Biomarkers for identifying cancer patients at high risk for chemotherapy-induced cardiotoxicity

Abstract

Biomarkers associated with being at risk of chemotherapy-induced cardiotoxicity are described. The identified biomarkers include biglycan (BGN), carbonic anhydrase 6 (CA6), CD109, thrombospondin-4 (TSP4), cadherin-5 (CDH5) and cystatin-F (CST7). Methods of identifying subjects diagnosed with cancer as being at risk for chemotherapy-induced cardiotoxicity, such as toxicity induced by an anthracycline, by measuring the level of one or more of the disclosed biomarkers, are also described. Kits for detecting the biomarkers associated with chemotherapy-induced cardiotoxicity are also described.

Description

[0001] BIOMARKERS FOR IDENTIFYING CANCER PATIENTS AT HIGH RISK FOR CHEMOTHERAPY-INDUCED CARDIOTOXICITY

[0002] CROSS REFERENCE TO RELATED APPLICATIONS

[0003] This application claims the benefit of U.S. Provisional Application No. 63 / 567,661, filed March 20, 2024, which is herein incorporated by reference in its entirety.

[0004] FIELD

[0005] This disclosure concerns methods and kits for identifying subjects at risk of chemotherapy- induced cardio toxicity by measuring the level of select biomarkers in the subject, and methods of treating subjects identified as being at risk of chemotherapy-induced cardiotoxicity with one or more cardioprotective agents (for example prior to or during treatment with the chemotherapy).

[0006] BACKGROUND

[0007] Anthracycline (AC)-based chemotherapy (e.g., doxorubicin, DOX) is one of the most effective and commonly used treatments for a wide range of cancers; however, the most significant adverse side effects of the treatment is cumulative dose-dependent cardiotoxicity, which may manifest as subclinical heart disease and asymptomatic decline in left ventricular ejection fraction (LVEF) to more serious functional decreases as left ventricular dysfunction (LVD) and congestive heart failure (CHF) (Singal, N Engl J Med 1998, 339 (13), 900-905; Floyd et al., J Clin Oncol 2005, 23 (30), 7685- 7696; Yeh, Anna Rev Med 2006, 57, 485-498; Todorova et al., Am J Cancer Res 2021, 11 (9), 4070- 4091). The timeline for development of AC-associated cardiotoxicity can range from days to years, and the early stage(s) of cardiotoxicity can be asymptomatic (Floyd et al., J Clin Oncol 2005, 23 (30), 7685-7696). Life-threatening cardiotoxicity significantly limits the clinical use of this class of chemotherapeutics .

[0008] Left ventricular dysfunction (LVD) is the most notable and frequent clinical manifestation of AC-induced cardiotoxicity. Imaging tests (e.g., echocardiography and multi-gated acquisition (MUGA) scan) are the most common monitoring tools for assessment of LVEF; however, imaging tools are costly and LVEF has not been qualified as a biomarker for early detection of cardiotoxicity. Cardiac troponin T (cTnT) and cardiac troponin I (cTnl) are sensitive biomarkers of cardiac tissue damage that have been suggested as alternatives to imaging methods; however, their ability to predict cardio toxicity is limited (Herman et al., Cancer Res 1998, 58 (2), 195-197; Monsuez, Arch Cardiovasc Dis 2012, 105 (11), 593-604) because cardiac troponins are released after tissue damage has already occurred and their levels in the blood are often transient.

[0009] Ideally, cancer patients are identified early during or even before initiation of chemotherapy so that treatment options are managed, or early intervention is introduced to mitigate LVD. However, the identification of early predictive biomarkers of cardiotoxicity that improve patient monitoring and facilitate early intervention is an unmet need.

[0010] SUMMARY

[0011] Disclosed herein are biomarkers associated with a cancer subject’s risk of chemotherapy- induced cardiotoxicity. Five biomarkers, biglycan (BGN), carbonic anhydrase 6 (CA6), CD109, thrombospondin-4 (TSP4), and cadherin-5 (CDH5), were identified as increased in subjects at risk for chemotherapy-induced cytotoxicity, while one biomarker, cystatin-F (CST7), was decreased in subjects at risk for chemotherapy-induced cytotoxicity. The disclosed biomarkers can be used, for example, to identify subjects who are at risk of chemotherapy-induced cytotoxicity, prior to treatment, in order to guide treatment decisions. In some aspects, the disclosed methods have a specificity of at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, or at least 90%. In some aspects, the disclosed methods have a sensitivity of at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, or at least 90%. For example, alternative anti-cancer drugs can be used for treatment, the subject can be treated with one or more cardioprotective agents (either before or concurrently with chemotherapy) to minimize cardiotoxicity, or both. Kits for detecting the biomarkers associated with chemotherapy-induced cardiotoxicity are also described.

[0012] Provided herein is a method of identifying a subject diagnosed with cancer as being at risk of chemotherapy-induced cardiotoxicity. The method includes measuring the level of one or more of (such as two, three or four of) BGN protein or nucleic acid, CA6 protein or nucleic acid, CD109 protein or nucleic acid, and TSP4 protein or nucleic acid in a biological sample obtained from the subject; comparing the level of one or more of BGN protein or nucleic acid, CA6 protein or nucleic acid, CD 109 protein or nucleic acid, and TSP4 protein or nucleic acid to a control (such as a sample(s) or reference value for a subject not at risk for chemotherapy-induced cardiotoxicity); and identifying the subject as being at risk of chemotherapy-induced cardiotoxicity if there is an increase in the level of BGN protein or nucleic acid, an increase in the level of CA6 protein or nucleic acid, an increase in the level of CD109 protein or nucleic acid and / or an increase in the level of TSP4 protein or nucleic acid compared to the control.

[0013] In some aspects, the method further includes measuring the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid; comparing the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid to a control; and identifying the subject as being at risk of chemotherapy-induced cardiotoxicity if there is an increase in the level of CDH5 protein or nucleic acid and / or a decrease in the level of CST7 protein or nucleic acid compared to the control (such as a sample(s) or reference value for a subject not at risk for chemotherapy-induced cardiotoxicity).

[0014] In some aspects, the method further includes measuring left ventricular ejection fraction (LVEF) in the subject, such as prior to or during cancer treatment. In some aspects, the chemotherapy is an anthracycline, such as, but not limited to, doxorubicin (DOX) or one of its liposomal forms (e.g., Doxil, Lipodox, or Lipodox 50).

[0015] In some aspects, when the subject is identified as being at risk of chemotherapy-induced cytotoxicity, the subject is treated with an anti-cancer treatment that is not a chemotherapeutic agent and / or the subject is treated with a cardioprotective agent. In some examples in which the at-risk subject is administered chemotherapy, the subject is treated with the cardioprotective agent prior to receiving the chemotherapy, concurrently with chemotherapy, and / or after treatment with chemotherapy.

[0016] In some aspects, when the subject is not identified as being at risk of chemotherapy-induced cytotoxicity, the subject is treated with chemotherapy, such as an anthracycline (e.g., DOX). In some examples, the subject is further administered one or more additional anti-cancer therapies.

[0017] Also provided is a method that includes obtaining a biological sample from a subject diagnosed with cancer: measuring the level of BGN protein or nucleic acid, CA6 protein or nucleic acid, CD109 protein or nucleic acid and / or TSP4 protein or nucleic acid in the biological sample; and comparing the level of BGN protein or nucleic acid, CA6 protein or nucleic acid, CD 109 protein or nucleic acid and TSP4 protein or nucleic acid to a control. In some aspects, the method further includes measuring the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid in the biological sample; and comparing the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid to a control.

[0018] Further provided are kits for detecting the biomarkers associated with chemotherapy-induced cytotoxicity. In some aspects, the kit includes a plurality of antibodies, wherein the plurality of antibodies includes at least two antibodies (or at least three, at least four, at least five or six antibodies) selected from the group consisting of an antibody specific for BGN, an antibody specific for CA6, an antibody specific for CD109, an antibody specific for TSP4, an antibody specific for CDH5 and an antibody specific for CST7, and further including no more than five additional antibodies specific for other proteins. In other aspects, the kit includes a plurality of aptamers including at least two aptamers (or at least three, at least four, at least five, or six aptamers) selected from the group consisting of an aptamer specific for BGN, an aptamer specific for CA6, an aptamer specific for CD 109, an aptamer specific for TSP4, an aptamer specific for CDH5 and an aptamer specific for CST7, and further including no more than five additional aptamers specific for other proteins. In other aspects, the kit includes a plurality of antibody pairs including at least two antibody pairs (or at least three, at least four, at least five, or six antibody pairs) selected from the group consisting of a pair of antibodies specific for BGN, a pair of antibodies specific for CA6, a pair of antibodies specific for CD109, a pair of antibodies specific for TSP4, a pair of antibodies specific for CDH5, and a pair of antibodies specific for CST7, and further including no more than five additional antibody pairs specific for other proteins. In some examples, the kits include a combination of an antibody and an aptamer specific for a marker (e.g., an antibody specific for BGN and an aptamer specific for BGN is an antibody / aptamer pair specific for BGN). In some examples, each pair of antibodies (or each antibody / aptamer pair) is linked to unique DNA sequences that hybridize only to each other, or one or both antibodies (or one or both the antibody and aptamer) are linked to an affinity tag (such as biotin) or a chromophore (such as a fluorophore).

[0019] In some examples of the kits disclosed herein, the kits further include recombinant BGN protein, recombinant CA6 protein, recombinant CD109 protein, recombinant TSP4 protein, recombinant CDH5 protein, recombinant CST7 protein, one or more control proteins, buffer! s), reagent(s), multi-well plate(s), tube(s), mixing bottles(s), a qPCR template, primers, nucleic acid probes, DNA polymerase, dNTPs, loading control dye(s), detection reagent(s), detection enzyme(s), chromophore-linked avidin or streptavidin or other affinity tags, substrate(s), adhesive plate sealer(s), DNA microarray slide(s), calibrator(s), quality control sample(s), sample matrix, next-generation sequencing (NGS) adapters, indexing primers, ligase, dA-tailing enzyme(s), DNA cleanup beads, instructions, labelled secondary antibodies, or any combination thereof.

[0020] The foregoing and other features of this disclosure will become more apparent from the following detailed description of several aspects which proceeds with reference to the accompanying figures.

[0021] BRIEF DESCRIPTION OF THE DRAWINGS

[0022] FIGS. 1A-1C: Study design showing inclusion and exclusion criteria (FIG. 1A), DOX treatment, blood sample collection, cardiac function monitoring for breast cancer patients (FIG. IB), and approaches for biomarker discovery using SOMASCAN assays and validation using OLINK assays (FIG. 1C). LVEF, left ventricular ejection fraction.

[0023] FIGS. 2A-2B: Differential baseline protein levels in plasma samples from breast cancer patients who experienced cardiotoxicity versus those who maintained normal cardiac functions after completion of doxorubicin-based chemotherapy in the discovery (FIG. 2A) and validation (FIG. 2B) cohorts. Samples were collected prior to initiation of the therapy. The volcano plots visualize the relationship between fold changes and statistical significance. The vertical lines correspond to 1.2- fold changes in up and down expression while the horizontal line represents a p- value of 0.05. The dots below the horizontal dotted line (1) represent proteins with no statistically significant differences (p > 0.05); the dots in the upper right corner (2) represent upregulated proteins with fold change > 1.2 and p < 0.05; the dots in the upper left corner (3) represent downregulated proteins with fold change >1.2 and p < 0.05; and the dots in the upper center area (4) represent proteins with fold change <1.2 and p < 0.05.

[0024] FIGS. 3A-3B: Violin plots showing baseline protein abundance differences in plasma samples from breast cancer patients who experienced cardiotoxicity (Cardiotox) versus those who maintained normal cardiac functions (Normal) after completion of doxorubicin-based chemotherapy. The plots demonstrate consistent differential protein levels between the discovery (upper panel) and validation (lower panel) cohorts for each protein. Samples were collected prior to initiation of the therapy. The protein abundance values (intensity) were log-transformed. The bars within each violin plot represent SD and the dot refers to the mean of protein abundance measurements.

[0025] FIG. 4: Positive correlation between the SOMASCAN and OLINK measured protein levels for each of biglycan (BGN), cadherin-5 (CDH5), carbonic anhydrase 6 (CA6), CD109, cystatin-F (CST7), and thrombospondin-4 (TSP4). Twenty-eight plasma samples were analyzed by both SOMASCAN and OLINK assays for analytical validation. The SOMASCAN and OLINK values were log-transformed.

[0026] FIG. 5: Correlation between SOMASCAN and OLINK measured protein levels for HER4. Twenty-eight plasma samples were analyzed by both SOMASCAN and OLINK assays for analytical validation. The SOMASCAN and OLINK values were log-transformed.

[0027] FIGS. 6A-6B: Confirmation of rabbit anti-human CA6 (FIG. 6A) and CDH5 (FIG. 6B) antibody coupling with magnetic beads. Each antibody was coupled to 2.5 million beads in two different amounts (5 pg and 12.5 pg). Antibody coupling was confirmed using a concentration series of anti-rabbit IgG-PE to detect fluorescence response of the beads.

