WO2011011453A2 - Phenotyping tumor-infiltrating leukocytes - Google Patents

Abstract

The present disclosure relates to an immune signature of tumor infiltrating leukocytes. In particular, the disclosure provides methods and kits for determining the immune signature of tumor infiltrating leukocytes for use in assessing risk of cancer recurrence and long term survival, and for developing a treatment regimen for a cancer patient.

Description

PHENOTYPING TUMOR-INFILTRATING LEUKOCYTES CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit under 35 USC 119(e) of U.S. Provisional Patent Application No. 61/227,035, filed July 20, 2009, which is incorporated herein by reference in its entirety for all purposes.

STATEMENT OF GOVERNMENT SUPPORT

[0002] This invention was made with United States Government support under

Department of Defense, United States Army Medical Research and Material Command grant W81XWH-06-1-0416. The United States Government has certain rights in the invention.

FIELD

[0003] The present disclosure relates to an immune signature of tumor infiltrating leukocytes. In particular, the disclosure provides methods and kits for determining the immune signature of tumor infiltrating leukocytes for use in assessing risk of cancer recurrence and long term survival, and for developing a treatment regimen for a cancer patient.

BACKGROUND

[0004] Cancer is a heterogeneous disease. There is widespread interest in identifying molecular characteristics of tumors indicative of their behavior and response to therapy in order to reduce morbidity and mortality. A better understanding of the molecular characteristics of tumors and their immune environment would also be valuable in reducing the use of toxic chemotherapy drugs on patients for whom these medicines would be ineffective and/or unnecessary for treating their cancers.

[0005] In the case of breast cancer, genome wide analyses of the associations between gene copy number (Bergamaschi et al., Genes Chromosomes Cancer, 45:1033-1040, 2006; and Chin et al., Cancer Cell, 10:529-541, 2006), gene expression (Sorlie et al., Proc Natl Acad Sci USA, 98:10869-10874, 2001; Perou et al., Nature, 406:747-752, 2000; Sorlie et al., Proc Natl Acad Sci USA, 100:8418-8423, 2003; van de Vijver et al., N Engl J Med, 347:1999-2009, 2002; van 't Veer, et al., Nature, 415:530-536, 2002; Rakha et al., Cancer, 109:25-32, 2007; Finnegan and Carey, Future Oncology, 3:55-63, 2007, and Cheang et al., Clin Cancer Res, 14:1368-1376, 2008), and clinical outcome have revealed two major breast cancer subtypes, "basal" and "luminal/amplifier," which are associated with distant recurrence and short survival duration. Basal subtype tumors are referred to as "triple negatives" because they do not express the estrogen receptor (ER), the progesterone receptor (PR) or the human epidermal growth factor receptor 2 (HER2/nue or ErbB2). As such basal subtype tumors are not expected to be sensitive to anti-estrogen therapies or trastuzumab.

[0006] Currently, in the clinical setting, the standard of care for breast cancer is to evaluate breast cancer tissue for prognostic markers including estrogen receptor (ER), progesterone receptor (PR), and HER2 expression levels. Independently, these biomarkers are used to help guide treatment decisions and overall disease management. More recently, gene signature approaches have been optimized by several groups: 21 -gene Oncotype DX (Genomic Health); 70-gene MammoPrint (Agendia); 76-gene "Rotterdam signature" (Veridex); and 41- gene signature (Ahr et al., J Pathol, 195:312-320, 2001). To varying degrees these approaches have been shown to have predictive value for improving risk assessment of patients with breast cancer. A significant percentage of patients however, develop recurrent disease despite having a tumor epithelial cell gene expression pattern that is predictive of a low risk of progression.

Moreover, some patients having a tumor epithelial cell gene expression pattern that is predictive of a high risk of disease progression, undergo chemotherapy, but may still not have experienced the expected rapid rate of disease progression with a less aggressive course of therapy. Thus, the determination of the gene expression profile of cancer cells alone is not sufficient to fully assess risk.

[0007] In the United States, breast cancer and ovarian cancer are the first and eighth most prevalent cancers in women, and the second and fifth most common cause of cancer- related death in women (U.S. Cancer Statistics Working Group, United States Cancer Statistics: 1999-2006 Incidence and Mortality Web-based Report, Atlanta: U.S. Department of Health and Human Services, Center for Disease Control and Prevention, and National Cancer Institute). The poor ratio of survival to incidence of ovarian cancer is a consequence of late diagnosis and the lack of effective therapies for advanced refractory disease. Despite improvements in surgical techniques and the advent of more targeted therapeutics, survival of ovarian cancer patients is 45% at five years (Jemal et al., CA Cancer J Clin, 58:71-96, 2008). New tools to facilitate epithelial ovarian cancer diagnosis and patient stratification are therefore needed to improve the efficacy of available treatment options.

[0008] Current management of ovarian cancer involves initial surgical debulking, which incorporates total abdominal hysterectomy, bilateral salpingo-oophorectomy and omentectomy with cytological evaluation of peritoneal fluid or washings. One of the most significant predictors of patient outcome to date is the extent of residual disease after primary surgery (Bristow et al., J Clin Oncol, 20:1248-1259, 2002), and current clinical recommendations include complete debulking leaving no macroscopic residuals (Guarneri et al., Gynecol Oncol, 117:152-158, 2010). Adjuvant systemic chemotherapy for ovarian cancer is empiric and initial treatment involves paclitaxel-platinum-based regimens that continue to show improved outcomes compared to other cytotoxic agents such as gecitabine, topotecan and liposomal doxorubicin (Bookman et al., J Clin Oncol, 27:1419-1425, 2009). Despite aggressive surgery and chemotherapy, the majority of ovarian cancer patients relapse within 3-5 years and the median time to relapse is 15 months post diagnosis (Hennessy et al., Lancet, 374:1371-1382, 2009).

[0009] There are separate histological subgroups of epithelial ovarian cancer, including serous subgroups and non-serous subgroups (e.g., endometroid, clear cell and mucinous tumors), which are characterized by different clinical behaviors. Approximately 70% of tumors are serous and have a distinctly worse prognosis than other forms of ovarian cancer (Kobel et al., PLoS Med, 5:e232, 2008). Patient stratification according to histological subtypes is therefore desirable.

[00010] Ovarian cancer patients are further stratified based on their treatment-free interval after platinum-based chemotherapy (Markman and Hoskins, J Clin Oncol, 10:513-514, 1992). Patients were classified as having platinum-resistant ovarian cancer if progression occurred on primary platinum-based therapy, less than a partial response to a platinum-based regimen (Therasse et al., J Natl Cancer Inst, 92:205-216, 2000), or recurrence within six months of completing a platinum-based regimen. Alternatively, patients were classified as having platinum-sensitive ovarian cancer if they demonstrated at least a partial response to a platinum- based regimen and had a treatment-free interval of more than six months. This classification is particularly relevant in determining treatment regimens for recurrent disease as the probability of response to platinum-retreatment is closely related to the duration of platinum-free interval (Markman, Trends Pharmacol Sci, 29:515-519, 2008).

[00011] Clinical and experimental studies have established that chronic infiltration of neoplastic tissue by leukocytes (e.g., chronic inflammation) promotes development and/or progression of various epithelial tumors (de Visser et al., Nature Reviews Cancer 6, 24-37, 2006; and Mantovani et al., Nature, 454:436-444, 2008). The organ- specific cellular and molecular programs that favor pro-tumor, as opposed to anti-tumor immunity, however, are incompletely understood. While some subsets of leukocytes exhibit anti-tumor activity, including cytotoxic CD8⁺ T lymphocytes (CTLs) and natural killer (NK) cells (Dunn et al., Immunity, 21:137-148, 2004), other leukocytes exhibit more bipolar roles. Most notably, mast cells, CD4⁺ T lymphocytes, B lymphocytes, dendritic cells, granulocytes and macrophages, have the capacity to either hinder or potentiate tumor progression (Ostrand-Rosenberg, Curr Opin Genet Dev, 18:11-18, 2008; and de Visser et al., Cancer Cell, 7:411-423, 2005).

[00012] Several recent studies have analyzed the influence of host immunity on disease prognosis. Tumor-infiltrating CD3+ T cells are strongly associated with favorable prognosis (Zhang et al., N Engl J Med, 348:203-213, 2003; Raspollini et al., Ann Oncol, 16:590-596, 2005; and Nelson, Immnol Rev, 222:101-116, 2008), with a particular emphasis on the CD8+ cytotoxic T cell subset (Hamanishi et al., Proc Natl Acad Sci USA, 104:3360-3365, 2007; Sato et al., Proc Natl Acad Sci USA, 102:18538-18543, 2005; Clarke et al., Mod Pathol, 22:393-402, 2009; and Milne et al., PLoS One, 4:e6412, 2009), suggesting that cytotoxic T lymphocytes (CTLs) play an important role in the antitumor immune response. Accordingly, other factors associated with CTL responses are also associated with an improved prognosis, including interferon-γ (IFN- γ) (Marth et al., AM J Obstet Gynecol, 191:1598-1605; and Kusuda et al., Oncol Rep, 12: 1153-1158, 2005), the IFN- γ receptor (Duncan et al., Clin Cancer Res, 13:4139- 4145, 2007), TNFα (Kusuda, 2005, supra) and MHC class I (Rolland et al., Clin Cancer Res, 13:3591-3596, 2007; and Leffers et al., Gynecol Oncol, 110:365-373, 2008).

