A method for determining a subject's responsiveness to treatment with a VEGF inhibitor

Abstract

The present invention generally relates to predicting a subject's responsiveness to treatment with an VEGF inhibitor based on a novel gene signature, to assist in the selection of an appropriate treatment regimen In particular, analysing the novel gene signature in a sample obtained from the subject is used for determining a subject's responsiveness to treatment with a VEGF inhibitor. More specifically, the invention relates to methods, uses and kits for predicting if a subject is likely to be responsive or non- responsive to treatment with a VEGF inhibitor, which may subsequently assist in the selection of a suitable cancer treatment regimen.

Description

A METHOD FOR DETERMINING A SUBJECT’S RESPONSIVENESS TO TREATMENT WITH A VEGF INHIBITORCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of priority of Singapore Patent Application No. 10202402067T filed 12 July 2024, the content of which being hereby incorporated by reference in its entirety for all purposes.FIELD OF INVENTION

[0002] The present invention generally relates to predicting a subject’s responsiveness to treatment with an VEGF inhibitor based on a novel gene signature, to assist in the selection of an appropriate treatment regimen. In particular, analysing the novel gene signature in a sample obtained from the subject is used for determining a subject’s responsiveness to treatment with a VEGF inhibitor. More specifically, the invention relates to methods, uses and kits for predicting if a subject is likely to be responsive or non-responsive to treatment with a VEGF inhibitor, which may subsequently assist in the selection of a suitable cancer treatment regimen.BACKGROUND

[0003] Bevacizumab is a monoclonal antibody targeting vascular endothelial growth factor (VEGF) that is used in the treatment of several different cancer types, often in the recurrent-metastatic setting. Identifying patients who will benefit most from bevacizumab treatment remains a challenge. Bevacizumab acts in blocking downstream effects of VEGF, including the modulation of endothelial cells, and is effective in prolonging survival across several cancers, including glioblastoma, ovarian cancer, lung cancer and colorectal cancer. More recently, bevacizumab has been used in the treatment of patients with advanced nasopharyngeal cancer (NPC). Since 2004, bevacizumab has been used to treat more than 2.2 million patients worldwide.

[0004] Unfortunately, despite the wide use of bevacizumab and the inclusion of bevacizumab in some clinical practice guidelines, it is not currently possible to identify which patients will respond to treatment with bevacizumab or other VEGF inhibitors. It remains a clinical dilemma whether or not to add bevacizumab or other VEGF inhibitors on top of already toxic chemotherapy regimens given to these patients.

[0005] Accordingly, there is still need for the development of a method capable of predicting a patient's responsiveness to treatment with VEGF inhibitors such as bevacizumab. These methods would allow physicians to accurately identify subjects that will be responsive to treatment with a VEGF inhibitor, and subjects that will be non-responsive to treatment with a VEGF inhibitor, and may thus be spared unnecessary treatment with a VEGF inhibitor that may have adverse side effects, whereby a suitable alternative cancer treatment regimen may be designed for the non-responsive subject.SUMMARY

[0006] The present invention meets this need by providing a method, use and kit that allow for the accurate determination and distinction between responsive and non-responsive subjects to treatment with a VEGF inhibitor.

[0007] In accordance with a first aspect of the present invention, there is provided a method for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, the method comprising: a) determining gene expression levels, in a sample obtained from the subject, to obtain a gene expression profile of the sample; b) analysing the gene expression profile based on a reduced response (RR) gene signature, and / or an improved response (IR) gene signature, wherein the RR gene signature comprises a RR gene set comprising at least two genes selected from the group consisting of: FAM110C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL11 A1 , ID2, ABCA2, HMOX1 , HLA-C, FTL, GJA5, COL5A2, MAN1A1 , CD248, LUM, FMO3, MTRNR2L1 , GUCY1A1 , and MCL1 , and wherein the IR gene signature comprises a IR gene set comprising at least two genes selected from the group consisting of: ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 ; and c) predicting the responsiveness of the subject to treatment with a VEGF inhibitor based on the analysis of the gene expression profile.

[0008] In various embodiments, the RR gene set comprises fibroblast cell-specific genes that define a fibroblast cell gene signature, preferably the fibroblast cell-specific genes comprise at least the following genes: o SPARC, GPC6, TIMP2, ISLR, COL5A2, and LUM; o SPARC, SFRP4, ISLR, COL11 A1 , COL5A2, and LUM; or o SPARC, HLA-A, CCL2, MYL9, HMOX1 , HLA-C, and LUM.

[0009] In various embodiments, the RR gene set consists of FAM1 10C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL1 1 A1 , ID2, ABCA2, HMOX1 , HLA-C, FTL, GJA5, COL5A2, MAN1 A1 , CD248, LUM, FMO3, MTRNR2L1 , GUCY1 A1 , and MCL1 ; and the IR gene set consists of ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 .

[0010] In various embodiments, the cancer is ovarian cancer, lung cancer, brain cancer (preferably glioblastoma), head and neck cancer (preferably nasopharyngeal) or colorectal cancer.

[0011] In various embodiments, the sample is a head and neck tumour, preferably a nasopharyngeal tumour, or a brain tumour, preferably a glioblastoma tumour, more preferably a proneural or non- proneural glioblastoma tumour.

[0012] In various embodiments, the glioblastoma tumour is a non-proneural glioblastoma tumour, preferably a mesenchymal, or proliferative (classical) glioblastoma tumour, or is an unclassified glioblastoma tumour.

[0013] In various embodiments, the sample is a tumour sample comprising tumor cells from the tumor epithelial compartment, which includes epithelial cells, as well as microenvironment cells (e.g. immune cells and stromal cells) admixed with epithelial cells within the same compartment.

[0014] In various embodiments, the sample is a tissue sample, preferably a fixed tissue sample, more preferably a formalin fixed paraffin embedded (FFPE) tissue sample, fresh, or frozen, optionally the tissue sample is microdissected by laser capture micro-dissection (LCM), preferably the microdissected tissue sample comprises epithelial cells.

[0015] In various embodiments, the VEGF inhibitor is an anti-VEGF antibody, preferably bevacizumab.

[0016] In various embodiments, the subject is a human who has been diagnosed with the cancer, and has either received treatment or is treatment-naive.

[0017] In various embodiments, the step of analysing comprises differential gene expression (DGE) analysis and / or gene set enrichment analysis, and / or gene set scoring analysis of the gene expression profile.

[0018] In various embodiments, the step of analysing comprises gene set enrichment analysis by one or more gene set enrichment methods, wherein the one or more gene set enrichment methods provide a gene set enrichment score that is predictive of the responsiveness of the subject to treatment with the VEGF inhibitor, optionally the significance of the gene set enrichment score of the RR and / or IR gene signature to predict the responsiveness of the subject to treatment with the VEGF inhibitor is based on a threshold enrichment score.

[0019] In various embodiments, the one or more gene set enrichment methods comprise ssGSEA and GSVA.

[0020] In various embodiments, the subject is predicted to have a reduced response (RR) to treatment with the VEGF inhibitor if a positive gene set enrichment score for the RR gene signature (i.e. enriched RR gene signature) is obtained; or the subject is predicted to have an improved response to treatment with the VEGF inhibitor if a positive gene set enrichment score for the IR gene signature (i.e. enriched IR gene signature) is obtained.

[0021] In various embodiments, the positive gene set enrichment score for the RR gene signature is correlated with a predicted length of survival or overall survival of the subject from the cancer, preferably wherein the length of survival or overall survival is provided in months, more preferably is at least 3 months, at least 4 months, at least 5 months, at least 6 months, at least 10 months, at least 15 months, at least 20 months, at least 25 months, at least 30 months, at least 35 months, or at least 40 months, or at least 50 months, or at least 60 months.

[0022] In various embodiments, the step of analysing comprises comparing the gene expression profile of the sample to a reference gene expression profile, wherein a differential gene expression profile in respect of the RR and / or IR gene signature compared to the reference gene expression profile in respect of the RR and / or IR gene signature may be used to predict the responsiveness of the subject to treatment with the VEGF inhibitor, preferably the step of analysing comprises differential gene expression analysis using one or more Differential Gene Expression (DGE) analysis methods.

[0023] In various embodiments, the subject is predicted to have a reduced response (RR) to treatment with the VEGF inhibitor if an increased gene expression profile of the RR gene signature compared to the reference gene expression profile is obtained; or the subject is predicted to have a improved response to treatment with the VEGF inhibitor if an increased gene expression profile of the IR gene signature compared to the reference gene expression profile is obtained.

[0024] In various embodiments, the step of analysing comprises gene set scoring analysis by one or more gene set scoring methods, wherein the one or more gene set scoring methods provide a gene set score that is predictive of the responsiveness of the subject to treatment with the VEGF inhibitor, optionally the significance of the gene set score of the RR and / or IR gene signature is determined based on a threshold score to predict the responsiveness of the subject to treatment with the VEGF inhibitor.

[0025] In accordance with a second aspect of the present invention, there is provided a method of treating, ameliorating, delaying or preventing cancer in a subject in need of such treatment, the method comprising: a) predicting the responsiveness of the subject to treatment with a VEGF inhibitor by the method disclosed herein; and b) treating the cancer with the VEGF inhibitor if, a decreased gene expression profile of the RR gene signature and / or a disenriched RR gene signature, and / or an absence of an enriched RR gene signature is obtained from the sample; and / oran increased gene expression profile of the IR gene signature and / or an enriched IR gene signature is obtained from the sample.

[0026] In accordance with a third aspect of the present invention, there is provided a method for selecting a suitable cancer treatment regimen in a subject with cancer, the method comprising: a) predicting the responsiveness of the subject to treatment with a VEGF inhibitor by the method disclosed herein: and b) selecting a cancer treatment regimen for the subject comprising administering a VEGF inhibitor to the subject if: a decreased gene expression profile of the RR gene signature and / or a disenriched RR gene signature, and / or an absence of an enriched RR gene signature is obtained from the sample; and / or an increased gene expression profile of the IR gene signature and / or an enriched IR gene signature is obtained from the sample.

