WO2015118353A1 - Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer - Google Patents

Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer Download PDF

Info

Publication number
WO2015118353A1
WO2015118353A1 PCT/GB2015/050352 GB2015050352W WO2015118353A1 WO 2015118353 A1 WO2015118353 A1 WO 2015118353A1 GB 2015050352 W GB2015050352 W GB 2015050352W WO 2015118353 A1 WO2015118353 A1 WO 2015118353A1
Authority
WO
WIPO (PCT)
Prior art keywords
sample
expression
biomarkers
subject
biomarker
Prior art date
Application number
PCT/GB2015/050352
Other languages
French (fr)
Inventor
Denis Paul Harkin
Richard Kennedy
Katherine E. Keating
Andrena MCCAVIGAN
Laura A HILL
Steve Deharo
Timothy Davison
Fionnuala Patterson
Sinead DONEGAN
Gera JELLEMA
Charlie GOURLEY
Original Assignee
Almac Diagnostics Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Almac Diagnostics Limited filed Critical Almac Diagnostics Limited
Priority to BR112016018044A priority Critical patent/BR112016018044A2/en
Priority to EP15705693.8A priority patent/EP3102700A1/en
Priority to CA2938807A priority patent/CA2938807A1/en
Priority to KR1020167024823A priority patent/KR20160117606A/en
Priority to AU2015213844A priority patent/AU2015213844A1/en
Priority to US15/116,641 priority patent/US10280468B2/en
Priority to CN201580016568.5A priority patent/CN106164296A/en
Priority to JP2016550631A priority patent/JP2017506506A/en
Publication of WO2015118353A1 publication Critical patent/WO2015118353A1/en

Links

Classifications

    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6876—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes
    • C12Q1/6883—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material
    • C12Q1/6886—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material for cancer
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61P—SPECIFIC THERAPEUTIC ACTIVITY OF CHEMICAL COMPOUNDS OR MEDICINAL PREPARATIONS
    • A61P35/00—Antineoplastic agents
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/106—Pharmacogenomics, i.e. genetic variability in individual responses to drugs and drug metabolism
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/118—Prognosis of disease development
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/158—Expression markers

Definitions

  • the present invention relates to a molecular diagnostic test useful for providing a prognosis and directing treatment of cancers from different anatomical sites.
  • the invention includes the derivation of a gene classification model from gene expression levels.
  • One application is the selection of whether to administer certain therapeutics, such as anti- angiogeneic therapeutics, to subjects receiving a standard of care cancer therapy.
  • Another application is the stratification of cancer patients into those that have a good clinical prognosis or poor clinical prognosis.
  • the present invention provides a test that can guide therapy selection as well as selecting patient groups for enrichment strategies during clinical trial evaluation of novel therapeutics.
  • the invention can be used as a prognostic indicator for certain cancers including ovarian cancer, breast cancer, colon, prostate, lung and glioblastoma.
  • the angiogenesis subtype can be identified from fresh/frozen (FF) or formalin fixed paraffin embedded FFPE patient samples.
  • Ovarian cancer is the leading cause of death among all gynecological cancers in western countries. This high death rate is due to the diagnosis at an advanced stage in most patients.
  • Epithelial ovarian cancer constitutes 90% of ovarian malignancies and is classified into distinct histologic categories including serous, mucinous, endometrioid, clear cell, transitional, mixed, and undifferentiated subtypes. There is increasing evidence that these differed histologies arise from different aetiologies.
  • microarrays and molecular genomics have the potential for a significant impact on the diagnostic capability and prognostic classification of disease, which may aid in the prediction of the response of an individual patient to a defined therapeutic regimen.
  • Microarrays provide for the analysis of large amounts of genetic information, thereby providing a genetic fingerprint of an individual. There is much enthusiasm that this technology will ultimately provide the necessary tools for custom-made drug treatment regimens.
  • Angiogenesis is a key component of neo-vascularisation of tumors and essential to tumorigenesis and metastatsis. As such, it is a key area for therapeutic intervention and has been correlated to poor prognosis and reduced survival. This has promoted the development of a number of agents that target angiogenesis related processes and pathways, including the market leader and first FDA-approved anti-angiogenic, bevacizumab (Avastin), produced by Genentech/Roche .
  • biomarker of response to bevacizumab would improve assessment of treatment outcomes and thus enable the identification of patient subgroups that would receive the most clinical benefit from bevacizumab treatment. This would be particularly relevant in the case of metastatic breast cancer, where the absence of a clinically beneficial biomarker has undermined the use of bevacizumab. Thus far, no such biomarker has been clinically validated to predict bevacizumab efficacy. Hypertension and VEGF polymorphisms are so far the only biomarkers to show potential, but important questions remain about their use in a clinical setting.
  • Another approach to anti-angiogenic therapy is simulataneous targeting of multiple angiogenic pathways rather than selective targeting of the VEGF pathway.
  • multitargeted anti-angiogenic agents should more completely inhibit angiogenesis than agents such as bevacizumab and thus may produce greater therapeutic benefit. It has been postulated that in some tumors, angiogenesis may require VEGF only in the early stages of disease but is driven by additional angiogenic pathways as the disease progresses. Therefore, by targeting multiple pathways, it may be possible to counteract compensatory escape mechanisms that could lead to resistance to VEGF inhibition.
  • biomarkers or a collection of biomarkers expressed in cancer that identify a subtype of cancer that is associated with an up- regulation in molecular signaling related to immune response and a down-regulation in molecular signaling related to angiogenesis and vasculature development, referred to herein as a "non-angiogenesis" or “immune” subtype.
  • the collection of biomarkers may be defined by an expression signature, and the expression signature is used to assign a cumulative score to the measured expression values of the collection of biomarkers.
  • biomarkers and expression signatures may form the basis of a single parameter or multiparametric predictive test that could be delivered using methods known in the art such as microarray, next generation sequencing (NGS), Q-PCR, immunohistochemistry, ELISA or other technologies that can quantify mRNA or protein expression.
  • NGS next generation sequencing
  • Q-PCR Q-PCR
  • immunohistochemistry immunohistochemistry
  • ELISA immunohistochemistry
  • the cancer subtypes described herein are common to many types of cancer and are not limited to a single cancer disease type. Accordingly, the expression signatures of the present invention are not limited to a single cancer type.
  • the non-angiogenesis expression signature comprises two or more biomarkers selected from the biomarkers listed in Tables 1. In another example embodiment, the non- angiogenesis expression signature comprises two or more biomarkers listed in Table 2 or 3. In certain other example embodiments, the expression signature comprises one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2PA, MFAP2, MATN3, or RTP4.
  • the expression signature comprises one or more of IGF2, CDR1, COL3A1, SPARC, TIMP3, INS, COL8A1, NUAK1, MATN3, and TMEM45A. In another example embodiment, the expression signature comprises one or more of INS, SPARC, COL8A1, COL3A1, CDR1, NUAK1, TIMP3, and MMP14.
  • the non-angiogenesis signature comprises the biomarkers listed in Table 2 and their corresponding weights as determined using a PLS classifier. In another example embodiment, the non-angiogenesis signature comprises the biomarkers listed in Table 3 and their corresponding ranks within a decision function.
  • the invention provides a method for selecting whether to administer an anti-angiogenic therapeutic agent to a subject, comprising: measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject; assessing from the expression level(s) of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature an anti-angiogenic therapeutic agent is contraindicated.
  • assessing whether the sample is positive or negative for the biomarker signature comprises: determining a sample expression score for the one or more biomarkers; comparing the sample expression score to a threshold score; and determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
  • the subject is receiving or has received treatment with a chemotherapeutic agent.
  • the invention provides a method of treating cancer comprising administering a chemotherapeutic agent and not administering an anti-angiogenic therapeutic agent to a subject wherein the subject is selected for treatment on the basis of a method as described herein and the subject is positive for the biomarker signature.
  • a method of treating cancer comprising administering a chemotherapeutic agent and not administering an anti-angiogenic therapeutic agent to a subject wherein the subject is selected for treatment by measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject; assessing from the expression levels of the biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is postive for the biomarker signature the subject is selected for treatment.
  • the present invention relates to a chemotherapeutic agent for use in treating cancer in a subject wherein the subject is selected for treatment on the basis of a method as described herein and is positive for the biomarker signature and wherein the subject is not treated with an anti-angiogenic therapeutic agent.
  • the invention also relates to a chemotherapeutic agent for use in treating cancer in a subject wherein the subject is selected for treatment by measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject; assessing from the expression levels of the biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature the subject is selected for treatment and wherein the subject is not treated with an anti-angiogenic therapeutic agent.
  • the present invention relates to a method of treating cancer comprising administering a chemotherapeutic agent to a subject wherein the subject is positive for a biomarker signature defined by the expression levels of one or more biomarkers selected from Table 2 or Table 3 and wherein an anti-angiogenic therapeutic agent is not administered.
  • the present invention relates to a chemotherapeutic agent for use in treating cancer in a subject wherein the subject is positive for a biomarker signature defined by the expression levels of one or more biomarkers selected from Table 2 or Table 3 and wherein the subject is not treated with an anti-angiogenic therapeutic agent.
  • the chemotherapeutic agent comprises a platinum-based chemotherapeutic agent, an alkylating agent, an anti-metabolite (such as 5FU), an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof.
  • the chemotherapeutic agent may comprise a platinum based-chemotherapeutic agent, a mitotic inhibitor, or a combination thereof.
  • the chemotherapeutic agent comprises carboplatin and/or paclitaxel.
  • the chemotherapeutic agent may reflect the standard of care treatment for the cancer. The standard of care treatment may differ for different types of cancer - for example carboplatin in ovarian cancer, 5FU in colorectal cancer, platinum in head and neck cancer.
  • assessing whether the sample is positive or negative for the biomarker signature may comprise determining a sample expression score for the one or more biomarkers; comparing the sample expression score to a threshold score; and determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
  • the subject may be suffering from cancer.
  • the cancer may be ovarian cancer, optionally high grade serous ovarian cancer.
  • administering an agent is used interchanging with “treating with” an agent.
  • a method for determing clinical prognosis of a subject with cancer comprising: measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject; assessing from the expression level(s) of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature the subject has a good prognosis.
  • Assessing whether the sample is positive or negative for the biomarker signature may comprise: determining a sample expression score for the biomarkers; comparing the sample expression score to a threshold score; and determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
  • the good prognosis indicates increased progression free survival or overall survival rates compared to samples that are negative for the biomarker signature, optionally compared to samples with a sample expression score below the threshold score.
  • the subject is receiving, has received and/or will receive chemotherapeutic treatment and/or will not receive treatment with an anti- angiogenic therapeutic agent.
  • the chemotherapeutic treatment may comprise administration of a platinum-based chemotherapeutic agent, an alkylating agent, an antimetabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof.
  • the chemotherapeutic treatment may comprise administration of a platinum-based chemotherapeutic agent, a mitotic inhibitor, or a combination thereof.
  • the chemotherapeutic treatment comprises administration of paclitaxel and carboplatin.
  • the cancer may be ovarian cancer or colorectal cancer
  • the present invention relates to a method for selecting whether to administer Bevacizumab to a subject, comprising: in a test sample obtained from a subject suffering from ovarian cancer, which subject is being, has been and/or will be treated using a platinum-based chemotherapeutic agent and/or a mitotic inhibitor; measuring expression levels of one, two or more, up to all of the, biomarkers selected from Table 2; assessing from the expression level(s) of the one, two or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, selecting a treatment based on whether the sample is positive for the biomarker signature, wherein is the sample is positive for the biomarker signature Bevacizumab is contraindicated.
  • assessing whether the sample is positive or negative for the biomarker signature comprises determining a sample expression score for the one, two or more biomarkers; comparing the sample expression score to a threshold score; and determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
  • the invention also relates to a method for determining clinical prognosis of a subject, comprising: (a) in a test sample obtained from a subject suffering from ovarian cancer, which subject is being, has been and/or will be treated using a platinum-based chemotherapeutic agent and/or a mitotic inhibitor; (b) measuring expression levels of one or more, up to all of the, biomarkers selected from Table 2 or Table 3; (c) assessing from the expression level(s) of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature the subject has a good prognosis.
  • Assessing whether the sample is positive or negative for the biomarker signature may comprise: (i) determining a sample expression score for the one or more biomarkers; (ii) comparing the sample expression score to a threshold score; and (iii) determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
  • the good prognosis indicates increased progression free survival or overall survival rates compared to samples that are negative for the biomarker signature, optionally compared to samples with a sample expression score below the threshold score.
  • the expression level(s) of two or more biomarkers selected from Table 2 or Table 3 may be measured in a sample from the subject.
  • a classification tree is built through a process called binary recursive partitioning, which is an iterative procedure of splitting the data into partitions/branches. The goal is to build a tree that distinguishes among pre-defined classes.
  • Each node in the tree corresponds to a variable. To choose the best split at a node, each variable is considered in turn, where every possible split is tried and considered, and the best split is the one which produces the largest decrease in diversity of the classification label within each partition.
  • a method for selecting whether to administer an anti-angiogenic therapeutic agent to a subject using the expression signatures disclosed herein comprising obtaining a test sample from the subject, measuring expression levels of a biomarker panel from the test sample, determining a sample expression score for the biomarker panel, comparing the sample expression score to a threshold score, and selecting a treatment based on whether the expression score is equal to or above the threshold score.
  • a sample expression score is equal to or above the threshold score indicates an anti-angiogenic agent is contraindicated and should not be administered to the subject.
  • a sample expression score below the threshold score indicates an anti-angiogenic agent is not contraindicated and can be administered to the subject.
  • a therapeutic agent is "contraindicated” or “detrimental” to a patient if the cancer's rate of growth is accelerated as a result of contact with the therapeutic agent, compared to its growth in the absence of contact with the therapeutic agent. Growth of a cancer can be measured in a variety of ways. For instance, the size of a tumor, or measuring the expression of tumor markers appropriate for that tumor type.
  • a therapeutic agent can also be considered “contraindicated” or “detrimental” if the patient' s overall prognosis (progression free survival and overall survival) is reduced by the administration of the therapeutic agent.
  • the expression signature disclosed herein may determine a patient' s clinical prognosis upon administration of an anti-angiogeneic agent following standard cancer therapy.
  • the subject suffers from cancer.
  • the cancer may include, but is not limited to, ovarian cancer, breast cancer, colon cancer, colorectal cancer, glioblastoma, kidney cancer, including renal cell carcinoma, heatocelluar cancer, thyroid cancer, pancreatic cancer, neuroendocrine cancer, esophageal cancer, gastrointestinal stromal tumors (GIST), gastric cancer, liver cancer, including adult primary liver cancer, lymphoma, melanoma, or multiple myeloma.
  • the cancer is ovarian cancer.
  • the ovarian cancer is high grade serous ovarian cancer.
  • the patient may have received, is receiving and/or will receive a treatment which may be a standard of care treatment for the cancer type of the subject.
  • that treatment which may be a standard of care treatment may include treatment with a chemotherapeutic agent.
  • the chemotherapeutic treatment may include administration of a platinum-based chemotherapeutic agent, an alkylating agent, an antimetabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof.
  • the chemotherapeutic treatment comprises administration of a platinum-based chemotherapeutic agent, a mitotic inhibitor, or a combination thereof. In certain other example embodiments, the chemotherapeutic treatment comprises administration of carboplatin and paclitaxel.
  • the subject has high grade serous ovarian cancer and has previously received a platinum-based chemotherapeutic agent and a mitotic inhibitor. In another example embodiment, the subject has high grade serous ovarian cancer and has previously received carboplatin and paclitaxel.
  • the anti-angiogenic therapeutic agent may be a VEGF-pathway-targeted therapeutic agent (such as bevacizumab or aflibercept), an angiopoietin-TIE2 pathway inhibitor, an endogenous angiogenic inhibitor, or an immunomodulatory agent.
  • the anti- angiogenic therapeutic agent is a VEGF-pathway-targeted therapeutic agent.
  • the anti-angiogenic therapeutic agent is bevacizumab.
  • a method for determining a clinical prognosis of a subject using the expression signatures disclosed herein comprising obtaining a test sample from the subject, measuring expression levels of a biomarker panel from the test sample, determining a sample expression score for the biomarker panel, comparing the sample expression score to the threshold expression score, wherein if the expression score is equal to or above the threshold expression score the clinical prognosis is a good prognosis.
  • a good prognosis indicates increased survival rates compared to a subject with an expression score below the threshold score.
  • the subject suffers from cancer.
  • the cancer may include, but is not limited to, ovarian cancer, breast cancer, colon cancer, colorectal cancer, or glioblastoma.
  • the cancer is ovarian cancer.
  • the ovarian cancer is high grade serous ovarian cancer.
  • the subject may receive, has received and/or will receive a chemotherapeutic treatment.
  • the chemotherapeutic treatment may include administration of a platinum-based chemotherapeutic agent, an alkylating agent, an anti-metabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof.
  • the chemotherapeutic treatment comprises administration of a platinum-based chemotherapeutic agent, a mitotic inhibitor, or a combination thereof. In certain other example embodiments, the chemotherapeutic treatment comprises administration of carboplatin and paclitaxel.
  • the subject has high grade serous ovarian cancer and may receive, has received and/or will receive a platinum-based chemotherapeutic agent and a mitotic inhibitor, such as taxane. In another example embodiment, the subject has high grade serous ovarian cancer and is may receive, has received, and/or will receive carboplatin and paclitaxel.
  • kits for conventional diagnostic uses listed above such as qPCR, NGS, microarray, and immunoassays such as immunohistochemistry, ELISA, Western blot and the like.
  • kits for conventional diagnostic uses listed above such as qPCR, NGS, microarray, and immunoassays such as immunohistochemistry, ELISA, Western blot and the like.
  • kits include appropriate reagents and directions to assay the expression of the genes or gene products and quantify rnPvNA or protein expression.
  • such methods may be used to identify patients that are sensitive to and respond to drugs that inhibit, either directly or indirectly, processes relating to angiogenesis.
  • such methods may be used to identify patients that are resistant to or do not respond or will respond in adverse fashion to drugs that inhibit, either directly or indirectly, processes relating to angiogenesis.
  • This invention also relates to guiding effective treatment of patients. Further, methods relating to selection of patient treatment regimens and selecting patients for clinical trials of current, or developmental stage drugs that directly or indirectly affect angiogenesis are provided.
  • FFPE formalin fixed paraffin-embedded
  • FF fresh/frozen
  • Figure 1 provides a heat map showing unsupervised hierarchical clustering of gene expression data in 265 high grade serous ovarian carcinomas. Each column represents the expression of these probe sets in one tumor. Probe set expression across all clusters is represented horizontally. The bar above the heat map is color-coded by cluster as described in the legend box. The second bar is color-coded for class label as described in the legend box. Functional processes corresponding to each probe set cluster are labeled to the right of the figure.
  • Figure 2 provides Kaplan-Meier analysis of overall survival by cluster from unsupervised analysis of gene expression in 265 high grade serous ovarian carcinomas.
  • Figure 3 provides Kaplan-Meier analysis of survival of the two classes defined by the 63 -gene signature classifier in the Edinburgh (discovery) dataset.
  • Proangiogenic group consists of Angio and Angioimmune subgroups.
  • Figure 4 provides Kaplan-Meier analysis of survival of the two classes defined by the 63-gene signature classifier in the Tothill (validation) dataset.
  • Proangiogenic group consists of Angio and Angioimmune subgroups.
  • Figure 5 provides Kaplan Meier curves for progression free survival in Immune
  • Figure 6 provides Kaplan Meier curves for overall survival in Immune ( Figure
  • Figure 7 provides Kaplan Meier curves for progression free survival (A) and overall survival (B) for carboplatin and paclitaxel treated ICON7 trail patients defined by the 63 gene signature.
  • Figure 8A and 8B are graphs demonstrating certain classification performance benchmarks of an example non-angiogenesis signature as applied to colorectal cancer samples.
  • Figure 9 Signature development: AUC of training set under CV.
  • Figure 10 Signature development: C-Index of training set under CV.
  • Figure 11 Signature development: HR of training set under CV.
  • Figure 12 Signature development: HR of ICON7 SOC samples under CV.
  • Figure 13 Signature development: C-Index of ICON7 SOC samples under CV.
  • Figure 14 Signature development: HR of ICON7 Immune samples under CV.
  • Figure 15 Signature development: HR of ICON7 ProAngio samples under CV.
  • Figure 16 Core set analysis: lmmune63GeneSig_CoreGenes_InternalVal.png.
  • Figure 17 Core set analysis: Immune63GeneSig_CoreGenes_Tothill.png.
  • Figure 18 Core set analysis: Immune63GeneSig_CoreGenes_ICON7_SOC.png.
  • Figure 22 Kaplan Meier to show the differences in progression free survival probability between the samples predicted as Angio-Off (inactive) versus those that are predicted as Angio-On (active) by the 63 gene signature
  • Figure 23 Semi-supervised hierarchical clustering of the 529 CRC samples published by Marrisa et al (2013) using the angiogenesis defining gene list.
  • Figure 24 ROC Curve showing the discrimination in 63 gene signature scores between the angiogenesis active subtype and angiogenesis inactive subtype in the Marissa CRC data.
  • Figure 25 Kaplan Meier Curve showing the survival differences between the angiogenesis active and angiogenesis inactive patients (treated only) as predicted by the 63 gene signature in the GSE14333 CRC data.
  • biomarker panel As used herein terms "biomarker panel,” “expression classifier,” “classifier,”
  • the panel typically includes a plurality of biomarkers but may include only a single biomarker where that biomarker is useful individually in the methods of the invention.
  • a major goal of current research efforts in cancer is to increase the efficacy of perioperative systemic therapy in patients by incorporating molecular parameters into clinical therapeutic decisions.
  • Pharmacogenetics/genomics is the study of genetic/genomic factors involved in an individuals' response to a foreign compound or drug. Agents or modulators which have a stimulatory or inhibitory effect on expression of a biomarker of the invention can be administered to individuals to treat (prophylactically or therapeutically) cancer in the patient. It is ideal to also consider the pharmacogenomics of the individual in conjunction with such treatment. Differences in metabolism of therapeutics may possibly lead to severe toxicity or therapeutic failure by altering the relationship between dose and blood concentration of the pharmacologically active drug.
  • understanding the pharmacogenomics of an individual permits the selection of effective agents (e.g., drugs) for prophylactic or therapeutic treatments.
  • Such pharmacogenomics can further be used to determine appropriate dosages and therapeutic regimens.
  • the level of expression of a biomarker of the invention in an individual can be determined to thereby select appropriate agent(s) for therapeutic or prophylactic treatment of the individual.
  • the present invention relates to a molecular diagnostic tests useful for diagnosing cancers from different anatomical sites that includes the use of one or more common subtypes related to angiogenesis.
  • the invention includes expression signatures that identify a subject as having a good or poor clinical prognosis, and expression signatures that indicate whether to administer an anti-angiogenic therapeutic agent to a subject.
  • the expression signature is derived by obtaining the expression profiles of samples from a sample set of known pathology and/or clinical outcome. The samples may originate from the same sample tissue type or different tissue types.
  • an "expression profile" comprises a set of values representing the expression level for each biomarker analyzed from a given sample.
  • the expression profiles from the sample set are then analyzed using a mathematical model.
  • Different mathematical models may be applied and include, but are not limited to, models from the fields of pattern recognition (Duda et al. Pattern Classification, 2 nd ed., John Wiley, New York 2001), machine learning (Scholkopf et al. Learning with Kernels, ⁇ Press, Cambridge 2002, Bishop, Neural Networks for Pattern Recognition, Clarendon Press, Oxford 1995), statistics (Hastie et al. The Elements of Statistical Learning, Springer, New York 2001), bioinformatics (Dudoit et al, 2002, J. Am. Statist. Assoc. 97:77-87, Tibshirani et al, 2002, Proc. Natl.
  • the mathematical model identifies one or more biomarkers expressed in the sample set that are most predictive of a given disease phenotype. These one or more biomarkers define an expression signature. Accordingly, an expression signature includes the biomarkers identified as most predictive of a given disease phenotype.
  • the mathematical model defines a variable, such as a weight, for each identified biomarker.
  • the mathematical model defines a decision function.
  • the decision function may further define a threshold score which separates the sample set into two disease phenotypes such as, but not limited to, samples that have a good and poor clinical prognosis.
  • the decision function and expression signature are defined using a linear classifier.
  • the biomarkers defined by the expression signature also referred to as a biomarker panel
  • the new sample biomarker panel expression profile is analyzed with the same mathematical model used to define the expression signature.
  • the biomarker panel may comprise one or more of the biomarkers defined by the expression signature.
  • the biomarker panel may comprise one or more of the biomarkers defined by the expression signature.
  • the biomarker panel comprises one or more of the biomarkers listed in Table 2.
  • the biomarker panel comprises all of the biomarkers listed in Table 2.
  • the biomarker panel comprises one or more of the biomarkers listed in Table 2.
  • the biomarker panel comprises all of the biomarkers listed in Table 2.
  • the mathematical model defines an expression score for the new sample.
  • the expression score may be determined by combining the expression values of the biomarkers with corresponding scalar weights using non-linear, algebraic, trigonometric or correlative means to derive a single scalar value.
  • the expression score is compared to the threshold score and the sample classified based on whether the expression score is greater than, or equal to, or less than the threshold score.
  • a sample expression score equal to or greater than the threshold score indicates a subject has a good clinical prognosis, and a sample expression score below the threshold score indicates a subject has a poor clinical prognosis.
  • a sample expression score equal to or greater than the threshold score indicates a subject has the signature. This may indicate a good clinical prognosis.
  • a sample expression score below the threshold score indicates a subject does not have the signature. This may indicate a poor clinical prognosis.
  • One application of the expression signatures disclosed herein is the identification of patients with a good and poor clinical prognosis.
  • the good or poor prognosis may be determined in the context of a certain treatment background (such as carboplatin/paclitaxel therapy as discussed herein).
  • the subject may be receiving or have received a standard chemotherapeutic treatment for the subject's cancer type.
  • the expression signatures disclosed herein may also be used to determine whether an additional therapeutic agent, such as an anti-angiogenic therapeutic agent, should be administered to the patient.
  • an additional therapeutic agent such as an anti-angiogenic therapeutic agent
  • the present invention relates to prediction of clinical prognosis using at least progression free survival or overall survival rates. Accordingly, a "good prognosis" indicates a subject population with a cancer subtype that demonstrates an increased survival rate compared to other cancer subtypes, whereas a “poor prognosis” or “bad prognosis” indicates a subject population with a cancer subtype that demonstrates decreased survival rate compared to other cancer subtypes.
  • a subject with an expression score equal to or above the threshold score is classified as having the non-angiogenesis subtype.
  • a subject with a sample expression score above the threshold score is classified as having a good clinical prognosis.
  • a subject with a sample expression score above or equal to the threshold score indicates the subject will likely experience a detrimental effect, or have a poorer clinical prognosis, if administered an anti-angiogenic therapeutic agent.
  • the determination of a subject's clinical prognosis or selection of an additional therapeutic agent may be made in the context of past, current, or planned chemotherapeutic treament.
  • the subject may set to start, be currently receiving, or have just completed, a standard of care chemotherapeutic treatment for the cancer type of the subject.
  • the chemotherapeutic treament may include administration of an alkylating agent, an anti-metabolite, a platinum-based drug, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, a corticosteroid, a hormone based therapeutic, or a combination thereof.
  • Example alkylating agents include nitrogen mustards, nitrosureas, alkyl sulfonates, triazines, and ethylenimines.
  • Example platinum drugs include cisplatin, carboplatin, and oxalaplatin.
  • Example anti-metabolites include 5-fluoruracil, 6-mercaptopurine, capecitabine, cladribine, clofarabine, cytarabine, floxuridine, fludarabine, gemcitabine, hydroxyurea, methotrexate, pemetrexed, pentostatin, and thioguanine.
  • Example anti-tumor antibiotics include daunorubicin, doxorubicin, epirubicin, idarubicin, actinomocyin- D, bleomycin, mitomycin-C, and mitoxantrone.
  • Example topoisomerase inhibtors include topotecan, irinotecan, etoposide, and teniposide.
  • Example mitotic inhibitors include taxanes, epothilones, vinca alkaloids, and estramustine.
  • Example corticosteroids include predisone, mefhylprednisolone, and dexamethasone.
  • the chemotherapy may include treatment with L-asparaginase, imatinib, gefitinib, sunitinib, bortezomib, retinoids, tretinoin, bexaroten, arsenic trioxide, fluvestrant, tamoxifen, toremifene, anastrozole, exemestane, letrozole, progestins, estrogens, bicalutamide, flutamide, nilutamide, gonadotropoin-releasing hormone agonists or analogs, rituximab, alemtuzumab, BCG, interleukin-2, interferon- alf a, thalidomide, and lenalidomide.
  • the chemotherapeutic treament may comprise a cyclosphoshamide, methotrexate, and fluorouracil (CMF) treatment regimen, a cyclophosphamide, doxorubicin, and fluorouracil (CAF) treatment regimen, an epirubicin and cyclophosphamide (EC) treatment regimen, a fluorouracil, epirubicin, and cyclophosphamide (FEC) treatment regimen, a paclitaxel and cyclophosphamide treatment regimen, a paclitaxel and carboplatin treatment regiment, a doxorubicin and cyclophosphamide treatment regiment, or a doxorubicin and paclitaxel treatment regimen.
  • the neoadjuvant cancer therapy comprises a platinum based chemotherapy treatment regimen.
  • the platinum-based chemotherapy treatement regimen comprises paclitaxel and carboplatin
  • Another application of the expression signatures disclosed herein is the stratification of response to, and selection of patients for therapeutic drug classes that encompass anti-angiogenic therapies.
