Metabolic vulnerabilities analyzed by NMR

Abstract

Methods and systems are disclosed for assessing various health markers for distinguishing various health risks. In one embodiment, the method includes assessing biomarkers to determine the relative risk of mortality due to any early cause in metabolic malnutrition-inflammation syndrome (MMIS). The method may include obtaining NMR-derived measurements of GlycA, S-HDLP, branched-chain amino acids (BCAAs), and citrate from a sample of the subject.

Description

Technical Field

[0001] Field The present disclosure generally relates to the analysis of samples. The present disclosure may be particularly suitable for nuclear magnetic resonance (NMR) analysis of samples.

Background Art

[0002] Background The current interest among medical professionals and patients is in techniques that can evaluate the risk of mortality in patients suffering from certain medical conditions. Some medical conditions can result in a higher risk of mortality, but sometimes it can be a combination of one or more medical conditions that leads to an increased risk of mortality. However, the risk of mortality need not be limited to only one medical condition or its combination. The increased risk of mortality can be due to various factors that can be defined herein as risk factors. Risk factors can include cholesterol levels, co-morbidities, family health history (e.g., diabetes, cancer, hypertension), weight, age, and other factors that can affect health. Thus, identifying health markers, also further referred to as biomarkers, that can correlate with a particular risk factor can assist in prediction models in the assessment of increased risk of mortality for a population. Many risk factors can be unpredictable in the assessment of the risk of mortality and sometimes may not correlate well with an increased risk of mortality. For example, a first person who smokes has a higher risk of lung cancer than a second person who does not smoke, which can result in a higher mortality rate, but the second person may have a higher risk of death for separate reasons. As a result, the mortality rate from all causes is used as a term that further indicates the mortality rate of a population regardless of the cause or risk. The interest in identifying certain biomarkers to aid in the prediction of risk and the assessment of the mortality rate from all causes has become natural.

[0003] Many medical conditions can pose a risk of death, but a syndrome known as malnutrition-inflammation complex syndrome (MICS) has attracted particular interest. MICS can be associated with accelerated atherosclerosis and can be a precursor to mortality. MICS can also be defined as malnutrition-inflammation related muscle wasting. In MICS, an individual may or may not exhibit appropriate risk factors that could lead a clinician to initiate a treatment process. Thus, it is necessary to analyze all-cause mortality and / or biomarkers that may be associated with MICS. Summary of the Invention Means for Solving the Problems

[0004] Overview Embodiments of the present disclosure include methods and systems that can evaluate the relative risk of early all-cause death in a person and / or provide mortality risk stratification by evaluating the NMR spectrum of a biological specimen derived from a patient sample. Biological specimens derived from patient samples can include at least blood, serum, saliva, urine, and sputum from which one or more biomarkers can be extracted for NMR testing. In certain embodiments, the analyte of interest can include biomarkers associated with MICS. In certain embodiments, the analyte is analyzed using NMR. NMR, particularly proton NMR, has the ability to detect the protons of a sample as compared to the structure of the compounds in the sample. In essence, NMR can enable a user to predict the specific structure of a compound or, alternatively, detect a specific analyte of interest using an internal standard.

[0005] It can be hypothesized that NMR biomarkers reflect, or can be the manifestation of, the metabolic derangements underlying protein-energy wasting or metabolic malnutrition, as opposed to the aspects of malnutrition and its clinical diagnostic phenotypes of undernutrition. Thus, in many cases, MICS can be similar to MMIS (metabolic malnutrition-inflammatory syndrome). Many biomarkers can be analyzed to deepen the understanding of MICS / MMIS. For example, biomarkers that can be evaluated can include low serum albumin, C-reactive protein, GlycA, S-HDLP, citrate, and branched-chain amino acids (BCAAs) including valine, leucine, and isoleucine. These biomarkers can be parameters that can help establish the boundaries of MICS in patients, and low serum albumin refers to a blood-binding globular protein that can be made by the liver and help maintain the volume in the bloodstream. Thus, low levels of serum albumin can be correlated with malnutrition and potentially liver failure. More specifically, simultaneous measurement of small molecule metabolites such as citrate and BCAAs along with GlycA and S-HDLP can define the parameters of the metabolic malnutrition-inflammatory syndrome (MMIS). C-reactive protein is a protein that is produced in response to high inflammation in the body and is also produced by the liver. GlycA is also a biomarker associated with systemic inflammation. S-HDLP can be defined as small high-density lipoprotein particles that are mainly synthesized in the liver. Citrate is an organic molecule that is produced in response to respiration at the level of the Krebs cycle and can be associated with pH changes in various body fluids including serum. Finally, valine, leucine, and isoleucine (e.g., branched-chain amino acids, BCAAs) are part of a larger group of nine amino acids that can be termed essential amino acids. The body does not produce the nine essential amino acids, meaning that eating habits can ultimately play a role in reducing MICS.

[0006] In one embodiment, a user can obtain a sample of blood, serum, or plasma from a subject for analysis by sampling or nuclear magnetic resonance (NMR) spectroscopy. After sample collection and analysis, deconvolution of the NMR spectrum can be performed. In certain embodiments, deconvolution includes determining the presence and area (i.e., corresponding to concentration) of signals corresponding to certain biomarkers. For example, the structure of GlycA can be difficult to extrapolate from the proton NMR spectrum, especially if one or more biomarkers overlap for a signal. Briefly, considering that GlycA maintains a consistent portion within its structure, the proton positions of that consistent portion can be determined and used in calculations to describe the molecule as a whole. Alternatively, if there is a possibility that no overlap occurs, consistent, individual protons or sets of protons for a second biomarker can be used to identify the presence of one or more biomarkers. If all present biomarkers are detected simultaneously, the proton spectrum can be overlapping, but the known consistent positions for portions within at least one biomarker can be easily detected, allowing the user to extrapolate the spectrum and detect the presence of all biomarkers simultaneously. The term position refers to the x-axis of the proton NMR spectrum read in parts per million (ppm) units.

[0007] In one embodiment of the present disclosure, a method for determining the risk of death associated with MVX can be used. MVX is known as a metabolic vulnerability index, and the MVX score can be correlated with the risk of death. Moreover, one embodiment of the present disclosure may include a method for calculating MVX, where MVX is a biomarker, the detection of low serum albumin, C-reactive protein, GlycA, S-HDLP, citrate, valine, leucine, and isoleucine, more specifically, it can be derived from the simultaneous measurement of GlycA, S-HDLP, citrate, valine, leucine, and isoleucine excluding low serum albumin and C-reactive protein. Another embodiment of the present disclosure may include a method for calculating MVX, where MVX is the detection of biomarkers including GlycA, S-HDLP, citrate, valine, leucine, and isoleucine, more specifically, it can be derived from the simultaneous measurement of six biomarkers, GlycA, S-HDLP, citrate, valine, leucine, and isoleucine. Further, the multimarker score can be in the range of 1 to 100, and by using citrate and three BCAAs for calculation, an index titled the metabolic malnutrition index (MMX) can be generated. Similarly, by using GlycA and S-HDLP, an index called the inflammatory vulnerability index (IVX) can be generated. By combining IVX and MMX, MVX, a composite MICS (or MMIS) multimarker can be calculated. By combining biomarkers of malnutrition and inflammation, a comprehensive, combined marker of the malnutrition (Malunion) - inflammation complex syndrome (MICS) called MVX can be obtained. MVX can have a strong correlation with mortality when evaluated over a 5-year period from one or more cohorts. Thus, it enables researchers to predict the mortality from all causes for MICS based on the calculated values of IVX and MMX generated from proton NMR spectroscopy.

[0008] Embodiments of the present disclosure may include collecting demographic data from at least one or more cohorts of a population. The cohorts may include cohorts from Catheterization Genetics known as CATHGEN, and may include cohorts from the Intermountain Heart Collaborative Study, for combinations of patients where the participants may total over 8,000. Having a large patient population may enable an increase in representation when generating statistical conclusions.

[0009] Further features, advantages, and details of the present disclosure will be appreciated by those of ordinary skill in the art by reading the subsequent figures and detailed description of the preferred embodiments, which are merely illustrative of the present disclosure. Features described with respect to one embodiment may be incorporated with other embodiments even if not specifically considered thereby. That is, it is noted that aspects of the present disclosure described with respect to one embodiment may be incorporated into different embodiments even if not specifically considered with respect thereto. That is, the features of any and / or all embodiments may be combined in any manner and / or combination. The foregoing and other aspects of the present disclosure are described in detail hereinbelow.

Brief Description of the Drawings

[0010]

Figure 1

[0011]

Figure 2

[0012]

Figure 3

[0013]

Figure 4

[0014]

Figure 5

[0015]

Figure 6

[0016]

Figure 7

[0017]

Figure 8

[0018]

Figure 9

[0019]

Figure 10

[0020]

Figure 11

[0021]

Figure 12

[0022]

Figure 13

[0023]

Figure 14

[0024]

Figure 15

[0025]

Figure 16

[0026]

Figure 17

[0027]

Figure 18

[0028]

Figure 19

[0029]

Figure 20-1

Figure 20-2

Figure 20-3

[0030]

Figure 21

[0031]

Figure 22

DETAILED DESCRIPTION OF THE INVENTION

[0032] The foregoing and other objects and aspects of the present disclosure are described in detail hereinbelow.

[0033] DETAILED DESCRIPTION The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises" and / or "comprising", when used herein, specify the presence of the described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items. As used herein, phrases such as "between X and Y" and "about between X and Y" should be construed to include X and Y. As used herein, phrases such as "about between X and Y" mean "about between X and about Y". As used herein, phrases such as "about X to Y" mean "about X to about Y".

[0034] Unless defined otherwise, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. Terms such as those defined in commonly used dictionaries should be construed to have a meaning that is consistent with their meaning in the context of the present specification and the relevant technical field, and should not be construed in an idealized or overly formal sense unless expressly so defined herein. Well-known functions or constructions may not be described in detail for the sake of brevity and / or clarity.

[0035] The terms "first", "second", etc. can be used herein to describe various elements, components, regions, layers and / or sections, but it will be understood that these elements, components, regions, layers and / or sections should not be limited by these terms. These terms are merely used to distinguish one element, component, region, layer or section from another. Thus, a first element, component, region, layer or section discussed below could be termed a second element, component, region, layer or section without departing from the teachings of the present disclosure. The order of operations (or steps) is not limited to the order presented in the claims or the figures unless otherwise specifically indicated.

[0036] The term "programmatically" means being executed using a computer program and / or software, an operation by a processor or an ASIC. The terms "electronic" and its derivatives refer to an automatic or semi-automatic operation that is not by mental steps but using a device having an electric circuit and / or module, typically referring to an operation executed programmatically. The terms "automatic" and "automatically" mean being able to execute an operation with minimal or no manual labor or input. The term "semi-automatic" refers to causing a certain degree of input or activation by an operator, but the calculation of the concentration of the ionized component(s) together with the calculation and signal acquisition is done electronically, typically programmatically, without requiring manual input.

