Biomarkers of CBL-b inhibition

Abstract

Disclosed herein are methods for predicting responsiveness to an immune checkpoint inhibitor, such as a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi), using a biomarker signature. Additionally disclosed herein are methods for determining a biomarker signature indicative of responsiveness to an immune checkpoint inhibitor, such as a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi). Additionally disclosed herein are methods for determining effects of a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi) administered to a subject.

Description

BIOMARKERS OF CBL-B INHIBITIONCross-Reference to Related Applications

[0001] This application claims the benefit of and priority to U.S. Provisional Application No. 63 / 662,552, filed on June 21, 2024; to U.S. Provisional Application No. 63 / 665,432, filed on June 28, 2024 and to U.S. Provisional Application No. 63 / 718,105, filed on November 8, 2024, the disclosures of each of which are hereby incorporated by reference in their entireties for all purposes.Background

[0002] Immune checkpoint inhibitors have shown promise as cancer immunotherapies. However, not all patients respond to current checkpoint inhibitors. Methods to determine patient responses to determine responders and non-responders to novel checkpoint inhibitors are needed to evaluate efficacy of therapeutics and to identify individuals who may benefit from treatment with checkpoint inhibitors. Therefore, methods for identifying biomarkers predictive of a response to checkpoint inhibitor treatment and methods for utilizing identified biomarkers to evaluate patient response to checkpoint inhibitors are needed. Methods for determining the effects of an immune checkpoint inhibitor administered to a subject, determining the immune checkpoint inhibitor dose or therapy or monitoring a cancer subject’s response to an immune checkpoint inhibitor are also needed.Summary

[0003] Disclosed herein are methods for predicting responsiveness to an immune checkpoint inhibitor using a biomarker signature. Additionally disclosed herein are methods for determining a biomarker signature indicative of responsiveness to an immune checkpoint inhibitor. Immune check point inhibitors disclosed herein include a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi). In various embodiments, methods disclosed herein are useful for monitoring a cancer subject (e.g., a subject diagnosed with a cancer, or a subject suspected of having cancer). For example, methods disclosed herein are useful for monitoring a cancer subject’s response to a CBL-Bi therapeutic. In various embodiments, if the cancer subject is predicted to be a responder to a CBL-Bi therapeutic, the treatment regimen for the cancer subject can be maintained. In various embodiments, if the cancer subject is predicted to be a non-responder to a CBL-Bi therapeutic, the treatment regimen forthe cancer subject can be altered (e.g., to include a different checkpoint inhibitor and / or include different therapies).

[0004] The disclosure provides, in one aspect, a method for determining responsiveness to an immune checkpoint inhibitor in a subject, the method comprising: a. obtaining or having obtained at least one dataset comprising expression levels of a plurality of biomarkers, wherein the plurality of biomarkers comprise four or more of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1 ; and b. generating a prediction of responsiveness to an immune checkpoint inhibitor by applying a predictive model to the expression levels of the plurality of biomarkers.

[0005] In some embodiments, the immune checkpoint inhibitor is Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi). In some embodiments, the plurality of biomarkers comprise five or more of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1. In some embodiments, the plurality of biomarkers comprise six or more of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1. In some embodiments, the plurality of biomarkers comprise seven or more of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1. In some embodiments, the plurality of biomarkers comprise each of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1. In some embodiments, the plurality of biomarkers are identified by analyzing data of a plurality of experimental studies, wherein the plurality of experimental studies comprises two or more, three or more, or four or more experimental studies. In some embodiments, the data of the plurality of experimental studies comprises gene expression data determined by one or more of multiplex nucleic acid hybridization analysis or RNA sequencing analysis. In some embodiments, the data of the plurality of experimental studies comprises gene expression data determined by multiplex nucleic acid hybridization analysis of murine cells. In some embodiments, the murine cells comprise tumor cells or PBMCs. In some embodiments, the murine cells were previously exposed to an immune checkpoint inhibitor. In some embodiments, the data of the plurality of experimental studies comprises gene expression data determined by multiplex nucleic acid hybridization analysis of human cells. In some embodiments, the human cells comprise PBMCs. In some embodiments, the human cells were previously exposed to an immune checkpoint inhibitor. In some embodiments, the plurality of experimental studies comprises a multiplex nucleic acid hybridization analysis of a plurality of tumor cells isolated from mice of at least five, at least six, at least seven, at least eight, at least nine, or at least ten murine tumor models. In some embodiments, the plurality of experimental studies comprises a multiplex nucleic acid hybridization analysis of PBMCs isolated from mice. In someembodiments, the mice were previously exposed to an immune checkpoint inhibitor. In some embodiments, the plurality of experimental studies comprises a multiplex nucleic acid hybridization analysis of human PBMCs. In some embodiments, the plurality of experimental studies comprises a RNA sequencing analysis of human PBMCs. In some embodiments, the human PBMCs were previously exposed to an immune checkpoint inhibitor. In some embodiments, the plurality of experimental studies comprise two or more, three or more, or four of: a. a multiplex nucleic acid hybridization analysis of a plurality of cells isolated from a plurality of murine tumor models wherein mice are treated with an immune checkpoint inhibitor; b. a multiplex nucleic acid hybridization analysis of PBMCs isolated from mice treated with an immune checkpoint inhibitor for pharmacodynamic studies; c. a multiplex nucleic acid hybridization analysis of human PBMCs treated with an immune checkpoint inhibitor; and d. a RNA sequencing analysis of human PBMCs treated with an immune checkpoint inhibitor.

[0006] In some embodiments, the murine tumor models are characterized as responders or non responders to treatment with an immune checkpoint inhibitor. In some embodiments, responders to treatment with an immune checkpoint inhibitor are characterized by tumor growth inhibition of >30%. In some embodiments, the plurality of cells is isolated from ten murine tumor models that are characterized as responders or as non-responders. In some embodiments, the method further comprises: administering the immune checkpoint inhibitor to the subject if the generated prediction of responsiveness indicates the subject is a responder to the immune checkpoint inhibitor.

[0007] In some embodiments, the method further comprises: modifying a treatment for the subject if the generated prediction of responsiveness indicates the subject is a non-responder to the immune checkpoint inhibitor.

[0008] The disclosure provides, in another aspect, a method for identifying a plurality of biomarkers for determining responsiveness to an immune checkpoint inhibitor in a subject, the method comprising: a. obtaining or having obtained a dataset from two or more experimental studies, wherein each experimental study identifies differentially expressed genes wherein the dataset comprises:i. gene expression data determined by multiplex nucleic acid hybridization analysis of nucleic acids isolated from mice of a plurality of murine tumor models wherein the mice are treated with an immune checkpoint inhibitor; ii. gene expression data determined by multiplex nucleic acid hybridization analysis of PBMCs isolated from mice treated with an immune checkpoint inhibitor for pharmacodynamic studies; iii. gene expression data determined by multiplex nucleic acid hybridization (NanoString) analysis of human PBMCs treated with an immune checkpoint inhibitor; and iv. gene expression data determined by RNA sequencing analysis of human PBMCs treated with an immune checkpoint inhibitor, and b. comparing differentially expressed genes across each of the two or more experimental studies, and wherein comparing differentially expressed genes across each of the four or more experimental studies further comprises identifying one or more commonly differentially expressed genes that are differentially expressed in at least one of the two or more experimental studies to form a plurality of biomarkers.

[0009] In some embodiments, the at least five, at least six, at least seven, at least eight, at least nine, or at least ten murine tumor models are characterized as responders or non responders to treatment with an immune checkpoint inhibitor, wherein responsiveness to treatment with an immune checkpoint inhibitor is characterized as tumor growth inhibition of >30%. In some embodiments, the method further comprises determining genes that have a significant differences in log2 fold change when comparing gene expression data, wherein log2 fold changes are calculated for each model by comparing gene expression data from animals treated with an immune checkpoint inhibitor to animals treated with vehicle control. In some embodiments, the method further comprises determining genes in mouse models with detected tumor growth inhibition (TGI) that have a log2 fold change that correlates with tumor growth inhibition. In some embodiments, the method further comprises determining genes upregulated in at least three of the responders. In some embodiments, the gene expression dataset determined by multiplex nucleic acid hybridization analysis of PBMCs isolated from mice treated with an immune checkpoint inhibitor for pharmacodynamic studies comprises expression data of upregulated genes collected from mice treated with 6 mpk of CBL-Bi. In some embodiments, the expression data of upregulated genes is compared to each of (i) gene expression data determined by multiplex nucleic acid hybridizationanalysis, (iii) gene expression data determined by multiplex nucleic acid hybridization analysis of human PBMCs, and (iv) gene expression data determined by RNA sequencing analysis of human PBMCs to determine a set of commonly differentially expressed genes. In some embodiments, the set of commonly differentially expressed genes comprises at least four, at least five, at least six, at least seven, or at least eight biomarkers. In some embodiments, the at least four biomarkers comprise CXCL9, IFIT3, OAS2, and STAT1. In some embodiments, the at least four biomarkers comprise CXCL9, GBP5, IRF1 and CD274. In some embodiments, the at least four biomarkers comprise CXCL9, GBP5, IRF1, CD274, STAT1, PSMB9, IFIT3, and IFNG. In some embodiments, the determined set of commonly differentially expressed genes form a gene signature, wherein at least 85% of the selected biomarkers of the gene signature are differentially expressed in at least two experimental studies. In some embodiments, the determined set of commonly differentially expressed genes form a gene signature, wherein at least 75% of the selected biomarkers of the gene signature are differentially expressed in at least three experimental studies.

[0010] Also disclosed herein are methods for determining effects of a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi) administered to a subject. Additionally disclosed herein is a method for determining a therapy for a subject, such as determining a dose of a CBL-Bi, or determining a therapy other than a CBL-Bi for the subject. In various embodiments, methods disclosed herein are useful for monitoring a cancer subject (e.g., a subject diagnosed with a cancer, or a subject suspected of having cancer). For example, methods disclosed herein are useful for monitoring a cancer subject’s response to a CBL-Bi therapeutic.

[0011] For example, a method disclosed herein is useful for determining effects of a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi), the method comprising: obtaining statuses of a plurality of biomarkers of a sample obtained from a subject that was previously administered the CBL-Bi, wherein the plurality of biomarkers comprise Notchl and IGF1R; using the obtained statuses, determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples; and determining the effects of the CBL-Bi based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples. In some embodiments, the statuses of the plurality of biomarkers are measured from the sample comprising peripheral blood mononuclear cells (PBMCs). In some embodiments, the PBMCs comprise CD4+ and / or CD 8+ T cells. In some embodiments, the method further comprises: obtaining the sample obtained from the subject; and providing one or both of anti-CD3antibodies and anti-CD28 antibodies to stimulate the PBMCs. In some embodiments, PBMCs are stimulated for at least 24 hours. In some embodiments, determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that one of Notch 1 and IGF1R are expressed at a higher level in the sample than in the one or more reference samples. In some embodiments, determining the effects of the CBL-Bi based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that the CBL-Bi is effective in response to determining that one of Notchl and IGF1R are expressed at a higher level in the sample. In some embodiments, determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that neither of Notchl or IGF1R are expressed at a higher level in the sample than in the one or more reference samples. In some embodiments, determining the effects of the CBL-Bi based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that the CBL-Bi is not effective in response to determining that neither of Notchl or IGF1R are expressed at a higher level in the sample. In some embodiments, the statuses of the plurality of biomarkers comprise expression levels of at least a subset of the plurality of biomarkers. In some embodiments, the plurality of biomarkers further comprise one or more of ZAP70, CD3E, LAT, and PLCG1. In some embodiments, the statuses of the plurality of biomarkers comprises phosphorylation statuses of one or more amino acids of one or more of ZAP70, CD3E, LAT, and PLCG 1 . In some embodiments, the one or more amino acids comprise one or more of tyrosine, serine, and threonine.

[0012] In another example of a method disclosed herein, the method comprises: obtaining statuses of a plurality of biomarkers of a sample obtained from a subject that was previously administered the CBL-Bi, wherein the plurality of biomarkers comprise Notchl and IGF1R; using the obtained statuses, determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples; and selecting a therapy for the subject based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples. In some embodiments, the statuses of the plurality of biomarkers are measured from the sample comprising peripheral blood mononuclear cells (PBMCs). In some embodiments, the PBMCs comprise CD4+ and / or CD8+ T cells. In some embodiments, the method further comprises: obtaining the sample obtained from the subject; and providing one or both of anti-CD3 antibodies and anti-CD28 antibodies to stimulate the PBMCs. In some embodiments, the PBMCs are stimulated for at least 24 hours. In some embodiments, determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that one of Notchl and IGF1R are expressed at a higher level in the sample than in the one or more reference samples. In some embodiments, selecting the therapy for the subject based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that the CBL-Bi is effective in response to determining that one of Notchl and IGF1R are expressed at a higher level in the sample. In some embodiments, determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that neither of Notchl or IGF1R are expressed at a higher level in the sample than in the one or more reference samples. In some embodiments, selecting the therapy for the subject based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that the CBL-Bi is not effective in response to determining that neither of Notchl or IGF1R are expressed at a higher level in the sample. In some embodiments, the statuses of the plurality of biomarkers comprise expression levels of at least a subset of the plurality of biomarkers. In some embodiments, the plurality of biomarkers further comprise one or more of ZAP70, CD3E, LAT, and PLCG1. In some embodiments, the statuses of the plurality of biomarkers comprises phosphorylation statuses of one or more amino acids of one or more of ZAP70, CD3E, LAT, and PLCG1 . In some embodiments, the one or more amino acids comprise one or more of tyrosine, serine, and threonine. In some embodiments, selecting the therapy for the subject comprises selecting a dose of CBL-Bi that differs from a dose of CBL-Bi previously administered to the subject.Brief Description of the Drawings

[0013] These and other features, aspects, and advantages of the present invention will become better understood with regard to the following description and accompanying drawings.