  • By examining the expression of a collection of the identified biomarkers in a tumor it is possible to determine which therapeutic agent or combination of agents will be most likely to reduce the growth rate of a cancer. It is also possible to determine which therapeutic agent or combination of agents will be the least likely to reduce the growth rate of a cancer and/or which may cause adverse affects and thus be contra-indicated.
  • By examining the expression of a collection of biomarkers it is therefore possible to eliminate ineffective or inappropriate therapeutic agents.
  • these determinations can be made on a patient-by-patient basis or on an agent-by- agent basis.
  • the present invention provides a test that can guide therapy selection as well as selecting patient groups for enrichment strategies during clinical trial evaluation of novel therapeutics. For example, when evaluating a putative anti-angiogeneic agent or treatment regime, the expression signatures and methods disclosed herein may be used to select individuals for clinical trials that have cancer subtypes that are responsive to anti-angiogenic agents.
  • a cancer is "responsive" to a therapeutic agent if its rate of growth is inhibited as a result of contact with the therapeutic agent, compared to its growth in the absence of contact with the therapeutic agent.
  • Growth of a cancer can be measured in a variety of ways. For instance, the size of a tumor or measuring the expression of tumor markers appropriate for that tumor type.
  • the expression signature disclosed herein may determine a patient's clinical prognosis upon administration of an anti-angiogeneic agent following standard of care chemotherapeutic therapy for the cancer type of the patient.
  • a cancer is "non-responsive" to a therapeutic agent if its rate of growth is not inhibited, or inhibited to a very low degree, as a result of contact with the therapeutic agent when compared to its growth in the absence of contact with the therapeutic agent.
  • growth of a cancer can be measured in a variety of ways, for instance, the size of a tumor or measuring the expression of tumor markers appropriate for that tumor type.
  • the quality of being non-responsive to a therapeutic agent is a highly variable one, with different cancers exhibiting different levels of "non-responsiveness" to a given therapeutic agent, under different conditions. Still further, measures of non-responsiveness can be assessed using additional criteria beyond growth size of a tumor such as, but not limited to, patient quality of life, and degree of metastases.
  • the angiogenesis subtype can be identified from a fresh/frozen (FF) or formalin fixed paraffin embedded (FFPE) patient sample.
  • the cancer type is ovarian cancer, breast cancer, colon cancer, colorectal cancer, lung cancer, prostate cancer, or glioblastoma.
  • the cancer type is an ovarian cancer.
  • the cancer type is breast cancer.
  • the cancer type is lung cancer.
  • the cancer type is colon cancer.
  • the cancer type is prostate cancer.
  • the cancer type is glioblastoma.
  • the expression signatures of the present invention are identified by analyzing the expression profiles of certain biomarkers in a patient sample set.
  • Biomarkers suitable for use in the present invention include DNA, RNA, and proteins.
  • biomarkers suitable for use in the present invention include RNA and cDNA. The biomarkers are isolated from a patient sample and their expression levels determined to derive a set of expression profiles for each sample analyzed in the patient sample set.
  • the expression signature identifies a non-angiogenesis phenotype observed in cancer tissues, identified as a signature score for a combination of biomarkers above or equal to a threshold, the phenotype characterized by an up-regulation of immune response related genes and a down-regulation of genes associated with angiogeneisis or vasculature development related processes.
  • the expression profile obtained is a genomic or nucleic acid expression profile, where the amount or level of one or more nucleic acids in the sample is determined.
  • the sample that is assayed to generate the expression profile employed in the diagnostic or prognostic methods is a nucleic acid sample.
  • the nucleic acid sample includes a population of nucleic acids that includes the expression information of the phenotype determinative biomarkers of the cell or tissue being analyzed.
  • the nucleic acid may include RNA or DNA nucleic acids, e.g., mRNA, cRNA, cDNA etc., so long as the sample retains the expression information of the host cell or tissue from which it is obtained.
  • the sample may be prepared in a number of different ways, as is known in the art, e.g., by mRNA isolation from a cell, where the isolated mRNA is used as isolated, amplified, or employed to prepare cDNA, cRNA, etc., as is known in the field of differential gene expression. Accordingly, determining the level of mRNA in a sample includes preparing cDNA or cRNA from the mRNA and subsequently measuring the cDNA or cRNA.
  • the sample is typically prepared from a cell or tissue harvested from a subject in need of treatment, e.g., via biopsy of tissue, using standard protocols, where cell types or tissues from which such nucleic acids may be generated include any tissue in which the expression pattern of the to be determined phenotype exists, including, but not limited to, disease cells or tissue, body fluids, etc.
  • the expression profile may be generated from the initial nucleic acid sample using any convenient protocol. While a variety of different manners of generating expression profiles are known, such as those employed in the field of differential gene expression/biomarker analysis, one representative and convenient type of protocol for generating expression profiles is array-based gene expression profile generation protocols. Such applications are hybridization assays in which a nucleic acid that displays "probe" nucleic acids for each of the genes to be assayed/profiled in the profile to be generated is employed.
  • a sample of target nucleic acids is first prepared from the initial nucleic acid sample being assayed, where preparation may include labeling of the target nucleic acids with a label, e.g., a member of a signal producing system.
  • a label e.g., a member of a signal producing system.
  • the sample is contacted with the array under hybridization conditions, whereby complexes are formed between target nucleic acids that are complementary to probe sequences attached to the array surface.
  • the presence of hybridized complexes is then detected, either qualitatively or quantitatively.
  • Specific hybridization technology which may be practiced to generate the expression profiles employed in the subject methods includes the technology described in U.S. Pat. Nos.
  • hybridization conditions e.g., stringent hybridization conditions as described above, and unbound nucleic acid is then removed.
  • the resultant pattern of hybridized nucleic acids provides information regarding expression for each of the biomarkers that have been probed, where the expression information is in terms of whether or not the gene is expressed and, typically, at what level, where the expression data, i.e., expression profile, may be both qualitative and quantitative.
  • the patient sample comprises a cancer tissue samples, such as an archived sample.
  • the patient sample is preferably derived from cancer tissue and may be from a sample having been characterized by prognosis, likelihood of recurrence, long term survival, clinical outcome, treatment response, diagnosis, cancer classification, or personalized genomics profile.
  • cancer includes, but is not limited to, leukemia, brain cancer, prostate cancer, liver cancer, ovarian cancer, stomach cancer, colorectal cancer, throat cancer, breast cancer, skin cancer, melanoma, lung cancer, sarcoma, cervical cancer, testicular cancer, bladder cancer, endocrine cancer, endometrial cancer, esophageal cancer, glioma, lymphoma, neuroblastoma, osteosarcoma, pancreatic cancer, pituitary cancer, renal cancer, and the like.
  • colorectal cancer encompasses cancers that may involve cancer in tissues of both the rectum and other portions of the colon as well as cancers that may be individually classified as either colon cancer or rectal cancer.
  • the methods described herein refer to cancers that are treated with anti-angiogenic agents, anti-angiogenic targeted therapies, inhibitors of angiogenesis signaling, but not limited to these classes. These cancers also include subclasses and subtypes of these cancers at various stages of pathogenesis.
  • the patient sample comprises an ovarian cancer sample.
  • the ovarian cancer sample is a serous ovarian cancer sample such as a high grade serous ovarian cancer sample.
  • the patient sample comprises a breast cancer sample.
  • the patient sample comprises a glioblastoma sample.
  • Biological sample “sample”, and “test sample” are used interchangeably herein to refer to any material, biological fluid, tissue, or cell obtained or otherwise derived from an individual.
  • a blood sample can be fractionated into serum or into fractions containing particular types of blood cells, such as red blood cells or white blood cells (leukocytes).
  • a sample can be a combination of samples from an individual, such as a combination of a tissue and fluid sample.
  • biological sample also includes materials containing homogenized solid material, such as from a stool sample, a tissue sample, or a tissue biopsy, for example.
  • biological sample also includes materials derived from a tissue culture or a cell culture, including tissue resection and biopsy samples.
  • any suitable methods for obtaining a biological sample can be employed; example methods include, e.g., phlebotomy, swab (e.g., buccal swab), and a fine needle aspirate biopsy procedure. Samples can also be collected, e.g., by micro dissection (e.g., laser capture micro dissection (LCM) or laser micro dissection (LMD)), bladder wash, smear (e.g., a PAP smear), or ductal lavage.
  • a "biological sample” obtained or derived from an individual includes any such sample that has been processed in any suitable manner after being obtained from the individual, for example, fresh frozen or formalin fixed and/or paraffin embedded.
  • the methods of the invention as defined herein may begin with an obtained sample and thus do not necessarily incorporate the step of obtaining the sample from the patient. The methods may be in vitro methods performed on an isolated sample.
  • the term "patient” includes human and non-human animals.
  • the preferred patient for treatment is a human.
  • "Patient,” “individual” and “subject” are used interchangeably herein.
  • biomarker can refer to a gene, an mRNA, cDNA, an antisense transcript, a miRNA, a polypeptide, a protein, a protein fragment, or any other nucleic acid sequence or polypeptide sequence that indicates either gene expression levels or protein production levels.
  • a biomarker indicates or is a sign of an abnormal process, disease or other condition in an individual, that biomarker is generally described as being either over- expressed or under-expressed as compared to an expression level or value of the biomarker that indicates or is a sign of a normal process, an absence of a disease or other condition in an individual.
  • Up-regulation “up-regulated”, “over-expression”, “over-expressed”, and any variations thereof are used interchangeably to refer to a value or level of a biomarker in a biological sample that is greater than a value or level (or range of values or levels) of the biomarker that is typically detected in similar biological samples from healthy or normal individuals.
  • the terms may also refer to a value or level of a biomarker in a biological sample that is greater than a value or level (or range of values or levels) of the biomarker that may be detected at a different stage of a particular disease.
  • Down-regulation Down-regulated
  • under-expression under-expressed
  • any variations thereof are used interchangeably to refer to a value or level of a biomarker in a biological sample that is less than a value or level (or range of values or levels) of the biomarker that is typically detected in similar biological samples from healthy or normal individuals.
  • the terms may also refer to a value or level of a biomarker in a biological sample that is less than a value or level (or range of values or levels) of the biomarker that may be detected at a different stage of a particular disease.
  • a biomarker that is either over-expressed or under-expressed can also be referred to as being “differentially expressed” or as having a “differential level” or “differential value” as compared to a "normal” expression level or value of the biomarker that indicates or is a sign of a normal process or an absence of a disease, disease subtype, or other condition in an individual.
  • "differential expression” of a biomarker can also be referred to as a variation from a "normal” expression level of the biomarker.
  • differential biomarker expression and “differential expression” are used interchangeably to refer to a biomarker whose expression is activated to a higher or lower level in a subject suffering from a specific disease, relative to its expression in a normal subject, or relative to its expression in a patient that responds differently to a particular therapy or has a different prognosis.
  • the terms also include biomarkers whose expression is activated to a higher or lower level at different stages of the same disease. It is also understood that a differentially expressed biomarker may be either activated or inhibited at the nucleic acid level or protein level, or may be subject to alternative splicing to result in a different polypeptide product.
  • Differential biomarker expression may include a comparison of expression between two or more genes or their gene products; or a comparison of the ratios of the expression between two or more genes or their gene products; or even a comparison of two differently processed products of the same gene, which differ between normal subjects and subjects suffering from a disease; or between various stages of the same disease.
  • Differential expression includes both quantitative, as well as qualitative, differences in the temporal or cellular expression pattern in a biomarker among, for example, normal and diseased cells, or among cells which have undergone different disease events or disease stages.
  • the biomarker is an RNA transcript.
  • RNA transcript refers to both coding and non-coding RNA, including messenger RNAs (mRNA), alternatively spliced mRNAs, ribosomal RNA (rRNA), transfer RNA (tRNA), small nuclear RNAs (snRNA), and antisense RNA. Measuring mRNA in a biological sample may be used as a surrogate for detection of the level of the corresponding protein and gene in the biological sample. Thus, any of the biomarkers or biomarker panels described herein can also be detected by detecting the appropriate RNA.
  • Methods of biomarker expression profiling include, but are not limited to quantitative PCR, NGS, northern blots, southern blots, microarrays, SAGE, immunoassays (ELISA, EIA, agglutination, nephelometry, turbidimetry, Western blot, immunoprecipitation, immunocytochemistry, flow cytometry, Luminex assay), and mass spectrometry.
  • the overall expression data for a given sample may be normalized using methods known to those skilled in the art in order to correct for differing amounts of starting material, varying efficiencies of the extraction and amplification reactions.
  • biomarkers useful for distinguishing between cancer subtypes that demonstrate a good clinical prognosis and a poor clinical prognosis can be determined by identifying biomarkers exhibiting the highest degree of variability between samples in the patient data set as determined using the expression detection methods and patient sample sets discussed above.
  • Standard statistical methods known in the art for identifying highly variable data points in expression data may be used to identify the highly variable biomarkers.
  • the significance threshold is 6.3 ⁇ 10 "5 . In another example embodiment, the significance threshold may be between 1.0 ⁇ 10 "7 to 1.0 ⁇ 10 "3 .
  • the highly variable biomarkers may be further analyzed to group samples in the patient data set into subtypes or clusters based on similar gene expression profiles.
  • biomarkers may be clustered based on how highly correlated the up-regulation or down-regulation of their expression is to one another.
  • clustering analysis techniques known in the art may be used.
  • hierarchical agglomerative clustering is used to identify the cancer subtypes.
  • the biomarkers within each cluster may be further mapped to their corresponding genes and annotated by cross-reference to one or more gene ontology databases containing information on biological activity and biological pathways associated with the gene.
  • biomarkers in clusters that are up regulated and enriched for immune response general functional terms are grouped into a putative non-angiongenesis sample group and used for expression signature generation.
  • biomarkers in clusters that are down regulated and enriched for angiogenesis and vasculature development and are up regulated and enriched for immune response general functional terms are grouped into a putative non-angiongenesis sample group and used for expression signature generation. Further details for conducting functional analysis of biomarker clusters is provided in the Examples section below.
  • the biomarkers useful in deriving an expression signature for use in the present invention are those biomarkers listed in Table 1. These biomarkers are identified as having predictive value to determine a patient response to a therapeutic agent and/or a prognostice value in identifying individuals with a good or poor clinical prognosis.
  • the expression of the biomarkers disclosed herein correlates with whether a patient will experience a detrimental or beneficial effect from administration of an anti-angiogenic therapeutic agent. By examining the expression of a collection of biomarkers, it is therefore possible to eliminate ineffective or inappropriate therapeutic agents. Importantly, in certain embodiments, these determinations can be made on a patient-by-patient basis or on an agent-by-agent basis. Thus, one can determine whether or not a particular therapeutic regimen is likely to benefit a particular patient or type of patient, and/or whether a particular regimen should be continued.
  • the expression of the biomarkers disclosed herein correlate with a patient's overall clinical prognosis.
  • a patient's overall clinical prognosis By examinaing the expression of a collection of biomarkers identified in a tumor, it is possible to determine whether the individual has a cancer subtype associated with good clinical prognosis or poor clinical prognosis.
  • these determinations can be made on a patient-by-patient basis.
  • one of ordinary skill in the art can use predicted prognosis to help select appropriate treatment regimens to treat the underlying disease while eliminating those treatment regimens most likely to produce undesired or medically unwarranted adverse side effects.
  • SEQ ID NOs listed in Table 1 refer to probe set identifiers used to measure the expression levels of the genes on an example transcriptome array.
  • Expression signatures of the present invention have been cross-validated using expression data from different arrays with differnt probe sets as detailed further in the Examples section below. Accordingly, the expression signatures and methods disclosed herein are not limited to expression values measured using the probe sets disclosed herein.
  • all or a portion of the biomarkers recited in Table 1 may be used in an expression signature.
  • expression signatures comprising the biomarkers in Table 1 can be generated using the methods provided herein and can comprise between one, and all of the markers set forth in Tables 1 and each and every combination in between (e.g., four selected markers, 16 selected markers, 74 selected markers, etc.).
  • the expression signature comprises at least 5, 10, 20, 40, 60, 100, 150, 200, or 300 or more markers.
  • the predictive biomarker panel comprises no more than 5, 10, 20, 40, 60, 100, 150, 200, 300, 400, 500, 600 or 700 markers.
  • the expression signature includes a plurality of markers listed in Table 1.
  • the expression signature includes at least about 1%, about 5%, about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80%, about 90%, about 95%, about 96%, about 97%, about 98%, or about 99% of the markers listed in Table 1.
  • Selected expression signatures can be assembled from the biomarkers provided using methods described herein and analogous methods known in the art.
  • the expression signature contains all genes or gene products in Table 1.
  • the following methods may be used to derive expression signatures for distinguishing between subjects that are responsive or non-responsive to anti-angiogenic therapeutics, or as prognostic indicators of certain cancer types, including expression signatures derived from the biomarkers disclosed above.
  • the expression signature is derived using a decision tree (Hastie et al. The Elements of Statistical Learning, Springer, New York 2001), a random forest (Breiman, 2001 Random Forests, Machine Learning 45:5), a neural network (Bishop, Neural Networks for Pattern Recognition, Clarendon Press, Oxford 1995), discriminant analysis (Duda et al.
  • Pattern Classification 2nd ed., John Wiley, New York 2001
  • PAM Prediction Analysis for Microarrays
  • SIMCA Soft Independent Modeling of Class Analogy analysis.
  • Biomarker expression values may be defined in combination with corresponding scalar weights on the real scale with varying magnitude, which are further combined through linear or non-linear, algebraic, trigonometric or correlative means into a single scalar value via an algebraic, statistical learning, Bayesian, regression, or similar algorithms which together with a mathematically derived decision function on the scalar value provide a predictive model by which expression profiles from samples may be resolved into discrete classes of responder or non-responder, resistant or non-resistant, to a specified drug, drug class, molecular subtype, or treatment regimen.
  • Such predictive models are developed by learning weights and the decision threshold, optimized for sensitivity, specificity, negative and positive predictive values, hazard ratio or any combination thereof, under cross-validation, bootstrapping or similar sampling techniques, from a set of representative expression profiles from historical patient samples with known drug response and/or resistance.
  • the biomarkers are used to form a weighted sum of their signals, where individual weights can be positive or negative.
  • the resulting sum (“expression score") is compared with a pre-determined reference point or value. The comparison with the reference point or value may be used to diagnose, or predict a clinical condition or outcome.
  • biomarkers included in the classifier provided in Tables 1 will carry unequal weights in a classifier for determining clinical prognosis. Therefore, while as few as one biomarker may be used to diagnose or predict an clinical prognosis or response to a therapeutic agent, the specificity and sensitivity or diagnosis or prediction accuracy may increase using more biomarkers.
  • weight refers to the absolute magnitude of an item in a statistical calculation.
  • the weight of each biomarker in a gene expression classifier may be determined on a data set of patient samples using learning methods known in the art.
  • bias or “offset” refers to a constant term derived using the mean expression of the signatures genes in a training set and is used to mean-center the each gene analyzed in the test dataset.
  • the expression signature is defined by a decision function.
  • a decision function is a set of weighted expression values derived using a linear classifier. All linear classifiers define the decision function using the following equation:
  • All measurement values such as the microarray gene expression intensities x for a certain sample are collected in a vector x. Each intensity is then multiplied with a corresponding weight w, to obtain the value of the decision function f(x) after adding an offset term b.
  • the linear classifier will further define a threshold value that splits the gene expression data space into two disjoint sections.
  • Example linear classifiers include but are not limited to partial least squares (PLS), (Nguyen et al, Bioinformatics 18 (2002) 39-50), support vector machines (SVM) (Scholkopf et al, Learning with Kernels, ⁇ Press, Cambridge 2002), and shrinkage discriminant analysis (SDA) (Ahdesmaki et al., Annals of applied statistics 4, 503-519 (2010)).
  • PLS partial least squares
  • SVM support vector machines
  • SDA shrinkage discriminant analysis
  • the linear classifier is a PLS linear classifier.
  • the decision function is empirically derived on a large set of training samples, for example from patients showing a good or poor clinical prognosis.
  • the threshold separates a patient group based on different characteristics such as, but not limited to, clinical prognosis before or after a given therapeutic treatment.
  • the interpretation of this quantity, i.e. the cut-off threshold is derived in the development phase ("training") from a set of patients with known outcome.
  • the corresponding weights and the responsiveness/resistance cut-off threshold for the decision score are fixed a priori from training data by methods known to those skilled in the art.
  • Partial Least Squares Discriminant Analysis (PLS-DA) is used for determining the weights. (L. Stahle, S. Wold, J. Chemom. 1 (1987) 185- 196; D. V. Nguyen, D.M. Rocke, Bioinformatics 18 (2002) 39-50).
  • the data space i.e. the set of all possible combinations of biomarker expression values
  • the data space is split into two mutually exclusive groups corresponding to different clinical classifications or predictions, for example, one corresponding to good clinical prognosis and poor clinical prognosis.
  • relative over- expression of a certain biomarker can either increase the decision score (positive weight) or reduce it (negative weight) and thus contribute to an overall decision of, for example, a good clinical prognosis.
  • the data is transformed non- linearly before applying a weighted sum as described above.
  • This non-linear transformation might include increasing the dimensionality of the data.
  • the non-linear transformation and weighted summation might also be performed implicitly, for example, through the use of a kernel function. (Scholkopf et al. Learning with Kernels, MIT Press, Cambridge 2002).
  • the patient training set data is derived by isolated RNA from a corresponding cancer tissue sample set and determining expression values by hybridizing the isolated RNA to a microarray.
  • the microarray used in deriving the expression signature is a transcriptome array.
  • a "transcriptome array" refers to a microarray containing probe sets that are designed to hybridize to sequences that have been verified as expressed in the diseased tissue of interest. Given alternative splicing and variable poly- A tail processing between tissues and biological contexts, it is possible that probes designed against the same gene sequence derived from another tissue source or biological context will not effectively bind to transcripts expressed in the diseased tissue of interest, leading to a loss of potentially relevant biological information.
  • telomere sequences are expressed in the disease tissue of interest before deriving a microarray probe set. Verification of expressed sequences in a particular disease context may be done, for example, by isolating and sequencing total RNA from a diseased tissue sample set and cross-referencing the isolated sequences with known nucleic acid sequence databases to verify that the probe set on the transcriptome array is designed against the sequences actually expressed in the diseased tissue of interest. Methods for making transcriptome arrays are described in United States Patent Application Publication No. 2006/0134663, which is incorporated herein by reference. In certain example embodiments, the probe set of the transcriptome array is designed to bind within 300 nucleotides of the 3' end of a transcript.
  • the microarray used in deriving the gene expression profiles of the present invention is the Almac Ovarian Cancer DSATM microarray (Almac Group, Craigavon, United Kingdom).
  • An optimal linear classifier can be selected by evaluating a linear classifier's performance using such diagnostics as "area under the curve” (AUC).
  • AUC refers to the area under the curve of a receiver operating characteristic (ROC) curve, both of which are well known in the art.
  • ROC receiver operating characteristic
  • AUC measures are useful for comparing the accuracy of a classifier across the complete data range. Linear classifiers with a higher AUC have a greater capacity to classify unknowns correctly between two groups of interest (e.g., ovarian cancer samples and normal or control samples).
  • ROC curves are useful for plotting the performance of a particular feature (e.g., any of the biomarkers described herein and/or any item of additional biomedical information) in distinguishing between two populations (e.g., individuals responding and not responding to a therapeutic agent).
  • the feature data across the entire population e.g., the cases and controls
  • the true positive and false positive rates for the data are calculated.
  • the true positive rate is determined by counting the number of cases above the value for that feature and then dividing by the total number of positive cases.
  • the false positive rate is determined by counting the number of controls above the value for that feature and then dividing by the total number of controls.
  • ROC curves can be generated for a single feature as well as for other single outputs, for example, a combination of two or more features can be mathematically combined (e.g., added, subtracted, multiplied, etc.) to provide a single sum value, and this single sum value can be plotted in a ROC curve. Additionally, any combination of multiple features, in which the combination derives a single output value, can be plotted in a ROC curve. These combinations of features may comprise a test. The ROC curve is the plot of the true positive rate (sensitivity) of a test against the false positive rate (1-specificity) of the test.
  • an angiogenesis expression signature is directed to the 63 biomarkers detailed in Table 2 with corresponding ranks, and weights and associated bias detailed in the table or alternative rankings, and weightings and bias, depending, for example, on the disease setting.
  • Table 2 ranks the biomarkers in order of absolute decreasing weight, in an example classifier, in the compound decision score function.
  • the methods of the invention may rely upon measuring one or more, up to all, of the biomarkers listed in table 2.
  • the methods of the invention may comprise measuring the expression levels of at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 40, 50, 60 or each of the biomarkers from Table 2.
  • the method may comprise measuring the expression levels of 2 to 5 of the biomarkers from Table 2.
  • an angiogenesis expression signature is directed to the 63 biomarkers detailed in Table 3 with corresponding ranks detailed in the table or alternative rankings depending, for example, on the disease setting.
  • Table 3 ranks the biomarkers in order of absolute decreasing weight, in an example classifier, in the compound decision score function.
  • the methods of the invention may rely upon measuring one or more, up to all, of the biomarkers listed in table 3.
  • the methods of the invention may comprise measuring the expression levels of at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 40, 50, 60 or each of the biomarkers from Table 3.
  • the methods may comprise measuring the expression levels of 2 to 5 of the biomarkers from Table
  • an expression signature comprises all or a portion of the following biomarkers; IGF2, SOXll, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, RTP4, COL3A1, CDRl, RARRES3, TNFSFIO, NUAKl, SNORD114-14, SRPX, SPARC, GJB1, ⁇ 3, ISLR, TUBA1A, DEXI, BASP1, PXDN, GBP4, SLC28A3, HLA-DRA, TAP2, ACSL5, CDHll, PSMB9, MMP14, CD74, LOXLl, CIITA, ZNF697, SH3RF2, MIR198, COL1A2, TNFRSF14, COL8A1, C21orf63, TAP1, PDPN, RHOBTB3, BCL11A, HLA-DOB, XAF1, ARHGAP26, POLD2, DPYSL2, COL4A1, ID
  • an expression signature comprises IGF2, SOX11, INS, and CXCL17 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,
  • an expression signature comprises IGF2, INS, SPARC, TMEM45A, COL8A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
  • an expression signature comprises IGF2, INS, SPARC, TMEM45A, COL8A1, COL3A1, CDRl, NUAKl, TIMP3, LOXL1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53.
  • an expression signature comprises IGF2, TIMP3, INS, CXCR2P1, NUAKl and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
  • an expression signature comprises IGF2, TIMP3, INS, CXCR2P1, NUAKl, CDRl, MATN3, SOX11, SNORD114-14, COL3A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53.
  • an expression signature comprises COL3A1, SPARC, CDRl, SRPX, MATN3 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
  • an expression signature comprises COL3A1, SPARC, CDR1, SRPX, MATN3, TIMP3, CDH11, COL8A1, BCL11A, MMP14 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53.
  • an expression signature comprises IGF2, CDR1, COL3A1, SPARC, TIMP3 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
  • an expression signature comprises IGF2, CDR1, COL3A1, SPARC, TIMP3, INS, COL8A1, NUAKl, MATN3, TMEM45A and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
  • an expression signature comprises INS, SPARC, COL8A1, COL3A1, CDR1, NUAKl, TIMP3, and MMP14 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, or 55.
  • an expression signature comprises at least INS, SPARC, COL8A1, COL3A1, CDR1, NUAKl, TIMP3, and MMP14.
  • the expression signature comprises at least IGF2, CDR1, COL3A1, SPARC, TIMP3, INS, COL8A1, NUAKl, MATN3, TMEM45A.
  • the expression signature comprises at least IGF2, CDR1, COL3A1, SPARC, TIMP3.
  • the expression signature comprises at least, COL3A1, SPARC, CDR1, SRPX, MATN3, TIMP3, CDH11, COL8A1, BCL11A, MMP14.
  • the expression signature comprises at least COL3A1, SPARC, CDR1, SRPX, MATN3. In another example embodiment, the expression signature comprises at least COL3A1, SPARC, CDR1, SRPX, MATN3. In another example embodiment, the expression signature comprises at least IGF2, TIMP3, INS, CXCR2P1, NUAKl, CDR1, MATN3, SOX11, SNORD114-14, COL3A1. In another example embodiment, the expression signature comprises at least IGF2, TIMP3, INS, CXCR2P1, NUAKl. In another example embodiment, the expression signature comprises at least IGF2, INS, SPARC, TMEM45A, COL8A1, COL3A1, CDRl, NUAKl, TIMP3, LOXLl. In another example embodiment, the expression signature comprises at least IGF2, INS, SPARC, TMEM45A, COL8A1. In another example embodiment, the expression signature comprises at least IGF2, SOX11, INS, and CXCL17.
  • an expression signature comprises IGF2 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises SOX11 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises INS and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises CXCL17 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises CDRl and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises COL3A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises SPARC and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises TIMP3 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises COL8A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises NUAK1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises MATN3 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises TMEM45A and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises CXCR2P1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
  • an expression signature comprises SRPX and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.