[0027] In accordance with a fourth aspect of the present invention, there is provided a kit for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, the kit comprising: i) reagents for determining a gene expression profile in a sample obtained from the subject: ii) reagents for analysing the gene expression profile based on a reduced response (RR) gene signature, and / or an improved response (IR) gene signature, wherein the RR gene signature comprises a RR gene set comprising at least two genes selected from the group consisting of: FAM110C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL11 A1 , ID2, ABCA2, HMOX1 , HLA-C, FTL, GJA5, COL5A2, MAN1A1 , CD248, LUM, FMO3, MTRNR2L1 , GUCY1A1 , and MCL1 , and wherein the IR gene signature comprises a IR gene set comprising at least two genes selected from the group consisting of: ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 ; and iii) instructions for using the reagents and interpreting the analysis to predict the responsiveness of the subject to treatment with a VEGF inhibitor.BRIEF DESCRIPTION OF DRAWINGS

[0028] FIG. 1A-C shows a laser-capture microdissected gene expression profiling of NPC tumors treated with bevacizumab: FIG.1A Principal component analysis of gene expression profiles and clusters of different microdissected cell types; FIG.1 B Differentially expressed genes in the microdissected tumor epithelial compartment, compared to the microenvironment compartment; and FIG.1C Geneset enrichment analysis of GO (geneontology) biological processes enriched in the tumor epithelial compartment compared to the microenvironment compartment.

[0029] FIG. 2A-F shows a discovery cohort of locally-advanced NPC tumors treated with bevacizumab: FIG. 2A-B Volcano plot and heatmap of differentially expressed genes in the microdissected tumor epithelial compartment in the discovery cohort comparing between primary NPC tumors that showed partial response, and tumors that showed improved response (n = 37 gene expression libraries); FIG. 2C-D Geneset enrichment analysis of Gavish 3GA cell types and GO cell types enriched in the NPC tumor epithelial compartment, comparing tumors that showed partial response and tumors that showed improved response; and FIG. 2E-F RR signature single sample geneset enrichment scores in NPC tumors that showed partial response compared to tumors that showed improved response, using the ssGSEA method (FIG. 2E), and the GSVA method (FIG. 2F). In these figures, the term “complete response” corresponds to, and is used interchangeably with, the term “improved response,”. Similarly, the term “partial response” corresponds to, and is used interchangeably with, the term “reduced response,”. The term “RRB signature” corresponds to the “RR gene signature” as disclosed herein, wherein the additional “B” refers to Bevacizumab.

[0030] FIG. 3A-F shows a validation cohort from recurrent GBM tumors treated with bevacizumab: FIG. 3A-B Correlation of the RR signature single sample geneset enrichment scores with overall survival using the ssGSEA method (FIG. 3A), and the GSVA method (FIG. 3B). FIG. 3C-D Correlation of the RR - cell type signature single sample geneset enrichment scores with overall survival ; and FIG. 3E-F Correlation of the RR - 3CA signature single sample geneset enrichment scores with overall survival. The RRB signature corresponds with the RR gene signature disclosed herein (the additional B refers to Bevacizumab).

[0031] FIG. 4A-D shows a validation cohort from the AVAglio clinical trial for GBM tumors treated with bevacizumab: FIG. 4A-B Correlation of the RR 7-gene subset signature single sample geneset enrichment scores with survival time in non-proneural GBM tumors that were not completely resected, in the bevacizumab-treated group (FIG. 4A), and the placebo group (FIG. 4B); and FIG. 4C-D Survival analysis based on Kaplan-Meier curve plots in patients stratified by their single sample geneset enrichment scores, in the bevacizumab-treated group (FIG. 4C), and the placebo group (FIG. 4D). The RRB signature corresponds with the RR gene signature disclosed herein (the additional B refers to Bevacizumab).DETAILED DESCRIPTION

[0032] It will be appreciated by a person skilled in the art that numerous variations and / or modifications may be made to the present invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects to be illustrative and not restrictive. Also, the invention includes any combination of features described for different embodiments, including in the summary section, even if the feature or combination of features is not explicitly specified in the summary section or the detailed description of the present embodiments.

[0033] The treatment of a variety of cancers utilize targeting strategies involving inhibition of the vascular endothelial growth factor pathway (VEGF). This may involve first-line therapies for primary disease, or second or subsequent-line therapies for recurrent / metastatic disease. There are currently no clinical tests available to predict the suitability of patients to treatment with VEGF inhibitors and / or predicting patient outcomes. This present invention provides a precision medicine solution so that patients who will benefit most from VEGF inhibition can be selected for treatment, while patients who are unresponsive, or have a reduced response, should be considered for alternative therapies. Therefore, the present invention may provide a precision medicine approach based on the provision of an accurate method of predicting the responsiveness of a cancer patient to treatment with a VEGF inhibitor.

[0034] The present invention is based on the inventors’ surprising identification of gene signatures correlating to expression levels of a gene set in patient tumour samples, which can be used to predict treatment response to a VEGF inhibitor, thereby allowing patients to be classified and categorised to help physicians select suitable treatment regimens, which may or may not include treatment with a VEGF inhibitor.

[0035] The findings of the inventors, which underpin the present invention, provide predictive methods that may be used in determining suitable cancer treatment regimens in a subject with cancer, with particular reference to whether the subject will gain benefit, and possibly improve their overall survival, from treatment with VEGF inhibitors, or an alternative treatment regimen that omits the use of VEGF inhibitors. The overall survival may refer to disease-specific survival or disease-free survival.

[0036] In particular, the present inventors identified gene sets including a combination of genes that were found to be differentially expressed and / or enriched in subjects with either a reduced response or improved response to a VEGF inhibitor (e.g. bevacizumab as a representative VEGF inhibitor). In this regard, the expression and / or enrichment of a gene set that is correlated with a reduced response may be designated as a reduced response (RR) gene signature, and the expression and / or enrichment of a gene set that is correlated with a improved response may be designated as an improved response (IR) gene signature, by applying one or both of these gene signatures to a gene expression profile of a sample derived from the subject with cancer, a prediction on the responsiveness of the subject to treatment with a VEGF inhibitor may be made, and subsequently the selection of suitable cancer treatment regimens for the subject decided by a medical practitioner.

[0037] The term "improved response" as used herein refers to the favorable clinical outcome of a patient being treated with a vascular endothelial growth factor (VEGF) inhibitor. The favourable clinical outcome of the improved response may refer to the treatment with a VEGF inhibitor being of greater benefit to the patient compared to the “reduced response’’ of a patient to the same treatment with the VEGF inhibitor. This outcome is characterized by measurable and clinically significant benefits as determined by specific medical criteria, which may include, but are not limited to, the following: TumorShrinkage (i.e. a significant reduction in tumor size as assessed by imaging techniques such as CT, MRI, or PET scans, following the RECIST (Response Evaluation Criteria in Solid Tumors) guidelines); Progression-Free Survival (PFS) (i.e. an extended period during and after treatment during which the patient shows no signs of disease progression; Overall Survival (OS) (i.e. an increase in the duration of survival from the start of the VEGF inhibitor treatment); Reduction in Metastasis (i.e. a decrease in the number and size of metastatic lesions); and / or Improved Quality of Life (i.e. enhanced patient- reported outcomes indicating better physical, functional, and symptomatic well-being).

[0038] The term "reduced response" as used herein refers to an unfavorable or less effective clinical outcome of a patient being treated with a VEGF inhibitor. The less effective clinical outcome of the reduced response may refer to the treatment with a VEGF inhibitor still being of benefit to the patient, but is less effective or beneficial compared to an ‘‘improved response” of a patient to the same treatment with the VEGF inhibitor. This outcome is characterized by minimal, reduced, or no measurable clinical benefits or a decline in clinical status as determined by specific medical criteria, which may include, but are not limited to, the following: Tumor Growth (i.e. an increase in tumor size or failure to achieve significant tumor shrinkage as assessed by imaging techniques); Disease Progression (i.e. a shorter period of progression-free survival (PFS), indicating rapid disease advancement despite treatment); Decreased Overall Survival (OS) (i.e. shorter duration of survival from the start of the VEGF inhibitor treatment); Increased Metastasis (i.e. an increase in the number and size of metastatic lesions); and / or Deterioration in Quality of Life (i.e. worsening patient-reported outcomes indicating a decline in physical, functional, and symptomatic well-being). The term “non-response" refers to cases where there is no measurable benefit, including disease progression or lack of any significant clinical improvement. Thus, unless otherwise specified, the term “reduced response” includes both partial responders and nonresponders in the analytical and predictive methods disclosed herein.

[0039] The term “gene expression profile" as used herein, refers to the measurement of gene expression levels and / or activity in the sample that provides a snapshot of the transcriptional activity of the genes expressed in the sample, that is the gene expression profile may refer to the collective expression levels and / or activity of the genes in the sample. In various embodiments, the gene expression profile may be a “transcriptome gene expression profile" which refers to the comprehensive measurement of the expression levels of all, or substantially all, or a large number of, RNA transcripts present in the sample. In various embodiments, the gene expression profile may refer to the measurement of expression levels of a subset of genes, for example, the genes within the IR and R R gene sets.

[0040] Accordingly, there is provided a method for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, generally comprising the steps of; determining gene expression levels, in a sample obtained from the subject, to obtain a gene expression profile of the sample;analysing the gene expression profile based on a reduced response (RR) gene signature, and / or an improved response (IR) gene signature; and predicting the responsiveness of the subject to treatment with a VEGF inhibitor based on the analysis of the gene expression profile.

[0041] The term “expression level” generally refers to the amount of a gene in a sample. “Expression” generally refers to the process by which information (e.g., gene-encoded and / or epigenetic information) is converted into the structures present and operating in the cell. Therefore, as used herein, “expression” may refer to transcription into a polynucleotide, translation into a polypeptide, or even polynucleotide and / or polypeptide modifications (e.g., posttranslational modification of a polypeptide). Fragments of the transcribed polynucleotide, the translated polypeptide, or polynucleotide and / or polypeptide modifications (e.g., posttranslational modification of a polypeptide) shall also be regarded as expressed whether they originate from a transcript generated by alternative splicing or a degraded transcript, or from a posttranslational processing of the polypeptide, e.g., by proteolysis. “Expressed genes” include those that are transcribed into a polynucleotide as mRNA and then translated into a polypeptide, and also those that are transcribed into RNA but not translated into a polypeptide (for example, transfer and ribosomal RNAs).

[0042] The “amount” or “level” of a gene is a detectable level in a sample. These can be measured by methods known to one skilled in the art and also disclosed herein.