[0028] FIG. 7: Graph showing human CDH5 concentration measured using antibody-coupled magnetic beads. CDH5 was detected in the range of 0.1-2000 ng / mL. LogCon, log concentration of CDH5.

[0029] DETAILED DESCRIPTION

[0030] I. Introduction

[0031] Anthracycline doxorubicin (DOX) treatment is associated with cumulative dose-dependent cardiotoxicity in a subset of cancer patients. However, there are currently no clinically-validated biomarkers to identify patients at greatest risk for cardiotoxicity associated with DOX-based chemotherapy. In the study disclosed herein, 83 breast cancer patients were treated with a combination of DOX (60 mg / m2) and cyclophosphamide (600 mg / m2) per cycle for 4 treatment cycles (FIG. IB). Cardiac function was assessed by a multi -gated acquisition (MUGA) scan before the initiation of treatment and at completion of chemotherapy. Blood samples were collected before initiation of treatment and processed into plasma for biomarker analysis. Thirty-nine patients, of which 9 experienced treatment-related cardiotoxicity (>10% absolute unit reduction in left ventricular ejection fraction), were randomly selected for biomarker discovery by the SOMASCAN assays. The remaining 44 patients, of which 10 experienced treatment-related cardiotoxicity, were assigned to the biomarker validation cohort. Forty-eight proteins had differential baseline levels prior to treatment in patients who experienced treatment-related cardiotoxicity, as compared to patients without cardiotoxicity in the discovery cohort. OLINK proteomic analysis of the verification cohort confirmed six proteins were associated with an increased likelihood of cardiotoxicity, including higher levels of biglycan (BGN), carbonic anhydrase 6 (CA6), cadherin-5 (CDH5), CD109, and thrombospondin-4 (TSP4), as well as lower levels of cystatin-F (CST7) in the patients with cardiotoxicity, as compared to those without cardiotoxicity. A predictive model using partial least squares discriminant analysis for these proteins and in combination with baseline LVEF, predicted cardiotoxicity with a sensitivity of 80%, a specificity of 88%, and an overall accuracy of prediction of 86%. These biomarkers and the model provide new tools for the prediction of cancer patients at higher risk for chemotherapy-induced cardio toxicity.

[0032] II. Abbreviations

[0033] AC anthracycline

[0034] AUC area under the curve

[0035] BGN biglycan

[0036] BMI body mass index

[0037] BNP brain natriuretic peptide

[0038] CA6 carbonic anhydrase 6

[0039] CDH5 cadherin-5

[0040] CHF congestive heart failure

[0041] CI confidence interval

[0042] CST7 cystatin-F cTnl cardiac troponin I cTnT cardiac troponin T

[0043] DOX doxorubicin dPCR digital polymerase chain reaction

[0044] ER estrogen receptor

[0045] HER2 receptor tyrosine-protein kinase erbB-2

[0046] HER4 receptor tyrosine-protein kinase erbB-4

[0047] LVD left ventricular dysfunction

[0048] LVEF left ventricular ejection fraction

[0049] MUGA multi-gated acquisition

[0050] NGS next generation sequencing

[0051] NPV negative predictive value

[0052] NPX Normalized Protein expression

[0053] NT-proBNP N-terminal proBNP OR odds ratio

[0054] PPV positive predictive value

[0055] PR progesterone receptor

[0056] RFU relative fluorescence unit SD standard deviation

[0057] SOMAMER slow off-rate modified aptamer TSP4 thrombospondin-4

[0058] III. Summary of Terms

[0059] Unless otherwise noted, technical terms are used according to conventional usage. Definitions of many common terms in molecular biology may be found in Krebs et al. (eds.), Lewin’s genes XII, published by Jones & Bartlett Learning, 2017. As used herein, the singular forms “a,” “an,” and “the,” refer to both the singular as well as plural, unless the context clearly indicates otherwise. For example, the term “a biomarker” includes singular or plural biomarkers and can be considered equivalent to the phrase “at least one biomarker.” As used herein, the term “comprises” means “includes.” It is further to be understood that any and all base sizes or amino acid sizes, and all molecular weight or molecular mass values, given for nucleic acids or polypeptides are approximate, and are provided for descriptive purposes, unless otherwise indicated. Although many methods and materials similar or equivalent to those described herein can be used, particular suitable methods and materials are described herein. In case of conflict, the present specification, including explanations of terms, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting. To facilitate review of the various aspects, the following explanations of terms are provided:

[0060] Administration: To provide or give a subject an agent, such as a therapeutic agent (e.g., an anti-cancer agent, such as chemotherapy), by any effective route. Exemplary routes of administration include, but are not limited to, injection (such as subcutaneous, intramuscular, intradermal, intraperitoneal, intravenous, intratumoral, or renal vein injection), oral, intraductal, sublingual, rectal, transdermal, intranasal, vaginal and inhalation routes.

[0061] Anthracy cline: A class of anti-cancer agents extracted from Streptomyces species. Anthracyclines induce cell death by intercalating in DNA, interfering with DNA metabolism, and inhibiting the ability of topoisomerase-II to repair DNA with double-strand breaks. In some aspects, the anthracycline is doxorubicin, daunorubicin, epirubicin, idarubicin, mitoxantrone or valrubicin.

[0062] Antibody: A polypeptide ligand comprising at least one variable region that recognizes and binds (such as specifically recognizes and specifically binds) an epitope of an antigen, such as a BGN, CA6, CD109, TSP4, CDH5, or CST7 protein. Mammalian immunoglobulin molecules are composed of a heavy (H) chain and a light (L) chain, each of which has a variable region, termed the variable heavy (VH) region and the variable light (VL) region, respectively. Together, the VH region and the VL region are responsible for binding the antigen recognized by the antibody. There are five main heavy chain classes (or isotypes) of mammalian immunoglobulin, which determine the functional activity of an antibody molecule: IgM, IgD, IgG, IgA and IgE. Antibody isotypes not found in mammals include IgX, IgY, IgW and IgNAR. IgY is the primary antibody produced by birds and reptiles, and is functionally similar to mammalian IgG and IgE. IgW and IgNAR antibodies are produced by cartilaginous fish, while IgX antibodies are found in amphibians.

[0063] A “single-domain antibody” refers to an antibody having a single domain (a variable domain) that is capable of specifically binding an antigen, or an epitope of an antigen, in the absence of an additional antibody domain. Single-domain antibodies include, for example, VH domain antibodies, VNAR antibodies, camelid VHH antibodies, and VL domain antibodies. VN R antibodies are produced by cartilaginous fish, such as nurse sharks, wobbegong sharks, spiny dogfish and bamboo sharks.

[0064] A “monoclonal antibody” is an antibody produced by a single clone of lymphocytes or by a cell into which the coding sequence of a single antibody has been transfected.

[0065] Antibody array: An arrangement of antibodies in assigned locations on a matrix. In some examples, the antibodies are attached covalently to the array.

[0066] Array: An arrangement of molecules, such as biological macromolecules (such as peptides, antibodies, or nucleic acid molecules) or biological samples (such as tissue sections), in addressable locations on or in a substrate. A “microarray” is an array that is miniaturized so as to require or be aided by microscopic examination for evaluation or analysis.

[0067] The array of molecules (“features”) makes it possible to carry out a large number of analyses on a sample at one time. Tn certain aspects, one or more molecules (such as an aptamer or antibody) will occur on the array a plurality of times (such as twice), for instance to provide internal controls. The number of addressable locations on the array can vary, for example from at least 4, to at least 6, at least 8, at least 10, at least 12, at least 16, at least 20, at least 30, at least 50, at least 100, or more. In some examples, an array includes 4 to 20 addressable locations. In particular examples, an array consists essentially of aptamers or antibodies (such as those that permit amplification or detection) specific for at least 2, at least 3, at least 4, at least 5 or six of BGN, CA6, CD109, TSP4, CDH5, and CST7, and in some examples, also 1 to 10, 1 to 8, 1 to 6, or 1 to 4 control molecules (such as housekeeping genes / proteins).

[0068] Within an array, each arrayed sample is addressable, in that its location can be reliably and consistently determined within at least two dimensions of the array. The feature application location on an array can assume different shapes. For example, the array can be regular (such as arranged in uniform rows and columns) or irregular. Thus, in ordered arrays the location of each sample is assigned to the sample at the time when it is applied to the array, and a key may be provided in order to correlate each location with the appropriate target or feature position. Often, ordered arrays are arranged in a symmetrical grid pattern, but samples could be arranged in other patterns (such as in radially distributed lines, spiral lines, or ordered clusters). Addressable arrays usually are computer readable, in that a computer can be programmed to correlate a particular address on the array with information about the sample at that position (such as hybridization or binding data, including for instance signal intensity). In some examples of computer readable formats, the individual features in the array are arranged regularly, for instance in a Cartesian grid pattern, which can be correlated to address information by a computer.

[0069] Protein-based arrays include probe molecules, such as antibodies or aptamers. In some examples, an array contains antibodies or aptamers to at least two, at least three, at least four, at least five or six different proteins associated with chemotherapy-induced cardiotoxicity (e.g., selected from BGN, CA6, CD109, TSP4, CDH5, and CST7) and in some examples also includes 1 to 10, 1 to 8, 1 to 6, or 1 to 4 control molecules (e.g., antibodies or aptamers specific for a housekeeping gene / protein). In some examples, the antibodies or aptamers are covalently attached to the array.

[0070] Biglycan (BGN): A member of the small leucine-rich proteoglycan family of proteins. The BGN protein plays a role in bone growth, muscle development and regeneration, and collagen fibril assembly. BGN is also known as PGI, MRLS, DSPG1, PG-S1, SEMDX and SLRR1A. Genomic, mRNA and protein sequences for human BGN are publicly available, such as under NCBI Gene ID 633.

[0071] Biological sample: A biological specimen containing genomic DNA, RNA (including mRNA), protein, or combinations thereof, obtained from a subject. Examples include, but are not limited to, blood, plasma, serum, urine, saliva, cerebral spinal fluid, bronchoalveolar lavage, tissue biopsy, fine needle aspirate, punch biopsy surgical specimen, and autopsy material.

[0072] Biomarker: A molecule, such as a protein or nucleic acid, indicative of the presence of a particular disease state, or susceptibility to particular disease or treatment side effects (such as chemotherapy-induced cytotoxicity).

[0073] Cadherin-5 (CDH5): A protein of the cadherin superfamily. CDH5 is a calcium-dependent cell-cell adhesion molecule comprised of five extracellular cadherin repeats, a transmembrane region and a highly conserved cytoplasmic tail. This protein plays a role in endothelial adherens junction assembly and maintenance. CDH5 is also known as 7B4 and CD144. Genomic, mRNA and protein sequences for human CDH5 are publicly available, such as under NCBI Gene ID 1003.

[0074] Carbonic anhydrase 6 (CA6): A protein found in salivary glands and saliva and thought to play a role in the reversible hydration of carbon dioxide. CA6 is also known as CA-VI and GUSTIN. Genomic, mRNA and protein sequences for human CA6 are publicly available, such as under NCBI Gene ID 765.

[0075] Cardioprotective agent: Drugs or compounds that provide protection against cardiovascular damage, such as damage induced by chemotherapeutic drugs (e.g., chemotherapy -induced cardiotoxicity). In some aspects, the cardioprotective agent is a beta blocker, an angiotensin receptor blocker, an angiotensin-converting-enzyme inhibitor (ACEI), a statin, an angiotensin receptor neprilysin inhibitor, a diuretic, an iron chelator, or a hyperpolarization-activated cyclic nucleotide- gated (HCN) channel blocker. Non-limiting examples of cardioprotective against are listed in the table below.

[0076] CD109: A glycosyl phosphatidylinositol (GPI)-linked glycoprotein that localizes to the surface of platelets, activated T cells, and endothelial cells. The CD109 protein binds to and negatively regulates signaling by transforming growth factor beta. CD109 is also known as pl80, rl50 and CPAMD7. Genomic, mRNA and protein sequences for human CD109 are publicly available, such as under NCBI Gene ID 135228.

[0077] Chemotherapeutic agent: Any chemical agent with therapeutic usefulness in the treatment of diseases characterized by abnormal cell growth. Such diseases include tumors, neoplasms, and cancer as well as diseases characterized by hyperplastic growth. A skilled person can readily identify a chemotherapeutic agent of use for treating cancer (see for example, Slapak and Kufe, Principles of Cancer Therapy, Chapter 86 in Harrison's Principles of Internal Medicine, 14th edition; Perry et al., Chemotherapy, Ch. 17 in Abeloff, Clinical Oncology 2nded., © 2000 Churchill Livingstone, Inc; Baltzer, L., Berkery, R. (edsf Oncology Pocket Guide to Chemotherapy, 2nd ed. St. Louis, Mosby- Year Book, 1995; Fischer, D.S., Knobf, M.F., Durivage, H.J. (eds): The Cancer Chemotherapy Handbook, 4th ed. St. Louis, Mosby- Year Book, 1993). In some aspects, the chemotherapeutic agent includes an anthracycline, such as doxorubicin (DOX) or a liposomal form of DOX (e.g., Doxil, Lipodox, or Lipodox 50). In other aspects, the chemotherapeutic agent is not an anthracycline.

