analytes identified in appendix 1; and

identifying the person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentrations of the one or more analytes identified in appendix 1.

22. A method for predicting a major adverse cardiovascular or cerebrovascular event in a human, said method comprising:

measuring in a sample from the person the absolute concentration of one or more analytes identified in appendix 1;

comparing the absolute concentration to a predetermined threshold;

identifying the person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentrations of the one or more analytes identified in appendix 1; and

transmitting, displaying, storing or outputting at least one of the severe adverse cardiovascular or cerebrovascular event likelihood, absolute concentration, predetermined threshold or equivalent thereof to a user's interface device, computer readable storage medium or local or remote computer system.

23. A method for predicting a major adverse cardiovascular or cerebrovascular event in a human, said method comprising:

obtaining at least one blood-based sample from a human;

measuring the relative concentration of one or more analytes identified in appendix 1 in a sample from the person; and

identifying the person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured relative concentrations of the one or more analytes identified in appendix 1.

24. A method for predicting a major adverse cardiovascular or cerebrovascular event in a human, said method comprising:

measuring the relative concentration of one or more analytes identified in appendix 1 in a sample from the person;

comparing the relative concentration to a predetermined threshold;

identifying the person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the relative concentrations of the one or more analytes identified in appendix 1; and

transmitting, displaying, storing or outputting at least one of the severe adverse cardiovascular or cerebrovascular event likelihood, absolute concentration, predetermined threshold or equivalent thereof to a user's interface device, computer readable storage medium or local or remote computer system.

25. The method of claim 2, 8, 13 or 15, wherein the MACCE index, the likelihood of a major adverse cardiovascular or cerebrovascular event, the measured concentration, the predetermined threshold, or an equivalent thereof is displayed on a screen or a tangible medium.

26. The method of claim 2, 8, 13 or 15, wherein the MACCE index, the likelihood of a major adverse cardiovascular or cerebrovascular event, the measured concentration, the predetermined threshold, or an equivalent thereof is transmitted to a pharmaceutical industry worker.

27. The method of claim 26, wherein the MACCE index, the likelihood of a major adverse cardiovascular or cerebrovascular event, the measured concentration, the predetermined threshold, or an equivalent thereof is transmitted to a medically insured person or physician.

28. A method of treatment, the method comprising:

identifying a person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentration of the one or more analytes identified in appendix 1 in a blood-based sample of the person; and

recommending, approving or administering treatment if the human is identified as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event.

29. A method of identifying a person who should not receive treatment, the method comprising:

identifying a person as having an increased or decreased likelihood of a major adverse cardiovascular or cerebrovascular event based on the measured concentration of the one or more analytes identified in appendix 1 in a blood-based sample of the person; and

unless the person is identified as having an increased likelihood of a major adverse cardiovascular or cerebrovascular event, the recommendation, approval, or administration of treatment is denied.