Biomarkers for predicting tumor response to and toxicity of immunotherapy

Abstract

To provide biomarkers for predicting tumor response to and toxicity of immunotherapy.SOLUTION: The invention is directed to biomarkers for predicting a patient's response, both therapeutic and toxic, to immunotherapy. The invention is based, in part, on the discovery that cancer patients carrying one or more specified mutations in their genome may respond to treatment with an immune modulating agent more effectively than other cancer patients, for example, patients homozygous for the wild-type allele. The invention is also based, in part, on the discovery that cancer patients carrying one or more specified mutations in their genome may respond to treatment with an immune modulating agent less effectively than other patients, for example, cancer patients homozygous for the wild-type allele.SELECTED DRAWING: None

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to and the benefit of U.S. Provisional Patent Application No. 62 / 520,459, filed June 15, 2017, which is incorporated herein by reference in its entirety.

[0002] The present invention is directed to methods for predicting a cancer patient's response to immunomodulatory agents using biomarkers present in the cancer patient's germline genome. [Background technology]

[0003] Most cancer drugs are effective in some patients but not in others. This result is at least in part due to genetic variation between patients. Variable patient responses are particularly pronounced with immunotherapy. Therefore, the full potential of immunotherapy for treating cancer cannot be realized unless appropriate testing is performed to determine whether or not each patient will benefit from a drug.

[0004] Many patients experience toxic responses to anti-cancer immunotherapy, leading to treatment discontinuation; however, it is difficult to predict whether a patient will have a toxic response to a therapy prior to administration.

[0005] According to the National Institutes of Health (NIH), the term "biomarker" is defined as "a characteristic that is objectively measured and evaluated as an indicator of normal biological or pathogenic processes, or of the pharmacological response to a therapeutic intervention" (Biomarkers Definitions Working Group, 2001, Clin. Pharmacol. Ther. 69:89-95 ).

[0006] The development of improved diagnostics based on biomarker discovery has the potential to accelerate new drug development by proactively identifying patients most likely to have a clinical or toxic response to a given drug. This would significantly reduce the size, length, and cost of clinical trials. Technologies such as genomics, proteomics, and molecular imaging now allow for the rapid, sensitive, and reliable detection of specific gene mutations, expression levels of specific genes, and other molecular biomarkers. However, the clinical use of cancer biomarkers to predict response or toxicity remains largely unrealized, as few cancer biomarkers have been discovered. For example, a recent review article stated: There is an urgent need for the rapid development of biomarkers and their use to improve cancer diagnosis and treatment (Cho, 2007, Molecular Cancer 6:25). Another recent review article on cancer biomarkers included the following comment: With the advent of genomic profiling technologies and selective molecular targeted therapies, biomarkers have come to play an increasingly important role in the clinical management of cancer patients. Assays based on single gene / protein or multigene "signatures" have been introduced to measure specific molecular pathway deregulation to guide treatment decision-making as predictive biomarkers. Genomic-based prognostic biomarkers are also available for several cancer types for potential incorporation into clinical prognostic staging systems or implementation guidelines. However, due to the difficult challenges in the process of cancer biomarker development, there remains a large gap between early biomarker discovery studies and their translation to the clinic (Goosens et al. al., Transl. Cancer Res. 2015 4(3):256-269). Comments such as these illustrate the recognized need for the discovery of clinically useful biomarkers that can assist physicians in determining the most appropriate course of treatment for cancer patients.

[0007] Regarding cancer immunotherapy, some, but not all, patients respond to certain drugs. Therefore, it is necessary to understand which biomarkers in a patient's genome may be relevant in determining whether a patient will respond to a particular immunotherapy. For example, PD-L1, a ligand for PD-1, is highly expressed in some cancers; PD-1 is expressed, for example, on T cells. Inhibition of the interaction between PD-1 and its ligand can enhance immune responses and improve anti-tumor activity. However, not all patients respond to treatment with anti-PDL1 or anti-PD1 therapy, for example, anti-PDL1 antibody or anti-PD1 antibody. Therefore, there is a need for a diagnostic method based on predictive biomarkers to identify patients with cancer who may (or may not) respond to immunotherapy, for example, treatment with a PDL1 or PD1 inhibitor, such as anti-PDL1 or anti-PD1 antibody.

[0008] Furthermore, there is a need in the art to identify biomarkers capable of predicting whether a patient is likely to have a toxic response to a given immunotherapy, so that medical professionals can determine the best course of treatment prior to administration and so that patients can avoid toxic responses to such therapies. For example, there is a need in the art to identify biomarkers that will assist in predicting the toxicity of a given immunotherapy, such as anti-PDL1 or anti-PD1 therapy (e.g., anti-PDL1 antibody or anti-PD1 antibody), or radiation, in a patient. Even if a patient responds to such a therapy, if the therapy is toxic to the patient, it would be helpful for a physician to know this in advance and take the possible toxic response into consideration when determining whether a given immunotherapy is appropriate for the patient. There is a particular need to identify biomarkers found in or inherited in a patient's germline that predict a patient's systemic response to immunotherapy (both treatment and toxicity related) without the need to characterize the patient's specific tumor DNA. [Prior art documents] [Non-patent literature]

[0009] [Non-Patent Document 1] Biomarkers Definitions Working Group, 2001, Clin. Pharmacol. Ther. 69:89-95 [Non-patent document 2] Cho, 2007, Molecular Cancer 6:25 [Non-patent document 3] Goosens et al., Transl. Cancer Res. 2015 4(3):256-269 Summary of the Invention [Means for solving the problem]

[0010] The present invention is based, in part, on the discovery that cancer patients who carry one or more specified mutations in their genome may respond more effectively to treatment with an immunomodulatory agent than other cancer patients, e.g., patients who are homozygous for the wild-type allele. The present invention is also based, in part, on the discovery that cancer patients who carry one or more specified mutations in their genome may respond less effectively to treatment with an immunomodulatory agent than other cancer patients, e.g., cancer patients who are homozygous for the wild-type allele.

[0011] The present invention is also based, in part, on the discovery that some cancer patients may have a toxic response to immunomodulatory therapy compared to other cancer patients, e.g., patients homozygous for the wild-type allele, who do not experience a toxic response. The present invention is also based, in part, on the discovery that cancer patients who carry one or more specified mutations in their genome may not experience a toxic response to immunomodulatory agents compared to other patients, e.g., patients homozygous for the wild-type allele, who experience a toxic response.

[0012] In one aspect, the invention provides a method of treating cancer, comprising administering an immunomodulatory agent to a patient identified as not possessing a G nucleotide at a position corresponding to position 101 of SEQ ID NO:2 (CD274 / rs4742098).

[0013] In another aspect, the invention provides a method of treating cancer, comprising administering an immunomodulatory agent to a patient identified as carrying an A nucleotide at a position corresponding to position 101 of SEQ ID NO:7 (IL18R1 / rs11465660).

[0014] In another aspect, the present invention provides a method of treating cancer, comprising administering an immunomodulatory agent to a patient identified as not carrying a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021).

[0015] In another aspect, the invention provides a method of treating cancer comprising administering an immunomodulatory agent to a cancer patient identified as not heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102). In a further embodiment, the patient is also identified as heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660).

[0016] In another embodiment, the invention provides a method of treating cancer comprising administering an immunomodulatory agent to a cancer patient identified as not heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO:1 (CD44 / rs11821102), wherein the patient is further identified as not heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO:7 (IL18R1 / rs11465660) and not homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO:8 (miR99a promoter). In a further embodiment, the patient is identified as heterozygous for a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO:19 (EXO1 / rs4150021), and in another embodiment, the patient is identified as not heterozygous for a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO:19 (EXO1 / rs4150021).

[0017] In another embodiment, the present invention provides a method of treating cancer, comprising administering to a patient a mutation in: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); or c) a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); f) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); g) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660); h) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); i) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); or j) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483) The present invention provides methods for treating cancer, comprising administering an immunomodulatory agent to a cancer patient identified as having or not having one or more of:

[0018] In one embodiment, the patient: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); c) a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); or f) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter) and have been identified as not carrying one or more mutations selected from:

[0019] In yet another embodiment, the patient is a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); c) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); or e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483) These have been identified as carrying one or more mutations selected from:

[0020] In yet another embodiment, the present invention provides a method for determining a cancer patient's responsiveness to treatment with an immunomodulatory agent. The method comprises determining whether the patient is heterozygous for an A nucleotide (CD44 / rs11821102) at a position corresponding to position 101 of SEQ ID NO:1. Not being heterozygous for an A nucleotide (CD44 / rs11821102) at a position corresponding to position 101 of SEQ ID NO:1 indicates that the patient has a high probability of responding to the immunomodulatory agent. In a further embodiment, the method also comprises determining whether the patient is heterozygous for an A nucleotide (IL18R1 / rs11465660) at a position corresponding to position 101 of SEQ ID NO:7. Being heterozygous for an A nucleotide (IL18R1 / rs11465660) at a position corresponding to position 101 of SEQ ID NO:7 indicates that the patient has a high probability of responding to the immunomodulatory agent.

[0021] In yet another embodiment, the invention provides a method of determining a cancer patient's responsiveness to treatment with an immunomodulatory agent, comprising determining whether the patient is heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102), whether the patient is heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660), and whether the patient is homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter). Not being heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102), not being heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660), and not being homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 indicates that the patient has a high probability of responding to the immunomodulatory agent. In a further embodiment, the method comprises determining whether the patient is heterozygous for a deletion of a T nucleotide (EXO1 / rs4150021) at a position corresponding to position 101 of SEQ ID NO: 19; in one embodiment, being heterozygous for the deletion of a T nucleotide (EXO1 / rs4150021) at a position corresponding to position 101 of SEQ ID NO: 19 indicates that the patient has a high probability of responding to the therapy, while in another embodiment, not being heterozygous for the deletion of a T nucleotide (EXO1 / rs4150021) occurring in the wild-type sequence at a position corresponding to position 101 of SEQ ID NO: 19 indicates that the patient has a high probability of responding to the immunomodulatory agent.

[0022] In a further embodiment, the present invention provides a method of determining the responsiveness of a cancer patient to treatment with an immunomodulatory agent, comprising determining whether the patient carries one or more of the following mutations: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); c) a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); f) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); g) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660); h) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); i) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); or j) A C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483).

[0023] In one embodiment, the patient has the following mutation: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); c) a deletion of a T nucleotide occurring in the wild-type sequence at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); or f) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (mir99a promoter) If a patient does not have one or more of the following, they have a high probability of responding to an immunomodulatory agent.

[0024] In another embodiment, the patient has the following mutation: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); c) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); or e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483) If a patient has one or more of the following, they have a high probability of responding to an immunomodulatory agent.

[0025] In any of the above-described embodiments of the invention, the cancer may be melanoma or lung cancer, or in other embodiments, the cancer is melanoma (including unresectable or metastatic melanoma), lung cancer (including non-small cell lung cancer and metastatic non-small cell lung cancer), adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, cancer of the brain or central nervous system, basal cell skin cancer, breast cancer, cervical cancer. cancer), colon cancer, colorectal cancer, endometrial cancer, esophageal cancer, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumor, gastrointestinal stromal tumor (GIST), gastric cancer, glioma, glioblastoma, head and neck cancer (including head and neck squamous cell carcinoma), Hodgkin's disease, classical Hodgkin's lymphoma, diffuse large B-cell lymphoma, follicular lymphoma, Kaposi's sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia (including acute myeloid leukemia), liver cancer (including hepatocellular carcinoma), lymphoma, malignant mesothelioma, metastatic colorectal ... The cancer may be: Luke's cell carcinoma, metastatic urothelial carcinoma, multiple myeloma, myeloma, myelodysplastic syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroendocrine cancer, neuroblastoma, non-Hodgkin's lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumor, prostate cancer, kidney cancer (including renal cell carcinoma), retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, squamous cell skin cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine cancer, or vaginal cancer.

[0026] In any of the above-described embodiments of the invention, the immunomodulatory therapy may be an anti-PDL1 or anti-PD1 antibody or portion thereof.

[0027] In any of the above-described embodiments of the invention, the patient may be progression-free six months after starting treatment with the immunomodulatory agent.

[0028] In any of the above-described embodiments of the invention, the patient is preferably a human patient. The human patient may be a male or female patient.

[0029] In another aspect, the present invention provides a method of treating cancer with reduced toxicity, comprising administering an immunomodulatory agent to a patient suffering from cancer, wherein the patient has been identified as not possessing an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374).

[0030] In another aspect, the present invention provides a method of treating cancer with reduced toxicity, comprising administering an immunomodulatory agent to a patient suffering from cancer, wherein the patient has been identified as carrying a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370).

[0031] In another aspect, the invention provides a method of treating cancer with reduced toxicity, comprising administering an immunomodulatory agent to a patient suffering from cancer, wherein the patient has been identified as being neither heterozygous nor homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033). In a further embodiment, the patient has also been identified as being neither heterozygous nor homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790).

[0032] In another aspect, the invention provides a method of treating cancer with reduced toxicity, comprising administering an immunomodulatory agent to a patient suffering from cancer, wherein the patient has been identified as neither heterozygous nor homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033) and has been identified as heterozygous or homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370). In a further embodiment, the patient has also been identified as neither heterozygous nor homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790).