[0037] The term "about" refers to + / - 10% (average value or average) of a specified value or number.

[0038] The term "patient" or "subject" is widely used and refers to an individual who provides a biological specimen sample or a biological sample for testing or analysis.

[0039] The term "biological sample" refers to in vitro blood, plasma, serum, CSF, saliva, lavage, sputum or tissue samples from humans or animals. Embodiments of the present disclosure may be particularly suitable for evaluating human blood, plasma or serum biological samples, particularly for GlycA (e.g., not found in urine). Blood, plasma or serum samples can be fasting or non-fasting. Human biological samples can be most preferably obtained from a patient by venipuncture.

[0040] The term "GlycA" refers to a biomarker derived from the measurement of a composite NMR signal from the carbohydrate moiety of an acute phase reactant glycoprotein containing N-acetylglucosamine and / or N-acetylgalactosamine moieties, more particularly from the protons of the 2-NAcGlc and 2-NAcGal methyl groups. The GlycA signal is centered at approximately 2.00 ppm in the plasma NMR spectrum at approximately 47 degrees Celsius (+ / - 0.5 degrees Celsius). The peak position is independent of the spectrometer's electric field but can vary depending on the analysis temperature of the biological sample and is not found in urine biological samples. Thus, the GlycA peak region can vary if the temperature of the test sample fluctuates. As discussed further below, the GlycA NMR signal can include a subset of the NMR signals in a defined peak region such that it includes only clinically relevant signal contributions and can exclude protein contributions to the signals in this region. See U.S. Patent Nos. 9,361,429, 9,470,771 and 9,792,410, which are hereby incorporated by reference as if their contents were fully set forth herein.

[0041] As used herein, chemical shift positions (ppm) refer to NMR spectra referenced internally or externally. In certain embodiments, the position can be referenced internally relative to the CaEDTA signal at 2.519 ppm. Thus, the described peak positions contemplated and / or claimed herein can vary depending on how the chemical shifts are generated or referenced, as is well known to those of skill in the art. Thus, for clarity, some of the described and / or claimed peak positions have equivalent different peak positions at other corresponding chemical shifts, as is well known to those of skill in the art.

[0042] The terms "population reference" and "standard" refer to values defined by large-scale study(ies) of higher-risk patients, such as those enrolled in the Catheterization Genetics (CATHGEN) cardiac catheterization procedure and the Intermountain Heart Collaborative study biorepository, or other studies having a sample size sufficient to represent the parent population or targeted patient population. Since the currently defined normal or low-risk and high-risk population values as levels can change over time, the present disclosure is not limited to the population values in the CATHGEN or Intermountain Heart Collaborative studies. Thus, reference ranges associated with values from defined populations in risk segments (e.g., quartiles or quintiles) can be provided and used to assess elevated or reduced levels and / or risks of having a clinical condition.

[0043] The term "clinical condition" is widely used and includes medical conditions at risk that can indicate that a medical intervention, treatment, treatment adjustment, or exclusion of a particular treatment (e.g., a pharmaceutical), and / or monitoring is appropriate. Identification of the potential for a clinical disease can enable the clinician to treat the condition accordingly, delay or inhibit its onset.

[0044] As used herein, the term "NMR spectral analysis" means obtaining data that can measure each parameter present in a biological sample, such as blood, plasma, or serum, using proton ( 1 H) nuclear magnetic resonance spectroscopy techniques. "Measuring" and its derivatives refer to determining levels or concentrations and / or, for certain lipoprotein subclasses, measuring their average particle size. The term "NMR-derived" means that the relevant measurement values are calculated using NMR signals / spectra from one or more scans of an in vitro biological sample in an NMR spectrometer.

[0045] The term "low field" refers to the region / position on the NMR spectrum that relates to the left side of a particular peak / position / point (higher ppm scale compared to the reference). Conversely, the term "high field" refers to the region / position on the NMR spectrum that relates to the right side of a particular peak / position / point.

[0046] The terms "mathematical model" and "model" are used interchangeably and, when used with "MVX", "metabolic vulnerability index", or "risk", refer to a statistical model of risk used to assess the risk of premature mortality in a subject in the future, typically within 1 to 12 years. The risk model can be or include any suitable model, including but not limited to one or more of a logistic regression model, a Cox proportional hazards regression model, a mixed model, or a hierarchical linear model. The risk model can provide a measure of risk based on the probability of premature mortality within a defined time frame, typically within 1 to 12 years. The risk model can be particularly suitable for providing risk stratification for patients with "intermediate risk" related to a low to moderate likelihood of having a clinical event based on traditional risk factors. The MVX risk model can stratify the relative risk of premature mortality as measured by standard χ2 and / or p-values (the latter by a sufficiently representative study population).

[0047] The term "interaction parameter" refers to at least two different defined parameters combined as a (multiplied) product and / or ratio. Examples of interaction parameters include, but are not limited to, (S-HDLP)(GlycA) and (protein)(citrate).

[0048] The term "multimarker" refers to a multi-component biomarker.

[0049] The term "lipoprotein constituent" refers to a constituent in a mathematical risk model related to lipoprotein particles that includes the size and / or concentration of one or more subclasses (subtypes) of lipoproteins. Lipoprotein constituents can include lipoprotein particle subclasses, concentrations, sizes, ratios, and / or mathematical products (multiplied) of lipoprotein parameters, and / or lipoprotein subclass measurements of defined lipoprotein parameters, or any combination with other parameters, such as GlycA.

[0050] The term "HDLP" refers to a measurement of the number of high-density lipoprotein particles (e.g., HDLP number) that sums the particle concentrations of defined HDL subclasses. Total HDLP can be generated using a total high-density lipoprotein particle measurement that sums the concentrations (μmol / L) of all HDL subclasses (which can be grouped into different size categories such as large, median, and small based on size) within a size range of about 7 nm (average) to about 14 nm (average), typically between 7.4 and 13.5 nm. In some embodiments, HDL can be identified as seven subpopulations (H1 - H7) of different sizes of HDLP that range from the smallest HDLP size associated with H1 to the largest HDLP size associated with H7. In some embodiments, the defined subclasses of HDL particles include small HDL particles (S-HDLP). In some embodiments, S-HDLP can include HDL particle subclasses having a diameter between about 7.3 nm (average) and about 9.0 nm (average).

[0051] The terms MICS (Malnutrition-Inflammation Complex Syndrome) and MMIS (Metabolic Malnutrition-Inflammation Syndrome) are used interchangeably and synonymously to explain the "paradoxical epidemiology" whereby an increase in conventional cardiovascular risk factors such as body mass index (BMI), serum cholesterol, and blood pressure may be associated with a decrease rather than an increase in cardiovascular and all-cause mortality.

[0052] The terms "sex" and "gender" are used interchangeably and synonymously herein. Thus, in many embodiments, examples, and figures, the terms "sex-specific" and "gender-specific" are used similarly to describe individuals from each of one or more cohorts. Moreover and more specifically, as used herein and for research and computational purposes, "sex" and "gender" are limited to one of two options, male and female.

[0053] Abbreviations can be defined as follows throughout, unless otherwise specified: ASCVD means atherosclerotic cardiovascular disease, CAD means coronary artery disease, CHF means chronic heart failure, CKD means chronic kidney disease, IVX means inflammation vulnerability index, MICS means Malnutrition-Inflammation Complex Syndrome, MMIS means Metabolic Malnutrition-Inflammation Syndrome, NMR means nuclear magnetic resonance, S-HDLP means small high-density lipoprotein particle, MMX means metabolic malnutrition index, MVX means metabolic vulnerability index. Lipoprotein

[0054] Lipoproteins can contain a wide variety of particles found in plasma, serum, whole blood, and lymph, including triglycerides, cholesterol, phospholipids, sphingolipids, and proteins in various types and amounts. These various particles enable the solubilization of inherently hydrophobic lipid molecules in the blood and perform various functions related to lipolysis, lipogenesis, and lipid transport between the intestine, liver, muscle tissue, and adipose tissue. In blood and / or plasma, lipoproteins can generally be classified in many ways based on physical properties such as density or electrophoretic mobility, or on measures of apolipoprotein content such as apoB or apoA-1, which are the major proteins in LDL and HDL, respectively.

[0055] Classification based on particle size determined by nuclear magnetic resonance can distinguish distinct lipoprotein particles based on size or size range. For example, NMR measurements can identify at least 15 distinct lipoprotein particle subtypes, including at least 7 subtypes of high-density lipoprotein (HDL), at least 3 subtypes of low-density lipoprotein (LDL), and at least 5 subtypes of very low-density lipoprotein (VLDL), which can also be named TRL (triglyceride-rich lipoprotein).

[0056] Current analytical methodologies can enable NMR measurements that can provide concentrations of subpopulations of VLDL, LDL, and HDL, resulting in measurements of groups of small and large subpopulations of each group. For example, different size groupings of HDL subpopulations can be used to optimize risk association with mortality from all causes, as discussed further below.

[0057] The NMR-derived estimated lipoprotein sizes described herein typically refer to average measurements, although other size classifications may be used.

[0058] In a preferred embodiment, the MVX risk assessment model parameters can include NMR-derived measurements of deconvolved signals related to the common NMR spectra of lipoproteins, particularly HDL, using a defined deconvolution model that characterizes the deconvolution components for proteins and lipoproteins, including HDL, LDL, VLDL / TRL. This type of analysis can provide rapid acquisition times of less than two minutes, typically between about 20 seconds and 90 seconds, and corresponding rapid programmatic calculations to generate measurements of the model components, followed by programmatic calculations of one or more MVX risk scores using one or more defined risk models.

[0059] Furthermore, while NMR measurements of lipoprotein particles may be contemplated as particularly suitable for the analyses described herein, it is contemplated that other techniques can be used to measure these parameters now or in the future, and it is noted that embodiments of the present disclosure are not limited to this measurement methodology. Different protocols using NMR can be used instead of the deconvolution protocol described herein (including, for example, different deconvolution protocols). See, for example, Kaess et al., The lipoprotein subfraction profile: heritability and identification of quantitative trait loci, J Lipid Res. Vol. 49 pp. 715-723 (2008); and Suna et al., 1H NMR metabolomics of plasma lipoprotein subclasses: elucidation of metabolic clustering by self-organizing maps, NMR Biomed. 2007; 20: 658-672. Flotation and ultracentrifugation using density-based separation techniques for evaluating lipoprotein particles and ion mobility analysis are alternative techniques for measuring lipoprotein subclass particle concentrations.

[0060] According to some specific embodiments of the present disclosure, lipoprotein subclass grouping can be, for example, cumulated to determine the number of HDL or LDL particles. It is noted that the "small, large, and median" size ranges described can vary or be redefined to widen or narrow their upper or lower end values, or even to exclude certain ranges within the described range. The particle sizes described above typically refer to average measured values, although other classifications may be used.