[0043] The term “VEGF inhibitor” may be used interchangeably with “anti-VEGF agent” herein, and refers to a molecule or therapeutic agent that inhibits, decreases, reduces or interferes with the VEGF signalling pathway and biological activities by either directly or indirectly acting on VEGF. The VEGF inhibitor may be an “VEGF antagonist” or “VEGF-specific antagonist” that refers to a molecule capable of binding to VEGF, reducing VEGF expression levels, or neutralizing, blocking, inhibiting, abrogating, reducing, or interfering with VEGF biological activities, including, but not limited to, VEGF binding to one or more VEGF receptors, VEGF signalling, and VEGF mediated angiogenesis and endothelial cell survival or proliferation. For example, a molecule capable of neutralizing, blocking, inhibiting, abrogating, reducing, or interfering with VEGF biological activities can exert its effects by binding to one or more VEGF receptor (VEGFR) (e.g., VEGFR1 , VEGFR2, VEGFR3, membrane-bound VEGF receptor (mbVEGFR), or soluble VEGF receptor (sVEGFR)). Included as VEGF-specific antagonists useful in the methods of the invention are polypeptides that specifically bind to VEGF, anti-VEGF antibodies and antigen-binding fragments thereof, receptor molecules and derivatives which bind specifically to VEGF thereby sequestering its binding to one or more receptors, fusions proteins (e.g., VEGF-Trap (Regeneron)), and VEGFi2i-gelonin (Peregrine). VEGF-specific antagonists also include antagonist variants of VEGF polypeptides, antisense nucleobase oligomers complementary to at least a fragment of a nucleic acid molecule encoding a VEGF polypeptide; small RNAs complementary to at least a fragment of a nucleic acid molecule encoding a VEGF polypeptide; ribozymes that target VEGF; peptibodies to VEGF; and VEGF aptamers. VEGF antagonists also include polypeptides that bind toVEGFR, anti-VEGFR antibodies, and antigen-binding fragments thereof, and derivatives which bind to VEGFR thereby blocking, inhibiting, abrogating, reducing, or interfering with VEGF biological activities (e.g., VEGF signalling), or fusions proteins. VEGF-specific antagonists also include nonpeptide small molecules that bind to VEGF or VEGFR and are capable of blocking, inhibiting, abrogating, reducing, or interfering with VEGF biological activities. Thus, the term “VEGF activities” specifically includes VEGF mediated biological activities of VEGF.

[0044] In various embodiments, the VEGF inhibitor may be administered in combination with one or more additional therapeutic modalities to enhance or complement the therapeutic effect. Such combination therapies may include, but are not limited to, immunotherapy (e.g., immune checkpoint inhibitors such as anti-PD-1 antibodies, anti-PD-L1 antibodies, or anti-CTLA-4 antibodies), chemotherapy (e.g., platinum-based agents, taxanes, antimetabolites), radiation therapy, or other targeted therapies (e.g., EGFR inhibitors, HER2 inhibitors, mTOR inhibitors, or tyrosine kinase inhibitors). The VEGF inhibitor and the additional therapeutic agent(s) may be administered simultaneously, sequentially, or in alternating regimens, and may be provided in the same or different formulations or routes of administration. The combination may result in additive or synergistic effects on inhibiting tumor growth, angiogenesis, or immune evasion, or in overcoming resistance to monotherapy.

[0045] In various embodiments, the VEGF inhibitor may be an anti-VEGF agent, or a VEGF receptor antagonist, including, but not limited to, anti-VEGFR2 antibodies and related molecules (e.g., ramucirumab, tanibirumab, aflibercept), anti-VEGFR1 antibodies and related molecules (e.g., icrucumab, aflibercept (VEGF Trap-Eye; EYLEA®), and ziv-aflibercept (VEGF Trap; ZALTRAP®)), multispecific VEGF antibodies (e.g., MP-0250, vanucizumab (VEGF-ANG2), and multispecific antibodies disclosed in US 2001 / 0236388), multispecific antibodies including combinations of two of anti-VEGF, anti-VEGFR1 , and anti-VEGFR2 arms, anti-VEGFA antibodies (e.g., bevacizumab, sevacizumab), anti-VEGFB antibodies, anti-VEGFC antibodies (e.g., VGX-100), anti-VEGFD antibodies, and nonpeptide small molecule VEGF antagonists (e.g., pazopanib, axitinib, vandetanib, stivarga, cabozantinib, lenvatinib, nintedanib, orantinib, telatinib, dovitinig, cediranib, motesanib, sulfatinib, apatinib, foretinib, famitinib, and tivozanib). In some examples, the VEGF antagonist may be a tyrosine kinase inhibitor, including a receptor tyrosine kinase inhibitors (e.g., a multi-targeted receptor tyrosine kinase inhibitor such as sunitinib or axitinib).

[0046] The term “anti-VEGF antibody” as used herein, refers to an antibody that binds to VEGF with sufficient affinity and specificity. In certain embodiments, the antibody will have a sufficiently high binding affinity for VEGF, for example, the antibody may bind hVEGF with a Kd value of between 100 nM-1 pM. Antibody affinities may be determined, e.g., by a surface plasmon resonance based assay (such as the BIAcore® assay as described in PCT Application Publication No. W02005 / 012359); enzyme-linked immunoabsorbent assay (ELISA); and competition assays (e.g. radioimmunoassays (RIAs)). The anti-VEGF antibody can be used as a therapeutic agent in targeting and interfering withdiseases or conditions wherein the VEGF activity is involved. The anti-VEGF antibody may be a recombinant humanized anti-VEGF monoclonal antibody generated according to Presta et al. (Cancer Res. 57:4593-4599, 1997), including but not limited to the antibody known as bevacizumab (BV:AVASTIN@). The anti-VEGF antibody “bevacizumab (BV),” also known as “rhuMAb VEGF’’ or “AVASTIN®,” is a recombinant humanized anti-VEGF monoclonal antibody generated according to Presta et al. (Cancer Res. 57:4593-4599, 1997). It comprises mutated human lgG1 framework regions and antigen-binding complementarity-determining regions from the murine anti-hVEGF monoclonal antibody A.4.6.1 that blocks binding of human VEGF to its receptors. Approximately 93% of the amino acid sequence of bevacizumab, including most of the framework regions, is derived from human IgG 1 , and about 7% of the sequence is derived from the murine antibody A4.6.1 . Bevacizumab has a molecular mass of about 149,000 daltons and is glycosylated. Bevacizumab and other humanized anti- VEGF antibodies are further described in U.S. Pat. No. 6,884,879 issued Feb. 26, 2005, the entire disclosure of which is expressly incorporated herein by reference. Additional preferred antibodies include the G6 or B20 series antibodies (e.g., G6-31 , B20-4.1 ), as described in PCT Application Publication No. WO 2005 / 012359. For additional preferred antibodies see U.S. Pat. Nos. 7,060,269, 6,582,959, 6,703,020; 6,054,297; WO98 / 45332; WO 96 / 30046; W094 / 10202; EP 0666868B1 ; U.S. Patent Application Publication Nos. 2006009360, 20050186208, 20030206899, 20030190317, 20030203409, and 200501 12126; and Popkov et al., (Journal of Immunological Methods 288:149-164, 2004). In various embodiments, the anti-VEGF antibody may be a multi-specific antibody that binds VEGF and at least one additional target antigen, such as CD3 or CD271 (e.g., for T cell engagement) or a tumor-associated antigen. Such multi-specific antibodies retain VEGF-neutralizing activity and may provide enhanced therapeutic efficacy through dual targeting mechanisms. Multi-specific formats may include full-length IgG-like structures, tandem single-chain variable fragments (scFvs), or other engineered antibody formats capable of simultaneously engaging VEGF and a secondary target.

[0047] In various embodiments, the VEGF inhibitor is an anti-VEGF antibody, preferably bevacizumab.

[0048] The term "subject", as used herein in the context of the methods, refers to a warm-blooded animal, preferably a mammal, more preferably a human. Said subject may be awaiting or receiving medical treatment, or is, or will become the subject of a medical procedure, or is being monitored for the development of cancer or any condition or disease treatable with a VEGF inhibitor. In the context of methods described herein, the subject may be termed as “non-healthy subject”.

[0049] In various embodiments, the subject has been diagnosed with the cancer prior to the method disclosed herein being carried out, and the subject has either received cancer treatment or is cancer treatment-naive or treated with cancer therapies excluding VEGF inhibitors, such that prediction of responsiveness to VEGF inhibitor therapy remains clinically relevant. The term “cancer treatment- naive” refers to individuals who have been diagnosed with cancer but have not yet received any form of therapeutic intervention for their disease.

[0050] In various embodiments, the subject with cancer has a cancer selected from the group consisting of: colorectal cancer; breast cancer; ovarian cancer, liver cancer, biliary duct cancer, prostate cancer, endometrial cancer, cervical cancer, lung cancer, gastric cancer, oesophageal cancer, pancreatic, bone cancer, bladder cancer, brain cancer, head and neck cancer, thyroid cancer, skin cancer, renal cancer, and oesophagus cancer and combinations thereof.

[0051] In various embodiments, the cancer comprises brain cancer ovarian cancer, lung cancer, colorectal cancer and / or head and neck cancer. In various embodiments, the brain cancer is a glioblastoma, preferably a proneural or non-proneural glioblastoma. In various embodiments, the glioblastoma may be a non-proneural glioblastoma, preferably a mesenchymal, or proliferative (classical) glioblastoma, or is an unclassified glioblastoma. In various embodiments, the head and neck cancer may be nasopharyngeal cancer.

[0052] The term “sample,” as used herein, refers to a composition that is obtained or derived from a subject and / or individual of interest that contains a cellular and / or other molecular entity that is to be characterized and / or identified, for example, based on physical, biochemical, chemical, and / or physiological characteristics. For example, the phrase “disease sample" and variations thereof (i.e. cancer or tumour sample) refers to any sample obtained from a subject of interest that would be expected or is known to contain the cellular and / or molecular entity that is to be characterized. Samples include, but are not limited to, tissue samples, primary or cultured cells or cell lines, cell supernatants, cell lysates, platelets, serum, plasma, vitreous fluid, lymph fluid, synovial fluid, follicular fluid, seminal fluid, amniotic fluid, milk, whole blood, blood-derived cells, urine, cerebro-spinal fluid, saliva, sputum, tears, perspiration, mucus, tumour lysates, and tissue culture medium, tissue extracts such as homogenized tissue, tumour tissue, cellular extracts, and combinations thereof.