[0014] Figure (FIG.) 1 A depicts an overview of an environment for determining a responder prediction to treatment with checkpoint inhibitors in an individual via an activity prediction system, in accordance with an embodiment.

[0015] FIG. IB is an example block diagram of the cancer prediction system, in accordance with an embodiment.

[0016] FIG. 1C depicts an example set of training data, in accordance with an embodiment.

[0017] FIG. 2 depicts an example flow diagram for predicting responsiveness to an immune checkpoint inhibitor for a subject, in accordance with an embodiment.

[0018] FIG. 3 depicts a schematic of the selection of biomarkers by identifying commonly differently expressed genes across two or more experiments, in accordance with an embodiment.

[0019] FIG. 4 illustrates an example computer for implementing the entities and processes shown in FIGS. 1A, IB, 1C, 2, and 3.

[0020] FIG. 5 depicts a graph of box plots of expression of score (gene signature) in responders (indicated by “R”) and non-responders (indicated by “NR”) suffering from kidney renal clear cell carcinoma (KIRC), skin cutaneous melanoma (SKCM), bladder cancer (BLCA), or stomach adenocarcinoma (STAD). A Wilcoxon test was used to determine significant differences in expression between responders and non-responders. P- values are indicated above each responder and non-responder pair.

[0021] FIG. 6 depicts a schematic of exemplary experimental studies used to generate the gene signature of the disclosure.

[0022] FIG. 7 depicts a graph of the correlation between tumor growth inhibition (TGI) and Log2 fold change (FC) in expression of IFI44L and RORA. Each circle or triangle represents data from individual mouse models indicated by the key.

[0023] FIG. 8 depicts a graph of box plots of Log2 FC in gene expression of CXCL10 and CXCL9 of responders (R) and non-responders (NR). Each circle or triangle represents data from individual mouse models indicated by the key. The percent of tumor growth inhibition in responder models is indicated in the table above the graph.

[0024] FIG. 9 depicts a Venn diagram of genes upregulated across the five responder models.

[0025] FIG. 10 depicts a Venn diagram of genes upregulated across gene sets 1, 2, 3, and 4 as described in the Examples.

[0026] FIG. 1 1A depicts a schematic of mouse and human models treated with a CBL-Bi and analyzed for the gene expression signature of the disclosure.

[0027] FIG. 1 IB depicts a bar graph of the mean gene signature (CBL-B signature) increase (in %) at Day 1 detected in PBMCs from an in vivo mouse model after treatment with a CBL- Bi.

[0028] FIG. 11C depicts a graph of box plots showing the acute maximum effect of the gene signature (CBL-B signature) changes (%) in human PBMCs from patients treated with the indicated doses a CBL-Bi on Day 1.

[0029] FIG. 12A depicts a graph of box plots of the Total tumor infiltrating lymphocyte (TIL) Score in screening or archival samples from patents showing no benefit or clinical benefit.

[0030] FIG. 12B depicts a graph of box plots of the gene signature (CBL-B signature) expression (geometric mean) on treatment samples from Day 15 from patients showing no benefit or clinical benefit.Detailed descriptionI. Definitions

[0031] Terms used in the claims and specification are defined as set forth below unless otherwise specified. Any terms not directly defined herein shall be understood to have the meanings commonly associated with them as understood within the art of the invention. Certain terms are discussed herein to provide additional guidance to the practitioner in describing the compositions, devices, methods and the like of aspects of the invention, and how to make or use them. It will be appreciated that the same thing may be said in more than one way. Consequently, alternative language and synonyms may be used for any one or more of the terms discussed herein. No signif icance is to be placed upon whether or not a term is elaborated or discussed herein. Some synonyms or substitutable methods, materials and the like are provided. Recital of one or a few synonyms or equivalents does not exclude use of other synonyms or equivalents, unless it is explicitly stated. Use of examples, including examples of terms, is for illustrative purposes only and does not limit the scope and meaning of the aspects of the invention herein.

[0032] The terms “subject”, “patient”, and “individual are used interchangeably and encompass a cell, tissue, or organism, human or non-human, whether in vivo, ex vivo, or in vitro, male or female.

[0033] The term “mammal” encompasses both humans and non-humans and includes but is not limited to humans, non-human primates, canines, felines, murines, bovines, equines, and porcines.

[0034] The term “sample” can include a single cell or multiple cells or fragments of cells or an aliquot of body fluid, such as a blood sample, taken from a subject, by means includingvenipuncture, excretion, ejaculation, massage, biopsy, needle aspirate, lavage sample, scraping, surgical incision, or intervention or other means known in the art. Examples of an aliquot of body fluid include amniotic fluid, aqueous humor, bile, lymph, breast milk, interstitial fluid, blood, blood plasma, cerumen (earwax), Cowper’s fluid (pre-ejaculatory fluid), chyle, chyme, female ejaculate, menses, mucus, saliva, urine, vomit, tears, vaginal lubrication, sweat, serum, semen, sebum, pus, pleural fluid, cerebrospinal fluid, synovial fluid, intracellular fluid, and vitreous humour.

[0035] The terms “marker,” “markers,” “biomarker,” and “biomarkers” encompass, without limitation, lipids, lipoproteins, proteins, cytokines, chemokines, growth factors, peptides, nucleic acids, genes, and oligonucleotides, together with their related complexes, metabolites, mutations, variants, polymorphisms, modifications, fragments, subunits, degradation products, elements, and other analytes or sample-derived measures. A marker can also include mutated proteins, mutated nucleic acids, variations in copy numbers, and / or transcript variants, in circumstances in which such mutations, variations in copy number and / or transcript variants are useful for generating a predictive model, or are useful in predictive models developed using related markers (e.g., non-mutated versions of the proteins or nucleic acids, alternative transcripts, etc.).

[0036] The term "antibody" is used in the broadest sense and specifically covers monoclonal antibodies (including full length monoclonal antibodies), polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), and antibody fragments that are antigen-binding so long as they exhibit the desired biological activity, e.g., an antibody or an antigen-binding fragment thereof.

[0037] "Antibody fragment", and all grammatical variants thereof, as used herein are defined as a portion of an intact antibody comprising the antigen binding site or variable region of the intact antibody, wherein the portion is free of the constant heavy chain domains (i.e. CH2, CH3, and CH4, depending on antibody isotype) of the Fc region of the intact antibody. Examples of antibody fragments include Fab, Fab', Fab'-SH, F(ab')2, and Fv fragments; diabodies; any antibody fragment that is a polypeptide having a primary structure consisting of one uninterrupted sequence of contiguous amino acid residues (referred to herein as a "single-chain antibody fragment" or "single chain polypeptide").

[0038] The term “responder” refers to an animal or individual who displays a response, responsiveness, or is responsive to treatment with a checkpoint inhibitor. The term “nonresponder” refers to an animal or human who displays a lack of response, non-responsiveness or is non-responsive to treatment with a checkpoint inhibitor. The responder or non-respondercan be a animal or individual suffering from a cancer. Responsiveness or non-responsiveness can be determined through evaluation of experimentally or clinically relevant measurements of response to a cancer treatment (e.g., tumor growth inhibition).

[0039] The term “tumor growth inhibition” can also be referred to as “tumor reduction” or and refers to a reduction in the size or quantity of tumors in an animal or individual suffering from a cancerous tumor.

[0040] The phrases “gene signature” and “biomarker signature” are used interchangeably, and hereby refer to a set biomarkers that are informative for determining responsiveness or non-responsiveness to an immune checkpoint inhibitor. For example, expression levels of the set of biomarkers in the biomarker panel can be informative for determining if subjects are responders or non-responders to treatment with an immune checkpoint inhibitor, e.g., determining if a subject is a responder to treatment with a Casitas B-lineage lymphoma protooncogene b inhibitor (CBL-Bi). In various embodiments, a biomarker panel can include three, four, five, six, seven, eight, nine, ten eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty, twenty one, twenty two, twenty three, twenty four, twenty five biomarkers, fifty biomarkers, or fifty three biomarkers. In some embodiments, a biomarker panel includes eight biomarkers.

[0041] As used herein, a gene signature may encompass any gene or genes, protein or proteins, or epigenetic element(s) whose expression profile or whose occurrence is associated with a responder or non-responder. Increased or decreased expression or activity or prevalence may be compared between different treatments and experiments to characterize or identify responders and non-responders. A gene signature as used herein, may thus refer to any set of up- and down-regulated genes derived from one or more sets of experimental gene expression data. For example, a gene signature may comprise a list of genes differentially expressed in a distinction of interest. It is to be understood that also when referring to proteins (e.g. differentially expressed proteins), such may fall within the definition of “gene” signature.

[0042] The term “obtaining or having obtained a dataset from two or more experimental studies” encompasses obtaining a set of data determined from at least two experimental studies. Obtaining a dataset encompasses obtaining a sample and processing the sample to experimentally determine the data. The phrase also encompasses receiving a set of data, e.g., from a third party that has processed the sample to experimentally determine the dataset. Additionally, the phrase encompasses mining data from at least one database or at least onepublication or a combination of databases and publications. A dataset can be obtained by one of skill in the art via a variety of known ways including stored on a storage memory.

[0043] The phrase “prediction of responsiveness to an immune checkpoint inhibitor” generally refers to a prediction of responsiveness for a specific patient to the immune checkpoint inhibitor. In various embodiments, the phrase “prediction of responsiveness to an immune checkpoint inhibitor” refers to a prediction of responsiveness for a patient who was previously administered the immune checkpoint inhibitor. Thus, methods disclosed herein can be useful for determining whether a patient responded or did not respond to the previously administered immune checkpoint inhibitor. In various embodiments, the phrase “prediction of responsiveness to an immune checkpoint inhibitor” refers to a prediction of future responsiveness of the patient to the immune checkpoint inhibitor. In such embodiments, the patient may not yet have received the immune checkpoint inhibitor and therefore, methods disclosed herein can be useful for preemptively distinguishing patients that are likely to respond to the immune checkpoint inhibitor and other patients that are unlikely to respond to the immune checkpoint inhibitor.

[0044] It must be noted that, as used in the specification, the singular forms “a,” “an” and “the” include plural referents unless the context clearly dictates otherwise.II. System Environment Overview

[0045] FIG. 1 A depicts an overview of a system environment 100 for generating a responder prediction for individuals who received, or are selected to receive, treatment with checkpoint inhibitors, in accordance with an embodiment. The system environment 100 provides context in order to introduce a marker quantification assay 120 and a cancer prediction system 130.

[0046] Referring first to the subject 110, in various embodiments, the subject 110 may have been diagnosed with a cancer. In various embodiments, the subject 110 may be suspected of having a cancer. In various embodiments, the subject 110 may be healthy and not diagnosed with a cancer. In various embodiments, the subject 110 was previously diagnosed with a cancer and provided a therapy. In various embodiments, the subject 110 was previously diagnosed with a cancer and provided a checkpoint inhibitor. In particular embodiments, the subject 110 was previously diagnosed with a cancer and provided a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi).

[0047] In various embodiments, a test sample is obtained from the subject 110. The sample can be obtained by the individual or by a third party, e.g., a medical professional. Examples of medical professionals include physicians, emergency medical technicians, nurses, firstresponders, psychologists, phlebotomist, medical physics personnel, nurse practitioners, surgeons, dentists, and any other obvious medical professional as would be known to one skilled in the art.

[0048] The test sample is tested to determine values of one or more markers by performing the marker quantification assay 120. The marker quantification assay 120 determines quantitative expression values of one or more biomarkers from the test sample. The marker quantification assay 120 may be an immunoassay, and more specifically, a multi-plex immunoassay, examples of which are described in further detail below. The expression levels of various biomarkers can be obtained in a single run using a single test sample obtained from the subject 110. The quantified expression values of the biomarkers are provided to the cancer prediction system 130.

[0049] Generally, the activity prediction system 130 includes one or more computers, embodied as a computer system 400 as discussed below with respect to FIG. 4. Therefore, in various embodiments, the steps described in reference to the cancer prediction system 130 are performed in silico. The cancer prediction system 130 analyzes the received biomarker expression values from the marker quantification assay 120 to generate a responder prediction 140 in the subject 1 10.

[0050] In various embodiments, the marker quantification assay 120 and the cancer prediction system 130 can be employed by different parties. For example, a first party performs the marker quantification assay 120 which then provides the results to a second party which implements the cancer prediction system 130. For example, the first party may be a clinical laboratory that obtains test samples from subjects 110 and performs the assay 120 on the test samples. The second party receives the expression values of biomarkers resulting from the performed assay 120 and analyzes the expression values using the cancer prediction system 130.

[0051] Reference is now made to FIG. IB which depicts a block diagram illustrating the computer logic components of the cancer prediction system 130, in accordance with an embodiment. Specifically, the cancer prediction system 130 may include a model training module 150, a model deployment module 160, and a training data store 170.