[00013] Since tumor growth and development is influenced by the phenotype of the immune response to neoplasia, as well as the genotype of malignant cells, there is a need for tools to characterize a patient's immune response to their tumor. Additional patient stratification criteria would provide valuable guidance for treatment regimen selection.

SUMMARY

[00014] The present disclosure relates to an immune signature of tumor infiltrating leukocytes. In particular, the disclosure provides methods and kits for determining the immune signature of tumor infiltrating leukocytes for use in assessing risk of cancer recurrence and long term survival, and for developing a treatment regimen for a cancer patient.

[00015] The present disclosure provides methods for assessing risk of poor clinical outcome for a human cancer patient, the method comprising: a) subjecting a tumor sample from the patient to a procedure for quantitation of expression of leukocyte biomarkers comprising CD4, CD8 and CD68; and b) detecting the presence of an immune signature of poor outcome comprising CD4^hl/CD8^lo/CD68^hl or an immune signature of favorable outcome comprising CD4^lo/CD8 ^h7CD68 ^lo in the tumor sample, wherein the immune signature of poor outcome is associated with an increased risk of poor clinical outcome as compared to the immune signature of favorable outcome. In some preferred embodiments, the sample is from a solid tumor. In some embodiments, the procedure for quantitation comprises an antibody-based technique. In some preferred embodiments, the antibody-based technique comprises a procedure selected from but not limited to immunohistochemistry, flow cytometry, antibody microarray, ELISA, western blotting, and magnetic resonance imaging. In other embodiments, the procedure for quantitation comprises a nucleic-acid based technique. In some preferred embodiments, the nucleic acid based technique comprises a procedure selected from but not limited to RT-PCR, nucleic acid microarray, serial analysis of gene expression, massively parallel signature sequencing, in situ hybridization, and northern blotting. In some embodiments, the poor clinical outcome comprises a relative reduction in one or more of overall survival, recurrence-free survival (cancer relapse), and distant recurrence-free survival (cancer metastasis). In some embodiments, the methods further comprise: c) treating the patient with an aggressive treatment regimen when the immune signature of poor outcome is detected. In some embodiments, the favorable clinical outcome comprises a relative increase in one or more of overall survival, recurrence-free survival, and distant recurrence-free survival. In some embodiments, the methods further comprise: c) treating the patient with a conservative treatment regimen when the immune signature of favorable outcome is detected. In some embodiments, the methods further comprise: c) treating the patient with a platinum-based chemotherapy drug when the immune signature of favorable outcome is detected (e.g., indicative of a platinum-sensitive tumor). In some embodiments, the methods further comprise: c) treating the patient with a taxane-based chemotherapy drug (e.g., paclitaxel, docetaxel, etc.) when the immune signature of poor outcome is detected (e.g., indicative of a platinum-resistant tumor). In some embodiments, the tumor sample is from a breast cancer biopsy, lumpectomy or resection. In other embodiments, the tumor sample is selected from but not limited to a breast (e.g., invasive ductal carcinoma, invasive lobular carcinoma, ductal carcinoma in situ, and lobular carcinoma in situ), bladder, cervical, colon, lung, mouth, ovarian, prostate, rectal, renal, testicular, and uterine cancer biopsy. In some embodiments, the methods further comprise determining subtype of the solid tumor. In some embodiments, when the solid tumor is breast cancer, the subtype is selected from the group consisting of basal, luminal A, and triple negative. In some embodiments, the ovarian cancer is epithelial ovarian cancer of a serous subtype. In other embodiments, the ovarian cancer is epithelial ovarian cancer of a non-serous subtype. In some embodiments, the methods further comprise: c) treating the patient with a regimen suitable for a basal or luminal A subtype of breast cancer. In some embodiments, the methods further comprise: c) treating the patient with a regimen suitable for a serous type of epithelial ovarian cancer. In some embodiments, the methods further comprise: c) treating the patient with an immune modulator. In some embodiments, the immune modulator is an anti-Th2 or a pro-Thl immune modulator. In some embodiments, the immune modulator is an inhibitor of colony stimulating factor 1 (CSFl, also known as monocyte colony stimulator factor, or M-CSF) or its receptor. In additional embodiments, the methods further comprise one or both steps before a) of obtaining the tumor sample from the patient, and establishing a cut-off value for distinguishing between high and low expression of the leukocyte biomarkers.

[00016] Moreover the present disclosure provides methods for assessing risk of poor clinical outcome for a human breast cancer or ovarian cancer patient, the method comprising: a) subjecting a breast cancer or epithelial ovarian cancer sample from the patient to an antibody- based technique for quantitation of expression of leukocyte biomarkers comprising CD4, CD8 and CD68; and b) detecting the presence of an immune signature of poor outcome comprising CD4^h7CD8^lo/CD68 ^hl or an immune signature of favorable outcome comprising

CD4^lo/CD8^hl/CD68^l0 in the sample, wherein the immune signature of poor outcome is associated with an increased risk of poor clinical outcome as compared to the immune signature of favorable outcome. In some embodiments, the antibody-based technique comprises

immunohistochemistry. In some embodiments, the poor clinical outcome comprises a relative reduction in one or more of overall survival, recurrence-free survival (cancer relapse), and distant recurrence-free survival (cancer metastasis). In some preferred embodiments, the methods further comprise: c) treating the patient with an aggressive treatment regimen when the immune signature of poor outcome is detected. In other embodiments, the favorable clinical outcome comprises a relative increase in one or more of overall survival, recurrence-free survival, and distant recurrence-free survival. In some embodiments, the breast cancer subtype is selected from the group consisting of basal, luminal A, and triple negative. In some embodiments, the epithelial ovarian cancer of a serous subtype. In other embodiments, the epithelial ovarian cancer of a non-serous subtype. In some preferred embodiments, the method further comprises: c) treating the patient with a conservative treatment regimen when the immune signature of favorable outcome is detected. In some embodiments, the methods further comprise: c) treating the patient with a regimen suitable for a basal or luminal A subtype of breast cancer. In some embodiments, the methods further comprise: c) treating the patient with a regimen suitable for a serous type of epithelial ovarian cancer. In some embodiments, the methods further comprise: c) treating the patient with an immune modulator. In some embodiments, the immune modulator is an anti-Th2 or a pro-Thl immune modulator. In some embodiments, the immune modulator is an inhibitor of colony stimulating factor 1 (CSFl, also known as monocyte colony stimulator factor, or M-CSF) or its receptor. In additional embodiments, the methods further comprise one or both steps before a) of obtaining the breast or epithelial ovarian cancer sample from the patient, and establishing a cut-off value for distinguishing between high and low expression of the leukocyte biomarkers. [00017] In addition, the present disclosure provides kits for assessing risk of poor clinical outcome for a human cancer patient, the kit comprising biomarker- specific reagents consisting essentially of: a) a CD4-specific reagent; b) a CD8-specific reagent; and c) a CD68-specific reagent. In some preferred embodiments, the CD4-specific reagent, the CD8-specific reagent and the CD68-specific reagent are antibodies. In other preferred embodiments, the CD4-specific reagent, the CD8-specific reagent and the CD68-specific reagent are nucleic acids.

BRIEF DESCRIPTION OF THE DRAWINGS

[00018] Figure 1 illustrated that a CD68/CD4/CD8 immune-based signature is a significant independent predictor of patient survival. A-C) Representative high power images (4Ox) with 6Ox inlays of representative human breast cancer specimens showing expression of CD68⁺ (A), CD4⁺ (B) and CD8⁺ (C). D-F) Automated analysis of CD68⁺ (D), CD4⁺ (E) and CD8⁺ (F) immuno-detection reveals a relationship between leukocyte density and overall survival. Kaplan- Meier estimate of overall survival comparing autoscore leukocyte high and low infiltration groups is shown. 179 samples were used in all analyses and log rank (mantel- cox) p values are denoted for difference in overall survival. G-H) A Kaplan-Meier estimate of overall survival comparing CD68^hl/CD4 ^h7CD8^l0 and CD68 ^lo/CD4 ^lo/CD8 ^hl immune profiles as assigned by random forest clustering, to identify optimum thresholds in Cohort I (179 samples). These results were validated in Cohort II (498 samples) In (G) and (H) the log rank (mantel- cox) p value is denoted for the observed difference in overall survival.

[00019] Figure 2 illustrates that a CD68/CD4/CD8 immune -based signature predicts patient survival independently. A-B) A Kaplan-Meier estimate of overall survival is shown comparing CD68^h7CD4 ^h7CD8^lD and CD68 ^lo/CD4 ^lo/CD8 ^hl immune profiles for Luminal A and Basal tumor subtypes in Cohort II.