[0053] In various embodiments, the sample is a tumour sample, preferably the tumour is a glioblastoma tumour, ovarian tumour, lung tumour, head and neck tumour or colorectal tumour. In various embodiments, the sample is obtained or derived from a head and neck tumour, preferably a nasopharyngeal tumour, or a glioblastoma tumour, preferably a proneural or non-proneural glioblastoma tumour. In various embodiments, the glioblastoma tumour is a non-proneural glioblastoma tumour, preferably a mesenchymal, or proliferative (classical) glioblastoma tumour, or is an unclassified glioblastoma tumour.

[0054] In various embodiments, the sample is a tumour sample comprising one or more cell types including but not limited to epithelial cells, immune cells, stromal cells, fibroblast cells or combinations thereof in an admixture of cells. In various embodiments, the sample is a tumour sample comprising epithelial cells. In various embodiments, the sample is a tumour sample comprising fibroblast cells. In various embodiments, the tumour sample comprises stromal cells (eg. endothelial cells or fibroblasts or macrophage cells) admixed with epithelial cells in the epithelial compartment.

[0055] In various embodiments, the sample is a tumour sample comprising tumour cells from the tumour epithelial compartment, which includes epithelial cells, as well as microenvironment cells (e.g. immune cells, stromal cells and fibroblast cells) admixed with epithelial cells within the same compartment.

[0056] The term “tissue sample” or “cell sample” as used herein, refers to a collection of similar cells obtained from a tissue of a subject or individual. The source of the tissue or cell sample may be solid tissue as from a fresh, frozen and / or preserved organ, tissue sample, biopsy, and / or aspirate; blood or any blood constituents such as plasma; bodily fluids such as cerebral spinal fluid, amniotic fluid, peritoneal fluid, or interstitial fluid; cells from any time in gestation or development of the subject. The tissue sample may also be primary or cultured cells or cell lines. Optionally, the tissue or cell sample is obtained from a disease tissue / organ. For instance, a “tumour sample” is a tissue sample obtained from a tumour (e.g., a brain or nasopharyngeal tumour) or other cancerous tissue. The tissue sample may contain a mixed population of cell types (e.g., tumour cells and non-tumour cells, cancerous cells and non-cancerous cells). In particular, the tumour sample may be obtained from one or more compartments within the tumour. For example, the tumour sample may comprise or consist of an epithelial compartment that contains epithelial cells.

[0057] In various embodiments, the sample may also contain compounds which are not naturally intermixed with the tissue in nature such as preservatives, anticoagulants, buffers, fixatives, nutrients, antibiotics, or the like.

[0058] The term “section” of a tissue sample as used herein, refers to a single part or piece of a tissue sample, for example, a thin slice of tissue or cells cut from a tissue sample (e.g., a tumour sample). It is to be understood that multiple sections of tissue samples may be taken and subjected to analysis, provided that it is understood that the same section of tissue sample may be analysed at both morphological and molecular levels, or analysed with respect to polypeptides and / or polynucleotides.

[0059] In various embodiments, the sample is a tissue sample, preferably a fixed tissue sample, more preferably a formalin fixed paraffin embedded (FFPE) tissue sample, fresh, or frozen. In this regard, it will be appreciated that clinical biopsy samples are routinely processed and stored as FFPE tissue blocks in hospitals. As such, the methods disclosed herein are compatible with existing patient material and hospital processes, such that no additional tissue is necessarily needed to be taken from the patient.

[0060] In various embodiments, the tissue sample is microdissected, preferably the microdissected tissue sample comprises epithelial cells, or epithelial cells admixed with stromal cells, or is a tumor epithelial compartment or a tumor microenvironment compartment. The term “microdissection” or “microdissected” as used herein refers to the use of a laboratory technique to isolate specific cells, tissues, or structures from a larger sample under a microscope. This process involves using fine toolsor laser technology to precisely excise areas of interest from a complex tissue, allowing for detailed analysis of particular cells or structures without contamination from surrounding tissues. Microdissection is widely used in research and clinical settings for various purposes, including genetic analysis, cancer research, and developmental biology. Microdissection methods include, but are not limited to, laser capture microdissection (LCM); manual microdissection using fine tools under magnification; and mechanical methods such as tissue punching (e.g., using a core biopsy punch or manual punch device) to extract defined tissue regions. Laser capture microdissection (LCM), uses a focused laser to cut and capture specific regions of tissue. Laser capture microdissection (LCM), may also be termed microdissection, laser microdissection (LMD), or laser-assisted microdissection (LMD or LAM). In the representative examples disclosed herein, LCM was used as a method to extract tissue for RNA-seq. LCM technology can harvest the cells of interest directly or can isolate specific cells by cutting away unwanted cells to give histologically pure enriched cell populations. A laser is coupled into a microscope and focuses onto the tissue on the slide. By movement of the laser an element is cut out and separated from the adjacent tissue.

[0061] Advantageously, the use of LCM improves purity of tumor epithelial and microenvironment compartments / regions obtained from samples separately, giving a stronger signal from each compartment / region. The purity of the tumor epithelial compartment in this instance is greater than techniques which do not use LCM. Moreover, LCM is able to separate a tumor epithelial compartment from the microenvironment compartment which allows the signal from each compartment of the cancer to be enriched, thus improving the ability of the measured gene expression levels / profile and correlated gene signatures to classify subjects with cancer as being responsive or non-responsive to VEGF inhibitors.

[0062] In various embodiments, the sample comprises or consists of a microdissection of the epithelial compartment in a tumour tissue, such that the sample comprises tumour epithelial cells. In particular, the sample comprises tumour cells from the tumour epithelial compartment comprising epithelial cells and other microenvironment cell types such as immune and stromal cells. In various embodiments, the the microdissected tissue sample comprises stromal cells (eg. endothelial cells or fibroblasts or macrophage cells) admixed with epithelial cells in the tumour epithelial compartment. Advantageously, determining the gene expression levels from a microdissected tumour sample, optionally in combination with RNA-seq technique, produces a greater signal purity and depth of information in generating the gene expression profile and allows for improved accuracy when the RR and / or IR gene signatures are applied for analysis of the gene expression profile and hence provides a more accurate prediction method of responsiveness disclosed herein.

[0063] In various embodiments, only a single sample obtained from the subject is used, or is necessary, for the method disclosed herein to be able to predict the responsiveness of the subject to treatment with a VEGF inhibitor.

[0064] The gene signatures disclosed herein comprise a gene set of at least two genes selected from the genes listed in Table 1 below.

[0065] The term “at least two”, as used herein, refers to 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 or 44 or more. It is understood that if used in relation to the gene set and combination of at least two genes, it relates to the number of genes to be included in the gene set.

[0066] In various embodiments, the gene set comprises at least two genes selected from the group consisting of: FAM1 10C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL1 1 A1 , ID2, ABCA2, HMOX1 , HLA- C, FTL, GJA5, COL5A2, MAN1 A1 , CD248, LUM, FMO3, MTRNR2L1 , GUCY1 A1 , and MCL1 . In various embodiments, the gene set may be termed as a Reduced Response (RR) gene set.

[0067] In various embodiments, the RR gene set comprises 2, 3, 4, 5, 6, 7, 8, 9, 10, 1 1 , 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 or 44 of the genes.

[0068] The enrichment of fibroblast related genes in tumour samples were shown herein to be correlated with the reduced response of a subject to treatment with a VEGF inhibitor. Accordingly, the genes selected for inclusion in the RR gene set may be fibroblast cell-specific genes that collectively define a fibroblast cell gene signature capable of identifying a subject with a reduced response to said treatment.

[0069] In various embodiments, the RR gene set comprises fibroblast cell-specific genes that define a fibroblast cell gene signature, preferably the fibroblast cell-specific genes comprise at least the following genes:o SPARC, SFRP4, ISLR, C0L11 A1 , COL5A2, and LUM; or o SPARC, HLA-A, CCL2, MYL9, HM0X1 , HLA-C, and LUM.

[0070] In various embodiments, the RR gene set consists of: FAM110C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL1 1 A1 , ID2, ABCA2, HMOX1 , HLA-C, FTL, GJA5, COL5A2, MAN1 A1 , CD248, LUM, FMO3, MTRNR2L1 , GUCY1 A1 , and MCL1 .

[0071] In various embodiments, the gene set comprises at least two genes selected from the group consisting of: ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 . In various embodiments, the gene set may be termed as an Improved Response (IR) gene set.

[0072] In various embodiments, the IR gene set comprises 2, 3, 4, 5, 6, 7, 8, 9, 10, 1 1 , 12, 13, or 14, of the genes.

[0073] In various embodiments, the IR gene set consists of: ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF1 1 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 .

[0074] Accordingly, the method disclosed herein for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, may comprise: determining gene expression levels, in a sample obtained from the subject, to obtain a gene expression profile, of the sample; analysing the gene expression profile based on a reduced response (RR) gene signature, and / or an improved response (IR) gene signature, wherein the RR gene signature comprises a RR gene set comprising at least two genes selected from the group consisting of: FAM110C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL11A1 , ID2, ABCA2, HMOX1 , HLA-C, FTL, GJA5, COL5A2, MAN1 A1 , CD248, LUM, FMO3, MTRNR2L1 , GUCY1A1 , and MCL1 , and wherein the IR gene signature comprises a IR gene set comprising at least two genes selected from the group consisting of: ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 ; and predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor based on the analysis of the gene expression profile.

[0075] In this regard, the combination of genes that form the gene set of the gene signature, and their respective gene expression levels / profile in a sample, may be described as a biomarker or indicator,e.g., predictive, and / or diagnostic, which can be detected in the sample. The gene signatures serve as indicators of the subject’s responsiveness to treatment of cancer, as well as particular subtypes of cancer characterized by certain, molecular, pathological, histological, and / or clinical features.

[0076] In the methods disclosed herein, the step of determining expression levels to obtain a gene expression profile, may comprise quantifying the amount (for example, absolute amount, relative amount, or concentration) of a gene product in a sample. A gene product may include, for example, a protein or RNA transcript encoded by the gene, or a fragment of the protein or RNA transcript. In various embodiments, determining gene expression levels includes obtaining quantitative gene expression levels of genes.