[0052] Each of the components of the cancer prediction system 130 is hereafter described in reference to two phases: 1) a training phase and 2) a deployment phase. More specifically, the training phase refers to the building and training of one or more predictive models based on training data that includes quantitative expression values of biomarkers obtained from individuals with a known classification (e.g., a cancer classification, a non-cancerclassification, a responder classification, or a non-responder classification). For example, a responder classification be assigned to an individual that is responsive to a cancer treatment (e.g., tumor reduction after treatment with a check-point inhibitor). A non-responder classification can be assigned to an individual that is non-responsive to cancer treatment (e.g., lack of tumor reduction after treatment with a check-point inhibitor). Therefore, in various embodiments, the predictive models are trained to distinguish responders and non-responders to a cancer treatment (e.g., a checkpoint inhibitor, such as a CBL-Bi) in a subject based on quantitative biomarker expression values. During the deployment phase, a predictive model is applied to quantitative biomarker expression values from a test sample obtained from a subject of interest to generate a responder prediction for the subject of interest.

[0053] In some embodiments, the components of the cancer prediction system 130 are applied during one of the training phase and the deployment phase. For example, the model training module 150 and training data store 170 (indicated by the dashed lines in FIG. IB) are applied during the training phase whereas the model deployment module 160 is applied during the deployment phase. In various embodiments, the training phase and the deployment phase can be performed to enable continuously trained models. For example, the model training module 150 can train a model that the model deployment module 160 can subsequently deploy. The same model can undergo additional training by the model training module 150 (e.g., continuously trained using, for example, new training data that is obtained). Therefore, as the model is continuously trained, it can exhibit improved prediction capacity when analyzing samples during deployment.

[0054] In various embodiments, the components of the cancer prediction system 130 can be performed by different parties depending on whether the components are applied during the training phase or the deployment phase. In such scenarios, the training and deployment of the predictive model are performed by different parties. For example, the model training module 150 and training data store 170 applied during the training phase can be employed by a first party (e.g., to train a predictive model) and the model deployment module 160 applied during the deployment phase can be performed by a second party (e.g., to deploy the predictive model).

[0055] Returning to FIG. 1 A, the cancer prediction system 130 outputs a responder prediction 140. The responder prediction 140 refers to a classification for the subject 110. In various embodiments, the responder prediction 140 is one of a responder, non-responder, individual that has a presence of cancer, individual that has an absence of cancer, individual that is healthy, or individual in a state of cancer remission. In particular embodiments, theresponder prediction 140 is one of a responder classification (e.g., to a CBL-Bi) or a nonresponder classification (e.g., to a CBL-Bi). Thus, the responder prediction 140 can be informative for guiding a therapy to be provided to the subject 110. In various embodiments, if the responder prediction 140 is a responder classification, which indicates that the subject 110 a responder to a checkpoint inhibitor (e.g., to a CBL-Bi), then the subject 110 can be further provided a checkpoint inhibitor (e.g., a CBL-Bi). In various embodiments, if the responder prediction 140 is a non-responder classification, which indicates that the subject 110 a non-responder to a checkpoint inhibitor (e.g., to a CBL-Bi), then the subject 110 can be provided a different therapy (e.g., a therapy other than a checkpoint inhibitor).III. Predictive modelIII. A. Training a Predictive model

[0056] During the training phase, the model training module 150 trains one or more predictive models using training data comprising expression values of biomarkers. Referring to FIG. IB, the training data may be stored in the training data store 170. In various embodiments, the cancer prediction system 130 generates the training data comprising expression values of biomarkers by analyzing biomarker expression values in test samples. In various embodiments, the cancer prediction system 130 obtains the training data comprising expression values of biomarkers from a third party. The third party may have analyzed test samples to determine the biomarker expression values.

[0057] In various embodiments, the training data comprising expression values of biomarkers are derived from clinical subjects. For example, the training data can be expression values of biomarkers that were measured from test samples obtained from clinical subjects. Examples of expression values of biomarkers derived from clinical subjects include biomarker expression values obtained through clinical studies such as those evaluating tumor response as in FIG. 5 described in the Examples section.

[0058] In various embodiments, the training data includes reference ground truths that identify individuals that are responders, non-responders, individuals that have a presence of cancer, individuals that have an absence of cancer, individuals that are healthy, individuals in a state of cancer remission. In particular embodiments, the training data includes reference ground truths that indicate whether an individual responded or did not respond to a treatment with a checkpoint inhibitor (e.g., a CBL-Bi). A responder to a treatment with a checkpoint inhibitor (e.g., a CBL-Bi) can be characterized as an individual that exhibits a reduction in tumor size after having been administered the checkpoint inhibitor (e.g., a CBL-Bi).

[0059] Reference is made to FIG. 1C, which depicts an example set of training data 190, in accordance with an embodiment. As shown in FIG. 1C, the training data 190 includes data corresponding to multiple individuals (e.g., column 1 depicting individual 1, 2, 3, 4...). For each individual, the training data 190 includes quantitative expression values (e.g., Al, Bl, A2, B2, etc.) for different biomarkers obtained from the corresponding individual. In some embodiments, the quantitative expression values are determined by the marker quantification assay 120 shown in FIG. 1. Although FIG. 1C depicts 4 individuals and 2 different markers (marker A and marker B), the training data 190 may include tens, hundreds, or thousands of individuals as well as tens, hundreds, or thousands of markers.

[0060] As shown in FIG. 1C, a first training example (e.g., first row) of the training data refers to individual 1 and corresponding quantitative expression values of marker A (e.g., Al) and the quantitative expression value of marker B (e.g., Bl). Similarly, the second training example (e.g., second row) of the training data refers to individual 2 and corresponding quantitative expression values of marker A (e.g., A2) and the quantitative expression value of marker B (e.g., B2). Individuals 3 and 4 have corresponding marker values as shown in FIG. 1C.

[0061] As shown in FIG. 1C, the training data 190 further includes a reference ground truth (“Indication” column) that identifies whether the corresponding individual is a responder or non-responder to a cancer treatment, such as a checkpoint inhibitor (e.g., a CBL-Bi). As an example, each indication may be an indication of treatment response in the patient. For example, referring to the first training example (e.g., first row), a “Responder” indication can reflect a presence of responsiveness to a checkpoint inhibitor in individual 1. For example, an image of individual 1 may have revealed a tumor shrinkage. Similarly, an indication of a “Non-responder” (e.g., individual 3 or individual 4) reflects a lack of responsiveness to a checkpoint inhibitor in the corresponding individual. For example, images of individual 3 and 4 may have revealed a lack of tumor shrinkage.

[0062] In some embodiments, the model training module 150 retrieves the training data from the training data store 170 and randomly partitions the training data into a training set and a test set. As an example, 80% of the training data may be partitioned into the training set and the other 20% can be partitioned into the test set. Other proportions of training set and test set may be implemented. As such, the training set is used to train predictive models whereas the test set is used to validate the predictive models.

[0063] In various embodiments, the predictive model is any one of a regression model e.g., linear regression, logistic regression, or polynomial regression), decision tree, random forest,support vector machine, Naive Bayes model, k-means cluster, or neural network (e.g., feedforward networks, convolutional neural networks (CNN), deep neural networks (DNN), autoencoder neural networks, generative adversarial networks, or recurrent networks (e.g., long short-term memory networks (LSTM), bi-directional recurrent networks, deep bidirectional recurrent networks), or any combination thereof. For example, the predictive model can be a stacked classifier that includes both a linear regression and decision tree.

[0064] The predictive model can be trained using a machine learning implemented method, such as any one of a linear regression algorithm, logistic regression algorithm, decision tree algorithm, support vector machine classification, Naive Bayes classification, K-Nearest Neighbor classification, random forest algorithm, deep learning algorithm, gradient boosting algorithm, and dimensionality reduction techniques such as manifold learning, principal component analysis, factor analysis, autoencoder regularization, and independent component analysis, or combinations thereof. In various embodiments, the cancer prediction model is trained using supervised learning algorithms, unsupervised learning algorithms, semisupervised learning algorithms (e. ., partial supervision), weak supervision, transfer, multitask learning, or any combination thereof.

[0065] In various embodiments, the predictive model has one or more parameters, such as hyperparameters or model parameters. Hyperparameters are generally established prior to training. Examples of hyperparameters include the learning rate, depth or leaves of a decision tree, number of hidden layers in a deep neural network, number of clusters in a k- means cluster, penalty in a regression model, and a regularization parameter associated with a cost function. Model parameters are generally adjusted during training. Examples of model parameters include weights associated with nodes in layers of neural network, support vectors in a support vector machine, and coefficients in a regression model. The model parameters of the cancer prediction model are trained (e.g., adjusted) using the training data to improve the predictive capacity of the cancer prediction model.

[0066] The model training module 150 trains one or more predictive models, each predictive model receiving, as input, one or more biomarkers. In various embodiments, the model training module 150 constructs a predictive model that receives, as input, expression values of two or more biomarkers. In various embodiments, the model training module 150 constructs a predictive model that receives, as input, expression values of three or more biomarkers. In various embodiments, the model training module 150 constructs a predictive model that receives, as input, expression values of four or more, five or more, six or more, seven or more, eight or more, nine or more, ten or more, eleven or more, twelve or more,thirteen or more, fourteen or more, fifteen or more, sixteen or more, seventeen or more, eighteen or more, nineteen or more, twenty or more, twenty one or more, twenty two or more, twenty three or more, twenty four or more, twenty five or more, twenty six or more, twenty seven or more, twenty eight or more, twenty nine or more, thirty or more, thirty one or more, thirty two or more, thirty three or more, thirty four or more, thirty five or more, thirty six or more, thirty seven or more, thirty eight or more, thirty nine or more, forty or more, forty one or more, forty two or more, forty three or more, forty four or more, forty five or more, forty six or more, forty seven or more, forty eight or more, forty nine or more, fifty or more, fifty one or more, fifty two or more, or fifty three or more biomarkers. In various embodiments, the model training module 150 constructs a predictive model that receives, as input, expression values of eight biomarkers. In some embodiments, the model training module 150 constructs a predictive model for more than eight biomarkers. For example, a predictive model receives, as input, expression values of 53 biomarkers.

[0067] In various embodiments, the model training module 150 identifies a set of biomarkers that are to be used to train a predictive model. The model training module 150 may begin with a list of candidate biomarkers that are promising for distinguishing responders and non-responders to treatment with checkpoint inhibitors. In one embodiment, candidate biomarkers may be biomarkers identified through a literature curation process. In some embodiments, candidate biomarkers may be biomarkers whose expression values in test samples obtained from individuals that are positive for response to checkpoint inhibitors (e.g., tumor reduction) are statistically significant in comparison to expression values of biomarkers in test samples obtained from individuals that are non-responsive to checkpoint inhibitors (e.g., lack of tumor reduction).

[0068] In various embodiments, a predictive model is iteratively trained using, as input, the quantitative expression values of the markers for each individual. For example, referring again to FIG. 1C, one iteration involves providing a training example (e.g., a row of the training data) that includes the quantitative expression value of biomarkers (e.g., “Al” and “B 1”) for a particular individual (e.g., individual 1). Each predictive model is trained on reference ground truth data that includes the indication (e.g., the responder or non-responder result). In various embodiments, over training iterations, each predictive model is trained (e.g., the parameters are tuned) to minimize a prediction error between a prediction of responder activity outputted by the predictive model and the ground truth data. In various embodiments, the prediction error is calculated based on a loss function, examples of whichinclude a LI regularization (Lasso Regression) loss function, a L2 regularization (Ridge Regression) loss function, or a combination of LI and L2 regularization (ElasticNet).III.B. Deploying a Predictive model

[0069] During the deployment phase, the model deployment module 160 (as shown in FIG. IB) analyzes quantitative biomarker expression values from a test sample obtained from a subject of interest by applying a trained predictive model. In some embodiments, the subject has not previously been evaluated and therefore, the deployment of the predictive model enables generation of an in silico responder prediction based on the quantitative biomarker expression values derived from the subject.

[0070] Reference is now made to FIG. 2, which depicts an example flow diagram for predicting responsiveness to an immune checkpoint inhibitor for a subject, in accordance with an embodiment. In various embodiments, the quantitative biomarker expression values are obtained from at least one data set and provided as input to the predictive model as in module 210. The predictive model analyzes the quantitative biomarker expression values and outputs an assessment of an individual’s response to treatment with a checkpoint inhibitor as in module 220.

[0071] In various embodiments, the responder prediction (e.g., responder prediction 140 described in FIG. 1 A) is a predicted score. In various embodiments, the predicted score outputted by the prediction model is compared to one or more reference scores to determine a measure of an individual’s response to treatment with a checkpoint inhibitor. Reference scores refer to previously determined scores, further described below as “responder scores” or “non-responder scores,” that correspond to responder patients or non-responder patients. For example, the one or more scores may be “responder scores” corresponding to patients that respond to treatment with a checkpoint inhibitor, a patient’s own baseline at a prior timepoint when the patient did not exhibit responder activity (e.g., a lack of tumor reduction), patients clinically diagnosed with the cancer but not exhibiting cancer activity, or a threshold score (e.g., a cutoff). As another example, the one or more scores may be “non-responder scores” corresponding to non-responder patients, a patient’s own score indicating non-responder activity at a prior timepoint, or a threshold score (e.g., a cutoff). As one example, the threshold score can correspond to responder patients and can be generated by training a predictive model using expression values of biomarkers from responder patients. As another example, the threshold score can correspond to non-responder patients and can be generatedby training a predictive model using expression values of biomarkers from the non-responder patients.