Landscapes

  • Chemical & Material Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Organic Chemistry (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Zoology (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Analytical Chemistry (AREA)
  • Genetics & Genomics (AREA)
  • Wood Science & Technology (AREA)
  • Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Microbiology (AREA)
  • Molecular Biology (AREA)
  • Biotechnology (AREA)
  • Biophysics (AREA)
  • Hospice & Palliative Care (AREA)
  • Biochemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Oncology (AREA)
  • General Chemical & Material Sciences (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Medicinal Chemistry (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Pharmacology & Pharmacy (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Pharmaceuticals Containing Other Organic And Inorganic Compounds (AREA)
  • Acyclic And Carbocyclic Compounds In Medicinal Compositions (AREA)
  • Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
  • Medicines That Contain Protein Lipid Enzymes And Other Medicines (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

Methods for selecting whether to administer an anti-angiogenic therapeutic agent to a subject include steps of measuring the expression levels of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject; assessing from the expression levels of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature an anti-angiogenic therapeutic agent is contraindicated. Related prognostic methods and treatment methods are also provided. The invention is particularly applicable in ovarian and colorectal cancers.

Description

MOLECULAR DIAGNOSTIC TEST FOR PREDICTING RESPONSE TO
ANTI-ANGIOGENIC DRUGS AND PROGNOSIS OF CANCER
FIELD OF THE INVENTION
[0001] The present invention relates to a molecular diagnostic test useful for providing a prognosis and directing treatment of cancers from different anatomical sites. The invention includes the derivation of a gene classification model from gene expression levels. One application is the selection of whether to administer certain therapeutics, such as anti- angiogeneic therapeutics, to subjects receiving a standard of care cancer therapy. Another application is the stratification of cancer patients into those that have a good clinical prognosis or poor clinical prognosis. The present invention provides a test that can guide therapy selection as well as selecting patient groups for enrichment strategies during clinical trial evaluation of novel therapeutics. The invention can be used as a prognostic indicator for certain cancers including ovarian cancer, breast cancer, colon, prostate, lung and glioblastoma. The angiogenesis subtype can be identified from fresh/frozen (FF) or formalin fixed paraffin embedded FFPE patient samples.
BACKGROUND
[0002] The pharmaceutical industry continuously pursues new drug treatment options that are more effective, more specific or have fewer adverse side effects than currently administered drugs. Drug therapy alternatives are constantly being developed because genetic variability within the human population results in substantial differences in the effectiveness of many established drugs. Therefore, although a wide variety of drug therapy options are currently available, more therapies are always needed in the event that a patient fails to respond.
[0003] Traditionally, the treatment paradigm used by physicians has been to prescribe a first-line drug therapy that results in the highest success rate possible for treating a disease. Alternative drug therapies are then prescribed if the first is ineffective. This paradigm is clearly not the best treatment method for certain diseases. For example, in diseases such as cancer, the first treatment is often the most important and offers the best opportunity for successful therapy, so there exists a heightened need to choose an initial drug that will be the most effective against that particular patient's disease.
[0004] Ovarian cancer is the leading cause of death among all gynecological cancers in western countries. This high death rate is due to the diagnosis at an advanced stage in most patients. Epithelial ovarian cancer (EOC) constitutes 90% of ovarian malignancies and is classified into distinct histologic categories including serous, mucinous, endometrioid, clear cell, transitional, mixed, and undifferentiated subtypes. There is increasing evidence that these differed histologies arise from different aetiologies. There have been recent advances in the methodology used to classify epithelial ovarian cancer (McCluggage, W.G. "Morphological subtypes of ovarian carcinoma: a review with emphasis on new developments and pathogenesis," PATHOLOGY 2011 Aug;43(5):420-32). One of the consequences of this is that many tumors that would previously been classified as endometrioid are now being classified as serous.
[0005] The current standard treatment for ovarian cancer is debulking surgery and standard platinum taxane based cytotoxic chemotherapy. However, not all patients respond to this, and of those that do, approximately 70% will experience a recurrence. Specific targeted therapies for ovarian cancer based on histological or molecular classification have not yet reached the marketplace. Similarly for other types of cancer, there is still no accurate way of selecting appropriate cytotoxic chemotherapeutic agents.
[0006] The advent of microarrays and molecular genomics has the potential for a significant impact on the diagnostic capability and prognostic classification of disease, which may aid in the prediction of the response of an individual patient to a defined therapeutic regimen. Microarrays provide for the analysis of large amounts of genetic information, thereby providing a genetic fingerprint of an individual. There is much enthusiasm that this technology will ultimately provide the necessary tools for custom-made drug treatment regimens.
[0007] Currently, healthcare professionals have few mechanisms to help them identify cancer patients who will benefit from chemotherapeutic agents. Identification of the optimal first-line drug has been difficult because methods are not available for accurately predicting which drug treatment would be the most effective for a particular cancer's physiology. This deficiency results in relatively poor single agent response rates and increased cancer morbidity and death. Furthermore, patients often needlessly undergo ineffective, toxic drug therapy.
[0008] Angiogenesis is a key component of neo-vascularisation of tumors and essential to tumorigenesis and metastatsis. As such, it is a key area for therapeutic intervention and has been correlated to poor prognosis and reduced survival. This has promoted the development of a number of agents that target angiogenesis related processes and pathways, including the market leader and first FDA-approved anti-angiogenic, bevacizumab (Avastin), produced by Genentech/Roche .
[0009] Treatment regimens that include bevacizumab have demonstrated broad clinical activity However, no overall survival (OS) benefit has been shown after the addition of bevacizumab to cytotoxic chemotherapy in most cancers 8 ' 1 -"13. This suggests that a substantial proportion of tumours are either initially resistant or quickly develop resistance to VEGF blockade (the mechanism of action of bevacizumab). In fact, 21% of ovarian, 10% of renal and monotherapy, suggesting that bevacizumab may be active in small subgroups of patients, but that such incremental benefits do not reach significance in unselected patients. 1 -"18 As such, the use of a biomarker of response to bevacizumab would improve assessment of treatment outcomes and thus enable the identification of patient subgroups that would receive the most clinical benefit from bevacizumab treatment. This would be particularly relevant in the case of metastatic breast cancer, where the absence of a clinically beneficial biomarker has undermined the use of bevacizumab. Thus far, no such biomarker has been clinically validated to predict bevacizumab efficacy. Hypertension and VEGF polymorphisms are so far the only biomarkers to show potential, but important questions remain about their use in a clinical setting.
[0010] Another approach to anti-angiogenic therapy is simulataneous targeting of multiple angiogenic pathways rather than selective targeting of the VEGF pathway. Theoretically, multitargeted anti-angiogenic agents should more completely inhibit angiogenesis than agents such as bevacizumab and thus may produce greater therapeutic benefit. It has been postulated that in some tumors, angiogenesis may require VEGF only in the early stages of disease but is driven by additional angiogenic pathways as the disease progresses. Therefore, by targeting multiple pathways, it may be possible to counteract compensatory escape mechanisms that could lead to resistance to VEGF inhibition.
[0011] As for other types of cancer there is still no accurate way of selecting which patients will or will not respond to standard of care with an anti-angiogenic therapeutic or single agent anti-angiogenic therapy.
[0012] What is therefore needed is a molecular diagnostic test that would facilitate the stratification of patients based upon their predicted response to anti-angiogenic therapeutics, either in combination with standard of care or as a single-agent therapeutic. This would allow for the rapid identification of those patients who should receive alternative therapies. Such a molecular diagnostic test should be predictive of therapeutic responsiveness across different cancer types with sufficient accuracy.
SUMMARY OF THE INVENTION
[0013] Disclosed are methods of using one or more biomarkers, or a collection of biomarkers expressed in cancer that identify a subtype of cancer that is associated with an up- regulation in molecular signaling related to immune response and a down-regulation in molecular signaling related to angiogenesis and vasculature development, referred to herein as a "non-angiogenesis" or "immune" subtype. The collection of biomarkers may be defined by an expression signature, and the expression signature is used to assign a cumulative score to the measured expression values of the collection of biomarkers. In different aspects, the biomarkers and expression signatures may form the basis of a single parameter or multiparametric predictive test that could be delivered using methods known in the art such as microarray, next generation sequencing (NGS), Q-PCR, immunohistochemistry, ELISA or other technologies that can quantify mRNA or protein expression.
[0014] In addition, the cancer subtypes described herein are common to many types of cancer and are not limited to a single cancer disease type. Accordingly, the expression signatures of the present invention are not limited to a single cancer type. In certain example embodiments, the non-angiogenesis expression signature comprises two or more biomarkers selected from the biomarkers listed in Tables 1. In another example embodiment, the non- angiogenesis expression signature comprises two or more biomarkers listed in Table 2 or 3. In certain other example embodiments, the expression signature comprises one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2PA, MFAP2, MATN3, or RTP4. In another example embodiment, the expression signature comprises one or more of IGF2, CDR1, COL3A1, SPARC, TIMP3, INS, COL8A1, NUAK1, MATN3, and TMEM45A. In another example embodiment, the expression signature comprises one or more of INS, SPARC, COL8A1, COL3A1, CDR1, NUAK1, TIMP3, and MMP14. In another example embodiment, the non-angiogenesis signature comprises the biomarkers listed in Table 2 and their corresponding weights as determined using a PLS classifier. In another example embodiment, the non-angiogenesis signature comprises the biomarkers listed in Table 3 and their corresponding ranks within a decision function.
[0015] In one aspect the invention provides a method for selecting whether to administer an anti-angiogenic therapeutic agent to a subject, comprising: measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject; assessing from the expression level(s) of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature an anti-angiogenic therapeutic agent is contraindicated. In certain embodiments assessing whether the sample is positive or negative for the biomarker signature comprises: determining a sample expression score for the one or more biomarkers; comparing the sample expression score to a threshold score; and determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature. In further embodiments the subject is receiving or has received treatment with a chemotherapeutic agent. [0016] In a further aspect the invention provides a method of treating cancer comprising administering a chemotherapeutic agent and not administering an anti-angiogenic therapeutic agent to a subject wherein the subject is selected for treatment on the basis of a method as described herein and the subject is positive for the biomarker signature. According to a further aspect of the invention there is provided a method of treating cancer comprising administering a chemotherapeutic agent and not administering an anti-angiogenic therapeutic agent to a subject wherein the subject is selected for treatment by measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject; assessing from the expression levels of the biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is postive for the biomarker signature the subject is selected for treatment. In yet a further aspect, the present invention relates to a chemotherapeutic agent for use in treating cancer in a subject wherein the subject is selected for treatment on the basis of a method as described herein and is positive for the biomarker signature and wherein the subject is not treated with an anti-angiogenic therapeutic agent. The invention also relates to a chemotherapeutic agent for use in treating cancer in a subject wherein the subject is selected for treatment by measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject; assessing from the expression levels of the biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature the subject is selected for treatment and wherein the subject is not treated with an anti-angiogenic therapeutic agent. In a further aspect, the present invention relates to a method of treating cancer comprising administering a chemotherapeutic agent to a subject wherein the subject is positive for a biomarker signature defined by the expression levels of one or more biomarkers selected from Table 2 or Table 3 and wherein an anti-angiogenic therapeutic agent is not administered. In yet a further aspect, the present invention relates to a chemotherapeutic agent for use in treating cancer in a subject wherein the subject is positive for a biomarker signature defined by the expression levels of one or more biomarkers selected from Table 2 or Table 3 and wherein the subject is not treated with an anti-angiogenic therapeutic agent. In certain embodiments the chemotherapeutic agent comprises a platinum-based chemotherapeutic agent, an alkylating agent, an anti-metabolite (such as 5FU), an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof. The chemotherapeutic agent may comprise a platinum based-chemotherapeutic agent, a mitotic inhibitor, or a combination thereof. In specific embodiments the chemotherapeutic agent comprises carboplatin and/or paclitaxel. The chemotherapeutic agent may reflect the standard of care treatment for the cancer. The standard of care treatment may differ for different types of cancer - for example carboplatin in ovarian cancer, 5FU in colorectal cancer, platinum in head and neck cancer. According to all aspects of the invention assessing whether the sample is positive or negative for the biomarker signature may comprise determining a sample expression score for the one or more biomarkers; comparing the sample expression score to a threshold score; and determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature. According to all aspects of the invention the subject may be suffering from cancer. The cancer may be ovarian cancer, optionally high grade serous ovarian cancer. Herein "administering" an agent is used interchanging with "treating with" an agent.
[0017] According to a further aspect of the invention there is provided a method for determing clinical prognosis of a subject with cancer, comprising: measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject; assessing from the expression level(s) of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature the subject has a good prognosis. Assessing whether the sample is positive or negative for the biomarker signature may comprise: determining a sample expression score for the biomarkers; comparing the sample expression score to a threshold score; and determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature. In certain embodiments the good prognosis indicates increased progression free survival or overall survival rates compared to samples that are negative for the biomarker signature, optionally compared to samples with a sample expression score below the threshold score. In certain embodiments the subject is receiving, has received and/or will receive chemotherapeutic treatment and/or will not receive treatment with an anti- angiogenic therapeutic agent. The chemotherapeutic treatment may comprise administration of a platinum-based chemotherapeutic agent, an alkylating agent, an antimetabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof. The chemotherapeutic treatment may comprise administration of a platinum-based chemotherapeutic agent, a mitotic inhibitor, or a combination thereof. In specific embodiments the chemotherapeutic treatment comprises administration of paclitaxel and carboplatin. The cancer may be ovarian cancer or colorectal cancer
[0018] In a further aspect the present invention relates to a method for selecting whether to administer Bevacizumab to a subject, comprising: in a test sample obtained from a subject suffering from ovarian cancer, which subject is being, has been and/or will be treated using a platinum-based chemotherapeutic agent and/or a mitotic inhibitor; measuring expression levels of one, two or more, up to all of the, biomarkers selected from Table 2; assessing from the expression level(s) of the one, two or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, selecting a treatment based on whether the sample is positive for the biomarker signature, wherein is the sample is positive for the biomarker signature Bevacizumab is contraindicated. In certain embodiments assessing whether the sample is positive or negative for the biomarker signature comprises determining a sample expression score for the one, two or more biomarkers; comparing the sample expression score to a threshold score; and determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
[0019] The invention also relates to a method for determining clinical prognosis of a subject, comprising: (a) in a test sample obtained from a subject suffering from ovarian cancer, which subject is being, has been and/or will be treated using a platinum-based chemotherapeutic agent and/or a mitotic inhibitor; (b) measuring expression levels of one or more, up to all of the, biomarkers selected from Table 2 or Table 3; (c) assessing from the expression level(s) of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature the subject has a good prognosis. Assessing whether the sample is positive or negative for the biomarker signature may comprise: (i) determining a sample expression score for the one or more biomarkers; (ii) comparing the sample expression score to a threshold score; and (iii) determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature. In certain embodiments the good prognosis indicates increased progression free survival or overall survival rates compared to samples that are negative for the biomarker signature, optionally compared to samples with a sample expression score below the threshold score.
[0020] According to all aspects of the invention the expression level(s) of two or more biomarkers selected from Table 2 or Table 3 may be measured in a sample from the subject. The expression level(s) of five or more biomarkers selected from Table 2 or Table 3 may be measured in a sample from the subject. Assessing whether the sample is positive or negative for the biomarker signature may comprise the use of classification trees or random forests.
Classification trees (Breiman, Leo; Friedman, J. H.; Olshen, R. A.; Stone, C. J. (1984).
Classification and regression trees. Monterey, CA: Wadsworth & Brooks/Cole Advanced Books & Software. ISBN 978-0-412-04841-8) provide a means of predicting outcomes based on logic and rules. A classification tree is built through a process called binary recursive partitioning, which is an iterative procedure of splitting the data into partitions/branches. The goal is to build a tree that distinguishes among pre-defined classes. Each node in the tree corresponds to a variable. To choose the best split at a node, each variable is considered in turn, where every possible split is tried and considered, and the best split is the one which produces the largest decrease in diversity of the classification label within each partition. This is repeated for all variables, and the winner is chosen as the best splitter for that node. The process is continued at the next node and in this manner, a full tree is generated. One of the advantages of classification trees over other supervised learning approaches such as discriminant analysis, is that the variables that are used to build the tree can be either categorical, or numeric, or a mix of both. In this way it is possible to generate a classification tree for predicting outcomes based on say the directionality of gene expression. Random forest algorithms (Breiman, Leo (2001). "Random Forests". Machine Learning 45 (1): 5-32. doi: 10.1023/A: 1010933404324) provide a further extension to classification trees, whereby a collection of classification trees are randomly generated to form a "forest" and an average of the predicted outcomes from each tree is used to make inference with respect to the outcome.
[0021] In one aspect, a method for selecting whether to administer an anti-angiogenic therapeutic agent to a subject using the expression signatures disclosed herein is provided, the method comprising obtaining a test sample from the subject, measuring expression levels of a biomarker panel from the test sample, determining a sample expression score for the biomarker panel, comparing the sample expression score to a threshold score, and selecting a treatment based on whether the expression score is equal to or above the threshold score. In certain example embodiments, a sample expression score is equal to or above the threshold score indicates an anti-angiogenic agent is contraindicated and should not be administered to the subject. In certain example embodiments, a sample expression score below the threshold score indicates an anti-angiogenic agent is not contraindicated and can be administered to the subject. A therapeutic agent is "contraindicated" or "detrimental" to a patient if the cancer's rate of growth is accelerated as a result of contact with the therapeutic agent, compared to its growth in the absence of contact with the therapeutic agent. Growth of a cancer can be measured in a variety of ways. For instance, the size of a tumor, or measuring the expression of tumor markers appropriate for that tumor type. A therapeutic agent can also be considered "contraindicated" or "detrimental" if the patient' s overall prognosis (progression free survival and overall survival) is reduced by the administration of the therapeutic agent. In one example embodiment, the expression signature disclosed herein may determine a patient' s clinical prognosis upon administration of an anti-angiogeneic agent following standard cancer therapy. [0022] In certain example embodiments, the subject suffers from cancer. The cancer may include, but is not limited to, ovarian cancer, breast cancer, colon cancer, colorectal cancer, glioblastoma, kidney cancer, including renal cell carcinoma, heatocelluar cancer, thyroid cancer, pancreatic cancer, neuroendocrine cancer, esophageal cancer, gastrointestinal stromal tumors (GIST), gastric cancer, liver cancer, including adult primary liver cancer, lymphoma, melanoma, or multiple myeloma. In certain example embodiments, the cancer is ovarian cancer. In certain other example embodiments, the ovarian cancer is high grade serous ovarian cancer. In certain example embodiments, the patient may have received, is receiving and/or will receive a treatment which may be a standard of care treatment for the cancer type of the subject. In certain example embodiments, that treatment which may be a standard of care treatment may include treatment with a chemotherapeutic agent. The chemotherapeutic treatment may include administration of a platinum-based chemotherapeutic agent, an alkylating agent, an antimetabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof. In certain example embodiments, the chemotherapeutic treatment comprises administration of a platinum-based chemotherapeutic agent, a mitotic inhibitor, or a combination thereof. In certain other example embodiments, the chemotherapeutic treatment comprises administration of carboplatin and paclitaxel. In one example embodiment, the subject has high grade serous ovarian cancer and has previously received a platinum-based chemotherapeutic agent and a mitotic inhibitor. In another example embodiment, the subject has high grade serous ovarian cancer and has previously received carboplatin and paclitaxel. The anti-angiogenic therapeutic agent may be a VEGF-pathway-targeted therapeutic agent (such as bevacizumab or aflibercept), an angiopoietin-TIE2 pathway inhibitor, an endogenous angiogenic inhibitor, or an immunomodulatory agent. In one example embodiment, the anti- angiogenic therapeutic agent is a VEGF-pathway-targeted therapeutic agent. In another example embodiment, the anti-angiogenic therapeutic agent is bevacizumab.
[0023] In another aspect, a method for determining a clinical prognosis of a subject using the expression signatures disclosed herein is provided, the method comprising obtaining a test sample from the subject, measuring expression levels of a biomarker panel from the test sample, determining a sample expression score for the biomarker panel, comparing the sample expression score to the threshold expression score, wherein if the expression score is equal to or above the threshold expression score the clinical prognosis is a good prognosis. In certain example embodiments, a good prognosis indicates increased survival rates compared to a subject with an expression score below the threshold score. In certain example embodiments, the subject suffers from cancer. The cancer may include, but is not limited to, ovarian cancer, breast cancer, colon cancer, colorectal cancer, or glioblastoma. In certain example embodiments, the cancer is ovarian cancer. In certain other example embodiments, the ovarian cancer is high grade serous ovarian cancer. In certain example embodiments, the subject may receive, has received and/or will receive a chemotherapeutic treatment. The chemotherapeutic treatment may include administration of a platinum-based chemotherapeutic agent, an alkylating agent, an anti-metabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof. In certain example embodiments, the chemotherapeutic treatment comprises administration of a platinum-based chemotherapeutic agent, a mitotic inhibitor, or a combination thereof. In certain other example embodiments, the chemotherapeutic treatment comprises administration of carboplatin and paclitaxel. In one example embodiment, the subject has high grade serous ovarian cancer and may receive, has received and/or will receive a platinum-based chemotherapeutic agent and a mitotic inhibitor, such as taxane. In another example embodiment, the subject has high grade serous ovarian cancer and is may receive, has received, and/or will receive carboplatin and paclitaxel.
[0024] In another aspect, the present invention relates to kits for conventional diagnostic uses listed above such as qPCR, NGS, microarray, and immunoassays such as immunohistochemistry, ELISA, Western blot and the like. Such kits include appropriate reagents and directions to assay the expression of the genes or gene products and quantify rnPvNA or protein expression.
[0025] Also disclosed are methods for identifying human tumors with or without the non- angiogenesis phenotype. In certain example embodiments, such methods may be used to identify patients that are sensitive to and respond to drugs that inhibit, either directly or indirectly, processes relating to angiogenesis. In certain other example embodiments, such methods may be used to identify patients that are resistant to or do not respond or will respond in adverse fashion to drugs that inhibit, either directly or indirectly, processes relating to angiogenesis.
[0026] This invention also relates to guiding effective treatment of patients. Further, methods relating to selection of patient treatment regimens and selecting patients for clinical trials of current, or developmental stage drugs that directly or indirectly affect angiogenesis are provided.
[0027] In addition, methods that accommodate the use of archived formalin fixed paraffin-embedded (FFPE) biopsy material, as well as fresh/frozen (FF) tissue, for assay of all transcripts, and are therefore compatible with the most widely available type of biopsy material, are described herein. A biomarker expression level may be determined using RNA obtained from FFPE tissue, fresh frozen tissue or fresh tissue that has been stored in solutions such as RNAlater®. BRIEF DESCRIPTION OF DRAWINGS
[0028] Figure 1 provides a heat map showing unsupervised hierarchical clustering of gene expression data in 265 high grade serous ovarian carcinomas. Each column represents the expression of these probe sets in one tumor. Probe set expression across all clusters is represented horizontally. The bar above the heat map is color-coded by cluster as described in the legend box. The second bar is color-coded for class label as described in the legend box. Functional processes corresponding to each probe set cluster are labeled to the right of the figure.
[0029] Figure 2 provides Kaplan-Meier analysis of overall survival by cluster from unsupervised analysis of gene expression in 265 high grade serous ovarian carcinomas.
[0030] Figure 3 provides Kaplan-Meier analysis of survival of the two classes defined by the 63 -gene signature classifier in the Edinburgh (discovery) dataset. Proangiogenic group consists of Angio and Angioimmune subgroups. A. Progression-free survival. B. Overall survival.
[0031] Figure 4 provides Kaplan-Meier analysis of survival of the two classes defined by the 63-gene signature classifier in the Tothill (validation) dataset. Proangiogenic group consists of Angio and Angioimmune subgroups. A. Progression-free survival. B. Overall survival.
[0032] Figure 5 provides Kaplan Meier curves for progression free survival in Immune
(Figure 5A) and Proangiogenic (Figure 5B) subgroups of the patients in the ICON7 trail cohort. Within each figure the survival differences are displayed across the 2 randomized treatment groups: 1) Carboplatin plus paclitaxel chemotherapy and 2) Carboplatin plus paclitaxel chemotherapy plus bevacizumab.
[0033] Figure 6 provides Kaplan Meier curves for overall survival in Immune (Figure
6A) and Proagniogenic (Figure 6A) subgroups of patients in the ICON7 trial cohort. Within each figure the survival differences are displayed across the 2 randomized treatment groups: 1) Carboplatin plus paclitaxel chemotherapy and 2) Carboplatin plus paclitaxel chemotherapy plus bevacizumab.
[0034] Figure 7 provides Kaplan Meier curves for progression free survival (A) and overall survival (B) for carboplatin and paclitaxel treated ICON7 trail patients defined by the 63 gene signature.
[0035] Figure 8A and 8B are graphs demonstrating certain classification performance benchmarks of an example non-angiogenesis signature as applied to colorectal cancer samples.
[0036] Figure 9: Signature development: AUC of training set under CV.
[0037] Figure 10: Signature development: C-Index of training set under CV. [0038] Figure 11 : Signature development: HR of training set under CV.
[0039] Figure 12: Signature development: HR of ICON7 SOC samples under CV.
[0040] Figure 13: Signature development: C-Index of ICON7 SOC samples under CV.
[0041] Figure 14: Signature development: HR of ICON7 Immune samples under CV.
[0042] Figure 15: Signature development: HR of ICON7 ProAngio samples under CV.
[0043] Figure 16: Core set analysis: lmmune63GeneSig_CoreGenes_InternalVal.png.
[0044] Figure 17: Core set analysis: Immune63GeneSig_CoreGenes_Tothill.png.
[0045] Figure 18: Core set analysis: Immune63GeneSig_CoreGenes_ICON7_SOC.png.
[0046] Figure 19: Minimum gene set analysis:
Immune63GeneSig_MinGenes_Tothill.png.
[0047] Figure 20: ICON7 SOC: Minimum gene set analysis:
Immune63 GeneSig_MinGenes_ICON7_S OC .png .
[0048] Figure 21 : ICON7 Immune: Minimum gene set analysis:
Immune63GeneSig_MinGenes_ICON7_Immune.png.
[0049] Figure 22: Kaplan Meier to show the differences in progression free survival probability between the samples predicted as Angio-Off (inactive) versus those that are predicted as Angio-On (active) by the 63 gene signature
[0050] Figure 23: Semi-supervised hierarchical clustering of the 529 CRC samples published by Marrisa et al (2013) using the angiogenesis defining gene list.
[0051] Figure 24: ROC Curve showing the discrimination in 63 gene signature scores between the angiogenesis active subtype and angiogenesis inactive subtype in the Marissa CRC data.
[0052] Figure 25: Kaplan Meier Curve showing the survival differences between the angiogenesis active and angiogenesis inactive patients (treated only) as predicted by the 63 gene signature in the GSE14333 CRC data.
DETAILED DESCRIPTION OF THE INVENTION
[0053] Unless defined otherwise, technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. Definitions of common terms in molecular biology may be found in Benjamin Lewin, Genes IX, published by Jones and Bartlet, 2008 (ISBN 0763752223); Kendrew et al. (eds.), The Encyclopedia of Molecular Biology, published by Blackwell Science Ltd., 1994 (ISBN 0632021829); Robert A. Meyers (ed.), Molecular Biology and Biotechnology: a Comprehensive Desk Reference, published by VCH Publishers, Inc., 1995 (ISBN 9780471185710); Singleton et al., Dictionary of Microbiology and Molecular Biology 2nd ed., J. Wiley & Sons (New York, N.Y. 1994), and March, Advanced Organic Chemistry Reactions, Mechanisms and Structure 4th ed., John Wiley & Sons (New York, N.Y. 1992).
[0054] The singular terms "a," "an," and "the" include plural referents unless context clearly indicates otherwise. Similarly, the word "or" is intended to include "and" unless the context clearly indicates otherwise. The term "comprises" means "includes." In case of conflict, the present specification, including explanations of terms, will control.
[0055] As used herein terms "biomarker panel," "expression classifier," "classifier,"
"expression signature," or "signature" may be used interchangeably. The panel typically includes a plurality of biomarkers but may include only a single biomarker where that biomarker is useful individually in the methods of the invention.
[0056] All publications, published patent documents, and patent applications cited in this application are indicative of the level of skill in the art(s) to which the application pertains. All publications, published patent documents, and patent applications cited herein are hereby incorporated by reference to the same extent as though each individual publication, published patent document, or patent application was specifically and individually indicated as being incorporated by reference.
Overview
[0057] A major goal of current research efforts in cancer is to increase the efficacy of perioperative systemic therapy in patients by incorporating molecular parameters into clinical therapeutic decisions. Pharmacogenetics/genomics is the study of genetic/genomic factors involved in an individuals' response to a foreign compound or drug. Agents or modulators which have a stimulatory or inhibitory effect on expression of a biomarker of the invention can be administered to individuals to treat (prophylactically or therapeutically) cancer in the patient. It is ideal to also consider the pharmacogenomics of the individual in conjunction with such treatment. Differences in metabolism of therapeutics may possibly lead to severe toxicity or therapeutic failure by altering the relationship between dose and blood concentration of the pharmacologically active drug. Thus, understanding the pharmacogenomics of an individual permits the selection of effective agents (e.g., drugs) for prophylactic or therapeutic treatments. Such pharmacogenomics can further be used to determine appropriate dosages and therapeutic regimens. Accordingly, the level of expression of a biomarker of the invention in an individual can be determined to thereby select appropriate agent(s) for therapeutic or prophylactic treatment of the individual.
[0058] The present invention relates to a molecular diagnostic tests useful for diagnosing cancers from different anatomical sites that includes the use of one or more common subtypes related to angiogenesis. The invention includes expression signatures that identify a subject as having a good or poor clinical prognosis, and expression signatures that indicate whether to administer an anti-angiogenic therapeutic agent to a subject. The expression signature is derived by obtaining the expression profiles of samples from a sample set of known pathology and/or clinical outcome. The samples may originate from the same sample tissue type or different tissue types. As used herein an "expression profile" comprises a set of values representing the expression level for each biomarker analyzed from a given sample.
[0059] The expression profiles from the sample set are then analyzed using a mathematical model. Different mathematical models may be applied and include, but are not limited to, models from the fields of pattern recognition (Duda et al. Pattern Classification, 2nd ed., John Wiley, New York 2001), machine learning (Scholkopf et al. Learning with Kernels, ΜΓΓ Press, Cambridge 2002, Bishop, Neural Networks for Pattern Recognition, Clarendon Press, Oxford 1995), statistics (Hastie et al. The Elements of Statistical Learning, Springer, New York 2001), bioinformatics (Dudoit et al, 2002, J. Am. Statist. Assoc. 97:77-87, Tibshirani et al, 2002, Proc. Natl. Acad. Sci. USA 99:6567-6572) or chemometrics (Vandeginste, et al, Handbook of Chemometrics and Qualimetrics, Part B, Elsevier, Amsterdam 1998). The mathematical model identifies one or more biomarkers expressed in the sample set that are most predictive of a given disease phenotype. These one or more biomarkers define an expression signature. Accordingly, an expression signature includes the biomarkers identified as most predictive of a given disease phenotype. In certain example embodiments, the mathematical model defines a variable, such as a weight, for each identified biomarker. In certain example embodiments, the mathematical model defines a decision function. The decision function may further define a threshold score which separates the sample set into two disease phenotypes such as, but not limited to, samples that have a good and poor clinical prognosis. In one example embodiment, the decision function and expression signature are defined using a linear classifier.