[0061] Embodiments of the present disclosure classify lipoprotein particles into subclasses grouped by size range based on functional / metabolic relevance as assayed by correlation with lipids and metabolic variables. Thus, as described above, the assay can measure over 15 distinct subpopulations (sizes) of lipoprotein particles. These distinct subpopulations can be grouped into defined subclasses for VLDL / TRL and HDL and LDL. Intermediate density lipoprotein (IDL) can be combined with VLDL / TRL or LDL, or can be a separate category in the size range between large LDL and small VLDL. For example, HDL subclass particles typically range (on average) between about 7 nm and about 15 nm, more typically between about 7.3 nm and about 14 nm (e.g., 7.4 nm - 13.5 nm). Total HDL concentration is the sum of the particle concentrations of each subpopulation of that HDL subclass. The different subpopulations of HDLP can be identified by numbers from 1 to 7, where "H1" represents the HDL subpopulation with the smallest size and "H7" is the HDL subpopulation with the largest size. In some embodiments, the defined subclasses of HDL particles include small HDL particles (S-HDLP). In some embodiments, S-HDLP can include a subclass of HDL particles having a diameter between about 7.3 nm (average) and about 9.0 nm (average). BCAA

[0062] In some embodiments, the MVX model includes measurements of at least one BCAA as described in U.S. Patent No. 9,361,429 and U.S. Patent Application 20150149095, which are incorporated herein by reference. The MVX model can include one or more BCAAs including one or more of isoleucine, leucine, and valine (as considered herein). In some embodiments, one or more of the three BCAAs (valine, leucine, and isoleucine) can be quantified by NMR. Ketone bodies

[0063] In some embodiments, the MVX model may include measurements of at least one ketone body (β-hydroxybutyric acid, acetoacetic acid, acetone) that can be obtained by NMR analysis of a biological sample NMR spectrum. The NMR quantification of each of the three ketone bodies is based on the amplitude of their NMR signals, which are derived from separate deconvolution models specific to three spectral regions where the ketone body NMR signals appear. Due to the large-scale overlap with signals from a number of lipoprotein subspecies as well as identified and unidentified small molecule metabolites, deconvolution analysis can be utilized instead of simple integration of the ketone body signals. The induced amplitudes of the β-hydroxybutyric acid, acetoacetic acid, and acetone signals can be converted to concentrations in μmol / L using conversion factors determined by spiking serum with stock ketone body solutions of known concentrations.

[0064] In one embodiment, the β-hydroxybutyric acid methyl signal doublets appearing at approximately 1.16 and 1.15 ppm can be quantified using a linear deconvolution model encompassing the spectral region of 1.07 - 1.33 ppm. This region may include overlapping interfering NMR signals from lipid fatty acid methylene protons of a number of TRL, LDL, and HDL lipoprotein subspecies, serum protein signals, a triplet signal from ethanol (at 1.13, 1.15, and 1.17 ppm), a doublet signal from lactate (at 1.29 and 1.31 ppm), and doublet signals from unidentified metabolites that rarely appear in human serum samples (at 1.10 and 1.11 ppm). In one embodiment, the deconvolution model may include a library of 83 spectral components to accurately account for the amplitudes of the NMR signals from β-hydroxybutyric acid and various interfering substances in serum.

[0065] In one embodiment, the singlet signal of methyl acetoacetate that appears at approximately 2.24 ppm can be quantified using a linear deconvolution model that encompasses the spectral region of 2.22 - 2.39 ppm. This region can include overlapping interfering NMR signals from lipid fatty acid methylene protons of a number of TRL, LDL, and HDL lipoprotein subclasses, serum protein signals, one or more octet signals (2.25 - 2.39 ppm) from β-hydroxybutyric acid, and signals from three unidentified metabolites that appear at 2.22, 2.30, and 2.35 - 2.41 ppm. In one embodiment, the deconvolution model can include a library of 82 spectral components in order to accurately account for the amplitudes of the NMR signals from acetoacetic acid and various interfering substances in the serum.

[0066] In one embodiment, the singlet signal of acetone methyl that appears at 2.19 ppm can be quantified using a linear deconvolution model that encompasses the spectral region of 2.14 - 2.22 ppm. This region can include overlapping interfering NMR signals from lipid fatty acid methylene protons of a number of TRL, LDL, and HDL lipoprotein subclasses, serum protein signals, and a singlet signal from an unidentified metabolite at 2.22 ppm. In one embodiment, the deconvolution model can include a library of 70 spectral components in order to accurately account for the amplitudes of the NMR signals from acetone and various interfering substances in the serum. GlycA

[0067] As described in U.S. Patent No. 9,470,771, which is hereby incorporated by reference in its entirety, GlycA can be measured using a defined mathematical linear deconvolution model. The GlycA measurement can be a unitless parameter assayed by NMR by calculating the area under the peak at defined peak positions in the NMR spectrum. In any case, levels or risks for subgroups having values within the upper half of a defined range, including values in certain subgroups, such as the third and fourth quartiles or the upper 3-5 fifths, etc., can be defined using the GlycA measure for a known population. Citrate

[0068] In some embodiments, the MVX model can include a measurement of citrate, which can be obtained by NMR analysis of the biological sample NMR spectrum. The NMR quantification of citrate can be based on the NMR signal amplitudes of three of the four members of the methylene proton quartet that appear at approximately 2.64, 2.60, and 2.48 ppm, which can be derived from a deconvolution model that assumes a linear baseline and variable offset. The fourth member of the citrate signal quartet can appear at approximately 2.52 ppm and can overlap with the singlet signal from CaEDTA that functions as an internal chemical shift reference. The derived amplitude of the citrate signal can be converted to a concentration in μmol / L using a conversion factor determined by spiking the serum with a stock citrate solution of known concentration. Serum protein

[0069] In some embodiments, the MVX model may include measurements of serum proteins that can be obtained by NMR analysis of the NMR spectrum of a biological sample. Alternatively, serum protein or serum albumin measurements, if used, can be obtained in a conventional manner. NMR quantification of serum proteins can be based on the amplitude of a broad NMR signal from non-lipoprotein proteins derived from a linear deconvolution model encompassing a spectral region of 0.71 - 1.03 ppm. This region includes overlapping interfering NMR signals from lipid fatty acid methyl protons of a number of TRL, LDL, and HDL lipoprotein subclasses, as well as signals from branched-chain amino acids valine, leucine, and isoleucine. In one embodiment, the deconvolution model may include a library of 66 spectral components to accurately account for the amplitudes of NMR signals from serum proteins and various interfering substances in serum. The induced amplitude of the serum protein signal can be reported in arbitrary units of signal amplitude or converted to molar concentration units of concentration using a conversion factor determined by spiking serum with a stock serum albumin solution of known concentration. Method

[0070] In certain embodiments, the method includes the step of evaluating the NMR spectrum of at least one biomarker including high density lipoprotein (HDL), GlycA, branched-chain amino acids, ketones, citrate, or protein derived from a patient sample. In further embodiments, a plurality of said biomarkers are evaluated. In another embodiment, the method includes the step of evaluating the NMR spectrum of at least one biomarker including high density lipoprotein (HDL), GlycA, branched-chain amino acids, or citrate derived from a patient sample. In additional embodiments, a score is created using the result(s) of the NMR evaluation of one or more of said biomarkers to account for the risk of death and / or relative risk of death of a patient compared to other patients.

[0071] In certain embodiments of the present invention, high-density lipoprotein (HDL) biomarkers can include a wide variety of particles found in plasma, serum, whole blood, and lymph fluid, including various types and amounts of triglycerides, cholesterol, phospholipids, sphingolipids, and proteins. HDL biomarkers can include, but are not limited to, large HDL particles (L-HDLP) having a diameter of 9.5 to 12.0 nm; small HDL particles (S-HDLP) having a diameter of 7.4 to 8.7 nm; and the like, including HDL subclasses.

[0072] In certain embodiments, amino acids can be utilized to contribute to the generation of the MVX index, and branched-chain amino acid biomarkers can include, but are not limited to, amino acids including valine, leucine, and / or isoleucine, or any combination thereof.

[0073] In certain embodiments, glycoprotein biomarkers can include one or more acute-phase glycoproteins that can include at least GlycA.

[0074] In certain embodiments, organic biomarkers can include at least citrate molecules.

[0075] In certain embodiments, lipoprotein particles can include high-density lipoprotein (HDL) biomarkers, more specifically, small high-density lipoprotein particles.

[0076] In certain embodiments, plasma protein biomarkers can include plasma protein subclasses including, but not limited to, albumin.

[0077] In certain embodiments, ketone body biomarkers can include a subset of ketone bodies including, but not limited to, acetone, acetoacetic acid, or beta-hydroxybutyric acid.

[0078] As described, the results of the evaluation of the biomarker(s) in certain embodiments of the present invention can be used to create a score that explains the patient's risk of death and / or relative risk of death. U.S. Patent No. 11,156,621, the disclosure of which is hereby incorporated by reference herein, includes consideration of a "metabolic vulnerability index" (MVX). The MVX score can be calculated using one or more NMR-derived measurements from certain embodiments of the present invention. The one or more NMR-derived measurements can include proton NMR spectra resulting from at least one BCAA biomarker, at least one HDL biomarker, at least one glycoprotein biomarker, a citrate biomarker, a plasma protein biomarker, and a ketone body biomarker, or combinations thereof.

[0079] In embodiments of the present disclosure, the at least one or more biomarkers, individually or in combination, provide data for explaining the patient's risk of death and / or relative risk of death when analyzed in NMR spectroscopy measurements. When calculating the MVX to demonstrate the risk of death and / or relative risk of death in a patient, biomarkers and subclasses of biomarkers can be used. In still other embodiments, when calculating the MVX to demonstrate the risk of death and / or relative risk of death, other biomarkers not included herein can be utilized and analyzed using NMR spectroscopy measurements to generate an MVX score. Enhanced multimarker performance can be achieved by including additional biomarkers.

[0080] Embodiments of the present disclosure include a method for determining the level of a marker associated with the risk of premature death in a person. The method may include obtaining a sample from the person and measuring GlycA, at least one high-density lipoprotein particle (HDLP) subclass, at least one branched-chain amino acid (BCAA), at least one ketone body, at least one serum protein, and citrate. The method may further include obtaining a sample from the person and measuring GlycA, at least one high-density lipoprotein particle (HDLP) subclass, at least one branched-chain amino acid (BCAA), and citrate. In certain embodiments, the high-density lipoprotein particle is a small HDL particle (S-HDLP).

[0081] In some embodiments, these measurements are used to generate an MVX score. In some embodiments, the MVX score is determined by using the following model: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β4*lnBCAA + β5*lnKetoneBody. In some embodiments, the MVX value is determined by using the following model: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP) + β4*lnBCAA + β5*lnKetoneBody.

[0082] Throughout the present disclosure, it is noted that the empirical values of A and β1-β n may vary depending on the model used. For example, β1 in the first above equation for MVX (i.e., the equation without the term β3*(lnGlycA*lnS-HDLP)) will generally be a different value than β1 in the second above equation (i.e., the equation including the product term).