[00020] Figure 3 illustrates that a CD68/CD4/CD8 immune-based signature is a significant independent predictor of recurrence-free survival in ovarian cancer. A-C)

Representative high power images (2Ox) are provided of representative human ovarian cancer specimens showing expression of CD4⁺ (A), CD68⁺ (B), and CD8⁺ (C). D-F) The automated analysis of CD4⁺ (D), CD68⁺ (E), and CD8⁺ (F) immuno-detection reveals a relationship between leukocyte density and overall survival. A Kaplan- Meier estimate of recurrence-free survival comparing autoscore leukocyte high and low infiltration groups is shown. 76 samples were used in all analyses and log rank (mantel-cox) p values are denoted for the difference in overall survival.

[00021] Figure 4 illustrates that a CD68/CD4/CD8 immune -based signature is a significant independent predictor of recurrence-free survival in ovarian cancer. A) A Kaplan- Meier estimate of overall survival comparing CD68^hl/CD4^hl/CD8^l0 and CD68¹7CD4^lo/CD8^hl immune profiles as assigned by decision tree analysis and 10-fold cross validation was employed. 76 samples were assessed; the log rank (mantel-cox) p value is denoted for difference in overall survival. H) Results from multivariate Cox regression analysis are provided for a 3-marker immune based "signature" considering tumor stage, grade, patient's age at diagnosis and residual disease after primary surgery.

[00022] Figure 5 illustrates that a CD68/CD4/CD8 immune -based signature is a predictor of recurrence-free survival in serous ovarian cancer tumors. A-B) A Kaplan- Meier estimate of overall survival comparing CD68^hl/CD4^hl/CD8^l0 and CD68^lo/CD4^lo/CD8^hl immune profiles as assigned by decision tree analysis and 10-fold cross validation was employed to identify optimum thresholds in non-serous tumors (n = 23) (A) and in serous tumors (n = 53) (B).

[00023] Figure 6 illustrates that immune infiltrates are different in serous compared to non-serous ovarian tumors. A-D) Analysis of mean cell density for CD68⁺ (A), CD20⁺ (B), CD8⁺ (C), and CD4⁺ (D). Error bars represent 2 SEM. * P < 0.05.

DETAILED DESCRIPTION

[00024] A three-marker (CD4, CD8 and CD68) immune-based profile is provided to robustly evaluate a cancer patient's immune response to malignancy at the time of biopsy and/or surgery. Characterization of tumor-infiltrating leukocytes permits predictions to be made regarding clinical outcome and likelihood of progression and/or relapse. As immune signatures are not apparently linked to the molecular and genetic heterogeneity of the tumor subtype; assessing immune responses to solid tumors provides an additional dimension to existing gene expression-based prognostic tools, which heretofore were solely reliant on the underlying tumor cell genetic and epigenetic features.

Prognostic Methods

[00025] In one embodiment, cancer patients with cancer samples with a

CD68^hl/CD4^hl/CD8^l0 immune profile are identified as being at a greater risk for cancer metastasis and/or relapse, and as having a reduced overall survival rate or rate of recurrence-free survival, as compared to cancer patients having cancer samples without a CD68 ^hl /CD4^hl/CD8 ° immune profile. In another embodiment, cancer patients with cancer samples with a

CD68^lo/CD4^lo/CD8^hl immune profile are identified as being at a reduced risk for cancer metastasis and/or relapse, and as having a greater overall survival rate or rate of recurrence-free ssuurrvviivvaall,, aass ccoomm]pared to cancer patients having cancer samples without a CD68^lo/CD4^lo/CD8^hl immune profile. [00026] In one embodiment, threshold levels of each marker are established to define a 'high' or 'low' level of expression of the marker. Depending on the cancer type analyzed, the technique used and the marker examined, different values may be used to define a 'high' or 'low' level of expression of the marker. In order to define 'high' or 'low' levels of expression of a marker, statistical analysis such as random forest clustering may be used in order to identify optimum threshold levels.

A. Antibody Based Methods

[00027] In some embodiments, CD4, CD8, and CD68 levels are determined by using antibody-based methods to determine the levels of each biomarker protein in the tumor sample. Antibody-based methods include various techniques that involve the recognition of CD4, CD8, and CD68 antigens using specific antibodies. For most techniques, monoclonal antibodies are used. However, for some techniques polyclonal antibodies can be used. Commonly used antibody -based techniques to detect the level of one or more proteins in a sample include immunohistochemistry, flow cytometry, antibody microarray, ELISA, western blotting, and magnetic resonance imaging.

[00028] Immunohistochemistry. Immunohistochemistry is the general process of determining the location and/or approximate level of one or more antigens in a tissue sample using antibodies directed against the antigens of interest. Typically, a thin slice of tumor tissue sample is cut from a larger tumor sample and mounted onto a slide, followed by treating the slice of tumor tissue with one or more reagents (including antibodies) to detect the antigens of interest. Immunohistochemistry can also be performed on tissue slices that are not mounted on a slide. In some instances, formalin-fixed and/or paraffin-embedded tissue samples are used for immunohistochemistry. Paraffinized samples can also be deparaffinized in order retrieve antigenicity of proteins.

[00029] During immunohistochemistry, antibody- antigen interactions can be detected through various mechanisms, including conjugating the antibody to an enzyme that can catalyze a color-producing reaction, such as a peroxidase, or conjugating the antibody to a fluorophore. A fluorophore is a molecule that will absorb energy at a specific wavelength and release energy at a different specific wavelength, e.g. fluorescein. The typical immunohistochemistry process involves treating first treating the thin tissue sample with blocking solution to reduce nonspecific background staining, followed by exposing the tissue sample the antibody or antibodies of interest, washing the tissue sample, and then visualizing the antibody-antigen complexes of interest. [00030] Flow Cytometry. To analyze tumor samples by flow cytometry, the tumor sample is processed to separate the tumor into individual cells. The cells are incubated with fluorophore-tagged antibodies of interest, and the collection of cells is processed through a flow cytometer. The flow cytometer uses different wavelengths of light to excite and detect different fluorophores. By analyzing a collection of cells from a tumor sample which have been incubated with fluorophore-tagged antibodies of interest, a measurement of level of the different antigens of interest in the tumor sample can be obtained.

[00031] Antibody Microarray. The general process for an antibody microarray is to bind a collection of antibodies against antigens of interest to a fixed surface (to create the

microarray), to incubate the microarray with a sample that may contain the antigen(s) of interest, and then to add one or more reagents that allow for the detection of antibody microarray-bound antigens of interest. For antibody microarray analysis, a tumor sample is prepared by a homogenization technique which eliminates large tumor particles which could interfere with the function of the antibody microarray, but which preserves the integrity of the antigens of interest. Reagents that can be used for detection of antibody microarray-bound antigens of interest include fluorophore or enzyme-tagged antibodies.

[00032] Enzyme-linked Immunosorbent Assay (ELISA). To detect protein levels in a sample by ELISA, what is commonly known as a 'Sandwich ELISA' is performed. For a Sandwich ELISA, antibodies against an antigen of interest are linked to a surface. The surface- linked antibodies are exposed to a non-specific blocking agent, and then they are incubated with a sample containing the antigens of interest (e.g. in this case, a tumor sample). After incubation, the antibodies are washed to remove unbound material, and then antibodies which bind to the antigen are added. These antibodies can be directly linked to a fluorophore or an enzyme to allow for their detection, or a secondary antibody linked to a fluorophore or an enzyme can be used to detect these antibodies. Through this technique, the level of one or more antigens in a sample can be determined.

[00033] Western Blotting. For western blotting, a tumor sample of interest is

homogenized, and a sample of the tumor is separated by polyacrylamide gel electrophoresis. The electrophoresis step separates proteins in the sample applied to the gel, and the proteins in the gel are next transferred to a membrane. Typically, PVDF or nitrocellulose membranes are used. After transferring proteins from the gel to the membrane, the membrane is treated with a non-specific blocking agent, and then incubated with antibodies against an antigen of interest. After incubation of the sample with the specific antibodies, the membrane is washed, and then treated with a secondary antibody which binds to the specific antibody. The secondary antibody is typically linked to an enzyme, which can be used to create a reaction to detect the location and approximate level of the antigen of interest on the membrane.

[00034] Molecular Resonance Imaging. Levels of proteins in a tumor inside a person's body can be determined through the use of specific molecular resonance imaging probes.

Probes include antibodies directed against antigens of interest in the tumor, which are linked to molecules which can be visualized in the body. Through the use of these techniques, the level of one or more proteins of interest in a tumor inside the body can be determined.

B. Nucleic Acid-Based Methods

[00035] In other embodiments, CD4, CD8, and CD68 levels are determined by using nucleic acid-based methods to determine the levels of each biomarker mRNA in the tumor sample. In general, methods of mRNA level and gene expression profiling can be divided into two large groups: methods based on hybridization analysis of polynucleotides, and methods based on sequencing of polynucleotides. The most commonly used methods known in the art for the quantification of mRNA expression in a sample include northern blotting and in situ hybridization (Parker and Barnes, Methods in Molecular Biology 106:247-283, 1999); RNAse protection assays (Hod, Biotechniques 13:852-854, 1992); and quantitative or semi-quantitative reverse transcription polymerase chain reaction (RT-PCR) (Weis et al., Trends in Genetics 8:263-264, 1992). Representative methods for sequencing-based gene expression analysis include Serial Analysis of Gene Expression (SAGE), and gene expression analysis by massively parallel signature sequencing (MPSS).