[0072] In various embodiments, the responder prediction (e.g., responder prediction 140 described in FIG. 1 A) is a predicted score that corresponds to the directional shift in response based on a decrease in tumor size or lack of decrease in tumor size. In one embodiment, the predicted score outputted by the prediction model can be compared to a score corresponding to individuals previously determined to have exhibited a reduction in tumor size (e.g., reduction in tumor size as determined by an imaging scan, such as a CT scan or MRI scan). The subject can be classified as likely to respond to treatment with a checkpoint inhibitor if the predicted score of the subject is significantly different (e.g., p-value <0.05) in comparison to the score corresponding to individuals previously determined to lack a reduction in tumor size. The subject can be classified as unlikely to respond to treatment with a checkpoint inhibitor if the predicted score of the subject is not significantly different (e.g., p-value > 0.05) in comparison to the score corresponding to individuals previously determined to lack a reduction in tumor size. In one embodiment, the predicted score outputted by the prediction score is compared to a score corresponding to individuals previously determined to have exhibited a response (e.g., reduction in tumor size as determined through imaging). The subject can be classified as likely to respond to treatment with a checkpoint inhibitor if the predicted score of the subject is not significantly different (e.g., p-value > 0.05) from the score corresponding to individuals that have shown a response (e.g., increasing reduction in tumor size as determined through imaging). The subject can be classified as likely to respond to treatment with a checkpoint inhibitor if the predicted score of the subject is significantly different (e.g., p-value < 0.05) from the score corresponding to individuals that have not exhibited a response (e.g., increasing reduction in tumor size as determined through imaging). In some embodiments, the predicted score outputted by the prediction model is compared to both a score corresponding to individuals previously determined to have exhibited a response (e.g., increasing reduction in tumor size as determined through imaging) and a score corresponding to individuals who have not exhibited a response (e.g., a lack of reduction in tumor size). For example, the subject can be classified as likely to respond to treatment with a checkpoint inhibitor if the predicted score of the subject is significantly different (e.g., p-value < 0.05) in comparison to the score corresponding to individuals who have not exhibited a response (e.g., a lack of reduction in tumor size) and not significantly different (e.g., p-value >0.05) in comparison to the score corresponding to individuals who have exhibited a response (e.g., increasing reduction in tumor size as determined through imaging). In variousembodiments, the subject can be classified as neither a responder or a non-responder if the predicted score of the subject is not significantly different (e.g., p-value > 0.05) in comparison to both the score corresponding to individuals who have exhibited a response to treatment with a checkpoint inhibitor and the score corresponding to individuals who have not exhibited a response to a checkpoint inhibitor.

[0073] In various embodiments, a measure of the response activity predicted by the predictive model provides additional utility for managing the cancer treatment in the patient. As one example, the measure of the response activity predicted by the predictive model is useful for selecting a candidate therapeutic (e.g., a checkpoint inhibitor) or for determining the effectiveness of a previously administered therapeutic (e.g., a checkpoint inhibitor).

[0074] In various embodiments, the measure of responder activity predicted by the predictive model for a patient can be compared to a prior measure of responder activity to determine whether a therapeutic administered to the patient is demonstrating efficacy. As one example, the prior measure of responder activity may be a prediction determined for the same patient (e.g., a baseline measure of responder activity). Thus in this example, the comparison of the measure of responder activity and the prior measure of responder activity is a longitudinal analysis of a patient that is undergoing treatment using the therapeutic (e.g., a checkpoint inhibitor). As such, a difference or lack of difference between the measure of responder activity and prior measure of responder activity can be an indication that the therapeutic is having an effect or lack of an effect. As another example, the prior measure of responder activity may be a measure determined for a population of patients (e.g., a reference set of patients). In this example, the comparison of the measure of responder activity and the prior measure of responder activity can reveal whether the patient is experiencing effects due to a therapeutic, as evidenced by the measure of responder activity, in comparison to the prior measure of responder activity for the population of patients.

[0075] In various embodiments, if the comparison between the measure of responder activity and prior measure of responder activity indicates that a currently administered therapeutic (e.g., a checkpoint inhibitor) is not exhibiting an effect, or is not exhibiting an effect to a desired extent, a change in the patient’s treatment can be undertaken. In one embodiment, the treatment dose of the currently administered therapeutic can be altered to effect a patient response. For example, the dosage of the currently administered therapeutic can be altered (e.g., increased or decreased). In one embodiment, a candidate therapeutic can be selected for administration to the patient. In various embodiments, a candidate therapeutic can be administered to the patient in place of the currently administered therapeutic or thecandidate therapeutic can be administered to the patient in addition to the currently administered therapeutic. For example, a different CBL-Bi can be administered to the patient in place of a currently administered therapeutic or in addition to the currently administered therapeutic.

[0076] As another example, a measure of the responder activity is useful for supporting symptom and medication tracking, nursing interventions, laboratory monitoring, and curated longitudinal tumor response reports. In such scenarios, the measure of responder activity can reduce unplanned healthcare utilization (e.g., unplanned visits to physician’s office), thereby improving patient and physician satisfaction.IV. Biomarker Panel

[0077] In various embodiments, generating a responder prediction (e.g., responder prediction 140) to treatment with a checkpoint inhibitor involves implementing a univariate biomarker panel. Therefore, the univariate biomarker panel includes one biomarker. In some embodiments, generating a responder prediction (e.g., responder prediction 140) involves implementing a multivariate biomarker panel. In such embodiments, the multivariate biomarker panel includes more than one biomarker. In various embodiments, the multivariate biomarker panel includes four or more biomarkers. In various embodiments, the multivariate biomarker panel includes 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59 biomarkers. . In particular embodiments, the multivariate biomarker panel includes 2 biomarkers. In particular embodiments, the multivariate biomarker panel includes 4 biomarkers. In particular embodiments, the multivariate biomarker panel includes 6 biomarkers. In particular embodiments, the multivariate biomarker panel includes 8 biomarkers. In particular embodiments, the multivariate biomarker panel includes 25 biomarkers. In particular embodiments, the multivariate biomarker panel includes 45 biomarkers.

[0078] In various embodiments described herein, a biomarker panel is implemented for generating a responder prediction to treatment with a checkpoint inhibitor. In various embodiments, generating a responder prediction with a checkpoint inhibitor involves implementing a univariate biomarker panel. Therefore, the univariate biomarker panel includes one biomarker. In other embodiments, the assessment of response to treatment with a checkpoint inhibitor involves implementing a multivariate biomarker panel. In such embodiments, the multivariate biomarker panel for assessing disease progression includesmore than one biomarker. In various embodiments, the multivariate biomarker panel for assessing disease progression includes four or more biomarkers. In various embodiments, the multivariate biomarker panel for assessing disease progression includes 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59 biomarkers. . In particular embodiments, the multivariate biomarker panel includes 2 biomarkers. In particular embodiments, the multivariate biomarker panel includes 4 biomarkers. In particular embodiments, the multivariate biomarker panel includes 6 biomarkers. In particular embodiments, the multivariate biomarker panel includes 8 biomarkers. In particular embodiments, the multivariate biomarker panel includes 25 biomarkers. In particular embodiments, the multivariate biomarker panel includes 45 biomarkers. In particular embodiments, the multivariate biomarker panel includes 53 biomarkers. In particular embodiments, the multivariate biomarker panel includes 59 biomarkers. In particular embodiments, the multivariate biomarker panel includes commonly differentially expressed genes that are determined from gene expression data generated from two or more experiments, as is described in further detail herein.

[0079] In various embodiments, the biomarkers in the biomarker panel can include one or more of biomarkers listed in Table 1. In various embodiments, the biomarkers in the biomarker panel can include 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 , 52, 53, 54, 55, 56, 57, 58, or 59 of the biomarkers listed in Table 1. In various embodiments, the biomarkers in the biomarker panel can include one or more of biomarkers listed in Table 2. In various embodiments, the biomarkers in the biomarker panel can include 2, 3, 4, 5, 6, 7, or 8 of the biomarkers listed in Table 2.Table 1: Example BiomarkersTable 2: Example Set of 8 Biomarkers

[0080] In some embodiments, the biomarkers can include one or more of: CXCL9, GBP5, IRF1, CD274, STAT1, PSMB9, IFIT3, IFNG. In some embodiments, the biomarkers are a gene-signature that includes two or more, three or more, four or more, five or more, six or more, seven or more, or each of: CXCL9, GBP5, IRF1, CD274, STAT1, PSMB9, IFIT3, IFNG.

[0081] In various embodiments, the biomarker panel for generating a prediction (e.g., a prediction for responsiveness to an immune checkpoint inhibitor) includes a minimal set of predictive biomarkers identified as a gene-signature that includes each of: CXCL9, GBP5, IRF1, CD274, STAT1, PSMB9, IFIT3, IFNG.

[0082] In some embodiments, the biomarkers can include one or more of: Notchl, IGF1R, ZAP70, CD3E, LAT, and PLCG1. In some embodiments, the biomarkers are Notchl and IGF1R. In some embodiments, the biomarkers are Notchl and IGF1R and further comprise one or more additional biomarkers selected from the group consisting of: ZAP70, CD3E, LAT, and PLCG1. In some embodiments, the phosphorylation status of one or more amino acids of the biomarker has been determined. In some embodiments the one or more amino acids comprise one or more of tyrosine, serine, and threonine.V. Biomarkers

[0083] The altered expression of biomarkers disclosed herein may contribute to the regression of cancer, such as reduction in tumor size or tumor growth inhibition.

[0084] CXCL9 is a 14.019 kDa biomarker involved in the immune and inflammatory response. Additionally, CXCL9 affects chemotaxis of activated T cell, growth and movement of cells, and activation of immune cells.

[0085] GBP5 is a 66.617 kDa biomarker that belongs to the TRAFAC class dynamin-like GTPase superfamily. GBP5 is an activator of inflammasome assembly and is induced by interferon and plays a role in innate immunity.

[0086] IRF1 is a 36.502 kDa biomarker that acts a transcriptional regulator and tumor suppressor. IRF1 is also involved in activation of genes in the innate and acquired immune response.

[0087] CD274 is a 33.275 kDa biomarker. CD274 interaction with its receptor inhibits T- cell activation and cytokine production. CD274 is also known as PDL1 or programed death ligand 1 and can bind to the receptor PD1. PD1 is known to be expressed in cancer and interaction of PD1 with PDL1 may be a mechanism for tumors to escape immune responses.

[0088] STAT1 is a 87.335 kDa biomarker that translocates to the nucleus to activate transcription. STAT1 activates expression of genes important for cell viability and immune response to pathogens.

[0089] PSMB9 is a 23.264 kDa biomarker that is part of the proteasome B-type family. PSMB9 expression is induced by gamma interferon to replace proteasome beta 6 subunit in the immunoproteasome.

[0090] IFIT3 is a 55.985 kDa biomarker that is induced by interferon and involved in responses to viral infection. Additionally, IFIT3 can inhibit cellular processes, cell migration, proliferation, and signaling.

[0091] IFNG is a 19.348 kDa biomarker that is a member of the type II interferon class. IFNG is secreted by cells involved in innate and adaptive immune responses. IFNG plays a role in cellular responses to viral and microbial infections as well as anti-tumor responses.VI. Assays for Determining Biomarkers

[0092] Reference is now made to FIG. 3 which depicts a schematic of the selection of biomarkers by identifying commonly differently expressed genes across two or more experiences, in accordance with an embodiment. As shown in FIG. 3, the two or more experiments (e.g., experiment 310A and 310B) are performed to generate gene expression data (e.g., gene expression data 320A and 320B). Although FIG. 3 depicts two experiments 310A and 310B, in various embodiments, the number of experiments may include three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, or twenty experiments. Furthermore, although FIG. 3 depicts one or more gene expression data 320 (e.g., a gene expression data 320 for each corresponding experiment 310), in various embodiments, the number of gene expression data 320 may include three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, or twenty or more different gene expression data 320.

[0093] In various embodiments, an experiment 310 involves an in vitro experiment. In various embodiments, an experiment 310 involves an in vivo experiment. In various embodiments, a first experiment 310A involves an in vitro experiment, and a second experiment 310B involves an in vitro experiment. In various embodiments, a first experiment 310A involves an in vitro experiment, and a second experiment 310B involves an in vivo experiment. In various embodiments, a first experiment 310 A involves an in vivo experiment, and a second experiment 310B involves an in vivo experiment. In various embodiments, a first experiment 310A involves an in vitro experiment, a second experiment 310B involves anin vivo experiment, and a third experiment involves an in vitro experiment. In various embodiments, a first experiment 310A involves an in vitro experiment, a second experiment 31 OB involves an in vivo experiment, and a third experiment involves an in vivo experiment. In various embodiments, a first experiment 310A involves an in vitro experiment, a second experiment 31 OB involves an in vivo experiment, a third experiment involves an in vivo experiment, and a fourth experiment involves an in vitro experiment.

[0094] In various embodiments, an experiment 310 involves analyzing gene expression in human cells. In various embodiments, an experiment 310 involves analyzing gene expression in murine cells. In various embodiments, a first experiment 310A involves analyzing gene expression in human cells, and a second experiment 310B involves analyzing gene expression in murine cells. In various embodiments, a first experiment 310A involves analyzing gene expression in human cells, a second experiment 310B involves analyzing gene expression in murine cells, and a third experiment involves analyzing gene expression in additional human cells. In various embodiments, a first experiment 310A involves analyzing gene expression in human cells, a second experiment 310B involves analyzing gene expression in murine cells, and a third experiment involves analyzing gene expression in additional murine cells. In various embodiments, a first experiment 310A involves analyzing gene expression in human cells, a second experiment 310B involves analyzing gene expression in murine cells, a third experiment involves analyzing gene expression in additional human cells, and a fourth experiment involves analyzing gene expression in additional murine cells.