[0083] In some embodiments, the method includes measuring at least one of citrate, at least one HDLP subclass, and at least one BCAA in addition to GlycA. In some embodiments, the method includes simultaneously measuring citrate, GlycA, at least one HDLP subclass, and at least one BCAA. In some other embodiments, the method includes measuring at least one of citrate and protein, at least one HDLP subclass, at least one BCAA, and at least one ketone body in addition to GlycA. In some embodiments, the step of measuring includes at least one of citrate (Citrate) and serum protein (Protein) and is performed in a subject considered to be at high risk of death. In some embodiments, the MVX value is determined using the following model: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP) + β4*lnBCAA + β5*lnKetoneBody + β6*lnCitrate + β7*lnProtein + β8*(lnCitrate*lnProtein).

[0084] In some embodiments, the MVX value is defined as including an Inflammatory Vulnerability Index (IVX) value and a Metabolic Malnutrition Index (MMX) value. The values MVX, IVX, and MMX can all be gender-specific. For example, MVX, which is a female-specific value F can be defined as including the female-specific value MMX F and IVX F . Alternatively, MVX M can be a male-specific value and can be defined as including the male-specific value MMX M and IVX M . Generally, the MVX value can be defined as including an Inflammatory Vulnerability Index (IVX) value and a Metabolic Malnutrition Index (MMX) value, regardless of gender.

[0085] In some embodiments, measurements of at least GlycA and at least one HDLP subclass are used to generate an Inflammatory Vulnerability Index (IVX) value. In some embodiments, the IVX value is determined using the following model: IVX = β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP). In other embodiments, the IVX F value can be determined using the following model: 9 + (GlycA * -0.000187) + (S-HDLP * -0.3585) + ((GlycA * S-HDLP) * 0.000348). The IVX F within which the resulting score can vary, but can generally include a range of 3.0 to 9.0 (inclusive of both ends), corresponding to scores of 1 to 100 respectively. Moreover, the IVX M can be determined using the following model: 9 + (GlycA * -0.00437) + (S-HDLP * -0.52307) + ((GlycA * S-HDLP) * 0.000817). The IVX M within which the resulting score can vary, but can generally include a range of 1.4 to 7.6 (inclusive of both ends), corresponding to scores of 1 to 100 respectively.

[0086] In some embodiments, measurements of at least one BCAA and at least one ketone body are used to generate a Metabolic Malnutrition Index (MMX) value. In some embodiments, the MMX value is determined using the following model: MMX = β4*lnBCAA + β5*lnKetoneBody. This model is denoted as MMX1 and can be a calculation used for populations / subjects considered to be at low risk, as discussed in detail herein.

[0087] In other embodiments, measurements of at least one BCAA, at least one ketone body, citrate, and protein are used to generate an alternative MMX value. For example, in some embodiments, the MMX value can be determined using the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein. Alternatively, the MMX value can be determined using the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein). In such embodiments, MMX can be described as follows: MMX = β9 * MMX1 + β10 * MMX2 (where MMX1 is as described above and MMX2 = β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein)).

[0088] Measurements used to generate an MMX value that includes measurements of citrate and protein are generally taken in subjects and / or populations (e.g., subjects who have had a cardiovascular event or symptom suggestive of underlying CVD) that are considered to be at high risk of CVD-related death.

[0089] Also, in other embodiments of the present disclosure, the MMX value can be determined using at least one BCAA and citrate. The MMX value can be gender-specific. Thus, the MMX F value can be found using the following model:

[0090] ((4 + (Leu * -0.03142) + (Leu 2 * 0.0000893)) * 0.353) + ((7 + (Val * -0.03362) + (Val 2 * 0.0000689)) * 0.684) + (Ileu * 0.00332) + ((1 + (Citr * -0.0072) + (Citr 2 * 0.0000573)) * 0.7135).

[0091] MMX F Among them, the resulting scores can vary but generally can include a range of ln(0.8) to ln(1.8) (including both ends), corresponding to scores of 1 to 100 respectively. Moreover, MMX m can be determined using the following model:

[0092] ((4 + (Leu * -0.01594) + (Leu 2 * 0.0000291)) * 1.076) + ((7 + (Val * -0.0239) + (Val 2 * 0.00005)) * 0.414) + (Ileu * 0.01265) + ((1 + (Citr * 0.00906) + (Citr 2 * -0.0000126)) * 0.5881).

[0093] MMX M Among them, the resulting scores can vary but generally can include a range of ln(1.59) to ln(2.1) (including both ends), corresponding to scores of 1 to 100 respectively.

[0094] Thus, in some embodiments, the metabolic vulnerability index can be described as MVX = βi * IVX + βm * MMX. For application to a normal (i.e., low) - risk patient population (e.g., people who have never had a known CV event), the metabolic vulnerability index can be described as MVX1 = βi * IVX + βm * MMX1 (where βi and βm can have specific values depending on the model used). Conversely, for application to a high - risk patient population (e.g., people who have had a known CV event), the metabolic vulnerability index can be described as MVX = βi * IVX + βm * MMX (where βi and βm can have specific values depending on the model used, and MMX = β9 * MMX1 + β10 * MMX2). In some cases, βm is the same for both the low - risk and high - risk models.

[0095] In some embodiments, the female metabolic vulnerability index is MVX F =(IVX F *2.27278)+(lnMMX F *12.13511)+(IVX F *lnMMX F )*-1.09312 and can be described as such, and the resulting scores can vary but generally can include 17 - 25 (including both ends) corresponding to scores from 1 - 100 respectively. Moreover, the male metabolic vulnerability index is MVX M =(IVX M *3.54601)+(lnMMX M *14.41428)+(IVX F *lnMMX M )*-1.43438 and can be described as such, and the resulting scores can vary but generally can include 27 - 34.2 (including both ends) corresponding to scores from 1 - 100 respectively.

[0096] For example, in some cases, the MVX score can be monitored in high - risk subjects or populations as a way to monitor the overall health of the subject and the risk of lethal CV events or death from other causes. Alternatively, the MVX score can be monitored in clinical trials of new pharmaceuticals being conducted in high - risk populations as a way to monitor the effectiveness of the test drug. For low - risk subjects, a clinician can choose to monitor the MVX value as a means of assessing overall health and wellness. The MVX value can, in certain embodiments, be used as a guide for lifestyle changes to promote heart health. System

[0097] Other embodiments are directed to systems. Some embodiments of the disclosure include systems capable of performing each of the methods described herein. In some embodiments, the system includes an NMR spectrometer configured to acquire an NMR spectrum (singular and / or plural) that includes at least one signal for GlycA, at least one signal for at least one subclass of small high-density lipoprotein particles (S-HDLP), at least one signal for at least one branched-chain amino acid (BCAA), and at least one signal for at least one citrate; and a processor for determining a metabolic vulnerability index (MVX) value based on the measured at least one signal for GlycA, at least one subclass of small high-density lipoprotein particles (S-HDLP), at least one branched-chain amino acid (BCAA), and at least one citrate, the processor including or in communication with a memory.

[0098] In some embodiments, the system includes an NMR spectrometer for acquiring at least one NMR spectrum of an in vitro biological sample, and at least one processor in communication with the NMR spectrometer. The at least one processor can use at least one NMR spectrum to determine a metabolic vulnerability index score for each biological sample, based on at least one defined mathematical model of the risk of early mortality that can take into account at least one HDL subclass component measurement, at least one branched-chain amino acid measurement, a GlycA measurement, a citrate measurement, and optionally ketone body and / or protein measurements, obtained from at least one in vitro biological sample of a subject. In one embodiment, the HDLP subclass is small HDLP (S-HDLP). In some embodiments, the processor can be configured to calculate an MVX score based on measurements of GlycA, at least one ketone body, at least one branched-chain amino acid, and at least one of the HDLP subclasses, using the following formula: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β4*lnBCAA + β5*lnKetoneBody. In some embodiments, the processor can be configured to calculate an MVX score using the following model: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP) + β4*lnBCAA + β5*lnKetoneBody. In some embodiments, the processor can be configured to calculate an MVX score using the following model: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP) + β4*lnBCAA + β5*lnKetoneBody + β6*lnCitrate + β7*lnProtein + β8*(lnCitrate*lnProtein). In some embodiments, the processor can be configured to calculate a gender-specific MVX score from one of the following models:

[0099] MVX F =(IVX F *2.27278)+(lnMMX F*12.13511)+(IVX F *lnMMX F )*-1.09312(The resulting scores can vary but generally include 17 - 25 (inclusive), corresponding to scores from 1 to 100 respectively). Alternatively, for males: MVX M =(IVX M *3.54601)+(lnMMX M *14.41428)+(IVX F *lnMMX M )*-1.43438(The resulting scores can vary but generally include 27 - 34.2 (inclusive), corresponding to scores from 1 to 100 respectively).

[0100] In some embodiments, the processor may be configured to calculate an MVX score that includes an Inflammatory Vulnerability Index (IVX) value and a Metabolic Malnutrition Index (MMX) value. Thus, in some embodiments, the processor may be configured to calculate the MVX score based on the following model: MVX = βi * IVX + βm * MMX. In these embodiments, the processor may be configured to calculate the IVX value using the following model: IVX = β1 * lnGlycA + β2 * lnS-HDLP + β3 * (lnGlycA * lnS-HDLP). In some embodiments, the processor may be configured to calculate the MMX value using the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody; this model may be denoted as MMX1. In some embodiments, the processor may be configured to calculate MMX using the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein. In another embodiment, the processor may be configured to calculate MMX using the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein). In such embodiments, MMX can be described as follows: MMX = β9 * MMX1 + β10 * MMX2 (where MMX1 is as described above and MMX2 = β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein)).

[0101] In some embodiments, the processor may be configured to calculate gender-specific MMX values using the following model: MMX F = ((4 + (Leu * -0.03142) + (Leu 2 * 0.0000893)) * 0.353) + ((7 + (Val * -0.03362) + (Val 2 * 0.0000689)) * 0.684) + (Ileu * 0.00332) + ((1+(Citr*-0.0072)+(Citr 2 *0.0000573))*0.7135)。

[0102] MMX F Within this range, the resulting scores can vary but generally can include the range of ln(0.8) to ln(1.8) (including both ends), corresponding to scores from 1 to 100 respectively. MMX m = ((4+(Leu*-0.01594)+(Leu 2 *0.0000291))*1.076)+ ((7+(Val*-0.0239)+(Val 2 *0.00005))*0.414)+(Ileu*0.01265)+ ((1+(Citr*0.00906)+(Citr 2 *-0.0000126))*0.5881)。

[0103] MMX M Within this range, the resulting scores can vary but generally can include the range of ln(1.59) to ln(2.1) (including both ends), corresponding to scores from 1 to 100 respectively.