[0033] In another aspect, the present invention provides a method of treating cancer with reduced toxicity, comprising administering an immunomodulatory agent to a patient suffering from cancer, wherein the patient has one or more of the following mutations: a. A G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b. A C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c. A nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); d. A G nucleotide at a position corresponding to position 101 of SEQ ID NO:6 (IL10RB / rs2834167); e. A G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370); f. A C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); g. An A nucleotide at a position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374); h. A G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); i. A G nucleotide at a position corresponding to position 101 of SEQ ID NO:2 (CD274 / rs4742098); or j. A C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) The present invention provides a method for treating a cancer, wherein the cancer is identified as having or not having one or more of:

[0034] In a further embodiment, the immunomodulatory therapy is an anti-PD1 or anti-PDL1 antibody. According to this embodiment, the patient has the following mutation: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374); d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); or e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098) and in still further embodiments, the patient is identified as not carrying one or more of the following mutations: a) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370); d) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); or e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) have been identified as carrying one or more of the following:

[0035] In a further embodiment, the immunomodulatory therapy is radiation. In such an embodiment, the patient has the following mutation: a. A G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); b. A G nucleotide at a position corresponding to position 101 of SEQ ID NO:2 (CD274 / rs4742098); or c. A G nucleotide at a position corresponding to position 101 of SEQ ID NO:6 (IL10RB / rs2834167) and in a further embodiment, the patient has one or more of the following genotypes: a) heterozygous or homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); b) is not homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO:2 (CD274 / rs4742098); or c) is not homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167) have been identified as having one or more of the following:

[0036] In one embodiment, the radiation may be external beam radiation therapy, while in another embodiment, the radiation is brachytherapy, while in another embodiment, the radiation may be stereotactic body radiation therapy (SBRT).

[0037] In another embodiment, the present invention provides a method for determining the toxicity of an immunomodulatory agent in a cancer patient. The method includes determining whether the patient is heterozygous or homozygous for a C nucleotide (FCGR2A / rs10919033) at a position corresponding to position 101 of SEQ ID NO: 12. If the patient is neither heterozygous nor homozygous for a C nucleotide (FCGR2A / rs10919033) at a position corresponding to position 101 of SEQ ID NO: 12, the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent. In a further embodiment, it is also determined whether the patient is heterozygous or homozygous for a G nucleotide (TLR4 / rs4986790) at a position corresponding to position 101 of SEQ ID NO: 16. If the patient is neither heterozygous nor homozygous for a G nucleotide (TLR4 / rs4986790) at a position corresponding to position 101 of SEQ ID NO: 16, the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent.

[0038] In another embodiment, the present invention provides a method for determining the toxicity of an immunomodulatory agent in a cancer patient. The method includes determining whether the patient is heterozygous or homozygous for a C nucleotide (FCGR2A / rs10919033) at a position corresponding to position 101 of SEQ ID NO: 12 and whether the patient is heterozygous or homozygous for a G nucleotide (KRAS / rs61764370) at a position corresponding to position 101 of SEQ ID NO: 14. If the patient is neither heterozygous nor homozygous for a C nucleotide (FCGR2A / rs10919033) at a position corresponding to position 101 of SEQ ID NO: 12, and is heterozygous or homozygous for a G nucleotide (KRAS / rs61764370) at a position corresponding to position 101 of SEQ ID NO: 14, the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent. In a further embodiment, it is determined whether the patient is heterozygous or homozygous for a G nucleotide (TLR4 / rs4986790) at a position corresponding to position 101 of SEQ ID NO: 16. If a patient is neither heterozygous nor homozygous for the G nucleotide at the position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790), the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent.

[0039] The present invention also provides a method for determining the toxicity of an immunomodulatory agent in a cancer patient, the method comprising: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370); f) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); g) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374); h) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); i) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098), and j) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) The method includes determining whether the individual possesses one or more mutations selected from the group consisting of:

[0040] In one embodiment, the immunomodulatory agent is an anti-PD1 or anti-PDL1 antibody. a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374); d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); or e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098) In another such embodiment, the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent if the patient does not possess one or more mutations selected from: a) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370); d) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); or e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) When the patient has one or more mutations selected from the following:

[0041] In yet another embodiment, the immunomodulatory agent is radiation therapy. In such embodiments, the patient has one or more of the following mutations: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and / or c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167) While in further such embodiments, it is determined whether the patient carries one or more of the following genotypes: a) heterozygous or homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); b) is not homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO:2 (CD274 / rs4742098); or c) not homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO:6 (IL10RB / rs2834167); and if the patient has one of these genotypes, the patient is considered to have a low likelihood of exhibiting a toxic response to radiation.

[0042] In one embodiment, the radiation is external beam radiation, while in another embodiment, the radiation is stereotactic body radiation therapy, while in another embodiment, the radiation is brachytherapy.

[0043] In any of the above-described embodiments of the invention, the cancer may be melanoma or lung cancer, or in other embodiments, the cancer is selected from the group consisting of melanoma (including unresectable or metastatic melanoma), lung cancer (including non-small cell lung cancer and metastatic non-small cell lung cancer), adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, cancer of the brain or central nervous system, basal cell skin cancer, breast cancer, cervical cancer, colon cancer, colorectal cancer, endometrial cancer, esophageal cancer, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumor, gastrointestinal stromal tumor (GIST), gastric cancer, glioma, glioblastoma, head and neck cancer (including head and neck squamous cell carcinoma), Hodgkin's disease, classical Hodgkin's lymphoma, diffuse large bowel disease ... The cancer may be follicular B-cell lymphoma, follicular lymphoma, Kaposi's sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia (including acute myeloid leukemia), liver cancer (including hepatocellular carcinoma), lymphoma, malignant mesothelioma, Merkel cell carcinoma, metastatic urothelial carcinoma, multiple myeloma, myeloma, myelodysplastic syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroendocrine carcinoma, neuroblastoma, non-Hodgkin's lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumor, prostate cancer, kidney cancer (including renal cell carcinoma), retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, squamous cell skin cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine cancer, or vaginal cancer.

[0044] In any of the above-described embodiments of the invention, the patient is preferably a human patient. The human patient may be a male or female patient. [Brief explanation of the drawings]

[0045] [Figure 1-1]Figure 1 is a table showing single nucleotide mutations found in various human genes and the corresponding wild-type sequences. As described herein, these mutations are biomarkers relevant for predicting a patient's systemic response (both therapeutic and toxic) to treatment with immunomodulatory agents. The sequences shown are segments of the nucleotide sequence of the human gene; 100 nucleotides upstream (5') and downstream (3') of the mutation are shown. The mutation is indicated in the variant sequence by a square bracket "[]" and is at position 101; the corresponding wild-type nucleotide is indicated in the wild-type sequence by a square bracket "[]" at position 101. SEQ ID NOS: 1-16 and 33 are variant sequences, while SEQ ID NOS: 17-32 and 34 are wild-type versions of the sequences (i.e., without the mutation). [Figure 1-2] Same as above. [Figure 1-3] Same as above. [Figure 1-4] Same as above. [Figure 1-5] Same as above.

[0046] [Figure 2] Figure 2 is a tree-based classification rule that provides a prognostic indicator of a patient's response to treatment with an immunomodulator. Each leaf of the tree provides a probability of response to treatment based on the presence or absence of a particular biomarker shown therein, namely, CD44 / rs11821102, IL18R1 / rs11465660, miR99a promoter, and EXO1 / rs4150021. "0" means the patient is homozygous wild-type. "1" means the patient carries one copy of the mutation, i.e., the patient is heterozygous for the mutation. "2" means the patient is homozygous for the mutation. "Missing" means that data is unavailable for the patient.

[0047] [Figure 3]3 is a tree-based classification rule that provides a prognostic indicator for whether a patient will have a toxic response to treatment with an immunomodulator. Each leaf of the tree provides a probability of a toxic response based on the presence or absence of a particular biomarker shown therein, namely, FCGR2A / rs10919033, TLR4 / rs4986790, and KRAS / rs61764370. "0" means that the patient is homozygous wild-type. "1" means that the patient carries one or two copies of the mutation, i.e., the patient is homozygous or heterozygous for the mutation. "Missing" means that data is unavailable for the patient. DETAILED DESCRIPTION OF THE INVENTION

[0048] Single-nucleotide germline variations as biomarkers The present invention is based in part on the discovery that cancer patients who have one or more designated mutations in their genome can respond to immunomodulatory agents more effectively than other patients, such as wild-type patients.The present invention is also based in part on the discovery that cancer patients who have one or more designated mutations in their genome can respond to immunomodulatory agents less effectively than other patients, such as wild-type patients.These mutations can generally be referred to as single nucleotide polymorphisms or "SNPs", but the mutations disclosed herein are functional mutations that exist in germline.Mutations are generally to single nucleotides, such as nucleotide substitutions or nucleotide deletions.

[0049] The mutations referred to herein include functional mutations that disrupt the microRNA pathway, including microRNA binding site mutations. MicroRNAs (miRNAs) are small non-coding RNA molecules containing approximately 22 nucleotides found in plants, animals, and some viruses, which function in RNA silencing and post-transcriptional regulation of gene expression. These functions are essential for the role of miRNAs as critical stress response mediators, including mediating immune and inflammatory responses. DNA damage is also known to cause changes in the global profile of miRNA expression (Weidhaas et al., Cancer Res, 2007 67:11111), and stress-induced miRNA deregulation has been observed at the levels of transcription, processing, subcellular localization, and function. Therefore, biomarkers that predict disruptions in the microRNA pathway may be useful for predicting systemic immune responses to immunomodulatory therapies and the toxicity of such therapies, particularly because of the way these pathways affect immune and inflammatory responses. Furthermore, because immunotherapy for treating cancer relies on modulating the immune response in the patient, and irAE toxicity to immunotherapy results from the immune response, the markers disclosed herein as being relevant to predicting response to immunomodulatory agents are also believed to be relevant to predicting the likelihood that a patient will have a toxic response to immunomodulatory agents.

[0050] As described in the Examples herein, several mutations have been identified that are relevant for determining the likelihood that a cancer patient will respond to an immunomodulatory agent for the treatment of cancer. These mutations (also referred to herein as "markers," "biomarkers," or "variants") are shown in SEQ ID NOS: 1-10 in Figure 1 as the nucleotides in square brackets.

[0051] One biomarker relevant to determining a cancer patient's response to an immunomodulatory agent is found in the human CD44 gene; the marker is a SNP defined as rs11821102. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is an A nucleotide (variant) substituted for a G nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 1 (variant sequence) or SEQ ID NO: 17 (wild-type sequence). In one embodiment, patients identified as not carrying this polymorphism are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, and patients identified as carrying the polymorphism are identified as having a low likelihood of responding to treatment with an immunomodulatory agent. In another embodiment, patients who do not carry a CD44 mutation or who are homozygous for the CD44 mutation have a higher likelihood of responding to an immunomodulatory agent compared to patients who are heterozygous for the CD44 mutation. In one embodiment, patients who are heterozygous for the CD44 mutation are considered non-responders.

[0052] Another biomarker relevant to determining a cancer patient's response to an immunomodulatory agent is found in the human CD274 gene; the marker is a SNP defined as rs4742098. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 2 (variant sequence) or SEQ ID NO: 18 (wild-type sequence). In one embodiment, patients identified as not carrying this polymorphism are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, and patients identified as carrying the polymorphism are identified as having a low likelihood of responding to treatment with an immunomodulatory agent.

[0053] Another biomarker relevant to determining a cancer patient's response to immunomodulatory agents is found in the human EXO1 gene; the marker is a SNP defined as rs4150021. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a deletion (variant) of a T nucleotide at position 101 of the wild-type sequence, SEQ ID NO: 19. In one embodiment, patients identified as not carrying this mutation are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, and patients identified as carrying the mutation are identified as having a low likelihood of responding to treatment with an immunomodulatory agent. In another embodiment, patients who do not carry an EXO1 mutation or who are homozygous for the EXO1 mutation have a higher likelihood of responding compared to patients who are heterozygous for the EXO1 mutation. In another embodiment, patients who are heterozygous for the EXO1 mutation would be predicted to be intermediate responders, as opposed to responders or strong responders.

[0054] Another biomarker relevant to determining a cancer patient's response to an immunomodulatory agent is found in the human IL8 gene; the marker is a SNP defined as rs4073. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is an A nucleotide (variant) substituted for a T nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 4 (variant sequence) or SEQ ID NO: 20 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, and patients identified as not carrying the mutation are identified as having a low likelihood of responding to treatment with an immunomodulatory agent.

[0055] Another biomarker relevant to determining a cancer patient's response to an immunomodulatory agent is found in the human IL10 gene; the marker is a SNP defined as rs3024496. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 5 (variant sequence) or SEQ ID NO: 21 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, and patients identified as not carrying the mutation are identified as having a low likelihood of responding to treatment with an immunomodulatory agent.

[0056] Another biomarker relevant to determining a cancer patient's response to an immunomodulatory agent is found in the human IL10RB gene; the marker is a SNP defined as rs2834167. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 6 (variant sequence) or SEQ ID NO: 22 (wild-type sequence). In one embodiment, patients identified as not carrying this mutation are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, and patients identified as carrying the mutation are identified as having a low likelihood of responding to treatment with an immunomodulatory agent.

[0057] Another biomarker relevant to determining a cancer patient's response to an immunomodulatory agent is found in the human IL18R1 gene; the marker is a SNP defined as rs11465660. In Figure 1, 100 nucleotides upstream (5') of the mutation and 100 nucleotides downstream (3') of the mutation are shown. The mutation is an A nucleotide (variant) substituted for a C nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 7 (variant sequence) or SEQ ID NO: 23 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, while patients identified as not carrying the mutation are identified as having a low likelihood of responding to treatment with an immunomodulatory agent. In another embodiment, patients who are heterozygous for an IL18R1 mutation have a higher likelihood of responding to an immunomodulatory agent compared to patients who do not carry the mutation or who are homozygous for the mutation.

[0058] Another biomarker relevant to determining a cancer patient's response to immunomodulatory agents is found in the promoter region of the human miR99a gene. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a C nucleotide (variant) substituted for a T nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 8 (variant sequence) or SEQ ID NO: 24 (wild-type sequence). In one embodiment, patients identified as not carrying this mutation are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, and patients identified as carrying the mutation are identified as having a low likelihood of responding to treatment with an immunomodulatory agent. In another embodiment, patients who do not carry the mir99a mutation or are heterozygous for the mir99 mutation have a higher likelihood of responding compared to patients who are homozygous for the mir99 mutation. In one embodiment, patients who are homozygous for the mir99 mutation are considered relative non-responders compared to patients who do not carry the mir99a mutation or who are heterozygous for the mir99 mutation.