[0077] In this regard, the sample may be subjected to various processing steps before being used in the methods of the invention, including common isolation and / or purification steps, for example steps to extract DNA or RNA from cells that may be present in the sample. In particular, the sample may be subjected to a variety of well-known post-collection preparative and storage techniques (for example, fixation, storage, freezing, lysis, homogenization, DNA or RNA extraction, ultrafiltration, concentration, evaporation, centrifugation, etc.) prior to assessing the gene expression levels in the sample.

[0078] In various embodiments, the amount of the gene product may be assessed by any suitable method known to a person having skill in the art. The skilled person will be aware of many suitable methods that may be used in determining and measuring the expression levels of genes in a sample, including but not limited to PCR based techniques (such as Taqman or sybr green gene expression assays) or mRNA / cDNA hybridisation techniques (such as gene expression micro array, chips, or NanoString technology) or Next-Generation Sequencing (NGS) techniques (such as RNA-seq) to yield the required gene expression profile. Standard assay normalization methods and batch effect correction methods suitable to each type of assay may, or may not, be employed. In various embodiments, the method includes normalizing the gene expression levels including, for example, normalizing the level of the RNA transcripts to obtain normalized gene expression levels.

[0079] In various embodiments, a suitable method or technique may be selected to allow expression levels of genes in the gene set to be measured, if present and expressed in the sample, for example a method or technique that is not limited in the number of genes, and their expression levels, that can be identified is preferable. In various embodiments, an overall expression level of the gene set may be measured for obtaining a gene expression profile of the gene signatures in the sample for use in subsequent analysis steps of the method disclosed herein.

[0080] In various embodiments, the gene expression levels in the sample may be measured by using one or more techniques comprising quantitative PCR (qPCR), Reverse Transcription PCR (RT-PCR), gene expression microarrays, Northern Blotting, In Situ Hybridization (ISH), RNA-Sequencing technique (RNA-seq), NanoString nCounter Analysis, and / or Digital Droplet PCR (ddPCR).

[0081] In various embodiments, the step of determining gene expression levels comprises measuring levels of RNA transcripts in the sample. The measured levels of RNA transcripts in the sample may provide a transcriptional gene expression profile of the subject’s sample.

[0082] In various embodiments, the step of determining gene expression levels may further comprise normalizing the level of the RNA transcripts to obtain normalized gene expression levels, however, normalizing gene expression levels may not be needed.

[0083] The term “RNA-seq,” also called “Whole Transcriptome Shotgun Sequencing (WTSS),” refers to the use of high-throughput sequencing technologies to sequence and / or quantify cDNA to obtain information about a sample’s RNA content. Publications describing RNA-seq include: Wang et al. Nature Reviews Genetics 10(1 ):57-63, 2009; Ryan et al. BioTechniques 45(1 ):81 -94, 2008; and Maher et al. Nature 458(7234) :97-101 , 2009.

[0084] In various embodiments, the gene expression levels in the sample may be measured by RNA- sequencing. In this regard, the step of determining gene expression levels generates a gene expression profile of the sample. Advantageously, gene expression profile data produced with the RNA-Seq technique may allow for high-quality data to be produced from degraded, archival tissue samples. Moreover, in embodiments where the sample is a formalin-fixed paraffin embedded (FFPE) tissue sample, the RNA-Seq technique may be optimised to identify genes throughout the entire transcriptome, where in comparison, gene expression microarrays are limited in the number of genes that can be identified.

[0085] In various embodiments, the step of determining gene expression levels further comprises providing a gene expression count based on RNA sequencing data, and providing a gene expression matrix based on the gene expression count, whereby the gene expression profile of the sample is based on gene expression matrix. In particular, the sequenced RNA data extracted from the sample may be used to generate raw sequence reads. These reads may then be aligned to a reference genome or transcriptome, and the number of reads mapping to each gene counted. This count data represents the raw expression levels of each gene in the sample. Next, this gene expression count may be used to create a gene expression matrix. The gene expression matrix is a structured data table where rows correspond to individual genes, and columns correspond to different samples. Each cell in the matrix contains the expression count for a particular gene in a specific sample. This matrix format facilitates easy comparison and analysis of gene expression levels in one or more samples. Finally, the gene expression profile of the sample may be derived from the gene expression matrix. The gene expression profile represents the gene expression levels within the sample, capturing the relative abundance of each gene’s transcripts. This gene expression profile can be used for various downstream analyses, such as identifying differentially expressed genes, conducting pathway analysis, and performing gene set enrichment analysis.

[0086] In various embodiments, the gene expression profile of the sample is obtained by using NanoString nCounter analysis, which may include the use of gene probes specific for a number of genes, which may include genes of the gene set described herein. In various embodiments, RNA-Seq technique may be combined with NanoString nCounter analysis to obtain the gene expression profile of the sample.

[0087] The step of determining gene expression levels provides data / information on the expression levels and / or expression profile of sample (including expression levels and / or expression profile of the gene set) for the subsequent analysis step of the method disclosed herein.

[0088] In the methods disclosed herein, the step of analysing may comprise the use of a suitable method / technique that analyses, interprets and correlates the gene expression profile of the sample, in respect of the RR and / or IR gene signatures, with the responsiveness of the subject to treatment with the VEGF inhibitor. A variety of suitable and well-known methods or techniques (i.e. computer programs / software / algorithms) are available for analysing gene expression profiles based on one or more gene sets that define gene signatures.

[0089] In various embodiments, the step of analysing may comprise differential gene expression (DGE) analysis and / or gene set enrichment analysis and / or gene set scoring analysis of the gene expression profile.

[0090] In various embodiments, the gene set scoring analysis may comprise one or more gene set scoring methods capable of generating a gene score (i.e. composite gene score) based on the combined expression levels of each gene in the gene set using logistic regression or other suitable statistical methods. The composite gene score is used to predict the responsiveness of the subject to treatment with the VEGF inhibitor.

[0091] In various embodiments, the gene set scoring methods may include, but are not limited to: Mean or Median Expression Scoring, Principal Component Analysis (PCA), Z-score Transformation, Pathway Analysis Methods (e.g., KEGG, Reactome), a logistic regression method, or Network-Based Methods.

[0092] This gene set score may be compared to a reference gene set score using statistical methods to assess the significance. It will be appreciated that the gene set score obtained from the gene set scoring methods are based on, and reflect differential expression levels, of the defined gene set in the RR and / or IR gene signatures. The reference gene set score may be obtained from the same sample or a reference sample from the subject or another individual who is not the subject. In various embodiments, the reference gene set score may be derived from a reference sample obtained from the subject.

[0093] In various embodiments, the significance of the gene set score of the RR and / or IR gene signature may be determined based on a threshold score (e.g. a cut-off score). A threshold score may be associated with a statistic. As used herein, a threshold score is a score above or below which the gene set score of the RR and / or IR gene signature is assigned significance, in particular with respect to a subject’s responsiveness to therapy with a VEGF inhibitor. In various embodiments, the gene set score may range -1 to 1 (or in the case of applying a Z-score method -3 to 3), whereby positive scores indicate upregulation, and negative scores indicate downregulation.

[0094] In various embodiments, the subject is predicted to have a reduced response (RR) to treatment with the VEGF inhibitor if a positive gene set score of the RR gene signature is obtained; or the subject is predicted to have an improved response to treatment with the VEGF inhibitor if a positive gene set score of the IR gene signature is obtained.

[0095] In various embodiments, the gene set enrichment analysis comprises one or more gene set enrichment methods, whereby the output of gene set enrichment methods is used to predict the responsiveness of the subject to treatment with the VEGF inhibitor. It will be appreciated that the gene set enrichment methods may be combined with one or more of the gene set scoring methods. In various embodiments, any suitable gene set enrichment method may be used that allows each sample to be analysed independently without having to normalize across an entire cohort of samples.

[0096] In various embodiments, the gene set enrichment methods may include, but are not limited to: Gene Set Enrichment Analysis (GSEA), DAVID (Database for Annotation, Visualization, and Integrated Discovery), Enrichr, GOseq, ClusterProfiler, ROAST (Rotation Gene Set Testing), Camera (Correlation Adjusted Mean Rank), TopGO, Single-Sample Gene Set Enrichment Analysis (ssGSEA), Gene Set Variation Analysis (GSVA), Pathway Level Analysis of Gene Expression (PLAGE), Pathway Analysis Using Expression Data (PADOG), QuSAGE (Quantitative Set Analysis for Gene Expression), or ROMER (Rotation Gene Set Testing).

[0097] The gene set enrichment methods provide an output, a gene set enrichment score indicating the responsiveness of the subject to treatment with the VEGF inhibitor. The gene set enrichment score may or may not be a normalized gene set enrichment score. In various embodiments, the gene set enrichment score may range -1 to 1 , whereby a positive enrichment score indicates that the gene set is enriched at the top of the ranked list of genes, while a negative enrichment score indicates enrichment at the bottom of the list. The magnitude of the enrichment score reflects the degree of enrichment: scores closer to 1 or -1 indicate stronger enrichment. This gene set enrichment score may be compared to a reference gene set enrichment score or distribution using statistical methods to assess the significance. It will be appreciated that the enrichment score obtained from the gene set enrichment methods are based on, and reflect differential expression levels, of the defined gene set in the RRand / or IR gene signatures. In various embodiments, the gene set enrichment methods are applied individually to each sample. In other words, each sample is scored independently of other samples.

[0098] The reference gene set enrichment score may be obtained from the same sample or a reference sample from the subject or another individual that is not the subject. In various embodiments, the reference gene set enrichment score may be derived from a reference sample obtained from the subject. In various embodiments, the sample and reference sample are obtained from the same tissue sample and tumour microenvironment in the subject, but are obtained from different microenvironment compartments. For example, the sample may comprise distinct cell types to the reference sample, preferably the sample comprises epithelial cells, immune cells, stromal cells etc, and the reference sample does not comprise epithelial cells and is obtained from a different compartment in the tumour microenvironment. In various embodiments, the sample is a tumour sample comprising one or more of epithelial cells, immune cells, stromal cells, fibroblast cells or combinations thereof. In various embodiments, the sample is a tumour sample comprising tumour cells from the tumour epithelial compartment, which includes epithelial cells, as well as microenvironment cells including immune cells, stromal cells and fibroblast cells admixed with epithelial cells within the same compartment.