[0060] To classify new samples using a defined expression signature, the biomarkers defined by the expression signature, also referred to as a biomarker panel, are isolated and an expression profile of the biomarker panel determined. The new sample biomarker panel expression profile is analyzed with the same mathematical model used to define the expression signature. The biomarker panel may comprise one or more of the biomarkers defined by the expression signature. The biomarker panel may comprise one or more of the biomarkers defined by the expression signature. In certain example embodiments, the biomarker panel comprises one or more of the biomarkers listed in Table 2. In certain other example embodiments, the biomarker panel comprises all of the biomarkers listed in Table 2. In certain example embodiments, the biomarker panel comprises one or more of the biomarkers listed in Table 2. In certain other example embodiments, the biomarker panel comprises all of the biomarkers listed in Table 2. In certain example embodiments, the mathematical model defines an expression score for the new sample. The expression score may be determined by combining the expression values of the biomarkers with corresponding scalar weights using non-linear, algebraic, trigonometric or correlative means to derive a single scalar value. The expression score is compared to the threshold score and the sample classified based on whether the expression score is greater than, or equal to, or less than the threshold score. In certain example embodiments, a sample expression score equal to or greater than the threshold score indicates a subject has a good clinical prognosis, and a sample expression score below the threshold score indicates a subject has a poor clinical prognosis. In certain example embodiments, a sample expression score equal to or greater than the threshold score indicates a subject has the signature. This may indicate a good clinical prognosis. A sample expression score below the threshold score indicates a subject does not have the signature. This may indicate a poor clinical prognosis.
[0061] One application of the expression signatures disclosed herein is the identification of patients with a good and poor clinical prognosis. The good or poor prognosis may be determined in the context of a certain treatment background (such as carboplatin/paclitaxel therapy as discussed herein). For example, the subject may be receiving or have received a standard chemotherapeutic treatment for the subject's cancer type. Given a treatment background, the expression signatures disclosed herein may also be used to determine whether an additional therapeutic agent, such as an anti-angiogenic therapeutic agent, should be administered to the patient. By examining the expression of at least one, optionally a collection of the identified biomarkers in a tumor, it is possible to determine the likely clinical outcomes of a patient. By examining the expression of at least one, optionally a collection of biomarkers, it is therefore possible to identify those patients in most need of more aggressive therapeutic regimens and likewise eliminate unnecessary therapeutic treatments or those unlikely to significantly improve or possibly harm a patient's clinical outcome. The present invention relates to prediction of clinical prognosis using at least progression free survival or overall survival rates. Accordingly, a "good prognosis" indicates a subject population with a cancer subtype that demonstrates an increased survival rate compared to other cancer subtypes, whereas a "poor prognosis" or "bad prognosis" indicates a subject population with a cancer subtype that demonstrates decreased survival rate compared to other cancer subtypes. Additional prognostic factors that may be considered are ethnicity and race, age, stage of disease, histology, tumor grade, tumor makers (for example, CA125), site-specific surgical treatment, size of residual disease, and tumor response. In certain example embodiments, a subject with an expression score equal to or above the threshold score is classified as having the non-angiogenesis subtype. In another example embodiment, a subject with a sample expression score above the threshold score is classified as having a good clinical prognosis. In yet another example embodiment, a subject with a sample expression score above or equal to the threshold score indicates the subject will likely experience a detrimental effect, or have a poorer clinical prognosis, if administered an anti-angiogenic therapeutic agent.
[0062] In certain example embodiments, the determination of a subject's clinical prognosis or selection of an additional therapeutic agent may be made in the context of past, current, or planned chemotherapeutic treament. For example, the subject may set to start, be currently receiving, or have just completed, a standard of care chemotherapeutic treatment for the cancer type of the subject. In certain example embodiments, the chemotherapeutic treament may include administration of an alkylating agent, an anti-metabolite, a platinum-based drug, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, a corticosteroid, a hormone based therapeutic, or a combination thereof. Example alkylating agents include nitrogen mustards, nitrosureas, alkyl sulfonates, triazines, and ethylenimines. Example platinum drugs include cisplatin, carboplatin, and oxalaplatin. Example anti-metabolites include 5-fluoruracil, 6-mercaptopurine, capecitabine, cladribine, clofarabine, cytarabine, floxuridine, fludarabine, gemcitabine, hydroxyurea, methotrexate, pemetrexed, pentostatin, and thioguanine. Example anti-tumor antibiotics include daunorubicin, doxorubicin, epirubicin, idarubicin, actinomocyin- D, bleomycin, mitomycin-C, and mitoxantrone. Example topoisomerase inhibtors include topotecan, irinotecan, etoposide, and teniposide. Example mitotic inhibitors include taxanes, epothilones, vinca alkaloids, and estramustine. Example corticosteroids include predisone, mefhylprednisolone, and dexamethasone. In certain example embodiments the chemotherapy may include treatment with L-asparaginase, imatinib, gefitinib, sunitinib, bortezomib, retinoids, tretinoin, bexaroten, arsenic trioxide, fluvestrant, tamoxifen, toremifene, anastrozole, exemestane, letrozole, progestins, estrogens, bicalutamide, flutamide, nilutamide, gonadotropoin-releasing hormone agonists or analogs, rituximab, alemtuzumab, BCG, interleukin-2, interferon- alf a, thalidomide, and lenalidomide.
[0063] In certain example embodiments the chemotherapeutic treament may comprise a cyclosphoshamide, methotrexate, and fluorouracil (CMF) treatment regimen, a cyclophosphamide, doxorubicin, and fluorouracil (CAF) treatment regimen, an epirubicin and cyclophosphamide (EC) treatment regimen, a fluorouracil, epirubicin, and cyclophosphamide (FEC) treatment regimen, a paclitaxel and cyclophosphamide treatment regimen, a paclitaxel and carboplatin treatment regiment, a doxorubicin and cyclophosphamide treatment regiment, or a doxorubicin and paclitaxel treatment regimen. In one example embodiment, the neoadjuvant cancer therapy comprises a platinum based chemotherapy treatment regimen. In one example embodiment, the platinum-based chemotherapy treatement regimen comprises paclitaxel and carboplatin.
[0064] Another application of the expression signatures disclosed herein is the stratification of response to, and selection of patients for therapeutic drug classes that encompass anti-angiogenic therapies. By examining the expression of a collection of the identified biomarkers in a tumor, it is possible to determine which therapeutic agent or combination of agents will be most likely to reduce the growth rate of a cancer. It is also possible to determine which therapeutic agent or combination of agents will be the least likely to reduce the growth rate of a cancer and/or which may cause adverse affects and thus be contra-indicated. By examining the expression of a collection of biomarkers, it is therefore possible to eliminate ineffective or inappropriate therapeutic agents. Importantly, in certain embodiments, these determinations can be made on a patient-by-patient basis or on an agent-by- agent basis. Thus, one can determine whether or not a particular therapeutic regimen is likely to benefit a particular patient or type of patient, and/or whether a particular regimen should be continued. The present invention provides a test that can guide therapy selection as well as selecting patient groups for enrichment strategies during clinical trial evaluation of novel therapeutics. For example, when evaluating a putative anti-angiogeneic agent or treatment regime, the expression signatures and methods disclosed herein may be used to select individuals for clinical trials that have cancer subtypes that are responsive to anti-angiogenic agents.
[0065] A cancer is "responsive" to a therapeutic agent if its rate of growth is inhibited as a result of contact with the therapeutic agent, compared to its growth in the absence of contact with the therapeutic agent. Growth of a cancer can be measured in a variety of ways. For instance, the size of a tumor or measuring the expression of tumor markers appropriate for that tumor type. In one example embodiment, the expression signature disclosed herein may determine a patient's clinical prognosis upon administration of an anti-angiogeneic agent following standard of care chemotherapeutic therapy for the cancer type of the patient.
[0066] A cancer is "non-responsive" to a therapeutic agent if its rate of growth is not inhibited, or inhibited to a very low degree, as a result of contact with the therapeutic agent when compared to its growth in the absence of contact with the therapeutic agent. As stated above, growth of a cancer can be measured in a variety of ways, for instance, the size of a tumor or measuring the expression of tumor markers appropriate for that tumor type. The quality of being non-responsive to a therapeutic agent is a highly variable one, with different cancers exhibiting different levels of "non-responsiveness" to a given therapeutic agent, under different conditions. Still further, measures of non-responsiveness can be assessed using additional criteria beyond growth size of a tumor such as, but not limited to, patient quality of life, and degree of metastases.
[0067] The angiogenesis subtype can be identified from a fresh/frozen (FF) or formalin fixed paraffin embedded (FFPE) patient sample. In one example embodiment, the cancer type is ovarian cancer, breast cancer, colon cancer, colorectal cancer, lung cancer, prostate cancer, or glioblastoma. In another example embodiment, the cancer type is an ovarian cancer. In a further example embodiment, the cancer type is breast cancer. In another example embodiment, the cancer type is lung cancer. In another example embodiment, the cancer type is colon cancer. In another example embodiment, the cancer type is prostate cancer. In another example embodiment, the cancer type is glioblastoma.
Identifying Expression Signatures
The expression signatures of the present invention are identified by analyzing the expression profiles of certain biomarkers in a patient sample set. Biomarkers suitable for use in the present invention include DNA, RNA, and proteins. In one example embodiment, biomarkers suitable for use in the present invention include RNA and cDNA. The biomarkers are isolated from a patient sample and their expression levels determined to derive a set of expression profiles for each sample analyzed in the patient sample set. In certain example embodiments the expression signature identifies a non-angiogenesis phenotype observed in cancer tissues, identified as a signature score for a combination of biomarkers above or equal to a threshold, the phenotype characterized by an up-regulation of immune response related genes and a down-regulation of genes associated with angiogeneisis or vasculature development related processes. a.
Expression Profiles
[0068] In certain embodiments, the expression profile obtained is a genomic or nucleic acid expression profile, where the amount or level of one or more nucleic acids in the sample is determined. In these embodiments, the sample that is assayed to generate the expression profile employed in the diagnostic or prognostic methods is a nucleic acid sample. The nucleic acid sample includes a population of nucleic acids that includes the expression information of the phenotype determinative biomarkers of the cell or tissue being analyzed. In some embodiments, the nucleic acid may include RNA or DNA nucleic acids, e.g., mRNA, cRNA, cDNA etc., so long as the sample retains the expression information of the host cell or tissue from which it is obtained. The sample may be prepared in a number of different ways, as is known in the art, e.g., by mRNA isolation from a cell, where the isolated mRNA is used as isolated, amplified, or employed to prepare cDNA, cRNA, etc., as is known in the field of differential gene expression. Accordingly, determining the level of mRNA in a sample includes preparing cDNA or cRNA from the mRNA and subsequently measuring the cDNA or cRNA. The sample is typically prepared from a cell or tissue harvested from a subject in need of treatment, e.g., via biopsy of tissue, using standard protocols, where cell types or tissues from which such nucleic acids may be generated include any tissue in which the expression pattern of the to be determined phenotype exists, including, but not limited to, disease cells or tissue, body fluids, etc.
[0069] The expression profile may be generated from the initial nucleic acid sample using any convenient protocol. While a variety of different manners of generating expression profiles are known, such as those employed in the field of differential gene expression/biomarker analysis, one representative and convenient type of protocol for generating expression profiles is array-based gene expression profile generation protocols. Such applications are hybridization assays in which a nucleic acid that displays "probe" nucleic acids for each of the genes to be assayed/profiled in the profile to be generated is employed. In these assays, a sample of target nucleic acids is first prepared from the initial nucleic acid sample being assayed, where preparation may include labeling of the target nucleic acids with a label, e.g., a member of a signal producing system. Following target nucleic acid sample preparation, the sample is contacted with the array under hybridization conditions, whereby complexes are formed between target nucleic acids that are complementary to probe sequences attached to the array surface. The presence of hybridized complexes is then detected, either qualitatively or quantitatively. Specific hybridization technology which may be practiced to generate the expression profiles employed in the subject methods includes the technology described in U.S. Pat. Nos. 5,143,854; 5,288,644; 5,324,633; 5,432,049; 5,470,710; 5,492,806; 5,503,980; 5,510,270; 5,525,464; 5,547,839; 5,580,732; 5,661,028; 5,800,992; the disclosures of which are herein incorporated by reference; as well as WO 95/21265; WO 96/31622; WO 97/10365; WO 97/27317; EP 373 203; and EP 785 280. In these methods, an array of "probe" nucleic acids that includes a probe for each of the biomarkers whose expression is being assayed is contacted with target nucleic acids as described above. Contact is carried out under hybridization conditions, e.g., stringent hybridization conditions as described above, and unbound nucleic acid is then removed. The resultant pattern of hybridized nucleic acids provides information regarding expression for each of the biomarkers that have been probed, where the expression information is in terms of whether or not the gene is expressed and, typically, at what level, where the expression data, i.e., expression profile, may be both qualitative and quantitative.
b. Diseases and Sample Tissue Sources
[0070] In certain example embodiments, the patient sample comprises a cancer tissue samples, such as an archived sample. The patient sample is preferably derived from cancer tissue and may be from a sample having been characterized by prognosis, likelihood of recurrence, long term survival, clinical outcome, treatment response, diagnosis, cancer classification, or personalized genomics profile. As used herein cancer includes, but is not limited to, leukemia, brain cancer, prostate cancer, liver cancer, ovarian cancer, stomach cancer, colorectal cancer, throat cancer, breast cancer, skin cancer, melanoma, lung cancer, sarcoma, cervical cancer, testicular cancer, bladder cancer, endocrine cancer, endometrial cancer, esophageal cancer, glioma, lymphoma, neuroblastoma, osteosarcoma, pancreatic cancer, pituitary cancer, renal cancer, and the like. As used herein, colorectal cancer encompasses cancers that may involve cancer in tissues of both the rectum and other portions of the colon as well as cancers that may be individually classified as either colon cancer or rectal cancer. In one embodiment, the methods described herein refer to cancers that are treated with anti-angiogenic agents, anti-angiogenic targeted therapies, inhibitors of angiogenesis signaling, but not limited to these classes. These cancers also include subclasses and subtypes of these cancers at various stages of pathogenesis. In certain example embodiments, the patient sample comprises an ovarian cancer sample. In certain example embodiments, the ovarian cancer sample is a serous ovarian cancer sample such as a high grade serous ovarian cancer sample. In another example embodiment, the patient sample comprises a breast cancer sample. In yet another example embodiment, the patient sample comprises a glioblastoma sample.
[0071] "Biological sample", "sample", and "test sample" are used interchangeably herein to refer to any material, biological fluid, tissue, or cell obtained or otherwise derived from an individual. This includes blood (including whole blood, leukocytes, peripheral blood mononuclear cells, buffy coat, plasma, and serum), sputum, tears, mucus, nasal washes, nasal aspirate, breath, urine, semen, saliva, meningeal fluid, amniotic fluid, glandular fluid, lymph fluid, nipple aspirate, bronchial aspirate, synovial fluid, joint aspirate, ascites, cells, a cellular extract, and cerebrospinal fluid. This also includes experimentally separated fractions of all of the preceding. For example, a blood sample can be fractionated into serum or into fractions containing particular types of blood cells, such as red blood cells or white blood cells (leukocytes). If desired, a sample can be a combination of samples from an individual, such as a combination of a tissue and fluid sample. The term "biological sample" also includes materials containing homogenized solid material, such as from a stool sample, a tissue sample, or a tissue biopsy, for example. The term "biological sample" also includes materials derived from a tissue culture or a cell culture, including tissue resection and biopsy samples. Any suitable methods for obtaining a biological sample can be employed; example methods include, e.g., phlebotomy, swab (e.g., buccal swab), and a fine needle aspirate biopsy procedure. Samples can also be collected, e.g., by micro dissection (e.g., laser capture micro dissection (LCM) or laser micro dissection (LMD)), bladder wash, smear (e.g., a PAP smear), or ductal lavage. A "biological sample" obtained or derived from an individual includes any such sample that has been processed in any suitable manner after being obtained from the individual, for example, fresh frozen or formalin fixed and/or paraffin embedded. The methods of the invention as defined herein may begin with an obtained sample and thus do not necessarily incorporate the step of obtaining the sample from the patient. The methods may be in vitro methods performed on an isolated sample.
[0072] As used herein, the term "patient" includes human and non-human animals. The preferred patient for treatment is a human. "Patient," "individual" and "subject" are used interchangeably herein.
c. Biomarkers
[0073] As used herein, the term "biomarker" can refer to a gene, an mRNA, cDNA, an antisense transcript, a miRNA, a polypeptide, a protein, a protein fragment, or any other nucleic acid sequence or polypeptide sequence that indicates either gene expression levels or protein production levels. When a biomarker indicates or is a sign of an abnormal process, disease or other condition in an individual, that biomarker is generally described as being either over- expressed or under-expressed as compared to an expression level or value of the biomarker that indicates or is a sign of a normal process, an absence of a disease or other condition in an individual. "Up-regulation", "up-regulated", "over-expression", "over-expressed", and any variations thereof are used interchangeably to refer to a value or level of a biomarker in a biological sample that is greater than a value or level (or range of values or levels) of the biomarker that is typically detected in similar biological samples from healthy or normal individuals. The terms may also refer to a value or level of a biomarker in a biological sample that is greater than a value or level (or range of values or levels) of the biomarker that may be detected at a different stage of a particular disease.
[0074] "Down-regulation", "down-regulated", "under-expression", "under-expressed", and any variations thereof are used interchangeably to refer to a value or level of a biomarker in a biological sample that is less than a value or level (or range of values or levels) of the biomarker that is typically detected in similar biological samples from healthy or normal individuals. The terms may also refer to a value or level of a biomarker in a biological sample that is less than a value or level (or range of values or levels) of the biomarker that may be detected at a different stage of a particular disease.
[0075] Further, a biomarker that is either over-expressed or under-expressed can also be referred to as being "differentially expressed" or as having a "differential level" or "differential value" as compared to a "normal" expression level or value of the biomarker that indicates or is a sign of a normal process or an absence of a disease, disease subtype, or other condition in an individual. Thus, "differential expression" of a biomarker can also be referred to as a variation from a "normal" expression level of the biomarker.
[0076] The terms "differential biomarker expression" and "differential expression" are used interchangeably to refer to a biomarker whose expression is activated to a higher or lower level in a subject suffering from a specific disease, relative to its expression in a normal subject, or relative to its expression in a patient that responds differently to a particular therapy or has a different prognosis. The terms also include biomarkers whose expression is activated to a higher or lower level at different stages of the same disease. It is also understood that a differentially expressed biomarker may be either activated or inhibited at the nucleic acid level or protein level, or may be subject to alternative splicing to result in a different polypeptide product. Such differences may be evidenced by a variety of changes including mRNA levels, miRNA levels, antisense transcript levels, or protein surface expression, secretion or other partitioning of a polypeptide. Differential biomarker expression may include a comparison of expression between two or more genes or their gene products; or a comparison of the ratios of the expression between two or more genes or their gene products; or even a comparison of two differently processed products of the same gene, which differ between normal subjects and subjects suffering from a disease; or between various stages of the same disease. Differential expression includes both quantitative, as well as qualitative, differences in the temporal or cellular expression pattern in a biomarker among, for example, normal and diseased cells, or among cells which have undergone different disease events or disease stages.
[0077] In certain example embodiments, the biomarker is an RNA transcript. As used herein "RNA transcript" refers to both coding and non-coding RNA, including messenger RNAs (mRNA), alternatively spliced mRNAs, ribosomal RNA (rRNA), transfer RNA (tRNA), small nuclear RNAs (snRNA), and antisense RNA. Measuring mRNA in a biological sample may be used as a surrogate for detection of the level of the corresponding protein and gene in the biological sample. Thus, any of the biomarkers or biomarker panels described herein can also be detected by detecting the appropriate RNA. Methods of biomarker expression profiling include, but are not limited to quantitative PCR, NGS, northern blots, southern blots, microarrays, SAGE, immunoassays (ELISA, EIA, agglutination, nephelometry, turbidimetry, Western blot, immunoprecipitation, immunocytochemistry, flow cytometry, Luminex assay), and mass spectrometry. The overall expression data for a given sample may be normalized using methods known to those skilled in the art in order to correct for differing amounts of starting material, varying efficiencies of the extraction and amplification reactions.
[0078] In certain example embodiments, biomarkers useful for distinguishing between cancer subtypes that demonstrate a good clinical prognosis and a poor clinical prognosis and can be determined by identifying biomarkers exhibiting the highest degree of variability between samples in the patient data set as determined using the expression detection methods and patient sample sets discussed above. Standard statistical methods known in the art for identifying highly variable data points in expression data may be used to identify the highly variable biomarkers. For example, a combined background and variance filter to the patient data set. The background filter is based on the selection of probe sets with expression E and expression variance var£ above the thresholds defined by background standard deviation aBg (from the Expression Console software) and quantile of the standard normal distribution za at a specified significance a probe sets were kept if:
E > log2((¾o Eg)); log2((var£) > 2 [log2(aBg) - E - log2(log(2))]
[0079] where a defines a significance threshold. In certain example embodiment, the significance threshold is 6.3 · 10"5. In another example embodiment, the significance threshold may be between 1.0 · 10"7 to 1.0 · 10"3.
[0080] In certain example embodiments, the highly variable biomarkers may be further analyzed to group samples in the patient data set into subtypes or clusters based on similar gene expression profiles. For examples, biomarkers may be clustered based on how highly correlated the up-regulation or down-regulation of their expression is to one another. Various clustering analysis techniques known in the art may be used. In one example embodiment, hierarchical agglomerative clustering is used to identify the cancer subtypes. To determine the biological relevance of each subtype, the biomarkers within each cluster may be further mapped to their corresponding genes and annotated by cross-reference to one or more gene ontology databases containing information on biological activity and biological pathways associated with the gene. In another example embodiment, biomarkers in clusters that are up regulated and enriched for immune response general functional terms are grouped into a putative non-angiongenesis sample group and used for expression signature generation. In another example embodiment, biomarkers in clusters that are down regulated and enriched for angiogenesis and vasculature development and are up regulated and enriched for immune response general functional terms are grouped into a putative non-angiongenesis sample group and used for expression signature generation. Further details for conducting functional analysis of biomarker clusters is provided in the Examples section below.
[0081] In one example embodiment, the biomarkers useful in deriving an expression signature for use in the present invention are those biomarkers listed in Table 1. These biomarkers are identified as having predictive value to determine a patient response to a therapeutic agent and/or a prognostice value in identifying individuals with a good or poor clinical prognosis. [0082] In certain example embodiments, the expression of the biomarkers disclosed herein correlates with whether a patient will experience a detrimental or beneficial effect from administration of an anti-angiogenic therapeutic agent. By examining the expression of a collection of biomarkers, it is therefore possible to eliminate ineffective or inappropriate therapeutic agents. Importantly, in certain embodiments, these determinations can be made on a patient-by-patient basis or on an agent-by-agent basis. Thus, one can determine whether or not a particular therapeutic regimen is likely to benefit a particular patient or type of patient, and/or whether a particular regimen should be continued.
[0083] In certain other example embodiments, the expression of the biomarkers disclosed herein correlate with a patient's overall clinical prognosis. By examinaing the expression of a collection of biomarkers identified in a tumor, it is possible to determine whether the individual has a cancer subtype associated with good clinical prognosis or poor clinical prognosis. Importantly, in certain embodiments, these determinations can be made on a patient-by-patient basis. Thus, one of ordinary skill in the art can use predicted prognosis to help select appropriate treatment regimens to treat the underlying disease while eliminating those treatment regimens most likely to produce undesired or medically unwarranted adverse side effects.
[0084] The SEQ ID NOs listed in Table 1 refer to probe set identifiers used to measure the expression levels of the genes on an example transcriptome array. Expression signatures of the present invention have been cross-validated using expression data from different arrays with differnt probe sets as detailed further in the Examples section below. Accordingly, the expression signatures and methods disclosed herein are not limited to expression values measured using the probe sets disclosed herein.
Table 1: Genes in Clusters of Figure 1
Figure imgf000025_0001
13 Sense (Fully Exonic) IGKC
14 Sense (includes Intronic) NFATC1
15 Sense (Fully Exonic) HMHA1
16 Sense (Fully Exonic) FCGR1B
17 Sense (Fully Exonic) EDA2R
18 Sense (Fully Exonic) COL8A1
19 Sense (Fully Exonic) COL12A1
20 Sense (Fully Exonic) HLA-B
21 Sense HLA-F
22 Sense (Fully Exonic) EGR1
23 Sense (Fully Exonic) SULF2
24 Sense (Fully Exonic) CERCAM
25 Sense (Fully Exonic) ATF3
26 Sense (Fully Exonic) MIR21
27 Sense (Fully Exonic) IFIT2
28 Sense (Fully Exonic) IGLC3
29 Sense (Fully Exonic) IGLC3
30 Sense (Fully Exonic) IGLC3
31 Sense (Fully Exonic) IGLC3
32 Sense (Fully Exonic) IGLC3
33 Sense (Fully Exonic) IGLC3
34 Sense (Fully Exonic) ANGPTL2
35 Sense (Fully Exonic) COL5A2
36 Sense (Fully Exonic) THY1
37 Sense (Fully Exonic) NDN
38 Sense (Fully Exonic) RGS2
39 Sense (Fully Exonic) MEIS3P2
40 Sense (Fully Exonic) GBP2
41 Sense (Fully Exonic) FAT1
42 Sense (Fully Exonic) COL1A1
43 Sense (Fully Exonic) MMP11
44 Sense (Fully Exonic) GADD45B
45 Sense (Fully Exonic) MMP14
46 Sense (Fully Exonic) IGHG4
47 Sense (Fully Exonic) HCLS1
48 No Transcript match
49 Sense (Fully Exonic) JAM3
50 Sense (Fully Exonic) TMEM49
51 Sense (Fully Exonic) LTBP2 52 Sense (Fully Exonic) IRS 1
53 Sense (Fully Exonic) C17orf91
54 Sense (Fully Exonic) GPNMB
55 Sense (Fully Exonic) FAM198B
56 Sense (Fully Exonic) CHST15
57 Sense (Fully Exonic) DCN
58 Sense (Fully Exonic) VCAM1
59 Sense (Fully Exonic) CIITA
60 Sense (Fully Exonic) GAS 7
61 Sense (Fully Exonic) COL3A1
62 Sense (Fully Exonic) ITGB2
63 Sense (Fully Exonic) ELN
64 Sense (Fully Exonic) CMTM3
65 Sense (Fully Exonic) ANTXR1
66 Sense (Fully Exonic) IL4I1
67 Sense (Fully Exonic) ANTXR2
68 Sense (Fully Exonic) IGLC2 /// IGLC3
69 Sense (Fully Exonic) IGLC3
70 Sense (Fully Exonic) BST2
71 Sense (Fully Exonic) COL10A1
72 Sense (Fully Exonic) IGLC3
73 Sense (Fully Exonic) FBP1
74 Sense (Fully Exonic) RHOBTB3
75 Sense (Fully Exonic) CD74
76 Sense (Fully Exonic) ISM1
77 Sense (Fully Exonic) CSRNP1
78 Sense (Fully Exonic) DCN
79 Sense (Fully Exonic) IGFBP4
80 Sense (Fully Exonic) CCDC80
81 Sense (Fully Exonic) COL3A1
82 Sense (Fully Exonic) ZFP36
83 Sense (Fully Exonic) MMP11
84 Sense (Fully Exonic) COL1A2
85 Sense (Fully Exonic) HLA-DPA1
86 Sense (Fully Exonic) TWIST 1
87 Sense (Fully Exonic) ZNF154
88 Sense (Fully Exonic) IGLC3
89 Sense (Fully Exonic) IGKC
90 Sense (Fully Exonic) IGHG1 91 Sense (Fully Exonic) COL1A2
92 Sense (Fully Exonic) APOC1
93 AntiSense EGR1
94 Sense (Fully Exonic) KIAA0146
95 Sense (Fully Exonic) TPM1
96 Sense (includes Intronic) DMD
97 No Transcript match
98 Sense (Fully Exonic) DUSP1
99 Sense (Fully Exonic) GBP1
100 Sense (includes Intronic) PDGFC
101 Sense (includes Intronic) MSN
102 Sense (includes Intronic) TPM1
103 Sense (Fully Exonic) EMB
104 Sense (Fully Exonic) FOS
105 Sense (includes Intronic) DPYSL3
106 AntiSense EGR1
107 AntiSense NRP2
108 Sense (Fully Exonic) MMP2
109 Sense (Fully Exonic) CTGF
110 Sense (Fully Exonic) ACTA2
111 Sense (Fully Exonic) LOXL1
112 Sense (Fully Exonic) CDH11
113 Sense (Fully Exonic) LUM
114 Sense (Fully Exonic) NNMT
115 Sense (Fully Exonic) GJA1
116 AntiSense CTHRC1
117 Sense (Fully Exonic) CTSB
118 Sense (Fully Exonic) PLAU
119 Sense (Fully Exonic) PDGFRA
120 Sense (Fully Exonic) VCAN
121 AntiSense --
Sense (Fully Exonic) IGHG4 /// IGHG2 /// IGHGl
122 llll GHGP
123 Sense (Fully Exonic) IGHG2
124 Sense (includes Intronic) C3orf26
125 AntiSense ATF3
126 AntiSense ATF3
127 Sense (Fully Exonic) FN1
128 AntiSense CALD1
129 AntiSense CALD1
130 AntiSense EGR1
131 AntiSense TWIST 1
132 AntiSense TWIST 1
133 AntiSense BATF2 134 AntiSense NFKBLZ
135 Sense (includes Intronic) C3orf26
136 AntiSense LOXL1
137 Sense (includes Intronic) --
138 AntiSense FN1
139 AntiSense COL1A1
140 Sense (Fully Exonic) TREH
141 AntiSense APOL1
142 Sense (Fully Exonic) COL10A1
143 Sense (Fully Exonic) GAL3ST4
144 Sense (Fully Exonic) TAGLN
145 Sense (Fully Exonic) TWIST 1
146 Sense (Fully Exonic) HCLS1
147 Sense (Fully Exonic) ITGB2
148 Sense (Fully Exonic) HLA-B
149 Sense (Fully Exonic) C17orf91
150 Sense (Fully Exonic) FBLIM1
151 Sense (Fully Exonic) COL15A1
152 Sense (Fully Exonic) AQP7P3
153 AntiSense IGFBP5
154 Sense (Fully Exonic) FANK1
155 AntiSense INS
156 Sense (Fully Exonic) COL27A1
157 Sense (Fully Exonic) COL5A1
158 Sense (Fully Exonic) PRICKLE2
159 Sense (Fully Exonic) N/A
160 Sense (Fully Exonic) GXYLT2
161 Sense (includes Intronic) KLF12
162 No Transcript match
163 Sense (Fully Exonic) FBX032
164 No Transcript match
165 Sense (Fully Exonic) ASAH2B
166 AntiSense PPFIBP1
167 AntiSense XIST
168 Sense (Fully Exonic) IGFBP6
169 Sense (Fully Exonic) ROB 01
170 Sense (Fully Exonic) TPM1
171 AntiSense N/A
172 AntiSense PLEKHG1 173 Sense (Fully Exonic) NR2F1
174 Sense (Fully Exonic) NPDC1
175 AntiSense INS
176 Sense (Fully Exonic) TRAF5
177 Sense (Fully Exonic) CALD1
178 Sense (includes Intronic) CHRM3
179 Sense (Fully Exonic) AMOTL1
180 Sense (includes Intronic) COL12A1
181 Sense (Fully Exonic) PLXNA4
182 Sense (includes Intronic) TMEM43
183 Sense (includes Intronic) RORA
184 AntiSense INS
185 Sense (Fully Exonic) TSPAN18
186 No Transcript match
187 Sense (Fully Exonic) TNC
188 Sense (Fully Exonic) TYR03
189 AntiSense EFNA5
190 Sense (Fully Exonic) MYL9
191 Sense (Fully Exonic) MIR198
192 Sense (includes Intronic) N/A
193 Sense (includes Intronic) PLA2R1
194 Sense (Fully Exonic) COL14A1
195 Sense (Fully Exonic) NRP1
196 Sense (Fully Exonic) FSCN1
197 Sense (includes Intronic) PDGFD
198 No Transcript match
199 Sense (includes Intronic) DOCK4
200 Sense (Fully Exonic) TRIM 13
201 Sense (Fully Exonic) IGFBP5
202 Sense (Fully Exonic) C19orf63
203 AntiSense KLF6
204 AntiSense TRIO
205 Sense (Fully Exonic) COL4A1
206 Sense (Fully Exonic) EPDR1
207 Sense (Fully Exonic) FNDC1
208 Sense (Fully Exonic) IL1R1
209 Sense (Fully Exonic) CES4
210 Sense (Fully Exonic) GPR176
211 Sense (includes Intronic) GXYLT2 212 AntiSense WHSC1L1
213 Sense (Fully Exonic) N/A
214 Sense (Fully Exonic) RGN
215 Sense (includes Intronic) CA3
216 Sense (Fully Exonic) TIMP3
217 Sense (Fully Exonic) EFNA5
218 Sense (Fully Exonic) RASGRF2
219 Sense (includes Intronic) RELL1
220 AntiSense ACS S3
221 Sense (Fully Exonic) STMN3
222 Sense (Fully Exonic) N/A
223 AntiSense C7orf29
224 Sense (Fully Exonic) HOXC6
225 Sense (Fully Exonic) KLF8
226 Sense (includes Intronic) SERINC5
227 Sense (Fully Exonic) AKT3
228 Sense (Fully Exonic) TGFB2
229 AntiSense WNT5A
230 No Transcript match
231 No Transcript match
232 AntiSense IGFBP7
233 No Transcript match
234 Sense (includes Intronic) SULT1C4
235 Sense (Fully Exonic) AASS
236 Sense (Fully Exonic) HEPH
237 Sense (Fully Exonic) ADH5
238 Sense (Fully Exonic) TIMP2
239 Sense (Fully Exonic) EMP1
240 Sense (Fully Exonic) CXCL14
241 Sense (Fully Exonic) ZNF548
242 Sense (Fully Exonic) SGCB
243 Sense (includes Intronic) ASH2L
244 Sense (includes Intronic) SERINC5
245 No Genome match
246 Sense (Fully Exonic) TMEM159
247 Sense (includes Intronic) RBMS3
248 Sense (Fully Exonic) TMEM49
249 Sense (includes Intronic) RORA
250 No Transcript match 251 AntiSense ZNF608
252 No Genome match
253 Sense (Fully Exonic) ADAMTS2
254 Sense (Fully Exonic) APCDD1
255 AntiSense GXYLT2
256 Sense (Fully Exonic) XIST
257 Sense (Fully Exonic) MBNL2
258 Sense (Fully Exonic) SHF
259 Sense (includes Intronic) APBB2
260 No Transcript match
261 Sense (Fully Exonic) COL14A1
262 Sense (Fully Exonic) IGFBP5
263 Sense (Fully Exonic) CREB5
264 AntiSense INS
265 Sense (Fully Exonic) BAHCC1
266 Sense (Fully Exonic) RFXAP
267 Sense (Fully Exonic) INS
268 Sense (Fully Exonic) DDR2
269 Sense (Fully Exonic) CA12
270 Sense (Fully Exonic) RHOB
271 Sense (Fully Exonic) N/A
272 Sense (Fully Exonic) SNORD 116-4
273 Sense (Fully Exonic) MEG3
274 Sense (Fully Exonic) WNT4
275 Sense (Fully Exonic) FBLN2
276 AntiSense DAAM1
277 No Transcript match
278 Sense (Fully Exonic) CHN1
279 Sense (includes Intronic) APBB2
280 Sense (Fully Exonic) PTRF
281 AntiSense IGF1
282 Sense (Fully Exonic) UST
283 Sense (Fully Exonic) SMARCA1
284 Sense (includes Intronic) N/A
285 Sense (Fully Exonic) IGLC3
286 AntiSense INS
287 Sense (Fully Exonic) KANK4
288 AntiSense IGF1
289 Sense (Fully Exonic) CYP27A1 290 AntiSense EIF2B5
291 No Transcript match
292 Sense (Fully Exonic) SNRNP25
293 Sense (Fully Exonic) SETD7
294 Sense (Fully Exonic) MSX1
295 Sense (Fully Exonic) HOPX
296 Sense (Fully Exonic) NID2
297 Sense (Fully Exonic) IGF1
298 Sense (Fully Exonic) PSD3
299 Sense (Fully Exonic) FGFR1
300 Sense (Fully Exonic) ETV1
301 Sense (Fully Exonic) ZNF655
302 No Genome match
303 AntiSense INS
304 Sense (Fully Exonic) SFRP2
305 Sense (Fully Exonic) SPAG16
306 AntiSense NR2F2
307 Sense (includes Intronic) SYNP02
308 Sense (Fully Exonic) FAM101B
309 AntiSense IGF2
310 Sense (Fully Exonic) CA3
311 Sense (Fully Exonic) XIST
312 No Transcript match
313 Sense (Fully Exonic) WNT7A
314 Sense (includes Intronic) N/A
315 Sense (Fully Exonic) FGFR1
316 AntiSense FXYD6
317 Sense (Fully Exonic) FGFR1
318 Sense (includes Intronic) IGFBP7
319 Sense (Fully Exonic) TIMP2
320 Sense (Fully Exonic) DUSP1
321 Sense (includes Intronic) SERINC5
322 No Transcript match
323 Sense (Fully Exonic) ABLIM1
324 Sense (Fully Exonic) ARL4A
325 AntiSense SH3TC2
326 AntiSense NR2F2
327 Sense (Fully Exonic) ENG
328 Sense (Fully Exonic) MGP 329 Sense (Fully Exonic) MEG3
330 AntiSense FAM115A
331 Sense (Fully Exonic) EGR1
332 Sense (Fully Exonic) SNORD 116-3
333 Sense (Fully Exonic) AEBP1
334 Sense (includes Intronic) SDK1
335 Sense (Fully Exonic) ENC1
336 Sense (Fully Exonic) SNORD 116-7
337 Sense (Fully Exonic) N/A
338 Sense (Fully Exonic) APOD
339 AntiSense N/A
340 AntiSense GAS 1
341 Sense (Fully Exonic) VPS 36
342 No Transcript match
343 Sense (Fully Exonic) SPHK2
344 Sense (Fully Exonic) SNORD116-8
345 Sense (Fully Exonic) MYO10
346 Sense (Fully Exonic) HOXC6
347 Sense (Fully Exonic) RNF149
348 Sense (Fully Exonic) BTG2
349 Sense (includes Intronic) MAP3K1