[0104] Still other embodiments are directed to an NMR system. The system includes an NMR spectrometer, a flow probe in communication with the spectrometer, and at least one processor in communication with the spectrometer. The at least one processor can be configured to obtain (i) at least one NMR signal of a defined GlycA compliant region of an NMR spectrum related to GlycA of a blood, plasma, or serum specimen in the flow probe; (ii) at least one NMR signal of a defined citrate compliant region of an NMR spectrum related to the specimen in the flow probe; (iii) at least one NMR signal of a defined BCAA compliant region of an NMR spectrum related to the specimen in the flow probe; and (iv) at least one NMR signal of an HDLP subclass parameter. The processor can be further configured to calculate measurements of GlycA, citrate, at least one branched chain amino acid, and HDLP subclass parameter. The system can be further configured to calculate an MVX score using a defined mathematical model of the risk of mortality from any cause that uses calculated measurements of at least one of GlycA, citrate, at least one branched chain amino acid, HDLP subclass parameter, and optionally serum protein and / or ketone bodies. In one embodiment, the HDLP subclass is small HDLP (S-HDLP). In some embodiments, the system can be further configured to calculate an MVX score based on measurements of at least one of GlycA, citrate, at least one branched chain amino acid, and HDLP subclass. In some embodiments, the system can be further configured to calculate an MVX score based on measurements of at least one of GlycA, at least one ketone body, at least one branched chain amino acid, and HDLP subclass using the following formula: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β4*lnBCAA + β5*lnKetoneBody.In some embodiments, the system may be further configured to calculate an MVX score using the following model: MVX = A + β1 * lnGlycA + β2 * lnS-HDLP + β3 * (lnGlycA * lnS-HDLP) + β4 * lnBCAA + β5 * lnKetoneBody. In some embodiments, the system may be further configured to calculate an MVX score using the following model: MVX = A + β1 * lnGlycA + β2 * lnS-HDLP + β3 * (lnGlycA * lnS-HDLP) + β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein). In some embodiments, the system may be further configured to calculate an MVX score that includes an Inflammatory Vulnerability Index (IVX) value and a Metabolic Malnutrition Index (MMX) value. In some embodiments, the system may be configured to calculate a gender-specific MVX score from one of the following models using gender-specific IVX and MMX: MVX F =(IVX F *2.27278)+(lnMMX F *12.13511)+(IVX F *lnMMX F )*-1.09312(The resulting scores can vary but generally can include 17 - 25 (inclusive), corresponding to scores from 1 - 100 respectively). Alternatively, for males: MVX M =(IVX M *3.54601)+(lnMMX M *14.41428)+(IVX F *lnMMX M )*-1.43438(The resulting scores can vary but generally can include 27 - 34.2 (inclusive), corresponding to scores from 1 - 100 respectively).

[0105] Thus, in some embodiments, the system can be further configured to calculate an MVX score based on the following model: MVX = βi * IVX + βm * MMX. In these embodiments, the system can be further configured to calculate an IVX value using the following model: IVX = β1 * lnGlycA + β2 * lnS-HDLP + β3 * (lnGlycA * lnS-HDLP). In some embodiments, the system can be further configured to calculate an MMX value using the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody; this model may be denoted as MMX1. In some embodiments, the system can be further configured to calculate an MMX using the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein. In another embodiment, the system can be further configured to calculate an MMX using the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein). In such embodiments, MMX can be described as follows: MMX = β9 * MMX1 + β10 * MMX2 (where MMX1 is as described above and MMX2 = β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein)). Additional methods

[0106] Additional aspects of the present disclosure are directed to methods of monitoring a patient to evaluate treatment or to determine whether the patient is at risk of early mortality. The method may include programming at least one defined MVX mathematical model disclosed herein, the model including a plurality of components including at least one NMR-derived measurement of at least one selected HDLP subclass, branched-chain amino acid, citrate, and GlycA, and optionally at least one of a protein or at least one ketone body. The method may further include programming to deconvolve a spectrum including the NMR-derived measurement. The method may include programming to calculate an MVX score for each patient using at least one defined model and corresponding patient sample measurements, and evaluating (i) whether the MVX score exceeds a defined level of population criteria associated with an increased risk of mortality from any cause; and / or (ii) at least one of whether the metabolic vulnerability index is increasing or decreasing over time in response to treatment. In some embodiments, the MVX score can be calculated based on measurements of GlycA, at least one ketone body, at least one branched-chain amino acid, and at least one of the HDLP subclasses using the following formula: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β4*lnBCAA + β5*lnKetoneBody. In some embodiments, the MVX score can be calculated using the following model: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP) + β4*lnBCAA + β5*lnKetoneBody. In some embodiments, the MVX score can be calculated using the following model: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP) + β4*lnBCAA + β5*lnKetoneBody + β6*lnCitrate + β7*lnProtein + β8*(lnCitrate*lnProtein). In some embodiments, the MVX score calculation may include an inflammation vulnerability index (IVX) value and a metabolic malnutrition index (MMX) value.Thus, in some embodiments, the calculation of the MVX score can be based on the following model: MVX = βi * IVX + βm * MMX. In these embodiments, the calculation of the IVX value can use the following model: IVX = β1 * lnGlycA + β2 * lnS-HDLP + β3 * (lnGlycA * lnS-HDLP). In some embodiments, the calculation of the MMX value can use the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody; this model may be denoted as MMX1. In some embodiments, the calculation of the MMX score can use the following model (may the following model): MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein. In another embodiment, the calculation of the MMX score can use the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein). In such embodiments, MMX can be described as follows: MMX = β9 * MMX1 + β10 * MMX2 (where MMX1 is as described above, and MMX2 = β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein)).

[0107] Methods and systems for determining a subject's metabolic vulnerability index (MVX) are disclosed herein. The method includes utilizing a multivariate model of defined biomarkers to predict the likelihood of mortality due to any early cause in a patient with a malnutrition-inflammation complex syndrome (MICS), also referred to as a malnutrition-inflammatory syndrome (MMIS).

[0108] One embodiment of the present disclosure is a method for determining the level of a marker associated with the relative risk of early death of a subject, the method comprising obtaining a sample from the subject; measuring one or more markers using NMR spectroscopy, wherein the markers can include GlycA, at least one high-density lipoprotein particle (HDLP) subclass, at least one branched-chain amino acid (BCAA), citrate, and optionally at least one ketone body, and at least one subset of serum proteins; determining a metabolic vulnerability index based on the NMR spectrum; repeating sample processing over time; and evaluating at least whether the MVX has increased or decreased over time.

[0109] In certain embodiments, the HDLP is small HDLP (S-HDLP).

[0110] In some embodiments, the MVX score can be gender-specific. In some embodiments, the MVX score can be calculated using IVX and MMX. In some embodiments, the MMX score can be calculated using at least one BCAA and citrate. In some embodiments, the IVX score can be calculated using GlycA measurements and HDLP measurements.

[0111] In some embodiments, the IVX is gender-specific. In the same embodiments, the MMX is also gender-specific. In some embodiments, the HDLP is small HDLP (S-HDLP), while in other embodiments, the HDLP can be medium or large HDLP (M-HDLP, L-HDLP).

[0112] In some embodiments, the MVX score can be calculated based on measurements of at least one of GlycA, at least one ketone body, at least one branched-chain amino acid, and HDLP subclass using the following formula: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β4*lnBCAA + β5*lnKetoneBody.

[0113] In some embodiments, the MVX score can be calculated using the following model: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP) + β4*lnBCAA + β5*lnKetoneBody. In these embodiments, the MVX value can be determined in subjects considered to have a low risk of cardiovascular events.

[0114] In some embodiments, the MVX value may include measurements of serum protein (Protein) and / or citrate (Citrate).

[0115] In some embodiments, the MVX value can be determined using the following model: MVX = A + β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP) + β4*lnBCAA + β5*lnKetoneBody + β6*lnCitrate + β7*lnProtein + β8*(lnCitrate*lnProtein). In these embodiments, the MVX value can be determined in subjects considered to have a high risk of cardiovascular events.

[0116] In alternative embodiments, the MVX value may include an inflammation vulnerability index (IVX) and / or a metabolic malnutrition index (MMX), as disclosed in more detail herein.

[0117] Thus, in some embodiments, the calculation of the MVX score can be based on the following model: MVX = βi*IVX + βm*MMX. In these embodiments, the calculation of the IVX value can use the following model: IVX = β1*lnGlycA + β2*lnS-HDLP + β3*(lnGlycA*lnS-HDLP).

[0118] In some embodiments, the calculation of the MMX value can use the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody; this model is denoted as MMX1 and can be used for subjects considered to have a low risk of cardiovascular disease-related events.

[0119] In some embodiments, the calculation of the MMX score can use the following model (may the following model): MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein.

[0120] In another embodiment, the calculation of the MMX score can use the following model: MMX = β4 * lnBCAA + β5 * lnKetoneBody + β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein).

[0121] In such embodiments, MMX can be described as follows: MMX = β9 * MMX1 + β10 * MMX2 (where MMX1 is as described above and MMX2 = β6 * lnCitrate + β7 * lnProtein + β8 * (lnCitrate * lnProtein)); in these embodiments, the MVX score can be determined for subjects considered to have a high risk of cardiovascular disease-related events.

[0122] In some embodiments, BCAA can be at least one of leucine, isoleucine, or valine.

[0123] In some embodiments, the ketone body can be at least one of acetone, acetoacetic acid, or beta-hydroxybutyric acid.

[0124] In some embodiments, the measuring step is performed by NMR. The metabolic vulnerability index can provide an early mortality risk assessment for short (1 year) to long (12 year) periods. Follow-up observations of the subject may be made at 12 years, at 10 years and / or every 6 years, every 5 years, every 4 years, every 3 years, every 2 years or every 1 year. These risk assessments can generate MVX values that are decoupled from traditional risk factors.

[0125] In some embodiments, gender may be included as a factor in the MVX model.

[0126] In some embodiments, age may be included as a factor in the MVX model. In other embodiments, the MVX model can exclude either consideration of gender or age so as to avoid the generation of false negatives or false positives based on data corruption of such auxiliary data that is not directly related to the biological sample, for example.

[0127] In some embodiments, the MVX score is provided to the clinician based on data electronically correlated to the sample or based on clinician or intake lab input, such as the patient's fasting "F" or non-fasting "NF" and statin "S" or non-statin "NS" characterizations, and this data can be provided in a label associated with the biological sample that is electronically associated with the sample in an NMR analyzer. Alternatively, the patient characterization data may be held in a computer database (by remote or server or other defined route), and may include patient identifiers, sample types, and test types, etc., input into an electronic correlation file by the clinician or intake lab and accessed or hosted by the intake lab in communication with the NMR analyzer. The patient characterization data can enable an appropriate MVX model to be used for a particular patient.

[0128] It is contemplated that the metabolic vulnerability index can be used to monitor subjects in clinical trials and / or drug treatment, identify drug contradictions, and / or monitor changes in the risk status (positive or negative) that may be specific to a particular drug and related to the patient's lifestyle, etc.

[0129] Referring to some embodiments of the present disclosure includes an NMR system capable of performing each method described herein.