[0095] In various embodiments, an experiment 310 involves exposing cells (e.g., human cells or murine cells) to a therapy (e.g., a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi), such as a CBL-Bi disclosed in Table 3) and determining the gene expression levels caused by the therapy. In various embodiments, the cells include human cells e.g., human cells obtained from a blood sample obtained from a subject. In various embodiments, the human cells include peripheral blood mononuclear cells (PBMCs). In various embodiments, the human cells include one or more of lymphocytes (T-cells, B cells, NK cells), monocytes, and granulocytes (neutrophils, basophils, and eosinophils). In various embodiments, the cells include murine cells e.g., murine cells obtained from a blood sample obtained from a murine subject. In various embodiments, the murine cells include peripheral blood mononuclear cells (PBMCs). In various embodiments, the murine cells include one or more of lymphocytes (T-cells, B cells, NK cells), monocytes, and granulocytes (neutrophils, basophils, and eosinophils).

[0096] In various embodiments, the cells are from a subject (e.g., human or murine subject) that has a cancer. For example, the subject may be a murine subject representing an animal tumor model. Such subjects may have been previously administered a therapy (e.g., a CBL- Bi) and therefore, the experiment 310 can involve determining the gene expression levels caused by the therapy within an animal tumor model. In various embodiments, the experiment 310 can involve determining the gene expression levels caused by the therapy within two or more animal tumor models. In various embodiments, the experiment 310 can involve determining the gene expression levels caused by the therapy within three or more, four or more, five or more, six or more, seven or more, eight or more, nine or more, ten or more, eleven or more, twelve or more, thirteen or more, fourteen or more, fifteen or more, sixteen or more, seventeen or more, eighteen or more, nineteen or more, or twenty or more animal tumor models. In particular embodiments, the experiment 10 can involve determining the gene expression levels caused by the therapy within ten tumor models. Example tumor models include A20, B16BL6, B16F10, CT26, EMT6, H22, Hepal-6, LL2, Renca, RM1 disclosed in Table 6.

[0097] In various embodiments, the two or more animal models can be previously determined to be responders or non-responders to the therapy (e.g., CBL-Bi). For example, an animal model that is a responder may exhibit tumor growth inhibition (TGI) due to the therapy that is greater than a threshold percentage. The animal model that is a non-responder may exhibit tumor growth inhibition (TGI) due to the therapy that is less than a threshold percentage. In various embodiments, the threshold percentage can be 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99%, or 100%. In particular embodiments, the threshold percentage can be 30%.

[0098] In various embodiments, an experiment 310 involves exposing cells (e.g., human cells or murine cells) to a therapy (e.g., a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi)) across two or more timepoints. For example, the experiment can involve exposing the cells to the therapy (e.g., CBL-Bi) across two timepoints, across three timepoints, across four timepoints, across five timepoints, across six timepoints, across seven timepoints, across eight timepoints, across nine timepoints, or across ten timepoints. As an example, the experiment can involve exposing the cells to the therapy (e.g., CBL-Bi) across two timepoints (e.g., at 6 hours and 24 hours after exposure to the therapy). As an example, the experiment can involve exposing the cells to the therapy (e.g., CBL-Bi) across four timepoints (e.g., at 6 hours, 24 hours, 34 hours, and / or 48 hours after exposure to the therapy), Following exposure of the cells to the therapy across the two or more timepoints,the experiment 310 can involve determining the gene expression levels caused by the therapy across the two or more timepoints.

[0099] In various embodiments, exposing cells to a therapy involves exposing the cells to a dosage of the therapy between 1 nM and 10 pM. In various embodiments, exposing cells to a therapy involves exposing the cells to a dosage of the therapy between 2 nM and 5 pM, between 3 nM and 4 pM, between 4 nM and 3 pM, between 5 nM and 2 pM, between 10 nM and 1 pM, between 20 nM and 500 nM, between 50 nM and 400 nM, between 75 nM and 300 nM, between 85 nM and 200 nM, or between 95 nM and 150 nM. In various embodiments, exposing cells to a therapy involves exposing the cells to a dosage of 10 nM. In various embodiments, exposing cells to a therapy involves exposing the cells to a dosage of 50 nM. In various embodiments, exposing cells to a therapy involves exposing the cells to a dosage of 100 nM. In various embodiments, exposing cells to a therapy involves exposing the cells to a dosage of 1 pM.

[0100] In various embodiments, an experiment 310 involves exposing cells (e.g., human cells or murine cells) to a therapy (e.g., a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi)), further stimulating the cells, and determining the resulting gene expression levels. In various embodiments, stimulating the cells involves exposing the cells to a CD3 stimulation (e.g., using anti-CD3 antibodies). In various embodiments, stimulating the cells involves exposing the cells to a CD28 stimulation (e.g., using anti-CD28 antibodies). In various embodiments, stimulating the cells involves exposing the cells to both a CD3 stimulation (e.g., using anti-CD3 antibodies) and a CD28 stimulation (e.g., using anti-CD28 antibodies).

[0101] In various embodiments, an experiment 310 involves performing a pharmacokinetics (PK) or pharmacodynamics (PD) study involving one or more animal models. In particular embodiments, an experiment 310 involves performing a pharmacodynamics (PD) study involving one or more animal models. For example, animals of an animal model can be administered a therapy (e.g., CBL-Bi) and samples can be obtained from the animals at set timepoints to measure the pharmacodynamics of the therapy. In various embodiments, the therapy (e.g., CBL-Bi) is administered at a dose between 0.1 mg / kg to 50 mg / kg. In various embodiments, the therapy (e.g., CBL-Bi) is administered at a dose between 0.5 mg / kg to 40 mg / kg, 1 mg / kg to 30 mg / kg, 2 mg / kg to 20 mg / kg, 3 mg / kg to 10 mg / kg, 4 mg / kg to 8 mg / kg, or 5 mg / kg to 7 mg / kg. In particular embodiments, the therapy (e.g., CBL-Bi) is administered at a dose of 1 mg / kg. In particular embodiments, the therapy (e.g., CBL-Bi) is administered at a dose of 6 mg / kg. In particular embodiments, the therapy (e.g., CBL-Bi) isadministered at a dose of 20 mg / kg. In various embodiments, samples can be obtained from the animals at 1 day after administration, 2 days after administration, 3 days after administration, 4 days after administration, 5 days after administration, 6 days after administration, 7 days after administration, 8 days after administration, 9 days after administration, 10 days after administration, 11 days after administration, 12 days after administration, 13 days after administration, 14 days after administration, 15 days after administration, 16 days after administration, 17 days after administration, 18 days after administration, 19 days after administration, or 20 days after administration. In particular embodiments, samples can be obtained from the animals at 1 day after administration. In particular embodiments, samples can be obtained from the animals at 3 days after administration. In particular embodiments, samples can be obtained from the animals at 7 days after administration.

[0102] In various embodiments, the animals in the PD study are tumor animal models. In various embodiments, samples can be obtained from the tumor in the animals to measure the PD across the timepoints in the tumor. In various embodiments, samples can be obtained from lymph nodes in the animals to measure the PD across the timepoints in the lymph nodes. In various embodiments, samples can be obtained from cells (e.g., PBMCs) from the animals to measure the PD across the timepoints in the cells.

[0103] In various embodiments, an experiment 310 involves exposing human PBMCs to a therapy (e.g., CBL-Bi), stimulating the cells (e.g., using anti CD3 stimulation, anti CD28 stimulation, or both anti CD3 and anti CD28 stimulation), and determining the resulting gene expression levels. In particular embodiments, the experiment 310 involves exposing human PBMCs to a CBL-Bi at two concentrations (e.g., 50 nM and 1 pM), and determining gene expression levels at four time points (6 hours, 24 hours, 34 hours, and 48 hours after exposure to the therapy).

[0104] In various embodiments, an experiment 310 involves exposing, in vivo, animals in at least 10 different tumor models to a therapy (e.g., CBL-Bi) and determining whether the animals are responders or non-responders (e.g., based on whether an animal exhibits at least a threshold percentage of tumor growth inhibition in response to the therapy). The resulting gene expression levels in the animals can be determined.

[0105] In various embodiments, an experiment 310 involves a pharmacodynamics (PD) experiment in which animals from animal tumor models are administered a therapy (e.g., CBL-Bi at any of 1 mg / kg, 6 mg / kg, or 20 mg / kg) and samples are obtained from the animals at set timepoints (e.g., Day 1, Day 3, and / or Day 7). Samples obtained from the animals caninclude samples from tumor tissue, samples including the lymph nodes, and / or samples including cells (e.g., PBMCs). The resulting gene expression levels in the obtained samples can be determined.

[0106] In various embodiments, an experiment 310 involves exposing human PBMCs to a therapy (e.g., CBL-Bi), stimulating the cells (e.g., using anti CD3 stimulation), and determining the resulting gene expression levels. In particular embodiments, the experiment 310 involves exposing human PBMCs to a CBL-Bi at three concentrations (e.g., 10 nM, 100 nM, and 1 M), and determining gene expression levels at two time points (6 hours and 24 hours after exposure to the therapy). The resulting gene expression levels in the obtained samples can be determined.

[0107] In various embodiments, an experiment 310 involves determining gene expression levels. In various embodiments, methods for determining gene expression levels from a sample or from cells can involve performing a multiplex nucleic acid hybridization analysis. In various embodiments, methods for determining gene expression levels from a sample or from cells can involve performing a RNA sequencing analysis.VII. Identification of Biomarkers and Gene Signature

[0108] Referring again to FIG. 3, methods for identifying biomarkers and / or a gene signature involve step 330 of comparing gene expression data e.g., the gene expression data 320A and gene expression data 320B to determine commonly differentially expressed genes 340 across the two or more experiments 310A and 310B. In particular embodiments, two gene expression data 320 sets derived from two different experiments are compared to identify commonly differentially expressed genes 340. In particular embodiments, three gene expression data 320 sets derived from three different experiments are compared to identify commonly differentially expressed genes 340. In particular embodiments, four gene expression data 320 sets derived from four different experiments are compared to identify commonly differentially expressed genes 340. In some embodiments, different gene expression data sets derived the same experiment are compared to identify commonly differentially expressed genes. For example, as shown in FIG. 3, experiment 310B can lead to generation of different sets of gene expression data 320B. Thus, the different sets of gene expression data 320B can be compared at step 330 to identify commonly differentially expressed genes 340. In some embodiments, the commonly differentially expressed genes 340 are used to create a gene signature of biomarkers.

[0109] In various embodiments, gene expression data 320 includes gene expression data determined by multiplex nucleic acid hybridization analysis of nucleic acids from mice of a plurality of murine tumor models wherein the mice are treated with an immune checkpoint inhibitor. In various embodiments, gene expression data 320 includes gene expression data determined by RNA sequencing analysis of PBMCs isolated from mice treated with an immune checkpoint inhibitor for pharmacodynamic studies. In various embodiments, gene expression data 320 includes gene expression data determined by multiplex nucleic acid hybridization analysis of human PBMCs treated with an immune checkpoint inhibitor. In various embodiments, gene expression data 320 includes gene expression data determined by RNA sequencing analysis of human PBMCs treated with an immune checkpoint inhibitor. In various embodiments, gene expression data 320 includes expression data of upregulated genes collected from mice treated with 1 mg / kg, 6 mg / kg, or 20 mg / kg of CBL-Bi.

[0110] In some embodiments of the disclosure, analyzing gene expression data 320 includes determining genes with altered expression (e.g., differentially expressed) when comparing control conditions to conditions in the presence of a checkpoint inhibitor. In some embodiments, analyzing gene expression data 320 includes determining genes with log2 fold changes that correlate with tumor growth inhibition in mouse models. In some embodiments, analyzing gene expression data 320 includes determining genes with log2 fold changes that differ significantly (e.g., p<0.05) between mouse responder and non-responder models. In some embodiments, analyzing gene expression data 320 includes determining genes upregulated in at least three mouse responder models. In some embodiments, analyzing gene expression data 320 includes determining genes with the greatest expression in response to treatment with checkpoint inhibitors.

[0111] In various embodiments, determining a set of commonly differentially expressed genes includes comparing each of (i) gene expression data determined by multiplex nucleic acid hybridization analysis, (iii) gene expression data determined by multiplex nucleic acid hybridization analysis of human PBMCs, and (iv) gene expression data determined by RNA sequencing analysis of human PBMCs.

[0112] In various embodiments, determining a set of commonly differentially expressed genes can include a multiple tiered analysis. For example, a first tier can involve determining a first set of commonly differentially expressed genes across two or more experiments and further determining a second set of commonly differentially expressed genes across an additional two or more experiments. Next, a second tier can involve determining the set of commonly differentially expressed genes as the union (e.g., overlap) between the first set ofcommonly differentially expressed genes and the second set of commonly differentially expressed genes.

[0113] In various embodiments, a first tier can involve determining a first set of commonly differentially expressed genes between gene expression data obtained from 1) a first experiment involving exposing human PBMCs to a therapy (e.g., CBL-Bi) at three concentrations (e.g., 10 nM, 100 nM, and 1 pM), stimulating the cells (e.g., using anti CD3 stimulation), and determining gene expression levels at two time points (6 hours and 24 hours after exposure to the therapy), and 2) a second experiment involving exposing, in vivo, animals in at least 10 different tumor models to a therapy (e.g., CBL-Bi) and determining whether the animals are responders or non-responders (e.g., based on whether an animal exhibits at least a threshold percentage of tumor growth inhibition in response to the therapy). As an example, the first set of commonly differentially expressed genes can include one or more of CXCL9, GBP5, IRF1 and CD274.