350 Sense (Fully Exonic) SNORD 116-23
351 Sense (includes Intronic) ACSL4
352 Sense (Fully Exonic) CYP27C1
353 Sense (includes Intronic) COL12A1
354 Sense (Fully Exonic) IGFBP5
355 Sense (Fully Exonic) DUSP4
356 Sense (Fully Exonic) PFKFB3
357 Sense (Fully Exonic) SDC2
358 AntiSense FXYD6
359 Sense (Fully Exonic) COL5A1
360 Sense (Fully Exonic) MARCKS
361 Sense (Fully Exonic) IRS2
362 Sense (Fully Exonic) N/A
363 AntiSense FSCN1
364 Sense (Fully Exonic) FYN
365 Sense (Fully Exonic) IGFBP5
366 Sense (Fully Exonic) NUDT4P1
367 Sense (Fully Exonic) NFKBLZ 368 Sense (Fully Exonic) N/A
369 Sense (Fully Exonic) C7orf41
370 Sense (Fully Exonic) MEG3
371 Sense (Fully Exonic) N/A
372 Sense (Fully Exonic) PLEKHG1
373 Sense (Fully Exonic) ZNF827
374 Sense (Fully Exonic) ZNF175
375 Sense (Fully Exonic) XIST
376 Sense (includes Intronic) GSN
377 Sense (includes Intronic) RORA
378 Sense (Fully Exonic) CA13
379 AntiSense TMX4
380 Sense (Fully Exonic) KIT
381 Sense (includes Intronic) WDR78
382 Sense (Fully Exonic) ECEL1
383 Sense (Fully Exonic) XIST
384 Sense (Fully Exonic) PROCR
385 Sense (Fully Exonic) C9orfl67
386 Sense (Fully Exonic) MUC6
387 Sense (includes Intronic) P4HA2
388 Sense (Fully Exonic) FAM69C
389 Sense (Fully Exonic) NOX4
390 Sense (includes Intronic) N/A
391 No Transcript match
392 Sense (Fully Exonic) SMOX
393 Sense (Fully Exonic) KIAA0922
394 No Transcript match
395 Sense (Fully Exonic) XIST
396 Sense (Fully Exonic) NPAS2
397 Sense (Fully Exonic) NAV1
398 Sense (includes Intronic) N/A
399 Sense (Fully Exonic) HLA-A
400 Sense (Fully Exonic) FAM46C
401 Sense (Fully Exonic) N/A
402 Sense (Fully Exonic) SLAMF7
403 Sense (Fully Exonic) FCER1G
404 Sense (Fully Exonic) CIS
405 Sense (Fully Exonic) NUPR1
406 AntiSense C1QC 407 AntiSense SAT1
408 Sense (Fully Exonic) SOD2
409 Sense (Fully Exonic) IRF1
410 Sense (Fully Exonic) SFN
411 AntiSense LTB
412 Sense (Fully Exonic) ARID5A
413 Sense (Fully Exonic) BST2
414 Sense (Fully Exonic) HLA-F
415 Sense (Fully Exonic) XAF1
416 Sense (Fully Exonic) TCOF1
417 Sense (Fully Exonic) RPL23AP1
418 Sense (Fully Exonic) IL1RN
419 Sense (Fully Exonic) IFIT5
420 Sense (Fully Exonic) B2M
421 AntiSense GBP1
422 Sense (Fully Exonic) HLA-F
423 Sense (Fully Exonic) DGKA
424 Sense (Fully Exonic) XBP1
425 Sense (Fully Exonic) PLCG2
426 Sense (Fully Exonic) FAM46C
427 No Genome match
428 Sense (Fully Exonic) TREM2
429 Sense (Fully Exonic) LGALS9
430 Sense (Fully Exonic) HLA-DPB 1
431 AntiSense ODF3B
432 Sense (Fully Exonic) MX1
433 Sense (Fully Exonic) STAT1
434 Sense (Fully Exonic) CTSB
435 Sense (Fully Exonic) FAM26F
436 Sense (includes Intronic) PARP14
437 AntiSense SAT1
438 Sense (Fully Exonic) CTSS
439 No Transcript match
440 Sense (Fully Exonic) CTSB
441 Sense (Fully Exonic) ADAM8
442 Sense (includes Intronic) B2M
443 Sense (Fully Exonic) FLVCR2
444 Sense (Fully Exonic) TYROBP
445 AntiSense SAMD9L 446 Sense (Fully Exonic) SAMD9L
447 Sense (Fully Exonic) SIGLECl
448 Sense (Fully Exonic) MMP7
449 Sense (Fully Exonic) APOL1
450 Sense (Fully Exonic) CYLD
451 Sense (Fully Exonic) HLA-B
452 Sense (Fully Exonic) SAT1
453 Sense (Fully Exonic) C1QB
454 Sense (Fully Exonic) HLA-DMB
455 Sense (Fully Exonic) NLRC5
456 Sense (Fully Exonic) FAM20A
457 AntiSense N/A
458 Sense (Fully Exonic) STAT1
459 Sense (includes Intronic) STAT1
460 Sense (Fully Exonic) STAT1
461 AntiSense N/A
462 Sense (Fully Exonic) DERL3
463 Sense (Fully Exonic) HLA-F
464 Sense (Fully Exonic) MAFB
465 Sense (Fully Exonic) CD4
466 Sense (Fully Exonic) HLA-A
467 Sense (Fully Exonic) UBE2L6
468 Sense (Fully Exonic) C1QC
469 Sense (Fully Exonic) CD163
470 Sense (Fully Exonic) LRMP
471 Sense (Fully Exonic) Cl lorfl7
472 Sense (Fully Exonic) XAF1
473 Sense (Fully Exonic) GLRX
474 Sense (Fully Exonic) IFIH1
475 Sense (Fully Exonic) CD44
476 Sense (Fully Exonic) LITAF
477 Sense (Fully Exonic) CCDC69
478 Sense (Fully Exonic) GBP5
479 Sense (Fully Exonic) PML
480 Sense (Fully Exonic) SAMD9
481 Sense (Fully Exonic) CBR3
482 Sense (Fully Exonic) RASGRP2
483 Sense (Fully Exonic) FCGR2A
484 Sense (Fully Exonic) BST2 485 Sense (Fully Exonic) HLA-A
486 Antisense COL1A1
487 No Genome Match
488 No Genome Match
[0085] In certain example embodiments, all or a portion of the biomarkers recited in Table 1 may be used in an expression signature. For example, expression signatures comprising the biomarkers in Table 1 can be generated using the methods provided herein and can comprise between one, and all of the markers set forth in Tables 1 and each and every combination in between (e.g., four selected markers, 16 selected markers, 74 selected markers, etc.). In some embodiments, the expression signature comprises at least 5, 10, 20, 40, 60, 100, 150, 200, or 300 or more markers. In other embodiments, the predictive biomarker panel comprises no more than 5, 10, 20, 40, 60, 100, 150, 200, 300, 400, 500, 600 or 700 markers. In one example embodiment, the expression signature includes a plurality of markers listed in Table 1. In some embodiments the expression signature includes at least about 1%, about 5%, about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80%, about 90%, about 95%, about 96%, about 97%, about 98%, or about 99% of the markers listed in Table 1. Selected expression signatures can be assembled from the biomarkers provided using methods described herein and analogous methods known in the art. In one embodiment, the expression signature contains all genes or gene products in Table 1.
4. Mathematical Models
[0086] The following methods may be used to derive expression signatures for distinguishing between subjects that are responsive or non-responsive to anti-angiogenic therapeutics, or as prognostic indicators of certain cancer types, including expression signatures derived from the biomarkers disclosed above. In certain other example embodiments, the expression signature is derived using a decision tree (Hastie et al. The Elements of Statistical Learning, Springer, New York 2001), a random forest (Breiman, 2001 Random Forests, Machine Learning 45:5), a neural network (Bishop, Neural Networks for Pattern Recognition, Clarendon Press, Oxford 1995), discriminant analysis (Duda et al. Pattern Classification, 2nd ed., John Wiley, New York 2001), including, but not limited to linear, diagonal linear, quadratic and logistic discriminant analysis, a Prediction Analysis for Microarrays (PAM, (Tibshirani et al, 2002, Proc. Natl. Acad. Sci. USA 99:6567-6572)) or a Soft Independent Modeling of Class Analogy analysis. (SIMCA, (Wold, 1976, Pattern Recogn. 8: 127-139)).
[0087] Biomarker expression values may be defined in combination with corresponding scalar weights on the real scale with varying magnitude, which are further combined through linear or non-linear, algebraic, trigonometric or correlative means into a single scalar value via an algebraic, statistical learning, Bayesian, regression, or similar algorithms which together with a mathematically derived decision function on the scalar value provide a predictive model by which expression profiles from samples may be resolved into discrete classes of responder or non-responder, resistant or non-resistant, to a specified drug, drug class, molecular subtype, or treatment regimen. Such predictive models, including biomarker membership, are developed by learning weights and the decision threshold, optimized for sensitivity, specificity, negative and positive predictive values, hazard ratio or any combination thereof, under cross-validation, bootstrapping or similar sampling techniques, from a set of representative expression profiles from historical patient samples with known drug response and/or resistance.
[0088] In one embodiment, the biomarkers are used to form a weighted sum of their signals, where individual weights can be positive or negative. The resulting sum ("expression score") is compared with a pre-determined reference point or value. The comparison with the reference point or value may be used to diagnose, or predict a clinical condition or outcome.
[0089] As described above, one of ordinary skill in the art will appreciate that the biomarkers included in the classifier provided in Tables 1 will carry unequal weights in a classifier for determining clinical prognosis. Therefore, while as few as one biomarker may be used to diagnose or predict an clinical prognosis or response to a therapeutic agent, the specificity and sensitivity or diagnosis or prediction accuracy may increase using more biomarkers.
[0090] As used herein, the term "weight" refers to the absolute magnitude of an item in a statistical calculation. The weight of each biomarker in a gene expression classifier may be determined on a data set of patient samples using learning methods known in the art. As used herein the term "bias" or "offset" refers to a constant term derived using the mean expression of the signatures genes in a training set and is used to mean-center the each gene analyzed in the test dataset.
[0091] In certain example embodiments, the expression signature is defined by a decision function. A decision function is a set of weighted expression values derived using a linear classifier. All linear classifiers define the decision function using the following equation:
f(x) = w' · x + b =∑ Wi » X; +b (1)
[0092] All measurement values, such as the microarray gene expression intensities x for a certain sample are collected in a vector x. Each intensity is then multiplied with a corresponding weight w, to obtain the value of the decision function f(x) after adding an offset term b. In deriving the decision function, the linear classifier will further define a threshold value that splits the gene expression data space into two disjoint sections. Example linear classifiers include but are not limited to partial least squares (PLS), (Nguyen et al, Bioinformatics 18 (2002) 39-50), support vector machines (SVM) (Scholkopf et al, Learning with Kernels, ΜΓΤ Press, Cambridge 2002), and shrinkage discriminant analysis (SDA) (Ahdesmaki et al., Annals of applied statistics 4, 503-519 (2010)). In one example embodiment, the linear classifier is a PLS linear classifier.
[0093] The decision function is empirically derived on a large set of training samples, for example from patients showing a good or poor clinical prognosis. The threshold separates a patient group based on different characteristics such as, but not limited to, clinical prognosis before or after a given therapeutic treatment. The interpretation of this quantity, i.e. the cut-off threshold, is derived in the development phase ("training") from a set of patients with known outcome. The corresponding weights and the responsiveness/resistance cut-off threshold for the decision score are fixed a priori from training data by methods known to those skilled in the art. In one example embodiment, Partial Least Squares Discriminant Analysis (PLS-DA) is used for determining the weights. (L. Stahle, S. Wold, J. Chemom. 1 (1987) 185- 196; D. V. Nguyen, D.M. Rocke, Bioinformatics 18 (2002) 39-50).
[0094] Effectively, this means that the data space, i.e. the set of all possible combinations of biomarker expression values, is split into two mutually exclusive groups corresponding to different clinical classifications or predictions, for example, one corresponding to good clinical prognosis and poor clinical prognosis. In the context of the overall classifier, relative over- expression of a certain biomarker can either increase the decision score (positive weight) or reduce it (negative weight) and thus contribute to an overall decision of, for example, a good clinical prognosis.
[0095] In certain example embodiments of the invention, the data is transformed non- linearly before applying a weighted sum as described above. This non-linear transformation might include increasing the dimensionality of the data. The non-linear transformation and weighted summation might also be performed implicitly, for example, through the use of a kernel function. (Scholkopf et al. Learning with Kernels, MIT Press, Cambridge 2002).
[0096] In certain example embodiments, the patient training set data is derived by isolated RNA from a corresponding cancer tissue sample set and determining expression values by hybridizing the isolated RNA to a microarray. In certain example embodiments, the microarray used in deriving the expression signature is a transcriptome array. As used herein a "transcriptome array" refers to a microarray containing probe sets that are designed to hybridize to sequences that have been verified as expressed in the diseased tissue of interest. Given alternative splicing and variable poly- A tail processing between tissues and biological contexts, it is possible that probes designed against the same gene sequence derived from another tissue source or biological context will not effectively bind to transcripts expressed in the diseased tissue of interest, leading to a loss of potentially relevant biological information. Accordingly, it is beneficial to verify what sequences are expressed in the disease tissue of interest before deriving a microarray probe set. Verification of expressed sequences in a particular disease context may be done, for example, by isolating and sequencing total RNA from a diseased tissue sample set and cross-referencing the isolated sequences with known nucleic acid sequence databases to verify that the probe set on the transcriptome array is designed against the sequences actually expressed in the diseased tissue of interest. Methods for making transcriptome arrays are described in United States Patent Application Publication No. 2006/0134663, which is incorporated herein by reference. In certain example embodiments, the probe set of the transcriptome array is designed to bind within 300 nucleotides of the 3' end of a transcript. Methods for designing transcriptome arrays with probe sets that bind within 300 nucleotides of the 3' end of target transcripts are disclosed in United States Patent Application Publication No. 2009/0082218, which is incorporated by reference herein. In certain example embodiments, the microarray used in deriving the gene expression profiles of the present invention is the Almac Ovarian Cancer DSA™ microarray (Almac Group, Craigavon, United Kingdom).
[0097] An optimal linear classifier can be selected by evaluating a linear classifier's performance using such diagnostics as "area under the curve" (AUC). AUC refers to the area under the curve of a receiver operating characteristic (ROC) curve, both of which are well known in the art. AUC measures are useful for comparing the accuracy of a classifier across the complete data range. Linear classifiers with a higher AUC have a greater capacity to classify unknowns correctly between two groups of interest (e.g., ovarian cancer samples and normal or control samples). ROC curves are useful for plotting the performance of a particular feature (e.g., any of the biomarkers described herein and/or any item of additional biomedical information) in distinguishing between two populations (e.g., individuals responding and not responding to a therapeutic agent). Typically, the feature data across the entire population (e.g., the cases and controls) are sorted in ascending order based on the value of a single feature. Then, for each value for that feature, the true positive and false positive rates for the data are calculated. The true positive rate is determined by counting the number of cases above the value for that feature and then dividing by the total number of positive cases. The false positive rate is determined by counting the number of controls above the value for that feature and then dividing by the total number of controls. Although this definition refers to scenarios in which a feature is elevated in cases compared to controls, this definition also applies to scenarios in which a feature is lower in cases compared to the controls (in such a scenario, samples below the value for that feature would be counted). ROC curves can be generated for a single feature as well as for other single outputs, for example, a combination of two or more features can be mathematically combined (e.g., added, subtracted, multiplied, etc.) to provide a single sum value, and this single sum value can be plotted in a ROC curve. Additionally, any combination of multiple features, in which the combination derives a single output value, can be plotted in a ROC curve. These combinations of features may comprise a test. The ROC curve is the plot of the true positive rate (sensitivity) of a test against the false positive rate (1-specificity) of the test.
[0098] In one example embodiment an angiogenesis expression signature is directed to the 63 biomarkers detailed in Table 2 with corresponding ranks, and weights and associated bias detailed in the table or alternative rankings, and weightings and bias, depending, for example, on the disease setting. Table 2 ranks the biomarkers in order of absolute decreasing weight, in an example classifier, in the compound decision score function. The methods of the invention may rely upon measuring one or more, up to all, of the biomarkers listed in table 2. The methods of the invention may comprise measuring the expression levels of at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 40, 50, 60 or each of the biomarkers from Table 2. In certain embodiments the method may comprise measuring the expression levels of 2 to 5 of the biomarkers from Table 2.
Table 2
Figure imgf000042_0001
23 DEXI 0.007271 5.5913
24 BASP1 -0.00724 8.4396
25 PXDN -0.00724 7.7570
26 GBP4 0.007226 3.1119
27 SLC28A3 0.007201 4.2125
28 HLA-DRA 0.007197 8.3089
29 TAP2 0.007189 4.8464
30 ACSL5 0.007155 6.8703
31 CDH11 -0.00708 4.9925
32 PSMB9 0.006962 4.1122
33 MMP14 -0.00683 10.1689
34 CD74 0.006825 9.2707
35 LOXL1 -0.00676 9.6429
36 CIITA 0.006623 5.5396
37 ZNF697 -0.00658 7.0319
38 SH3RF2 0.006549 5.0029
39 MIR198 -0.00654 5.1935
40 COL1A2 -0.00645 6.0427
41 TNFRSF14 0.006421 9.0366
42 COL8A1 -0.00642 6.4565
43 C21orf63 0.006261 5.9811
44 TAP1 0.006215 8.6458
45 PDPN -0.00612 5.3198
46 RHOBTB3 -0.00597 3.5609
47 BCL11A 0.005943 4.3818
48 HLA-DOB 0.005851 4.6075
49 XAF1 0.005742 7.9229
50 ARHGAP26 0.005632 4.3991
51 POLD2 -0.00558 9.4183
52 DPYSL2 -0.00533 8.3469
53 COL4A1 -0.0052 7.0317
54 ID3 -0.00516 7.5673
55 CFB 0.005077 5.7951
56 NIDI -0.00494 4.7186
57 FKBP7 -0.00489 2.9437
58 TIMP2 -0.00468 7.5253
59 RCBTB1 -0.00458 7.4491
60 ANGPTL2 -0.00448 5.6807
61 ENTPD7 -0.00442 7.3772
62 SHISA4 -0.00403 6.0601
63 HINT1 0.003651 6.0724
[0099] In another example embodiment an angiogenesis expression signature is directed to the 63 biomarkers detailed in Table 3 with corresponding ranks detailed in the table or alternative rankings depending, for example, on the disease setting. Table 3 ranks the biomarkers in order of absolute decreasing weight, in an example classifier, in the compound decision score function. The methods of the invention may rely upon measuring one or more, up to all, of the biomarkers listed in table 3. The methods of the invention may comprise measuring the expression levels of at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 40, 50, 60 or each of the biomarkers from Table 3. In certain embodiments the methods may comprise measuring the expression levels of 2 to 5 of the biomarkers from Table
3.
[00100]
Table 3
Gene Total Delta HR Rank
IGF2 0.048910407 1
CD 1 0.045335288 2
COL3A1 0.044869217 3
SPARC 0.043434096 4
TIMP3 0.042053053 5
I NS 0.04013658 6
COL8A1 0.026780907 7
NUAK1 0.026752491 8
MATN3 0.02402318 9
TM EM45A 0.016999761 10
SRPX 0.016372168 11
CDH11 0.015604812 12
M M P14 0.014583388 13
LOXL1 0.010315358 14
PXDN 0.009728534 15
COL1A2 0.009267887 16
ANGPTL2 0.006071504 17
POLD2 0.004297935 18
NIDI 0.00408724 19
ISLR 0.003014488 20
SNORD114-14 0.002992636 21
CXCR2P1 0.002804432 22
M IR198 0.002173041 23
BCL11A 0.001258286 24
PDPN 0.000989109 25
TN FRSF14 0.000132838 26
ENTPD7 6.25143E-05 27
HI NT1 -0.000113156 28
TAP1 -0.000379242 29
ID3 -0.000452476 30
RCBTB1 -0.000695459 31
SOX11 -0.001068812 32
SHISA4 -0.001470801 33
COL4A1 -0.001714442 34
TU BA1A -0.001817696 35
TIMP2 -0.004079263 36
FKBP7 -0.004575097 37
TAP2 -0.004597761 38
TNFSF10 -0.005307314 39
ZN F697 -0.007733496 40 CIITA -0.008785689 41
BASP1 -0.009340492 42
XAF1 -0.009760794 43
DEXI -0.009798099 44
SH3 F2 -0.009856754 45
HLA-DOB -0.009987248 46
RHOBTB3 -0.010264542 47
GBP4 -0.010747831 48
DPYSL2 -0.012042179 49
ARHGAP26 -0.012380203 50
M FAP2 -0.013981916 51
CD74 -0.016415304 52
ACSL5 -0.016912224 53
SLC28A3 -0.016996213 54
GJ B1 -0.018395345 55
C21orf63 -0.019853038 56
PSM B9 -0.020314379 57
HLA-DRA -0.020436677 58
CFB -0.022202886 59
RARRES3 -0.034723666 60
CXCL17 -0.038523986 61
SLC5A1 -0.042034346 62
RTP4 -0.045259104 63
Probesets that can be used to measure the expression of the biomarkers are shown
Table 4
Figure imgf000045_0001
OC S2.9952_x_at BASP1 509
OCRS.854_s_at BCL11A 510
OC3P.14938.Cl_s_at BCL11A 511
OCMX.12290.Cl_at BCL11A 512
OCADA.10230_s_at BCL11A 513
OC3SNGnh.4343_at BCL11A 514
OC3SNGnh.l6766_x_at BCL11A 515
OCMX.1680.Cl_s_at BCL11A 516
OC3P.14938.Cl-334a_s_at BCL11A 517
OCMX.12290.Cl_x_at BCL11A 518
OCADA.2850_s_at BCL11A 519
OCADA.1135_s_at C21orf63 520
OCMX.14248.Cl_s_at C21orf63 521
OC3P.14091.Cl_s_at C21orf63 522
OC3P.14431.Cl_s_at C21orf63 523
OCADA.8368_x_at CD74 524
OC3SNGnh.l9144_s_at CD74 525
OC3P.104.CBl_x_at CD74 526
OCADN P.1805_s_at CD74 527
OC3SNG.3064-21a_x_at CD74 528
OC3P.14147.Cl_s_at CDH11 529
OCADNP.10024_s_at CDH11 530
OCHP.148_s_at CDH11 531
OCADA.6210_s_at CDH11 532
OC3SNGnh.5056_x_at CDH11 533
OC3SNGnh.4032_s_at CDH11 534
OCHPRC.58_s_at CDH11 535
OCMX.1718.Cl_s_at CDH11 536
OCADA.8067_x_at CDH11 537
OCADN P.8007_s_at CDR1 538
OC3P.295.Cl_s_at CFB 539
ADXStrongB56_at CFB N/A
OC3P.295.C2_x_at CFB 540
OC3SNGnh.l4167_at CFB 541
OC3SNGn.5914-165a_s_at CFB 542
OC3SNGn.970-10a_s_at CFB 543
OCADN P.9683_s_at CFB 544
OC3P.295.C2_at CFB 545
OC3SNGnh.l4167_s_at CFB 546
OCADNP.17538_s_at CIITA 547
OC3P.805.Cl_s_at CIITA 548
OCEM .1780_s_at CIITA 549
OC3SNGnh.l6892_s_at CIITA 550
OCADA.6540_s_at CIITA 551
OCH P.1927_s_at CIITA 552
OC3SNGn.354-123a_s_at CIITA 553
OC3SNGnh.4794_at CIITA 554
OC3SNGn.8474-50a_x_at C0L1A2 555
OCMX.184.Cll_s_at C0L1A2 556
OC3SNG.115-2502a_at C0L1A2 557
OC3SNG.116-9169a_s_at C0L1A2 558
OC3P.60.CB2_x_at C0L1A2 559 OC3P.6454.Cl_s_at COL1A2 560
OC3SNG.115-2502a_x_at COL1A2 561
OCMX.184.C16_x_at COL1A2 562
OCHP.173_x_at COL1A2 563
OC3P.60.CBl_x_at COL1A2 564
OC3SNGn.2538-539a_x_at COL1A2 565
OCMX.184.C16_s_at COL1A2 566
OCADNP.4048_s_at COL3A1 567
OC3P.81.CB2_s_at COL3A1 568
OC3SNGnh.l9127_s_at COL3A1 569
OC3SNGn.l211-6a_s_at COL3A1 570
OCADNP.11975_s_at COL4A1 571
OC3P.850.Cl-1145a_s_at COL4A1 572
OCHP C.29_s_at COL4A1 573
OC3SNGnh.276_x_at COL4A1 574
OC3SNGnh.l8844_at COL8A1 575
OC3P.1087.Cl_s_at COL8A1 576
OC3P.13652.Cl_s_at COL8A1 577
OCADNP.14932_s_at COL8A1 578
OC3P.10562.Cl_s_at COL8A1 579
OCHPRC.94_s_at CXCL17 580
OC3SNG.3604-23a_at CXCR2P1 581
OC3SNG.3604-23a_x_at CXCR2P1 582
OC3SNGnh.l3095_at DEXI 583
OC3P.7366.Cl_s_at DEXI 584
OCADA.2531_s_at DEXI 585
OC3SNGnh.3527_at DEXI 586
OC3P.10489.Cl_s_at DEXI 587
OCADNP.10600_s_at DEXI 588
OCADA.1911_s_at DPYSL2 589
OC3P.7322.Cl_s_at DPYSL2 590
OC3SNG.366-35a_s_at ENTPD7 591
OC3SNGnh.5644_s_at FKBP7 592
OC3SNGnh.l7831_at FKBP7 593
OCADNP.7326_s_at FKBP7 594
OC3P.12003.Cl_x_at FKBP7 595
OC3P.4378.Cl_s_at GBP4 596
OC3SNGnh.5459_s_at GBP4 597
OCADNP.3694_s_at GBP4 598
OC3SNG.3671-13a_s_at GJB1 599
2874688_at HINT1 N/A
2874689_at HINT1 N/A
Adx-200093_s_at HINT1 600
OC3SNGnh.5235_x_at HINT1 601
2874702_at HINT1 N/A
2874727_at HINT1 N/A
200093_s_at HINT1 602
2874697_at HINT1 N/A
2874725_at HINT1 N/A
2874696_at HINT1 N/A
2874737_at HINT1 N/A
2874735_at HINT1 N/A Adx-200093-up_s_at HINT1 603
OC3P.14829.Cl_s_at HLA-DOB 604
ADXBad55_at HLA-DOB N/A
OC3P.674.Cl_s_at HLA-DRA 605
OCADNP.8307_s_at HLA-DRA 606
OC3P.2407.Cl_s_at ID3 607
ADXGoodl00_at IGF2 N/A
OC3SNG.899-20a_s_at IGF2 608
OC3SNGn.5728-103a_x_at IGF2 610
OC3P.4645.Cl_s_at IGF2 613
OC3SNGnh.l9773_s_at IGF2 614
OCADNP.10122_s_at IGF2 615
OCADNP.7400_s_at IGF2 616
ADXGoodl00_at INS N/A
OCADNP.17017_s_at INS 609
OC3SNGn.5728-103a_x_at INS 610
OCEM.2174_s_at INS 611
OCEM.2035_x_at INS 612
OC3P.4645.Cl_s_at INS 613
OC3SNGnh.l9773_s_at INS 614
OCADNP.10122_s_at INS 615
OCADNP.7400_s_at INS 616
OCEM.2035_at INS 617
OC3P.9976.Cl_x_at ISLR 618
OCHP.1306_s_at LOXL1 619
OCADA.10621_s_at MATN3 620
OC3P.2576.Cl_x_at MFAP2 621
OCHP.1079_s_at MFAP2 622
OC3P.11139.Cl_s_at MIR198 623
OC3P.211.Cl_x_at MIR198 624
ADXBad7_at MIR198 N/A
OCHP.462_s_at MIR198 625
OC3SNGn.8954-766a_s_at MIR198 626
OCADNP.4997_s_at MIR198 627
OCHP.228_s_at MMP14 628
OC3P.4123.Cl_x_at MMP14 629
OC3P.4123.Cl_s_at MMP14 630
OCADA.1433_x_at NIDI 631
OCADNP.7347_s_at NIDI 632
OC3P.3404.Cl_s_at NIDI 633
OC3SNGn.3328-664a_s_at NIDI 634
OCADNP.9225_s_at NUAK1 635
ADXStrongB87_at NUAK1 N/A
OC3SNGn.2676-391a_s_at NUAK1 636
OCHPRC.lll_s_at PDPN 637
OCADNP.10047_s_at PDPN 638
OCHPRC.96_s_at PDPN 639
OC3P.13523.Cl_s_at PDPN 640
OC3SNG.4571-22a_x_at POLD2 641
OCEM.1126_s_at POLD2 642
ADXGood4_at POLD2 N/A OC3SNGn.890-5a_s_at POLD2 643
OC3P.14770.Cl_s_at PSMB9 644
OC S.920_s_at PSMB9 645
OC3P.4627.Cl_s_at PSMB9 646
OC3SNGnh.8187_at PSMB9 647
OCMX.15283.Cl_x_at PSMB9 648
OCADNP.804_s_at PSMB9 649
OC3SNGnh.8187_x_at PSMB9 650
OCMX.14440.Cl_x_at PSMB9 651
OC3P.1307.Cl_s_at PXDN 652
OC3P.8838.Cl_s_at PXDN 653
OCHP.1891_s_at RARRES3 654
OC3P.8963.Cl_s_at RCBTB1 655
OC3SNGnh.6721_x_at RH0BTB3 656
OC3SNGnh.6912_x_at RH0BTB3 657
OC3SNGnh.957_s_at RH0BTB3 658
OC3SNG.2402-2883a_s_at RH0BTB3 659
OCHPRC.1436_at RH0BTB3 660
OC3SNGn.5382-76a_s_at RH0BTB3 661
OC3SNGnh.957_x_at RH0BTB3 662
OC3SNGnh.957_at RH0BTB3 663
OC3P.12862.Cl_s_at RH0BTB3 664
OC3SNG.2401-1265a_x_at RH0BTB3 665
OC3P.5737.Cl_s_at RH0BTB3 666
OCHP.1722_s_at RTP4 667
OC3P.9552.Cl-496a_s_at RTP4 668
OC3P.9552.Cl_x_at RTP4 669
OC3P.9552.Cl_at RTP4 670
OC3SNGnh.865_s_at SH3RF2 671
OC3SNGnh.l6695_s_at SH3RF2 672
OCADNP.12161_s_at SH3RF2 673
OC3SNGn.439-184a_s_at SH3RF2 674
OCHPRC.86_s_at SH3RF2 675
OCADNP.2340_s_at SHISA4 676
OC3SNG.6118-43a_s_at SHISA4 677
OCADNP.8940_s_at SLC28A3 678
OC3SNGnh.971_s_at SLC28A3 679
OCADA.4025_s_at SLC28A3 680
OC3P.9666.Cl_s_at SLC28A3 681
OC3P.5726.Cl_s_at SLC5A1 682
OCADNP.7872_s_at SLC5A1 683
OCRS2.10331_x_at SNORD114-14 684
OCRS2.8538_x_at SNORD114-14 685
OCRS2.10331_at SNORD114-14 686
OC3SNGn.2110-23a_s_at S0X11 687
OCHP.1171_s_at SOX11 688
OCHP.1523_s_at SOX11 689
OC3SNGnh.l9157_x_at SPARC 690
OCHP.508_s_at SPARC 691
OC3P.148.CBl-990a_s_at SPARC 692
OCEM.2143_at SPARC 693
OC3SNG.2614-40a_s_at SPARC 694 OC3P.148.CBl_x_at SPARC 695
OCEM.2143_x_at SPARC 696
OC3SNG.1657-20a_s_at SRPX 697
ADXGoodB4_at TAP1 N/A
OC3SNG.2665-23a_s_at TAP1 698
OC3P.5602.Cl_s_at TAP2 699
OCADNP.2260_s_at TAP2 700
OCADNP.8242_s_at TAP2 701
OC3SNGnh.l8127_s_at TAP2 702
OC3P.14195.Cl_s_at TIMP2 703
OCHP.320_s_at TIMP2 704
OC3P.543.CBl_x_at TI P2 705
OC3SNGnh.l9238_s_at TI P2 706
OC3P.543.CBl-699a_s_at TI P2 707
OCADNP.14191_s_at TI P2 708
OCADNP.13017_s_at TIMP3 709
OCADA.9324_s_at TIMP3 710
OCHP.1200_s_at TIMP3 711
ADXGood73_at TIMP3 N/A
OC3P.10470.Cl_s_at TIMP3 712
OC3P.15327.Cl_at TIMP3 713
OCHP.112_s_at TIMP3 714
OC3P.5348.Cl_s_at TMEM45A 715
OC3P.4028.Cl_at TNFRSF14 716
OC3SNGn.2230-103a_s_at TNFRSF14 717
OC3P.4028.Cl_x_at TNFRSF14 718
OC3SNG.1683-90a_s_at TNFSFIO 719
OC3P.2087.Cl_s_at TNFSFIO 720
OCHP.318_x_at TNFSFIO 721
OC3SNGn.6279-343a_s_at TNFSFIO 722
OC3SNGn.5842-826a_x_at TNFSFIO 723
OCADNP.9180_s_at TNFSFIO 724
OCHP.1136_s_at TUBA1A 725
OCADNP.7771_s_at XAF1 726
ADXStrongB9_at XAF1 N/A
OC3SNG.2606-619a_x_at XAF1 727
OC3SNGnh.l0895_at XAF1 728
OC3P.4873.Cl_s_at XAF1 729
OC3SNGnh.l0895_x_at XAF1 730
OC3SNG.2605-236a_x_at XAF1 731
OC3SNG.5460-81a_x_at XAF1 732
OCADA.154_s_at ZNF697 733
OCADA.3112_s_at ZNF697 734
[00102] In one example embodiment, an expression signature comprises all or a portion of the following biomarkers; IGF2, SOXll, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, RTP4, COL3A1, CDRl, RARRES3, TNFSFIO, NUAKl, SNORD114-14, SRPX, SPARC, GJB1, ΤΓΜΡ3, ISLR, TUBA1A, DEXI, BASP1, PXDN, GBP4, SLC28A3, HLA-DRA, TAP2, ACSL5, CDHll, PSMB9, MMP14, CD74, LOXLl, CIITA, ZNF697, SH3RF2, MIR198, COL1A2, TNFRSF14, COL8A1, C21orf63, TAP1, PDPN, RHOBTB3, BCL11A, HLA-DOB, XAF1, ARHGAP26, POLD2, DPYSL2, COL4A1, ID3, CFB, NIDI, FKBP7, TIMP2, RCBTB 1, ANGPTL2, ENTPD7, SHISA4, and HINT1,
[00103] In another example embodiment, an expression signature comprises IGF2, SOX11, INS, and CXCL17 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,
18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43,
44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
[00104] In another example embodiment, an expression signature comprises IGF2, INS, SPARC, TMEM45A, COL8A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44,
45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
[00105] In another example embodiment, an expression signature comprises IGF2, INS, SPARC, TMEM45A, COL8A1, COL3A1, CDRl, NUAKl, TIMP3, LOXL1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53.
[00106] In another example embodiment, an expression signature comprises IGF2, TIMP3, INS, CXCR2P1, NUAKl and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
[00107] In another example embodiment, an expression signature comprises IGF2, TIMP3, INS, CXCR2P1, NUAKl, CDRl, MATN3, SOX11, SNORD114-14, COL3A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53.
[00108] In another example embodiment, an expression signature comprises COL3A1, SPARC, CDRl, SRPX, MATN3 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59. [00109] In another example embodiment, an expression signature comprises COL3A1, SPARC, CDR1, SRPX, MATN3, TIMP3, CDH11, COL8A1, BCL11A, MMP14 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53.
[00110] In another example embodiment, an expression signature comprises IGF2, CDR1, COL3A1, SPARC, TIMP3 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
[00111] In another example embodiment, an expression signature comprises IGF2, CDR1, COL3A1, SPARC, TIMP3, INS, COL8A1, NUAKl, MATN3, TMEM45A and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
[00112] In another example embodiment, an expression signature comprises INS, SPARC, COL8A1, COL3A1, CDR1, NUAKl, TIMP3, and MMP14 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equalsl, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, or 55.
[00113] In another example embodiment, an expression signature comprises at least INS, SPARC, COL8A1, COL3A1, CDR1, NUAKl, TIMP3, and MMP14. In another example embodiment, the expression signature comprises at least IGF2, CDR1, COL3A1, SPARC, TIMP3, INS, COL8A1, NUAKl, MATN3, TMEM45A. In another example embodiment, the expression signature comprises at least IGF2, CDR1, COL3A1, SPARC, TIMP3. In another example embodiment, the expression signature comprises at least, COL3A1, SPARC, CDR1, SRPX, MATN3, TIMP3, CDH11, COL8A1, BCL11A, MMP14. In another example embodiment, the expression signature comprises at least COL3A1, SPARC, CDR1, SRPX, MATN3. In another example embodiment, the expression signature comprises at least COL3A1, SPARC, CDR1, SRPX, MATN3. In another example embodiment, the expression signature comprises at least IGF2, TIMP3, INS, CXCR2P1, NUAKl, CDR1, MATN3, SOX11, SNORD114-14, COL3A1. In another example embodiment, the expression signature comprises at least IGF2, TIMP3, INS, CXCR2P1, NUAKl. In another example embodiment, the expression signature comprises at least IGF2, INS, SPARC, TMEM45A, COL8A1, COL3A1, CDRl, NUAKl, TIMP3, LOXLl. In another example embodiment, the expression signature comprises at least IGF2, INS, SPARC, TMEM45A, COL8A1. In another example embodiment, the expression signature comprises at least IGF2, SOX11, INS, and CXCL17.
[00114] In another example embodiment, an expression signature comprises IGF2 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[00115] In another example embodiment, an expression signature comprises SOX11 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[00116] In another example embodiment, an expression signature comprises INS and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[00117] In another example embodiment, an expression signature comprises CXCL17 and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[00118] In another example embodiment, an expression signature comprises CDRl and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[00119] In another example embodiment, an expression signature comprises COL3A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62. [00120] In another example embodiment, an expression signature comprises SPARC and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[00121] In another example embodiment, an expression signature comprises TIMP3 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[00122] In another example embodiment, an expression signature comprises COL8A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[00123] In another example embodiment, an expression signature comprises NUAK1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[00124] In another example embodiment, an expression signature comprises MATN3 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0100] In another example embodiment, an expression signature comprises TMEM45A and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0101] In another example embodiment, an expression signature comprises CXCR2P1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0102] In another example embodiment, an expression signature comprises SRPX and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0103] In another example embodiment, an expression signature comprises CDH11 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0104] In another example embodiment, an expression signature comprises BC11A and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0105] In another example embodiment, an expression signature comprises LOXL1 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0106] In another example embodiment, an expression signature comprises MMP14 and at least N additional biomarkers selected from the list of biomarkers in Table 2, and at least N additional biomarkers selected from the list of biomarkers in Table 2, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0107] In another example embodiment, an example expression signature comprises the biomarkers and corresponding biomarker weighted values listed in Table 2. In another example embodiment, an example expression signature consists of the biomarkers and corresponding biomarker weighted values listed in Table 2. In another example embodiment, an example expression score comprises the biomarkers and rank listed in Table 3. In another example embodiment, an example expression signature consists of the biomarkers and corresponding ranks listed in Tabl3.
[0108] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising all or a portion of the biomarkers listed in Table 2 or 3.
[0109] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising IGF2, SOXl 1, INS, and CXCL17 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
[0110] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising IGF2, INS, SPARC, TMEM45A, COL8A1, and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
[0111] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising IGF2, INS, SPARC, TMEM45A, COL8A1, COL3A1, CDRl, NUAKl, TIMP3, LOXLl and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53. [0112] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising IGF2, TIMP3, INS, CXCR2P1, NUAKl and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
[0113] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising IGF2, TIMP3, INS, CXCR2P1, NUAKl, CDRl, MATN3, SOX11, SNORD114-14, COL3A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53.
[0114] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising COL3A1, SPARC, CDRl, SRPX, MATN3, and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59.
[0115] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising COL3A1, SPARC, CDRl, SRPX, MATN3, ΤΓΜΡ3, CDHl l, COL8A1, BCLl lA, MMP14 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53.
[0116] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising IGF2, CDRl, COL3A1, SPARC, ΤΓΜΡ3, and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59. [0117] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising IGF2, CDRl, COL3A1, SPARC, TIMP3, INS, COL8A1, NUAKl, MATN3, TMEM45A and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or 53.
[0118] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising INS, SPARC, COL8A1, COL3A1, CDRl, NUAKl, TIMP3, and MMP14 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, or 55.
[0119] In another example embodiment, an biomarker panel comprises at least INS, SPARC, COL8A1, COL3A1, CDRl, NUAKl, TIMP3, and MMP14. In another example embodiment, the biomarker panel comprises at least IGF2, CDRl, COL3A1, SPARC, TTMP3, INS, COL8A1, NUAKl, MATN3, TMEM45A. In another example embodiment, the biomarker panel comprises at least IGF2, CDRl, COL3A1, SPARC, TTMP3. In another example embodiment, the biomarker panel comprises at least, COL3A1, SPARC, CDRl, SRPX, MATN3, TIMP3, CDH11, COL8A1, BCL11A, MMP14. In another example embodiment, the biomarker panel comprises at least COL3A1, SPARC, CDRl, SRPX, MATN3. In another example embodiment, the biomarker panel comprises at least COL3A1, SPARC, CDRl, SRPX, MATN3. In another example embodiment, the biomarker panel comprises at least IGF2, ΤΓΜΡ3, INS, CXCR2P1, NUAKl, CDRl, MATN3, SOX11, SNORD114-14, COL3A1. In another example embodiment, the biomarker panel comprises at least IGF2, TIMP3, INS, CXCR2P1, NUAKl. In another example embodiment, the biomarker panel comprises at least IGF2, INS, SPARC, TMEM45A, COL8A1, COL3A1, CDRl, NUAKl, TTMP3, LOXLl. In another example embodiment, the biomarker panel comprises at least IGF2, INS, SPARC, TMEM45A, COL8A1. In another example embodiment, the biomarker panel comprises at least IGF2, SOX11, INS, and CXCL17.
[0120] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising IGF2 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0121] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising SOXl l and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0122] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising INS and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0123] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising CXCL17 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0124] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising SPARC and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0125] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising TMEM45A and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0126] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising COL8A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1,
2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0127] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising COL3A1 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0128] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising CDR1 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0129] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising NUAK1 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0130] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising ΤΓΜΡ3 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0131] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising LOXL1 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0132] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising CXCR2P1 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0133] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising SPARC and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1,