[0130] In some embodiments, the NMR system includes an NMR spectrometer, a flow probe in communication with the spectrometer, and (i) at least one NMR signal in a defined GlycA-compliant region of the NMR spectrum related to GlycA in a blood, plasma, or serum sample in the flow probe; (ii) at least one NMR signal in a defined citrate-compliant region of the NMR spectrum related to the sample in the flow probe; (iii) at least one NMR signal in a defined BCAA-compliant region of the NMR spectrum related to the sample in the flow probe; and (iv) at least one NMR signal for at least one HDLP subclass; and optionally, a processor in communication with the spectrometer configured to obtain at least one NMR signal for serum protein and / or ketone bodies.

[0131] In some embodiments, the processor is further configured to calculate an MVX score based on measurements obtained by the spectrometer according to any of the embodiments of the invention disclosed herein.

[0132] Next, further embodiments of the present disclosure are described as the following non-limiting examples.

Example

[0133] The present disclosure can be more deeply understood by referring to the following non-limiting examples. (Example 1)

[0134] A composite biomarker score named the metabolic vulnerability index (MVX) was derived from six metabolites measured simultaneously by a clinically deployed nuclear magnetic resonance (NMR) blood test, which may reflect different etiological aspects of the metabolic malnutrition - inflammation syndrome (plural possible). The MVX score provided a remarkably strong stratification of the risk of death from any cause in two large independent cohorts of cardiac catheterization patients in whom the susceptibility to these dysmetabolic wasting syndromes had not previously been suspected. The contribution of MVX to the multivariable prediction model of 5-year mortality was superior to that of 15 risk factors including age. The risk association was equally strong in men and women, younger and older individuals, overweight and underweight persons, and patients with and without comorbid conditions such as heart failure, renal dysfunction, diabetes, and hypertension. The uniformity of the MVX risk association in higher- and lower-risk patient subgroups suggests that the influence of the metabolic malnutrition - inflammation syndrome on survival may be more universal than previously thought.

[0135] Factors that can affect the risk of mortality may include, among other potential factors, malnutrition and inflammation. Furthermore, there may be a correlation between malnutrition - inflammation and muscle wasting, which can lead to an additional risk of mortality. Malnutrition - inflammation-related muscle wasting can often be described in patients with chronic kidney disease (CKD) or chronic heart failure (CHF). Malnutrition - inflammation-related muscle wasting may also be noted in both healthy and frail individuals, as well as in patients with malignancies and liver diseases. Examples of names given to the syndrome (plural possible) may include protein - energy wasting (PEW), of which cachexia can be a severe form, the malnutrition - inflammation atherosclerosis syndrome, and the malnutrition - inflammation complex syndrome (MICS), and terms that support an implied synergistic relationship between systemic inflammation and protein - energy malnutrition.

[0136] Subpopulations with a high prevalence of MICS may exhibit a "risk factor paradox," also known as "reverse epidemiology," where increases in traditional cardiovascular risk factors such as body mass index (BMI), serum cholesterol, and blood pressure are associated with a decrease rather than an increase in cardiovascular and all-cause mortality. The pathophysiological mechanisms contributing to cardiovascular disease are not necessarily ineffective in the subpopulation, but different overlapping etiologies may be dominant in causing the reversal of multiple conventional risk factors associated with mortality at first glance. Thus, in patients with MICS, the conventional risk factors for lethal versus non-lethal outcomes may be governed by different influences, and these may respond optimally to different therapeutic interventions.

[0137] What may contribute to the further hindrance of the recognition and understanding of MICS and its involvement in the frailty syndrome can be the lack of simple quantitative and objective clinical assessment tools. Malnutrition components can typically be particularly difficult to assess using patient history, physical examination, and anthropomorphic evaluation. The main laboratory biomarkers of MICS, low serum albumin (reflecting both malnutrition and inflammation) and elevated C-reactive protein (CRP), are non-specific and thus their clinical applicability may be limited.

[0138] Mortality from any cause in the large, high-risk CATHGEN (CATHeterization GENetics) cardiac catheterization cohort may be independently associated with two novel biomarkers that plausibly reflect the inflammatory contribution to MICS etiology. Both can be measured by nuclear magnetic resonance (NMR) spectroscopy as part of a clinical NMR LipoProfile® scan. The first may be GlycA, whose composite NMR signal arises from the glycan residues of several acute-phase glycoproteins, thereby providing a sensitive and stable measure of systemic inflammation. The second may be the content of smaller-sized high-density lipoprotein particles (S-HDLP), which may appear to mediate several protective functions carried out by (among other things) bound anti-inflammatory and immune response proteins. Laboratory Methods

[0139] Analysis of the LipoProfile® of fasting EDTA plasma samples is performed on the NMR Profiler platform at LipoScience (now Labcorp, Morrisville, NC) using the LP-4 algorithm. The “scan” (proton NMR spectrum) can generate particle concentrations of one or more subclasses of different sizes of triglyceride-rich lipoproteins (TRL), low-density lipoproteins (LDL), high-density lipoproteins (HDL), mean TRL, LDL, and HDL particle sizes, as well as derived lipids (triglycerides, and total, LDL, and HDL cholesterol), the inflammatory marker GlycA, three branched-chain amino acids (valine, leucine, isoleucine), citrate, plasma proteins, ketone bodies, plus several other small molecule metabolites. In some cases, seven HDL particle subclasses with the indicated estimated diameter (nm) were quantified: H7P (12 nm), H6P (10.8 nm), H5P (10.3 nm), H4P (9.5 nm), H3P (8.7 nm), H2P (7.8 nm), and H1P (7.4 nm). The identified subparticles were grouped into “large” (L-HDLP = H4P + H5P + H6P + H7P) and “small” (S-HDLP = H1P + H2P + H3P) HDL subclasses for analysis purposes. In the Intermountain Heart cohort, lipids were measured by standard chemical analysis. Lipids and creatinine were measured by standard chemical analysis, and the estimated glomerular filtration rate (eGFR) was calculated using the 2021 CKD-EPI equation. Chronic kidney disease (CKD) was defined as eGFR < 60 ml / min / 1.73m 2 as defined. Study population

[0140] For this study, consecutive patients undergoing cardiac catheterization at Duke University Medical Center for suspected ischemic heart disease, registered in the CATHGEN biorepository between 2001 and 2011 with sufficient available frozen EDTA plasma, were identified (n = 6,969). The final study cohort (n = 5,876) excluded those with missing angiography (n = 889), heart failure (n = 98), creatinine (n = 78), and BMI (n = 28) information. Demographic, medical history, and angiography data were obtained from the Duke Databank for Cardiovascular Disease. Follow-up included determination of mortality (confirmed by the National Death Index and Social Security Death Index) and myocardial infarction (MI). All individuals were followed longitudinally by contact at 6 months after the procedure and annually thereafter. The CATHGEN biorepository was monitored and approved by the Duke University Institutional Review Board on March 18, 2011. Prior to blood sample collection, all study participants provided written informed consent. Incident events were defined as death from any cause or cause-specific death or non-fatal MI at any point during the follow-up period, and time-to-event was defined as from the time of enrollment in cardiac catheterization until death at any point after enrollment. The median (IQR) follow-up time was 6.2 (4.4 - 8.9) years. Cardiovascular death was defined as death due to one of the following causes: MI, heart failure, sudden death, post-resuscitation, vascular causes, during or after cardiac surgery, or during cardiac catheterization. Non-cardiovascular death was defined as death due to non-cardiac medical causes or non-cardiac causes related to the procedure. Unknown cause of death was defined as unobserved or unknown cause of death. Coronary artery disease (CAD) was defined as the presence of at least one epicardial coronary vessel with clinically significant stenosis (≥75%) at the time of the index catheterization.

[0141] The second study group (n = 2,998) was drawn from the Intermountain Heart Catheterization Registry of patients who underwent coronary angiography (09 / 2000–09 / 2006) at LDS Hospital (Salt Lake City, Utah)(2). Consecutive patients were included if they were ≥18 years old, had at least 5 years of follow-up, and had sufficient available frozen EDTA plasma and no missing clinical and laboratory variables. Patients provided informed consent prior to angiography, and the study was approved by the Intermountain Urban Central Region Institutional Review Board on March 23, 2012. The outcome event was death from any cause as determined by hospital records, Utah State Health Department records (death certificates), and the Social Security Administration Death Master File. The time to event was defined as the time from cardiac catheterization registry to death at any point after enrollment. The median (IQR) follow-up time was 8.2 (6.9–9.2) years. Statistical analysis

[0142] Continuous variables can be presented as mean ± SD or median and interquartile range (IQR), and binary variables as percentages (may presented). Chi-square statistics and Student's t-test are used to compare baseline characteristics between those who died and those who did not die during 5 years of follow-up. Spearman correlation coefficient is used to assess the correlation between the selected variables. The associations of all-cause mortality (and, furthermore, cause-specific mortality and non-fatal MI in CATHGEN) with NMR-measured lipoprotein and metabolite variables can be assessed using a Cox proportional hazards model adjusted for age, sex, race, smoking, diabetes, hypertension, BMI, total cholesterol, HDL cholesterol, triglycerides, eGFR, CAD, heart failure, previous MI, and family history of CAD. The assumption of proportional hazards is tested by including covariates that vary over time in the model. Of several NMR measures that were found to have a significant association with all-cause mortality in CATHGEN when tested individually, including plasma proteins (reverse) and ketone bodies, only six (S-HDLP, GlycA, citrate, valine, leucine, isoleucine) made a significant independent contribution to the joint prediction model. These may be combined into sex-specific IVX, MMX, and MVX multimarker scores. Estimation of the relative importance of each predictor can be provided by its chi-square value as a percentage of the total chi-square of the model. The estimation of relative importance can be very similar to that obtained by comparing the discrimination c-index of the full model with that of the model with each variable omitted. CATHGEN and Intermountain Heart statistical analyses can be performed by I.S. using SAS version 9.4 and by H.T.M. using SPSS version 22.0, respectively. All reported P-values can be two-sided. Results

[0143] A study was conducted to assess the relative risk of mortality in two large patient cohorts. Patient-derived samples were collected from 5,876 patients in the CATHGEN cohort and 2,998 patients in the Intermountain Heart cohort over a 5-year period. The samples optimally included blood samples taken by venipuncture. NMR spectroscopy was used to evaluate the samples for the presence and amount of eight health markers referred to as biomarkers. The eight biomarkers included GlycA, at least one high-density lipoprotein particle (HDLP) subclass, valine, leucine and isoleucine, three branched-chain amino acids (BCAAs), at least one ketone body, citrate, and at least one subset of serum proteins. Using spectra from previously obtained NMR instruments, three indices were calculated from these biomarkers and at least six biomarkers including GlycA, citrate, at least one high-density lipoprotein particle (HDLP) subclass and at least one BCCAA (at least on BCCAA): the metabolic malnutrition index (MMX), the inflammation vulnerability index (IVX), and the malnutrition vulnerability index (MVX) as a result of the two previous indices. The biomarkers were assayed over a 5-year period, and thereby the indices were also assayed. The MVX obtained from MMX and IVX over a 5-year period in two large cohorts was followed to assess the risk of mortality in patients with the malnutrition-inflammation syndrome (MICS). Interestingly, of the eight biomarkers, only six analyzed biomarkers (S-HDLP, GlycA, citrate, valine, leucine, isoleucine) made a significant independent contribution to the joint prediction model.