[0078] Chemotherapy-induced cardiotoxicity: Damage to cardiac cells, cardiac tissues, or the heart as a result of treatment with a chemotherapeutic agent.

[0079] Control: A sample or standard used for comparison with an experimental sample. In some aspects, the control is a sample obtained from a healthy subject or from a patient who is not at risk of chemotherapy-induced cytotoxicity. In some aspects, the control is a historical control or standard reference value or range of values (e.g., a previously tested control sample, or an average or median value obtained from a group of patients who are not at risk of chemotherapy-induced cytotoxicity) or a group of healthy subjects. In other aspects, the control is a reference value based on the baseline blood level of a group of healthy subjects or breast cancer patients, such as patients who have not yet been treated with chemotherapy.

[0080] Cystatin-F (CST7): A member of the cystatin superfamily. The CST7 protein is a glycosylated cysteine inhibitor with a putative role in immune regulation through the inhibition of a unique target in the hematopoietic system. CST7 is also known as CMAP. Genomic, mRNA and protein sequences for human CST7 are publicly available, such as under NCBI Gene ID 8530.

[0081] Decrease: A reduction in the quantity, quality, or strength of something. In some aspects herein, the level of CST7 protein or nucleic acid is decreased in a subject at risk of chemotherapy- induced cytotoxicity compared to the level of CST7 protein or nucleic acid in a subject who is not at risk of chemotherapy-induced cytotoxicity. Alternatively, in some aspects, the level of CST7 protein or nucleic acid is decreased compared to a reference value (such as a reference value of a particular patient population) or a historical control value.

[0082] Immunomodulator: An agent that stimulates or suppresses the immune system. Immunosuppressive agents can be used to reduce immune responses against foreign antigens, transplanted tissue / organs and to treat some types of autoimmune disease. Exemplary immunosuppressive agents include, for example, cyclosporine A, tacrolimus, sirolimus, prednisone, dexamethasone, azathioprine, cyclophosphamide, and certain types of monoclonal antibodies. Conversely, immunostimulatory agents enhance the immune system, such as for promoting immune responses against infectious agents and tumors. Exemplary immunostimulatory agents include, but are not limited to, BCG, LPS, recombinant cytokines (e.g., IL-2, IL-1, IL-12 and IFN-y), and antigenspecific antibodies (such as tumor-specific antibodies, such as 3F8, Abagovomab, Adecatumumab, Afutuzumab, Alacizumab , Alemtuzumab, Altumomab pentetate, Anatumomab mafenatox, Apolizumab, Arcitumomab, basiliximab, Bavituximab, Bectumomab, Belimumab, Besilesomab, Bevacizumab, Bivatuzumab mertansine, Blinatumomab, Brentuximab vedotin, Cantuzumab mertansine, Capromab pendetide, Catumaxomab, CC49, Cetuximab, Citatuzumab bogatox, Cixutumumab, Clivatuzumab tetraxetan, Conatumumab, Dacetuzumab, Detumomab, Ecromeximab, Eculizumab, Edrecolomab, Epratuzumab, Ertumaxomab, Etaracizumab, Farletuzumab, Figitumumab, Galiximab, Gemtuzumab ozogamicin, Girentuximab, Glembatumumab vedotin, Ibritumomab tiuxetan, Igovomab, Imciromab, Intetumumab, Inotuzumab ozogamicin, Ipilimumab, Iratumumab, Labetuzumab, Lexatumumab, Lintuzumab, Lorvotuzumab mertansine, Lucatumumab, Lumiliximab, Mapatumumab, Matuzumab, Mepolizumab, Metelimumab, Milatuzumab, Mitumomab, Morolimumab, Nacolomab tafenatox, Naptumomab estafenatox, Necitumumab, Nimotuzumab, Nofetumomab merpentan, Ofatumumab, Olaratumab, Oportuzumab monatox, Oregovomab, Panitumumab, Pemtumomab, Pertuzumab, Pintumomab, Pritumumab, Ramucirumab, Rilotumumab, Rituximab, Robatumumab, Satumomab pendetide, Sibrotuzumab, Sonepcizumab, Tacatuzumab tetraxetan, Taplitumomab paptox, Tenatumomab, TGN1412, Ticilimumab (tremelimumab), Tigatuzumab, TNX-650, Trastuzumab, Tremelimumab, Tucotuzumab celmoleukin, Veltuzumab, Volociximab, Votumumab, and Zalutumumab).

[0083] Increase: An expansion in the quantity, quality, or strength of something. In some aspects herein, the level of BGN, CA6, CD109, TSP4, or CDH5 protein or nucleic acid is increased in a subject at risk of chemotherapy-induced cytotoxicity compared to the level of BGN, CA6, CD109, TSP4, or CDH5 protein or nucleic acid in a subject who is not at risk of chemotherapy-induced cytotoxicity. Alternatively, in some aspects, the level of BGN, CA6, CD 109, TSP4, or CDH5 protein or nucleic acid is increased compared to a reference value (such as a reference value of a particular patient population) or a historical control value.

[0084] Left ventricular ejection fraction (LVEF): A measure of how much blood is pumped out of the left ventricle of the heart after each contraction. LVEF, which is expressed as the percentage of blood pumped out of the ventricle after a contraction, is calculated by dividing the amount of blood pumped out of the ventricle with each contraction by the end-diastolic volume (the total amount of blood in the ventricle) x 100. Normal LVEF for healthy adults ranges from about 52% to about 74%.

[0085] Measuring the level: As used herein, measuring or determining the level of a particular protein or nucleic acid (such as a BGN, CA6, CD 109, TSP4, CDH5 or CST7 protein or nucleic acid) refers to quantifying the amount of the protein or nucleic acid (such as mRNA) present in a sample (such as a blood or serum sample). Quantification can be either numerical or relative. Measuring the level of a protein can be achieved using any known method, such as by immunoassay (<?.g., ELISA), antibody array, aptamers that specifically bind each protein (e.g., SOMASCAN assay), or proximity extension assay (e.g., OLINK assay). Measuring the level of a nucleic acid can be accomplished using any known method, such as by quantitative PCR (qPCR), digital PCR (dPCR), next generation sequencing (NGS), Northern blotting, nuclease protection assay, RNA-seq, or in situ hybridization.

[0086] Solid support: Any solid surface to which biological molecules (e.g., oligonucleotides, antibodies, or other proteins) can be adhered. In some aspects, the solid support includes a slide (e.g., a glass slide or a plastic slide), a microtiter / multi-well plate, a test tube, a dipstick, a membrane (such as a polypropylene membrane), a filter, or a bead (such as a magnetic or glass bead).

[0087] Subject: Living multi-cellular vertebrate organisms, a category that includes human and nonhuman mammals. In some aspects, the subject has been diagnosed with cancer, such as but not limited to, breast cancer.

[0088] Thrombospondin-4 (TSP4): A member of the thrombospondin protein family, which are adhesive glycoproteins that mediate cell-to-cell and cell-to-matrix interactions. The TSP4 protein forms a pentamer and can bind to heparin and calcium. This protein is involved in local signaling in the developing and adult nervous system, and may also play a role in inflammatory responses in Alzheimer's disease. TSP4 is also known as THBS4. Genomic, mRNA and protein sequences for human TSP4 are publicly available, such as under NCBI Gene ID 7060. IV. Biomarkers Associated with Chemotherapy-Induced Cytotoxicity

[0089] The present disclosure describes the identification of biomarkers associated with a cancer patient’ s risk of chemotherapy-induced cardiotoxicity, such cardiotoxicity caused by an anthracycline (e.g., DOX). Five biomarkers, biglycan (BGN), carbonic anhydrase 6 (CA6), CD109, thrombospondin-4 (TSP4), and cadherin-5 (CDH5), were identified as increased in subjects at risk for chemotherapy-induced cytotoxicity, while a sixth biomarker, cystatin-F (CST7), was identified as decreased in subjects at risk for chemotherapy-induced cytotoxicity. The disclosed biomarkers can be used, for example, to identify subjects, prior to treatment, who are at risk of chemotherapy-induced cytotoxicity, such as to guide treatment decisions. For example, alternative anti-cancer agents (e.g., tyrosine kinase inhibitors, angiogenesis inhibitors, immunomodulators, immunotherapy, etc.) can be used for treatment instead of chemotherapy and / or the subject can be treated with one or more cardioprotective agents before, during and / or after treatment to minimize cardiac cytotoxicity. Kits for detecting the biomarkers associated with chemotherapy-induced cardiotoxicity are also described herein.

[0090] Provided herein is a method of identifying a subject diagnosed with cancer as being at risk of chemotherapy-induced cardiotoxicity. The method includes measuring the level of one or more of (such as two, three or four of) BGN protein or nucleic acid, CA6 protein or nucleic acid, CD109 protein or nucleic acid, and TSP4 protein or nucleic acid in a biological sample obtained from the subject; comparing the level of the one or more of BGN protein or nucleic acid, CA6 protein or nucleic acid, CD109 protein or nucleic acid, and TSP4 protein or nucleic acid to a control; and identifying the subject as being at risk of chemotherapy-induced cardiotoxicity if there is an increase in the level of BGN protein or nucleic acid, an increase in the level of CA6 protein or nucleic acid, an increase in the level of CD109 protein or nucleic acid and / or an increase in the level of TSP4 protein or nucleic acid compared to the control.

[0091] In some aspects, the method further includes measuring the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid; comparing the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid to a control; and identifying the subject as being at risk of chemotherapy-induced cardiotoxicity if there is an increase in the level of CDH5 protein or nucleic acid and / or a decrease in the level of CST7 protein or nucleic acid compared to the control.

[0092] In some aspects, the method further includes using a computational algorithm to identify the subject as being at risk of chemotherapy-induced cardio toxicity. The algorithms are trained on patients’ data with known occurrence (or non-occurrence) of cardiotoxicity and the measurements for protein or nucleic acids. The performance of the algorithms is optimized on the accuracy for predicting the occurrences of cardiotoxicity with protein or nucleic acid levels. Once trained, the algorithm is used to predict the occurrence of cardiotoxicity for patient with only protein or nucleic acid measurements. Some examples for machine learning algorithms include various classification approaches such as PLS-DA, support vector machines (SVM), neural networks, logistic regression, classification and regression trees, and boosting.

[0093] In some aspects, the chemotherapy includes an anthracy cline. In some examples, the anthracycline includes doxorubicin, daunorubicin, epirubicin, idarubicin, mitoxantrone or valrubicin. In specific examples, the anthracycline is doxorubicin (DOX), or a liposomal form thereof, such as Doxil, Lipodox, or Lipodox 50.

[0094] Also provided herein is a method that includes obtaining a biological sample from a subject diagnosed with cancer: measuring the level of one or more of (such as two, three or four of) BGN protein or nucleic acid, CA6 protein or nucleic acid, CD109 protein or nucleic acid and / or TSP4 protein or nucleic acid in the biological sample; and comparing the level of the one or more of BGN protein or nucleic acid, CA6 protein or nucleic acid, CD109 protein or nucleic acid and TSP4 protein or nucleic acid to a control. In some aspects, the method further includes measuring the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid in the biological sample; and comparing the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid to a control.

[0095] In some aspects of the disclosed methods, the subject has not yet received treatment for the cancer. In other aspects, the subject has not yet received treatment with a chemotherapeutic agent, but has received at least one other anti-cancer treatment, such as surgery, radiation, or administration of a tyrosine kinase inhibitor, an angiogenesis inhibitor, an immunomodulatory agent, or immunotherapy (such as a monoclonal antibody or chimeric antigen receptor targeting a tumor antigen).

[0096] In some aspects of the disclosed methods, the control is a reference standard or a historical control. For example, the control can be a sample obtained from a healthy subject or from a patient who is not at risk of chemotherapy-induced cytotoxicity or a group of healthy subjects. In other examples, the control is an average or median value obtained from a group of patients who are not at risk of chemotherapy-induced cytotoxicity. In other examples, the control is a reference value based on the baseline blood level of a group of cancer patients, such as patients who have not yet been treated with chemotherapy.

[0097] In some aspects of the disclosed methods, the biological sample includes blood, serum, plasma, urine, cells, or tissue (such as a tumor biopsy or fine needle aspirate). In particular examples, the biological sample includes plasma.

[0098] In some aspects, the biological sample is collected before initiation of chemotherapy and / or other anti-cancer therapies. In some examples, the biological sample is collected before and after treatment with chemotherapy and / or another anti-cancer treatment.

[0099] The level of each biomarker in the biological sample can be measured using any known method for quantifying protein or nucleic acid. In some aspects, the level of BGN protein, CA6 protein, CD109 protein, TSP4 protein, CDH5 protein and / or CST7 protein is measured by immunoassay, antibody array, aptamers that specifically bind each protein (e.g., SOMAMER® aptamers), or proximity extension assay (OLINK). In some aspects, the level of BGN nucleic acid, CA6 nucleic acid, CD 109 nucleic acid, TSP4 nucleic acid, CDH5 nucleic acid, and / or CST7 nucleic acid is measured by quantitative PCR (qPCR), digital PCR (dPCR), next generation sequencing (NGS), Northern blotting, nuclease protection assay, RNA-seq, or in situ hybridization.