[0059] Another biomarker relevant to determining a cancer patient's response to an immunomodulatory agent is found in the human RAD23A gene; the marker is a SNP defined as rs8240. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is an A nucleotide (variant) substituted for a G nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 9 (variant sequence) or SEQ ID NO: 25 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, and patients identified as not carrying the mutation are identified as having a low likelihood of responding to treatment with an immunomodulatory agent.

[0060] Another biomarker relevant to determining a cancer patient's response to an immunomodulatory agent is found in the human STAT3 gene; the marker is a SNP defined as rs3744883. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a C nucleotide (variant) substituted for a T nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 10 (variant sequence) or SEQ ID NO: 26 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a high likelihood of responding to treatment with an immunomodulatory agent, and patients identified as not carrying the mutation are identified as having a low likelihood of responding to treatment with an immunomodulatory agent.

[0061] According to one embodiment of the present invention, the above-mentioned biomarkers are used to identify patients who will respond to immunomodulatory agents. Thus, in one embodiment, to determine a patient's predicted response to immunomodulatory therapy, the patient is screened for the following mutations: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and c) a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); f) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); g) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660); h) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); i) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); and j) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483) It is decided whether to have one or more of the following:

[0062] In a further embodiment, the patient is a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and c) a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); or f) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter)

[0033] It is determined whether a patient does not possess one or more mutations selected from the following. If the patient does not possess one or more of these mutations, the patient is predicted to have a high probability of responding to an immunomodulatory agent. In one embodiment, to predict a patient's response to an immunomodulatory agent, it is determined whether the patient possesses one, two, three, four, five, or all six of the above-mentioned biomarkers. For example, a determination as to whether a patient has a high probability of responding to an immunomodulatory agent can be based on whether the patient does not possess only one, two, three, four, five, or all six markers, but may not require evaluation of all six markers. For example, a determination as to whether a patient has a high probability of responding to an immunomodulatory agent can be based on whether the patient does not possess at least one, at least two, at least three, or at least four, or at least five of the six biomarkers described above.

[0063] In still a further embodiment, the patient is a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); c) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); or e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483)

[0013] It is determined whether a patient possesses one or more mutations selected from the following: If the patient possesses one or more of these mutations, the patient is predicted to have a high probability of responding to an immunomodulatory agent. In one embodiment, to predict the patient's response to an immunomodulatory agent, it is determined whether the patient possesses one, two, three, four, or all five of the biomarkers described above. For example, a determination as to whether a patient has a high probability of responding to an immunomodulatory agent can be based on whether the patient possesses only one, only two, only three, only four, or all five markers, but may not require evaluation of all five markers. In a further embodiment, a determination as to whether a patient has a high probability of responding to an immunomodulatory agent can be based on whether the patient possesses at least one, at least two, at least three, or at least four of the five biomarkers described above.

[0064] The predictive biomarkers can be used alone or in combination with one or more other biomarkers disclosed herein that predict response to immunomodulatory therapy to provide a method for predicting a patient's likely response to immunomodulatory therapy. In particular, knowing whether a patient is homozygous or heterozygous for a particular marker associated with response to immunotherapy (or does not carry the marker, i.e., is wild-type) can be useful in determining a patient's likely response to immunomodulatory therapy.

[0065] For example, with respect to CD44 / rs11821102, if a patient is not heterozygous for the A nucleotide at the position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102), the patient has a high likelihood of responding to immunomodulatory therapy.

[0066] With respect to IL18R1 / rs11465660, if a patient is heterozygous for the A nucleotide at the position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660), the patient has a high likelihood of responding to immunomodulatory therapy.

[0067] In particular, if a patient is not heterozygous for the A nucleotide at the position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102) and is not heterozygous for the A nucleotide at the position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660), the patient has a high likelihood of responding to immunomodulatory therapy.

[0068] With respect to the mutations disclosed in SEQ ID NO: 8 (mir99a promoter) in the human mir99a promoter, if the patient is not homozygous for a C nucleotide (mir99a promoter) at the position corresponding to position 101 of SEQ ID NO: 8, the patient has a high likelihood of exhibiting a response to immunomodulatory therapy. If the patient is not heterozygous for an A nucleotide (IL18R1 / rs11465660) at the position corresponding to position 101 of SEQ ID NO: 7, this genotype is particularly indicative of a response. Furthermore, if the patient is not heterozygous for an A nucleotide (CD44 / rs11821102) at the position corresponding to position 101 of SEQ ID NO: 1, this genotype is also particularly indicative of a response.

[0069] With respect to EXO1 / rs4150021, if a patient is heterozygous for the deletion of a T nucleotide (EXO1 / rs4150021) at a position corresponding to position 101 of SEQ ID NO: 19, this indicates a high likelihood of showing a response. However, if a patient is not heterozygous for the deletion of a T nucleotide (EXO1 / rs4150021) occurring in the wild-type sequence at a position corresponding to position 101 of SEQ ID NO: 19, the likelihood of showing a response will be higher than if the patient is heterozygous for the deletion. These markers are particularly predictive of a high likelihood of showing a response in patients who are not homozygous for a C nucleotide (miR99a promoter) at a position corresponding to position 101 of SEQ ID NO: 8. These markers are particularly predictive of a high likelihood of response in patients who are not heterozygous for the A nucleotide at the position corresponding to position 101 of SEQ ID NO:7 (IL18R1 / rs11465660) and are not heterozygous for the A nucleotide at the position corresponding to position 101 of SEQ ID NO:1 (CD44 / rs11821102).

[0070] In one embodiment of the present invention, a patient's response to an immunomodulatory agent is predicted by determining the patient's zygosity status with respect to an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102), an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660), a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter), and a deletion of a T nucleotide occurring in the wild-type sequence at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021). The zygosity status of one, two, three, or all four of these markers can be determined and assessed to predict a patient's response to an immunomodulatory agent.

[0071] For example, in one embodiment, a patient is evaluated to determine whether they are heterozygous or not heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102). For example, in another embodiment, a patient is evaluated to determine whether they are heterozygous or not heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660). For example, in another embodiment, a patient is evaluated to determine whether they are homozygous or not homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter). In yet another embodiment, a patient is evaluated to determine whether they are heterozygous or not heterozygous for a deletion of a T nucleotide occurring in the wild-type sequence at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021).

[0072] Biomarkers disclosed herein that predict patient response to immunomodulatory agents may also be useful in predicting patient response to treatment with combinations of immunomodulatory agents, e.g., two, three, or more immunomodulatory agents. Thus, these biomarkers may be useful in determining what combinations of immunomodulatory agents may be effective in treating a patient's cancer.

[0073] In certain embodiments, if a patient is progression-free at 6 months (i.e., progression-free 6 months after initiating treatment with one or more immunomodulatory agents), the biomarkers disclosed herein to be predictive of patient response to immunomodulatory agents may be useful in predicting patient response to treatment.

[0074] As described in the Examples herein, several mutations have been identified that are relevant for determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent used to treat cancer. These mutations (also referred to herein as "markers," "biomarkers," or "variants") are shown in Figure 1 as SEQ ID NOS: 2, 6, 8, 11-16, and 33, as nucleotides in square brackets.

[0075] One biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is found in the human EREG gene; the marker is a SNP defined as rs1460008. In Figure 1, 100 nucleotides upstream (5') of the mutation and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 11 (variant sequence) or SEQ ID NO: 27 (wild-type sequence). In one embodiment, patients identified as not carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, and patients identified as carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent.

[0076] Another biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is found in the human FCGR2A gene; the marker is a SNP defined as rs10919033. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a C nucleotide (variant) substituted for a T nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 12 (variant sequence) or SEQ ID NO: 28 (wild-type sequence). In one embodiment, patients identified as not carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, while patients identified as carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent.

[0077] Another biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is also found in the human FCGR2A gene; the marker is a SNP defined as rs1801274. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a C nucleotide (variant) substituted for a T nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 13 (variant sequence) or SEQ ID NO: 29 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, and patients identified as not carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent.

[0078] Another biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is found in the human IL10RB gene; the marker is a SNP defined as rs2834167. In Figure 1, 100 nucleotides upstream (5') of the mutation and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 6 (variant sequence) or SEQ ID NO: 22 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, and patients identified as not carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent. In a further embodiment, patients identified as carrying this mutation are identified as having a high likelihood of exhibiting a toxic response to radiation therapy. In yet a further embodiment, patients homozygous for this mutation are identified as having a high likelihood of exhibiting a toxic response to radiation therapy.

[0079] Another biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is found in the human KRAS gene; the marker is a SNP defined as rs61764370. In FIG. 1 , 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for a T nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 14 (variant sequence) or SEQ ID NO: 30 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, while patients identified as not carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent.

[0080] Another biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is found in the human miR99a promoter gene. In Figure 1, 100 nucleotides upstream (5') of the mutation and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a C nucleotide (variant) substituted for a T nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 8 (variant sequence) or SEQ ID NO: 24 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, while patients identified as not carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent.

[0081] Another biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is found in the human RAC1 gene; the marker is a SNP defined as rs9374. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is an A nucleotide (variant) substituted for a G nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 15 (variant sequence) or SEQ ID NO: 31 (wild-type sequence). In one embodiment, patients identified as not carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, while patients identified as carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent.

[0082] Another biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is found in the human TLR4 gene; the marker is a SNP defined as rs4986790. In Figure 1, 100 nucleotides upstream (5') of the mutation and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 16 (variant sequence) or SEQ ID NO: 32 (wild-type sequence). In one embodiment, patients identified as not carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, while patients identified as carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent. In a further embodiment, patients carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to radiation therapy. In yet a further embodiment, patients identified as homozygous or heterozygous for this mutation are identified as having a low likelihood of exhibiting a toxic response to radiation therapy.

[0083] Another biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is found in the human MSH2 gene; the marker is a SNP defined as rs2303428. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a C nucleotide (variant) substituted for a T nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 33 (variant sequence) or SEQ ID NO: 34 (wild-type sequence). In one embodiment, patients identified as carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, while patients identified as not carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent.

[0084] Another biomarker relevant to determining the likelihood that a cancer patient will have a toxic response to an immunomodulatory agent is found in the human CD274 gene; the marker is a SNP defined as rs4742098. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 2 (variant sequence) or SEQ ID NO: 18 (wild-type sequence). In one embodiment, patients identified as not carrying this mutation are identified as having a low likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent, and patients identified as carrying the mutation are identified as having a high likelihood of exhibiting a toxic response to treatment with an immunomodulatory agent. In a further embodiment, patients carrying this mutation are identified as having a high likelihood of exhibiting a toxic response to radiation therapy. In yet a further embodiment, patients homozygous for this mutation are identified as having a high likelihood of exhibiting a toxic response to radiation therapy.

[0085] Each of the biomarkers disclosed herein to predict toxicity can be used alone or in combination with one or more of the other markers disclosed herein to predict toxicity to predict whether a patient will have a toxic response to immunomodulatory therapy. For example, one, two, three, four, five, six, seven, eight, nine, or ten of the biomarkers can be used in determining a patient's predicted toxicity to immunomodulatory therapy. In particular, determining whether a patient is homozygous or heterozygous (or does not possess the marker, i.e., the patient is wild-type) for a particular marker associated with toxicity to immunotherapy can be useful in determining a patient's predicted toxicity to immunomodulatory therapy. For example, to determine a patient's predicted toxicity to immunomodulatory therapy, the patient's zygosity can be determined for one, two, three, four, five, six, seven, eight, nine, or ten of the biomarkers disclosed herein to predict toxicity. For example, the zygosity of at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, or at least nine of the above-mentioned markers associated with toxicity can be determined.

[0086] For example, with respect to FCGR2A / rs10919033, if a patient is neither heterozygous nor homozygous for the C nucleotide at the position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033) (e.g., the patient is homozygous wild-type), the patient has a low likelihood of exhibiting a toxic response to immunomodulatory therapy.

[0087] With respect to TLR4 / rs4986790, if a patient is neither heterozygous nor homozygous for the G nucleotide (TLR4 / rs4986790) at the position corresponding to position 101 of SEQ ID NO: 16 (e.g., the patient is homozygous wild-type), the patient has a low likelihood of exhibiting a toxic response to immunomodulatory therapy. In a further embodiment, if a patient is also neither heterozygous nor homozygous for the C nucleotide (FCGR2A / rs10919033) at the position corresponding to position 101 of SEQ ID NO: 12, the patient has a low likelihood of exhibiting a toxic response to immunomodulatory therapy.

[0088] With respect to KRAS / rs61764370, if a patient is heterozygous or homozygous for a G nucleotide (KRAS / rs61764370) at the position corresponding to position 101 of SEQ ID NO: 14 (e.g., the patient is homozygous wild-type), the patient has a low likelihood of exhibiting a toxic response to immunomodulatory therapy. In a further embodiment, if a patient is also neither heterozygous nor homozygous for a C nucleotide (FCGR2A / rs10919033) at the position corresponding to position 101 of SEQ ID NO: 12, and neither heterozygous nor homozygous for a G nucleotide (TLR4 / rs4986790) at the position corresponding to position 101 of SEQ ID NO: 16, the patient has a low likelihood of exhibiting a toxic response to immunomodulatory therapy.

[0089] In one embodiment of the present invention, a patient's toxic response to an immunomodulatory agent is predicted by determining the patient's zygosity status with respect to a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033), a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790), and / or a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370). The zygosity status of one, two, or all three of these markers can be determined and assessed to predict a patient's likely toxic response to an immunomodulatory agent. Alternatively, only one or two of these markers can be assessed.

[0090] For example, in one embodiment, a patient is evaluated to determine whether the patient is neither heterozygous nor homozygous for a C nucleotide (FCGR2A / rs10919033) at a position corresponding to position 101 of SEQ ID NO: 12. For example, in another embodiment, a patient is evaluated to determine whether the patient is neither heterozygous nor homozygous for a G nucleotide (TLR4 / rs4986790) at a position corresponding to position 101 of SEQ ID NO: 16. For example, in another embodiment, a patient is evaluated to determine whether the patient is heterozygous or homozygous for a G nucleotide (KRAS / rs61764370) at a position corresponding to position 101 of SEQ ID NO: 14.