[0099] In various embodiments, the significance of the gene set enrichment score of the RR and / or IR gene signature to predict the responsiveness of the subject to treatment with the VEGF inhibitor is based on a threshold enrichment score. As used herein, a threshold enrichment score is a score above or below which the gene set enrichment score of the RR and / or IR gene signature is assigned significance, in particular with respect to a subject’s responsiveness to therapy with a VEGF inhibitor. The threshold enrichment score may be derived from a reference sample or the same sample used in the methods of the subjects (i.e. in the case of a single-sample gene set enrichment method being applied). The threshold may be based on statistical metrics such as normalized enrichment scores (NES), p-values, or false discovery rates (FDR). In various embodiments, a p-value less than 0.05 (p < 0.05) or an adjusted p-value (FDR) less than 0.25 for gene set enrichment may be considered statistically significant. In various embodiments, the threshold enrichment score may be derived from comparison of responders and non-responders in a discovery cohort, and may be selected to optimize separation of clinical outcomes, such as progression-free or overall survival. In one embodiment, for single-sample gene set enrichment scores (ssGSEA), a threshold may be defined based on a statistically significant difference in scores observed between clinical outcome groups (e.g., ssGSEA score threshold of approximately 23000 based on the midpoint between 20670 for improved response and 25569 for partial response in NPC). Alternatively, in methods using across-sample normalization (e.g., PLAGE), the threshold may correspond to the y-axis intercept of a linear regression model correlating enrichment score with survival, or other empirically derived cut-offs. Thresholds may be validated or refined across multiple datasets and tumor types and need not be fixed; they may be dynamically defined based on internal controls or reference distributions within a clinical or experimental cohort.

[0100] In this regard, the gene set enrichment methods apply the RR and / or IR gene signature to the gene expression profile to analyse the enrichment of the genes within the gene set of the gene signatures, across the gene expression profile of the sample. The result is an enrichment score correlating with the relative overrepresentation and enrichment of the gene set within the sample’s gene expression profile. In particular, a gene set's enrichment score represents the activity level of the biological process in which the gene set's members are co-ordinately up-regulated or down-regulated. A positive enrichment score indicates that the genes in the gene set are collectively increased / upregulated in the sample (i.e. enriched), whereas a negative enrichment score indicates that the genes in the gene set are collectively decreased / downregulated in the sample (i.e. disenriched). The statistical significance of the enrichment scores may be determined by calculating p-values to assess the likelihood that the enrichment scores could have occurred by chance.

[0101] In various embodiments, the one or more gene set enrichment methods comprise a singlesample gene set enrichment method designed to assess the expression / activity of the gene sets in individual samples, rather than comparing groups of samples. These single-sample gene set enrichment methods quantify the extent to which a gene set is upregulated or downregulated in a single sample. The person skilled in the art would be able to suitable select and adapt an appropriate singlesample gene set enrichment method for use with the methods disclosed herein.

[0102] In various embodiments, the one or more gene set enrichment methods comprise ssGSEA and GSVA. In various embodiments, normalization across multiple samples is not needed for the ssGSEA and GSVA methods. The single-sample gene set enrichment approach, is advantageous in that every patient sample is scored independently, without the need to reference a larger cohort of samples, which may prove crucial in a real-life clinical testing scenario, where patient samples are received one at a time. In various embodiments, a PLAGE (Pathway Level Analysis of Gene Expression) method may be applied for single-sample gene set enrichment scoring by evaluating the expression levels of genes within the gene sets, and then calculating a score that represents the activity of the gene set.

[0103] In various embodiments, a positive gene set enrichment score for the RR gene signature, indicates that the RR gene set is overrepresented and enriched in the sample, suggesting that the genes are upregulated, and it may be predicted that the subject has a reduced response (RR) to treatment with the VEGF inhibitor. Similarly, a positive gene set enrichment score for the IR gene signature, indicates that the IR gene set is overrepresented in the sample, suggesting that the genes are upregulated, and it may be predicted that the subject has an improved response (IR) to treatment with the VEGF inhibitor. The significance of the enrichment score may be assessed using a p-value.

[0104] In various embodiments, if the enrichment score indicates an enrichment of expressed genes in the RR gene set, the subject may be predicted to be non-responsive to treatment with the VEGF inhibitor (i.e. enriched RR gene signature). It will be appreciated that the absence of an enriched RR gene signature in the subject sample may indicate that the subject is responsive to the treatment. Thus,in various embodiments, if the enrichment score fails to indicate an enrichment of expressed genes in the RR gene set, the subject may be predicted to be responsive to treatment with the VEGF inhibitor.

[0105] As used herein, a responsive subject is a subject with cancer who is predicted to benefit from treatment with a VEGF inhibitor. Responsiveness is indicated by an improved response, which refers to a favorable clinical outcome characterized by measurable and clinically significant benefits. These benefits may include tumor shrinkage, longer progression -free survival (PFS), increased overall survival (OS), reduction in metastasis, and / or improved quality of life. A responsive subject typically shows an enriched IR (improved response) gene signature and / or a disenriched or absent RR (reduced response) gene signature based on analysis of the subject’s gene expression profile. As used herein, a non- responsive subject is a subject with cancer who is predicted to derive little or no meaningful benefit from treatment with a VEGF inhibitor. Non-responsiveness includes reduced responses, which may involve minimal or no tumor shrinkage, disease progression, shortened PFS or OS, increased metastasis, or deteriorated quality of life. A non-responsive subject typically shows an enriched RR gene signature in their gene expression profile. The term “non-responsive” encompasses both partial responders and non-responders, and these classifications are used interchangeably with “reduced response” in the methods described.

[0106] In various embodiments, the gene set enrichment score may be compared to a reference gene set enrichment score, whereby a higher RR gene set enrichment score may correlate with the subject having a greater reduction in response to treatment with the VEGF inhibitor (i.e. high-RR), and a lower RR gene set enrichment score may correlate with the subject having a lesser reduction in response to treatment with the VEGF inhibitor (i.e. low-RR).

[0107] In various embodiments, based on the enrichment score, a y-axis intercept of a linear regression model may be used to further classify a non-responsive subject as having either a high-RR or a low-RR to the treatment with a VEGF inhibitor, which may be used to predict a length of survival or overall survival of the subject from the cancer. As used herein, “overall survival” and “OS” refer to the length of time from either the date of diagnosis or the start of treatment for cancer that the patient is still alive. In this regard, a positive gene set enrichment score for the RR gene signature is negatively correlated with the length and overall survival in the subject.

[0108] In various embodiments, the enrichment score may be evaluated to predict a length of survival of the subject from the cancer. In various embodiments, the gene set enrichment score is correlated with a predicted length of survival of the subject, wherein a high enrichment score (i.e. of the RR gene signature) predicts a reduced length of survival compared to a low enrichment score. In various embodiments, the length of survival is at least 5 months, at least 10 months, at least 15 months, at least 20 months, at least 25 months, at least 30 months, at least 35 months, or at least 40 months.

[0109] In various embodiments, a Kaplan-Meier curve survival analysis of the gene set enrichment score may be performed to confirm if the subject is predicted to be responsive to treatment with a VEGF inhibitor as well as predicting the length of survival or overall survival of the subject from the cancer.

[0110] In various embodiments, if the enrichment score indicates an enrichment of expressed genes in the IR gene set, the subject may be predicted to be responsive to treatment with the VEGF inhibitor (i.e. enriched IR gene signature). It will be appreciated that the absence of an enriched IR gene signature in the subject sample may indicate that the subject is non-responsive or have a reduced response to the treatment. Thus, in various embodiments, if the enrichment score fails to indicate an enrichment of expressed genes in the IR gene set, the subject may be predicted to be non-responsive or have a reduced response to treatment with the VEGF inhibitor.

[0111] In various embodiments, the gene set enrichment score may be compared to a reference gene set enrichment score, whereby a higher IR gene set enrichment score may correlate with the subject having a greater response to treatment with the VEGF inhibitor (i.e. high-IR), and a lower IR gene set enrichment score may correlate with the subject having a lesser improved response to treatment with the VEGF inhibitor (i.e. low-IR). Accordingly, the enrichment score may correlate with the degree of treatment response, and the reference gene set enrichment score may be derived from reference samples (e.g. samples in the discovery cohort and / or validation cohorts of patients as shown in the working examples disclosed herein).

[0112] Alternatively, or in addition to the gene set enrichment methods and / or gene set scoring methods, the step of analysing may comprise comparing the gene expression profile of the sample to a reference gene expression profile, wherein a differential gene expression profile in respect of the RR and / or IR gene signature compared to the reference gene expression profile in respect of the RR and / or IR gene signature, may be used to predict the responsiveness of the subject to treatment with the VEGF inhibitor. In other words, a differential expression level of the RR or IR gene set in the sample may indicate the responsiveness of the subject to treatment with the VEGF inhibitor.

[0113] In various embodiments, the reference expression profile may be derived from a reference sample or control sample obtained from the subject. The term “reference sample,” “reference cell,” “reference tissue,” “control sample,” “control cell,” or “control tissue,” as used herein, refers to a sample, cell, tissue, standard, or level that is used for comparison purposes. In various embodiments, a reference sample, reference cell, reference tissue, control sample, control cell, control tissue, or reference level is obtained from a healthy and / or nondiseased part of the body (e.g., tissue or cells) of the same subject or individual. For example, the reference sample, reference cell, reference tissue, control sample, control cell, control tissue, or reference level may be healthy and / or non-diseased cells or tissue adjacent to the diseased cells or tissue (e.g., cells or tissue adjacent to a tumour). In another embodiment, a reference sample is obtained from an untreated tissue and / or cell of the body of the same subject or individual. In yet another embodiment, a reference sample, reference cell, referencetissue, control sample, control cell, control tissue, or reference level is obtained from a healthy and / or non-diseased part of the body (e.g., tissues or cells) of an individual who is not the subject or individual. In even another embodiment, a reference sample, reference cell, reference tissue, control sample, control cell, control tissue, or reference level is obtained from an untreated tissue and / or cell of the body of an individual who is not the subject or individual.