[0100] In some aspects of the disclosed methods, the method further includes measuring left ventricular ejection fraction (LVEF) in the subject. A decrease in LVEF is an indicator of chemotherapy-induced cardiotoxicity, particularly an LVEF reduction of 10% absolute unit or more compared to the subject’s own baseline LVEF, or LVEF of less than about 50%, less than about 45%, less than about 40%, less than about 35%, or less than about 30%.

[0101] In some aspects of the disclosed methods, the subject has been diagnosed with a solid tumor, a leukemia, or a lymphoma. In particular examples, the subject has been diagnosed with a breast cancer.

[0102] In some aspects, when the subject is identified as being at risk of chemotherapy-induced cytotoxicity, the method further includes treating the subject with a chemotherapeutic agent that is not an anthracy cline. In some examples, the subject is treated with a chemotherapeutic agent that is not DOX.

[0103] In some aspects, when the subject is not identified as being at risk of chemotherapy-induced cytotoxicity, the method further includes treating the subject with a chemotherapeutic agent, such as an anthracy cline. In some examples, the subject is administered an anthracycline selected from DOX, daunorubicin, epirubicin, idarubicin, mitoxantrone and valrubicin. In specific examples, the subject is administered DOX. In one example, the subject has been diagnosed with breast cancer, is not at risk for chemotherapy-induced cytotoxicity and is administered DOX.

[0104] In some aspects of the disclosed methods (when the subject is either identified as at risk or is identified as not being at risk of chemotherapy-induced cardiotoxicity), the subject is further treated with an anti-cancer agent that is not a chemotherapeutic agent. In some examples, the subject is administered an immunomodulator or a monoclonal antibody (such as a monoclonal antibody specific for a tumor antigen expressed by the subject’s cancer). In some examples, the subject is administered a tyrosine kinase inhibitor, an angiogenesis inhibitor, or an immunotherapy (such as a chimeric antigen receptor targeted to a tumor antigen expressed by the subject’s cancer).

[0105] In some aspects of the methods, the subject (at risk or not at risk of chemotherapy-induced cardiotoxicity) is further treated with surgery (such as surgical resection of a tumor) or radiation therapy.

[0106] In some aspects of the disclosed methods, when the subject is identified as being at risk of chemotherapy-induced cardiotoxicity, the method further includes administering a cardioprotective agent to the subject. The cardioprotective agent can be administered before chemotherapy, during chemotherapy, and / or after chemotherapy. In some examples, the cardioprotective agent includes a beta blocker (e.g., Carvedilol, Metoprolol, Nebivolol, or Bisoprolol), an angiotensin receptor blocker (e.g., Candesartan or Telmisartan), an angiotensin-converting-enzyme inhibitor (ACEI) (e.g., Enalapril, Lisinopril, or Perindopril), a statin (e.g., Atorvastatin), an angiotensin receptor neprilysin inhibitor (e.g., Sacubitril / Valsartan), a diuretic (e.g., Spironolactone), an iron chelator (e.g., Dexrazoxane), a hyperpolarization-activated cyclic nucleotide-gated (HCN) channel blocker (e.g., Ivabradine), or combinations thereof.

[0107] Also provided herein are kits, such as kits for measuring the level of biomarkers associated with chemotherapy-induced cardiotoxicity. In some aspects, the kit includes a plurality of antibodies, wherein the plurality of antibodies includes at least two antibodies (or at least three, at least four, at least live or six antibodies) selected from the group consisting of an antibody specific for BGN, an antibody specific for CA6, an antibody specific for CD109, an antibody specific for TSP4, an antibody specific for CDH5 and an antibody specific for CST7, and further including no more than five additional antibodies specific for other proteins.

[0108] In other aspects, the kit includes a plurality of aptamers including at least two aptamers (or at least three, at least four, at least five, or six aptamers) selected from the group consisting of an aptamer specific for BGN, an aptamer specific for CA6, an aptamer specific for CD109, an aptamer specific for TSP4, an aptamer specific for CDH5 and an aptamer specific for CST7, and further including no more than five additional aptamers specific for other proteins.

[0109] In other aspects, the kit includes a plurality of antibody pairs including at least two antibody pairs (or at least three, at least four, at least five, or six antibody pairs) selected from the group consisting of a pair of antibodies specific for BGN, a pair of antibodies specific for CA6, a pair of antibodies specific for CD109, a pair of antibodies specific for TSP4, a pair of antibodies specific for CDH5, and a pair of antibodies specific for CST7, and further including no more than five additional antibody pairs specific for other proteins. In some examples, each pair of antibodies is linked to unique DNA sequences that hybridize only to each other. In other examples, each of pair of antibodies is linked to unique chromophores (such as fluorophores). In other examples, one of the pair of antibodies is linked to colored solid beads and the other of the pair is linked to an affinity tag (such as biotin) or a chromophore (such as a fluorophore). A skilled person can select appropriate antibody pairs for biomarker detection (see, e.g., xMAP® Cookbook, Luminex Corporation; Dunbar, Clin Chim Acta 363(l):71-82, 2006).

[0110] In other aspects, the kit includes a plurality of antibody / aptamer pairs including at least two antibody / aptamer pairs (at least three, at least four, at least five, or six antibody / aptamer pairs) selected from the group consisting of an antibody / aptamer pair specific for BGN, an antibody / aptamer pair specific for CA6, an antibody / aptamer pair specific for CD 109, an antibody / aptamer pair specific for TSP4, an antibody / aptamer pair specific for CDH5, and an antibody / aptamer pair specific for CST7, and further including no more than five additional antibody / aptamer pairs specific for other proteins. In some examples, the antibody is linked to solid beads and / or an affinity tag (such as biotin) and / or a chromophore (such as a fluorophore). In some examples, the aptamer is linked to solid beads and / or an affinity tag (such as biotin) and / or a chromophore (such as a fluorophore). As used herein, an antibody / aptamer pair refers to an antibody and an aptamer that are both specific for the same marker.

[0111] In some examples, the kit includes an antibody specific for BGN, an antibody specific for CA6, an antibody specific for CD109, and an antibody specific for TSP4. In other examples, the kit includes an aptamer specific for BGN, an aptamer specific for CA6, an aptamer specific for CD109 and an aptamer specific for TSP4. In other examples, the kit includes a pair of antibodies specific for BGN, a pair of antibodies specific for CA6, a pair of antibodies specific for CD 109, and a pair of antibodies specific for TSP4, wherein each pair of antibodies is linked to unique DNA sequences that hybridize only to each other; each pair of antibodies is linked to unique chromophores; or one of the pair of antibodies is linked to colored solid beads and the other of the pair is linked to an affinity tag or a chromophore.

[0112] In other examples, the kit includes an antibody specific for BGN, an antibody specific for CA6, an antibody specific for CD109, an antibody specific for TSP4, an antibody specific for CDH5, and an antibody specific for CST7. In other examples, the kit includes an aptamer specific for BGN, an aptamer specific for CA6, an aptamer specific for CD 109, an aptamer specific for TSP4, an aptamer specific for CDH5, and an aptamer specific for CST7. In other examples, the kit includes a pair of antibodies specific for BGN, a pair of antibodies specific for CA6, a pair of antibodies specific for CD 109, a pair of antibodies specific for TSP4, a pair of antibodies specific for CDH5, and a pair of antibodies specific for TSP4, wherein each pair of antibodies is linked to unique DNA sequences that hybridize only to each other; each pair of antibodies is linked to unique chromophores; or one of the pair of antibodies is linked to colored solid beads and the other of the pair is linked to an affinity tag or a chromophore.

[0113] In other examples, the kit includes an antibody specific for BGN, an aptamer specific for BGN, an antibody specific for CA6, an aptamer specific for CA6, an antibody specific for CD 109, an aptamer specific for CD 109, an antibody specific for TSP4, an aptamer specific for TSP4, an antibody specific for CDH5, an aptamer specific for CDH5, an antibody specific for CST7, and an aptamer specific for CST7. In some instances, the antibodies are linked to solid beads, an affinity tag, and / or a chromophore, and the aptamers are linked to solid beads, an affinity tag, and / or a chromophore.

[0114] In some aspects of the disclosed kits, the antibodies are immobilized on a solid support. In some examples, the solid support includes a slide, a membrane, or beads (such as magnetic beads).

[0115] In some aspects, the kit further includes recombinant BGN protein, recombinant CA6 protein, recombinant CD 109 protein, recombinant TSP4 protein, recombinant CDH5 protein and / or recombinant CST7 protein. In some examples, the recombinant BGN, CA6, CD109, TSP4, CDH5 and / or CST7 proteins are human proteins.

[0116] In some aspects, the kit further includes one or more control proteins, buffer(s), reagent(s), multi-well plate(s), tube(s), mixing bottles(s), a qPCR template, primers, nucleic acid probes, DNA polymerase, dNTPs, loading control dye(s), detection reagent(s), detection enzyme(s), chromophore - linked avidin or streptavidin (or other affinity tags), substrate(s), adhesive plate sealer(s), DNA microarray slide(s), calibrator(s), quality control sample(s), sample matrix, next-generation sequencing (NGS) adapters, indexing primers, ligase, dA-tailing enzyme(s), DNA cleanup beads, instructions, labelled secondary antibodies, or any combination thereof.

[0117] V. Methods for Detection of Protein Biomarkers

[0118] The protein biomarkers disclosed herein can be detected in a biological sample, such as a plasma or blood sample, using any one of a number of known assay methods for detection and quantitation of proteins. For example, protein biomarkers can be detected using antibody-based immunoassays, antibody array, protein-specific aptamers, or proximity extension assays.

[0119] A. Antibody-based detection

[0120] Antibodies specific for a particular protein biomarker can be used for detection and quantification by one of a number of known immunoassay methods, such as those described in Antibodies, A Laboratory Manual, Second Edition (Cold Spring Harbor Laboratories, 2014). For example, any standard immunoassay format, such as ELISA, Western blot, cytometric bead assay, or radioimmunoassay (RIA) can be used to measure protein levels. Thus, biomarker protein levels in a sample (such as a blood or plasma sample) can readily be evaluated using these methods. General guidance regarding such techniques can be found in Bancroft and Gamble (Theory and Practice of Histological Techniques, Churchill Livingstone, 2008) and Ausubel et al. (Current Protocols in Molecular Biology, John Wiley & Sons, New York, 1998).

[0121] For immunoassay methods, commercially available antibodies to a target protein can be utilized. The table below provides a list of exemplary commercially available antibodies that can be used to detect the protein biomarkers disclosed herein.

[0122] In some aspects, protein biomarkers associated with chemotherapy-induced cardiotoxicity are detected using an antibody array. Antibody arrays are available for a variety of different commercial sources, including PROTEOME PROFILER™ from R&D Systems (Minneapolis, MN), antibody arrays from RayBiotech (Peachtree Corners, GA), and antibody arrays from Abeam (Boston, MA).

[0123] In some aspects, protein biomarkers associated with chemotherapy-induced cardiotoxicity are detected using a multiplex immunoassay, that is, two or more proteins arc detected simultaneously in a single immunoassay. A skilled person can select an appropriate multiplex immunoassay known in the art. An exemplary method is provided in Example 8.

[0124] Moreover, methods of making polyclonal and monoclonal antibodies are well-known. Polyclonal antibodies, antibodies which consist essentially of pooled monoclonal antibodies with different epitopic specificities, as well as distinct monoclonal antibody preparations are included. The preparation of polyclonal antibodies is well-known (see, for example, Green et al., “Production of Polyclonal Antisera,” in: Immunochemical Protocols, pages 1-5, Manson, ed., Humana Press, 1992; Coligan et al., “Production of Polyclonal Antisera in Rabbits, Rats, Mice and Hamsters,” in: Current Protocols in Immunology, section 2.4.1, 1992).

[0125] The preparation of monoclonal antibodies likewise is conventional (see, for example, Kohler & Milstein, Nature 256:495, 1975; Coligan et al., sections 2.5.1-2.6.7; and Antibodies. A Laboratory Manual, Second Edition, Cold Spring Harbor Laboratories, 2014). Briefly, monoclonal antibodies can be obtained by injecting mice with a composition that includes an antigen, verifying the presence of antibody production by removing a serum sample, removing the spleen to obtain B lymphocytes, fusing the B lymphocytes with myeloma cells to produce hybridomas, cloning the hybridomas, selecting positive clones that produce antibodies to the antigen, and isolating the antibodies from the hybridoma cultures. Monoclonal antibodies can be isolated and purified from hybridoma cultures by a variety of well-established techniques. Such isolation techniques include affinity chromatography with Protein-A Sepharose, size-exclusion chromatography, and ion-exchange chromatography (see, e.g., Coligan etal., sections 2.7.1-2.7.12 and sections 2.9.1-2.9.3; Barnes et al., Purification of Immunoglobulin G (IgG), in: Methods in Molecular Biology, Vol. 10, pages 79-104, Humana Press, 1992).

[0126] Monoclonal antibodies specific for a particular protein can also be generated and selected using an antibody display library, such as a phage display library.