[0091] In one embodiment of the present invention, when determining whether a patient is likely to have a toxic or non-toxic response to an immunomodulatory agent, the patient is evaluated for the following mutations: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370); f) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); g) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374); h) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); i) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and j) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) It is decided whether to have one or more of the following:

[0092] Assessment of the likelihood that a patient will have a toxic response to an immunomodulatory agent can be based on determining the presence or absence of only one, two, three, four, five, six, seven, eight, nine, or all ten of these markers, but does not necessarily require assessment of all ten markers. For example, assessment can be based on determining the presence or absence of at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, or at least nine of the above-mentioned markers associated with toxicity.

[0093] When the immunomodulatory agent is an anti-PD1 or anti-PDL1 antibody therapy, a particular subset of biomarkers has been determined to be relevant in determining whether a patient will have a toxic response to the antibody therapy. In particular, whether a patient has or has not had any of the following mutations: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374); d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); or e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098) If a patient is identified as not carrying one or more of the following, the patient will be considered at low risk of a toxic response to antibody therapy.

[0094] If a patient does not have one or more of these mutations, the patient is predicted to have a low probability of having a toxic response to the immunomodulator. In one embodiment, it is determined whether the patient has one, two, three, four, or all five of the above-mentioned biomarkers. For example, the determination of whether a patient has a low probability of having a toxic response to the immunomodulator can be based on whether the patient does not have only one, two, three, four, or all five biomarkers, but may not require the evaluation of all five markers. For example, the evaluation can be based on the determination of the presence or absence of at least one, at least two, at least three, or at least four of the above-mentioned markers associated with toxicity.

[0095] When the immunomodulatory agent is an anti-PD1 or anti-PDL1 antibody therapy, a particular subset of biomarkers has been determined to be relevant in determining whether a patient will have a toxic response to the antibody therapy. In particular, whether a patient has or has not had any of the following mutations: a) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370); d) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); or e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) If identified as carrying one or more of the following, the patient will be considered at low risk of a toxic response to antibody therapy.

[0096] If a patient has one or more of these mutations, the patient is predicted to have a low probability of having a toxic response to an immunomodulator. In one embodiment, it is determined whether the patient has one, two, three, four, or all five of the above-mentioned biomarkers. For example, the determination of whether a patient has a low probability of having a toxic response to an immunomodulator can be based on whether the patient has only one, two, three, four, or all five biomarkers, but may not require the evaluation of all four markers. For example, the evaluation can be based on the determination of the presence or absence of at least one, at least two, at least three, or at least four of the above-mentioned markers associated with toxicity.

[0097] When the immunomodulatory agent is radiation therapy, a particular subset of biomarkers has been determined to be relevant in determining whether a patient will have a toxic response to radiation therapy.

[0098] One biomarker relevant to determining whether a cancer patient will have a toxic response to radiation therapy is found in the human CD274 gene; the marker is a SNP defined as rs4742098. In Figure 1, 100 nucleotides upstream (5') and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 2 (variant sequence) or SEQ ID NO: 18 (wild-type sequence). In one embodiment, patients who do not carry this mutation or are identified as heterozygous for this mutation are identified as having a high likelihood of exhibiting a non-toxic response to radiation therapy, and patients who are identified as homozygous for this mutation are identified as being at high risk for a toxic response to radiation therapy.

[0099] Another biomarker relevant to determining whether a cancer patient will have a toxic response to radiation therapy is found in the human IL10RB gene; the marker is a SNP defined as rs2834167. In Figure 1, 100 nucleotides upstream (5') of the mutation and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 6 (variant sequence) or SEQ ID NO: 22 (wild-type sequence). In one embodiment, patients identified as not carrying this mutation or as being heterozygous for this mutation are identified as having a high likelihood of exhibiting a non-toxic response to radiation therapy, while patients identified as being homozygous for this mutation are identified as being at high risk for a toxic response to radiation therapy.

[0100] Another biomarker relevant to determining whether a cancer patient will have a toxic response to radiation therapy is found in the human TLR4 gene; the marker is a SNP defined as rs4986790. In Figure 1, 100 nucleotides upstream (5') of the mutation and 100 nucleotides downstream (3') of the mutation are shown. The mutation is a G nucleotide (variant) substituted for an A nucleotide (wild-type). The mutation occurs at position 101 of SEQ ID NO: 16 (variant sequence) or SEQ ID NO: 32 (wild-type sequence). In one embodiment, patients identified as carrying this mutation (whether heterozygous or homozygous) are identified as having a high likelihood of exhibiting a non-toxic response to radiation therapy, and patients identified as not carrying the mutation are identified as having a high likelihood of having a toxic response to treatment with an immunomodulatory agent.

[0101] In one embodiment, to determine whether a patient has a toxic response to radiation therapy, the patient is monitored for the following markers: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and / or c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167) It is decided whether to hold one or more of the following:

[0102] In one embodiment, a patient has a low risk of a toxic response to radiation therapy if they possess a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790). In another embodiment, a patient has an increased risk of a toxic response to radiation therapy if they possess a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098). In yet another embodiment, a patient has an increased risk of a toxic response to radiation therapy if they possess a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167).

[0103] In further embodiments, the zygosity status of a patient for certain biomarkers is determined to determine the patient's risk of a toxic response to radiation therapy. For example, in one embodiment, if the patient is heterozygous or homozygous for a G nucleotide (TLR4 / rs4986790) at a position corresponding to position 101 of SEQ ID NO: 16, the patient has a low risk of a toxic response to radiation therapy. In another embodiment, if the patient is not homozygous for a G nucleotide (CD274 / rs4742098) at a position corresponding to position 101 of SEQ ID NO: 2, the patient has a low risk of a toxic response to radiation therapy. In yet another embodiment, if the patient is not homozygous for a G nucleotide (IL10RB / rs2834167) at a position corresponding to position 101 of SEQ ID NO: 6, the patient has a low risk of a toxic response to radiation therapy. definition

[0104] As used herein, the term "response" or "responding" in the context of a patient's response to a therapy or treatment for cancer refers to RECIST (Response Evaluation Criteria in Solid Tumors) criteria for assessing the response of a target lesion to a cancer therapy. According to the RECIST criteria, a response Patients who undergo treatment are categorized as either "complete responders" (disappearance of all target lesions) or "partial responders" (at least a 30% reduction in the sum of the longest diameters of the target lesions, referenced to the baseline sum longest diameter); non-responders are placed into one of two categories: stable disease (neither a sufficient reduction in the smallest sum longest diameter since treatment began to qualify as a partial response nor a sufficient increase in the smallest sum longest diameter since treatment began to qualify as progressive disease) or progressive disease (at least a 20% increase in the sum of the longest diameters of the target lesions, referenced to the smallest sum longest diameter recorded since treatment began, or the appearance of one or more new lesions). RECIST criteria are discussed in detail, for example, in Eisenhauer et al., Eur. J. Cancer, 2009: 25:228-247. Thus, as described herein, a response to a therapy refers to a patient who falls within the RECIST categories of complete or partial responder, while a non-response refers to a patient who falls within the RECIST categories of stable disease or progressive disease.

[0105] As used herein, the terms "treat," "treating," or "treatment," in the context of cancer, refer to (a) slowing of tumor growth, (b) cessation of tumor growth, (c) regression, or (d) improvement in one or more patient symptoms. According to one embodiment of the present invention, "treat" or "treating" can refer to a patient outcome in which a patient receiving immunomodulatory therapy exhibits a response to the therapy.

[0106] As used herein, the term "toxicity" or "toxic response" refers to the occurrence of one or more immune response adverse reactions (irAEs), a specific class of adverse reactions that patients may experience in response to cancer therapy, most commonly cancer immunotherapy. irAEs are thought to occur as a result of stimulation of the immune system by cancer therapy and include different forms of autoimmunity induced by the administration of these therapies, such as pneumonitis, hepatitis, pancreatitis, and colitis. For example, irAEs are particularly observed in patients treated with checkpoint inhibitor therapy. irAEs are described, for example, in Abdel-Wahab et al., PLOS ONE, 11(7):e0160221 (2016) This is discussed more fully in [1]. irAE toxicity is generally graded on a scale of 0 to 5, with 0 representing no adverse event or an event within the normal range, 1 representing a mild adverse event, 2 representing a moderate adverse event, 3 representing a severe and undesirable adverse event, 4 representing a life-threatening or disabling adverse event, and 5 representing death related to the adverse event. Thus, a "toxic response" or "high toxicity" as referred to herein refers to an irAE of grade 2 or higher on this scale, while a grade of 0 or 1 is considered a "non-toxic response" or "low toxicity."

[0107] The term "high probability / likelihood" or "low probability / likelihood" in the context of the present invention can refer to the "absolute" probability or "relative" probability of a subject, relating to the high or low probability that a certain event will occur over a certain period of time. Absolute probability can be measured by referring to the actual observation after measurement over a relevant time period, or by referring to the index value created from a statistically valid historical cohort that has been followed over a relevant period of time. Relative probability refers to the ratio of a subject's absolute probability compared to either the absolute probability of a low-probability cohort or the average population probability, and this can vary depending on how clinical probability is assessed. Odds ratio, which is the ratio of positive events to negative events for a given test result, is also commonly used without transformation (odds follow the formula p / (1-p), where p is the probability of an event, and (1-p) is the probability of no event).

[0108] A "high likelihood or probability" of a patient responding to a therapy can be based on a comparison with the response rate (or predicted response rate) of a cohort of patients having a different genotype from the patient's genotype for a particular biomarker or biomarkers. A "high likelihood or probability" of a patient responding to a therapy can also be based on a comparison with the response rate (or predicted response rate) of a cohort of patients without taking into account the marker (or markers).

[0109] A patient's "low likelihood or probability" of responding to a therapy can be based on a comparison with the response rate (or expected response rate) of a cohort of patients having a different genotype from the patient's genotype for a particular biomarker or biomarkers. A "low likelihood or probability" of responding to a therapy can also be based on a comparison with the response rate (or expected response rate) of a cohort of patients without taking into account the marker (or markers).

[0110] A "high likelihood or probability" of a patient having a toxic response to a therapy can be based on a comparison with the toxic response rate (or predicted toxic response rate) of a cohort of patients having a different genotype from the patient's genotype for a particular biomarker or biomarkers. A "high likelihood or probability" of a patient having a toxic response to a therapy can also be based on a comparison with the toxic response rate (or predicted response rate) of a cohort of patients without taking into account the marker (or markers).

[0111] A "low likelihood or probability" of a patient having a toxic response to a therapy can be based on a comparison with the toxic response rate (or predicted toxic response rate) of a cohort of patients having a different genotype from the patient's genotype for a particular biomarker or biomarkers. A "low likelihood or probability" of a patient having a toxic response to a therapy can also be based on a comparison with the toxic response rate (or predicted response rate) of a cohort of patients without taking into account the marker (or markers).

[0112] Biomarkers disclosed herein that predict patient toxicity to immunomodulatory agents may also be useful in predicting patient toxicity to combinations of immunomodulatory agents, e.g., treatment with two, three, or more immunomodulatory agents. Thus, these biomarkers may be useful in determining what combinations of immunomodulatory agents can be used without obtaining a toxic response in the patient. immunomodulators

[0113] The mutations described herein are relevant biomarkers for determining a cancer patient's therapeutic response to treatment with an immunomodulatory agent, or for determining whether a patient will have a toxic response to treatment with an immunomodulatory agent. In one embodiment, the immunomodulatory agent is designed to enhance a patient's immune system. Immunomodulatory agents that function to initially stimulate a weakened immune system are preferred over agents that rely on a fully functional immune system for their benefit. Immunomodulatory agents can include antibodies, cytokines, adoptive cell transfer, anti-cancer vaccines, checkpoint inhibitors, or non-biologic drugs that stimulate the immune system. The terms "immunomodulatory agent" and "immunomodulatory therapy" are used interchangeably herein.

[0114] Immunomodulators include cetuximab, panitumumab, nimotuzumumab, and matsutake. The immunomodulator may be an anti-cancer antibody such as trastuzumab, futuximab, imgatuzumab, necitumumab, alemtuzumab, trastuzumab, ibritumomab, brentuximab, blinatumomab, bevacizumab, cetuximab, or ipilimumab. The immunomodulator may be an anti-cancer antibody conjugated to a drug. For example, the immunomodulator may be ibritumomab tiuxetan, brentuximab vedotin, ado-trastuzumab emtansine, or denileukin diftitox. The immunomodulator may be a cytokine such as IL-2, IL-12, IL-21, IFN-α, IFN-β, or IFN-γ. The cytokine may be conjugated to an antibody. The immunomodulatory agent may be thalidomide, lenalidomide, pomalidomide, or ibuprofen. It can also be a drug that stimulates the immune system, such as miquimod.

[0115] The immunomodulatory agent can be radiation. The radiation can be external beam radiation therapy, such as photon beams of x-rays or gamma rays, electron beams, or proton therapy. The radiation can be internal radiation therapy (brachytherapy), in which case the radiation is delivered from a radioactive source placed inside or on the body, even inside tumor tissue. The radioactive source can be a radioisotope in the form of a seed or pellet implanted in or placed on the patient. The radiation can be systemic radiotherapy administered orally or by injection. The radioactive material can be radioactive iodine ( 131 I), or an antibody conjugated to a radioactive substance, such as ibritumomab tiuxetan or I131-tositumomab.