[0114] In various embodiments, a second tier can involve determining a second set of commonly differentially expressed genes between gene expression data obtained from 1) a first experiment involving exposing human PBMCs to a therapy (e.g., CBL-Bi) at two concentrations (e.g., 50 nM and 1 pM), stimulating the cells (e.g., using anti CD3 stimulation, anti CD28 stimulation, or both anti CD3 and anti CD28 stimulation) and determining gene expression levels at four time points (6 hours, 24 hours, 34 hours, and 48 hours after exposure to the therapy), and 2) a second experiment involving a pharmacodynamics (PD) experiment in which animals from animal tumor models are administered a therapy (e.g., CBL-Bi at any of Img / kg, 6 mg / kg, or 20 mg / kg) and samples are obtained from the animals at set timepoints (e.g., Day 1, Day 3, and / or Day 7) from tumor tissue, lymph nodes, and / or cells (e.g., PBMCs). As an example, the second set of commonly differentially expressed genes can include one or more of CXCL9, IFIT3, OAS2 and STAT1.

[0115] In various embodiments, the commonly differentially expressed genes 340 can serve as a biomarker signature e.g., for determining responders and non-responders to a therapy (e.g., a CBL-Bi). In various embodiments, one or more additional biomarkers can be included with the commonly differentially expressed genes 340 to serve as a biomarker signature e.g., for determining responders and non-responders to a therapy (e.g., a CBL-Bi). In various embodiments, an additional biomarker can be selected for exhibiting the most significant differential expression in an experiment. In various embodiments, an additional biomarker can be selected for exhibiting the strongest dose-dependent response in an experiment. For example, an additional biomarker can be selected for exhibiting thestrongest dose-dependent response in an experiment involving exposing human PBMCs to a therapy (e.g., CBL-Bi) at two concentrations (e.g., 50 nM and 1 pM), stimulating the cells (e.g., using anti CD3 stimulation, anti CD28 stimulation, or both anti CD3 and anti CD28 stimulation) and determining gene expression levels at four time points (6 hours, 24 hours, 34 hours, and 48 hours after exposure to the therapy), In various embodiments, an additional biomarker includes IFNG.VIII. Computer Implementation

[0116] The methods of the invention, including the methods of predicting responders and non-responders to treatment with checkpoint inhibitors, are, in some embodiments, performed on one or more computers.

[0117] For example, the building and deployment of a predictive model and database storage can be implemented in hardware or software, or a combination of both. In one embodiment of the invention, a machine-readable storage medium is provided, the medium comprising a data storage material encoded with machine readable data which, when using a machine programmed with instructions for using said data, is capable of displaying any of the datasets and execution and results of a predictive model of this invention. The invention can be implemented in computer programs executing on programmable computers, comprising a processor, a data storage system (including volatile and non-volatile memory and / or storage elements), a graphics adapter, a pointing device, a network adapter, at least one input device, and at least one output device. A display is coupled to the graphics adapter. Program code is applied to input data to perform the functions described above and generate output information. The output information is applied to one or more output devices, in known fashion. The computer can be, for example, a personal computer, microcomputer, or workstation of conventional design.

[0118] Each program can be implemented in a high level procedural or object oriented programming language to communicate with a computer system. However, the programs can be implemented in assembly or machine language, if desired. In any case, the language can be a compiled or interpreted language. Each such computer program is preferably stored on a storage media or device (e.g., ROM or magnetic diskette) readable by a general or special purpose programmable computer, for configuring and operating the computer when the storage media or device is read by the computer to perform the procedures described herein. The system can also be considered to be implemented as a computer-readable storagemedium, configured with a computer program, where the storage medium so configured causes a computer to operate in a specific and predefined manner to perform the functions described herein.

[0119] The signature patterns and databases thereof can be provided in a variety of media to facilitate their use. “Media” refers to a manufacture that contains the signature pattern information of the present invention. The databases of the present invention can be recorded on computer readable media, e.g. any medium that can be read and accessed directly by a computer. Such media include, but are not limited to: magnetic storage media, such as floppy discs, hard disc storage medium, and magnetic tape; optical storage media such as CD-ROM; electrical storage media such as RAM and ROM; and hybrids of these categories such as magnetic / optical storage media. One of skill in the art can readily appreciate how any of the presently known computer readable mediums can be used to create a manufacture comprising a recording of the present database information. "Recorded" refers to a process for storing information on computer readable medium, using any such methods as known in the art. Any convenient data storage structure can be chosen, based on the means used to access the stored information. A variety of data processor programs and formats can be used for storage, e.g. word processing text file, database format, etc.

[0120] In some embodiments, the methods of predicting responders or non-responders to treatment with checkpoint inhibitors, are performed on one or more computers in a distributed computing system environment (e.g., in a cloud computing environment). In this description, “cloud computing” is defined as a model for enabling on-demand network access to a shared set of configurable computing resources. Cloud computing can be employed to offer on-demand access to the shared set of configurable computing resources. The shared set of configurable computing resources can be rapidly provisioned via virtualization and released with low management effort or service provider interaction, and then scaled accordingly. A cloud-computing model can be composed of various characteristics such as, for example, on-demand self-service, broad network access, resource pooling, rapid elasticity, measured service, and so forth. A cloud-computing model can also expose various service models, such as, for example, Software as a Service (“SaaS”), Platform as a Service (“PaaS”), and Infrastructure as a Service (“laaS”). A cloud-computing model can also be deployed using different deployment models such as private cloud, community cloud, public cloud, hybrid cloud, and so forth. In this description and in the claims, a “cloud-computing environment” is an environment in which cloud computing is employed.VIII.A. Example Computer

[0121] FIG. 4 illustrates an example computer 400 for implementing the entities shown in FIGS. 1 and 3. The computer 400 includes at least one processor 402 coupled to a chipset 404. The chipset 404 includes a memory controller hub 420 and an input / output (I / O) controller hub 422. A memory 406 and a graphics adapter 412 are coupled to the memory controller hub 420, and a display 418 is coupled to the graphics adapter 412. A storage device 408, a keyboard 410, an input interface 414, and network adapter 416 are coupled to the I / O controller hub 422. Other embodiments of the computer 400 have different architectures.

[0122] The storage device 408 is a non-transitory computer-readable storage medium such as a hard drive, compact disk read-only memory (CD-ROM), DVD, or a solid-state memory device. The memory 406 holds instructions and data used by the processor 402. The input interface 414 is a touch-screen interface, a mouse, track ball, or other type of pointing device, a keyboard 410, or some combination thereof, and is used to input data into the computer 400. In some embodiments, the computer 400 may be configured to receive input (e.g., commands) from the input interface 414 via gestures from the user. The graphics adapter 412 displays images and other information on the display 418. The network adapter 416 couples the computer 400 to one or more computer networks.

[0123] The computer 400 is adapted to execute computer program modules for providing functionality described herein. As used herein, the term “module” refers to computer program logic used to provide the specified functionality. Thus, a module can be implemented in hardware, firmware, and / or software. In one embodiment, program modules are stored on the storage device 408, loaded into the memory 406, and executed by the processor 402.

[0124] The types of computers 400 used by the entities of FIG. 1 can vary depending upon the embodiment and the processing power required by the entity. For example, the activity prediction system 130 can run in a single computer 400 or multiple computers 400 communicating with each other through a network such as in a server farm. The computers 400 can lack some of the components described above, such as graphics adapters 412, and displays 418.IX. Cancers

[0125] In some embodiments, the present disclosure provides a method for predicting responsiveness in an individual of a tumor to treatment by an immune checkpoint inhibitor. In some embodiments, the tumor is a cancerous tumor. In some embodiments, the cancer is a B cell lymphoma, subcutaneous melanoma, a highly aggressive metastatic subcutaneousmelanoma, colorectal carcinoma, a cancer that undergoes epithelial to mesenchymal transition, hepatocellular carcinoma, Lewis lung carcinoma, a renal malignancy, prostate adenocarcinoma, kidney renal clear cell carcinoma (KIRC), skin cutaneous melanoma (SKCM), bladder cancer (BLCA), or stomach adenocarcinoma (STAD). In some embodiments, the cancer is s squamous cell carcinoma, basal cell carcinoma, breast cancer, head and neck carcinoma, thyroid carcinoma, soft tissue sarcoma, bone sarcoma, testicular cancer, prostatic cancer, ovarian cancer, bladder cancer, skin cancer, brain cancer, glioblastoma, medulloblastoma, ependymoma, angiosarcoma, hemangiosarcoma, mast cell tumor, primary hepatic cancer, small cell lung cancer, non-small-cell lung cancer, pancreatic cancer, gastrointestinal cancer, renal cell carcinoma, hematopoietic neoplasia, lymphoma, mesothelioma, glioblastoma, low-grade glioma, high-grade glioma, pediatric brain cancer, medulloblastoma, or a metastatic cancer thereof. In some embodiments, the cancer is caused by an oncolytic virus.X. Therapies

[0126] In some embodiments of the present disclosure, individuals predicted to be responders to checkpoint inhibitors by evaluation of the expression levels of the gene signature biomarker panel are administered a therapy. In some embodiments the therapy is one or more checkpoint inhibitors. In some embodiments the one or more checkpoint inhibitors are selected from a group of checkpoint inhibitors comprising a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bis), an antagonist of programmed death 1 (PD- 1), a programmed death ligand 1 (PD-L1) inhibitor, a cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) inhibitor, V-domain Ig suppressor of T cell activation (VISTA) inhibitor, a programmed death ligand 2 (PD-L2) inhibitor, an indoleamine 2,3 -dioxygenase (IDO) inhibitor, an arginase, a B7 family inhibitory ligand, a lymphocyte activation gene 3 (LAG3) inhibitor, a B and T lymphocyte attenuator (BTLA), a T cell membrane protein 3 (TIM3) inhibitor, and an adenosine A2a receptor (A2a) inhibitor.

[0127] In some embodiments, the checkpoint inhibitor is a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi). Example CBL-Bi include either Compound 1, or Compound 2, as shown in Table 3. Further example CBL-Bi are disclosed in W020220221704, which is incorporated by reference in its entirety.Table 3: Exemplary CBL-Bi compounds

[0128] In some embodiments of the present disclosure, the checkpoint inhibitor is administered prior to, subsequent to, or concurrent with another cancer therapy. In some embodiments, the checkpoint inhibitor is administered zero days, one day, two days, three days, four days, five days, six days, one week, two weeks, three weeks, four weeks, five weeks, or six weeks after completion of another cancer therapy. In some embodiments, the checkpoint inhibitor is administered zero days, one day, two days, three days, four days, five days, six days, one week, two weeks, three weeks, four weeks, five weeks, or six weeks prior to administration of another cancer therapy. In some embodiments, the checkpoint inhibitor is administered concurrent with another cancer therapy. In some embodiments, other cancer therapies comprise chemotherapy, radiation therapy, or hematopoietic stem cell therapy. In some embodiments, the checkpoint inhibitor is administered at a therapeutically effective amount.XI. Kit Implementation

[0129] Also disclosed herein are kits for predicting responders or non-responders to treatment with checkpoint inhibitors (e.g., tumor reduction) in an individual. Such kits can include reagents for detecting expression levels of one or biomarkers and instructions for assessing predicting responders or non-responders to treatment with checkpoint inhibitors based on the detected expression levels.

[0130] The detection reagents can be provided as part of a kit. Thus, the invention further provides kits for detecting the presence of a panel of biomarkers of interest in a biological test sample. A kit can comprise a set of reagents for generating a dataset via at least one protein detection assay (e.g., immunoassay) that analyzes the test sample from the subject. In various embodiments, the set of reagents enable detection of quantitative expression levels of biomarkers from any one of Tables 1 or 2. In particular embodiments, the set of reagents enable detection of quantitative expression levels of biomarkers of Table 1. In particular embodiments, the set of reagents enable detection of quantitative expression levels of biomarkers of Table 2. In certain aspects, the reagents include one or more antibodies that bind to one or more of the markers. The antibodies may be monoclonal antibodies or polyclonal antibodies. In some aspects, the reagents can include reagents for performing ELISA including buffers and detection agents.

[0131] A kit can include instructions for use of a set of reagents. For example, a kit can include instructions for performing at least one biomarker detection assay such as an immunoassay, a protein-binding assay, an antibody-based assay, an antigen-binding proteinbased assay, a protein-based array, an enzyme-linked immunosorbent assay (ELISA), flow cytometry, a protein array, a blot, a Western blot, nephelometry, turbidimetry, chromatography, mass spectrometry, enzymatic activity, proximity extension assay, and an immunoassay selected from RIA, immunofluorescence, immunochemiluminescence, immunoelectrochemiluminescence, immunoelec trophoretic, a competitive immunoassay, and immunoprecipitation.

[0132] In various embodiments, the kits include instructions for practicing the methods disclosed herein (e.g., methods for training or deploying a predictive model to predict responders or non-responders to treatment with checkpoint inhibitors). These instructions can be present in the subject kits in a variety of forms, one or more of which can be present in the kit. One form in which these instructions can be present is as printed information on a suitable medium or substrate, e.g., a piece or pieces of paper on which the information is printed, in the packaging of the kit, in a package insert, etc. Yet another means would be a computer readable medium, e.g., diskette, CD, hard-drive, network data storage, etc., on which the information has been recorded. Yet another means that can be present is a website address which can be used via the internet to access the information at a removed site. Any convenient means can be present in the kits.XII. Systems

[0133] Further disclosed herein are system for analyzing quantitative expression levels of biomarkers for predicting responders or non-responders to treatment with checkpoint inhibitors. In various embodiments, such a system can include a set of reagents for detecting expression levels of biomarkers in the biomarker panel, an apparatus configured to receive a mixture of the set of reagents and a test sample obtained from a subject to measure the expression levels of the soluble mediators, and a computer system communicatively coupled to the apparatus to obtain the measured expression levels and to implement the predictive model to assess the disease activity.