[0127] Antibodies include intact molecules as well as fragments thereof, such as Fab, F(ab')2, and Fv which are capable of binding the epitopic determinant. These antibody fragments retain the ability to selectively bind with their antigen and are defined as follows:

[0128] (1) Fab, the fragment which contains a monovalent antigen-binding fragment of an antibody molecule, can be produced by digestion of whole antibody with the enzyme papain to yield an intact light chain and a portion of one heavy chain;

[0129] (2) Fab', the fragment of an antibody molecule can be obtained by treating whole antibody with pepsin, followed by reduction, to yield an intact light chain and a portion of the heavy chain: two Fab' fragments are obtained per antibody molecule;

[0130] (3) (Fab'):, the fragment of the antibody that can be obtained by treating whole antibody with the enzyme pepsin without subsequent reduction; F(ab')2 is a dimer of two Fab' fragments held together by two disulfide bonds;

[0131] (4) Fv, defined as a genetically engineered fragment containing the variable region of the light chain and the variable region of the heavy chain expressed as two chains; and

[0132] (5) Single chain antibody, defined as a genetically engineered molecule containing the variable region of the light chain, the variable region of the heavy chain, linked by a suitable polypeptide linker as a genetically fused single chain molecule.

[0133] Methods of making these fragments are known in the art (see for example, Antibodies, A Laboratory Manual, Second Edition, Cold Spring Harbor Laboratories, 2014). For example, antibody fragments can be prepared by proteolytic hydrolysis of the antibody or by expression in E. coli of DNA encoding the fragment. Antibody fragments can be obtained by pepsin or papain digestion of whole antibodies by conventional methods. For example, antibody fragments can be produced by enzymatic cleavage of antibodies with pepsin to provide a 5S fragment denoted F(ab')2. This fragment can be further cleaved using a thiol reducing agent, and optionally a blocking group for the sulfhydryl groups resulting from cleavage of disulfide linkages, to produce 3.5S Fab' monovalent fragments. Alternatively, an enzymatic cleavage using pepsin produces two monovalent Fab' fragments and an Fc fragment directly (see U.S. Patent No. 4,036,945 and U.S. Patent No. 4,331,647, and references contained therein; Nisonhoff et al., Arch Biochem Biophys 89:230, 1960; Porter, Biochem J 73: 119, 1959; Edelman et al., Methods in Enzymology, Vol. 1, page 422, Academic Press, 1967; and Coligan et al. at sections 2.8.1-2.8.10 and 2.10.1-2.10.4). B. Aptamer-based detection

[0134] Aptamers, small nucleic acid molecules that specifically bind a target protein, can be used for detection and quantification of protein biomarkers in a biological sample. Aptamer-based arrays can be used for simultaneous detection of multiple biomarkers, even from very small sample volumes. A technology developed by SOMALOGIC (Boulder, CO) utilizes a class of aptamer termed Slow Off- rate Modified Aptamer (SOM AMER® aptamers). These aptamers are high-affinity aptamers that can be selected for almost any protein target (Gold et al., PLoS ONE 5(12):el5004), allowing for high- throughput proteomic analyses, such as for identifying protein biomarkers of disease (Webber et al. , Molecular & Cellular Proteomics 13:10.1074 / mcp.Ml 13.032136, 1050-1064, 2014).

[0135] The SOMALOGIC proteomics assay is described in detail in Gold et al. (PLoS ONE 5(12):el5004) and Gold et al. (Cold Spring Harb Perspect Biol 4:a003582, 2012). As described herein, biological samples (e.g., plasma samples) from patients diagnosed with breast cancer were evaluated using the SOMALOGIC aptamer-based proteomics platform (see Example 1). In brief, quantitative proteome profiling of plasma samples was performed by SOMASCAN assay (version 1.3k) as described previously (Gold et al., PLoS One 2010, 5 (12), el5004; Yu et al., Kidney Int Rep 2018, 3 (5), 1202-1213). The assay quantified 1,305 low, middle, and high abundance proteins using single-stranded DNA slow off-rate modified aptamers (SOMAMER® aptamers) with 5 calibration samples and 2 quality control samples for each assay. Samples (50 pL) were incubated with preimmobilized SOMAMER® aptamers, which were subjected to a series of washes to remove nonspecific binders. SOMAMER® aptamers specifically bound to their cognate proteins were released and hybridized to custom DNA microarrays. The microarrays were scanned using an Agilent C scanner (G2505C, Agilent Technologies, Palo Alto, CA). The raw data were processed as described previously (Yu et al., Kidney Int Rep 2018, 3 (5), 1202-1213). Hybridization control normalization was applied to relative fluorescence units (RFUs) to remove variation introduced during the hybridization and scanning processes, followed by median signal normalization to eliminate intra-run bias, and finally calibration to account for inter-run differences.

[0136] Aptamers for detection of the protein biomarkers disclosed herein can also be obtained from other commercial sources, such as LC Sciences (Houston, TX), TriLink Biotechnologies (San Diego, CA), and Aptagen (Jacobus, PA).

[0137] C. Proximity extension assay

[0138] Proximity extension assays detect secreted proteins in biological samples, such as in serum and plasma samples. This technology is highly sensitive and specific, requires only very small sample volumes and can detect picogram quantities of protein. A proximity extension assay uses matched pairs of oligonucleotide-labelled antibodies, which bind to target proteins in a pairwise manner. The oligonucleotides include regions that are complementary to each other. Binding of the antibodies to their target brings the oligonucleotides into proximity with each other, allowing the complementary regions to hybridize. Using a DNA polymerase, the complementary regions are extended, producing a target sequence that can be amplified by PCR, allowing for detection and quantification (Wik et al., Mol Cell Proteomics 20: 100168, 2021). The target sequence can also be amplified using next generation sequencing (NGS).

[0139] Proximity extension assays are commercially available, such as from Olink Proteomics (Boston, MA). As described in Example 1, plasma samples from breast cancer patients were analyzed using the proximity extension assay of Olink Proteomics. The assay required only a few microliters of sample, as described in Assarsson et al. (PLoS One 2014, 9 (4), e95192). In brief, a total of 1463 proteins were measured using four OLINK Explore 384-plex panels. Quality control was performed by adding four internal controls into all samples and running external controls in every assay plate. More information regarding the OLINK platform, including the full list of proteins in the panel, is available online at olink.com.

[0140] Proximity extension assays are described in, for example, PCT Publication Nos. WO 2012 / 104261, WO 2013 / 113699, and WO 2015 / 035087; and U.S. Patent Application Publication Nos. 2018 / 0292581 and 2023 / 0107654.

[0141] VI. Methods for Detection of Nucleic Acid Encoding Biomarkers

[0142] Nucleic acid molecules (such as mRNA) encoding the protein hiomarkers disclosed herein can be detected in a biological sample, such as a plasma or blood sample, using any one of a number of known assay methods for detection and quantitation of nucleic acid. For example, mRNA encoding the protein biomarkers can be detected using PCR-based methods, such as quantitative PCR (qPCR) or digital PCR (dPCR), next generation sequencing (NGS), Northern blotting, nuclease protection assay, RNA-seq, or in situ hybridization.

[0143] In some aspects herein, biomarker expression is determined by detecting mRNA encoding a biomarker of interest. Thus, the disclosed methods can include evaluating mRNA encoding the BGN, CA6, CD109, TSP4, CDH5, and CST7 proteins. In some examples, the mRNA is detected and quantified. RNA can be isolated from a biological sample obtaining from a subject with cancer (such as breast cancer) using well-known methods, including commercially available kits. General methods for mRNA extraction are disclosed in standard textbooks of molecular biology, including Ausubel et al., Current Protocols of Molecular Biology, John Wiley and Sons (1997). In one example, RNA isolation can be performed using a purification kit, buffer set and protease from commercial manufacturers, such as QIAGEN, according to the manufacturer's instructions. For example, total RNA from biological sample (such as a plasma or blood sample obtained from a subject) can be isolated using QIAGEN RNeasy mini-columns. Other commercially available RNA isolation kits include MASTERPURE and Complete DNA and RNA Purification Kit (EPICENTRE, Madison, WL).

[0144] Methods of measuring the level of biomarker nucleic acid (e.g., BGN, CA6, CD109, TSP4, CDH5, and CST7 nucleic acid) include, for example, methods based on hybridization analysis of polynucleotides or sequencing of polynucleotides. In some aspects, the level of biomarker mRNA in a sample is quantified using Northern blotting (Yang et al., BMC Genomics 23:66, 2022; Streit et al., Nat Protocols 4:37-43, 2009), in situ hybridization (Parker & Barnes, Methods in Molecular Biology 106:247-283, 1999); nuclease (such as RNAse) protection assay (Hod, Biotechniques 13:852-4, 1992); next generation sequencing (Qin, Cancer Biol Med 16( 1 ) :4- 10, 2019); RNA-seq (Wang et al., Nat Rev Genet 10(l):57-63, 2009) or PCR-based methods, such as reverse transcription polymerase chain reaction (RT-PCR) (Weis et al., Trends in Genetics 8:263-4, 1992).

[0145] A. Northern blot

[0146] In some aspects herein, biomarker RNA is detected by Norther blotting. This method is used to detect and quantify RNA, including mRNA. In this assay method, RNA is purified from a biological sample (such as a blood or plasma sample) and separated using electrophoresis. The separated RNA is transferred from the electrophoretic gel onto a solid membrane and exposed to a DNA probe complementary to a target RNA (such as a DNA probe complementary to a BGN, CA6, CD109, TSP4, CDH5, and CST7 mRNA). The probe is labelled with a tag (such as a radioactive, fluorescent, or chemical tag) to enable identification of a target RNA.

[0147] B. In situ hybridization (ISH)

[0148] In some aspects, biomarker nucleic acid is detected using in situ hybridization (ISH). This method applies and extrapolates the technology of nucleic acid hybridization to the single cell level. ISH permits the maintenance of morphology and the identification of cellular markers to be maintained and identified, and allows the localization of sequences to specific cells within populations, such as in tissues or blood samples. ISH is a type of hybridization that uses a complementary nucleic acid to localize one or more specific nucleic acid sequences in a portion or section of tissue (in situ), or, if the tissue is small enough, in the entire tissue (whole mount ISH). RNA ISH can be used to assay expression patterns in a tissue, such as the expression of biomarkers associated with chemotherapy-induced cardiotoxicity.

[0149] C. Nuclease protection assay

[0150] In other aspects herein, the level of biomarker nucleic acid is measured using a nuclease protection assay (such as an RNase protection assay). RNase protection assays are highly sensitive and can detect and measure the level of specific mRNA molecules in biological samples (Ma et al., Methods 10(3):273-278, 1996). Nuclease protection assays can identify one or more target RNA molecules of known sequence, even at very low concentrations. Isolated RNA is mixed with RNA or DNA probes that are complementary to a target RNA (such as a BGN, CA6, CD109, TSP4, CDH5, and CST7 mRNA). When present in a biological sample, the target RNA hybridizes to the complementary RNA / DNA to form a double stranded molecule. A ribonuclease that degrades only single-stranded RNA is added to the mixture, enabling detection of only the hybridized target RNA. D. Next generation sequencing (NGS)

[0151] In other aspects, biomarker nucleic acid (such as BGN, CA6, CD 109, TSP4, CDH5, and CST7 mRNA) is detected and measured using NGS. NGS can be used for DNA or RNA sequencing, as well as for the detection of variant or mutant nucleic acid. This technology combines the advantages of unique sequencing chemistries, different sequencing matrices, and bioinformatics technology. Such a combination allows a massive parallel sequencing of various lengths of DNA or RNA sequences or even whole genome within a relatively short period of time (Qin, Cancer Biol Med 16(1):4- 10, 2019). Methods of performing NGS are described in, for example, PCT publication Nos. WO 2016 / 077602 and WO 2017 / 147294; U.S. Publication No. 2023 / 0144391; and Shendure, Nat Biotechnol 26: 1135-1145, 2008).

[0152] E. RNA sequencing (RNA-Seq)

[0153] In other aspects herein, biomarker nucleic acid is detected and measured using RNA-seq. This method uses NGS to identify and quantify RNA molecules in a biological sample. RNA-seq methods include isolating RNA from a biological sample and mixing it with a DNase to reduce the amount of genomic DNA in the sample. Particular types of RNA can either be selected for or depleted depending upon the application. For example, the isolated RNA can be enriched for RNAs having a polyA tail, resulting in isolation / concentration of mRNA. After any desired enrichment or depletion steps, cDNA is produced by reverse transcription. The cDNA is then fragmented and adapters are added to each of the fragments. The adaptors include functional elements that allow for sequencing. Following amplification, size selection, clean-up and quality checking, the cDNA library is analyzed by NGS. RNA-seq is described in Wang et al. (Nat Rev Genet 10(l):57-63, 2009) and Owens et al. (Cold Spring Harb Protoc 2019(6), doi: 10.1101 / pdb.prot098368, 2019).