[0116] In another embodiment, the immunomodulatory agent can be a checkpoint inhibitor. For example, in one embodiment, the immunomodulatory agent is an anti-PD1 or anti-PDL1 antibody. According to one aspect of the present invention, the patient treated according to the method of the present invention is treated with an anti-PDL1 or anti-PD1 therapy. In one embodiment, the therapy is an anti-PDL1 antibody. In another embodiment, the therapy is an anti-PD1 antibody. In one embodiment, the anti-PD1 antibody is Opdivo® (nivolumab) or Keytruda® (pembrolizumab). In another embodiment, the anti-PDL1 antibody is BMS-936559 (MDX-1105), Tecentriq® (atezolizumab), Imfinzi® (durvalumab), or Bavencio® (avelumab). In other embodiments, the anti-PD1 antibody is an antibody capable of binding to PD1. In yet other embodiments, the anti-PDL1 antibody is an antibody capable of binding to PDL1. In another embodiment, the immunomodulatory agent can be an anti-CTLA4 antibody. For example, in one embodiment, the immunomodulatory agent is ipilimumab.

[0117] In another embodiment, the immunomodulatory agent can be a cell-based therapy delivered by adoptive cell transfer. For example, the immunomodulatory agent can be a chimeric antigen receptor T cell (CAR-T cell).

[0118] In yet another embodiment, the immunomodulatory agent can be an anti-cancer vaccine. For example, in one embodiment, the immunomodulatory agent can be a dendritic cell vaccine.

[0119] As used herein, the term "antibody" refers to an intact antibody (e.g., an intact monoclonal antibody). In some embodiments, "antibody" includes antigen-binding fragments of antibodies. Antigen-binding fragments include Fab, Fab', F(ab')2, Fv, single-chain antibodies (e.g., scFv), minibodies, diabodies, and monoclonal antibodies. Single domain antibodies ("sdAbs" or "nanobodies" or "camelids") "). In yet other embodiments, the antibody comprises an optimized, engineered, or chemically conjugated intact antibody or antigen-binding fragment of an antibody (e.g., a phage-displayed antibody, including a fully human antibody, a semi-synthetic antibody, or a fully synthetic antibody). An example of an optimized antibody is an affinity-matured antibody. Examples of engineered antibodies are Fc-optimized antibodies and multispecific antibodies (e.g., bispecific antibodies). An antibody conjugated to a toxin moiety is an example of a chemically conjugated antibody.

[0120] Methods for producing antibodies such as anti-PDL1 and anti-PD1 antibodies or other anti-cancer antibodies are known in the art. For example, DNA molecules encoding the light chain variable region and heavy chain variable region can be chemically synthesized. The synthetic DNA molecules can be ligated to other appropriate nucleotide sequences, including, for example, constant region coding sequences and expression control sequences, to produce a conventional gene expression construct encoding the desired antibody. The production of a defined gene construct is within the routine skill of those in the art. Alternatively, the sequences provided herein can be cloned from hybridomas by conventional hybridization or polymerase chain reaction (PCR) techniques, taking into account the genes encoding the heavy and light chains of mouse antibodies in hybridoma cells and using synthetic nucleic acid probes whose sequences are based on known sequences.

[0121] Nucleic acids encoding the antibodies disclosed herein can be incorporated (ligated) into expression vectors, which can be introduced into host cells by conventional transfection or transformation techniques. Exemplary host cells are E. coli cells, Chinese hamster ovary (CHO) cells, HeLa cells, baby hamster kidney (BHK) cells, monkey kidney (COS) cells, human hepatocellular carcinoma cells (e.g., Hep G2), and myeloma cells that do not otherwise produce IgG protein. Transformed host cells can be grown under conditions that allow the host cells to express genes encoding immunoglobulin light and / or heavy chain variable regions.

[0122] Specific expression and purification conditions will vary depending on the expression system used.For example, if a gene is to be expressed in E. coli, it is first cloned into an expression vector by placing a suitable bacterial promoter, such as Trp or Tac, and the engineered gene downstream from a prokaryotic signal sequence.The expressed secretory protein accumulates in refractile bodies or inclusion bodies, and can be collected after cell disruption by French press or sonication.The refractile bodies are then solubilized, and the protein is refolded and cleaved by methods known in the art.

[0123] When a DNA construct encoding an antibody disclosed herein is to be expressed in a eukaryotic host cell, such as a CHO cell, it is first inserted into an expression vector containing a suitable eukaryotic promoter, secretion signal, IgG enhancer, and various introns. This expression vector optionally contains sequences encoding all or part of the constant region, allowing the entire heavy and / or light chain to be expressed. In some embodiments, a single expression vector contains both the heavy and light chain variable regions to be expressed.

[0124] Gene constructs can be introduced into eukaryotic host cells using conventional techniques. Host cells express variable light (VL) or variable heavy (VH) chain fragments, VL-VH heterodimers, VH-VL or VL-VH single-chain polypeptides, complete immunoglobulin heavy or light chains, or portions thereof, each of which can be attached to a moiety having another function (e.g., cytotoxicity). In some embodiments, host cells are transfected with a single vector expressing polypeptides expressing all or part of a heavy chain (e.g., a heavy chain variable region) or a light chain (e.g., a light chain variable region). In other embodiments, host cells are transfected with a single vector encoding (a) a polypeptide comprising a heavy chain variable region and a polypeptide comprising a light chain variable region, or (b) an entire immunoglobulin heavy chain and an entire immunoglobulin light chain. In yet other embodiments, the host cell is co-transfected with more than one expression vector (e.g., one expression vector expressing a polypeptide comprising all or part of a heavy chain or heavy chain variable region and another expression vector expressing a polypeptide comprising all or part of a light chain or light chain variable region).

[0125] A method for producing a polypeptide comprising an immunoglobulin heavy chain variable region or a polypeptide comprising an immunoglobulin light chain variable region can include growing a host cell transfected with an expression vector under conditions that allow expression of the polypeptide comprising an immunoglobulin heavy chain variable region or the polypeptide comprising an immunoglobulin light chain variable region. The polypeptide comprising the heavy chain variable region or the polypeptide comprising the light chain variable region can then be purified using techniques well known in the art, for example, affinity tags such as glutathione-S-transferase (GST) and histidine tags.

[0126] Human monoclonal antibodies can be isolated or selected from phage display libraries, including immune, naive, and synthetic libraries. Antibody phage display libraries are known in the art and are described, for example, in Hoet et al., NATURE BIOTECH. 23:344-348, 2005; Soderlind et al., NATURE BIOTECH. 18:852-856, 2000. ;Rothe et al., J. MOL. BIOL. 376:1182-1200, 2008;Knappik et al., J. See MOL. BIOL. 296:57-86, 2000; and Krebs et al., J. IMMUNOL. METH. 254:67-84, 2001. When used as therapeutic agents, phage display The isolated human antibodies can then be optimized (e.g., affinity matured) to improve biochemical characteristics, including affinity and / or specificity, improve biophysical properties, including aggregation, stability, precipitation, and / or nonspecific interactions, and / or reduce immunogenicity. Affinity maturation procedures are within the routine skill of those in the art. For example, diversity can be introduced into the immunoglobulin heavy chain and / or immunoglobulin light chain by DNA shuffling, chain shuffling, CDR shuffling, random mutagenesis, and / or site-directed mutagenesis.

[0127] In some embodiments, the isolated human antibody contains one or more somatic mutations. In these cases, the antibody can be modified to human germline sequences to optimize the antibody (i.e., a process called germlining).

[0128] Generally, an optimized antibody has at least the same or substantially the same affinity for the antigen as the non-optimized (or parent) antibody from which it is derived. Preferably, the optimized antibody has a higher affinity for the antigen when compared to the parent antibody.

[0129] Human antibody fragments (e.g., parental and optimized variants) can be engineered to contain certain constant (i.e., Fc) regions with designated effector functions (e.g., antibody-dependent cellular cytotoxicity (ADCC)). Human constant regions are known in the art.

[0130] Antibodies can be conjugated to effector moieties such as small molecule toxins or radionuclides using standard in vitro conjugation chemistry. If the effector moiety is a polypeptide, the antibody can be chemically conjugated to the effector or linked to the effector as a fusion protein. Construction of fusion proteins is within the ordinary skill in the art.

[0131] Methods for reducing or eliminating the antigenicity of antibodies and antibody fragments are known in the art. When an antibody is to be administered to a human, the antibody is preferably "humanized" to reduce or eliminate its antigenicity in humans. Preferably, a humanized antibody has the same or substantially the same affinity for an antigen as the non-humanized mouse antibody from which it is derived.

[0132] In one type of humanization approach, chimeric proteins are produced in which mouse immunoglobulin constant regions are replaced with human immunoglobulin constant regions. See, e.g., Morrison et al., 1984, PROC. NAT. ACAD. SCI. 81:6851-6855; Neuberger et al., 1984, NATURE 312:604-608; U.S. Patent No. 6,893,625 (Robinson); U.S. Patent No. 5,50 See US Pat. No. 0,362 (Robinson); and US Pat. No. 4,816,567 (Cabilly). stomach.

[0133] In an approach known as CDR grafting, the CDRs of the light and heavy chain variable regions are grafted onto frameworks derived from another species.For example, mouse CDRs can be grafted onto human FRs.In some embodiments, the CDRs of the light and heavy chain variable regions of anti-ErbB3 antibodies are grafted onto human FRs or consensus human FRs.To create consensus human FRs, FRs from several human heavy or light chain amino acid sequences are aligned to identify a consensus amino acid sequence.CDR grafting is described in U.S. Patent No. 7,022,500 (Queen); U.S. Patent No. 6,982,321 (Winter ... No. 6,180,370 (Queen); No. 6,054,297 (Carter); No. 5,69 No. 3,762 (Queen); No. 5,859,205 (Adair); No. 5,693,761 (Queen); No. 5,565,332 (Hoogenboom); No. 5,585,089 (Queen); No. 5,530,101 (Queen); Jones et al. (1986) NATURE 321: 522-525;Riechmann et al. (1988) NATURE 332: 323-327;Verhoeyen et al. (1988) SCIENCE 239: 1534-1536; and Winter (1998) FEBS LETT 430: 92-94.

[0134] In an approach termed "SUPERHUMANIZATION™," human CDR sequences are selected from human germline genes based on the structural similarity of the human CDRs to the CDRs of the mouse antibody to be humanized. See, e.g., U.S. Patent No. 6,881,557 (Foote); and Tan et al., 2002, J. IMMUNOL 169:1119-1125. I want to be done that.

[0135] Other methods to reduce immunogenicity include "reshaping," "hyperchimerization," and "Hyperchimerization" and "veneering / resurfacing" are included. See, e.g., Vaswami et al., 1998, ANNALS OF ALLERGY, ASTHMA, & IMMUNOL. 81:105; Roguska et al., 1996, PROT. ENGINEER 9:895-904; and U.S. Patent No. 6,072,035 (Hardman). Veneering / Resurfacing In the resurfacing approach, surface-accessible amino acid residues in a murine antibody are replaced with amino acid residues that are more frequently found at the same positions in human antibodies. This type of antibody resurfacing is described, for example, in U.S. Patent No. 5,639,641 (Pedersen).

[0136] Another approach to convert mouse antibodies into a form suitable for human medical use is known as ACTIVMAB™ technology (Vaccinex, Inc., Rochester, NY), which involves the use of vaccinia virus-based vectors to express antibodies in mammalian cells. It is said that high levels of combinatorial diversity of IgG heavy and light chains are produced. See, for example, U.S. Patent No. 6,706,477 (Zauderer); U.S. Patent No. 6,800,442 (Zauderer); and U.S. Patent No. 6,872,518 (Zauderer).

[0137] Another approach to converting murine antibodies into a form suitable for use in humans is a technology commercially practiced by KaloBios Pharmaceuticals, Inc. (Palo Alto, CA). This technology involves the use of a proprietary human "acceptor" library to generate an "epitope-focused" library for antibody selection.

[0138] Another approach to modifying mouse antibodies into a form suitable for human medical use is the HUMAN ENGINEERING™ technology, which is commercially implemented by XOMA (US) LLC. See, e.g., PCT Publication No. WO93 / 11794 and U.S. Patent Nos. 5,766,886; 5,770,196; 5,821,123; and 5,869,619.

[0139] Any suitable approach, including any of the approaches described above, can be used to reduce or eliminate the human immunogenicity of the antibodies disclosed herein.

[0140] Methods for producing multispecific antibodies are known in the art. Multispecific antibodies include bispecific antibodies. Bispecific antibodies are antibodies that have binding specificities for at least two different epitopes. Exemplary bispecific antibodies bind to two different epitopes of a target antigen. Bispecific antibodies are described, for example, in Milstein et al., NATURE 305:537-539 (1983) , WO93 / 08829, Traunecker et al., EMBO J., 10:3655-3659 (1991), WO94 / 04690, Suresh et al., METHODS IN ENZYMOLOGY, 121:210 (1986), WO96 / 27011, Brennan et al., SCIENCE, 229: 81 (1985), Shalaby et al., J. EXP. MED., 175: 217-225 (1992), Kostelny et al., J. IMMUNOL., 148(5):1547-1553 (1992), Hollinger et al., PNAS, 90:6444-6448, Gruber et al., J. IMMUNOL., 152:5368 (1994), Wu et al., NAT. BIOTECHNOL., 25(11): 1290-1297, U.S. Patent Application Publication No. 2007 / 0071675, and Bostrom They can be prepared as full length antibodies or antibody fragments (e.g., F(ab')2 bispecific antibodies and diabodies) as described in [End Page 110] et al., SCIENCE 323:1640-1644 (2009). SNP genotyping methods

[0141] The process of determining which specific nucleotide (i.e., allele) is present at each of one or more SNP positions is referred to as SNP genotyping. The present invention provides SNP genotyping methods to determine whether a patient has a particular genotype with respect to the variants disclosed herein as useful biomarkers in predicting a patient's therapeutic and toxic response to immunomodulatory agents.