2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0134] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising MATN3 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0135] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising SNORDl 14-14 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0136] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising SRPX and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0137] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising CDH11 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0138] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising BCl lA and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0139] In a further aspect, the methods of the present invention comprise conducting an assay on a biological sample from an individual to determine the expression levels of one or more biomarkers in a biomarker panel, the biomarker panel comprising MMP14 and at least N additional biomarkers selected from the list of biomarkers in Table 2 or 3, wherein N equals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, or 62.
[0140] Classifying New Test Samples using an Expression Signature
[0141] To classify new test samples using an expression signature, such as those described above, the relative expression levels of one or more biomarkers in a cancer tissue are measured to form a test sample expression profile. In certain example embodiments, the test sample expression profile is summarized in the form of a compound decision score ("expression score") and compared to a threshold score that is mathematically derived from a training set of patient data. The score threshold is established with the purpose of maximizing the ability to separate a patient group into different groups based on characteristics such as, but not limited to, good/poor clinical prognosis. The patient training set data is preferably derived from cancer tissue samples having been characterized by prognosis, likelihood of recurrence, long term survival, clinical outcome, treatment response, diagnosis, cancer classification, or personalized genomics profile. Expression profiles, and corresponding decision scores from patient samples may be correlated with the characteristics of patient samples in the training set that are on the same side of the mathematically derived score decision threshold. In certain example embodiments, the threshold of the linear classifier scalar output is optimized to maximize the sum of sensitivity and specificity under cross-validation as observed within the training dataset.
[0142] The overall expression data for a given sample is normalized using methods known to those skilled in the art in order to correct for differing amounts of starting material, varying efficiencies of the extraction and amplification reactions, etc.
[0143] In one embodiment, the biomarker expression profile of a patient tissue sample is evaluated by a linear classifier. As used herein, a linear classifier refers to a weighted sum of the individual biomarker intensities into a compound decision score ("decision function"). The decision score is then compared to a pre-defined cut-off score threshold, corresponding to a certain set-point in terms of sensitivity and specificity which indicates if a sample is equal to or above the score threshold (decision function positive) or below (decision function negative).
[0144] Using a linear classifier on the normalized data to make a diagnostic or prognostic call (e.g. good or poor clinical prognosis) effectively means to split the data space, i.e. all possible combinations of expression values for all genes in the classifier, into two disjoint segments by means of a separating hyperplane. This split is empirically derived on a large set of training examples, for example from patients showing responsiveness or resistance to a therapeutic agent. Without loss of generality, one can assume a certain fixed set of values for all but one biomarker, which would automatically define a threshold value for this remaining biomarker where the decision would change from, for example, responsiveness or resistance to a therapeutic agent. Expression values above this dynamic threshold would then either indicate a poor clinical prognosis (for a biomarker with a negative weight) or a good clinical prognosis (for a biomarker with a positive weight). The precise value of this threshold depends on the actual measured expression profile of all other biomarkers within the classifier, but the general indication of certain biomarkers remains fixed, i.e. high values or "relative over-expression" always contributes to either a good clinical prognosis (genes with a positive weight) or a poor clinical prognosis (genes with a negative weights). Therefore, in the context of the overall gene expression classifier, relative expression can indicate if either up- or down-regulation of a certain biomarker is indicative of a good or bad clinical prognosis. In certain example embodiments, a sample expression score above the threshold expression score indicates the subject has the non-angiogenesis subtype. In certain other example embodiments, a sample expression score above a threshold score indicates the subject has a good clinical prognosis compared to a subject with a sample expression score below the threshold score. In certain other example embodiments, a sample expression score above the threshold score indicates the subject will likely experience a detrimental effect, or have a poor prognosis, if an anti- angiogenic therapeutic agent is administered.
[0145] There are a number of suitable methods for measuring expression profiles of test samples depending on the type of biomarker to be assayed. Measuring mRNA in a biological sample may be used as a surrogate for detection of the level of the corresponding protein in the biological sample. Thus, any of the biomarkers or biomarker panels described herein can also be detected by detecting the appropriate RNA. Methods of gene expression profiling include, but are not limited to, microarray, RT-PCT, qPCR, NGS, northern blots, SAGE, mass spectrometry.
[0146] mRNA expression levels are measured by reverse transcription quantitative polymerase chain reaction (RT-PCR followed with qPCR). RT-PCR is used to create a cDNA from the mRNA. The cDNA may be used in a qPCR assay to produce fluorescence as the DNA amplification process progresses. By comparison to a standard curve, qPCR can produce an absolute measurement such as number of copies of mRNA per cell. Northern blots, microarrays, Invader assays, and RT-PCR combined with capillary electrophoresis have all been used to measure expression levels of mRNA in a sample. See Gene Expression Profiling: Methods and Protocols, Richard A. Shimkets, editor, Humana Press, 2004.
[0147] miRNA molecules are small RNAs that are non-coding but may regulate gene expression. Any of the methods suited to the measurement of mRNA expression levels can also be used for the corresponding miRNA. Recently many laboratories have investigated the use of miRNAs as biomarkers for disease. Many diseases involve widespread transcriptional regulation, and it is not surprising that miRNAs might find a role as biomarkers. The connection between miRNA concentrations and disease is often even less clear than the connections between protein levels and disease, yet the value of miRNA biomarkers might be substantial. Of course, as with any RNA expressed differentially during disease, the problems facing the development of an in vitro diagnostic product will include the requirement that the miRNAs survive in the diseased cell and are easily extracted for analysis, or that the miRNAs are released into blood or other matrices where they must survive long enough to be measured. Protein biomarker s have similar requirements, although many potential protein biomarkers are secreted intentionally at the site of pathology and function, during disease, in a paracrine fashion. Many potential protein biomarkers are designed to function outside the cells within which those proteins are synthesized.
[0148] Gene expression may also be evaluated using mass spectrometry methods. A variety of configurations of mass spectrometers can be used to detect biomarker values. Several types of mass spectrometers are available or can be produced with various configurations. In general, a mass spectrometer has the following major components: a sample inlet, an ion source, a mass analyzer, a detector, a vacuum system, and instrument-control system, and a data system. Difference in the sample inlet, ion source, and mass analyzer generally define the type of instrument and its capabilities. For example, an inlet can be a capillary-column liquid chromatography source or can be a direct probe or stage such as used in matrix-assisted laser desorption. Common ion sources are, for example, electrospray, including nanospray and microspray or matrix-assisted laser desorption. Common mass analyzers include a quadrupole mass filter, ion trap mass analyzer and time-of-flight mass analyzer. Additional mass spectrometry methods are well known in the art (see Burlingame et al., Anal. Chem. 70:647 R- 716R (1998); Kinter and Sherman, New York (2000)).
[0149] Protein biomarkers and biomarker values can be detected and measured by any of the following: electrospray ionization mass spectrometry (ESI-MS), ESI-MS/MS, ESI-MS/(MS)n, matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS), surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF- MS), desorption/ionization on silicon (DIOS), secondary ion mass spectrometry (SIMS), quadrupole time-of-flight (Q-TOF), tandem time-of-flight (TOF/TOF) technology, called ultraflex III TOF/TOF, atmospheric pressure chemical ionization mass spectrometry (APCI- MS), APCI-MS/MS, APCI-(MS).sup.N, atmospheric pressure photoionization mass spectrometry (APPI-MS), APPI-MS/MS, and APPI-(MS).sup.N, quadrupole mass spectrometry, Fourier transform mass spectrometry (FTMS), quantitative mass spectrometry, and ion trap mass spectrometry.
[0150] Sample preparation strategies are used to label and enrich samples before mass spectroscopic characterization of protein biomarkers and determination biomarker values. Labeling methods include but are not limited to isobaric tag for relative and absolute quantitation (iTRAQ) and stable isotope labeling with amino acids in cell culture (SILAC). Capture reagents used to selectively enrich samples for candidate biomarker proteins prior to mass spectroscopic analysis include but are not limited to aptamers, antibodies, nucleic acid probes, chimeras, small molecules, an F(ab')2 fragment, a single chain antibody fragment, an Fv fragment, a single chain Fv fragment, a nucleic acid, a lectin, a ligand-binding receptor, affybodies, nanobodies, ankyrins, domain antibodies, alternative antibody scaffolds (e.g. diabodies etc) imprinted polymers, avimers, peptidomimetics, peptoids, peptide nucleic acids, threose nucleic acid, a hormone receptor, a cytokine receptor, and synthetic receptors, and modifications and fragments of these.
[0151] The foregoing assays enable the detection of biomarker values that are useful in methods for determining a patient's clinical prognosis and selecting appropriate treatment regimens, where the methods comprise detecting, in a biological sample from an individual, at least N biomarker values that each correspond to a biomarker selected from the group consisting of the biomarkers provided in Tables 1 or Table 2, wherein a classification, as described in detail below, using the biomarker values indicates whether the individual has a good prognosis or a bad progonosis, or will receive a detrimental or beneficial effect if a certain therapeutic agent is administered. While some of the described predictive biomarkers are useful alone for predicting clinical prognosis, methods are also described herein for the grouping of multiple subsets of the biomarkers that are each useful as a panel of two or more biomarkers. Thus, various embodiments of the instant application provide combinations comprising N biomarkers, wherein N is at least three biomarkers. It will be appreciated that N can be selected to be any number from any of the above-described ranges, as well as similar, but higher order, ranges. In accordance with any of the methods described herein, biomarker values can be detected and classified individually or they can be detected and classified collectively, as for example in a multiplex assay format.
b) Microarray methods
[0152] In one embodiment, the present invention makes use of "oligonucleotide arrays" (also called herein "microarrays"). Microarrays can be employed for analyzing the expression of biomarkers in a cell, and especially for measuring the expression of biomarkers of cancer tissues.
[0153] In one embodiment, biomarker arrays are produced by hybridizing detectably labeled polynucleotides representing the mRNA transcripts present in a cell (e.g., fluorescently-labeled cDNA synthesized from total cell mRNA or labeled cRNA) to a microarray. A microarray is a surface with an ordered array of binding (e.g., hybridization) sites for products of many of the genes in the genome of a cell or organism, preferably most or almost all of the genes. Microarrays can be made in a number of ways known in the art. However produced, microarrays share certain characteristics. The arrays are reproducible, allowing multiple copies of a given array to be produced and easily compared with each other. Preferably the microarrays are small, usually smaller than 5 cm2, and they are made from materials that are stable under binding (e.g., nucleic acid hybridization) conditions. A given binding site or unique set of binding sites in the microarray will specifically bind the product of a single gene in the cell. In a specific embodiment, positionally addressable arrays containing affixed nucleic acids of known sequence at each location are used.
[0154] It will be appreciated that when cDNA complementary to the RNA of a cell is made and hybridized to a microarray under suitable hybridization conditions, the level of hybridization to the site in the array corresponding to any particular gene will reflect the prevalence in the cell of mRNA transcribed from that gene/biomarker. For example, when detectably labeled (e.g., with a fluorophore) cDNA or cRNA complementary to the total cellular mRNA is hybridized to a microarray, the site on the array corresponding to a gene (i.e., capable of specifically binding the product of the gene) that is not transcribed in the cell will have little or no signal (e.g., fluorescent signal), and a gene for which the encoded mRNA is prevalent will have a relatively strong signal. Nucleic acid hybridization and wash conditions are chosen so that the probe "specifically binds" or "specifically hybridizes' to a specific array site, i.e., the probe hybridizes, duplexes or binds to a sequence array site with a complementary nucleic acid sequence but does not hybridize to a site with a non-complementary nucleic acid sequence. As used herein, one polynucleotide sequence is considered complementary to another when, if the shorter of the polynucleotides is less than or equal to 25 bases, there are no mismatches using standard base-pairing rules or, if the shorter of the polynucleotides is longer than 25 bases, there is no more than a 5% mismatch. Preferably, the polynucleotides are perfectly complementary (no mismatches). It can be demonstrated that specific hybridization conditions result in specific hybridization by carrying out a hybridization assay including negative controls using routine experimentation .
[0155] Optimal hybridization conditions will depend on the length (e.g., oligomer vs. polynucleotide greater than 200 bases) and type (e.g., RNA, DNA, PNA) of labeled probe and immobilized polynucleotide or oligonucleotide. General parameters for specific (i.e., stringent) hybridization conditions for nucleic acids are described in Sambrook et ah, supra, and in Ausubel et ah, "Current Protocols in Molecular Biology", Greene Publishing and Wiley- interscience, NY (1987), which is incorporated in its entirety for all purposes. When the cDNA microarrays are used, typical hybridization conditions are hybridization in 5xSSC plus 0.2% SDS at 65C for 4 hours followed by washes at 25 ^ in low stringency wash buffer (lxSSC plus 0.2% SDS) followed by 10 minutes at 25 °C in high stringency wash buffer (0.1SSC plus 0.2% SDS) (see Shena et al, Proc. Natl. Acad. Sci. USA, Vol. 93, p. 10614 (1996)). Useful hybridization conditions are also provided in, e.g., Tijessen, Hybridization With Nucleic Acid Probes", Elsevier Science Publishers B.V. (1993) and Kricka, "Nonisotopic DNA Probe Techniques", Academic Press, San Diego, Calif. (1992).
c) Immunoassay methods
[0156] Immunoassay methods are based on the reaction of an antibody to its corresponding target or analyte and can detect the analyte in a sample depending on the specific assay format. To improve specificity and sensitivity of an assay method based on immunoreactivity, monoclonal antibodies are often used because of their specific epitope recognition. Polyclonal antibodies have also been successfully used in various immunoassays because of their increased affinity for the target as compared to monoclonal antibodies. Immunoassays have been designed for use with a wide range of biological sample matrices. Immunoassay formats have been designed to provide qualitative, semi-quantitative, and quantitative results.
[0157] Quantitative results may be generated through the use of a standard curve created with known concentrations of the specific analyte to be detected. The response or signal from an unknown sample is plotted onto the standard curve, and a quantity or value corresponding to the target in the unknown sample is established.
[0158] Numerous immunoassay formats have been designed. ELISA or EIA can be quantitative for the detection of an analyte/biomarker. This method relies on attachment of a label to either the analyte or the antibody and the label component includes, either directly or indirectly, an enzyme. ELISA tests may be formatted for direct, indirect, competitive, or sandwich detection of the analyte. Other methods rely on labels such as, for example, radioisotopes (I 125 ) or fluorescence. Additional techniques include, for example, agglutination, nephelometry, turbidimetry, Western blot, immunoprecipitation, immunocytochemistry, immunohistochemistry, flow cytometry, Luminex assay, and others (see ImmunoAssay: A Practical Guide, edited by Brian Law, published by Taylor & Francis, Ltd., 2005 edition).
[0159] Example assay formats include enzyme-linked immunosorbent assay (ELISA), radioimmunoassay, fluorescent, chemiluminescence, and fluorescence resonance energy transfer (FRET) or time resolved-FRET (TR-FRET) immunoassays. Examples of procedures for detecting biomarkers include biomarker immunoprecipitation followed by quantitative methods that allow size and peptide level discrimination, such as gel electrophoresis, capillary electrophoresis, planar electrochromatography, and the like.
[0160] Methods of detecting and/or quantifying a detectable label or signal generating material depend on the nature of the label. The products of reactions catalyzed by appropriate enzymes (where the detectable label is an enzyme; see above) can be, without limitation, fluorescent, luminescent, or radioactive or they may absorb visible or ultraviolet light. Examples of detectors suitable for detecting such detectable labels include, without limitation, x- ray film, radioactivity counters, scintillation counters, spectrophotometers, colorimeters, fluorometers, luminometers, and densitometers.
[0161] Any of the methods for detection can be performed in any format that allows for any suitable preparation, processing, and analysis of the reactions. This can be, for example, in multi-well assay plates (e.g., 96 wells or 384 wells) or using any suitable array or microarray. Stock solutions for various agents can be made manually or robotically, and all subsequent pipetting, diluting, mixing, distribution, washing, incubating, sample readout, data collection and analysis can be done robotically using commercially available analysis software, robotics, and detection instrumentation capable of detecting a detectable label.
Kits
[0162] Reagents, tools, and/or instructions for performing the methods described herein can be provided in a kit. For example, the kit can contain reagents, tools, and instructions for determining an appropriate therapy for a cancer patient. Such a kit can include reagents for collecting a tissue sample from a patient, such as by biopsy, and reagents for processing the tissue. The kit can also include one or more reagents for performing a gene or gene product expression analysis, such as reagents for performing nucleic acid amplification (e.g RT-PCR, qPCR), sequencing (e.g. next generation sequencing), northern blot, proteomic analysis, or immunohistochemistry to determine expression levels of gene or gene product markers in a sample of a patient. For example, primers for performing RT-PCR, probes for performing northern blot analyses, and/or antibodies for performing proteomic analysis such as Western blot, immunohistochemistry and ELISA analyses can be included in such kits. Appropriate buffers for the assays can also be included. Detection reagents required for any of these assays can also be included. The appropriate reagents and methods are described in further detail below. The kits may include suitable primers and/or probes to detect the expression levels of at least one (up to all) of the biomarkers of in Table 2. Where expression is determined at the protein level the kit may contain binding reagents specific for the proteins of interest. The binding reagents may comprise antibodies to include all fragments and derivatives thereof. In the context of the various embodiments of the present invention the term "antibody" includes all immunoglobulins or immunoglobulin-like molecules with specific binding affinity for the relevant protein (including by way of example and without limitation, IgA, IgD, IgE, IgG and IgM, combinations thereof, and similar molecules produced during an immune response in any vertebrate, for example, in mammals such as humans, goats, rabbits and mice). Specific immunoglobulins useful in the various embodiments of the invention include IgG isotypes. The antibodies useful in the various embodiments of the invention may be monoclonal or polyclonal in origin, but are typically monoclonal antibodies. Antibodies may be human antibodies, non- human antibodies, or humanized versions of non-human antibodies, or chimeric antibodies. Various techniques for antibody humanization are well established and any suitable technique may be employed. The term "antibody" also refers to a polypeptide ligand comprising at least a light chain or heavy chain immunoglobulin variable region which specifically recognizes and binds an epitope of an antigen, and it extends to all antibody derivatives and fragments that retain the ability to specifically bind to the relevant protein. These derivatives and fragments may include Fab fragments, F(ab')2 fragments, Fv fragments, single chain antibodies, single domain antibodies, Fc fragments etc. The term antibody encompasses antibodies comprised of both heavy and light chains, but also heavy chain (only) antibodies (which may be derived from various species of cartilaginous fish or camelids). In specific embodiments, the antibodies may be engineered so as to be specific for more than protein, for example bi-specific to permit binding to two different target proteins as identified herein (see Tables 2).
[0163] In some embodiments, the kits may also contain the specific anti-angiogenic therapeutic agent to be administered in the event that the test predicts responsiveness. This agent may be provided in a form, such as a dosage form, that is tailored to the specific treatment. The kit may be provided with suitable instructions for administration according to an appropriate treatment regimen.
[0164] The kits featured herein can also include an instruction sheet describing how to perform the assays for measuring gene or gene product expression. The instruction sheet can also include instructions for how to determine a reference cohort, including how to determine expression levels of gene or gene product markers in the reference cohort and how to assemble the expression data to establish a reference for comparison to a test patient. The instruction sheet can also include instructions for assaying gene or gene product expression in a test patient and for comparing the expression level with the expression in the reference cohort to subsequently determine the appropriate chemotherapy for the test patient. Methods for determining the appropriate chemotherapy are described above and can be described in detail in the instruction sheet.
[0165] Informational material included in the kits can be descriptive, instructional, marketing or other material that relates to the methods described herein and/or the use of the reagents for the methods described herein. For example, the informational material of the kit can contain contact information, e.g., a physical address, email address, website, or telephone number, where a user of the kit can obtain substantive information about performing a gene expression analysis and interpreting the results, particularly as they apply to a human's likelihood of having a positive response to a specific therapeutic agent. [0166] The kits featured herein can also contain software necessary to infer a patient's likelihood of having a positive response to a specific therapeutic agent from the gene product marker expression.
Therapeutic agents
[0167] As described above, the methods described herein permit the classification of a patient as having a good or bad clinical prognosis prior to, upon or following administration of an anti-angiogeneic therapeutic agent following, or in combination with, a chemotherapeutic treatment. Some current such anti-angiogenic therapeutics used to treat cancer include, but are not limited to, the following agents; VEGF pathway-targeted therapeutic agent, including multi- targeted pathway inhibitors (VEGF/PDGF/FGF/EGFT/FLT-3/c-KIT), Angiopoietin-TIE2 pathway inhibitors, endogenous angiogenic inhibitors, immunomodulatory Agents. VEGF specific inhibitors include, but are not limited to, Bevacizumab (Avastin), Aflibercept (VEGF Trap), IMC-1121B (Ramucirumab). Multi-targeted pathway inhibitors include, but are not limited to, Imatinib (Gleevec), Sorafenib (Nexavar), Gefitinib (Iressa), Sunitinib (Sutent), Erlotinib, Tivozinib, Cediranib (Recentin), Pazopanib (Votrient), BIBF 1120 (Vargatef), Dovitinib, Semaxanib (Sugen), Axitinib (AGO 13736), Vandetanib (Zactima), Nilotinib (Tasigna), Dasatinib (Sprycel), Vatalanib, Motesanib, ABT-869, TKI-258. Angiopoietin-TIE2 pathway inhibitors include, but are not limited to, AMG-386, PF-4856884 CVX-060, CEP- 11981, CE-245677, MEDI-3617, CVX-241, Trastuzumab (Herceptin). Endogenous angiogenic inhibitors include, but are not limited to, Thombospondin, Endostatin, Tumstatin, Canstatin, Arrestin, Angiostatin, Vasostatin, Interferon alpha. Immunomodulatory Agents include, but are not limited to, Thalidomide and Lenalidomide. In one example embodiment, the anti- angiogenic agent is bevacizumab.
[0168] The invention is further defined in the following numbered clauses:
[0169] 1. A method for selecting whether to administer an anti-angiogenic therapeutic agent to subjects, comprising:
obtaining a test sample from a subject;
measuring expression levels of a biomarker panel from the test sample obtained from the subject, wherein the biomarker panel comprises one or more biomarkers selected from Table 2 or Table 3 ;
determining a sample expression score for the biomarker panel;
comparing the sample expression score to a threshold score; and
selecting a treatment based on whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above the threshold score an anti-angiogenic agent is contraindicated. [0170] 2. The method of clause 1, wherein the subject is suffering from cancer.
[0171] 3. The method of clause 2, wherein the cancer is ovarian cancer.
[0172] 4. The method of clause 3, wherein the ovarian cancer is high grade serous ovarian cancer.
[0173] 5. The method of any one of clauses 1 to 4, wherein the subject is receiving or has received chemotherapeutic treatment.
[0174] 6. The method of clause 5, wherein the chemotherapeutic treatment comprises administration of a platinum-based chemotherapeutic agent, an alkylating agent, an antimetabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof.
[0175] 7. The method of clause 6, wherein the chemotherapeutic treatment comprises administration of a platinum based-chemotherapeutic agent, a mitotic inhibitor, or a combination thereof.
[0176] 8. The method of clause 6, wherein the chemotherapeutic treatment comprises administration of carboplatin and paclitaxel.
[0177] 9. The method of any one of clauses 1 to 8, wherein the anti-angiogenic therapeutic agent is a VEGF-pathway-targeted therapeutic agent, an angiopoietin-TIE2 pathway inhibitor, an endogenous angiogenic inhibitor, or an immunomodulatory agent.
[0178] 10. The method of clause 9, wherein the VEGF pathway-targeted therapeutic agents include Bevacizumab (Avastin), Afibercept (VEGF Trap), 1MC-1121B (Ramucirumab), Imatinib (Gleevec), Sorafenib (Nexavar), Gefitinib (Iressa), Sunitinib (Sutent), Erlotinib, Tivozinib, Cediranib (Recentin), Pazopanib (Votrient), BIBF 1120 (Vargatef), Dovitinib, Semaxanib (Sugen), Axitinib (AG013736), Vandetanib (Zactima), Nilotinib (Tasigna), Dasatinib (Sprycel), Vatalanib, Motesanib, ABT-869, TKI-258 or a combination thereof.
[0179] 11. The method of clause 16, wherein the angiopoietin-TIE2 pathway inhibitor includes AMG-386, PF-4856884 CVX-060, CEP-11981, CE-245677, MEDI-3617, CVX-241, Trastuzumab (Herceptin) or a combination thereof.
[0180] 12. The method of clause 9, wherein the endogenous angiogenic inhibitors include Thombospondin, Endostatin, Tumstatin, Canstatin, Arrestin, Angiostatin, Vasostatin, Interferon alpha or a combination thereof.
[0181] 13. The method of clause 9, wherein the immunomodulatory agents include thalidomide and lenalidomide.
[0182] 14. The method of clause 10, wherein the VEGF pathway-targeted therapeutic agent is bevacizumab. [0183] 15. The method of any one of clauses 1 to 14, wherein the biomarker panel comprises one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, or RTP4.
[0184] 16. The method of any one of clauses 1 to 14, wherein the biomarker panel comprises the biomarkers listed in Table 2.
[0185] 17. The method of clause 16, wherein the expression score is calculated using a weight value and a bias value for each biomarker in the biomarker panel, and wherein the weight value and the bias value are defined for each biomarker in Table 2.
[0186] 18. The method of clause 16, wherein the expression score is calculated using a weight value for each biomarker in the biomarker panel, and wherein the weight for each biomarker is ranked in decreasing absolute value as defined in Table 3.
[0187] 19. The method of any one of clauses 1 to 14, wherein the biomarker panel comprises one or more of IGF2, CDR1, COL3A1, SPARC, ΤΓΜΡ3, INS, COL8A1, NUAK1, MATN3, TMEM45A.
[0188] 20. The method of any one of clauses 1 to 14, wherein the biomarker panel comprises one or more INS, SPARC, COLA1, COL3A1, CDR1, NUAK1, TEVIP3, and MMP14.
[0189] 21. A method for determing clinical prognosis of subjects, comprising:
obtaining a test sample from a subject suffering from cancer;
measuring expression levels of a biomarker panel from the test sample obtained from the subject, wherein the biomarker panel comprises one or more biomarkers selected from Table 2; determining a sample expression score for the biomarker panel;
comparing the sample expression score to a threshold score; and
determining a clinical prognosis for the subject based on whether the sample expression score is above the threshold expression score, wherein if the sample expression score is above or equal to the threshold expression score the clinical prognosis is a good prognosis.
[0190] 22. The method of clause 21, wherein the good prognosis indicates increased progression free survival or overall survival rates compared to samples with a sample expression score below the threshold score.
[0191] 23. The method of clause 21 or clause 22, wherein the cancer is ovarian cancer.
[0192] 24. The method of clause 23, wherein the ovarian cancer is high grade serous ovarian cancer.
[0193] 25. The method of any one of clauses 21 to 24, wherein the chemotherapeutic treatment comprises administration of a platinum-based chemotherapeutic agent, an alkylating agent, an anti-metabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof. [0194] 26. The method of clause 25, wherein the chemotherapeutic treatment comprises administration of a platinum-based chemotherapeutic agent, a mitotic inhibitor, or a combination thereof.
[0195] 27. The method of clause 25, wherein the chemotherapeutic treatment comprises administration of paclitaxel and carboplatin.
[0196] 28. The method of any one of clauses 21 to 27, wherein the biomarker panel comprises one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, or RTP4.
[0197] 29. The method any one of clauses 21 to 28, wherein the biomarker panel comprises the biomarkers listed in Table 2.
[0198] 30. The method of clause 29, wherein the expression score is calculated using a weight value and a bias value for each biomarker in the biomarker panel, and wherein the weight value and bias value for each biomarker are defined in Table 2.
[0199] 31. The method of clause 29, wherein the expression score is calculated using a weight value for each biomarker in the biomarker panel, and wherein the weight for each biomarker is ranked in decreasing absolute value as defined in Table 3.
[0200] 32. The method of any one of clauses 21 to 28, wherein the biomarker panel comprises one or more of IGF2, CDR1, COL3A1, SPARC, ΤΓΜΡ3, INS, COL8A1, NUAK1, MATN3, TMEM45A.
[0201] 33. The method of any one of clauses 21 to 28, wherein the biomarker panel comprises one or more INS, SPARC, COLA1, COL3A1, CDR1, NUAK1, TEVIP3, and MMP14.
[0202] 34. A method for selecting whether to administer Bevacizumab to a subject, comprising:
in a test sample obtained from a subject suffering from ovarian cancer, which subject is being, has been and/or will be treated using a platinum-based chemotherapeutic agent and/or a mitotic inhibitor;
measuring expression levels of one or more, up to all of the, biomarkers selected from Table 2;
determining a sample expression score for the one or more biomarkers;
comparing the sample expression score to a threshold score; and
selecting a treatment based on whether the sample expression score is above the threshold expression score, wherein if the sample expression score is above or equal to the threshold score Bevacizumab is contraindicated. [0203] 35. The method of clause 34 wherein the ovarian cancer comprises serous ovarian cancer.
[0204] 36. The method of clause 35 wherein the serous ovarian cancer is high grade serous ovarian cancer.
[0205] 37. The method of any one of clauses 34 to 35 wherein if Bevacizumab is contraindicated the patient is treated with a platinum-based chemotherapeutic agent and/or a mitotic inhibitor.
[0206] 38. The method of any one of clauses 34 to 37 wherein if the sample expression score is below the threshold score the patient is treated with a platinum-based chemotherapeutic agent and/or a mitotic inhibitor together with Bevacizumab.
[0207] 39. The method of any one of clauses 34 to 38, wherein the platinum-based chemotherapeutic agent comprises carboplatin.
[0208] 40. The method of any one of clauses 34 to 39, wherein the mitotic inhibitor comprises a taxane, optionally paclitaxel.
[0209] 41. The method of any one of clauses 34 to 40, wherein the biomarker panel comprises one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, or RTP4.
[0210] 42. The method any one of clauses 34 to 40, wherein the biomarker panel comprises the biomarkers listed in Table 2.
[0211] 43. The method of clause 42, wherein the expression score is calculated using a weight value and a bias value for each biomarker in the biomarker panel, and wherein the weight value and bias value for each biomarker are defined in Table 2.
[0212] 44. The method of clause 42, wherein the expression score is calculated using a weight value for each biomarker in the biomarker panel, and wherein the weight for each biomarker is ranked in decreasing absolute value as defined in Table 3.
[0213] 45. The method of any one of clauses 34 to 40, wherein the biomarker panel comprises one or more of IGF2, CDR1, COL3A1, SPARC, ΤΓΜΡ3, INS, COL8A1, NUAK1, MATN3, TMEM45A.
[0214] 46. The method of any one of clauses 34 to 40, wherein the biomarker panel comprises one or more INS, SPARC, COLA1, COL3A1, CDR1, NUAK1, TEVIP3, and MMP14.
[0215] 47. A method for determining clinical prognosis of a subject, comprising:
a. in a test sample obtained from a subject suffering from ovarian cancer, which subject is being, has been and/or will be treated using a platinum-based chemotherapeutic agent and/or a mitotic inhibitor; b. measuring expression levels of one or more, up to all of the, biomarkers selected from Table 2;
c. determining a sample expression score for the one or more biomarkers; d. comparing the sample expression score to a threshold score; and
e. determining the clinical prognosis based on whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the clinical prognosis is a good clinical prognosis.
[0216] 48. The method of clause 47, wherein the ovarian cancer comprises serous ovarian cancer.
[0217] 49. The method of clause 48, wherein the serous ovarian cancer is high grade serous ovarian cancer.
[0218] 50. The method of of any one of clauses 47 to 48, wherein if the patient has a good prognosis, treatment using Bevacizumab is contraindicated.
[0219] 51. The method of any one of clauses 47 to 50 wherein if the sample expression score is below the threshold score the patient is treated with a platinum-based chemotherapeutic agent and/or a mitotic inhibitor together with Bevacizumab.
[0220] 52. The method of any one of clauses 47 to 51, wherein the platinum-based chemotherapeutic agent comprises carboplatin.
[0221] 53. The method of any one of clauses 47 to 52, wherein the mitotic inhibitor comprises a taxane, optionally paclitaxel.
[0222] 54. The method of any one of clauses 47 to 53, wherein the biomarker panel comprises one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, or RTP4.
[0223] 55. The method any one of clauses 47 to 53, wherein the biomarker panel comprises the biomarkers listed in Table 2.
[0224] 56. The method of clause 55, wherein the expression score is calculated using a weight value and a bias value for each biomarker in the biomarker panel, and wherein the weight value and bias value for each biomarker are defined in Table 2.
[0225] 57. The method of clause 55, wherein the expression score is calculated using a weight value for each biomarker in the biomarker panel, and wherein the weight for each biomarker is ranked in decreasing absolute value as defined in Table 3.
[0226] 58. The method of any one of clauses 47 to 52, wherein the biomarker panel comprises one or more of IGF2, CDR1, COL3A1, SPARC, ΤΓΜΡ3, INS, COL8A1, NUAK1, MATN3, TMEM45A. [0227] 59. The method of any one of clauses 47 to 52, wherein the biomarker panel comprises one or more INS, SPARC, COLA1, COL3A1, CDR1, NUAK1, ΤΓΜΡ3, and MMP14.
[0228] This invention is further illustrated by the following examples, which are not to be construed in any way as imposing limitations upon the scope thereof. On the contrary, it is to be clearly understood that resort may be had to various other embodiments, modifications, and equivalents thereof which, after reading the description herein, may suggest themselves to those skilled in the art without departing from the spirit of the present invention and/or scope of the appended claims.
EXAMPLES
Example 1: Tissue processing, hierarchical clustering, subtype identification and classifier development
Tumor Material.