[0154] F. PCR-based methods

[0155] In some aspects herein, the level of biomarker nucleic acid is measured using a PCR-based method. In some examples, the method utilizes RT-PCR. Generally, the first step in gene expression profiling by RT-PCR is the reverse transcription of the RNA template into cDNA, followed by its exponential amplification in a PCR reaction. The reverse transcription step is typically primed using specific primers, random hexamers, or oligo-dT primers, depending on the circumstances and the goal of expression profiling. For example, extracted RNA can be reverse-transcribed using a GeneAmp RNA PCR kit (Perkin Elmer, Calif., USA), following the manufacturer's instructions. The derived cDNA can then be used as a template in the subsequent PCR reaction.

[0156] Although the PCR step can use a variety of thermostable DNA-dependent DNA polymerases, it typically employs the Taq DNA polymerase. TaqMan® PCR typically utilizes the 5'-nuclease activity of Taq polymerase to hydrolyze a hybridization probe bound to its target amplicon, but any enzyme with equivalent 5' nuclease activity can be used. Two oligonucleotide primers are used to generate an amplicon typical of a PCR reaction. A third oligonucleotide, or probe, is designed to detect nucleotide sequence located between the two PCR primers. The probe is non-extendible by Taq DNA polymerase enzyme, and in some examples, is labeled with a reporter fluorescent dye and a quencher fluorescent dye. Any laser-induced emission from the reporter dye is quenched by the quenching dye when the two dyes are located close together as they are on the probe. During the amplification reaction, the Taq DNA polymerase enzyme cleaves the probe in a template-dependent manner. The resultant probe fragments disassociate in solution, and signal from the released reporter dye is free from the quenching effect of the second fluorophore. One molecule of reporter dye is liberated for each new molecule synthesized, and detection of the unquenched reporter dye provides the basis for quantitative interpretation of the data.

[0157] To minimize errors and the effect of sample-to-sample variation, RT-PCR can be performed using an internal standard. RNAs commonly used to normalize patterns of gene expression are mRNAs for the housekeeping genes glyceraldehyde-3-phosphate-dehydrogenase (GAPDH), betaactin, and 18S ribosomal RNA.

[0158] A variation of PCR is real time quantitative PCR (qPCR), which measures PCR product accumulation through a dual-labeled fluorogenic probe (e.g., TAQMAN® probe). Real time PCR is compatible both with quantitative competitive PCR, where internal competitor for each target sequence is used for normalization, and with quantitative comparative PCR using a normalization gene contained within the sample, or a housekeeping gene for RT-PCR (see Held et al., Genome Research 6:986 994, 1996).

[0159] In some examples, the PCR-based method is digital PCR (dPCR), which allows for absolute quantification through partitioning the reaction. qPCR is highly sensitive and accurate for molecular detection and requires only very small sample volumes (see, e.g., Mao et al., Am J Transl Res l l(12):7209-7222, 2019; and Quan et al., Sensors (Basel) 18(4): 1271, 2018).

[0160] EXAMPLES

[0161] The following examples are provided to illustrate particular features of certain aspects of the disclosure, but the scope of the claims should not be limited to those features exemplified.

[0162] Example 1: Methods

[0163] This example describes the materials and experimental procedures for the studies described in Examples 1-7.

[0164] Study Design (Patients and plasma samples)

[0165] Eighty-three female breast cancer patients were enrolled for DOX-based chemotherapy (FIG. 1 A). All patients were treated with a combination of 60 mg / m2DOX and 600 mg / m2cyclophosphamide every 2-3 weeks for four cycles (240 mg / m2cumulative dose of DOX and 2400 mg / m2cumulative dose of cyclophosphamide). Blood samples were collected into EDTA blood collection tubes immediately prior to the first (TO), the second (T 1), and the third (T2) doses of DOX- based treatment (FIG. IB). Blood samples were immediately processed into plasma by centrifugation. The plasma samples were de-identified and stored at -80°C until further analysis.

[0166] Cardiac function of the patients was assessed by measuring LVEF using MUGA scans before the first dose and after completion of the 4thcycle of DOX-based treatment. An asymptomatic reduction of the LVEF of >10% or LVEF <50%, or a reduction of LVEF >5% to LVEF<55% with symptoms of heart failure was considered as cardiotoxicity (cardiotox) (Mookadam et al. , Front Oncol 2014, 4, 259; Seidman et al., J Clin Oncol 2001, 19 (10), 2587-2595).

[0167] For this study, plasma samples collected prior to initiation of DOX-based chemotherapy (TO) were analyzed to identify and verify proteomic biomarkers for the prediction of chemotherapy-related cardiotoxicity. From a total of 83 female breast cancer patients, 39 patients were randomly assigned to the biomarker discovery cohort and the remaining 44 patients were grouped into the biomarker validation cohort (FIG. 1C). Both cohorts kept similar incidence of treatment-induced cardiotoxicity.

[0168] ELISA of Plasma cTnT Concentrations

[0169] Patients’ plasma concentrations of cTnT were measured using enzyme-linked immunosorbent assay (ELISA) (CAT# LS-F26593, USA), a sandwich chemiluminescent immunoassay (CLIA) with a detection range of 2.74-2000 pg / mL. The assay was performed per manufacturer’s standard procedure. Assay data were acquired on a Cytation 5 cell imaging multimode reader (BioTek, Winooski, VT).

[0170] SOMASCAN Proteomic Profiling

[0171] Quantitative proteome profiling of plasma samples was performed for 39 patients in the discovery cohort by SOMASCAN assay (version 1.3k) developed by SomaLogic Inc. (Boulder, CO) as described previously (Gold et al., PLoS One 2010, 5 (12), el5004; Yu et al., Kidney Int Rep 2018, 3 (5), 1202-1213). The assay quantified 1,305 low, middle, and high abundance proteins using singlestranded DNA SOMAMER® aptamers with 5 calibration samples and 2 quality control samples for each assay. Briefly, serum samples (50 pL) were incubated with pre-immobilized SOMAMER® aptamers, which were subjected to a series of washes to remove non-specific bindings. SOMAMER® aptamers specifically bound to their cognate proteins were released and hybridized to custom DNA microarrays. The microarrays were scanned using an Agilent C scanner (G2505C, Agilent Technologies, Palo Alto, CA). The raw data were processed as described previously (Yu et al., Kidney Int Rep 2018, 3 (5), 1202-1213). Hybridization control normalization was applied to relative fluorescence units (RFUs) to remove variation introduced during the hybridization and scanning processes, followed by median signal normalization to eliminate intra-run bias, and finally calibration to account for inter-run differences. OLINK Assay of Plasma Proteins

[0172] Plasma samples from 44 patients in the validation cohort were measured by Olink Proteomics (Boston, MA) using the Proximity Extension Assay technology with consumption of a few microliters of sample, as previously described (Assarsson et al., PLoS One 2014, 9 (4), e95192). A total of 1463 proteins were measured using 4 OLINK Explore 384-plex panels, including cardiometabolic, inflammation, neurology, and oncology. Quality control was performed by adding four internal controls into all samples and running external controls in every assay plate. Assay results were reported in arbitrary, relative units as Normalized Protein expression (NPX) on a log2 scale. More information regarding the OLINK platform, NPX, assay validation data, and the full list of proteins in the panel is available online (olink.com).

[0173] Analytical Correlation of SOMASCAN and OLINK Assays

[0174] To evaluate whether protein measurements are comparable and consistent between the SOMASCAN and OLINK assays, 28 samples were analyzed by both assays. The data were log transformed and correlation of common proteins between SOMASCAN and OLINK assays was assessed using Pearson correlation analysis under R (version 4.1.3) to calculate the correlation coefficient and p-value.

[0175] Identification of Differential Proteins

[0176] Patient plasma samples were stratified based on cardiotoxicity status after completion of chemotherapy. Welch’s z-test was performed for log-transformed SOMASCAN RFUs and OLINK NPX values to find significantly changed proteins between the cardiotoxicity (cardiotox) and non- cardiotoxicity (normal) groups. A fold change of >1.2 and p < 0.05 were set as significant changes as the criteria were evaluated and validated (Yu et al., Kidney Int Rep 2018, 3 (5), 1202-1213). The t-test ^-values were also adjusted for this multiplex assay to calculate false discovery rate (FDR) using the Benjamini and Hochberg method (Benjamini and Hochberg, JR Statist Soc B 57(l):289-300, 1995). For protein expression mean, standard deviation (SD), and fold changes, RFUs were used for SOMASCAN data while log scale NPX values were converted to linear scale values (non-log transformed) for OLINK data.

[0177] Statistical Analysis

[0178] Continuous variables of patient characteristics were expressed as mean with standard deviation or median (25th, 75thpercentile) and categorical variables as frequencies and percentages. The differences between patients with and without cardiotoxicity were evaluated using t-test for continuous variables and Chi-square test (or Fisher’s exact test) for categorical variables. The differential proteins (cardiotox vs. normal groups) in the validation cohort that were consistent with those in the discovery cohort were further analyzed statistically. The association of plasma levels of each protein with the probability of cardiotoxicity was evaluated in the univariate logistic regression initially, and then in the multivariable logistic regression with adjustment for baseline LVEF, ER / PR / HER2 status, hypertension, and vitamin D deficiency status. Logistic regression analysis was also performed for patient characteristics covariates per se for their potential association with cardiotoxicity. All statistical analyses were performed using the SAS software version 9.4 (SAS Institute, Cary, NC).

[0179] Predictive Models of Cardiotoxicity and Biomarker Validation

[0180] To explore the feasibility to predict cardiotoxicity, a classifier based on partial least squares discriminant analysis (PLS-DA; Barker and Rayens, Journal of Chemometrics 17(3): 166-173, 2003) using the biomarker discovery data (SOMASCAN data) was trained. Performance of the model was evaluated with the OL1NK data for validation. As protein abundance levels were measured with different assays for the biomarker discovery and validation cohorts, the measurements of each protein were first normalized by the corresponding median value of the protein in each cohort. The PLS-DA classifier was then trained with the normalized protein abundance (BGN, CA6, CD109, CDH5, CST7, TSP4) as well as the baseline LVEF value as predictors and with the status of cardiotoxicity (normal or abnormal) as the response. The PLS-DA classifier was fitted with the mixOmics package in the R statistical software following Rohart et al., PLoS Comput Biol 13(1 l):el005752, 2017. The fitted classifier was then used to predict the cardiotoxicity for patients in the validation cohort. Performance metrics such as sensitivity and specificity were computed with R.

[0181] Example 2: Patient Characteristics and Cardiotoxicity

[0182] The overall characteristics of 83 patients as well as those stratified by the status of cardiac function after completion of the therapy (cardiotoxicity or non-cardiotoxicity) are shown in Table 1. The mean age of the patient cohort was 52.0+11.4 years with an average body mass index (BMI) of 32.0+7.5 kg / m2. Twenty-six patients (31.3%) were diagnosed as triple negative breast cancer (ER- / PR- / HER2-), while the remaining 57 patients (68.7%) were HER2- with either or both of ER+ and PR+. Thirty-eight, 13, and 21 patients had hypertension (45.8%), diabetes (15.7%), and vitamin D deficiency (25.3%), respectively. The cohort of patients had an overall baseline LVEF (%) of 63.8+6.7. After completion of chemotherapy, 19 patients (22.9%) experienced decreased cardiac function (LVEF) as defined in the methods with average LVEF (%) change of -14.7±4.0 while the remaining 64 (77.1%) patients maintained normal cardiac function with average LVEF (%) change of 0.2+6.1 compared to corresponding baseline levels of the same group. It was observed that patients in the cardiotoxicity group had higher baseline LVEF (%) (67.7+5.5) than those in the non- cardiotoxicity group (62.6+6.6), and this difference was statistically significant ( / i=0.0015). Table 1. Characteristics of breast cancer patients stratified by cardiotoxicity status after doxorubicin-based chemotherapy

[0183] “Continuous variables are expressed as mean ± SD, categorical variables as %;bp- values comparing patients who experienced cardiotoxicity versus those who maintained normal cardiac functions were calculated using / -test for continuous variables and Chi-square test (or Fisher’s exact test if the count is <5) for categorical variables. ND, not detectable (below the detection limit); NA, not applicable.

[0184] Example 3: SOMASCAN Proteomic Analysis of the Biomarker Discovery Cohort

[0185] In the biomarker discovery cohort of 39 patients, 9 patients experienced cardiotoxicity after chemotherapy. Characteristics of the cohort and groups of patients stratified by cardiotoxicity status are presented in Table 2. SOMASCAN analysis of plasma samples collected from the discovery cohort prior to initiation of chemotherapy showed that most of the proteins did not present statistically significant difference in baseline abundance levels between the cardiotoxicity and non-cardiotoxicity groups (black dots, FIG. 2A). The volcano plot of SOMASCAN analysis demonstrates fold changes (FC cutoff=1.2) and statistical significance (p<0.05) of those proteins that were higher (red dots) or lower (green dots) in plasma samples from patients who experienced treatment-related cardiotoxicity compared to those who maintained normal cardiac functions after chemotherapy. A total of 48 proteins exhibited differential baseline levels between the two groups, 41 were higher and 7 were lower in the cardiotoxicity group (FIG. 2A). Table 2. Characteristics of the biomarker discovery patient cohort stratified by cardiotoxicity status after doxorubicin-based chemotherapy

[0186] “Continuous variables are expressed as mean + SD, categorical variables as %;bp- values comparing patients who experienced cardiotoxicity versus those who maintained normal cardiac functions were calculated using t-test for continuous variables and Chi-square test (or Fisher’s exact test if the count is <5) for categorical variables. ND, not detectable (below the detection limit); NA, not applicable.