[00036] RT-PCR. Of the techniques listed above, the most sensitive and most flexible quantitative method is RT-PCR, which can be used to compare mRNA levels in different sample populations, in normal and tumor tissues, with or without drug treatment, to characterize patterns of gene expression, to discriminate between closely related mRNAs, and to analyze RNA structure.

[00037] The first step is the isolation of mRNA from a target sample. The starting material is typically total RNA isolated from human tumors or tumor cell lines, and

corresponding normal tissues or cell lines, respectively. Thus RNA can be isolated from a variety of primary tumors, including breast, lung, colon, prostate, brain, liver, kidney, pancreas, spleen, thymus, testis, ovary, uterus, etc., tumor, or tumor cell lines, with pooled DNA from healthy donors. If the source of mRNA is a primary tumor, mRNA can be extracted, for example, from frozen or archived paraffin-embedded and fixed (e.g. formalin-fixed) tissue samples. [00038] General methods for mRNA extraction are well known in the art and are disclosed in standard textbooks of molecular biology, including Ausubel et al., Current Protocols of Molecular Biology, John Wiley and Sons (1997). Methods for RNA extraction from paraffin embedded tissues are disclosed, for example, in Rupp and Locker, Lab Invest. 56:A (1987), and De Andres et al., BioTechniques 18:42044 (1995). In particular, RNA isolation can be performed using purification kit, buffer set and protease from commercial manufacturers, such as Qiagen, according to the manufacturer's instructions. Other commercially available RNA isolation kits include MasterPure™ Complete DNA and RNA Purification Kit (EPICENTRE®, Madison, Wis.), and Paraffin Block RNA Isolation Kit (Ambion, Inc.). Total RNA from tissue samples can be isolated using RNA Stat-60 (Tel-Test). RNA prepared from tumor can be isolated, for example, by cesium chloride density gradient centrifugation.

[00039] As RNA cannot serve as a template for PCR, the first step in gene expression profiling by RT-PCR is the reverse transcription of the RNA template into cDNA, followed by its exponential amplification in a PCR reaction. The two most commonly used reverse transcriptases are avilo myeloblastosis virus reverse transcriptase (AMV-RT) and Moloney murine leukemia virus reverse transcriptase (MMLV-RT). The reverse transcription step is typically primed using specific primers, random hexamers, or oligo-dT primers, depending on the circumstances and the goal of expression profiling. For example, extracted RNA can be reverse-transcribed using a GeneAmp RNA PCR kit (Perkin Elmer, Calif., USA), following the manufacturer's instructions. The derived cDNA can then be used as a template in the subsequent PCR reaction.

[00040] Although the PCR step can use a variety of thermostable DNA-dependent DNA polymerases, it typically employs the Taq DNA polymerase, which has a 5'-3' nuclease activity but lacks a 3'-5' proofreading endonuclease activity. Thus, TaqMan® PCR typically utilizes the 5 '-nuclease activity of Taq or Tth polymerase to hydrolyze a hybridization probe bound to its target amplicon, but any enzyme with equivalent 5' nuclease activity can be used. Two oligonucleotide primers are used to generate an amplicon typical of a PCR reaction. A third oligonucleotide, or probe, is designed to detect nucleotide sequence located between the two PCR primers. The probe is non-extendible by Taq DNA polymerase enzyme, and is labeled with a reporter fluorescent dye and a quencher fluorescent dye. Any laser-induced emission from the reporter dye is quenched by the quenching dye when the two dyes are located close together as they are on the probe. During the amplification reaction, the Taq DNA polymerase enzyme cleaves the probe in a template-dependent manner. The resultant probe fragments disassociate in solution, and signal from the released reporter dye is free from the quenching effect of the second fluorophore. One molecule of reporter dye is liberated for each new molecule synthesized, and detection of the unquenched reporter dye provides the basis for quantitative interpretation of the data.

[00041] TAQMAN® RT-PCR can be performed using commercially available equipment, such as, for example, ABI PRISM 7700™ Sequence Detection System™ (Perkin- Elmer-Applied Biosystems, Foster City, Calif., USA), or Lightcycler (Roche Molecular Biochemicals, Mannheim, Germany). In a preferred embodiment, the 5' nuclease procedure is run on a real-time quantitative PCR device such as the ABI PRISM 7700™ Sequence Detection System™. The system consists of a thermocycler, laser, charge-coupled device (CCD), camera and computer. The system amplifies samples in a 96-well format on a thermocycler. During amplification, laser-induced fluorescent signal is detected at the CCD. The system includes software for running the instrument and for analyzing the data.

[00042] 5 '-Nuclease assay data are initially expressed as CT, or the threshold cycle. As discussed above, fluorescence values are recorded during every cycle and represent the amount of product amplified to that point in the amplification reaction. The point when the fluorescent signal is first recorded as statistically significant is the threshold cycle (C_T).

[00043] To minimize errors and the effect of sample-to-sample variation, RT-PCR is usually performed using one or more reference genes as internal standards. The ideal internal standard is expressed at a constant level among different tissues. RNAs most frequently used to normalize patterns of gene expression are mRNAs for the housekeeping genes glyceraldehyde- 3-phosphate-dehydrogenase (GAPDH) and β-actin (ACTB).

[00044] A more recent variation of the RT-PCR technique is the real time quantitative PCR, which measures PCR product accumulation through a dual-labeled fluorigenic probe (i.e., TAQMAN® probe). Real time PCR is compatible both with quantitative competitive PCR, where internal competitor for each target sequence is used for normalization, and with quantitative comparative PCR using a normalization gene contained within the sample, or a housekeeping gene for RT-PCR. For further details see, e.g. Held et al., Genome Research 6:986-994 (1996).

[00045] The steps of a representative protocol for profiling gene expression using fixed, paraffin-embedded tissues as the RNA source, including mRNA isolation, purification, primer extension and amplification are given in various published journal articles for example: T. E. Godfrey et al. J. Molec. Diagnostics 2: 84-91 (2000); K. Specht et al., Am. J. Pathol. 158: 419- 29 (2001); Cronin et al., Am J Pathol 164:35-42 (2004). Briefly, a representative process starts with cutting about 10 μm thick sections of paraffin-embedded tumor tissue samples. The RNA is then extracted, and protein and DNA are removed. After analysis of the RNA concentration, RNA repair and/or amplification steps may be included, if necessary, and RNA is reverse transcribed using gene specific promoters followed by RT-PCR.

[00046] Microarrays. Differential gene expression can also be identified, or confirmed using the microarray technique. Thus, the expression profile of breast cancer-associated genes can be measured in either fresh or paraffin-embedded tumor tissue, using microarray technology. In this method, polynucleotide sequences of interest (including cDNAs and oligonucleotides) are plated, or arrayed, on a microchip substrate. The arrayed sequences are then hybridized with specific probes from cells or tissues of interest. Just as in the RT-PCR method, the source of mRNA typically is total RNA isolated from human tumors or tumor cell lines, and

corresponding normal tissues or cell lines. Thus RNA can be isolated from a variety of primary tumors or tumor cell lines. If the source of mRNA is a primary tumor, mRNA can be extracted, for example, from frozen or archived paraffin-embedded and fixed (e.g. formalin-fixed) tissue samples, which are routinely prepared and preserved in everyday clinical practice.

[00047] In a specific embodiment of the microarray technique, the microarrayed genes are suitable for hybridization under stringent conditions. Fluorescently labeled cDNA probes may be generated through incorporation of fluorescent nucleotides by reverse transcription of RNA extracted from tissues of interest. Labeled cDNA probes applied to the chip hybridize with specificity to each spot of DNA on the array. After stringent washing to remove non-specifically bound probes, the chip is scanned by confocal laser microscopy or by another detection method, such as a CCD camera. Quantitation of hybridization of each arrayed element allows for assessment of corresponding mRNA abundance. With dual color fluorescence, separately labeled cDNA probes generated from two sources of RNA are hybridized pairwise to the array. The relative abundance of the transcripts from the two sources corresponding to each specified gene is thus determined simultaneously. The miniaturized scale of the hybridization affords a convenient and rapid evaluation of the expression pattern for large numbers of genes. Such methods have been shown to have the sensitivity required to detect rare transcripts, which are expressed at a few copies per cell, and to reproducibly detect at least approximately two-fold differences in the expression levels (Schena et al., Proc. Natl. Acad. ScL USA 93(2): 106-149 (1996)). Microarray analysis can be performed by commercially available equipment, following manufacturer's protocols, such as by using the Affymetrix GenChip technology, or Incyte's microarray technology.

[00048] Serial Analysis of Gene Expression (SAGE). Serial analysis of gene expression (SAGE) is a method that allows the simultaneous and quantitative analysis of a large number of gene transcripts, without the need of providing an individual hybridization probe for each transcript. First, a short sequence tag (about 10-14 bp) is generated that contains sufficient information to uniquely identify a transcript, provided that the tag is obtained from a unique position within each transcript. Then, many transcripts are linked together to form long serial molecules, that can be sequenced, revealing the identity of the multiple tags simultaneously. The expression pattern of any population of transcripts can be quantitatively evaluated by determining the abundance of individual tags, and identifying the gene corresponding to each tag. For more details see, e.g. Velculescu et al., Science 270:484-487 (1995); and Velculescu et al., Cell 88:243-51 (1997).