[0229] Example expression signatures were identified from gene expression analysis of a cohort of macrodissected epithelial serous ovarian tumor FFPE tissue samples sourced from the NHS Lothian and University of Edinburgh.
Table 3: results of pathology review of 357 epithelial ovarian cancer samples.
Figure imgf000077_0001
Gene Expression Profiling from FFPE
[0230] Total RNA was extracted from macrodissected FFPE tissue using the High Pure RNA Paraffin Kit (Roche Diagnostics GmbH, Mannheim, Germany). RNA was converted into complementary deoxyribonucleic acid (cDNA), which was subsequently amplified and converted into single-stranded form using the SPIA® technology of the WT-Ovation™ FFPE RNA Amplification System V2 (NuGEN Technologies Inc., San Carlos, CA, USA). The amplified single-stranded cDNA was then fragemented and biotin labeled using the FL- Ovation™ cDNA Biotin Module V2 (NuGEN Technologies Inc.). The fragmented and labeled cDNA was then hybridized to the Almac Ovarian Cancer DSA™. Almac's Ovarian Cancer DSA™ research tool has been optimised for analysis of FFPE tissue samples, enabling the use of valuable archived tissue banks. The Almac Ovarian Cancer DSA™ research tool is an innovative microarray platform that represents the transcriptome in both normal and cancerous ovarian tissues. Consequently, the Ovarian Cancer DSA™ provides a comprehensive representation of the transcriptome within the ovarian disease and tissue setting, not available using generic microarray platforms. Arrays were scanned using the Affymentrix Genechip® Scanner 7G (Affymetrix Inc., Santa Clara, CA).
Data preparation
[0231] Quality Control (QC) of profiled samples was carried out using MAS5 preprocessing algorithm. Different technical aspects were addressed: average noise and background homogeneity, percentage of present call (array quality), signal quality, RNA quality and hybridization quality. Distributions and Median Absolute Deviation of corresponding parameters were analyzed and used to identify possible outliers.
[0232] Almac's Ovarian Cancer DSA™ contains probes that primarily target the area within 300 nucleotides from the 3' end. Therefore standard Affymetrix RNA quality measures were adapted - for housekeeping genes intensities of 3' end probe sets with ratios of 3' end probe set intensity to the average background intensity were used in addition to usual 375' ratios. Hybridization controls were checked to ensure that their intensities and present calls conform to the requirements specified by Affymetrix.
Hierarchical Clustering and Functional Analysis
[0233] Sample pre-processing was carried out using Robust Multi-array Average (RMA) [16]. The data matrix was sorted by decreasing variance, decreasing intensity and increasing correlation to cDNA yield. Following filtering of probe sets correlated with cDNA yield, incremental subsets of the data matrix were tested for cluster stability: the GAP statistic [17] was applied to calculate the number of sample and probe set clusters while the stability of cluster composition was assessed using partition comparison methods [18,19] . The final most variable probe set list was determined based on the smallest and most stable data matrix for the selected number of sample cluster. [0234] Following standardization of the data matrix to the median probe set expression values, agglomerative hierarchical clustering was performed using Euclidean distance and Ward's linkage method [20] . The optimal number of sample and probe set clusters was determined using the GAP statistic [17]. The significance of the distribution of clinical parameter factor levels across sample clusters was assessed using ANOVA (continuous factor) or chi-squared analysis (discrete factor) and corrected for false discovery rate (product of p- value and number of tests performed). A corrected p-value threshold of 0.05 was used as criterion for significance.
®
[0235] Ovarian Cancer DSA probe sets were remapped to genes using an annotation pipeline based on Ensembl v60 [http://oct2012.archive.ensembl.org/]. Functional enrichment analysis was conducted to identify and rank biological entities which were found to be associated with the clustered gene sets using the Gene Ontology biological processes classification [21] . Entities were ranked according to a statistically derived enrichment score [22] and adjusted for multiple testing [23] . A corrected p-value of 0.05 was used as significance threshold. The identified enriched processes were summarised into an overall group function for each probe set/gene cluster.
Signature development and evaluation
[0236] Following subtype identification, a gene signature was developed for predicting the molecular group. To facilitate application of a signature to samples profiled on different platforms, probe sets were remapped to genes by summarizing all probe sets to their median expression and log2 transforming the data. Signature generation was performed using the partial least squares method [24] with the selection of features/genes based on filter feature selection during 10 repeats of five-fold cross-validation.
[0237] Univariate and multivariable survival analysis was performed using the survival package [25] in R 2.15.0. Multivariable analysis corrected for the following factors: High grade serous: Debulking status, Stage, Chemotherapy and Age at diagnosis; Tothill: Grade, Stage, Neoadjuvant treatment and Residual disease. All Kaplan-Meier graphs are a univariate representation of survival data.
Results
[0238] 265 HGS tumors underwent unsupervised hierarchical clustering based on 1400 most variable probe sets (corresponding to 1040 genes). Three sample clusters and four gene clusters were identified (Figure 1). There was no significant association between HGS clusters and clinico-pathological features. Functional analysis (Figure 1) revealed that cluster HGS3 was characterized by up regulation of genes associated with immune response and angiogenesis/vascular development (cluster referred to as Angioimmune forthwith). Cluster HGS1 was associated with upregulation of angiogenesis/vascular development (although apparently to a lesser extent than cluster HGS3) but without high expression of genes involved in immune response (cluster referred to as Angio forthwith). Cluster HGS2 was characterized by upregulation of genes involved in immune response without upregulation of genes involved in angiogenesis or vascular development (cluster referred to as Immune forthwith).
[0239] Multivariable survival analysis according to subgroup revealed that the patients in the Immune cluster had significantly prolonged OS compared to both patients in the Angioimmune (HR=0.58 [0.41-0.82], padj=0.001) and Angio clusters (HR=0.55 [0.37- 0.80], Padj=0.001). Kaplan-Meier curves are shown in Figure 2 (univariable HR and p- values are shown).
[0240] Since patients in the Immune cluster had a significantly better outcome than those in the other clusters we proceeded to develop an assay to prospectively identify these patients in the clinic. In addition, given the low expression of angiogenic genes in the immune cluster, we hypothesized that this assay may identify a population that would not benefit from therapies targeting angiogenesis, although it would require additional datasets to test this theory. For the purpose of signature generation the Angio and Angioimmune clusters were grouped together and labeled as the "pro- angiogenic" group.
[0241] A 63-gene biomarker assay was then developed that could identify patients in the immune cluster (Table 2). Consistent with the hierarchical clustering analysis, patients classified by the assay as being in the Immune cluster had a significantly improved progression free survival (PFS) (multivariable analysis; HR=0.72 [0.52-0.99], p=0.043) and OS (multivariable analysis; HR=0.61 [0.44-0.86], p=0.004) compared to the other HGS patients. These multivariate analyses corrected for debulking status, stage, chemotherapy and age at diagnosis. Kaplan-Meier curves for PFS and OS according to signature call in the Edinburgh dataset are shown in Figures 3A and 3B respectively with univariate HR performance displayed on each figure.
[0242] In order to independently validate our biomarker as a prognostic assay, it was applied to the HGS ovarian tumors within the dataset of Tothill et al. Clinical Cancer Research 2008: 14(16):5198-208. The patients identified as being in the Immune cluster had a significantly improved PFS (multivariable analysis; HR=0.62 [0.41-0.95], p=0.029) and OS (multivariable analysis; HR=0.32 [0.19-0.54], p=0.00001) compared to the other HGS patients. These multivariate analyses corrected for grade, stage, neoadjuvant treatment and residual disease. Kaplan-Meier curves for PFS and OS according to signature call in the Tothill dataset are shown in Figures 4A and 4B respectively, with univariate HR performance displayed on each figure. Example 2; Independent validation of the predictive utility of the "immune" signature Background
[0243] The International Collaboration on Ovarian Neoplasms 7 (ICON7) trial is a Gynecologic Cancer Intergroup phase 3 trial that assessed the effects of adding bevacizumab, concurrently and as a continuation, to standard chemotherapy with carboplatin and paclitaxel in patients with primary peritoneal carcinoma, fallopian tube carcinoma, and epithelial ovarian carcinoma (Perren TJ, Swart AM, Pfisterer J, Ledermann IA, Pujade-Lauraine E, Kristensen G, et al. A phase 3 trial of bevacizumab in ovarian cancer. N Engl J Med. 365(26): 2484-96, Aghajanian C, Blank SV, Goff BA, Judson PL, Teneriello MG, Husain A, et al. OCEANS: A randomized, double-blind, placebo-controlled phase III trial of chemotherapy with or without bevacizumab in patients with platinum-sensitive recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancer. Journal of Clinical Oncology. 2012; 30(17): 2039-45).
[0244] Patient characteristics, progression-free survival, toxicity, and preliminary overall survival data and a summary of quality-of-life (QoL) data have been reported from ICON7. In the standard chemotherapy group, 696 (91%) of 764 women received 18 weeks of chemotherapy by protocol. In the bevacizumab group, 719 (94%) of 764 women received 18 weeks of chemotherapy and bevacizumab and 472 (62%) continued bevacizumab to protocol completion at 54 weeks. The hazard ratio for progression-free survival with standard chemotherapy and bevacizumab was 0.81 (95% CI 0-70-0-94, p=0-004). In patients at high risk of progression, defined as International Federation of Gynecology and Obstetrics (FIGO) stage IV disease or stage III disease with greater than 1-0 cm of residual disease after debulking surgery, the hazard ratio for death in the bevacizumab group was 0-64 (95% CI 0-48-0-85; p=0-002).
Methods
[0245] Access was obtained to the ICON7 trail samples via the Medical Research Council (MRC). An honest broker held the associated clinical data from the MRC. A randomization strategy for profiling the samples has been performed based on clinical factors. All reagents, arrays, and reference samples were previously tested and passed qualification criteria.
[0246] To confirm diagnosis and histological type, all samples were independently reviewed using H+E slides by two specialist gynecological pathologists and WT1 staining was used to confirm serous histology in problematic cases. Sections were taken from FFPE blocks (almost exclusively from an adnexal mass rather than peritoneal or omental disease) and macrodissected under bright field microscopy to minimize stromal contamination (<10%). The number of ΙΟμιη sections used was dependent on the percentage of tumor in the block: two, three and four for >50%, 25- 50% and <25% tumor content in the block respectively. [0247] Total RNA was extracted from macrodissected FFPE tissue using the High Pure RNA Paraffin Kit (Roche Diagnostics GmbH, Mannheim, Germany). RNA was converted into complementary deoxyribonucleic acid (cDNA), which is subsequently amplified and converted into single- stranded form using the SPIA® technology of the WT-Ovation™ FFPE RNA Amplification System V2 (NuGEN Technologies Inc., San Carlos, CA, USA). The amplified single-stranded cDNA is then fragemented and biotin labeled using the FL-Ovation™ cDNA Biotin Module V2 (NuGEN Technologies Inc.). The fragmented and labeled cDNA was then hybridized to the Almac Ovarian Cancer DSA, on this the signature was developed. Arrays are scanned using the Affymentrix Genechip® Scanner 7G (Affymetrix Inc., Santa Clara, CA).
[0248] After pathology review there were 286 patients in in the sample with high grade serous ovarian cancer; 144 patients from the bevacizumab arm, and 142 from the no bevacizumab arm. It was estimated that 69% of patients would be in the pro-angiogenic patient group and that this would be the same in both study arms.
[0249] The primary study hypothesis was that within the 'pro-angiogenic" subgroup there will be a marked effect of bevacizumab, corresponding to at least halving of the hazard ratio for progression-free survival compared to the immune subgroup. By contrast the expectation is that in the 'immune' subgroup bevacizumab will have at best no effect or may even be slightly disadvantageous.
[0250] Five samples failed processing QC such that there were 238 progression-free survival events (83%) in high grade serous clinical study data set. The estimated study power (using formula 6 in C Schmoor, W Sauerbrei, and M Schumacher Sample size considerations for the evaluation of prognostic factors in survival analysis Statist. Med. 2000; 19:441-452) to detect detect i3>2 (corresponding to the differential effect of bevacizumab outlined in the previous paragraph) is 88% at the 10% one-sided level of statistical significance. There were 147 deaths in the data set and the power for the same analysis on survival was 75%.
[0251] Progression free survival was the primary end-point; this is the MRC calculated time provided in the data set. This is time from randomization to progression or death (from any cause) whichever occurs first. Overall survival was a secondary study end-point. This is time from randomization to death from any cause.
[0252] A stratified Cox-proportional hazards model was fitted to the progression-free survival data initially. The model had a single effect term for randomised study arm. A second stratified Cox-proportional hazards model was then fitted to the progession-free survival data. This model was also be stratified, but had separate terms for the effect of randomised study arm within each strata. [0253] The log-likelihood of the two fitted models were compared to determine whether the effect of randomised study arm depends on pro-angiogenic status (chi-square test with degrees of freedom corresponding to number of strata- 1). If the above test is statistically significant at the 5% level of statistical significance then the appropriateness of the proportional hazards assumption for the model with separate terms for randomised study arm within each strata would be assessed.
[0254] The test test for proportional hazards was done via the Grambsch-Therneau test [P. Grambsch and T. Therneau (1994), Proportional hazards tests and diagnostics based on weighted residuals. Biometrika, 81, 515-26]. Progression-free survival times were transformed to a log scale for the test. The tests on each of the terms for study arm within each strata were assessed seperately using the 5% level of statistical asignificance. If the test for proportional hazard was rejected within one or more of the strata then a restricted mean survival model would be fitted within each strata using flexible parametric survival models [P Royston, MKB Parmar The use of restricted mean survival time to estimate the treatment effect in randomized clinical trials when the proportional hazards assumption is in doubt Stat Med, 30 (2011), pp. 2409- 2421]. These models would use 3 degrees of freedom to estimate the baseline distribution function and 1 degree of freedom for the time dependent treatment effect. The maximum time over which the retricted mean would be calculated in each case was 3 years.
[0255] The above analysis will be repeated for overall survival.
Results
[0256] For patients classified as in the immune subtype (39%), the addition of bevacizumab conferred a worse progression free survival (HR 1.73 (1.12-2.68)) and overall survival (HR 2.00 (1.11-3.61)) when compared to proangiogenic patients. See Figures 5A-B and 6A-B. Accordingly, subjects with the Immune cancer subtype exhibited a poor prognosis when bevacizumab was added to their treatment regimen compared to subjects that did not have the Immune cancer subtype.
Example 3; Independent validation of the prognostic utility of immune signature
[0257] The primary objections of the study were: 1. predication of individual risk of tumour recurrence using the 63-gene signature on patient's gene expression data; 2. evaluation of the performance of the signature prognositic prediction with regards to paitents progressive free survival (PFS) outcome as well as overall survival (OS); 3. investigation of the influence of clinical covariates on the signature prognostic performance in relation to the progressive free survival event. [0258] This exploratory study included all the 139 patients in the control arm of the study. It was estimated that 85 (61.2%) of the patients would be categorized as pro-angiogenic (signature negative) by the gene signature. The chart and follow-up review indicated that 72 had progression-free survival events, with 46 deaths (overall survival events) occurring in the arm (information on patient numbers and percentages were provided by Jim Paul).
[0259] A retrospective power calculation using a sample size and power calculation method (Freedman, L.S. (1982). Tables of the number of patients required in clinical trials using the log- rank test. Statistics in Medicine. 1: 121-129) under a Cox proportional hazards regression showed that the above study figures will provide approximately 85% power to detect an hazard ratio (HR) of 0.5 when comparing progression free survival of 'immune only' with 'pro- angiogenic' molecular subgroups patients at 2-sided 5% level of significance.
[0260] Time to event (survival) analysis using the progression free survival as outcome was performed to evaluate the prognostic effects of the signature. The survival distributions of patient groups defined by the angiogenic status ('pro-angiogenic or signature negative' and 'immune-only or signature positive') were visualized using the Kaplan-Meier (KM) curve.
[0261] The Cox proportional hazards regression was performed to relate the patients' angiogenic status (negative or positive) to progression free survival event. In addition to the univariate (unadjusted) exploration, the multivariable (adjusted) Cox model was performed to explore the effect of the signature molecular subgroups (positive or negative) on the PFS and OS adjusting for other important clinical covariates. All estimated effects were reported with 95% confidence intervals from an analysis in which the signature and standard prognostic variables are included, regardless of their significance (P Royston, MKB Parmar The use of restricted mean survival time to estimate the treatment effect in randomized clinical trials when the proportional hazards assumption is in doubt Stat Med, 30 (2011), pp. 2409-2421). Due to size limitation only a few important covariates were considered; these included FIGO stage, tumour grade, debulking status, performance status (ECOG) and patient age.
[0262] The appropriateness of the proportional hazards assumption across the molecular subgroup was investigated before interpreting the Cox model results. In line with the previous analysis of study trial data, a restricted mean survival model was fitted for each molecular subgroup If the test for proportional hazard is rejected. The maximum time over which the retricted mean was to be calculated in 4 years.
Results
[0263] The 63-gene signature is prognostic in high grade serous (HGS) ovarian patients in the control arm (receiving carboplatin plus pacitaxel chemotherapy treatment) of ICON7 trial data. The results using Cox proportional hazards regression show that patients classified to the 'immune' molecular subgroup by the gene signature have a statistically significant improved progression free survival compared to those classified as pro-angiogenic before (Univariate HR = 0.48, 95%CI=0.32, 0.72; p < 0.001) and after (Multivariable HR = 0.50; 95%CI=0.32, 0.79; p = 0.003) adjusting for other clinical covariates including age, grade, ECOG, debulking status and stage see Figure 7A. Similarly, patients classified into immune molecular subgroup had a statistically significant better overall survival compared to the pro-angiogenic before (Univariate HR=0.46; 95%CI=0.26, 0.80; p = 0.006) and after (Multivariable HR=0.53; 95% CI=0.29, 0.97; p = 0.041) adjusting for other clinical covariates see Figure 7B. The data show no serious departure of the proportionality assumption.
Example 4: Predictive utility of non-angiogenesis signature in colorectal cancer
[0264] A public array data set obtained from the Gene Expression Omninbus database for a cohort of recurrent or metastatic colorectal cancer responders and non-responders to Bevacizumab on plus 2 arrays (E-GEOD- 19862) was obtained and analyzed using the example 63 gene signature of Table 2. The 63 gene ovarian immune signatures predicts response to bevacizumab with an AUC: 0.86 (0.60-1.00). See Figure 8.
Example 5; Summary of 63 gene signature
Samples;
■ Internal training samples : This sample set comprised of 193 High Grade Serous Ovarian samples retrieved from the Edinburgh Ovarian Cancer Database
■ Tothill samples: This is a publically available dataset, from which 152 High Grade
Serous Ovarian samples were used for analysis
■ ICON7 samples: This sample set comprises of 284 High Grade Serous samples from a phase III randomized trial of carboplatin and paclitaxel with or without bevacizumab first line cancer treatment which were accessed through the MRC (Medical Research Council).
o ICON7 SOC (Standard of Care) - 140 samples - refers to patients who did not receive the addition of bevacizumab
o ICON7 Immune group - 116 samples: this refers to the ICON7 samples predicted in the Immune group by the Immune 63 gene signature
o ICON7 ProAngio group - 168 samples: this refers to the ICON7 samples predicted in the ProAngiogenesis group by the Immune 63 gene signature
Methods;
Signature development A balanced sample set of 193 Ovarian HGS samples were used to develop the signature using the PLS (Dejong S. Simpls - an Alternative Approach to Partial Least-Squares Regression. Chemometr Intell Lab 1993;18:251-63) (Partial Least Squares) method during 10 repeats of 5- fold cross validation (CV). The following steps were used within signature development:
• Probesets mapped to genes and gene expression measured using the log2 transformation of the median probeset expression for each gene
• Within nested CV, quantile normalization was performed following a pre-filtering to remove 75% of genes with low variance, low intensity, and high correlation to cDNA yield
• Genes/features were ranking based on correlation adjusted t-scores and feature reduction involved discarding 10% of the least important genes until 5 genes remained
• The 63 gene signature was identified as the feature set for which the hazard ratio (HR) predicting Progression free survival (PFS) under cross-validation was optimal
The following datasets have been evaluated within CV to determine the performance of the 63 gene signature:
• Internal training set - 193 samples
• ICON7 SOC (Standard of Care) - 140 samples
• ICON7 Immune group - 116 samples
• ICON7 ProAngio group - 168 samples
Core gene analysis
The purpose of evaluating the core gene set of the signature is to determine a ranking for the genes based upon their impact on performance when removed from the signature.
This analysis involved 1,000,000 random samplings of 10 signature genes from the original 63 signature gene set. At each iteration, 10 randomly selected signature genes were removed and the performance of the remaining 53 genes was evaluated using the PFS endpoint to determine the impact on HR performance when these 10 genes were removed in the following 3 datasets:
• Internal Validation - 72 samples
• Tothill HGS (Tothill RW, Tinker AV, George J, et al. Novel molecular subtypes of serous and endometrioid ovarian cancer linked to clinical outcome. Clin Cancer Res 2008;14:5198-208) (High Grade Serous) - 152 samples
• ICON7 SOC (Standard of Care) - 140 samples
Within each of these 3 datasets, the signature genes were weighted based upon the change in HR performance (Delta HR) based upon their inclusion or exclusion. Genes ranked T have the most negative impact on performance when removed and those ranked '63' have the least impact on performance when removed.
Minimum gene analysis
The purpose of evaluating the minimum number of genes is to determine if significant performance can be achieved within smaller subsets of the original signature.
This analysis involved 10,000 random samplings of the 63 signature genes starting at 1 gene/feature, up to a maximum of 25 genes/features. For each randomly selected feature length, the signature was redeveloped using the PLS machine learning method under CV and model parameters derived. At each feature length, all randomly selected signatures were applied to calculate signature scores for the following 3 datasets:
• Tothill HGS (High Grade Serous) - 152 samples
• ICON7 SOC (Standard of Care) - 140 samples
• ICON7 Immune group - 116 samples
Continuous signature scores were evaluated with PFS (Progression Free Survival) to determine the HR (Hazard Ratio) effect. The HR for all random signatures at each feature length was summarized and figures generated to visualize the performance over CV.
Results
Signature development
This section presents the results of signature development within CV.
• Internal training set: Figures 9, 10 & 11 show the AUC (Area under the receiver operating curve), C-Index (Concordance Index) & HR of the training set, from which the 63 gene signature was identified.
• ICON7 SOC: Figures 12 & 13 show the HR and C-Index of the ICON7 SOC samples under CV.
• ICON7 Immune group: Figure 14 shows the HR of the ICON7 Immune samples (as identified by the 63 gene signature) under CV.
• ICON7 ProAngio group: Figure 15 shows the HR of the ICON7 ProAngio samples (as identified by the 63 gene signature) under CV.
Core gene analysis
The results for the core gene analysis of the 63 gene signature in 3 datasets is provided in this section.
• Internal Validation: Delta HR performance measured in this dataset for the 63 signature genes is shown in Figure 16. This figure highlights the top 10 ranked genes in the signature which are the most important in retaining a good HR performance within this dataset.
• Tothill HGS: Delta HR performance measured in this dataset for the 63 signature genes is shown in Figure 17. This figure highlights the top 10 ranked genes in the signature which are the most important in retaining a good HR performance within this dataset.
• ICON7 SOC: Delta HR performance measured in this dataset for the 63 signature genes is shown in Figure 18. This figure highlights the top 10 ranked genes in the signature which are the most important in retaining a good HR performance within this dataset.
• Delta HR across these 3 datasets was evaluated to obtain a combined gene ranking for each of the signature genes. The ranks assigned to the signature genes based on the core set analysis have been outlined in Immune63GeneSig_CoreGenes_HR.txt.
Minimum gene analysis
The results for the minimum gene analysis of the 63 gene signature in 3 datasets is provided in this section.
• Tothill HGS: The average HR performance measured in this dataset using the random sampling of the signature genes from a feature length of 1..25 is shown in Figure 19. This figure shows that to retain a significant HR performance (i.e. HR < 1) a minimum of 5 of the signature genes must be selected.
• ICON7 SOC: The average HR performance measured in this dataset using the random sampling of the signature genes from a feature length of 1..25 is shown in Figure 20. This figure shows that to retain a significant HR performance (i.e. HR < 1) a minimum of 2 of the signature genes must be selected.
• ICON7 Immune: The average HR performance measured in this dataset using the random sampling of the signature genes from a feature length of 1..25 is shown in Figure 21. This figure shows that to retain a significant HR performance (i.e. HR < 1) a minimum of 5 of the signature genes must be selected.
In summary, it is recommended that a minimum of at least 5 genes can be used and significant performance will be retained.
Example 6: Colon cancer samples
Samples and Methods
Samples
529 fresh-frozen (FF) primary tumour samples from patients with stage II, III or IV disease, 232 of which had progression free survival data followed up after receiving adjuvant chemotherapy, were assessed with the 63-gene signature. Microarray data from gene expression profiling on the Affymetrix U133 Plus 2.0 platform was obtained from the public domain (GEO accession number GSE40967).
Signature score calculation
Signature scores were calculated using the following steps:
• Robust Multi-array Analysis (RMA) background correction.
• Summarisation of probes to probesets using median expression.
• Summarisation of probesets to genes using median expression
• A quantile normalisation model is applied to the gene level matrix (or vector) on a per sample basis, which transforms the distribution of individual patient gene profiles to a similar distribution as the training data.
• The signature score is calculated per sample using a weighted sum of the expression of each gene in the signature:
5igsiaias¾ Score = »¾ Gr¾— £¾) 4- k
Where w¾ is a weight for each gene, ί¾ is a gene-specific bias, is the observed gene expression level after pre-processing and k = o.2¾S3 is a constant offset
Statistical analysis
Cox's proportional hazard regression model is used to estimate the univariate hazard ratio (HR) effect of the 63-gene signature on progression free survival following adjuvant chemotherapy. The p-value for the HR estimate is calculated using the log-rank test.
Results
Figure 22 provides the Kaplan Meier curve for the signature predictions in 232 patients that received adjuvant chemotherapy, showing that the 63 gene signature can be used to predict prognosis. The univariate HR calculated from Cox's proportional hazard regression is 0.49 with log-rank p = 0.001. "Angio on" is equivalent to negative for the biomarker signature and "Angio off is equivalent to positive for the biomarker signature.
Example 7: Prognostic utility of subtype in colorectal cancer
Samples
A public array data set was obtained from the Gene Expression Omnibus database (GSE40967) comprising a cohort of 529 patients with colorectal cancer (Marrisa et al, 2013). Samples were profiled on the Affymetrix Plus 2.0 array platform. The data comprises patients with stage II, III and IV disease, and 232 patients received adjuvant chemotherapy with follow up recurrence free survival (RFS) data. Methods
All samples were pre-processed using RMA and semi- supervised hierarchical clustering was performed using the Entrez Gene IDs defining the "angiogenesis" gene cluster (cluster 4) in Figure 1 to filter the data matrix. Following standardization of the filtered data matrix to the median gene expression values, agglomerative hierarchical clustering was performed using Euclidean distance and Ward's linkage method. The optimal number of sample and gene clusters was determined using the GAP statistic. As the gene list used to cluster the samples was highly enriched for the angiogenesis biology, samples clusters with up-regulation of these genes were class labelled angiogenesis active (or angio on); and those with down-regulation of these genes were class labelled angiogenesis inactive (or angio off).
The samples were further tested with the 63-gene signature and an association between the 63 gene signature score and the sample clusters (angiogenesis active or angiogenesis inactive) was evaluated using the Area Under the Receiver Operating Characteristic (ROC) Curve (AUC). The threshold for classifying patients as angiogenesis active/inactive within colorectal cancer was optimised to maximise the sensitivity + specificity with respect to predicting the subtype as defined by the hierarchical clustering analysis.
The clinical significance of the 63-gene signature predictions was evaluated using Kaplan Meier curves and Cox-proportional hazard survival analysis. The endpoint was defined as a progression free survival, and the Cox-proportional hazards modelling included an adjustment for clinical covariates: age; gender; stage; tumour location and MSI status.
Results
520 samples underwent semi-supervised clustering based on 219 genes (the genes defining cluster 4 in Figure 23). Two sample clusters and 3 gene clusters were identified (Figure 23). Sample cluster 1 (273 patients) was characterised by up-regulation of expression of angiogenesis genes, therefore was labelled angiogenesis active; and sample cluster 2 (256 patients) was characterised by down-regulation of expression of angiogenesis genes, therefore was labelled angiogenesis inactive.
The AUC calculated for the 63-gene signature with respect to predicting the two sample clusters was 0.86 [0.83-0.89], as depicted in the ROC curve in Figure 24. The maximum sum of sensitivity and specificity for predicting the subtype was defined at a signature score threshold of 0.6604; and this is the threshold that is applied to predict patients as being in the angiogenesis active or angiogenesis inactive subtype. Patients with a signature score > 0.6604 are classified as angiogenesis active and patients with a signature score < 0.6604 are classified as angiogenesis inactive. Cox-proportional hazard modelling of progression free survival predicted by the signature (adjusting for clinical covariates) revealed that patients in the angiogenesis inactive group had significantly improved progression free survival compared to patients in the angiogenesis active group (Table 4: HR=0.47 [0.30-0.76]). Kaplan Meier curve is shown in Figure 22.
Table 4: Multivariable survival analysis results using the 232 adjuvant treated patients
Predictive variable HR [95% CI]
Assay Positive 0.47 [0.30 - 0.76]
Stage (II) 1.37 [0.74 - 2.55]
III 3.24 [1.45 - 7.23]
IV
TWiOUr tecatiofi 1.68 [0-99 - Z i
Age 1.01 0.99 - 1.03]
Figure imgf000091_0001
Example 8: Independent validation of prognostic utility of 63-gene signature
Samples
A public array data set was obtained from the Gene Expression Omnibus database (GSE14333) comprising a cohort of 290 patients with colorectal cancer (Jorissen et al, 2009). Samples were profiled on the Affymetrix Plus 2.0 platform. 87 patients received adjuvant chemotherapy with follow up recurrence free survival (RFS)
Methods
The samples were tested with the 63-gene signature classified as being angiogenesis active or angiogenesis inactive using the threshold of 0.6604 as defined in Example 5.
The clinical significance of the 63-gene signature predictions was evaluated using Kaplan Meier curves and Cox-proportional hazard survival analysis. The endpoint was defined as a progression free survival, and the Cox-proportional hazards modelling included an adjustment for all available clinical covariates: age; gender; stage and tumour location.
Results
159 patients received a signature score > 0.6604 and were classified as angiogenesis active; and 131 patients received a signature score < 0.6604 and were classified as angiogenesis inactive. Cox-proportional hazard modelling of progression free survival predicted by the signature (adjusting for clinical covariates) revealed that patients in the angiogenesis inactive group had significantly improved progression free survival compared to patients in the angiogenesis active group (Table 5: HR=0.33 [0.14-0.83]). Kaplan Meier curve is shown in Figure 25.
Table 5: Multivariable survival analysis results using the 87 adjuvant treated patients
Predictive variable H [95% CI]
A"in ; Pn-i t.vr 0 ¾3 C.14 - 0,83}
Stage (II I) 1.03 [0.41 - 2.59]
1 and II
Figure imgf000092_0001
Sex 1.13 [0.47 : - 2.6Bj
References
1. Friedman HS, Prados MD, Wen PY, et al. Bevacizumab alone and in combination with irinotecan in recurrent glioblastoma. J Clin Oncol; 27.-4733-40 (2009).
2. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med; 350:2335-42 (2004).
3. Rini BI, Halabi S, Rosenberg JE, et al. Bevacizumab plus interferon alfa compared with interferon alfa monotherapy in patients with metastatic renal cell carcinoma: CALGB 90206. J Clin Oncol; 26/5422-8 (2008).
4. Sandler A, Gray R, Perry MC, et al. Paclitaxel-carboplatin alone or with bevacizumab for non- small-cell lung cancer. Ή Engl J Med; 355: 2542-50 (2006).
5. Wolmark N, Yothers G, O'Connell MJ, et al. A phase III trial comparing mFOLFOX6 to mFOLFOX6 plus bevacizumab in stage II or III carcinoma of the colon: results of NSABP protocol C-08. / Clin Oncol; 27. LBA4 (2009).
6. Yang JC, Haworth L, Sherry RM, et al., A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer, N Engl J Med 349 427-434 (2003).
7. WiUett CG, Boucher Y, di Tomaso E, et al, Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer, Nat. Med. 10, 145-147 (2004).
8. Miller K, Wang M, Gralow J, et al, Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer, N Engl J Med 357 2666-2676 (2007).
9. Miller KD, Chap LI , Holmes FA, et al, Randomized phase ΠΙ trial of capecitabine compared with bevacizumab plus capecitabine in patients with previously treated metastatic breast cancer, / Clin Oncol 23 792-799 (2005).
10. O'Shaughnessy J, Miles D, Gray RJ, et al, A meta-analysis of overall survival data from three randomized trials of bevacizumab (BV) and first-line chemotherapy as treatment for patients with metastatic breast cancer (MBC), / Clin Oncol 28 (suppl) (abstr 1005) (2010).
11. Reck M, von Pawel J, Zatloukal P, et al, Phase III trial of cisplatin plus gemcitabine with either placebo or bevacizumab as first-line therapy for nonsquamous non- small-cell lung cancer: AVAil, J Clin Oncol 27, 1227-1234 (2009).
12. Escudier B, Bellmunt J, Negrier S et al, Phase III trial of bevacizumab plus interferon alfa- 2a in patients with metastatic renal cell carcinoma (AVOREN): final analysis of overall survival, J Clin Oncol 28, 2144-2150 (2010)
13. Burger RA, Sill MW, Monk BJ, et al. Phase II trial of bevacizumab in persistent or recurrent epithelial ovarian cancer or primary peritoneal cancer: a Gynecologic Oncology Group Study. J Clin Oncol; 20;25(33):5165-71 (2007). 14 Tothill RW et al. Novel molecular subtypes of serous and endometrioid ovarian cancer linked to clinical outcome. Clin Cancer Res. 14(16), 5198-208 (2008).
15. Bagri A, Berry L, Gunter B, et a. (2010) Effects of anti-VEGF treatment duration on tumor growth, tumor regrowth, and treatment efficacy. Clin Cancer Res 16:3887-3900.
16. Ebos JM, Lee CR, Cruz-Munoz W, et al. (2009) Accelerated metastisis after short-term treatment with a potent inhibitor of tumor angiogenesis. Cancer Cell 15:232-239.
17. Grepin R, Guyot M, Jacquin M, Durivault J, Chamorey E. Sudaka A, Serdjebi C, Lacarelle B. Scoazec JY, Negrier S, Simmonet H, Pages G. Acceleration of clear cell renal cell carcinoma growth in mice following bevacizumab/Avastin treatment: the role of CXCL cytokines. Oncogene 2012 Mar 29:31(13).
18. Ma J. Pulfer S, Li S, Chu J, Reed K, Gallo JM. Pharmacodynamic-mediated reducation of temozolomide tumor concentrations by the angiogenesis inhibitor TNP-470. Cancer Res. 2001;61:5491-5498.
19. Marissa et al. (2013) Gene Expression Classification of Colon Cancer into Molecular Subtypes: Characterization, Validation, and Prognostic Value PLoS Med 2013; 10(5):el001453
20. Jorissen et al. (2009) Metastasis-associated gene expression changes predict poor outcomes in patients with Dukes stage B and C colorectal cancer. Clin Cancer Res 2009;15(24):7642-51