[0187] Example 4: OLINK Proteomic Analysis of the Biomarker Validation Cohort

[0188] In the biomarker validation cohort of 44 patients, 10 patients experienced cardiotoxicity after chemotherapy. Characteristics of the cohort and groups of patients stratified by cardiotoxicity status are presented in Table 3. OLINK proteomic analysis of plasma samples collected from the validation cohort prior to initiation of chemotherapy resulted in detection of 61 proteins with differential baseline levels between the cardiotoxicity and non-cardiotoxicity groups, 53 were higher and 8 were lower in the group of patients who experienced treatment-related cardiotoxicity (FIG. 2B, Table 4).

[0189] Table 3. Characteristics of the biomarker validation patient cohort stratified by cardiotoxicity status after doxorubicin-based chemotherapy

[0190] “Continuous variables are expressed as mean ± SD, categorical variables as %;bp-values comparing patients who experienced cardiotoxicity versus those who maintained normal cardiac functions were calculated using t-test for continuous variables and Chi-square test (or Fisher’s exact test if the count is <5) for categorical variables. ND, not detectable (below the detection limit); NA, not applicable.

[0191] Table 4. Proteins with significant changes in baseline levels (1.2-fold and p <0.05) as measured by OLINK assays in the biomarker validation patient cohort who experienced cardiotoxicity compared to those who did not experience cardiotoxicity after chemotherapy

[0192] Ct, cardiotoxicity, decreased LVEF after chemotherapy; N, normal, remained normal LVEF after chemotherapy; FDR, Benjamini & Hochberg adjusted p-value; SD, standard deviation.

[0193] Results from the OLINK analysis confirmed the differential baseline levels of 7 proteins from the SOMASCAN analysis of the discovery cohort, including higher levels of biglycan, cadherin-5, carbonic anhydrase 6 (CA6), CD 109, receptor tyrosine-protein kinase erbB-4 (HER4), and thrombospondin-4 (TSP4), as well as lower levels of cystatin-F in the patients who experienced treatment-related cardiotoxicity as shown in Table 5 and demonstrated in the violin plots in FIGS. 3A- 3B.

[0194] Table 5. Common proteins in both the biomarker discovery and validation cohorts that showed significantly different baseline levels in plasma between cardiotoxicity and non-cardiotoxicity groups

[0195] SOMASCAN

[0196] Discovery OLINK Validation

[0197] P-

[0198] Protein UniProt Gene Ratio (Ct / N) value Ratio (Ct / N) P-value

[0199] Ct, cardiotoxicity, decreased LVEF after chemotherapy; N, normal, remained normal LVEF after chemotherapy

[0200] Example 5: Analytical Correlation of SOMASCAN and OLINK Assays

[0201] To evaluate whether protein measurements between the SOMASCAN and OLINK assays were comparable and analytically consistent for the 7 proteins, 28 samples were analyzed by both assays aimed for analytical validation. Pearson correlation analysis of protein measurements for the same samples by both assays demonstrated strong positive correlations between the two assays for cadherin-5 (r=0.93, p=l . le-12), CA6 (r=0.93, p=7.0e-13), cystatin-F (r=0.93, p=5.4e-13), and TSP4 (r=0.8, p=2.8e-7). SOMASCAN and OLINK intensities of biglycan (r=0.45, yj=0.017) and CD109 (r=0.65, p-1.9e-4) showed moderate positive correlation (FIG. 4). However, HER4 did not show positive and significant correlation (r=-0.32, p=0. 1 ) (FIG. 5). Thus, 6 proteins, including biglycan, cadherin-5, CA6, CD109, cystatin-F, and TSP4, were further analyzed as potential biomarkers.

[0202] Example 6: Association of Baseline Covariates and Protein Biomarkers with Cardiotoxicity

[0203] Prior to evaluating whether the 6 SOMASCAN and OLINK correlated proteins were associated with chemotherapy-related cardiotoxicity, logistic regression analysis was performed for baseline covariates in the validation cohort, including age, body mass index (BMI), baseline LVEF, triple negative type breast cancer, vitamin D deficiency, and hypertension. The analysis showed that the odds ratios (OR) were close to 1 for age and BMI while patients with triple negative type breast cancer, vitamin D deficiency, or hypertension showed either decreased or increased odds ratios but not statistically significant, suggestion these covariates per se were not associated with cardiotoxicity. However, slightly increased odds of cardiotoxicity (OR=1.15, 95% confidence interval [CI]=1.01- 1.32, =0.04) was observed for the patients with higher baseline LVEF (Table 6).

[0204] Table 6. Association of baseline covariates with DOX-related cardiotoxicity as analyzed by logistic regression for the validation cohort aOR, odds ratio; CI, confidence interval.

[0205] When considered as a continuous variable, univariate analyses of baseline plasma levels of each of the 6 proteins in the validation cohort revealed that patients with higher baseline levels for each of biglycan, cadherin-5, CA6, CD109, and TSP4 had greater odds of cardiotoxicity (reduced LVEF) compared to the patients with lower levels of each corresponding protein, while higher levels of cystatin-F were associated with lower odds of cardiotoxicity. These associations were statistically significant for biglycan, cadherin-5, CD109, and TSP4 (p<0.05) but the p- values were slightly over 0.05 for CA6 and cystatin-F (Table 7). After multivariable adjustment for baseline LVEF, triple negative type breast cancer, vitamin D deficiency, and hypertension, the associations remained significant (p<0.05) for biglycan, CD109, and TSP4 (Table 7).

[0206] Table 7. Association of baseline protein levels in plasma with DOX-related cardiotoxicity as analyzed by logistic regression for the validation cohort

[0207] > Univariate Analysis Multivariable Analysisc

[0208] Protein3OR (95% CI)b / '-value OR (95% CI) / ‘-value aCA6, carbonic anhydrase 6; TSP4, thrombospondin-4.bOR, odds ratio; CI, confidence interval.cMultivariable logistic regression analysis was adjusted for baseline LVEF, triple negative, vitamine D deficiency, and hypertension.

[0209] Example 7: Predictive Models and Validation of Biomarkers of Cardiotoxicity

[0210] To verify the 6 identified protein biomarkers for the prediction of chemotherapy-induced c rdio toxicity, predictive models based on algorithms of partial least squares discriminant analysis were generated. The models were trained and optimized using the normalized abundance data of the 6 proteins measured by SOMASCAN assays for the discovery cohort. Since logistic regression analysis found baseline LVEF was associated with treatment-related cardiotoxicity, it was included as an independent predictor or in combination with protein biomarkers during training. The models were applied to the OLINK data for predicting treatment-related cardiotoxicity in the validation cohort. When each protein biomarker or baseline LVEF was tested as an individual predictor, TSP4 achieved the highest accuracy of cardiotoxicity prediction (79.5% [95% CI, 64.7%-90.2%]) while cystatin-F showed the lowest accuracy of prediction of (63.6% [95% CI, 47.8%-77.6%]) (Table 8). Highest sensitivity was achieved by both biglycan and cystatin-F with a value of 80.0% (95% CI, 44.4-97.5%) while highest specificity was achieved by CD109 with a value of 88.2% (95% CI, 72.5-96.7%). Positive predictive value (PPV) was at a range of 36.4-54.5% and negative predictive value (NPV) was at a range of 81.1-91.3% among these individual biomarkers. As an individual predictor, a sensitivity of 50.0% (95% CI, 18.7-81.3%) and specificity of 79.4% (95% CI, 62.1-91.3%) was obtained for baseline LVEF (Table 8).

[0211] Table 8. Validation of biomarkers for predicting cardiotoxicity using models of partial least squares discriminant analysis CA6, carbonic anhydrase 6; TSP4, thrombospondin-4. Values in brackets are 95% confidence intervals (CI).

[0212] Models with multiple biomarkers resulted in better prediction in general as examples of 4, 5 or all 6 proteins in terms of sensitivity, specificity, PPV and NPV. Addition of baseline LVEF to the multiple protein panels increased the overall accuracy of prediction compared to the corresponding protein panels (Table 8). In general, sensitivity of 70.0-90.0%, specificity of 73.5-88.2%, PPV of 43.8-69.2%, NPV of 89.3-96.8%, and accuracy of 72.7-88.6% were achieved for multiple biomarker panels. A model based on the panel of biglycan, CA6, CD109, TSP4, and baseline LVEF, 9 out of 10 patients experienced cardiotoxicity were correctly predicted (sensitivity of 90.0% [95% CI, 55.5- 99.7%]), and 30 out of 34 patients without cardiotoxicity were correctly predicted (specificity of 88.2% [95% CI, 72.5-96.7%]). PPV of 69.2% (95% CI, 38.6-90.9%), NPV of 96.8% (95% CI, 83.3- 99.9%) and overall accuracy of cardiotoxicity prediction of 88.6% (95% CI, 75.4-96.2%) were obtained from this model. The model based on all the 6 proteins and baseline LVEF resulted in comparable accuracy of 86.4% (95% CI, 72.6-94.8%) in the prediction of cardiotoxicity with sensitivity of 80.0% (95% CI, 44.4-97.5%), specificity of 88.2% (95% CI, 72.5-96.7%), PPV of 66.7% (95% CI, 34.9-90.1%), NPV of 93.8% (95% CI, 79.2-99.2%) (Table 8).

[0213] Example 8: Multiplex biomarker assay development

[0214] This example describes multiplex assays for detection and validation of markers associated with a risk for chemotherapy-induced cardiotoxicity.

[0215] Materials

[0216] • MagPlex-C Microspheres, 1 mL (various regions, Luminex Corporation)

[0217] • Antibody coupling kit (Cat# 40-50016, Luminex Corporation)

[0218] • Protein biomarker antibody pairs including capture and detection antibodies (available from various commercial sources); detection antibodies are biotinylated

[0219] • Phycoerythrin (PE)-labeled goat anti-rabbit IgG detection antibody (for validation of antibody coupling)

[0220] • R-phycoerythrin conjugated streptavidin (SAPE) (ThermoFisher or other sources)

[0221] • Assay buffer: phosphate buffered saline (PBS), pH 7.4, 1% bovine serum albumin (BSA), and 0.05% sodium azide. One pack of PBS powder (Sigma- Aldrich Cat# P-3813), pH 7.4, is mixed with 10 grams of BSA (Sigma-Aldrich Cat# A7888) and 0.5 grams of sodium azide (Sigma- Aldrich Cat# S8032). Water is added to make 1 liter of the buffer.

[0222] Antibody coupling to colored magnetic beads

[0223] Magnetic beads (microspheres) from Luminex with different colors (regions) are coupled with capture antibodies against human protein biomarkers (e.g., BGN, CA6, CD109, TSP4, CDH5, and / or CST7) using the Luminex antibody coupling kit per manufacturer’s instruction. Each antibody is coupled with beads with a unique region. Stock beads are resuspended by vortexing for 10-20 seconds and 200 pL of beads (-2.5 million beads) from the stock vial are dispensed into each microcentrifuge tube (“reaction tube”). The magnetic beads are washed twice with 500 L of activation buffer. After washing, the activation buffer is removed and 480 pL of activation buffer is added to each tube. The reaction tubes are vortexed and 10 pL of N-hydroxysulfosuccinimide (Sulfo-NHS) is added to each reaction tube.

[0224] Activation buffer (250 pL) is added to a 10 mg vial of EDC (l-Ethyl-3-[3- dimethylaminopropyl]carbodiimide hydrochloride), the mixture is vortexed, and 10 pL of this EDC solution is added to each reaction tube. The reaction is incubated by rotating the tubes at 30 rpm for 20 minutes at room temperature in the dark. The reaction buffer is removed and the beads are washed 3 times with 500 pL of activation buffer. After washing, the activation buffer is removed.

[0225] Activation buffer and each capture antibody are added to the appropriate reaction tube for a total reaction volume of 500 pL. The volume of activation buffer is adjusted based on the volume of antibody to be added. The reaction is then incubated by rotating the tubes at 30 rpm for 2 hours at room temperature in the dark. The reaction buffer is removed and the beads are washed 3 times with 500 pL of wash buffer. After washing, the wash buffer is removed. One mL of wash buffer is added to each reaction tube. The tubes are vortexed and stored with beads at 2-8°C in the dark.

[0226] Validation of antibody coupling

[0227] Antibody-coupled magnetic beads with different regions can be mixed for the validation test if the antibody host species and the type of antibodies are the same (e.g., the capture antibodies are rabbit IgG antibodies against human CA6 and CDH5, but they are coupled with magnetic beads with different regions). The same anti -rabbit IgG detection antibody can be used for detection of both CA6 and CDH5 antibodies on the mixed beads.