[0142] Nucleic acid samples can be genotyped to determine which allele(s) are present in any given genetic region of interest (e.g., SNP position) by methods well known in the art. Nearby sequences can be used to design SNP detection reagents, such as oligonucleotide probes, which can be implemented in kit format if necessary. Exemplary SNP genotyping methods are described in Chen et al. (2003) PHARMACOGENOMICS J. 3(2):77-96; Kwok et al. (2003) CURR ISSUES MOL. BIOL. 5(2):43-60; Shi (2002) AM J PHARMACOGENOMICS 2(3):197-205; and Kwok (2001) ANNU REV GENOMICS HUM GENET 2:235-58. Exemplary techniques for high-throughput SNP genotyping are described in Marnellos (2003) CURR OPIN DRUG DISCOV DEVEL. 6(3):317-21. Common SNP genotyping methods include, but are not limited to, quantitative PCR, TaqMan assay, molecular beacon assay, nucleic acid array, allele-specific primer extension, allele-specific PCR, arrayed primer extension, homogeneous primer extension assay, primer extension with mass spectrometry detection, pyrosequencing, multiplex primer extension screened on a genetic array, ligation by rolling circle amplification, homogeneous ligation, OLA (U.S. Pat. No. 4,988,167), multiplex ligation reaction screened on a genetic array, restriction fragment length polymorphism, single base extension-tag assay, and Invader assay. Such methods can be used in combination with detection mechanisms such as, for example, luminescence or chemiluminescence detection, fluorescence detection, time-resolved fluorescence detection, fluorescence resonance energy transfer, fluorescence polarization, mass spectrometry, and electrical detection, which methods are well known in the art.

[0143] The biological sample for determining the presence or absence of any of the mutations disclosed herein can be any tissue or bodily fluid from a patient that contains nucleic acid. Various embodiments include paraffin-embedded tissue, frozen tissue, surgical fine needle aspirate, and cells from various tissues of a subject, such as blood cells or buccal swabs.

[0144] In one embodiment, the determination of whether a patient is a carrier of a particular germline mutation or has a particular genotype or zygosity status with respect to that mutation is based on genetic evaluation of normal cells (as opposed to tumor cells) from the patient, e.g., blood cells or cells from a buccal swab. Administration of immunomodulatory agents

[0145] The immunomodulators of the present invention can be administered to a patient in a therapeutically effective amount. When the immunomodulator is radiation, the radiation can be administered by external beam radiation therapy (e.g., x-rays or gamma rays, photon beams, proton therapy, electron therapy) or stereotactic body radiation therapy (SBRT), or by brachytherapy, i.e., by internally placed radioactive material. Radiation can also be administered systemically, for example, by mouth or intravenous injection. Therapeutically effective doses for external beam radiation therapy or SBRT can be in the range of 20-80 Gy / Kg, 40-70 Gy / Kg, or 60-80 Gy / Kg. Therapeutic effective doses for brachytherapy can deliver doses of up to 150 Gy over several months or 60 Gy in 7 days.

[0146] Generally, a therapeutically effective amount of a drug, e.g., an antibody or non-biological drug, is in the range of 0.1 mg / kg to 100 mg / kg, e.g., 1 mg / kg to 100 mg / kg, e.g., 1 mg / kg to 10 mg / kg. The amount administered will depend on variables such as the type and severity of the disease or indication being treated, the patient's overall health, the in vivo efficacy of the drug, the pharmaceutical formulation, and the route of administration. The initial dosage can be increased above an upper level to rapidly achieve the desired blood or tissue level. Alternatively, the initial dosage can be lower than the optimal amount, and the dosage can be gradually increased over the course of treatment. The optimal dosage can be determined by routine experimentation. For parenteral administration, a dose of between 0.1 mg / kg and 100 mg / kg, alternatively between 0.5 mg / kg and 50 mg / kg, alternatively between 1 mg / kg and 25 mg / kg, alternatively between 2 mg / kg and 10 mg / kg, alternatively between 5 mg / kg and 10 mg / kg, is administered, for example, once weekly, once every other week, once every three weeks, or once monthly per treatment cycle. In one embodiment, the dose is 200 mg intravenously every three weeks, while in another embodiment, the dose is 2 mg / kg intravenously every three weeks. In another embodiment, the dose is 240 mg intravenously every two weeks, while in yet another embodiment, the dose is 3 mg / kg intravenously every two weeks. In yet another embodiment, the dose is 1200 mg intravenously every three weeks.

[0147] For therapeutic use, the immunomodulators of the present invention are preferably combined with a pharmaceutically acceptable carrier. As used herein, "pharmaceutically acceptable carrier" refers to buffers, carriers, and excipients suitable for use in contact with human and animal tissues without excessive toxicity, irritation, allergic responses, or other problems or complications, commensurate with a reasonable benefit / risk ratio. The carrier(s) should be "acceptable" in the sense of being compatible with the other ingredients of the formulation and not harmful to the recipient. Pharmaceutically acceptable carriers include buffers, solvents, dispersion media, coatings, isotonic and absorption delaying agents, and the like, that are compatible with pharmaceutical administration. The use of such media and agents for pharmaceutically active substances is known in the art.

[0148] Pharmaceutical compositions containing immunomodulatory agents can be presented in unit dosage form and can be prepared by any suitable method. Pharmaceutical compositions should be formulated to be compatible with their intended route of administration. Examples of routes of administration are intravenous (IV), intradermal, inhalation, transdermal, topical, transmucosal, and rectal administration. Pharmaceutical compositions can be administered intravenously or intravenously for therapeutic treatment. For parenteral, intranasal, topical, oral or local administration, such as by transdermal means It is contemplated that the pharmaceutical compositions can be administered parenterally (e.g., by intravenous, intramuscular, or subcutaneous injection), or by oral ingestion, or by local application in a blood vessel or in an area affected by a cancerous condition, or by intraarticular injection. Additional routes of administration include intravascular, intraarterial, intratumoral, intraperitoneal, intraventricular, intraepidural, as well as nasal, ophthalmic, intramuscular, and intravenous. This includes intrathecal, intraorbital, rectal, topical or aerosol inhalation administration.

[0149] The present invention provides compositions for parenteral administration comprising the above-described agents dissolved or suspended in an acceptable carrier, preferably an aqueous carrier such as water, buffered water, saline, PBS, etc. The compositions can contain pharmaceutically acceptable auxiliary substances required to approximate physiological conditions, such as pH adjusting and buffering agents, tonicity adjusting agents, wetting agents, detergents, etc. The present invention also provides compositions for oral delivery, which can contain inactive ingredients, such as binders or fillers for tablets, capsules, and other formulations. Furthermore, the present invention provides compositions for topical administration, which can contain inactive ingredients, such as solvents or emulsifiers for creams, ointments, and other formulations.

[0150] The preferred route of administration for antibodies is IV infusion. Useful formulations can be prepared by methods well known in the pharmaceutical arts. See, for example, Remington's Pharmaceutical Sciences, 18th ed. (Mack Publishing Company, 1990). Suitable formulations for parenteral administration include: The formulation components include a sterile diluent such as distilled water for injection, saline solution, fixed oils, polyethylene glycol, glycerin, propylene glycol, or other synthetic solvents; an antibacterial agent such as benzyl alcohol or methylparaben; an antioxidant such as ascorbic acid or sodium bisulfite; a chelating agent such as EDTA; a buffer such as acetate buffer, citrate buffer, or phosphate buffer; and an agent for adjusting tonicity such as sodium chloride or dextrose.

[0151] For intravenous administration, suitable carriers include physiological saline, bacteriostatic water, Cremophor EL™ (BASF, Parsippany, NJ) or phosphate buffered saline (PBS). The carrier should be stable under the conditions of manufacture and storage and preserved against microorganisms. The carrier can be, for example, a solvent or dispersion medium containing water, ethanol, polyol (e.g., glycerol, propylene glycol, and liquid polyethylene glycol), and suitable mixtures thereof.

[0152] The pharmaceutical formulation is preferably sterile. Sterilization can be achieved, for example, by filtration through a sterile filtration membrane. If the composition is lyophilized, sterilization by filtration can be performed prior to or after lyophilization and reconstitution. The aqueous solution can be packaged for immediate use or can be lyophilized, with the lyophilized preparation being combined with a sterile aqueous carrier prior to administration. The pH of the preparation will typically be between 3 and 11, more preferably between 5 and 9 or between 6 and 8, and most preferably between 7 and 8, such as between 7 and 7.5. The resulting solid form of the composition can be packaged in a plurality of single-dose units, each containing a fixed amount of one or more of the above-mentioned agents, such as in a sealed package of tablets or capsules.

[0153] In one embodiment, the pharmaceutical carrier for the antibody formulation is, for example, mannitol, penetetic acid, polysorbate 80, sodium chloride, sodium citrate dihydrate. The formulation may contain sodium citrate dehydrate and sterile water. An acid, such as hydrochloric acid, and / or sodium hydroxide may be added as needed to adjust the pH of the formulation. In one embodiment, the antibody is formulated in a liquid solution, each mL of which contains mannitol (30 mg), pentetic acid (0.008 mg), polysorbate 80 (0.2 mg), sodium chloride (2.92 mg), sodium citrate dihydrate (5.88 mg), and sterile water for injection (USP). In one particular embodiment, the antibody is nivolumab and is contained in a liquid formulation, wherein each mL of solution contains nivolumab (10 mg), mannitol (30 mg), pentetic acid (0.008 mg), polysorbate 80 (0.2 mg), sodium chloride (2.92 mg), sodium citrate dihydrate (5.88 mg) and sterile water for injection (USP), and hydrochloric acid and / or sodium hydroxide, if needed, to adjust the pH to 6.

[0154] In another embodiment, the pharmaceutical carrier for the antibody formulation can include, for example, L-histidine, polysorbate, and sucrose. Acids such as hydrochloric acid or sodium hydroxide can be added as needed to adjust the pH of the formulation. In a specific embodiment, the antibody is formulated in a liquid solution, where each mL of the solution contains L-histidine (1.55 mg), polysorbate 80 (0.2 mg), sucrose (70 mg), and sterile water for injection, USP. In a specific embodiment, the antibody is pembrolizumab, and is contained in a liquid formulation, where each mL of the solution contains 25 mg of pembrolizumab, and is formulated in L-histidine (1.55 mg), polysorbate 80 (0.2 mg), sucrose (70 mg), and sterile water for injection, USP.

[0155] In another embodiment, a pharmaceutical carrier for an antibody formulation can include, for example, glacial acetic acid, L-histidine, sucrose, and polysorbate 80. In one particular embodiment, the antibody is formulated in a liquid solution at pH 5.8, where each mL of the solution contains glacial acetic acid (16.5 mg), L-histidine (62 mg), sucrose (821.6 mg), polysorbate 20 (8 mg). In a further embodiment, the antibody is atezolizumab and is contained in a liquid formulation, where each mL of the solution contains glacial acetic acid (16.5 mg), L-histidine (62 mg), sucrose (821.6 mg), polysorbate 20 (8 mg), pH 5.8. cancer

[0156] Cancers that can be treated according to the methods of the present invention, and cancers for which a patient's response to a treatment regimen can be determined according to the methods of the present invention, include melanoma (including unresectable or metastatic melanoma), prostate cancer, lung cancer (including non-small cell lung cancer and metastatic non-small cell lung cancer), adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, cancer of the brain or central nervous system, basal cell skin cancer, breast cancer, cervical cancer, colorectal cancer, endometrial cancer, esophageal cancer, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumors, gastrointestinal stromal tumors (GIST), gastric cancer, glioma, glioblastoma, head and neck cancer (including head and neck squamous cell carcinoma), Hodgkin's disease, classical Hodgkin's lymphoma, and the like. lymphoma, diffuse large B-cell lymphoma, follicular lymphoma, Kaposi's sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia (including acute myeloid leukemia), liver cancer (including hepatocellular carcinoma), lymphoma, malignant mesothelioma, Merkel cell carcinoma, metastatic urothelial carcinoma, multiple myeloma, myeloma, myelodysplastic syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroendocrine carcinoma, neuroblastoma, non-Hodgkin's lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumor, kidney cancer (including renal cell carcinoma), retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, squamous cell skin cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine cancer, or vaginal cancer. [Example]

[0157] Example 1: Patient response based on chi-square analysis of biomarkers Genetic analysis of 85 cancer patients was performed to determine the germline variants carried by these patients. Patients were categorized as responders (complete, partial) or non-responders (progressive, stable disease) to various anti-PD1 and anti-PDL1 antibody therapies, including nivolumab, pembrolizumab, durvalumab, and atezolizumab. An extensive panel of biomarkers was tested for each patient, and biomarkers were analyzed to assess response to therapy and the correlation between each of these biomarkers. Chi-square analysis was performed to determine the significance of the p-values ​​of various biomarkers correlated with patient response or non-response to treatment. Results for a subset of these tested biomarkers are shown in Table 1 below. [Table 1]

[0158] A p-value close to 0.05, preferably lower than 0.05, indicates greater significance of the correlation between a patient's genotype and their response to anti-PDL1 or anti-PD1 therapy. As shown in Table 1, other markers were also found to be relevant in determining a patient's likelihood of responding to immunomodulatory therapy, but IL18R1 / rs14465660, CD44 / rs11821102, and IL10RB / rs11821102 had the lowest p-values, indicating that these biomarkers had the strongest correlation with response. Example 2: Biomarker-driven identification of positive responders

[0159] The goal of this example was to find prognostic rules that would assign patients a probability of responding to treatment given a panel of genetic signatures. Two classification techniques readily applicable to the case of categorical predictors were used and compared: classification trees and random forests.

[0160] Table 2 below summarizes the comparison of classification strategies, showing the accuracy, sensitivity and specificity for each model based on the number of patients, the number of markers analyzed and the mutations considered. [Table 2] *0 = wild type; 1 = heterozygous for the mutation; 2 = homozygous for the mutation **0 = wild type; 1 = heterozygous or homozygous for the mutation Example 3: Tree-based classification rules for predicting response

[0161] Tree-based classification provides intuitive and easily interpreted prognostic rules. Each leaf of the tree provides the probability of response to treatment and the proportion of samples that fall into each leaf category. Figure 2 shows the tree-based classification rules for response to anti-PD1 or anti-PDL1 antibody therapy based on the analysis of 22 markers for all 85 patients in this study. The tree provides a method for prognosticating patient response to treatment based on four genes: CD44, IL18R1, miR99a promoter, and EXO1.