[0134] The set of reagents enable the detection of quantitative expression levels of the biomarkers in the biomarker panel. In various embodiments, the set of reagents involve reagents used to perform an assay, such as an assay or immunoassay as described above. For example, the reagents include one or more antibodies that bind to one or more of the biomarkers. The antibodies may be monoclonal antibodies or polyclonal antibodies. As another example, the reagents can include reagents for performing ELISA including buffers and detection agents.

[0135] The apparatus is configured to detect expression levels of biomarkers in a mixture of a reagent and test sample. For example, the apparatus can determine quantitative expression levels of biomarkers through an immunologic assay or assay for nucleic acid detection. The mixture of the reagent and test sample may be presented to the apparatus through various conduits, examples of which include wells of a well plate (e.g., 96 well plate), a vial, a tube, and integrated fluidic circuits. As such, the apparatus may have an opening (e.g., a slot, a cavity, an opening, a sliding tray) that can receive the container including the reagent test sample mixture and perform a reading to generate quantitative expression values of biomarkers. Examples of an apparatus include a plate reader (e.g., a luminescent plate reader, absorbance plate reader, fluorescence plate reader), a spectrometer, and a spectrophotometer.

[0136] The computer system, such as example computer 400 described in FIG. 4, communicates with the apparatus to receive the quantitative expression values of biomarkers. The computer system implements, in silico, a predictive model to analyze the quantitative expression values of the biomarkers to predict an assessment of the disease activity.Examples

[0137] Below are examples of specific embodiments for carrying out the present invention. The examples are offered for illustrative purposes only and are not intended to limit the scope of the present invention in any way. Efforts have been made to ensure accuracy with respect to numbers used (e.g., amounts, temperatures, etc.), but some experimental error and deviation should be allowed for.Example 1: Exemplary Biomarker Panel Successfully Distinguishes Responders and Non-responders

[0138] A biomarker panel of an eight-gene gene signature (Table 2) was predictive of a response of tumors to checkpoint inhibitors. Datasets analyzing molecular response to treatment with immune checkpoint inhibitors (ICIs) were obtained from the CRI Atlas Portal (https: / / cri-iatlas.org / ) for patients with kidney renal clear cell carcinoma (KIRC), skin cutaneous melanoma (SKCM), bladder cancer (BLCA), and stomach adenocarcinoma (ST AD) were treated with an anti-PDl checkpoint inhibitor treatment. The datasets are summarized in Table 4. Patients were categorized as responders with mRECIST of Partial Response or Complete Response or categorized as non-responders if they had Progressive Disease or Stable Disease (FIG. 5). Datasets without response information or without responding patients were excluded from the analysis. Evaluation of the expression score of the gene signature showed a significant difference in expression of the gene signature between responders and non-responders to the anti-PDl treatment. The p-values as determined by a Wilcoxon test are shown in Table 5.Table 4: Summary of immune checkpoint inhibitor datasetsTable 5: Significance determined by Wilcoxon testExample 2; Exemplary Experimental Studies Identify Biomarkers that Distinguish Responders and Non-Responders

[0139] Multiple animal and human studies were used for the development of this biomarker panel as illustrated in FIG. 6. Gene expression was measured in exemplary syngeneic mouse tumor models and human PBMCs treated with checkpoint inhibitors by multiplex nucleic acid hybridization analysis and in mouse PBMCs isolated from a murine pharmacodynamic study and human PBMCs treated with checkpoint inhibitors by RNA sequencing analysis. Genes with altered differential expression common across two or more assays were used to generate an eight-gene gene signature.Example 2A: Gene Expression Analysis of Murine Tumor ModelsCorrelation of tumor response with gene expression analysis

[0140] Ten exemplary syngeneic mouse tumor models covering nine different tumor types (Table 6) were analyzed for tumor growth inhibition after treatment with the checkpointinhibitor, Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi; Compound 1 shown in Table 3). Changes in gene expression levels under different conditions were analyzed by multiplex nucleic acid hybridization analysis using a mouse pan cancer immune profiling panel that includes 770 genes. Correlation analysis of two exemplary genes, Ifi-441 and Rora, from a differential expression dataset indicated that gene expression alterations (Logs fold change) correlated with tumor growth inhibition (TGI; FIG. 7). Based on tumor growth inhibition, the mouse models were categorized as responder or non-responder models.Table 6: Mouse Tumor ModelsGene expression differences between responders and non- responders

[0141] The Logz FC was determined for Cxcl-10 and Cxcl9 in each responder and non- responder mouse model. For both genes there was a significant difference in the Log2 FC when comparing the non-responders to responders indicating analysis of gene expression is predictive of the tumor response to treatment with CBL-Bi. FIG. 8 depicts a graph of box plots of Log2FC in gene expression of CXCL10 and CXCL9 of responders (R) and non- responders (NR). Each circle or triangle represents data from individual mouse models indicated by the key. The percent of tumor growth inhibition (TGI) in responder models is indicated in the table above the graph.Commonly upregulated genes in responder models

[0142] The differentially expressed genes in the responder models were compared in a Venn diagram to determine genes common to all or subsets of the responder models (FIG. 9). Across all responder models there were 6 common genes (Thyl , Irgm2, H2, T23, Zbpl ,Serpingl, and Cxcl9). Additional differentially expressed genes were common to subsets of the responder models.Example 2B: Gene Expression in mPBMCs from a Tumor Model Treated with a Checkpoint InhibitorPharmacodynamic study of a mouse tumor model

[0143] Using the mouse tumor model H22 (Hepatocellular carcinoma; responder) a pharmacodynamic study was performed to analyze gene expression alterations in peripheral blood and relate expression changes to changes in tumor and lymph nodes. Mice were treated with either vector (V), 1 milligrams per kilogram (mpk), 6 mpk, or 20 mpk of a CBL-Bi (Compound 2 shown in Table 3) at days 1, 3, and 7. Tumor tissue, tumor draining lymph node and mPBMC samples were harvested 4 hours after the last dose. Gene expression analysis of the mPBCMs treated on day 1 with 6 mpk CBL-Bi by RNA-sequencing identified 133 altered genes (set 1 ; S I).Example 2C: Gene Expression in Human PBMCs Treated with a Checkpoint InhibitorAnalysis of Human PBMCs by multiplex nucleic acid hybridization analysis

[0144] Human PBMCs were treated with 50 nM or 1 uM CBL-Bi (Compound 2 shown in Table 3) at 6 h, 24 h, 34 h, and 48 h under three different stimulation conditions (no stimulation, aCD3 stimulation, or aCD3CD28 stimulation). Changes in gene expression levels under different conditions were analyzed by multiplex nucleic acid hybridization analysis using a human immune exhaustion panel that includes 785 genes covering exhaustion in lymphocytes. The gene with the greatest dose-dependent response was IFNG.Example 2D: Analysis of Gene Expression in Human PBMCs Treated With a Checkpoint InhibitorAnalysis of Human PBMCs by RNA-sequencing

[0145] Human PBMCs were treated with 10 nM, 100 nM, or 1 uM CBL-Bi (Compound 2 shown in Table 3) at 6 h and 24 h under two different stimulation conditions (no stimulation or aCD3 stimulation). Changes in gene expression levels under different conditions were analyzed by RNA-sequencing.Example 2E: Identifying Commonly Differentially Expressed Biomarkers from the Experimental Studies

[0146] Three gene sets were developed by further analysis of the genes identified by multiplex nucleic acid hybridization analysis: genes with a log2 fold change after CBL-Bi treatment that correlated with tumor growth inhibition (set 2; S2), genes with a log2 foldchange after CBL-Bi treatment that differed significantly between responder and nonresponder models (set 3; S3), and genes that were upregulated after CBL-Bi treatment across at least three of the responder models (set 4; S4).

[0147] A Venn diagram was used to compare genes common to sets 1, 2, 3, and 4 (FIG. 10). In total there were 53 genes found in set 1 from analysis of the mPBCMs treated on day 1 with 6 mpk CBL-Bi that were also found in at least one of sets 2, 3, or 4 (Table 1).

[0148] Of the 53 genes identified from the mouse tumor study and in the pharmacodynamic studies, four genes were also upregulated in the human PBMC RNA-sequencing dataset: CXCL9, GBP5, IRF1, and CD274. An additional four genes were found to be commonly upregulated between the pharmacodynamic studies and the hPBMC study: CXCL9, IFIT3, OAS2, and STAT1. A combination of genes was selected based on commonality among the exemplary human and mouse studies as shown below in Table 7. This eight-gene gene signature for determining responsiveness to treatment checkpoint inhibitors is: CXCL9, GBP5, IRF1, CD274, STAT1, PSMB9, IFIT3, and IFNG.Table 7: Selection of Commonly Differentially Expressed Genes Across MultipleExperimentsExample 3A: Treatment with a Checkpoint Inhibitor Results in a Consistant Gene Signature Expression Response Across Species

[0149] As shown in the schematic of FIG. 11 A, mouse PBMCs, mouse tumor models, and human PBMCs were treated with a CBL-Bi (Compound 2) and evaluated by multiplex nucleic acid hybridization analysis for expression of the gene signature of Example 2E. The response of seven of the eight genes in the eight-gene signature (CXCL9, GBP5, IRF1, CD274, STAT1, PSMB9, and IFNG) that were included for detection on the commercially available Nanostring Immunology V2 Chip was evaluated.

[0150] As shown in FIG. 11B, the mean signature expression increased on a pre-clinical in vivo mouse hepatocellular carcinoma model (H22) treated with 1 mpk, 6 mpk, or 12 mpk of the CBL-Bi demonstrating a dose response.

[0151] Human patients were treated with 5 mg, 10 mg, 20 mg, 30 mg, or 40 mg of a CBL-Bi during a clinical study. RNA was extracted from patient peripheral blood samples taken at various timepoints during the first cycle of treatment with the CBL-Bi (Compound 2) and tested on a Nanostring Immunology V2 chip for expression analysis of the gene signature. FIG. 11C shows a dose related increase of the gene signature expression in patient peripheral samples.Example 3B: Tumor Infiltrating Lymphocytes Score Corresponds with Clinical Benefit and Increased Gene Signature Expression

[0152] Human patient tumor samples were collected from patients treated with a CBL-Bi as described in Example 3A during a clinical study. Next generation RNA transcriptome sequencing was performed. The RNA transcriptome data was analyzed to derive the specific gene signature responses.

[0153] The total tumor infiltrating lymphocyte (TIL) score signature was evaluated on baseline tumor samples. The TIL score signature was determined in accordance with the methods described in Danaher, P et al. Gene expression markers of tumor infiltrating leukocytes. J. Immunol. Of Cancer. 2017; (5): 18, which is incorporated by reference in its entirety. Patients who tended to benefit clinically from CBL-Bi treatment showed a higher TIL score in their tumors at baseline (FIG. 12A).

[0154] The CBL-Bi response signature was evaluated in Day 15 biopsies from patients. Expression of the eight genes in the eight-gene signature of Example 2E (CXCL9, GBPS, IRF1, CD274, STAT1, PSMB9, IFIT3, and IFNG) was analyzed by RNA sequencing. FIG. 12B shows that those patients who benefited from CBL-Bi treatment exhibited an increase in the eight-gene signature as early as Day 15 of the first cycle of treatment compared to thosewho did not benefit. Altogether, these results indicate that the CBL-Bi gene signature appropriately distinguishes between individuals that are likely to exhibit a clinical benefit in comparison to individuals that are likely to exhibit no benefit.Example 4; CBL-B Pathway proximal biomarkers can be identified by evaluation of the effect of CBL-Bi on TCR signaling

[0155] This Example has been presented for the purpose of illustration; it is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Persons skilled in the relevant art can appreciate that many modifications and variations are possible in light of the teachings.

[0156] It should be noted that the language used in Example 4 has been principally selected for readability and instructional purposes, and it may not have been selected to delineate or circumscribe the inventive subject matter. Accordingly, the disclosure of Example 4 is intended to be illustrative, but not limiting, of the scope of the invention.Background

[0157] Casitas B-lineage lymphoma proto-oncogene b (CBL-B) is an E3 ubiquitin-protein ligase that serves as a critical regulator of immunity. Substantial preclinical evidence supports CBL-B inhibition, as a potent driver of anti-tumor immunotherapy. An example of an orally bioavailable, selective, small molecule, allosteric CBL-B inhibitor (CBL-Bi) includes Compound 2, which has advanced into Phase 1 / 2 clinical trial (NCT05662397). As part of the translation of mechanism of action to clinical exploration, efforts were made to identify and characterize proximal biomarkers capable of monitoring CBL-Bi clinically.Methods

[0158] A phosphoproteomics study investigated alterations in phosphorylation sites on proteins in Jurkat cells activated by anti-CD3 antibody cross-linking (anti-CD3), in the presence and absence of CBL-Bi. Selected phosphorylation changes were validated using a Meso Scale Discovery (MSD) assay, and flow cytometry in human T cells. A flow-based assay was employed to validate the biomarkers in mouse peripheral blood mononuclear cells (PBMCs) from animals treated with CBL-Bi. CBL-Bi's impact on potential substrates of CBL-B ’s E3 ligase activity were monitored for changes in treated human PBMCs using flow cytometry.Results

[0159] When comparing Jurkat cells treated with anti-CD3 to those treated with anti-CD3 and CBL-Bi, significant differences were observed in tyrosine, serine, and threoninephosphorylation. Among them were phosphorylation sites on ZAP70, CD3E, LAT, and PLCG1 , key proteins downstream of T cell receptor (TCR) signaling. Examination of ZAP70 revealed dose-dependent increases in tyrosine phosphorylation in activated human T cells. This effect was also measured by flow cytometry in mouse PBMCs treated in vivo. Despite the desirability of ZAP70 phosphorylation as a proximal biomarker for CBL-Bi, quantifying phosphoproteins clinically poses significant challenges. Increased abundance of potential substrates of CBL-B’s E3 ligase activity was also observed upon CBL-Bi. Treatment of TCR- activated human PBMCs demonstrated dose-dependent increases in Notch 1 and IGF1R on the surface of CD4 and CD8+ T cells, measured by flow cytometry. CBL-Bi concentration- related changes in these markers following immune stimulation, measured by absolute levels and increases from baseline, suggest that one or both proteins could be used as proximal biomarkers for monitoring CBL-B pathway inhibition in patients treated with a CBL-Bi.Conclusion

[0160] Evaluating the effect of CBL-Bi on TCR signaling enabled identification of a repertoire of proximal biomarkers with potential to monitor and optimize CBL-Bi clinically. We believe clinical validation and exploration are warranted to assess the utility of these proximal biomarkers and their association with downstream pharmacodynamics following treatment with a CBL-Bi in the clinic.