[0114] In various embodiments, a “reference gene expression profile” refers to the expression level of a collection of genes or the gene set which is indicative of the responsiveness of the subject to treatment with a VEGF inhibitor. The “reference expression gene profile” may be obtained from publicly available gene expression datasets, whereby such datasets can provide a broad context for normalization.

[0115] In various embodiments, the reference expression levels may be derived from a reference sample obtained from the subject. In various embodiments, the sample and reference sample are obtained from the same tumour microenvironment in the subject, but are obtained from different compartments. For example, the sample may comprise distinct cell types to the reference sample, preferably the sample comprises epithelial cells and the reference sample does not comprise epithelial cells.

[0116] In various embodiments, the reference expression levels is derived from a control sample obtained from a healthy subject. In various embodiments, the control sample is obtained from an individual that is not the subject, and the individual may be a healthy subject with no cancer.

[0117] As will be appreciated, the term “differential gene expression profile" refers to a significant variation in the expression levels of a gene or collection of genes in the gene set of the gene signatures disclosed herein, relative to a reference expression profile. Genes identified as differentially expressed may be upregulated (higher expression) or downregulated (lower expression) relative to the reference expression level. Statistically significant differences may be determined by applying statistical tests well- known to those skilled in the art.

[0118] Accordingly, the step of analysing may comprise differential gene expression analysis using methods readily known to those skilled in the art, which may typically involve normalization, statistical testing, and interpretation of results. In various embodiments, differential gene expression analysis comprises one or more differential gene expression methods, whereby the output of differential gene expression methods is used to predict the responsiveness of the subject to treatment with the VEGF inhibitor.

[0119] In various embodiments, the one or more Differential Gene Expression (DGE) analysis methods includes but is not limited to: DESeq2, edgeR, voom (Variance Modeling at the Observational Level), Cuffdiff, EBSeq, T-tests, MAST (Model-based Analysis of Single-cell Transcriptomics), BaySeq, and Limma-voom.

[0120] In various embodiments, a differential gene expression profile of the RR or IR gene signature in the sample, when compared to a reference expression profile, indicates a reduced responsiveness (RR) or improved responsiveness (IR) to treatment with the VEGF inhibitor.

[0121] In various embodiments, an increased gene expression profile of the RR gene signature compared to the reference gene expression profile may be used to predict a reduced responsiveness (RR) of a subject to treatment with the VEGF inhibitor, optionally the increased gene expression profile of the RR gene signature may be above a threshold value.

[0122] In various embodiments, an increased gene expression profile of the IR gene signature compared to the reference gene expression profile may be used to predict an improved responsiveness (IR) of a subject to treatment with the VEGF inhibitor, optionally the increased gene expression profile of the IR gene signature may be above a threshold value.

[0123] The term “increased gene expression profile”, may refer to a higher or increase of expression level of a gene or collective gene set (e.g.. gene signature) in a sample above or greater than a reference gene expression profile. In contrast, the term “decreased expression profile” may refer to a lower or decrease of expression level of a gene or collective gene set (e.g. gene signature) in a sample below or less than a reference expression profile. The increase or decrease may be relative to a reference expression profile.

[0124] Accordingly, the method disclosed herein for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, may comprise: determining gene expression levels, in a sample obtained from the subject, to obtain a gene expression profile of the sample; analysing the gene expression profile using one or more differential gene expression methods and / or one or more gene set enrichment methods based on a reduced response (RR) gene signature, and / or an improved response (IR) gene signature, wherein the RR gene signature comprises a RR gene set comprising at least two genes selected from the group consisting of: FAM110C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL11A1 , ID2, ABCA2, HMOX1 , HLA-C, FTL, GJA5, COL5A2, MAN1 A1 , CD248, LUM, FMO3, MTRNR2L1 , GUCY1A1 , and MCL1 , and wherein the IR gene signature comprises a IR gene set comprising at least two genes selected from the group consisting of: ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 ; and predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor based on the analysis of the gene expression profile,wherein a differential gene expression profile of the RR and / or IR gene signature, and / or an gene set score of the RR and / or IR gene signature, and / or an enrichment score of the RR and / or IR gene signature is predictive of the responsiveness of the subject to treatment with the VEGF inhibitor, more preferably an increased gene expression profile of the RR gene signature, or positive RR gene set score, or enriched RR gene signature indicates a reduced responsiveness (RR) of the subject to treatment with the VEGF inhibitor and the subject may be predicted to be a non-responder, or an increased gene expression profile of the IR gene signature, or positive IR gene set score, or enriched IR gene signature indicates an improved responsiveness (IR) of the subject to treatment with the VEGF inhibitor and the subject may be predicted to be a responder.

[0125] As will be appreciated, the method and steps may be carried out using one or more computational methods / models, software / programs or algorithms, as well as any subsequent interpretation of the output of the analysis steps to provide a prediction. As such, the method disclosed herein may apply the use of a computer-processing device to perform the steps of the methods disclosed herein, optionally connected to a computer network. Accordingly, there is also provided a computer-implemented method for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, wherein the computer and its components may carry out one or more steps and aspects of the method disclosed herein.

[0126] In various embodiments, an electronic memory may be used for capturing and storing the measured gene expression levels and optionally gene expression profile. A software module executed by the computer-processing device may be used to generate an gene expression profile based on the gene expression levels. The software module executed by the computer-processing device may also analyse the gene expression profile using one or more DGE or gene set enrichment methods and transmit such an analysis, and / or any interpretation and / or prediction to the subject or a medical professional treating the subject.

[0127] Considering that the methods disclosed herein are capable of predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, and thus distinguish and classify if the subject will benefit from treatment with a VEGF inhibitor, the present invention further provides a method of treating, ameliorating, delaying or preventing cancer in a subject in need of such treatment, the method comprising: predicting the responsiveness of the subject to treatment with a VEGF inhibitor by the method disclosed herein; and treating the cancer with the VEGF inhibitor if, an increased expression profile of the IR gene signature and / or an enriched IR gene signature is in the sample, and optionally a decreased gene expression profile of the RR gene signature and / or absence of an enriched IR gene signature is in the sample.

[0128] The terms “treating, ameliorating, delaying or preventing”, as used herein refers to achieving one or more of the following in the subject: (a) reducing the severity of a given condition; (b) limiting or preventing the development of a condition; (c) removing a given condition; (d) limiting or preventing the recurrence of a given condition; (e) alleviation of the condition and / or its symptoms; and (f) delay the onset of a condition. Any one or more of these effects may be achieved in a subject who has or is suspected to develop cancer. In particular, the therapeutic terms “treating, ameliorating or preventing”, may refer to reducing the likelihood of a particular condition or disease state (e.g., cancer) from occurring in a subject not presently experiencing or afflicted with the condition or disease state. The terms do not necessarily indicate complete or absolute prevention or treatment.

[0129] In various embodiments, the treating of the cancer may comprise administering a therapeutically effective amount of the VEGF inhibitor to the subject, if the subject is identified as a responder to the treatment. The term “administering” and variations of that term including “administer” and “administration”, includes contacting, applying, delivering or providing a VEGF inhibitor to the subject by any appropriate means.

[0130] The present invention further provides an anti-VEGF therapy for use in treating cancer in a subject, wherein the subject has been identified as being responsive to treatment with the anti-VEGF therapy by the method disclosed herein. Similarly, there is also provided the use of an anti-VEGF therapy in the manufacture of a medicament for treating cancer in a subject, wherein the subject has been identified as being responsive to treatment with the anti-VEGF therapy by the method disclosed herein.

[0131] The present invention further provides a method for selecting a suitable cancer treatment regimen in a subject with cancer, the method comprising: predicting the responsiveness of the subject to treatment with a VEGF inhibitor by the method disclosed herein; and selecting a cancer treatment regimen for the subject comprising administering a VEGF inhibitor to the subject if: an increased expression profile of the IR gene signature and / or an enriched IR gene signature is in the sample obtained from the subject, and optionally, a decreased gene expression profile of the RR gene signature and / or absence of an enriched IR gene signature is in the sample.

[0132] The term "cancer treatment regimen" as used herein refers to a cancer treatment regimen the selection of which may be indicated by the methods or kits of the invention comprises provision to the subject of an anti-VEGF agent. Examples of VEGF inhibitors that may be employed in such treatment regimens are set out above. Suitably the anti-VEGF agent may be provided as the only (or primary) anti-cancer agent in a treatment regimen. Details of treatment regimens using VEGF inhibitors in this manner will be known to those skilled in the art, and known regimens of this type may be used in accordance with the present invention. As referred to above, in suitable embodiments, a cancertreatment regimen may comprise, the provision of anti-cancer agents such as EGFR inhibitors (for example Erbitux) as combination partner, in addition to the provision of an anti-VEGF agent and / or a chemotherapeutic agent.

[0133] In various embodiments, the cancer treatment regimen further comprises provision of one or more chemotherapeutic agents.

[0134] The term "chemotherapeutic agent" as used herein refers to a chemotherapeutic agent suitable for use in a cancer treatment regimen selection may be one that is conventionally used for treatment of the cancer in question. For example, in the case that the cancer to be treated is brain cancer, head and neck cancer, colorectal cancer or ovarian cancer, a suitable chemotherapeutic agent for used in a cancer treatment regimen selected in accordance with the invention may be one that is conventionally used for treatment of these cancers.

[0135] All embodiments disclosed herein in relation to the methods for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor similarly apply to the method of treating cancer and method for selecting a suitable cancer treatment regimen in a subject with cancer disclosed and vice versa.

[0136] The invention further features kits for carrying out the methods of the invention. Such kits typically comprise a number of agents and reagents useful and necessary to perform the methods of the invention.

[0137] Accordingly, there is provided a kit for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, the kit comprising: reagents for determining a gene expression profile in a sample obtained from the subject; reagents for analysing the gene expression profile based on a reduced response (RR) gene signature, and / or an improved response (IR) gene signature, wherein the RR gene signature comprises a RR gene set comprising at least two genes selected from the group consisting of: FAM110C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL11A1 , ID2, ABCA2, HMOX1 , HLA-C, FTL, GJA5, COL5A2, MAN1A1 , CD248, LUM, FMO3, MTRNR2L1 , GUCY1A1 , and MCL1 , and wherein the IR gene signature comprises a IR gene set comprising at least two genes selected from the group consisting of: ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 ; and instructions for using the reagents and interpreting the analysis to predict the responsiveness of the subject to treatment with a VEGF inhibitor.