[00049] Gene Expression Analysis by Massively Parallel Signature Sequencing (MPSS). This method, described by Brenner et al., Nature Biotechnology 18:630-634 (2000), is a sequencing approach that combines non-gel-based signature sequencing with in vitro cloning of millions of templates on separate 5 μm diameter microbeads. First, a microbead library of DNA templates is constructed by in vitro cloning. This is followed by the assembly of a planar array of the template-containing microbeads in a flow cell at a high density (typically greater than 3x10 microbeads/cm ). The free ends of the cloned templates on each microbead are analyzed simultaneously, using a fluorescence-based signature sequencing method that does not require DNA fragment separation. This method has been shown to simultaneously and accurately provide, in a single operation, hundreds of thousands of gene signature sequences from a yeast cDNA library.

[00050] General Description of the mRN A Isolation, Purification and Amplification. The steps of a representative protocol for profiling gene expression using fixed, paraffin-embedded tissues as the RNA source, including mRNA isolation, purification, primer extension and amplification are provided in various published journal articles (for example: T. E. Godfrey et al,. /. Molec. Diagnostics 2: 84-91 [2000]; K. Specht et al., Am. J. Pathol. 158: 419-29 [2001]). Briefly, a representative process starts with cutting about 10 μm thick sections of paraffin- embedded tumor tissue samples. The RNA is then extracted, and protein and DNA are removed. After analysis of the RNA concentration, RNA repair and/or amplification steps may be included, if necessary, and RNA is reverse transcribed using gene specific promoters followed by RT-PCR. Finally, the data are analyzed to identify the best treatment option(s) available to the patient on the basis of the characteristic gene expression pattern identified in the tumor sample examined, dependent on the predicted likelihood of cancer recurrence.

Treatment Methods

[00051] In yet another embodiment, determination of CD4, CD8, and CD68 levels in a cancer patient is used to determine an optimal cancer treatment regimen for the patient. For a cancer patient with a cancer sample with a CD68^hl/CD4^h7CD8 ° immune profile, an aggressive cancer treatment regimen may be indicated. Aggressive cancer treatment regimens include but are not limited to local surgical resection of regional tumor tissue, and adjuvant systemic therapies. Surgical resection of regional disease includes lumpectomy, modified mastectomy or total mastectomy. Adjuvant systemic therapies include external radiation, combinatorial chemotherapy (for example six cycles of fluorouracil, doxorubicin and cyclophosphamide), as well as hormone and/or growth factor targeted therapy for patients with ER, PR or HER2 positive disease (e.g., tamoxifen or trastuzumab). For a cancer patient with a cancer sample with a CD68^lo/CD4^lo/CD8^hl immune profile, a conservative cancer treatment regimen may be indicated. Conservative cancer treatment regimens include but are not limited to local surgical resection and hormonal therapy, and would be similar to patients with low risk or Stage I disease with no lymph node involvement.

Table A. Risk Categories for Women with Node-Negative Breast Cancer

Table B. Adjuvant Systemic Treatment Options for Women with Axillary Node-Negative Breast Cancer

*This treatment option is under clinical evaluation. Table C. Treatment Options for Women with Axillary Node-Positive Breast Cancer

* This treatment option is under clinical evaluation.

[00052] In the case of epithelial ovarian cancer, aggressive cancer treatment regimens include initial surgical debulking, which includes total abdominal hysterectomy, bilateral salpingo-oophorectomy and omentectomy with cytological evaluation of peritoneal fluid or washings. Adjuvant systemic therapies include combinatorial chemotherapy (e.g. paclitaxel- platinum-based regimens, gecitabine, topotecan, liposomal doxorubicin).

[00053] Examples of drugs that can be used for invasive breast cancer or epithelial ovarian cancer treatment include but are not limited to: bevacizumab, carboplatin, chlorambucil, cisplatin, cetuximab, cyclophosphamide, cytarabine liposomal, docetaxel, epirubicin, erlotinib, fluorouracil, gefitinib, imatinib, mechlorethamine, methotrexate, mitomycin, mitoxantrone, PARP (poly ADP-ribose polymerase) inhibitors, paclitaxel, thiotepa, vinblastine, vinorelbine, trastuzumab, abraxane, doxorubicin, pamidronate disodium, anastrozole, exemestane, raloxifene, toremifene, letrozole, megestrol, tamoxifen, capecitabine, goserelin acetate, and zoledronic acid.

Identification of Tumor Subtypes

[00054] In some embodiments, the prognostic and treatment methods further comprise the classification of solid tumor subtype, using methods known in the art.

A. Breast Cancer

[00055] Breast cancers are categorized as basal, luminal A, luminal B, or triple-negative subtypes using immunohistochemistry as previously described (Carey et al., JAMA, 295, 2492- 2502, 2006). Basal tumors are defined as estrogen receptor (ER) negative, progesterone receptor (PR) negative, HER2 negative and epidermal growth factor receptor (EGFR) positive. Luminal A tumors are defined as ER, PR and HER2 positive. Luminal B tumors are defined as ER and PR positive, and HER2 negative. Triple negative tumors are as ER, PR and HER2 negative. Basal tumors are a subset of triple negative tumors.

B. Ovarian Cancer

[00056] The classification of ovarian epithelial tumors currently used by pathologists is based entirely on tumor cell morphology (see e.g., Tavassoli, World Health Organization:

Tumours of the Breast and Female Genital Organs, IARC WHO Classification of Tumours, 2003; and Gilks et al., Hum Pathol, 39:1239-1251, 2004). The four major types of epithelial tumors (serous, endometrioid, clear cell, and mucinous) bear strong resemblance to the normal cells lining different organs in the female genital tract. For example, serous, endometrioid, and mucinous tumor cells exhibit morphological features similar to non- neoplastic epithelial cells in the fallopian tube, endometrium, and endocervix. respectively. Λs approximately 70% of ovarian cancers are serous, ovarian tumors can be categorized as serous or non-serous (Cho et al Annu Rev Pathol, 4:287-313, 2009).

Kits

[00057] In another embodiment, a kit of reagents capable of detecting CD4, CD8, and CD68 molecules in a tumor sample is provided. Reagents capable of detecting CD4, CD8 and CD68 molecules include but are not limited to anti-CD4, anti-CD8, and anti-CD68 antibodies, and nucleic acids capable of forming duplexes with CD4, CD8, or CD68 mRNA. Nucleic acids capable of forming duplexes with CD4, CD8 or CD68 mRNA include DNA or RNA sequences which are complementary to the respective mRNA sequence. Reagents capable of detecting CD4, CD8 and CD68 molecules are also typically directly or indirectly linked to a molecule such as a fluorophore or an enzyme which can catalyze detectable reaction, in order to indicate the binding of the reagents to their respective targets.

DEFINITIONS

[00058] To facilitate an understanding of the embodiments disclosed herein, a number of terms and phrases are defined below.

[00059] Aggressive treatment regimen: A cancer treatment regimen in which the emphasis is on killing and/or removing the cancer from the body as thoroughly as possible, to the possible detriment of patient comfort and/or safety. As compared to a conservative treatment regimen, an aggressive treatment regimen may, for example, use higher doses of anticancer therapeutics, higher total treatment times, and more radical surgeries. [00060] Conservative treatment regimen: A cancer treatment regimen in which efforts are made to kill and/or remove the cancer from the body, but which also heavily takes into account patient comfort and/or safety. As compared to an aggressive treatment regimen, a conservative treatment regimen may, for example, use lower doses of anti-cancer therapeutics, lower total treatment times, and less radical surgeries.

[00061] Immuno-based / Antibody-based: Any technique that involves the use of an antibody to detect an antigen. Immuno-based techniques include immunostaining, ELISA, antibody microarray, flow cytometry, and western blotting.

[00062] Nucleic acid-based: Any technique that involves the use of a nucleic acid to detect another nucleic acid. Nucleic acid includes both DNA and RNA. Nucleic acid-based techniques include nucleic acid microarray, RT-PCR, northern blotting, nuclease protection assays, and in situ hybridization.

EXPERIMENTAL

[00063] The present disclosure is described in further detail in the following examples which are not in any way intended to limit the scope of the disclosure as claimed. The attached figures are meant to be considered as integral parts of the specification and description of the disclosure. The following examples are offered to illustrate, but not to limit the claimed disclosure.

[00064] In the experimental disclosure which follows, the following abbreviations apply: M (molar); mM (millimolar); μM (micromolar); nM (nanomolar); mol (moles); mmol

(millimoles); μmol (micromoles); nmol (nanomoles); gm (grams); mg (milligrams); μg

(micrograms); pg (picograms); L (liters); ml and mL (milliliters); μl and μL (microliters); cm (centimeters); mm (millimeters); μm (micrometers); nm (nanometers); U (units); V (volts); MW (molecular weight); sec (seconds); min(s) (minute/minutes); h(s) and hr(s) (hour/hours); ⁰C (degrees Centigrade); QS (quantity sufficient); ND (not done); NA (not applicable); rpm (revolutions per minute); H₂O (water); dt^O (deionized water); aa (amino acid); bp (base pair); kb (kilobase pair); kD (kilodaltons); cDNA (copy or complementary DNA); DNA

(deoxyribonucleic acid); ssDNA (single stranded DNA); dsDNA (double stranded DNA); dNTP (deoxyribonucleotide triphosphate); RNA (ribonucleic acid); OD (optical density); PCR (polymerase chain reaction); RT-PCR (reverse transcription PCR).