[0162] As shown in Figure 2, CD44 is the first gene in predictive value. If the patient is heterozygous for CD44 / rs11821102 (1), the patient is considered a non-responder (by following "Yes" and proceeding left in the tree). This node in the tree indicates a 0% chance of response based on 5 patients (6% of 85). In contrast, if the patient is not heterozygous (0, 2), other markers should be considered before determining whether the patient is predicted as a responder or non-responder.

[0163] If the patient is not heterozygous for CD44 / rs11821102 (0, 2), the presence or absence of IL18R1 / rs11465660 should next be determined. If the patient is homozygous wild-type (0) or homozygous for IL18R1 / rs11465660 (2), other markers should be considered before determining whether the patient is predicted as a responder or non-responder. In contrast, if the patient is heterozygous for IL18R1 / rs11465660 (1), the patient is considered a responder. At this node in the tree, the likelihood of response is 75% based on 12 patients (14% of 85 patients).

[0164] If the patient is homozygous wild-type (0) or homozygous for IL18R1 / rs11465660 (2), the presence or absence of a miR99a promoter mutation should be determined. If the patient is homozygous for the miR99a promoter mutation (or data is unavailable) (2), the patient is considered a non-responder because this node in the tree indicates a 25% (based on 12 of 85 patients) likelihood of response. If the patient is heterozygous for the miR99a promoter mutation (1) or homozygous wild-type (0), other markers should be considered before determining whether the patient is predicted as a responder or non-responder.

[0165] If a patient is heterozygous for the miR99a promoter mutation (1) or homozygous wild-type (0), the presence or absence of an EXO1 mutation should be determined. If a patient is heterozygous for the EXO1 mutation (or data is unavailable) (1), the patient is considered an intermediate responder because this node in the tree indicates a 43% chance of response (based on 19 of 85 patients). If a patient is homozygous for the EXO1 mutation (2) or homozygous wild-type (0), the patient is considered a responder because this node in the tree indicates a 55% chance of response (based on 37 of 85 patients). Therefore, patients who are homozygous wild-type or homozygous for the EXO1 mutation are considered to have a better response rate than patients who are heterozygous for the EXO1 mutation, provided they meet the upstream parameters. Example 4: Variable Importance of Response Biomarkers

[0166] The variable importance of each biomarker was assessed. When each marker was added to the classification tree, variable importance was measured as the normalized decrease in cross-validated entropy. The results are shown in Table 3. The higher the value, the more important the variant is as a relevant indicator for predicting response to immunomodulatory agents. Thus, some biomarkers not predicted by the tree method of Example 3 may still be important in predicting patient response to immunomodulatory therapy.

[0167] Based on the overall values ​​generated, the variants in Table 3 can be considered significant for predicting a cancer patient's response to immunomodulatory agents, including but not limited to anti-PDL1 or anti-PD1 antibody therapy. [Table 3]

[0168] Based on these values, the disclosed IL8, RAD23A, CD274 and STAT3 biomarkers can be considered significant for predicting a patient's response to immunomodulatory agents, such as, but not limited to, anti-PDL1 or anti-PD1 antibody therapy. Example 5: Confirmation of the significance of response biomarkers

[0169] In a separate study, genetic analyses were performed to determine germline variants carried by 55 cancer patients treated with anti-PD1 and anti-PDL1 antibody therapies, including nivolumab, pembrolizumab, durvalumab, and atezolizumab. A panel of 29 biomarkers was tested for each patient, and marginal chi-square analysis was used to assess the correlation between response to therapy and each of these biomarkers.

[0170] As in the previous examples, CD274 / rs4742098, IL18R1 / rs11465660, EXO1 / rs4150021, STAT3 / rs3744483, miR99a promoter, and IL10RB / rs2834167 were relevant in determining the likelihood that a patient would respond to immunomodulatory therapy. CD274 / rs4742098, IL18R1 / rs11465660, and EXO1 / rs4150021 had the lowest p-values, indicating that these biomarkers had the strongest correlation with response. [Table 4-1] [Table 4-2]

[0171] Of the 55 cancer patients analyzed, several statistical classifiers were trained on 36 cancer patients who were progression-free after 6 months of treatment. Biomarkers were treated as either categorical variables (i.e., wild-type, heterozygous mutant, or homozygous mutant) or continuous variables by assigning a value of 0 to wild-type, 1 to heterozygous mutant, and 2 to homozygous mutant. Predictive models were fitted for each type of biomarker data. Two sets of classification trees were separately trained at minimum splits and minimum node size over separate grids ranging from 1 to 30 and 1 to 20, respectively. LASSO (least absolute shrinkage and selection) operator was used. A logistic regression model with a penalty operator) was adjusted for the regularization parameter lambda. During training, the adjustment parameters were selected to maximize predictive accuracy using leave-one-out cross-validation. Classification tree and LASSO models were fitted in R (version 3.3.2) using rpart (version 4.1-11) and glmnet (version 2.0-10), respectively. The respective accuracy, sensitivity, and specificity are shown in Table 5 below. [Table 5] *0 = wild type; 1 = heterozygous for the mutation; 2 = homozygous for the mutation **0 = wild type; 1 = heterozygous or homozygous for the mutation

[0172] These results indicate that the biomarkers listed above, and in particular CD274 / rs4742098, are particularly useful in predicting patients who will have a durable response to immunotherapy versus those who will not. Example 6: Toxicity prediction based on chi-square analysis of biomarkers

[0173] Genetic analysis of 90 cancer patients was performed to determine the SNP biomarkers carried by these patients. Patients were categorized as having a toxic response (grade 2 or higher based on RECIST criteria) or a non-toxic response (grade 0 or 1 based on RECIST criteria) to various anti-PD1 and anti-PDL1 antibody therapies, including nivolumab, pembrolizumab, durvalumab, and atezolizumab. An extensive panel of biomarkers was tested for each patient, and the biomarkers were analyzed to assess the correlation between the patient's toxicity score and each of these biomarkers. Chi-square analysis was performed to determine the significance of the p-values ​​of the various biomarkers, correlated with the patient's level of toxic response, on a scale of grades 0 to 5 (based on the RECIST criteria described herein). Results for a subset of these tested biomarkers are shown in Table 6 below. [Table 6]

[0174] A p-value close to 0.05, preferably lower than 0.05, indicates greater significance of the correlation between a patient's genotype and their toxic response to anti-PDL1 or anti-PD1 therapy. As shown in Table 4, other markers were also found to be relevant in determining the likelihood that a patient would have a toxic response to immunomodulatory therapy, but KRAS / rs61764370, FCGR2A / rs10919033, CD274 / rs4742098, and TLR4 / rs4986790 had the lowest p-values, indicating that these biomarkers had the strongest correlation with toxicity.

[0175] Regarding CD274 / rs4742098, being homozygous for this marker correlates with a high likelihood of toxicity. Example 7: Biomarker-driven identification of toxicity responses

[0176] The objective of this example was to find prognostic rules that would assign patients a probability of having a toxic response (grade 2 or higher) to immunomodulatory agents given a panel of genetic signatures. Two classification techniques readily applicable to the case of categorical predictors were used and compared: classification trees and random forests.

[0177] Table 5 below summarizes the comparison of classification strategies, showing the accuracy, sensitivity and specificity for each model based on the number of patients, the number of markers analyzed, and the mutations considered. [Table 7] 0 = wild type; 1 = heterozygous or homozygous for the mutation Example 8: Tree-based classification rules for predicting toxicity

[0178] Tree-based classification provides intuitive and easily interpreted prognostic rules. Each leaf of the tree provides the probability of response to treatment and the proportion of samples that fall into each leaf category. Figure 3 shows the tree-based classification rules for response to anti-PD1 or anti-PDL1 antibody therapy based on the analysis of 50 markers for all 90 patients in this study. The tree provides a method for prognosticating the likelihood that a patient will have a toxic response to treatment based on three genes - KRAS, TLR4, and FCGR2A.

[0179] As shown in Figure 2, FCGR2A is the first gene in predictive value. If a patient is heterozygous or homozygous for FCGR2A / rs10919033 (1), the patient is considered to have a strong likelihood of a toxic response (by following "No" and proceeding right up the tree). This node in the tree indicates a 67% likelihood of a toxic response based on 12 patients (13% of 90). In contrast, if the patient is homozygous wild-type (0), e.g., the patient does not carry the mutation (or the data is not available), other markers should be considered before determining the likelihood of the patient having a toxic response.

[0180] If a patient is homozygous wild-type (0) for FCGR2A / rs10919033, e.g., the patient does not carry a mutation (or data is unavailable), the presence or absence of TLR4 / rs4986790 should be determined. If a patient is homozygous wild-type (0) for FCGR2A / rs10919033 (or data is unavailable), other markers should be considered before determining whether the patient is predicted to have a toxic vs. non-toxic response. In contrast, if a patient is heterozygous or homozygous for TLR4 / rs4986790 (1), the patient is considered a toxic responder. At this node in the tree, the likelihood of a toxic response is 67% based on 9 patients (10% of 90 patients).

[0181] If the patient is homozygous wild-type (0) for TLR4 / rs4986790 (or data is unavailable), the presence or absence of KRAS / rs61764370 should be determined. If the patient is homozygous for wild-type (0), this node of the tree indicates a 29% likelihood of a toxic response (based on 53 of 90 patients), and the patient is therefore not considered to have a low likelihood of a toxic response. If the patient is homozygous or heterozygous for KRAS / rs61764370 (1), this node of the tree indicates a 6% likelihood of a toxic response (based on 16 of 90 patients), and the patient is therefore considered to have a low likelihood of a toxic response. Such patients may be predicted to have a non-toxic response. Example 9: Variable Importance of Toxicity Biomarkers

[0182] The variable importance of each biomarker was assessed. When each marker was added to the classification tree, variable importance was measured as the normalized decrease in cross-validated entropy. The results are shown in Table 6. The higher the value, the more important the variant is as an indicator of a toxic response to immunomodulatory agents. Thus, some biomarkers not predicted by the tree method of Example 7 may still be important in predicting whether a patient will have a toxic response to immunomodulatory therapy.

[0183] Based on the overall values ​​generated, the mutations in Table 8 can be considered significant for predicting the toxic response of cancer patients to immunomodulatory agents, e.g., anti-PDL1 or anti-PD1 antibody therapy. [Table 8]

[0184] Based on these values, the disclosed FCGR2A, IL10RB and MSH2 mutations can be considered significant for predicting a patient's toxic response to immunomodulatory agents, such as, but not limited to, anti-PDL1 or anti-PD1 antibody therapy. Example 10: Prediction of toxicity to radiation therapy

[0185] Genetic analysis of 90 cancer patients was performed to determine the biomarkers possessed by these patients. Patients receiving radiation therapy as treatment for cancer were categorized as having a toxic response (grade 2 or higher) or a non-toxic response (grade 0 or 1). An extensive panel of biomarkers was tested for each patient, and the biomarkers were analyzed to assess the correlation between the patient's radiation toxicity score and each of these biomarkers. Chi-square analysis was performed to determine the significance by p-value of various biomarkers correlated with the patient's level of toxic response to radiation. The results for a subset of these tested biomarkers are shown in Table 9 below. [Table 9]

[0186] A p-value close to 0.05, preferably lower than 0.05, indicates a greater significance of the correlation between a patient's genotype and whether the patient had a toxic response to radiotherapy. As shown in Table 7, the p-values ​​of CD274 / rs4742098, IL10RB / rs2834167, and TLR4 / rs4986790 demonstrate that these markers have a strong correlation with a patient's toxicity score for radiation, and are therefore relevant in assessing the likelihood that a patient will have a toxic response to radiotherapy.

[0187] Analysis of these markers also showed that for TLR4 / rs4986790, the marker acts in a dominant manner, and therefore patients experience a protective effect from radiation toxicity if they are heterozygous or homozygous for the marker.

[0188] Analysis of these markers also showed that, for CD274 / rs4742098, the marker acts in a recessive manner, so patients are at risk for a toxic response to radiation if they are homozygous for the marker.

[0189] Analysis of these markers also showed that for IL10RB / rs2834167, the marker acts in a recessive manner, so patients are at risk for a toxic response to radiation if they are homozygous for the marker. Example 11: Biomarker-driven identification of toxicity responses is not cancer type specific

[0190] Analysis of biomarkers associated with toxic responses to anti-PD1 and anti-PDL1 antibody therapies (including nivolumab, pembrolizumab, durvalumab, and atezolizumab) was performed across various cancers, including melanoma and prostate. As shown in previous examples, RAC1 / rs9374, KRAS / rs61764370, and FCGR2A / rs10919033 were relevant in determining the likelihood that patients would have a toxic response to immunomodulatory therapy across various cancers.

[0191] Of the biomarkers tested, RAC1-rs9374 was most strongly associated with toxicity across all cancer types. Using a marginal chi-square test comparing the non-toxic and toxic groups, the p-value for RAC1-rs9374 was calculated and found to be 0.001 (54 patients) for the melanoma dataset, 0.046 (30 patients) for the prostate cancer dataset, and 0.001 (49 patients) for the all other cancer dataset, for an overall p-value across cancer types of 0.000.

[0192] Several statistical classifiers were trained on the entire set of cancer patients evaluated for toxicity. Subjects were classified as experiencing high toxicity (maximum grade of 2 or higher) versus low toxicity (maximum grade of less than 2). Prediction models for high-grade toxicity were based on common markers between the training and test samples. Biomarkers were treated as either categorical variables (i.e., wild-type, heterozygous mutant, or homozygous mutant) or continuous variables by assigning a value of 0 to wild-type, 1 to heterozygous mutant, and 2 to homozygous mutant. Prediction models were fitted for each type of biomarker data. Two sets of classification trees were separately trained with minimum splits and minimum node size over distinct grids ranging from 1 to 30 and 1 to 20, respectively. Logistic regression models with lasso penalties were trained on the regularization parameter lambda. During training, training parameters were selected to maximize predictive accuracy using leave-one-out cross-validation. The CT and LASSO models were fitted in R (version 3.3.2) calling rpart (version 4.1-11) and glmnet (version 2.0-10), respectively. A total of six missing SNP values ​​were imputed by linkage equations with the SNPs treated as categorical variables. Imputation was performed in R calling mi (version 1.0) with a maximum of 20 iterations. Final misclassification error rates were estimated on validation data not used in training. The respective accuracy, sensitivity, and specificity are shown in Table 12 below.