[0144] Of the 5,876 CATHGEN and 2,888 Intermountain Heart patients evaluated, 1,000 (17%) and 441 (15.3%) died within 5 years, respectively. Baseline characteristics of participants stratified by 5-year survival status are presented in Figure 1. In both cohorts, there was a high prevalence of hypertension, diabetes, heart failure and CAD defined by angiography, plus signs of the "risk factor paradox" (lower cholesterol, triglycerides and BMI in decedents). Among NMR-measured metabolites hypothesized to reflect MICS, levels of GlycA and citrate were higher and S-HDLP, valine and leucine were lower in patients who died. Correlations among these biomarkers and selected risk factors were similar in both cohorts (Figure 9). In CATHGEN, six putative MICS-related biomarkers made significant contributions to a multivariate prediction model of mortality, increasing the c-index from 0.690 to 0.761 (Figures 4, 10 and 11). As estimated by the percent contribution to the total chi-square of the model, the strongest base model predictors were age (28%), renal function as assessed by eGFR (27%) and heart failure (16%). Addition of the MMIS biomarkers to the model substantially reduced the predictive contributions of heart failure and eGFR (to 3% each) and replaced them in importance with S-HDLP (25%) and GlycA (17%). The strong protective (inverse) association observed for small HDL particles applied only to particles with diameters of less than approximately 8.8 nm; larger-sized HDL particles and the concentration of total HDL cholesterol (HDL-C) had negligible mortality associations (Figure 14). In the Intermountain Heart replication cohort, the mortality associations of the six MMIS variables were similar to those in CATHGEN, but the major contributions of heart failure and eGFR were lower and attenuated (from 19% to 9% and from 32% to 13%, respectively), and GlycA (12%) and S-HDLP (12%) had relatively reduced importance (Figure 21).Considering the complexity and multifactorial etiology of the metabolic dysfunctions underlying cachexia / sarcopenia / malnutrition and their related mortality risks, it is reasonable to expect utility in generating the following two mortality multimarker “subscores”: an Inflammatory Vulnerability Index (IVX) that combines GlycA and S-HDLP, and a Metabolic Malnutrition Index (MMX) that combines valine, leucine, isoleucine, and citrate (Figure 8). These were calculated sex-specifically to numerically similar scores in men and women. Otherwise, due to sex differences in metabolite levels unrelated to mortality risk, the scores would not have been similar. Specifically, citrate and GlycA levels are higher and BCAA levels are lower in women (Figure 15).

[0145] In CATHGEN, the predictive contribution of IVX (44%) was approximately twice that of MMX (20%) (Figures 4 and 12), while in Intermountain Heart, IVX and MMX made comparable contributions to the mortality risk of 18% and 12%, respectively (Figure 21). In contrast to mortality prediction, the risk of non-fatal MI was much less affected by MMIS-related NMR biomarkers and much more dependent on comorbid conditions, and the combination of prevalent angiographic CAD and a history of previous MI made a dominant contribution (62 - 75%) in CATHGEN (Figures 4 and 10).

[0146] By combining IVX and MMX, an overall mortality risk multimarker named MVX (Metabolic Vulnerability Index) is produced. The calculation of MVX includes a product term (IVX*MMX) that takes into account the observed interaction (synergy) between the inflammatory and malnutrition parts of MICS (Figure 8). The cross-classification graph of 5-year mortality by IVX and MMX tertiles (Figure 5) explains this interaction by showing that a higher mortality risk in those with a high MMX score occurs only when the IVX score is also high. MVX in CATHGEN contributed far more to mortality prediction (69%) than any of the other 15 risk factors in the model including age (17%) (Figures 4 and 13). In Intermountain Heart, MVX was also lower but dominant, and its predictive contribution (31%) exceeded that of all variables except age (35%) (Figure 21).

[0147] There was a deviation from the proportional hazards assumption (p<0.0001) in the models of 5-year mortality for IVX and MVX, but not for MMX. Therefore, a sensitivity analysis was performed comparing models limited to mortality occurring within 1 year (n = 255), within 3 years (n = 651), within 5 years (n = 1000), and between 6 and 10 years (n = 529) of enrollment in CATHGEN (Figures 16 - 17). IVX and MVX were most strongly associated with 1-year mortality (IVX HR: 2.48; 95% CI 2.13 - 2.89; MVX HR: 3.00; 95% CI 2.60 - 3.45) and remained dominant predictors of death occurring after 5 years (IVX HR: 1.57; 95% CI 1.42 - 1.72; MVX HR: 1.53; 95% CI 1.39 - 1.68). In contrast, MMX showed an almost constant mortality association during the first 5 years of follow-up and substantially weakened over a longer period. The contributions of MVX (82%) and age (10%) were dominant over the prediction of death occurring after 5 years (36% and 42% respectively) along with 1-year mortality.

[0148] Figure 2 shows the (risk factor-adjusted) associations of IVX, MMX, and MVX with all-cause mortality over 5 years in the CATHGEN and Intermountain Heart cohorts, tested per quintile and per 1 SD. The significant associations of all three multimarkers in CATHGEN were replicated in Intermountain Heart, but were slightly weaker (CATHGEN: MVX HR: 2.18; 95% CI 2.03–2.34, per 1 SD of 12.7; Intermountain: MVX HR: 1.67; 95% CI 1.50–1.87, per 1 SD of 12.5). Plots of cumulative mortality in subgroups of CATHGEN participants stratified by small differences in baseline MVX scores showed a strikingly stepwise relationship, with clinically significant mortality differences seen not only at high but also at lower values of MVX (Figure 6). Risk stratification by MVX quintiles in the Intermountain Heart cohort was comparable to that observed in CATHGEN (Figure 22).

[0149] Of the 1000 deaths that occurred within 5 years in CATHGEN, 379 (38%) and 507 (51%) were due to cardiovascular and non-cardiovascular causes, respectively, while 114 (11%) were of unknown cause (Figures 18–19). Some risk factors such as heart failure and diabetes predicted cardiovascular death but not non-cardiovascular death, while IVX, MMX, and MVX were similarly associated with mortality risk regardless of cause (MVX HR: 2.02; 95% CI 1.80–2.26 for cardiovascular death; HR: 2.30; 95% CI 2.08–2.53 for non-cardiovascular death).

[0150] Figure 3 and Figure 20 provide the comparative associations of all-cause mortality with MVX in subgroups differing by sex, age, and baseline BMI, hypertension, smoking, diabetes, heart failure, previous MI, CAD, or CKD status. The MVX associations were very similar in men (HR: 2.18; 95% CI 2.03–2.34) and women (HR: 2.22; 95% CI 1.97–2.50), and also in those with and without risk factors that influenced 5-year survival. This was most notably true for approximately doubled mortality in patients with and without existing heart failure and CKD, and also in patients in the lowest (<22) vs highest (≥30) BMI categories. In the two lowest-risk patient subgroups, the youngest (≤50 years) and the most “healthy” (without CAD, heart failure, diabetes, and CKD), the MVX-mortality associations were equally strong (HR per 1 SD: 2.61; 95% CI 2.10–3.24 and 2.67; 95% CI 2.18–3.28, respectively). Conclusion

[0151] The complex and incompletely understood metabolic derangements associated with inflammation and protein-energy wasting are significant contributing factors to the increased mortality risk of elderly patients and those with chronic organ diseases who are stressed by the syndromes of cachexia, sarcopenia, malnutrition, and frailty. However, these wasting syndromes have an uncertain association with cardiovascular patients or lower-risk populations. Research has been hampered by the lack of objective clinical assessment tools for these intertwined metabolic malnutrition and inflammation syndromes. This study sought to determine the mortality risk associated with the metabolic vulnerability index (MVX), a multimarker derived from six simultaneously measured serum biomarkers most plausibly associated with these metabolic abnormal syndromes, in two independent cohorts of cardiac catheterization patients. The results are stimulating in suggesting that survival may depend on an as yet unrecognized etiological factor separate from the diseases or vulnerabilities that are considered to be the "causes" of death. If borne out by future research, treating the metabolic derangements underlying the overlapping syndromes of cachexia, sarcopenia, malnutrition, and frailty with anti-inflammatory, nutritional, or alternative therapies could provide a superior survival benefit compared to targeting traditional disease risk factors.

[0152] Using clinical NMR analysis, we efficiently quantified six metabolites in plasma that were plausibly related to MMIS. The resulting IVX, MMX, and MVX multimarker scores were shown to have a strong, graded association with all-cause mortality in two large cardiac catheterization cohorts. The associations were equally strong for death due to cardiovascular and non-cardiovascular causes, in men and women, in younger and older individuals, and in underweight and overweight patients, and in patients with and without angiographic evidence of CAD or other comorbid conditions such as heart failure, renal dysfunction, diabetes, and hypertension. These observations, along with the surprising finding that the MVX-mortality association was at least as strong in two of the lowest-risk patient subgroups, those aged ≤50 years and phenotypically “healthy” individuals without CAD, heart failure, CKD, or diabetes, suggest that MMIS may have a general relevance to mortality risk by contributing to greater or lesser metabolic vulnerability or resilience.

[0153] Figure 7 conceptualizes this suggestion that there are commonalities in the impact of MICS on survival regardless of the underlying cause of death. On the left, a long-term trajectory is depicted that is influenced by disease-specific risk factors (red) from health to conditions such as atherosclerotic cardiovascular disease (ASCVD). Targeting these is fundamental to the success of preventive strategies to reduce the risk of both lethal and non-lethal outcomes (such as lowering serum cholesterol). This risk-factor-targeted disease prevention strategy is exemplified by the canoe paddler who avoids being washed away by the waterfall using the most appropriate tools (such as a large paddle). On the right (blue), a superimposed influence of MICS in metabolic vulnerability is shown as a dominant consideration in the latter part of the trajectory where the mortality risk is closer. The canoe paddler under these altered circumstances (high MVX score) where survival is the top priority may be most helped by different risk-reduction approaches (such as a helmet) that target the inflammatory and / or nutritional status to promote metabolic resilience.

[0154] This hypothesis was previously proposed to explain the "paradoxical epidemiology" in hemodialysis patients and other vulnerable patient subpopulations, and the failed preventive efforts had focused on traditional risk factors such as obesity, hypertension, and hypercholesterolemia to substantially improve survival. This finding in patient subpopulations not previously associated with MICS suggests the degree of biological universality of the contribution of malnutrition / inflammation to mortality risk. Of particular note is that the predictive contribution of MVX was dominant over the multivariable model of 5-year mortality, exceeding the contribution of age and making the contributions of risk factors such as smoking, diabetes, and heart failure appear small. These latter variables were not unimportant in these patients; together, they accounted for approximately 55% of the mortality prediction in the Cox model without MVX in CATHGEN. Rather, the addition of MVX to the model greatly improved the mortality prediction, so that the 69% predictive contribution of MVX made the contributions of the covariates appear small. In contrast, the risk of nonfatal MI was little affected by the MVX score and its component parts.