[00065] Additional abbreviations include: CTL (cytotoxic T lymphocyte); Th (helper T lymphocyte); NK (natural killer cell); EOC (epithelial ovarian cancer); ER (estrogen receptor); PR (progesterone receptor); OS (overall survival); IHC (immunohistochemistry); and TMA (tissue microarrays).

EXAMPLE 1

Antibody-Based Method for Determining the Immune Signature of

Tumor-Infiltrating Leukocytes in Breast Cancer Patients

[00066] This example describes methods for immunophenotyping tumor-infiltrating leukocytes, and provides results of immune signature analyses in two cohorts of breast cancer patients.

Materials and Methods

[00067] Patients and tumor samples. Tissue microarray studies were conducted on two separate patient cohorts. The screening cohort, described elsewhere in detail (Brennan et al., Clin Cancer Res, 14:2681-2689, 2008), was constructed from 179 cases of invasive breast cancer diagnosed at the Department of Pathology, Malmo University Hospital, Malmo, Sweden, between 2001 and 2002. The median age at diagnosis was 65 and the median follow-up time for overall survival was 52 months. Patients did not receive neo-adjuvant treatment and were treated with either modified radical mastectomy or wide local excision. The median tumor size was 2.2 cm and 62% of the tumors were PR positive and 72% were ER positive. The second (validation) cohort includes 498 patients with primary invasive breast cancer diagnosed at the Malmo University Hospital between 1988 and 1992 These cases belonged to an original cohort of 512 patients as previously described in detail (Paulsson et al., Am J Pathol, 175:334-341, 2009; and Borgquist et al., J Clin Pathol, 61:197-203, 2008). The median age at diagnosis was 65 years and median follow-up time to first breast cancer event was 128 months. Information regarding the date of death was obtained from the regional cause-of-death registries for all patients in both cohorts. Complete treatment data was available for 379 patients, 160 of whom had received adjuvant tamoxifen. Information on adjuvant chemotherapy was available for 382 patients, of which 23 patients received treatment. 200 patients received no adjuvant treatment.

[00068] Immunohistochemistry. Tissue microarray slide sections (4.0 μm) were deparaffinized in xylene, and re-hydrated through descending concentrations of ethanol. For CD4 and CD8 detection, heat-mediated antigen retrieval was performed using microwave treatment for 7 min in a citrate buffer (BioGenex) followed by 3 serial 5 minute washes in phosphate buffered saline (PBS). Antigen retrieval for CD68 detection was accomplished by treatment of slides with Protinase XXV (Lab Vision Inc.) for 5 minutes followed by 4 serial 5 minute washes in PBS. Endogenous peroxidase activity was blocked by incubating slides in 5% hydrogen peroxide (Sigma) diluted in methanol for 20 min, followed by 4 serial washes in PBS. To reduce nonspecific background, slides were pretreated with blocking buffer (10% goat serum (Invitrogen), 1.0 % bovine serum albumen (Sigma) dissolved in PBS for 30 min. Primary antibodies were pre-diluted in blocking buffer to 1: 100 for CD68 (KP-I, NeoMarkers) and CD8 (C8/144B, NeoMarkers) or 1:25 for CD4 (368, Novocastra) and applied to tissue section for 16 hours at 4°C. Signal amplification and development were accomplished utilizing the Ultravision LP Detection System (Thermo Scientific) according to manufactures guidelines. Additional CD4 antibodies were tested including clone 34930 (R&D systems) and clone 4B.12

(Noemarkers) but yielded sub-optimum staining and thus were not used for analysis.

[00069] Automated Image acquisition, management, and analysis. Fully automated image acquisition was employed herein. The Aperio ScanScope XT Slide Scanner (Aperio Technologies, Vista, CA) system was used to capture whole slide digital images with a 2OX objective. Slides were de-arrayed to visualize individual cores, using Spectrum software (Aperio). A tumor nuclear algorithm (IHCMark) was developed in house to quantify DAB

- 99 - positive immune cells(Rexhepaj et al., Breast Cancer Res, 10:R89, 2008). The algorithm calculated the density of immune cells/mm².

[00070] Manual and semi-automated evaluation of immunohistochemical staining. As an alternative to fully automated signal assessment, semi- automated and manual assessment of leukocyte infiltration in tumor specimens is possible. Briefly, CD68, CD8 and CD4 positive cells are scored at 2OX by two independent pathologists and averaged for continuity. In studies using TMAs leukocyte density is quantitated by counting all high power fields (20X) per tissue section (1.1 mm) / 2 sections / patient. In studies using human whole tissue samples, 10 high power fields (20X) are accessed and averaged to generate a leukocyte density score. This can be done completely manually, using a bright light microscope. Alternatively leukocyte infiltration can be assessed by semi-automated image capture at 1OX magnification by OpenLab

(Improvision/PerkinElmer) and quantitation of positive cells utilizing ImageJ (NIH). Briefly, image capture is accomplished on standard Leica DC500 microscope equipped with digital camera. Images are then exported as tiff files and loaded into ImageJ (NIH). ImageJ software is utilized to record manual quantitation and allows for samples to be processed in bulk. For these manually quantitated data, leukocyte infiltration characterized as "high" using a 75th percentile cut-off from the mean leukocyte infiltration for the assessed sample population. All other samples are deemed to fall in the "low" infiltration group.

[00071] Image and Statistical analysis. Image analysis and statistical modeling was accomplished using methods described in detail elsewhere (Brennan et al., Clin Cancer Res, 14:2681-2689, 2008; and Rexhepaj et al., Breast Cancer Res, 10:R89, 2008). Briefly, the leukcocyte infiltration was scored as the total number of leukocytes expressing the marker of interest (e.g., CD68, CD8, CD4) over the total number of tumor cells. The final value of leukocyte infiltration quantitated for each patient was the maximum out of both tissue cores. Patients with less than 200 cells in both tissue cores were discarded from analysis. A classification tree procedure was used to create a tree-based classification model and cases were classified into groups based on a the values of dependent predictor variable(target). Patient survival was used as the target variable for building the predictions trees. These tree models were evaluated in terms prediction accuracy using a 10-fold cross-validation approach. Kaplan- Meier analysis and the log-rank test were used to illustrate differences between overall survival (OS) according to individual CD68, CD4, and CD8 expression. A Cox regression proportional hazards model was employed to estimate the relationship to OS of the CD68/CD4/CD8 immune profile, lymph node status, tumor grade, and HER2, PR and ER status in the patient cohorts. Multivariate models included any variable that displayed a significant association with outcome following univariate analysis. A p-value of <0.05 was considered statistically significant and all calculations were performed using Statistical Package for the Social Sciences (SPSS, Inc.). Random forest clustering (RFC) was performed using R software.

Results

[00072] Single Immune Marker Analysis of Leukocyte Infiltrates. Previous work utilizing transgenic mouse models of mammary carcinogenesis has revealed a tumor-promoting role for TH2-CD4+ T lymphocytes that elicit pro-tumor, as opposed to cytotoxic bioactivities of tumor- associated macrophages and enhancement of pro-metastatic epidermal growth factor receptor signaling programs in malignant mammary epithelial cells (DeNardo et al., Cancer Cell, 16:91- 102, 2009). Without being bound by theory, in addition to the underlying genetic alterations in malignant epithelial cells, the host immune response to neoplasia in the breast was determined during development of the present disclosure to contribute to the histopathologic features of breast cancer. CD4⁺, CD8⁺ and CD68⁺ leukocyte density was assessed by

immunohistochemistry using a tissue microarray (TMA) consisting of tumor tissue representing two independent cohorts of breast cancers (FIG. IA-F). Following digital scanning of stained TMA slides using an Aperio ScanScope XT slide scanner, a fully automated nuclear algorithm was used to discriminate tumor from "normal" tissue, and to quantify CD4⁺, CD8⁺ and CD68⁺ cells. Random forest clustering was employed to identify optimum thresholds for survival analysis. Kaplan Myer analysis for overall survival demonstrates that as single variables "high" infiltration by CD4⁺ cells and low CD8⁺ cell density predict reduced overall survival while CD68⁺ cell density alone showed no statistical difference in overall survival (FIG. ID-F).