Claims

CLAIMS1. A method for determining responsiveness to an immune checkpoint inhibitor in a subject, the method comprising: a. obtaining or having obtained at least one dataset comprising expression levels of a plurality of biomarkers, wherein the plurality of biomarkers comprise four or more of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1 ; and b. generating a prediction of responsiveness to an immune checkpoint inhibitor by applying a predictive model to the expression levels of the plurality of biomarkers.

2. The method of claim 1 , wherein the immune checkpoint inhibitor is Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi).

3. The method of any one of claims 1-2, wherein the plurality of biomarkers comprise five or more of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1.

4. The method of any one of claims 1-2, wherein the plurality of biomarkers comprise six or more of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1.

5. The method of any one of claims 1-2, wherein the plurality of biomarkers comprise seven or more of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1.

6. The method of any one of claims 1-2, wherein the plurality of biomarkers comprise each of CD274, CXCL9, GBP5, IFIT3, IFNG, IRF1, PSMB9, and STAT1.

7. The method of any one of claims 1-6, wherein the plurality of biomarkers are identified by analyzing data of a plurality of experimental studies, wherein the plurality of experimental studies comprises two or more, three or more, or four or more experimental studies.

8. The method of claim 7, wherein the data of the plurality of experimental studies comprises gene expression data determined by one or more of multiplex nucleic acid hybridization analysis or RNA sequencing analysis.

9. The method of claim 8, wherein the data of the plurality of experimental studies comprises gene expression data determined by multiplex nucleic acid hybridization analysis of murine cells.

10. The method of claim 9, wherein the murine cells comprise tumor cells or PBMCs.

11. The method of claim 9 or 10, wherein the murine cells were previously exposed to an immune checkpoint inhibitor.

12. The method of claim 8, wherein the data of the plurality of experimental studies comprises gene expression data determined by multiplex nucleic acid hybridization analysis of human cells.

13. The method of claim 12, wherein the human cells comprise PBMCs.

14. The method of claim 12 or 13, wherein the human cells were previously exposed to an immune checkpoint inhibitor.

15. The method of claim 7, wherein the plurality of experimental studies comprises a multiplex nucleic acid hybridization analysis of a plurality of tumor cells isolated from mice of at least five, at least six, at least seven, at least eight, at least nine, or at least ten murine tumor models.

16. The method of claim 7, wherein the plurality of experimental studies comprises a multiplex nucleic acid hybridization analysis of PBMCs isolated from mice.

17. The method of claim 15 or 16, wherein the mice were previously exposed to an immune checkpoint inhibitor.

18. The method of claim 7, wherein the plurality of experimental studies comprises a multiplex nucleic acid hybridization analysis of human PBMCs.

19. The method of claim 7, wherein the plurality of experimental studies comprises a RNA sequencing analysis of human PBMCs.

20. The method of claim 18 or 19, wherein the human PBMCs were previously exposed to an immune checkpoint inhibitor.

21. The method of claim 7, wherein the plurality of experimental studies comprise two or more, three or more, or four of: a. a multiplex nucleic acid hybridization analysis of a plurality of cells isolated from a plurality of murine tumor models wherein mice are treated with an immune checkpoint inhibitor; b. a multiplex nucleic acid hybridization analysis of PBMCs isolated from mice treated with an immune checkpoint inhibitor for pharmacodynamic studies; c. a multiplex nucleic acid hybridization analysis of human PBMCs treated with an immune checkpoint inhibitor; and d. a RNA sequencing analysis of human PBMCs treated with an immune checkpoint inhibitor.

22. The method of claim 21, wherein the murine tumor models are characterized as responders or non responders to treatment with an immune checkpoint inhibitor.

23. The method of claim 22, wherein responders to treatment with an immune checkpoint inhibitor are characterized by tumor growth inhibition of >30%.

24. The method of claim 22 or 23, wherein the plurality of cells is isolated from ten murine tumor models that are characterized as responders or as non-responders.

25. The method of any one of claims 1-24, further comprising: administering the immune checkpoint inhibitor to the subject if the generated prediction of responsiveness indicates the subject is a responder to the immune checkpoint inhibitor.

26. The method of any one of claims 1-25, further comprising: modifying a treatment for the subject if the generated prediction of responsiveness indicates the subject is a non-responder to the immune checkpoint inhibitor.

27. A method for identifying a plurality of biomarkers for determining responsiveness to an immune checkpoint inhibitor in a subject, the method comprising: a. obtaining or having obtained a dataset from two or more experimental studies, wherein each experimental study identifies differentially expressed genes wherein the dataset comprises: i. Gene expression data determined by multiplex nucleic acid hybridization analysis of nucleic acids isolated from mice of a plurality of murine tumor models wherein the mice are treated with an immune checkpoint inhibitor; ii. Gene expression data determined by multiplex nucleic acid hybridization analysis of PBMCs isolated from mice treated with an immune checkpoint inhibitor for pharmacodynamic studies; hi. Gene expression data determined by multiplex nucleic acid hybridization (NanoString) analysis of human PBMCs treated with an immune checkpoint inhibitor; and iv. Gene expression data determined by RNA sequencing analysis of human PBMCs treated with an immune checkpoint inhibitor, and b. comparing differentially expressed genes across each of the two or more experimental studies, and wherein comparing differentially expressed genes across each of the four or more experimental studies further comprises identifying one or more commonly differentially expressed genes that are differentially expressed in at least one of the two or more experimental studies to form a plurality of biomarkers.

28. The method of claim 27, wherein the at least five, at least six, at least seven, at least eight, at least nine, or at least ten murine tumor models are characterized as responders or non responders to treatment with an immune checkpoint inhibitor, wherein responsiveness to treatment with an immune checkpoint inhibitor is characterized as tumor growth inhibition of >30%.

29. The method of claim 28, further comprising determining genes that have a significant differences in log2 fold change when comparing gene expression data, wherein log2 fold changes are calculated for each model by comparing gene expression data from animals treated with an immune checkpoint inhibitor to animals treated with vehicle control.

30. The method of claim 28, further comprising determining genes in mouse models with detected tumor growth inhibition (TGI) that have a log2 fold change that correlates with tumor growth inhibition.

31. The method of claim 28, further comprising determining genes upregulated in at least three of the responders.

32. The method of claim 27, wherein the gene expression dataset determined by multiplex nucleic acid hybridization analysis of PBMCs isolated from mice treated with an immune checkpoint inhibitor for pharmacodynamic studies comprises expression data of upregulated genes collected from mice treated with 6 mpk of CBL-Bi.

33. The method of claim 27, wherein the expression data of upregulated genes is compared to each of (i) gene expression data determined by multiplex nucleic acid hybridization analysis, (iii) gene expression data determined by multiplex nucleic acid hybridization analysis of human PBMCs, and (iv) gene expression data determined by RNA sequencing analysis of human PBMCs to determine a set of commonly differentially expressed genes.

34. The method of claim 27, wherein the set of commonly differentially expressed genes comprises at least four, at least five, at least six, at least seven, or at least eight biomarkers.

35. The method of claim 34, wherein the at least four biomarkers comprise CXCL9, IFIT3, OAS2, and STAT1.

36. The method of claim 34, wherein the at least four biomarkers comprise CXCL9, GBP5, IRF1 and CD274.

37. The method of claim 34, wherein the at least four biomarkers comprise CXCL9,GBP5, IRF1, CD274, STAT1, PSMB9, and IFNG.

38. The method of claim 34, wherein the at least four biomarkers comprise CXCL9, GBP5, IRF1, CD274, STAT1, PSMB9, IFIT3, and IFNG.

39. The method of any one of claims 27-38, wherein the determined set of commonly differentially expressed genes form a gene signature, wherein at least 85% of the selected biomarkers of the gene signature are differentially expressed in at least two experimental studies.

40. The method of any one of claims 27-38, wherein the determined set of commonly differentially expressed genes form a gene signature, wherein at least 75% of the selected biomarkers of the gene signature are differentially expressed in at least three experimental studies.

41. A method for determining effects of a Casitas B-lineage lymphoma proto-oncogene b inhibitor (CBL-Bi), the method comprising: obtaining statuses of a plurality of biomarkers of a sample obtained from a subject that was previously administered the CBL-Bi, wherein the plurality of biomarkers comprise Notch 1 and IGF1R; using the obtained statuses, determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples; and determining the effects of the CBL-Bi based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples.

42. The method of claim 41 , wherein the statuses of the plurality of biomarkers are measured from the sample comprising peripheral blood mononuclear cells (PBMCs).

43. The method of claim 42, wherein the PBMCs comprise CD4+ and / or CD8+ T cells.

44. The method of claim 43, further comprising: obtaining the sample obtained from the subject; and providing one or both of anti-CD3 antibodies and anti-CD28 antibodies to stimulate the PBMCs.

45. The method of claim 44, wherein the PBMCs are stimulated for at least 24 hours.

46. The method of any one of claims 41-45, wherein determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that one of Notchl and IGF1R are expressed at a higher level in the sample than in the one or more reference samples.

47. The method of claim 46, wherein determining the effects of the CBL-Bi based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that the CBL-Bi is effective in response to determining that one of Notchl and IGF1R are expressed at a higher level in the sample.

48. The method of any one of claims 41-45, wherein determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that neither of Notchl or IGF1R are expressed at a higher level in the sample than in the one or more reference samples.

49. The method of claim 48, wherein determining the effects of the CBL-Bi based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that the CBL-Bi is not effective in response to determining that neither of Notchl or IGF1R are expressed at a higher level in the sample.

50. The method of any one of claims 41-49, wherein the statuses of the plurality of biomarkers comprise expression levels of at least a subset of the plurality of biomarkers.

51. The method of any one of claims 41-50, wherein the plurality of biomarkers further comprise one or more of ZAP70, CD3E, LAT, and PLCG1.

52. The method of claim 51, wherein the statuses of the plurality of biomarkers comprises phosphorylation statuses of one or more amino acids of one or more of ZAP70, CD3E, LAT, and PLCGl.

53. The method of claim 52, wherein the one or more amino acids comprise one or more of tyrosine, serine, and threonine.

54. A method for determining a therapy for a subject, the method comprising: obtaining statuses of a plurality of biomarkers of a sample obtained from a subject that was previously administered the CBL-Bi, wherein the plurality of biomarkers comprise Notchl and IGF1R; using the obtained statuses, determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples; and selecting a therapy for the subject based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples.

55. The method of claim 54, wherein the statuses of the plurality of biomarkers are measured from the sample comprising peripheral blood mononuclear cells (PBMCs).

56. The method of claim 55, wherein the PBMCs comprise CD4+ and / or CD8+ T cells.

57. The method of claim 56, further comprising: obtaining the sample obtained from the subject; and providing one or both of anti-CD3 antibodies and anti-CD28 antibodies to stimulate the PBMCs.

58. The method of claim 57, wherein the PBMCs are stimulated for at least 24 hours.

59. The method of any one of claims 54-58, wherein determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that one of Notchl and IGF1R are expressed at a higher level in the sample than in the one or more reference samples.

60. The method of claim 59, wherein selecting the therapy for the subject based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that the CBL-Bi is effective in response to determining that one of Notchl and IGF1R are expressed at a higher level in the sample.

61. The method of any one of claims 54-58, wherein determining whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that neither of Notchl or IGF1R are expressed at a higher level in the sample than in the one or more reference samples.

62. The method of claim 61, wherein selecting the therapy for the subject based on the determination of whether the plurality of biomarkers are differentially expressed in comparison to one or more reference samples comprises determining that the CBL-Bi is not effective in response to determining that neither of Notchl or IGF1R are expressed at a higher level in the sample.

63. The method of any one of claims 54-62, wherein the statuses of the plurality of biomarkers comprise expression levels of at least a subset of the plurality of biomarkers.

64. The method of any one of claims 54-63, wherein the plurality of biomarkers further comprise one or more of ZAP70, CD3E, LAT, and PLCG1.

65. The method of claim 64, wherein the statuses of the plurality of biomarkers comprises phosphorylation statuses of one or more amino acids of one or more of ZAP70, CD3E, LAT, and PLCGl.

66. The method of claim 65, wherein the one or more amino acids comprise one or more of tyrosine, serine, and threonine.

67. The method of any one of claims 54-66, wherein selecting the therapy for the subject comprises selecting a dose of CBL-Bi that differs from a dose of CBL-Bi previously administered to the subject.

Images