[0155] If MMIS plays an important role in the relatively short-term survival of cardiovascular patients, the prominent results would question the customary use in ASCVD clinical trials of composite endpoints that combine fatal and nonfatal events. Previous criticisms of composite endpoints have focused on pitfalls in clinical interpretation and the weighting of "hard" and "soft" outcomes, but did not question the underlying assumption that fatal and nonfatal ASCVD events share a common etiology. Doubtless for them, but if stronger MMIS etiologies overlap and are dominant over relatively short-term survival, this would explain why MVX has a top-priority impact on both cardiovascular and non-cardiovascular mortality, but much less impact on nonfatal MI. Viewed through this lens, the results of past clinical trials in which the intervention affected the fatal and nonfatal components of the composite endpoint unevenly may be worthy of reexamination.

[0156] Given the strongly elevated mortality associated with MVX and its MMX and IVX components, the reasons why MICS largely eluded the clinical radar are unclear. A possible reason is the lack of clinically accessible serum markers specific for the detection of co-existing protein-energy malnutrition / wasting and inflammation. The challenge is the inherent complexity of the overlapping syndromes, with uncertainty regarding which of the intertwined metabolic stresses are cause and which are effect. These include chronic inflammation, endocrine disruption, oxidative stress, enhanced protein catabolism, acidosis, muscle anabolism resistance, impaired mTOR signaling, elevated resting energy expenditure, and endothelial dysfunction. In light of this complexity, it is unsurprising that a multimarker index, which aggregates six simultaneously measured metabolites reflecting different aspects of the syndrome, may have advantages over the conventional clinical markers serum albumin and CRP. Dividing MVX into its IVX and MMX components may serve potential future clinical aims of differentiating and treating MMIS-related mortality risk, depending on the reason(s) underlying the MVX elevation. Considering their apparent synergy, and in particular the uncertainty regarding how branched-chain amino acids and citrate are mechanistically involved, it is worth recognizing that it is arbitrary to ascribe IVX and MMX to "inflammatory" and "metabolic malnutrition" etiologies, respectively.

[0157] The major impact of S-HDLP on IVX and MVX scores is intriguing. Understanding how this small HDL subspecies is involved with MICS can help in determining which aspects of HDL multifunctionality affect longevity and metabolic resilience. The current understanding of the clinical importance (or lack thereof) of HDL is based almost exclusively on studies of HDL cholesterol (HDL-C), which is just one HDL biomarker. Since far more cholesterol is present in the larger HDL particle subpopulations compared to the smaller HDL particle subpopulations, the clinical relevance of HDL-C can be misleading about the protective role(s) played by small HDL particles. This study presents a telling example: adding HDL-C to a multivariable model for mortality in CATHGEN did not increase the c-index at all, while adding S-HDLP instead substantially increased the c-index. The concentration of larger-sized HDL was not associated with mortality at all. This distinct partitioning of HDL bioactivity by particle size is consistent with evidence that certain HDL functional activities, such as antioxidant, anti-inflammatory, and anti-infection, are mediated by proteins and / or lipid species that preferentially or exclusively reside in the smaller-sized HDL particle subpopulations.

[0158] Although it is too early to predict how the MVX and its components, the MMX and IVX indices, can ultimately be used clinically, several possibilities invite future research. The MVX has an obvious mortality prognostic value, but it is not known whether reducing MVX by anti-inflammatory, nutritional, or alternative therapies prolongs survival time. Currently, the most clinically relevant use for MVX could be to complement or expand the "disease burden / inflammatory state" etiological criteria used for the diagnosis and grading of the severity of malnutrition, as this is related to the syndromes of cachexia, sarcopenia, and frailty. In this context, the MVX is not yet needed both for prognosis and to help evaluate the effectiveness and safety of therapeutic interventions, but it can provide a simple, quantitative, and objective measure of metabolic dysfunction(s) affecting survival. Regarding real-world clinical utility, the MVX, IVX, and MMX scores are calculated using data from the same NMR LipoProfile "scan" that is currently being rolled out in the United States for routine patient testing for cardiometabolic risk by co-assessing lipid panels, apolipoprotein B, GlycA, and the LP-IR insulin resistance score. The analysis can produce the MVX index with little or no incremental analytical cost, as it uses no assay-specific reagents or other consumables.

[0159] Overall, the MVX score, which aggregates six NMR-measured biomarkers reflecting different aspects of malnutrition–inflammation syndrome, was superior to the contributions made by standard cardiovascular risk factors and comorbidities to the 5-year mortality prediction model in two large cohorts of cardiac catheterization patients. Although speculative, these results suggest that lethal and non-lethal outcomes can be affected by different etiologies and that improved survival may potentially be achieved using treatments that target one or more components of the MMIS. MVX measured in high- and low-risk subjects for cardiovascular events made a superior contribution to mortality prediction in overall cardiovascular patients and in a relatively low-risk subgroup without existing disease, suggesting that the metabolic malnutrition–inflammation syndrome may play a more universal etiological role in affecting survival than previously thought.

[0160] The foregoing is illustrative of the present disclosure and should not be construed as limiting thereof. Although several exemplary embodiments of the present disclosure have been described, those skilled in the art will readily and fully understand that many modifications are possible in the exemplary embodiments without substantially departing from the novel teachings and advantages of the present disclosure. Accordingly, it is intended that all such modifications be included within the scope of the present disclosure as defined in the claims. In the claims, means-plus-function clauses, when used, are intended to cover not only the structures described herein as performing the recited function but also structural equivalents as well as equivalent structures. Thus, it should be understood that the foregoing is illustrative of the present disclosure and should not be construed as limited to the specific embodiments disclosed, and that modifications to other embodiments along with the disclosed embodiments are intended to be included within the scope of the appended claims.

Claims

**Claim 1** A method for monitoring the risk of metabolic malnutrition and inflammatory syndrome in a subject, comprising: (a) obtaining a sample of blood, serum or plasma from the subject; (b) measuring at least two biomarkers simultaneously from the sample; (c) generating a metabolic vulnerability index (MVX) value from the at least two simultaneously measured biomarkers; (d) determining the relative risk of early death of the subject based at least on the MVX value; (e) repeating steps (a) to (d) at a later time point; and (f) evaluating the relative risk of early death and the MVX value of the subject over time. A method comprising the above steps. **Claim 2** The method according to claim 1, wherein the at least two biomarkers include GlycA, at least one of the high-density lipoprotein particle (HDLP) subclasses, citrate, and two or more of the branched-chain amino acids (BCAAs). **Claim 3** The method according to claim 2, wherein the HDLP subclass is small HDLP (S-HDLP). **Claim 4** The method according to claim 2, wherein the BCAA is at least one of leucine, isoleucine or valine. **Claim 5** The method according to claim 1, wherein the MVX value is gender-specific and includes a gender-specific inflammatory vulnerability index (IVX) and a gender-specific metabolic malnutrition index (MMX). **Claim 6** The method according to claim 1, wherein the MVX value is gender-specific for females and is determined using the following model: MVX Ｆ = (IVX Ｆ * 2.27278) + (lnMMX Ｆ * 12.13511) + (IVX Ｆ * lnMMX Ｆ ) * -1.09312 (description of the model for females) **Claim 7** The method according to claim 1, wherein the MVX value is gender-specific for males and is determined using the following model: MVX Ｍ = (IVX Ｍ * 3.54601) + (lnMMX Ｍ * 14.41428) + (IVX Ｆ * lnMMX Ｍ ) * -1.43438 (description of the model for males) **Claim 8** The method according to claim 5, wherein the measured values of the at least one of the BCAAs and citrate are used to generate the gender-specific metabolic malnutrition index (MMX) value. **Claim 9** The method according to claim 5, wherein the measured values of the at least one HDLP subclass and GlycA are used to generate the gender-specific inflammatory vulnerability index (IVX). **Claim 10** The method according to claim 8, wherein the gender-specific MMX value for females is determined using the following model: MMX Ｆ = ((4 + (Leu * -0.03142) + (Leu ２ * 0.0000893)) * 0.353) + ((7 + (Val * -0.03362) + (Val ２ * 0.0000689)) * 0.684) + (Ileu * 0.00332) + ((1 + (Citr* - 0.0072) + (Citr ２ *0.0000573)) * 0.7135) (description of the model for females' MMX) **Claim 11** The method according to claim 8, wherein the gender-specific MMX value for males is determined using the following model: fff MMX ｍ = ((4 + (Leu * -0.01594) + (Leu ２ * 0.0000291)) * 1.076) + ((7 + (Val * -0.0239) + (Val ２ * 0.00005)) * 0.414) + (Ileu * 0.01265) + ((1 + (Citr * 0.00906) + (Citr ２ * -0.0000126)) * 0.5881) (description of the model for males' MMX) **Claim 12** The method according to claim 9, wherein the sex-specific IVX value for females is determined using the following model: IVX Ｆ = 9 + (GlycA* - 0.000187) + (S-HDLP* - 0.3585) + ((GlycA*S-HDLP)*0.000348) The method according to claim 9, which is determined using the following model:

13. wherein said sex-specific IVX value for males is determined using the following model: IVX Ｍ = 9 + (GlycA* - 0.00437) + (S-HDLP* - 0.52307) + ((GlycA*S-HDLP)*0.000817), the method according to claim 9

14. The method according to claim 1, wherein the MVX value is determined in a subject considered to be sex-specific and at low risk of cardiovascular events or a subject considered to be at high risk of cardiovascular events.

15. The method according to claim 1, wherein the measuring step is performed by nuclear magnetic resonance (NMR) spectroscopy.

16. The metabolic vulnerability index (MVX) can provide risks for short (1 year) to long (12 years) periods of early death assessment, and the step of evaluating the subject over time includes observing the subject at 12 years, 10 years, and / or every 6 years, every 5 years, every 4 years, every 3 years, every 2 years, or every 1 year. The method according to claim 1.

17. An NMR spectrometer configured to simultaneously acquire NMR spectra (singular and / or plural) of two or more biomarkers from a blood, serum, or plasma sample derived from a subject, A processor for determining a metabolic vulnerability index (MVX) value based on the NMR spectra (singular and / or plural) of the two or more biomarkers, the processor including or communicating with a memory A system comprising.

18. The system according to claim 17, wherein the processor is configured to determine a sex-specific inflammatory vulnerability index (IVX) and a sex-specific metabolic malnutrition index (MMX).

19. The system according to claim 17, wherein the NMR spectra (singular and / or plural) of the at least two biomarkers include two or more of a signal for a GlycA biomarker, at least one signal for at least one high-density lipoprotein particle (HDLP) subclass biomarker, at least one signal for at least one branched-chain amino acid (BCAA) biomarker, and at least one signal for a citrate biomarker.

20. The system according to claim 19, wherein the at least one HDLP class is a small HDLP (S-HDLP) class, and the BCAA is at least one of leucine, isoleucine, or valine.

Images