[0228] Antibody-coupled beads are vortexed for 10-20 seconds. Each type of antibody -coupled beads is diluted to a final concentration of 50 beads / L in assay buffer as a working bead solution. Up to 4 different bead sets in the same mixture in a multiplex format can be prepared. At least 2 mL of the bead working solution is used for a mixture of 2 types of beads (e.g., CA6 and CDH5 antibody- coupled beads). A solution of phycoerythrin (PE)-labeled anti-species IgG detection antibody is prepared at 4 pg / mL in assay buffer (e.g., PE-labeled goat anti -rabbit IgG antibody). A 1 :2 dilution series of the detection antibody solution is prepared to a concentration of 0.0625 pg / mL (a series of 7 concentrations from 4 pg / mL-0.0625 pg / mL). 50 pL of the bead solution is aliquoted into two columns of wells of a 96-well plate (duplicate sets of 8 wells, 16 wells total). Assay buffer (50 pL), as a blank sample, is added into two wells. 50 pL of each of the diluted detection antibody solutions is added into the appropriate wells of the plate. The plate is covered with an aluminized foil seal and incubated for 30 minutes at room temperature on a plate shaker at 800 rpm. The liquid is removed from each well and the wells are washed twice with 100 pL of assay buffer. The beads in each well are resuspended in 100 pL of assay buffer. The plate is covered with an aluminized foil seal and incubated for 1 minute at room temperature on a plate shaker at 800 rpm. The plate is read on a Luminex xMAP instrument.

[0229] To validate this method, an assay was performed to confirm rabbit anti-human CA6 and CDH5 antibody coupling with magnetic beads. Each antibody was coupled to 2.5 million of beads in two different amounts (5 pg and 12.5 pg). As shown in FIGS. 6A-6B, antibody coupling was confirmed using a concentration series of anti-rabbit IgG-PE to detect fluorescence response of the beads.

[0230] Protein concentration measurement

[0231] The assay format is a capture sandwich immunoassay. Biotinylated detection antibodies against target protein biomarkers are used (e.g., CHD5). Alternatively, the assay is modified by using biotin-coupled aptamers that can specifically bind to the target protein biomarkers as detection reagents.

[0232] Antibody-coupled magnetic beads are resuspended by vortexing for 10-20 seconds. A working bead mixture is prepared by diluting the bead stocks to a final concentration of 50 beads of each set / pL in assay buffer. 50 pL of the working bead mixture is aliquoted into the appropriate wells of a 96-well plate. The beads are washed twice with 100 pL of assay buffer. After washing, the assay buffer is removed and 50 pL of assay buffer is added to each background well. Standard or sample (50 pL) is added to the appropriate wells. The plate is covered with an aluminized foil seal and incubated for 1 hour at room temperature on a plate shaker at 800 rpm. The liquid is removed and the plate is washed three times with 100 pL of assay buffer. After washing, the assay buffer is removed.

[0233] The biotinylated detection antibody is diluted to 2 pg / mL in assay buffer and 50 pL of the diluted detection antibody is added to each well. The plate is covered with an aluminized foil seal and incubated for 30 minutes at room temperature on a plate shaker at 800 rpm. The antibody solution is removed and the plate is washed three times with 100 pL of assay buffer. After washing, the assay buffer is removed.

[0234] R-phycoerythrin conjugated streptavidin (SAPE) reporter stock solution is diluted to 4 pg / mL in assay buffer and 50 pL of the diluted reporter solution is added to each well. The plate is covered with an aluminized foil seal and incubated for 30 minutes at room temperature on a plate shaker at 800 rpm. The liquid is removed and the plate is washed three times with 100 pL of assay buffer. After washing, the assay buffer is removed and 100 pL of assay buffer is added to each well of the plate. The plate is covered with an aluminized foil seal and incubated for 1 minute at room temperature on a plate shaker at 800 rpm. The plate is read on the Luminex analyzer using 50 pL of beads.

[0235] Following this method, the concentration of human CDH5 was measured using antibody- coupled magnetic beads. As shown in FIG. 7, CDH5 was detected in the range of 0.1-2000 ng / mL. It will be apparent that the precise details of the methods or compositions described may be varied or modified without departing from the spirit of the described aspects of the disclosure. We claim all such modifications and variations that fall within the scope and spirit of the claims below.

Claims

CLAIMS1. A method of identifying a subject diagnosed with cancer as being at risk of chemotherapy-induced cardiotoxicity, comprising: measuring the level of biglycan (BGN) protein or nucleic acid, carbonic anhydrase 6 (CA6) protein or nucleic acid, CD109 protein or nucleic acid, and thrombospondin-4 (TSP4) protein or nucleic acid in a biological sample obtained from the subject; comparing the level of BGN protein or nucleic acid, CA6 protein or nucleic acid, CD109 protein or nucleic acid, and TSP4 protein or nucleic acid to a control; and identifying the subject as being at risk of chemotherapy-induced cardiotoxicity if there is an increase in the level of BGN protein or nucleic acid, an increase in the level of CA6 protein or nucleic acid, an increase in the level of CD109 protein or nucleic acid and an increase in the level of TSP4 protein or nucleic acid compared to the control.

2. The method of claim 1, further comprising: measuring the level of cadherin-5 (CDH5) protein or nucleic acid and / or cystatin-F (CST7) protein or nucleic acid; comparing the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid to a control; and identifying the subject as being at risk of chemotherapy-induced cardiotoxicity if there is an increase in the level of CDH5 protein or nucleic acid and / or a decrease in the level of CST7 protein or nucleic acid compared to the control.

3. The method of claim 1 or claim 2, wherein the chemotherapy comprises an anthracycline.

4. The method of claim 3, wherein the anthracycline is doxorubicin, daunorubicin, epirubicin, idarubicin, mitoxantrone or valrubicin.

5. The method of claim 3 or claim 4, wherein the anthracycline is doxorubicin or a liposomal form thereof.

6. A method, comprising: obtaining a biological sample from a subject diagnosed with cancer; measuring the level of biglycan (BGN) protein or nucleic acid, carbonic anhydrase 6 (CA6) protein or nucleic acid, CD 109 protein or nucleic acid and thrombospondin-4 (TSP4) protein or nucleic acid in the biological sample; andcomparing the level of BGN protein or nucleic acid, CA6 protein or nucleic acid, CD109 protein or nucleic acid and TSP4 protein or nucleic acid to a control.

7. The method of claim 6, further comprising: measuring the level of cadherin-5 (CDH5) protein or nucleic acid and / or cystatin-F (CST7) protein or nucleic acid in the biological sample; and comparing the level of CDH5 protein or nucleic acid and / or CST7 protein or nucleic acid to a control.

8. The method of any one of claims 1-7, wherein the subject has not yet received treatment for the cancer.

9. The method of any one of claims 1-8, wherein the control is a reference standard or a historical control.

10. The method of any one of claims 1-9, wherein the biological sample comprises blood, serum or plasma.

11. The method of any one of claims 1-10, wherein the level of BGN protein, CA6 protein, CD109 protein, TSP4 protein, CDH5 protein and / or CST7 protein is measured by immunoassay, antibody array, aptamers that specifically bind each protein, or proximity extension assay.

12. The method of any one of claims 1-10, wherein the level of BGN nucleic acid, CA6 nucleic acid, CD 109 nucleic acid, TSP4 nucleic acid, CDH5 nucleic acid, and / or CST7 nucleic acid is measured by quantitative PCR (qPCR), digital PCR (dPCR), next generation sequencing (NGS), Northern blotting, nuclease protection assay, RNA-seq, or in situ hybridization.

13. The method of any one of claims 1-12, further comprising measuring left ventricular ejection fraction (LVEF) in the subject.

14. The method of any one of claims 1-13, wherein the subject has been diagnosed with a solid tumor, a leukemia, or a lymphoma.

15. The method of any one of claims 1-14, further comprising treating the subject with a chemotherapeutic agent, wherein the chemotherapeutic agent is not an anthracycline when the subject is identified as being at risk of chemotherapy-induced cytotoxicity.

16. The method of any one of claims 1-14, further comprising treating the subject with a chemotherapeutic agent, wherein the chemotherapeutic agent is not doxorubicin.

17. The method of any one of claims 1-16, further comprising treating the subject with an anti-cancer agent, wherein the anti-cancer agent is not a chemotherapeutic agent.

18. The method of claim 17, wherein the anti-cancer agent is an immunomodulator or a monoclonal antibody.

19. The method of any one of claims 1-18, further comprising treating the subject with radiation therapy, surgery, or both.

20. The method of any one of claims 1-19, further comprising treating the subject with one or more cardioprotective agents.

21. The method of claim 20, wherein the one or more cardioprotective agents comprise a beta blocker, an angiotensin receptor blocker, an angiotensin-converting-enzyme inhibitor (ACEI), a statin, an angiotensin receptor neprilysin inhibitor, a diuretic, an iron chelator, or a hyperpolarization- activated cyclic nucleotide-gated (HCN) channel blocker.

22. A kit, comprising:(i) a plurality of antibodies comprising at least two antibodies selected from the group consisting of an antibody specific for biglycan (BGN), an antibody specific for carbonic anhydrase 6 (CA6), an antibody specific for CD109, an antibody specific for thrombospondin-4 (TSP4), an antibody specific for cadherin-5 (CDH5) and an antibody specific for cystatin-F (CST7), and further comprising no more than five additional antibodies specific for other proteins;(ii) a plurality of aptamers comprising at least two aptamers selected from the group consisting of an aptamer specific for BGN, an aptamer specific for CA6, an aptamer specific for CD109, an aptamer specific for TSP4, an aptamer specific for CDH5 and an aptamer specific for CST7, and further comprising no more than five additional aptamers specific for other proteins;(iii) a plurality of antibody pairs comprising at least two antibody pairs selected from the group consisting of a pair of antibodies specific for BGN, a pair of antibodies specific for CA6, a pair of antibodies specific for CD109, a pair of antibodies specific for TSP4, a pair of antibodies specific for CDH5, and a pair of antibodies specific for CST7, and further comprising no more than five additional antibody pairs specific for other proteins, optionally wherein each pair of antibodies is linked to unique DNA sequences that hybridize only to each other, or one or both antibodies are linked to an affinity tag or a chromophore; or(iv) a plurality of antibody / aptamer pairs comprising at least two antibody / aptamer pairs selected from the group consisting of an antibody / aptamer pair specific for BGN, an antibody / aptamer pair specific for CA6, an antibody / aptamer pair specific for CD109, an antibody / aptamer pair specific for TSP4, an antibody / aptamer pair specific for CDH5, and an antibody / aptamer pair specific for CST7, and further comprising no more than five additional antibody / aptamer pairs specific for other proteins.

23. The kit of claim 22, comprising:(i) an antibody specific for BGN, an antibody specific for CA6, an antibody specific for CD109, and an antibody specific for TSP4;(ii) an aptamer specific for BGN, an aptamer specific for CA6, an aptamer specific for CD109 and an aptamer specific for TSP4; or(iii) a pair of antibodies specific for BGN, a pair of antibodies specific for CA6, a pair of antibodies specific for CD109, and a pair of antibodies specific for TSP4, optionally wherein each pair of antibodies is linked to unique DNA sequences that hybridize only to each other, or one or both antibodies are linked to an affinity tag or a chromophore.

24. The kit of claim 22, comprising:(i) an antibody specific for BGN, an antibody specific for CA6, an antibody specific for CD109, an antibody specific for TSP4, an antibody specific for CDH5, and an antibody specific for TSP4;(ii) an aptamer specific for BGN, an aptamer specific for CA6, an aptamer specific for CD109, an aptamer specific for TSP4, an aptamer specific for CDH5, and an aptamer specific for TSP4; or(iii) a pair of antibodies specific for BGN, a pair of antibodies specific for CA6, a pair of antibodies specific for CD109, a pair of antibodies specific for TSP4, a pair of antibodies specific for CDH5, and a pair of antibodies specific for TSP4, optionally wherein each pair of antibodies is linked to unique DNA sequences that hybridize only to each other, or one or both antibodies are linked to an affinity tag or a chromophore.

25. The kit of any one of claims 22(i), 23(i) and 24(1), wherein the antibodies are immobilized on a solid support.

26. The kit of claim 25, wherein the solid support comprises a slide, a membrane, or beads.

27. The kit of any one of claims 22-26, further comprising recombinant BGN protein, recombinant CA6 protein, recombinant CD 109 protein, recombinant TSP4 protein, recombinantCDH5 protein and / or recombinant CST7 protein.

28. The kit of any one of claims 22-27, further comprising one or more control proteins, buffer(s), reagent(s), multi-well plate(s), tube(s), mixing bottles(s), a qPCR template, primers, nucleic acid probes, DNA polymerase, dNTPs, loading control dye(s), detection reagent(s), detection enzyme(s), chromophore-linked avidin or streptavidin or other affinity tags, substrate(s), adhesive plate sealer(s), DNA microarray slide! s), calibrator(s), quality control sample(s), sample matrix, next- generation sequencing (NGS) adapters, indexing primers, ligase, dA-tailing enzyme(s), DNA cleanup beads, instructions, labelled secondary antibodies, or any combination thereof.

29. The method of any one of claims 1-20, further comprising using a computational algorithm to identify the subject as being at risk of chemotherapy-induced cardiotoxicity.

Images