Claims

1. A method for selecting whether to administer an anti-angiogenic therapeutic agent to a subject, comprising:
measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject;
assessing from the expression level(s) of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature,
wherein if the sample is positive for the biomarker signature an anti-angiogenic therapeutic agent is contraindicated.
2. The method of claim 1 wherein assessing whether the sample is positive or negative for the biomarker signature comprises:
determining a sample expression score for the one or more biomarkers;
comparing the sample expression score to a threshold score; and
determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
3. The method of claim 1 or 2, wherein the subject is suffering from cancer
4. The method of any of claims 1-3, wherein the cancer is ovarian cancer or colorectal cancer.
5. The method of claim 4, wherein the ovarian cancer is high grade serous ovarian cancer.
6. The method of any one of claims 1 to 5, wherein the subject is receiving or has received treatment with a chemotherapeutic agent.
7. A method of treating cancer comprising administering a chemotherapeutic agent and not administering an anti-angiogenic therapeutic agent to a subject wherein the subject is selected for treatment on the basis of a method as claimed in any previous claim and the subject is positive for the biomarker signature.
8. A method of treating cancer comprising administering a chemotherapeutic agent and not administering an anti-angiogenic therapeutic agent to a subject wherein the subject is selected for treatment by
measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject;
assessing from the expression levels of the biomarkers whether the sample from the subject is positive or negative for a biomarker signature,
wherein if the sample is postive for the biomarker signature the subject is selected for treatment.
9. A chemotherapeutic agent for use in treating cancer in a subject
wherein the subject is selected for treatment on the basis of a method as claimed in any previous claim and wherein if the subject is positive for the biomarker signature the subject is not treated with an anti-angiogenic therapeutic agent.
10. A chemotherapeutic agent for use in treating cancer in a subject
wherein the subject is selected for treatment by
measuring the expression level(s) of one or more biomarkers selected from Table 2 or Table 3 in a sample from the subject;
assessing from the expression levels of the biomarkers whether the sample from the subject is positive or negative for a biomarker signature,
wherein if the sample is positive for the biomarker signature the subject is selected for treatment
and wherein the subject is not treated with an anti-angiogenic therapeutic agent.
11. A method of treating cancer comprising administering a
chemotherapeutic agent to a subject wherein the subject is positive for a biomarker signature defined by the expression levels of one or more biomarkers selected from Table 2 or Table 3 and wherein an anti-angiogenic therapeutic agent is not administered.
12. A chemotherapeutic agent for use in treating cancer in a subject
wherein the subject is positive for a biomarker signature defined by the expression levels of one or more biomarkers selected from Table 2 or Table 3 and wherein the subject is not treated with an anti-angiogenic therapeutic agent.
13. The method of any of claims 6 to 8 or 11 or chemotherapeutic agent for use of any of claims 9, 10 or 12 wherein the chemotherapeutic agent comprises a platinum-based chemotherapeutic agent, an alkylating agent, an anti-metabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof.
14. The method of any of claims 6 to 8 or 11 or chemotherapeutic agent for use of any of claims 9, 10 or 12, wherein the chemotherapeutic agent comprises a platinum based- chemotherapeutic agent, a mitotic inhibitor, or a combination thereof.
15. The method of any of claims 6 to 8 or 11 or chemotherapeutic agent for use of any of claims 9, 10 or 12, wherein the chemotherapeutic agent comprises carboplatin and/or paclitaxel.
16. The method of any of claims 1 to 8, 11 or 13 to 15 or chemotherapeutic agent for use of any of claims 9, 10 or 12 to 15, wherein the anti-angiogenic therapeutic agent is a VEGF-pathway-targeted therapeutic agent, an angiopoietin-TIE2 pathway inhibitor, an endogenous angiogenic inhibitor, or an immunomodulatory agent.
17. The method or chemotherapeutic agent for use of claim 16, wherein the VEGF pathway- targeted therapeutic agent is selected from Bevacizumab (Avastin), Aflibercept (VEGF Trap), IMC-1121B (Ramucirumab), Imatinib (Gleevec), Sorafenib (Nexavar), Gefitinib (Iressa), Sunitinib (Sutent), Erlotinib, Tivozinib, Cediranib (Recentin), Pazopanib (Votrient), BIBF 1120 (Vargatef), Dovitinib, Semaxanib (Sugen), Axitinib (AG013736), Vandetanib (Zactima), Nilotinib (Tasigna), Dasatinib (Sprycel), Vatalanib, Motesanib, ABT-869, T I-258 or a combination thereof.
18. The method or chemotherapeutic agent for use of claim 16, wherein the angiopoietin- TIE2 pathway inhibitor is selected from AMG-386, PF-4856884 CVX-060, CEP-11981, CE-245677, MEDI-3617, CVX-241, Trastuzumab (Herceptin) or a combination thereof.
19. The method or chemotherapeutic agent for use of claim 16, wherein the endogenous angiogenic inhibitor is selected from Thombospondin, Endostatin, Tumstatin, Canstatin, Arrestin, Angiostatin, Vasostatin, Interferon alpha or a combination thereof.
20. The method or chemotherapeutic agent for use of claim 16, wherein the immunomodulatory agent is selected from thalidomide and lenalidomide.
21. The method or chemotherapeutic agent for use of claim 16, wherein the VEGF pathway- targeted therapeutic agent is bevacizumab.
22. The method of any of claims 1 to 8, 11 or 13 to 21 or chemotherapeutic agent for use of any of claims 9, 10 or 12 to 21, wherein the biomarkers comprise one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, or RTP4.
23. The method of any of claims 1 to 8, 11 or 13 to 21 or chemotherapeutic agent for use of any of claims 9, 10 or 12 to 21, wherein the biomarkers comprise the biomarkers listed in Table 2.
24. The method of any of claims 7 to 8, 11 or 13 to 23 or chemotherapeutic agent for use of any of claims 9, 10 or 12 to 23 wherein assessing whether the sample is positive or negative for the biomarker signature comprises:
determining a sample expression score for the one or more biomarkers;
comparing the sample expression score to a threshold score; and
determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
25. The method of any of claims 7 to 8, 11 or 13 to 24 or chemotherapeutic agent for use of any of claims 9, 10 or 12 to 24 wherein the subject is suffering from cancer.
26. The method or chemotherapeutic agent for use of claim 25 wherein the cancer is ovarian cancer, optionally high grade serous ovarian cancer, or colorectal cancer.
27. The method of claim 2 or 24 or chemotherapeutic agent for use of claim 24, wherein the expression score is calculated using a weight value and a bias value for each biomarker, and wherein the weight value and the bias value are defined for each biomarker in Table 2.
28. The method of claim 2 or 24 or chemotherapeutic agent for use of claim 24, wherein the expression score is calculated using a weight value for each biomarker, and wherein the weight for each biomarker is ranked in decreasing absolute value as defined in Table 3.
29. The method of any of claims 1 to 8, 11 or 13 to 28 or chemotherapeutic agent for use of any of claims 9, 10 or 12 to 28, wherein the biomarker panel comprises one or more of IGF2, CDR1, COL3A1, SPARC, ΤΓΜΡ3, INS, COL8A1, NUAK1, MATN3, TMEM45A.
30. The method of any of claims 1 to 8, 11 or 13 to 28 or chemotherapeutic agent for use of any of claims 9, 10 or 12 to 28, wherein the biomarker panel comprises one or more INS, SPARC, COLA1, COL3A1, CDR1, NUAK1, TIMP3, and MMP14.
31. A method for determing clinical prognosis of a subject with cancer, comprising:
measuring the expression level(s) of one or more biomarkers selected from Table 2 or
Table 3 in a sample from the subject;
assessing from the expression level(s) of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature,
wherein if the sample is positive for the biomarker signature the subject has a good prognosis.
32. The method of claim 28 wherein assessing whether the sample is positive or negative for the biomarker signature comprises:
determining a sample expression score for the biomarkers;
comparing the sample expression score to a threshold score; and
determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
33. The method of claim 31 or 32, wherein the good prognosis indicates increased progression free survival or overall survival rates compared to samples that are negative for the biomarker signature (with a sample expression score below the threshold score).
34. The method of any of claims 31-33, wherein the cancer is ovarian cancer or colorectal cancer.
35. The method of claim 34, wherein the ovarian cancer is high grade serous ovarian cancer.
36. The method of any one of claims 31- 35, wherein the subject is receiving, has received and/or will receive chemotherapeutic treatment and/or will not receive treatment with an anti-angiogenic therapeutic agent.
37. The method of claim 36 wherein the chemotherapeutic treatment comprises administration of a platinum-based chemotherapeutic agent, an alkylating agent, an antimetabolite, an anti-tumor antibiotic, a topoisomerase inhibitor, a mitotic inhibitor, or a combination thereof.
38. The method of claim 36, wherein the chemotherapeutic treatment comprises administration of a platinum-based chemotherapeutic agent, a mitotic inhibitor, or a combination thereof.
39. The method of claim 36, wherein the chemotherapeutic treatment comprises administration of paclitaxel and carboplatin.
40. The method of any one of claims 31 to 39, wherein the biomarkers comprise one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, or RTP4.
41. The method any one of claims 31 to 40, wherein the biomarkers comprise the biomarkers listed in Table 2.
42. The method of any of claims 32 to 41, wherein the expression score is calculated using a weight value and a bias value for each biomarker, and wherein the weight value and bias value for each biomarker are defined in Table 2.
43. The method of any of claims 32 to 41, wherein the expression score is calculated using a weight value for each biomarker, and wherein the weight for each biomarker is ranked in decreasing absolute value as defined in Table 3.
44. The method of any one of claims 31 to 43, wherein the biomarkers comprise one or more of IGF2, CDR1, COL3A1, SPARC, ΤΓΜΡ3, INS, COL8A1, NUAK1, MATN3, TMEM45A.
45. The method of any one of claims 31 to 44, wherein the biomarkers comprise one or more INS, SPARC, COLA1, COL3A1, CDR1, NUAK1, TIMP3, and MMP14.
46. A method for selecting whether to administer Bevacizumab to a subject, comprising: in a test sample obtained from a subject suffering from ovarian cancer, which subject is being, has been and/or will be treated using a platinum-based chemotherapeutic agent and/or a mitotic inhibitor;
measuring expression levels of one, two or more, up to all of the, biomarkers selected from Table 2 or Table 3;
assessing from the expression level(s) of the one, two or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature
selecting a treatment based on whether the sample is positive for the biomarker signature, wherein is the sample is positive for the biomarker signature Bevacizumab is contraindicated.
47. The method of claim 46 wherein assessing whether the sample is positive or negative for the biomarker signature comprises
determining a sample expression score for the one, two or more biomarkers;
comparing the sample expression score to a threshold score; and
determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
48. The method of claim 46 ot 47 wherein the ovarian cancer comprises serous ovarian cancer.
49. The method of claim 48 wherein the serous ovarian cancer is high grade serous ovarian cancer.
50. The method of any one of claims 46 to 49 wherein if Bevacizumab is contraindicated the patient is treated with a platinum-based chemotherapeutic agent and/or a mitotic inhibitor.
51. The method of any one of claims 46 to 49 wherein if the sample expression score is below the threshold score the patient is treated with a platinum-based chemotherapeutic agent and/or a mitotic inhibitor together with Bevacizumab.
52. The method of any one of claims 46 to 51, wherein the platinum-based chemotherapeutic agent comprises carboplatin.
53. The method of any one of claims 46 to 52, wherein the mitotic inhibitor comprises a taxane, optionally paclitaxel.
54. The method of any one of claims 46 to 453, wherein the biomarkers comprise one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, or RTP4.
55. The method any one of claims 46 to 53, wherein the biomarkers comprise the biomarkers listed in Table 2.
56. The method of any of claims 47 to 55, wherein the expression score is calculated using a weight value and a bias value for each biomarker, and wherein the weight value and bias value for each biomarker are defined in Table 2.
57. The method of any of claims 47 to 55, wherein the expression score is calculated using a weight value for each biomarker, and wherein the weight for each biomarker is ranked in decreasing absolute value as defined in Table 3.
58. The method of any one of claims 46 to 57, wherein the biomarker panel comprises one or more of IGF2, CDR1, COL3A1, SPARC, TIMP3, INS, COL8A1, NUAK1, MATN3, TMEM45A.
59. The method of any one of claims 46 to 58, wherein the biomarker panel comprises one or more INS, SPARC, COLA1, COL3A1, CDR1, NUAK1, TIMP3, and MMP14.
60. A method for determining clinical prognosis of a subject, comprising: a. in a test sample obtained from a subject suffering from ovarian cancer or colorectal cancer, which subject is being, has been and/or will be treated using a platinum-based chemotherapeutic agent and/or a mitotic inhibitor;
b. measuring expression levels of one or more, up to all of the, biomarkers selected from Table 2 or Table 3;
c. assessing from the expression level(s) of the one or more biomarkers whether the sample from the subject is positive or negative for a biomarker signature, wherein if the sample is positive for the biomarker signature the subject has a good prognosis.
61. The method of claim 60 wherein assessing whether the sample is positive or negative for the biomarker signature comprises:
a. determining a sample expression score for the one or more biomarkers;
b. comparing the sample expression score to a threshold score; and
c. determining whether the sample expression score is above or equal to the threshold expression score, wherein if the sample expression score is above or equal to the threshold score the sample is positive for the biomarker signature.
62. The method of claim 60 or 61, wherein the ovarian cancer comprises serous ovarian cancer.
63. The method of claim 62, wherein the serous ovarian cancer is high grade serous ovarian cancer.
64. The method of of any one of claims 60 to 63, wherein if the patient has a good prognosis, treatment using Bevacizumab is contraindicated.
65. The method of any of claims 60 to 63, wherein if the sample expression score is below the threshold score the patient is treated with a platinum-based chemotherapeutic agent and/or mitotic inhibitor together with Bevacizumab.
66. The method of any one of claims 60 to 65, wherein the platinum-based chemotherapeutic agent comprises carboplatin.
67. The method of any one of claims 60 to 66, wherein the mitotic inhibitor comprises a taxane, optionally paclitaxel.
68. The method of any one of claims 60 to 67, wherein the biomarker panel comprises one or more of IGF2, SOX11, INS, CXCL17, SLC5A1, TMEM45A, CXCR2P1, MFAP2, MATN3, or RTP4.
69. The method any one of claims 60 to 67, wherein the biomarker panel comprises the biomarkers listed in Table 2.
70. The method of any one of claims 61 to 69, wherein the expression score is calculated using a weight value and a bias value for each biomarker, and wherein the weight value and bias value for each biomarker are defined in Table 2.
71. The method of any one of claims 61 to 69, wherein the expression score is calculated using a weight value for each biomarker, and wherein the weight for each biomarker is ranked in decreasing absolute value as defined in Table 3.
72. The method of any one of claims 60 to 71, wherein the biomarkers comprise one or more of IGF2, CDR1, COL3A1, SPARC, ΤΓΜΡ3, INS, COL8A1, NUAK1, MATN3, TMEM45A.
73. The method of any one of claims 60 to 72, wherein the biomarkers comprise one or more of INS, SPARC, COLA1, COL3A1, CDR1, NUAK1, TIMP3, and MMP14.
74. The method of any one of claims 60 to 73 wherein the good prognosis indicates increased progression free survival or overall survival rates compared to samples that are negative for the biomarker signature (with a sample expression score below the threshold score).
75. The method of any of claims 1 to 8, 11 or 13 to 74 or chemo therapeutic agent for use of any of claims 9, 10 or 12 to 30 comprising measuring the expression level(s) of two or more biomarkers selected from Table 2 or Table 3 in a sample from the subject.
76. The method of any of claims 1 to 8, 11 or 13 to 74 or chemotherapeutic agent for use of any of claims 9, 10 or 12 to 30 comprising measuring the expression level(s) of five or more biomarkers selected from Table 2 or Table 3 in a sample from the subject.
77. The method of any of claims 1 to 8, 11 or 13 to 76 or chemotherapeutic agent for use of any of claims 9, 10 or 12 to 30 wherein assessing whether the sample is positive or negative for the biomarker signature comprises the use of classification trees or random forests.
PCT/GB2015/050352 2014-02-07 2015-02-09 Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer WO2015118353A1 (en)

Priority Applications (8)

Application Number Priority Date Filing Date Title
BR112016018044A BR112016018044A2 (en) 2014-02-07 2015-02-09 Molecular diagnostic test to predict antiangiogenic drug response and cancer prognosis
EP15705693.8A EP3102700A1 (en) 2014-02-07 2015-02-09 Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer
CA2938807A CA2938807A1 (en) 2014-02-07 2015-02-09 Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer
KR1020167024823A KR20160117606A (en) 2014-02-07 2015-02-09 Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer
AU2015213844A AU2015213844A1 (en) 2014-02-07 2015-02-09 Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer
US15/116,641 US10280468B2 (en) 2014-02-07 2015-02-09 Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer
CN201580016568.5A CN106164296A (en) 2014-02-07 2015-02-09 For prediction, the response of anti-angiogenic drugs and the molecular diagnosis of cancer prognosis are tested
JP2016550631A JP2017506506A (en) 2014-02-07 2015-02-09 Molecular diagnostic tests for response to anti-angiogenic drugs and prediction of cancer prognosis

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US201461937224P 2014-02-07 2014-02-07
US61/937,224 2014-02-07
GBGB1409479.1A GB201409479D0 (en) 2014-05-28 2014-05-28 Molecular diagnostic test for cancer
GB1409479.1 2014-05-28

Publications (1)

Publication Number Publication Date
WO2015118353A1 true WO2015118353A1 (en) 2015-08-13

Family

ID=51177582

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB2015/050352 WO2015118353A1 (en) 2014-02-07 2015-02-09 Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer

Country Status (10)

Country Link
US (1) US10280468B2 (en)
EP (1) EP3102700A1 (en)
JP (1) JP2017506506A (en)
KR (1) KR20160117606A (en)
CN (1) CN106164296A (en)
AU (1) AU2015213844A1 (en)
BR (1) BR112016018044A2 (en)
CA (1) CA2938807A1 (en)
GB (1) GB201409479D0 (en)
WO (1) WO2015118353A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018124153A1 (en) * 2016-12-28 2018-07-05 公益財団法人がん研究会 Biomarker for predicting therapeutic effect of anti-vegfr-2 antibody drug, screening method, and screening kit
WO2020092808A1 (en) * 2018-10-31 2020-05-07 Regents Of The University Of Minnesota Methods for predicting a response to bevacizumab or platinum-based chemotherapy or both in patients with ovarian cancer
WO2022204438A1 (en) * 2021-03-25 2022-09-29 Oncxerna Therapeutics, Inc. Targeted therapies in cancer

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11568982B1 (en) 2014-02-17 2023-01-31 Health at Scale Corporation System to improve the logistics of clinical care by selectively matching patients to providers
US10650325B2 (en) * 2015-07-31 2020-05-12 Microsoft Technology Licensing, Llc Deterministic message distribution
US10496691B1 (en) * 2015-09-08 2019-12-03 Google Llc Clustering search results
WO2018062862A1 (en) * 2016-09-28 2018-04-05 가톨릭대학교 산학협력단 Gender-specific markers for diagnosing prognosis and determining treatment strategy for renal cancer patients
CN107301328B (en) * 2017-05-19 2021-04-06 浙江工业大学 Cancer subtype accurate discovery and evolution analysis method based on data flow clustering
US20180357361A1 (en) 2017-06-13 2018-12-13 Feliks Frenkel Systems and methods for identifying responders and non-responders to immune checkpoint blockade therapy
JP6977965B2 (en) * 2017-10-26 2021-12-08 公立大学法人福島県立医科大学 Ovarian cancer histological differentiation method
CN110880356A (en) * 2018-09-05 2020-03-13 南京格致基因生物科技有限公司 Method and apparatus for screening, diagnosing or risk stratification for ovarian cancer
US11610679B1 (en) 2020-04-20 2023-03-21 Health at Scale Corporation Prediction and prevention of medical events using machine-learning algorithms
CN112164474B (en) * 2020-07-14 2024-05-24 中国矿业大学 Drug sensitivity prediction method based on self-expression model
US12094582B1 (en) 2020-08-11 2024-09-17 Health at Scale Corporation Intelligent healthcare data fabric system
US12080428B1 (en) 2020-09-10 2024-09-03 Health at Scale Corporation Machine intelligence-based prioritization of non-emergent procedures and visits
CN113667752A (en) * 2021-09-09 2021-11-19 菲思特(上海)生物科技有限公司 Detection kit for daunorubicin metabolic marker and detection method and application thereof

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012092336A2 (en) * 2010-12-28 2012-07-05 Caris Mpi, Inc. Molecular profiling for cancer
WO2012167278A1 (en) * 2011-06-02 2012-12-06 Almac Diagnostics Limited Molecular diagnostic test for cancer
WO2013106765A1 (en) * 2012-01-13 2013-07-18 Genentech, Inc. Biological markers for identifying patients for treatment with vegf antagonists
WO2014087156A1 (en) * 2012-12-03 2014-06-12 Almac Diagnostics Limited Molecular diagnostic test for cancer

Family Cites Families (46)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5525464A (en) 1987-04-01 1996-06-11 Hyseq, Inc. Method of sequencing by hybridization of oligonucleotide probes
US5202231A (en) 1987-04-01 1993-04-13 Drmanac Radoje T Method of sequencing of genomes by hybridization of oligonucleotide probes
GB8810400D0 (en) 1988-05-03 1988-06-08 Southern E Analysing polynucleotide sequences
US5143854A (en) 1989-06-07 1992-09-01 Affymax Technologies N.V. Large scale photolithographic solid phase synthesis of polypeptides and receptor binding screening thereof
US6040138A (en) 1995-09-15 2000-03-21 Affymetrix, Inc. Expression monitoring by hybridization to high density oligonucleotide arrays
US5800992A (en) 1989-06-07 1998-09-01 Fodor; Stephen P.A. Method of detecting nucleic acids
US5547839A (en) 1989-06-07 1996-08-20 Affymax Technologies N.V. Sequencing of surface immobilized polymers utilizing microflourescence detection
EP0430881A3 (en) 1989-11-29 1991-10-23 Ciba-Geigy Ag Photochromic compounds, process for their preparation and their use
US5288644A (en) 1990-04-04 1994-02-22 The Rockefeller University Instrument and method for the sequencing of genome
US5324633A (en) 1991-11-22 1994-06-28 Affymax Technologies N.V. Method and apparatus for measuring binding affinity
DE69322266T2 (en) 1992-04-03 1999-06-02 Perkin-Elmer Corp., Foster City, Calif. SAMPLES COMPOSITION AND METHOD
US5503980A (en) 1992-11-06 1996-04-02 Trustees Of Boston University Positional sequencing by hybridization
US5858659A (en) 1995-11-29 1999-01-12 Affymetrix, Inc. Polymorphism detection
US5470710A (en) 1993-10-22 1995-11-28 University Of Utah Automated hybridization/imaging device for fluorescent multiplex DNA sequencing
GB9401833D0 (en) 1994-02-01 1994-03-30 Isis Innovation Method for discovering ligands
WO1997027317A1 (en) 1996-01-23 1997-07-31 Affymetrix, Inc. Nucleic acid analysis techniques
GB9507238D0 (en) 1995-04-07 1995-05-31 Isis Innovation Detecting dna sequence variations
US5661028A (en) 1995-09-29 1997-08-26 Lockheed Martin Energy Systems, Inc. Large scale DNA microsequencing device
US6998234B2 (en) 2000-11-03 2006-02-14 Oncotech, Inc. Methods for cancer prognosis and diagnosis relating to tumor vascular endothelial cells
MXPA05012723A (en) 2003-05-30 2006-02-08 Genentech Inc Treatment with anti-vegf antibodies.
US7659062B2 (en) 2003-06-03 2010-02-09 The Board of Trustee of the University of Arkansas System Gene expression profiling of uterine serous papillary carcinomas and ovarian serous papillary tumors
US7700280B2 (en) 2003-12-31 2010-04-20 The Penn State Research Foundation Methods for assessing cisplatin resistance, disease progression, and treatment efficacy in ovarian cancer
US20080113874A1 (en) 2004-01-23 2008-05-15 The Regents Of The University Of Colorado Gefitinib sensitivity-related gene expression and products and methods related thereto
WO2005100606A2 (en) 2004-04-09 2005-10-27 Genomic Health, Inc. Gene expression markers for predicting response to chemotherapy
MXPA06013079A (en) 2004-05-14 2007-10-08 Ludwig Inst Cancer Res Identification and characterization of a subset of glioblastomas sensitive to treatment with imatinib.
MX2007000944A (en) 2004-07-23 2007-04-13 Astrazeneca Ab Method of predicting the responsiveness oa a tumor to erbb receptor drugs.
EP1815021A2 (en) 2004-11-03 2007-08-08 Almac Diagnostics Limited Transcriptome microarray technology and methods of using the same
US7862995B2 (en) * 2004-12-10 2011-01-04 Targeted Molecular Diagnostics Methods and materials for predicting responsiveness to treatment with dual tyrosine kinase inhibitor
EP1937837A2 (en) 2005-07-29 2008-07-02 Siemens Healthcare Diagnostics Inc. Methods and kits for the prediction of therapeutic success, recurrence free and overall survival in cancer therapies
US8445198B2 (en) 2005-12-01 2013-05-21 Medical Prognosis Institute Methods, kits and devices for identifying biomarkers of treatment response and use thereof to predict treatment efficacy
EP1963862A4 (en) 2005-12-05 2010-03-03 Merck & Co Inc Methods for predicting treatment response based on the expression profiles of protein and transcription biomakers
JP2009524438A (en) 2006-01-27 2009-07-02 トライパス イメージング インコーポレイテッド Methods and compositions for identifying patients likely to have ovarian cancer
US8768629B2 (en) 2009-02-11 2014-07-01 Caris Mpi, Inc. Molecular profiling of tumors
KR20090071603A (en) 2006-09-19 2009-07-01 노파르티스 아게 Biomarkers of target modulation, efficacy, diagnosis, and/or prognosis for raf inhibitors
EP2201142A1 (en) 2007-08-13 2010-06-30 Almac Diagnostics Limited A 3' -based sequencing approach for microarray manufacture
RU2010123381A (en) * 2007-11-09 2011-12-20 Дженентек, Инк. (Us) METHOD AND COMPOSITIONS FOR DIAGNOSTIC USE IN CANCER PATIENTS
US20110159492A1 (en) 2008-06-04 2011-06-30 Rimsza Lisa M Diffuse Large B-Cell Lymphoma Markers and Uses Therefore
CA2726811C (en) 2008-06-06 2019-11-26 The United States Of America, As Represented By The Secretary, Dept. Of Health And Human Services Survival predictor for diffuse large b cell lymphoma
US20110265197A1 (en) 2008-07-16 2011-10-27 Dana-Farber Cancer Institute, Inc. Signatures and PCDeterminants Associated with Prostate Cancer and Methods of Use Thereof
WO2010010153A1 (en) 2008-07-23 2010-01-28 F. Hoffmann-La Roche Ag Identification of subjects being susceptible to anti-angiogenesis therapy
MX2011006675A (en) 2008-12-23 2011-07-12 Merck Patent Gmbh Biomarkers for inhibitors with anti-angiogenic activity.
WO2010088688A2 (en) 2009-02-02 2010-08-05 Biotheranostics, Inc. Diagnosis of in situ and invasive breast cancer
CA2760333A1 (en) 2009-05-01 2010-11-04 Genomic Health Inc. Gene expression profile algorithm and test for likelihood of recurrence of colorectal cancer and response to chemotherapy
US7615353B1 (en) * 2009-07-06 2009-11-10 Aveo Pharmaceuticals, Inc. Tivozanib response prediction
CA2772670A1 (en) 2009-09-17 2011-03-24 F. Hoffmann-La Roche Ag Methods and compositions for diagnostics use in cancer patients
EP2975399B1 (en) 2010-09-15 2022-05-11 Almac Diagnostic Services Limited Molecular diagnostic test for cancer

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012092336A2 (en) * 2010-12-28 2012-07-05 Caris Mpi, Inc. Molecular profiling for cancer
WO2012167278A1 (en) * 2011-06-02 2012-12-06 Almac Diagnostics Limited Molecular diagnostic test for cancer
WO2013106765A1 (en) * 2012-01-13 2013-07-18 Genentech, Inc. Biological markers for identifying patients for treatment with vegf antagonists
WO2014087156A1 (en) * 2012-12-03 2014-06-12 Almac Diagnostics Limited Molecular diagnostic test for cancer

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
DIRK O BAUERSCHLAG ET AL: "Evaluation of Potentially Predictive Markers for Anti-Angiogenic Therapy with Sunitinib in Recurrent Ovarian Cancer Patients", TRANSLATIONAL ONCOLOGY, vol. 6, no. 3, 1 June 2013 (2013-06-01), United States, pages 305 - 310, XP055141223, ISSN: 1936-5233, DOI: 10.1593/tlo.13205 *
REINMUTH N ET AL: "Current data on predictive markers for antiangiogenic therapy in thoracic tumours", vol. 36, no. 4, 1 January 2010 (2010-01-01), pages 915 - 924, XP008146564, ISSN: 0903-1936, Retrieved from the Internet <URL:http://erj.ersjournals.com/> [retrieved on 20150424], DOI: 10.1183/09031936.00074009 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018124153A1 (en) * 2016-12-28 2018-07-05 公益財団法人がん研究会 Biomarker for predicting therapeutic effect of anti-vegfr-2 antibody drug, screening method, and screening kit
JPWO2018124153A1 (en) * 2016-12-28 2019-11-07 公益財団法人がん研究会 Biomarker, test method and test kit for predicting therapeutic effect of anti-VEGFR-2 antibody drug
WO2020092808A1 (en) * 2018-10-31 2020-05-07 Regents Of The University Of Minnesota Methods for predicting a response to bevacizumab or platinum-based chemotherapy or both in patients with ovarian cancer
US20220017965A1 (en) * 2018-10-31 2022-01-20 Regents Of The University Of Minnesota Methods for predicting a response to bevacizumab or platinum-based chemotherapy or both in patients with ovarian cancer
WO2022204438A1 (en) * 2021-03-25 2022-09-29 Oncxerna Therapeutics, Inc. Targeted therapies in cancer

Also Published As

Publication number Publication date
GB201409479D0 (en) 2014-07-09
US20170073761A1 (en) 2017-03-16
JP2017506506A (en) 2017-03-09
AU2015213844A1 (en) 2016-09-15
BR112016018044A2 (en) 2018-02-20
EP3102700A1 (en) 2016-12-14
KR20160117606A (en) 2016-10-10
CN106164296A (en) 2016-11-23
US10280468B2 (en) 2019-05-07
CA2938807A1 (en) 2015-08-13

Similar Documents

Publication Publication Date Title
US10280468B2 (en) Molecular diagnostic test for predicting response to anti-angiogenic drugs and prognosis of cancer
US11091809B2 (en) Molecular diagnostic test for cancer
US11254986B2 (en) Gene signature for immune therapies in cancer
AU2012261820B2 (en) Molecular diagnostic test for cancer
US10378066B2 (en) Molecular diagnostic test for cancer
AU2012261820A1 (en) Molecular diagnostic test for cancer
US20160222459A1 (en) Molecular diagnostic test for lung cancer
US20160222460A1 (en) Molecular diagnostic test for oesophageal cancer
WO2017216559A1 (en) Predicting responsiveness to therapy in prostate cancer
US20170088902A1 (en) Expression profiling for cancers treated with anti-angiogenic therapy

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15705693

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2938807

Country of ref document: CA

WWE Wipo information: entry into national phase

Ref document number: 15116641

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2016550631

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112016018044

Country of ref document: BR

REEP Request for entry into the european phase

Ref document number: 2015705693

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2015705693

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 20167024823

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2015213844

Country of ref document: AU

Date of ref document: 20150209

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 112016018044

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20160803