[00073] Three-Marker Profile of Leukocyte Infiltrates. Heterotypic interactions between diverse leukocytes populations often determine the outcome of an immune response. Thus combining the predictive power of CD4, CD8 and CD68 should significantly stratify patients represented on the TMA for overall survival by accessing both anti-tumor immune response (e.g., CD8⁺ density), as well as pro-tumor immune responses capable of promoting metastatic spread (e.g., high CD4⁺ and CD68⁺ density). Accordingly an immune profile of

CD68^lo/CD4^lo/CD8^hl was predicted to represent individuals whose breast tumors were controlled by local resection of primary tumor and adjuvant therapy (e.g., patients exhibiting longer relapse-free survival). In contrast, patients bearing an immune response characterized as CD68^h7CD4^h7CD8^l0 were predicted to represent a population of patients at risk for metastasis, relapse and reduced overall survival. Random forest clustering was employed to identify optimum thresholds for discriminating CD68^hl/CD4^h7CD8^l0 and CD68^lo/CD4^lo/CD8^hl in survival analysis using patient cohort I. Utilizing thresholds set for discriminating CD68^h7CD4^h7CD8^l0 and CD687CD47CD8^hl from cohort I this signature was verified in a second independent cohort II including 498 samples with a median follow-up time of 128 months. Kaplan Meier analysis for these two groups demonstrated significantly reduced overall survival in patients bearing the CD68^hl/CD4^hl/CD8^l0 IHC signature (p = 0.008, p = 0.001; FIG IG-H). In addition, Log Rank (Mantel-Cox) analysis revealed the CD68/CD4/CD8 signature had a increased survival hazard ratio of 3.51 and 1.38 respectively, in cohorts one and two (Table 1-1 and Table 1-2). There was no statistical correlation with disease grade, node status, tumor size, ER expression, PR expression or HER2 positivety, indicating that the immune signature predicts breast cancer survival independently of the typical clinical histopathological markers currently employed.

Table 1-1. Log Rank Analysis of Breast Cancer Cohort One

Table 1-2. Log Rank Analysis of Breast Cancer Cohort Two

[00074] Immune Signature Predicts Breast Cancer Survival Independently of Tumor Subtype. The CD68/CD4/CD8 signature was further studied to determine if it correlated with an individual tumor subtype (such as basal or luminal, etc). Multivariate Cox regression analysis of tumor subtypes and the CD68/CD4/CD8 signature demonstrated that indeed the three-marker based signature was independent of luminal B, HER2-positive, basal type or even un- typed triple negative (ERa, PR, HER2 negative) tumors. The CD68/CD4/CD8 signature significantly predicted survival in luminal A and basal tumors, but not luminal B and HER2 positive tumors (FIG. 2A-B and Table 1-3). These analyses indicate that the three-marker immune -based signature is a useful predictor of overall survival for multiple breast cancer subtypes, and as such is an improvement to existing gene expression-based prognostic profiling methods to evaluate risk. In particular this immune signature is predictive for patients with low-risk ER⁺ tumors (such as in the luminal A subtype) or alternatively high-risk triple negative breast tumors.

Table 1-3. Log-Rank (Mantel-Cox) Analysis For The CD68/CD4/CD8 Signature in Various Breast Tumor Subtypes

[00075] Tissue microarrays containing specimens representing various grades of ductal carcinoma in situ (DCIS) are also prepared and examined. The three marker immune-based signature is also expected to stratify patients with this common noninvasive type of breast cancer.

EXAMPLE 2

Nucleic Acid-Based Method for Determining the Immune Signature of

Tumor-Infiltrating Leukocytes in Cancer Patients

[00076] This example describes methods for assessing mRNA levels of three biomarkers of tumor-infiltrating leukocytes, and for retrospectively analyzing the immune signature in published gene expression data sets.

[00077] Assessing mRNA of biomarkers of tumor-infiltrating leukocytes. Isolation and measurement of mRNA from tumor samples can be performed as known in the art (Micke et. al., Laboratory Investigation, 86, 202-211, 2006). Tumor samples are removed from a patient and frozen in liquid nitrogen. The tumor samples are thawed, and cut into approximately 20 micrometer-thick sections. Each section of tumor is placed in a tube containing 300 microliters Trizol (Invitrogen). Sixty microliters of chloroform-isoamylalcohol is then added to the Trizol- tumor sample mixture. The tube is mixed, and centrifuged at 10,000 rpm for 10 minutes. Two liquid phases form during the centrifugation, and after the end of the spin, the aqueous phase is transferred to a new tube and 2 microliters of co-precipitant (PelletPaint, Novagen) and 1 equivalent volume (approximately 160 microliters) isopropanol is added. The tube is mixed and incubated at room temperature for 5 minutes, and then centrifuged at 13,000 rpm for 10 minutes at room temperature, which precipitates the RNA. The supernatant is poured off, and the RNA pellet is then washed twice with 70% ethanol. The 70% ethanol is poured off, and the RNA pellet is dried. The RNA pellet is then resuspended in 20 microliters RNAse-free water.

[00078] Next, the re-suspended RNA is used a template for cDNA synthesis. One microliter of oligo-dT primers (dT17) and 1 microliter 10 nM dNTPmix (Clonetech) is added to 5 microliters of each resuspended RNA solution. The mixtures are heated for 10 minutes at 65 degrees Celsius, and then immediately chilled on ice. To each tube, 4 microliters of first strand buffer, 2 microliters DTT, 1 microliter RNAsin (Promega) and 1 microliter Superscript II reverse transcriptase (Invitrogen) are added, and then the reactions are kept at 42 degrees Celsius for 1 hour to allow reverse transcription to occur. After 1 hour, the reaction is stopped by moving the tubes to 65 degrees Celsius for 20 minutes. Next, the RNA is digested with 1 microliter RNAseH (Clonetech) at 37 degrees Celsius for 30 minutes. [00079] For the PCR, 2 microliters of the cDNA solution (10%) is used for each 40-cycle Sybgreen PCR assay using the Sybrgreen Universal PCR Master Mix (Applied Biosystems) with forward and backward primers for the CD4, CD8, or CD68 cDNA. The PCR reaction is performed with an ABI PRISM 7000HT real-time PCR cycler (Applied Biosystems) using conditions recommended by the manufacturer. Gene expression levels are normalized to expression of the housekeeping gene glyceraldehyde-3 -phosphate dehydrogenase (GAPDH).

[00080] A profile of a combination of high CD68, high CD4, and low CD8 mRNA levels is contemplated to be associated with a greater risk of tumor relapse and/or metastasis as compared to a profile of a combination of low CD68, low CD4, and high CD8 mRNA levels.

[00081] Validation of Immune Profile in Existing Expression Data Bases. Data base mining is done to assess the predictive power of the CD68/CD4/CD8 signature in published data sets. Assessments of single gene expression changes have demonstrated that both CD68 and CD4 genes are enriched in tissue from patients who have relapsed within 5 years, while CD8 is down regulated in tissue of patients who have been in remission during the 5-year follow up time period. The CD68/CD4/CD8 gene expression prognostic signature is assessed by multi- variant analysis using random forest clustering to evaluate cut off points, and the signature is used to improve risk stratification independently of known clinicopathologic factors and previously established prognostic signatures based on unsupervised hierarchical clustering ("molecular subtypes") or supervised predictors of metastasis ("multi-gene prognosis signature").

EXAMPLE 3

Antibody-Based Method for Determining the Immune Signature of Tumor-Infiltrating Leukocytes in Ovarian Cancer Patients

[00082] This example describes methods for immunophenotyping tumor-infiltrating leukocytes, and provides results of immune signature analyses in a cohort of epithelial ovarian cancer (EOC) patients.

Materials and Methods

[00083] Patients and tumor samples. The tissue microarray (TMA) used in this study, described elsewhere in detail (Brennan et al., Eur J Cancer, 45:1510-1517, 2009), was constructed from a consecutive cohort of 76 patients diagnosed with primary invasive epithelial ovarian cancer at the National Maternity Hospital, Dublin, with a median follow-up of 4.3 years. The standard surgical management was a total abdominal hysterectomy, bilateral salpingo- oophorectomy and omentectomy with cytological evaluation of peritoneal fluid or washings. Residual disease was resected to less than 2 cm where possible. Stage and volume of residual disease (no residual disease, residual disease greater or less than 2 cm) were recorded in all cases. All patients received adjuvant chemotherapy consisting of cisplatin or carboplatin prior to 1992 and combined with paclitaxel from 1992 to 2002. No patient received neo-adjuvant chemotherapy. Benign or borderline ovarian cancers, non-epithelial ovarian cancer and cases with histological features typical of secondary ovarian cancer were excluded from the study. Diagnostic specimens were all formalin fixed and paraffin embedded in the Department of Pathology at the National Maternity Hospital, Dublin, Ireland. All tissue blocks were stored in that department prior to construction of the tissue microarray. Full ethical approval was obtained from the Ethics Committee of the National Maternity Hospital, Dublin and informed consent was obtained from living patients and relatives of deceased patients.

[00084] Tissue microarrays. Seventy six paraffin-embedded tumor specimens were used for tissue microarray construction as previously described (Brennan et al., 2009, supra). Areas representative of invasive cancer were marked on haematoxylin and eosin-stained slides and the tissue microarray was constructed using a manual tissue arrayer (MTA-I, Beecher Inc, WI). The array consisted of four cores per patient. Two 1.0 mm cores were extracted from each donor block and assembled in a recipient block. Recipient blocks were limited to approximately 100 cores each. In general, cores were taken from the peripheral part of the tumor in cases where the tumor had well-defined borders. In more diffusely growing tumors, areas with the highest tumor cell density were primarily targeted. Necrotic tissue was avoided.

[00085] Immunohistochemistry. Tissue microarray slide sections were prepared and immunohistochemistry was performed as described in Example 1.