[0193] [Table 10]

[0194] These results provide further evidence of the significance of the biomarkers identified above in predicting whether a patient will have a low likelihood of exhibiting a toxic response to cancer treatment with an immunomodulatory agent, and show that this prediction is valid regardless of the type of cancer being treated.

[0195] In a specific embodiment, these results indicate that patients who do not possess an A nucleotide at the position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374) have a reduced likelihood of exhibiting a toxic response to cancer treatment with an immunomodulatory agent, regardless of the type of cancer being treated. The present invention provides, for example, the following items. (Item 1) 1. A method of treating cancer, comprising administering an immunomodulatory agent to a patient identified as not possessing a G nucleotide at a position corresponding to position 101 of SEQ ID NO:2 (CD274 / rs4742098). (Item 2) 1. A method of treating cancer, comprising administering an immunomodulatory agent to a patient identified as carrying an A nucleotide at a position corresponding to position 101 of SEQ ID NO:7 (IL18R1 / rs11465660). (Item 3) A method of treating cancer, comprising administering an immunomodulatory agent to a patient identified as not carrying a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021). (Item 4) 1. A method of treating cancer, comprising administering an immunomodulatory agent to a patient identified as not heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO:1 (CD44 / rs11821102). (Item 5) 5. The method of item 4, wherein the patient is further identified as heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660). (Item 6) 5. The method of item 4, wherein the patient is further identified as not heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO:7 (IL18R1 / rs11465660) and not homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO:8 (miR99a promoter). (Item 7) 7. The method of item 6, wherein the patient is further identified as heterozygous for a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021). (Item 8) 7. The method of item 6, wherein the patient is further identified as not heterozygous for a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021). (Item 9) 1. A method of treating cancer, comprising: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102); b) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); f) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); and g) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483) administering an immunomodulatory agent to a patient identified as carrying or not carrying one or more mutations selected from the group consisting of: (Item 10) The patient: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102); b) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); and d) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter) The method of item 9, wherein the patient is identified as not carrying one or more mutations selected from the group consisting of: (Item 11) The patient: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); c) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); and d) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483) The method of item 6, wherein the patient is identified as carrying one or more mutations selected from the group consisting of: (Item 12) A method for determining the responsiveness of a cancer patient to treatment with an immunomodulatory agent, comprising the step of determining whether the patient is heterozygous for an A nucleotide (CD44 / rs11821102) at a position corresponding to position 101 of SEQ ID NO: 1, wherein not being heterozygous for an A nucleotide (CD44 / rs11821102) at a position corresponding to position 101 of SEQ ID NO: 1 indicates that the patient has a high probability of responding to the immunomodulatory agent. (Item 13) 13. The method of claim 12, further comprising determining whether the patient is heterozygous for an A nucleotide (IL18R1 / rs11465660) at a position corresponding to position 101 of SEQ ID NO:7, wherein being heterozygous for an A nucleotide (IL18R1 / rs11465660) at a position corresponding to position 101 of SEQ ID NO:7 indicates that the patient has a high probability of responding to the immunomodulatory agent. (Item 14) 13. The method of claim 12, further comprising determining whether the patient is heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO:7 (IL18R1 / rs11465660) and whether the patient is homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO:8 (miR99a promoter), wherein being not heterozygous for an A nucleotide at a position corresponding to position 101 of SEQ ID NO:7 (IL18R1 / rs11465660) and not homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO:8 (miR99a promoter) indicates that the patient has a high probability of responding to the immunomodulatory agent. (Item 15) 15. The method of claim 14, further comprising determining whether the patient is heterozygous for a deletion of a T nucleotide (EXO1 / rs4150021) at a position corresponding to position 101 of SEQ ID NO: 19, wherein being heterozygous for a deletion of a T nucleotide (EXO1 / rs4150021) at a position corresponding to position 101 of SEQ ID NO: 19 indicates that the patient has a high probability of responding to the immunomodulatory agent. (Item 16) 17. The method of claim 14, further comprising determining whether the patient is heterozygous for a deletion of a T nucleotide sequence (EXO1 / rs4150021) at a position corresponding to position 101 of SEQ ID NO: 19, wherein determining that the patient is not heterozygous for a deletion of a T nucleotide (EXO1 / rs4150021) at a position corresponding to position 101 of SEQ ID NO: 19 indicates that the patient has a high probability of responding to the immunomodulatory agent. 1. A method for determining the responsiveness of a cancer patient to treatment with an immunomodulatory agent, comprising: a) A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102 / ; b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and c) a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); f) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); g) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660); h) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); i) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); and j) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483) A method comprising determining whether the individual possesses one or more mutations selected from the group consisting of: (Item 18) The patient, a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 1 (CD44 / rs11821102); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and c) a deletion of a T nucleotide at a position corresponding to position 101 of SEQ ID NO: 19 (EXO1 / rs4150021); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); or f) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter) 18. The method of claim 17, wherein the patient has a high probability of responding to the immunomodulatory agent if the patient does not carry one or more mutations selected from: (Item 19) The patient: a) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 4 (IL8 / rs4073); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 5 (IL10 / rs3024496); c) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 7 (IL18R1 / rs11465660); d) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 9 (RAD23A / rs8240); or e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 10 (STAT3 / rs3744483) 18. The method of claim 17, wherein the patient has a high probability of responding to the immunomodulatory agent if the patient carries one or more mutations selected from: (Item 20) 20. The method of any one of items 1 to 19, wherein the cancer is melanoma. (Item 21) 20. The method according to any one of items 1 to 19, wherein the cancer is prostate cancer. (Item 22) 20. The method according to any one of items 1 to 19, wherein the cancer is lung cancer. (Item 23) The cancer is selected from the group consisting of adrenal gland cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain / CNS, basal cell skin cancer, breast cancer, cervical cancer, colorectal cancer, endometrial cancer, esophageal cancer, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumor, gastrointestinal stromal tumor (GIST), gastric cancer, glioma, glioblastoma, head and neck cancer, Hodgkin's disease, Kaposi's sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia, liver cancer, lymphoma, malignant mesothelioma, Merkel cell carcinoma, metastatic urothelial carcinoma, 20. The method of any one of items 1 to 19, wherein the cancer is selected from multiple myeloma, myeloma, myelodysplastic syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroendocrine cancer, neuroblastoma, non-Hodgkin's lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumor, renal cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, squamous cell skin cancer, small intestine cancer, gastric cancer, testicular cancer, thymic cancer, thyroid cancer, uterine cancer, or vaginal cancer. (Item 24) 24. The method of any one of items 1 to 23, wherein the immunomodulatory agent is an anti-PDL1 or anti-PD1 antibody or portion thereof. (Item 25) 25. The method of any one of items 1 to 24, wherein the patient is progression-free 6 months after starting the treatment with the immunomodulatory agent. (Item 26) 26. The method of any one of items 1 to 25, wherein the patient is a human. (Item 27) A method for treating cancer with reduced toxicity, comprising administering an immunomodulatory agent to a patient suffering from cancer, wherein the patient has been identified as not possessing an A nucleotide (RAC1 / rs9374) at a position corresponding to position 101 of SEQ ID NO: 15. (Item 28) A method for treating cancer with reduced toxicity, comprising administering an immunomodulatory agent to a patient suffering from cancer, wherein the patient has been identified as carrying a G nucleotide (KRAS / rs61764370) at a position corresponding to position 101 of SEQ ID NO: 14. (Item 29) A method for treating cancer with reduced toxicity, comprising administering an immunomodulatory agent to a patient suffering from cancer, wherein the patient has been identified as being neither heterozygous nor homozygous for a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033). (Item 30) 30. The method of item 29, wherein the patient is further identified as being neither heterozygous nor homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790). (Item 31) 31. The method of item 29 or 30, wherein the patient is further identified as heterozygous or homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370). (Item 32) 1. A method of treating cancer with reduced toxicity, comprising administering an immunomodulatory agent to a patient suffering from cancer, wherein the patient: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); f) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); g) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and h) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) The method of claim 1, wherein the subject is identified as carrying or not carrying one or more mutations selected from the group consisting of: (Item 33) 33. The method of claim 32, wherein the immunomodulatory agent is an anti-PD1 or anti-PDL1 antibody. (Item 34) The patient has the following mutation: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); or d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098) 34. The method of claim 32 or 33, wherein the patient is identified as not carrying one or more of: (Item 35) The patient has the following mutation: a) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); c) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); or d) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) The method according to item 32 or 33, wherein the antibody is identified as having one or more of the following: (Item 36) 33. The method of claim 32, wherein the immunomodulatory agent is radiation. (Item 37) the one or more mutations are a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167) 37. The method of claim 36, selected from the group consisting of: (Item 38) said patient having the following genotype: a) being heterozygous or homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); b) not being homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO:2 (CD274 / rs4742098); or c) not being homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167) 38. The method of claim 36 or 37, wherein the patient is identified as having one or more of: (Item 39) 39. The method according to any one of items 36 to 38, wherein the radiotherapy is external beam radiotherapy, stereotactic body radiotherapy or brachytherapy. (Item 40) 1. A method for determining the toxicity of an immunomodulatory agent in a cancer patient, comprising a step of determining whether the patient is heterozygous or homozygous for a C nucleotide (FCGR2A / rs10919033) at a position corresponding to position 101 of SEQ ID NO: 12, wherein if the patient is neither heterozygous nor homozygous for a C nucleotide (FCGR2A / rs10919033) at a position corresponding to position 101 of SEQ ID NO: 12, the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent. (Item 41) 41. The method of claim 40, further comprising determining whether the patient is heterozygous or homozygous for a G nucleotide (TLR4 / rs4986790) at a position corresponding to position 101 of SEQ ID NO: 16, wherein if the patient is neither heterozygous nor homozygous for a G nucleotide (TLR4 / rs4986790) at a position corresponding to position 101 of SEQ ID NO: 16, the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent. (Item 42) 42. The method of claim 40 or 41, further comprising determining whether the patient is heterozygous or homozygous for a G nucleotide (KRAS / rs61764370) at a position corresponding to position 101 of SEQ ID NO: 14, wherein if the patient is heterozygous or homozygous for a G nucleotide (KRAS / rs61764370) at a position corresponding to position 101 of SEQ ID NO: 14, the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent. (Item 43) 1. A method for determining the toxicity of an immunomodulatory agent in a cancer patient, comprising: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370); f) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); g) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374); h) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); i) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and j) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) A method comprising determining whether the individual possesses one or more mutations selected from the group consisting of: (Item 44) 44. The method of claim 43, wherein the immunomodulatory agent is an anti-PD1 or anti-PDL1 antibody. (Item 45) The patient: a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 11 (EREG / rs1460008); b) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 12 (FCGR2A / rs10919033); c) an A nucleotide at a position corresponding to position 101 of SEQ ID NO: 15 (RAC1 / rs9374); d) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); or e) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098) 44. The method of claim 42 or 43, wherein the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent if the patient does not possess one or more mutations selected from: (Item 46) The patient: a) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 13 (FCGR2A / rs1801274); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167); c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 14 (KRAS / rs61764370); d) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 8 (miR99a promoter); or e) a C nucleotide at a position corresponding to position 101 of SEQ ID NO: 33 (MSH2 / rs2303428) 44. The method of claim 42 or 43, wherein the patient has a low likelihood of exhibiting a toxic response to the immunomodulatory agent if the patient carries one or more mutations selected from: (Item 47) 44. The method of claim 43, wherein the immunomodulatory agent is radiation. (Item 48) the one or more mutations are a) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); b) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 2 (CD274 / rs4742098); and / or c) a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167) Item 48. The method according to Item 47, wherein (Item 49) said patient having the following genotype: d) being heterozygous or homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 16 (TLR4 / rs4986790); e) not being homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO:2 (CD274 / rs4742098); or f) not being homozygous for a G nucleotide at a position corresponding to position 101 of SEQ ID NO: 6 (IL10RB / rs2834167) and if the patient has one of these genotypes, the patient is considered to have a low likelihood of exhibiting a toxic response to the radiation. (Item 50) 50. The method according to any one of items 47 to 49, wherein the radiation is external beam radiation, stereotactic body radiotherapy or brachytherapy. (Item 51) 51. The method of any one of items 25 to 50, wherein the cancer is melanoma. (Item 52) 51. The method according to any one of items 25 to 50, wherein the cancer is prostate cancer. (Item 53) 51. The method according to any one of items 25 to 50, wherein the cancer is lung cancer. (Item 54) The cancer is selected from the group consisting of adrenal gland cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain / CNS, basal cell skin cancer, breast cancer, cervical cancer, colorectal cancer, endometrial cancer, esophageal cancer, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumor, gastrointestinal stromal tumor (GIST), gastric cancer, glioma, glioblastoma, head and neck cancer, Hodgkin's disease, Kaposi's sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia, liver cancer, lymphoma, malignant mesothelioma, Merkel cell carcinoma, metastatic urothelial carcinoma, 51. The method of any one of items 25 to 50, wherein the cancer is selected from multiple myeloma, myeloma, myelodysplastic syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroendocrine cancer, neuroblastoma, non-Hodgkin's lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumor, renal cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, squamous cell skin cancer, small intestine cancer, gastric cancer, testicular cancer, thymic cancer, thyroid cancer, uterine cancer, or vaginal cancer. (Item 55) 55. The method of any one of items 25 to 54, wherein the patient is a human.

Claims

[Claim 1] The invention as set forth in the drawings.

Images