[00086] Image acquisition and analysis. The Aperio ScanScope XT Slide Scanner (Aperio Technologies) system with a 2Ox objective was used to capture whole-slide digital images. Slides were de-arrayed to visualize individual cores, using Spectrum software (Aperio). A tumor nuclear algorithm (IHCMark) was developed in-house to quantify the density of DAB positive immune cells/mm (Rexhepaj et al., Breast Cancer Res., 10:R89, 2008).

[00087] Statistics. Decision tree analysis was performed on the tissue microarray data. For this purpose, all patients were randomly divided in 10 subsets. A decision tree model was selected using a 10-fold cross-validation approach. Ten consecutive decision tree models were independently constructed using the CD4, CD8 and CD68 continuous output from 9 subsets. The prognostic accuracy of each decision tree model was tested using the remaining set of patients, with the model displaying the highest accuracy being selected as the optimal model for the dataset. Kaplan-Meier analysis and the log-rank test were used to illustrate differences between recurrence free survival according to individual CD68, CD4, and CD8 expression. A Cox regression proportional hazards models was employed to estimate the relationship between recurrence free survival and the CD68/CD4/CD8 immune profile, disease stage, tumor grade, age at diagnosis and residual disease after primary surgery. A P value of <0.05 was considered statistically significant and all calculations were performed using SPSS version 12.0 (SPSS Inc, Chicago, IL).

[00088] Validation of Immune Signature in a Second Large Cohort. A validation cohort of 154 ovarian cancer cases collected from two prospective, population-based cohorts, the Malmo Diet and Cancer Study (MDCS) and Malmo Preventive Project (MPP) have now been screened. The MDCS was initiated in 1991 and enrolled 17,035 healthy women (Berglund et al., J Intern Med, 223:45-51, 1993). The MPP was established in 1974 for screening with regard to cardiovascular risk factors and enrolled 10,902 women (Berglund et al., J Intern Med, 239:489-497, 1996). Median follow-up for this cohort is 2.67 years (range 0 - 21.1 years) at which point 105 patients were dead, 98 from ovarian cancer. Results from these studies are expected to extend the predictive power of the CD68/CD4/CD8 signature to different histological subtypes.

Results

[00089] Single Immune Marker Analysis of Leukocyte Infiltrates. After finding a correlation between the survival of breast cancer patients and the CD68/CD4/CD8 signature of their tumor infiltrating leukocytes (Example 1), the applicability of this signature to epithelial ovarian cancer was assessed. First, individual CD4⁺, CD8⁺ and CD68⁺ leukocyte densities were analyzed by immunohistochemistry using a tissue microarray (TMA) consisting of tumor tissue representing 76 epithelial ovarian cancers (FIG. 3). After digital scanning of stained TMA slides using an Aperio ScanScope XT slide scanner, a fully automated nuclear algorithm was employed to quantify CD4⁺, CD8⁺ and CD68⁺ cells. For survival analysis, high and low thresholds for each marker were established by decision tree analysis with 10-fold cross- validation of tree models. Finally, Kaplan Myer analysis of recurrence free survival (RFS) demonstrated that as single variables high infiltration by CD68⁺ cells and low CD8⁺ cell density predict reduced RFS, while CD4⁺ cell density alone showed no statistical difference in overall survival (FIG. 3D-E).

[00090] Three-marker Profile of Leukocyte Infiltrates. Next, the predictive power of CD4, CD8, and CD68 was combined to stratify EOC patients for RFS. A classification and regression trees algorithm was used to define the signature. High and low thresholds for each marker were established through decision tree analysis with 10-fold cross-validation of tree models. All patients were categorized as having 1) CD68^hl/CD4^h7CD8^l0 or 2) CD68^lo/CD4^lo/CD8^hl. Kaplan Myer analysis of these two groups demonstrated significantly reduced RFS in patients bearing the CD68^hl/CD4^hl/CD8^l0 immunohistochemistry signature (p < 0.001; FIG. 4A). In addition, multivariate Cox regression analysis revealed the CD68/CD4/CD8 signature had an increased survival hazard ratio of 2.26 and no statistical correlation with disease grade, stage, age at diagnosis or post-operative residual disease (FIG. 4B), indicating that the immune signature predicts epithelial ovarian cancer survival independently of the typical clinical and histopathological markers currently employed.

[00091] Immune Signature Predicts Epithelial Ovarian Cancer Survival in Serous Subtype. Epithelial ovarian cancer is known to be a heterogeneous disease, the entities of which are in part reflected in traditional histopathological characteristics. Therefore, biomarkers were assessed separately in histological subgroups, as well as across the entire patient cohort.

Consequently, it is critical for the utility of the CD68/CD4/CD8 signature to predict the outcome in individual tumor histological subtypes of ovarian cancer (e.g., serous versus non-serous).

[00092] Kaplan Meier analysis demonstrated that the CD68/CD4/CD8 signature was associated with a reduced RFS in serous (n=53) tumors (p = 0.001) (FIG 5B), but not non-serous tumors (n =23) (FIG 5A). Analysis of the leukocyte infiltrate in serous compared to non-serous tumors further revealed that the mean CD68 cell density (p=0.037) was higher and the mean CD20 cell density (p=0.004) was lower in serous tumors. A non- significant trend towards a lower CD8 and higher CD4 cell density in serous tumors was also detected (FIG 6).

[00093] Various modifications and variations of the present disclosure will be apparent to those skilled in the art without departing from the scope and spirit of the disclosure. Although the disclosure has been described in connection with specific preferred embodiments, it should be understood that the disclosure as claimed should not be unduly limited to such specific embodiments. Indeed, various modifications of the described modes for carrying out the disclosure which are understood by those skilled in the art are intended to be within the scope of the claims.

Claims

CLAIMS We claim:

1. A method for assessing risk of poor clinical outcome for a human cancer patient, said method comprising:

a) subjecting a sample from a solid tumor of said patient to a procedure for quantitation of expression of leukocyte biomarkers comprising CD4, CD8 and CD68;

b) detecting the presence of an immune signature of poor outcome comprising

CD4^hl,CD8^lo,CD68^hl or an immune signature of favorable outcome comprising

CD4^lo,CD8^hl,CD68^l0 in said sample, wherein said immune signature of poor outcome is associated with an increased risk of poor clinical outcome as compared to said immune signature of favorable outcome.

2. The method of claim 1, wherein said poor clinical outcome comprises a relative reduction in one or more of overall survival, recurrence-free survival, and distant recurrence-free survival.

3. The method of claim 1, further comprising:

c) treating said patient with an aggressive treatment regimen when said immune signature of poor outcome is detected.

4. The method of claim 1, wherein said favorable clinical outcome comprises a relative increase in one or more of overall survival, recurrence-free survival, and distant recurrence-free survival.

5. The method of claim 1, further comprising:

c) treating said patient with a conservative treatment regimen when said immune signature of favorable outcome is detected.

6. The method of claim 1, further comprising:

a step before a) of obtaining said sample of said solid tumor from said patient.

7. The method of claim 1, further comprising determining subtype of the solid tumor.

8. The method of claim 1, wherein said solid tumor is breast cancer.

9. The method of claim 8, wherein said breast cancer is of a subtype selected from the group consisting of basal, luminal A, and triple negative.

10. The method of claim 1, wherein said solid tumor is epithelial ovarian cancer.

11. The method of claim 10, wherein said epithelial ovarian cancer is of a serous subtype.

12. A method for assessing risk of poor clinical outcome for a human cancer patient, said method comprising:

a) subjecting a breast cancer sample or an epithelial ovarian cancer sample from said patient to an antibody-based technique for quantitation of expression of leukocyte biomarkers comprising CD4, CD8 and CD68;

b) detecting the presence of an immune signature of poor outcome comprising

CD4^hl,CD8 °,CD68^hl or an immune signature of favorable outcome comprising

CD4^lo,CD8^hl,CD68^l0 in said sample, wherein said immune signature of poor outcome is associated with an increased risk of poor clinical outcome as compared to said immune signature of favorable outcome.

13. The method of claim 12, wherein said poor clinical outcome comprises a relative reduction in one or more of overall survival, recurrence-free survival, and distant recurrence-free survival.

14. The method of claim 12, further comprising:

c) treating said patient with an aggressive treatment regimen when said immune signature of poor outcome is detected.

15. The method of claim 12, wherein said favorable clinical outcome comprises a relative increase in one or more of overall survival, recurrence-free survival, and distant recurrence-free survival.

16. The method of claim 12, further comprising:

c) treating said patient with a conservative treatment regimen when said immune signature of favorable outcome is detected.

17. The method of claim 12, further comprising:

a step before a) of obtaining said sample from said patient.

18. The method of claim 12, wherein said sample is said breast cancer sample of a subtype selected from the group consisting of basal, luminal A, and triple negative.

19. The method of claim 12, wherein said sample is said epithelial ovarian cancer sample of a serous subtype.

20. The method of claim 12, wherein said antibody-based technique is

immunohistochemistry.

21. A kit for assessing risk of poor clinical outcome for a human cancer patient, said kit comprising biomarker- specific reagents consisting essentially of:

a) a CD4- specific reagent;

b) a CD8-specific reagent; and

c) a CD68-specific reagent.

22. The kit of claim 21, wherein said CD4-specific reagent, said CD8-specific reagent and said CD68-specific reagent are antibodies.