[0138] In various embodiments, the kit may comprise means for determining gene expression profile in a sample, means for analysing the gene expression profile.

[0139] In various embodiments, the instructions comprise gene signature interpretation thresholds and treatment predictions.

[0140] The kit may comprise a number of further agents that may selected from, without limitation, a polymerase, buffers, mononucleotides, and all other agents necessary for a performing one or all necessary steps of the methods disclosed herein.

[0141] All embodiments disclosed herein in relation to the methods of the invention similarly apply to the kits and uses disclosed and vice versa.EXAMPLES

[0142] The following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how the methods, uses and kits disclosed herein may be carried out and applied, and are intended to be purely exemplary and are not intended to limit the disclosure.MATERIALS AND METHODSProcessing of genomic libraries from NPC samples

[0143] Subjects were bevacizumab-treated patients enrolled on a clinical trial (NCT01309633) for locally advanced NPC

[0010] . Formalin-fixed, paraffin-embedded biopsy tissue samples were sectioned onto membrane frame slides and stained with hematoxylin. Tissue slides were microdissected on an ArcturusXT laser-capture microdissection microscope.

[0144] RNA sequencing libraries were prepared from the microdissected tissue by Smart-3SEQ

[0011] Libraries were sequenced with single-end reads on an Illumina NextSeq 500 instrument.

[0145] Single-read FASTQ files were adapter-trimmed by CutAdapt version 4.4 [3] and mapped to the T2T-CHM13v2.0 reference genome concatenated with the Epstein-Barr virus reference genome, by STAR version 2.7.10b [5], Human gene annotations from UCSC GENCODEv35 CAT / Liftoff v2 were used [4]. Reads per gene were counted by featureCounts version 2.0.6 from the Subread package [6].

[0146] Differential gene expression analysis was performed using DESeq2 version 1.32.0 [7], Gene set enrichment analysis was performed with Fgsea version 1.18.0

[0014] , with gene sets downloaded from the Molecular Signatures Database v2023.2 [8],Processing of gene expression matrix for GBM samples [1]

[0147] RNA-Seq libraries were retrieved from the NCBI Gene Expression Omnibus (accession GSE79671 ). Single-read FASTQ files were adapter-trimmed by CutAdapt version 4.4 [3] and mapped to the T2T-CHM13v2.0 reference genome with UCSC GENCODEv35 CAT / Liftoff v2 gene annotations [4] by STAR version 2.7.10b [5]. Reads per gene in the same annotation were counted by featurecounts version 2.0.6 from the Subread package [6].Processing of gene expression matrix for GBM samples [2]

[0148] Nanostring z-scores from GSE84010 were mapped to gene symbols: scores from multiple probes matching the same gene symbol were combined by mean and probes without a gene symbol were discarded.Single sample gene set enrichment analysis

[0149] All single sample gene set enrichment analysis was performed in R (version 4.1 .0) using the GSVA package version 1.40.1 , applying the “ssGSEA”

[0012] , “GSVA”[9], and “PLAGE”

[0013] methods. For the ssGSEA method, normalization of scores was disabled.EXAMPLE 1 : DISCOVERY COHORT

[0150] A discovery cohort of locally advanced NPC patients treated with bevacizumab was utilized, and enrolled on a clinical trial (NCT01309633). These were treatment-naive locally advanced NPC patients treated with three cycles of 3-weekly cisplatin and gemcitabine preceded by 1 week of bevacizumab for each cycle, followed by standard concurrent chemoradiation. The majority of patients treated with bevacizumab had either partial clinical response or an improved clinical response at the end of treatment.

[0151] Primary, pre-treatment tumors from formalin-fixed, paraffin embedded (FFPE) of NPC patients were profiled. There were 11 patients in the improved response group and 8 patients in the partial response group. Tumor epithelial and microenvironment compartments were separately microdissected using laser-capture microdissection. The tumor epithelial compartments targeted for laser-capture microdissection were epithelial-rich areas, which included admixed microenvironment cells such as immune cells and stromal cells. In contrast, the microenvironment compartments targeted for lasercapture microdissection were epithelial-poor areas, comprising largely of microenvironment cells including immune cells and stromal cells. Normal tissue were also profiled, and samples were prepared in biological duplicates. Library preparation was performed with Smart-3SEQ, a specialized RNA-Seq technique for FFPE tissue. Quality control was performed using tapestation and libraries quantified by qPCR. Libraries were pooled, sequenced, and demultiplexed. Raw fastq files were mapped to a concatenated human CHM13 and Epstein-Barr virus genome, followed by quantification of gene expression.

[0152] On principal component analysis it was observed that there was clustering of global gene expression profiles based on the cell types that were microdissected, confirming that these geneexpression profiles were indeed enriched for purity of cells (FIG. 1 A). Differential gene expression analysis between tumor epithelial and microenvironment compartments showed clear differences in gene expression patterns and relevant biological gene sets enriched in each compartment (FIG. 1 B, 1C).

[0153] Subsequently, differential gene expression analysis was performed, comparing the gene expression of tumors from patients that had only partial clinical response to treatment (Group A) and tumors that had improved clinical response (Group B). From this analysis in the tumor epithelial compartment, 58 differentially expressed genes were identified in the tumor epithelial compartment at p-adjusted < 0.05. Unsupervised hierarchical clustering showed the ability of these 58 genes to cluster tumors into their respective response groups (FIG. 2A, 2B). The majority of differentially expressed genes (44 genes) were enriched in tumors with partial response, including several mesenchymal genes related to vascular endothelial, tissue matrix and fibroblast processes (CD248, TIMP2, COL11 A1 , COL5A2, SPARC, etc.), as well as genes related to inflammation and immune response (CCL2, HLA- A, HLA-C, etc.). In tumors with partial response, gene set enrichment analysis showed enrichment of inflamed fibroblasts, endothelial and stromal cell types (e.g. “Gavish 3CA Metaprogram Fibroblasts CAF 1", “Cui Developing Heart C3 Fibroblast Like Cell”, FIG. 2C, 2D). In tumors with improved response, 14 genes were enriched, including cell cycle related biological processes.

[0154] The list of differentially expressed genes are as follows:> Enriched in the partial (reduced) response group (44 genes, RR signature): FAM1 10C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL11A1 , ID2, ABCA2, HMOX1 , HLA-C, FTL, GJA5, COL5A2, MAN1 A1 , CD248, LUM, FMO3, MTRNR2L1 , GUCY1 A1 , MCL1 ; and> Enriched in the improved response group (14 genes, IR signature): ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , SRPK1 .

[0155] The 44 genes enriched in partial response tumors were used to generate a RR gene signature predicting reduced response to bevacizumab (“the reduced response to bevacizumab signature”). When gene set enrichment scores were applied to the discovery cohort using single-sample gene set enrichment methods including ssGSEA and GSVA, the RR signature score was significantly different between tumors with improved and partial response (FIG. 2E, 2F; ssGSEA score 20670 vs 25569, p = 2.41 x 10’7; GSVA score -0.550 vs 0.238, p = 1.83 x 109). Importantly, both the ssGSEA and SGVA methods are single-sample gene set-enrichment methods that are scored based on the gene expression of a single sample only and do not require normalization across multiple samples.EXAMPLE 2: VALIDATION COHORTSFirst Validation Cohort

[0156] The ability of the RR signature was validated to identify patients with reduced response to bevacizumab treatment in other clinical cohorts.

[0157] Apart from nasopharyngeal cancer, bevacizumab has been applied to glioblastoma (GBM), a highly-aggressive form of brain cancer with poor survival. The gene expression of glioblastoma tumors from patients who had tumor tissue obtained before treatment with bevacizumab[1] was evaluated. RNA-Seq libraries were retrieved from the NCBI Gene Expression Omnibus (accession GSE79671 ), mapped to the human genome and gene expression counts obtained.

[0158] The tumors evaluated in this validation cohort included gene expression libraries from tumors obtained either from the primary GBM diagnosis, or from the recurrence before treatment with bevacizumab (n = 20). The outcome measure was overall survival in months.

[0159] The RR signature to the gene expression profiles in this cohort was applied to perform singlesample gene set enrichment with both ssGSEA and GSVA. A consistent negative correlation with overall survival in this clinical cohort was observed (FIG. 3A, 3B).

[0160] By further filtering the RR signature based on overlapping leading-edge genes from the most disenriched biological process in the GO Cell Type (“Cui Developing Heart C3 Fibroblast Like Cell”) and Gavish 3CA (“Gavish 3CA Metaprogramme Fibroblasts CAF1") gene sets, the correlation of the RR signature with overall survival was able to be enhanced FIG. 3C-F). Leading-edge genes are the core group of genes that account for the gene set’s enrichment signal. The following overlapping gene signatures were used:> RR signature biologically-informed by GO cell type gene set (RR - cell type): SPARC, GPC6, TIMP2, ISLR, COL5A2, LUM; and> RR signature biologically-informed by 3CA gene set (RR - 3CA): SPARC, SFRP4, ISLR, COL11 A1 , COL5A2, LUM.Second Validation Cohort

[0161] The gene expression of GBM tumors from patients enrolled on the AVAglio clinical trial were also evaluated, in which patients were treated with radiotherapy and temozolomide, with or without bevacizumab[2]. The gene expression matrix was obtained from the NCBI Gene Expression Omnibus (accession GSE84010).

[0162] The gene expression of these samples were profiled using the NanoString nCounter Gene Expression Assay, which comprises a limited number of 786 gene probes. Among the 44 genes of the RBB signature, seven genes were represented in the nCounter gene expression assay (SPARC, HLA- A, CCL2, MYL9, HMOX1 , HLA-C, LUM). As the nCounter platform analysis typically requires the geneexpression matrix to be normalized to reference probes and across multiple samples, we utilized the PLAGE method for single sample gene set enrichment scoring, as PLAGE uses across-sample normalization.

[0163] In the AVAglio clinical trial, GBM tumors with the proneural subtype were observed to have better response to bevacizumab treatment, however, it was unclear whether patients with other subtypes of GBM (e.g. mesenchymal, proliferative or unclassified) would benefit from bevacizumab treatment. It was observed that non-proneural tumors in this cohort that were not completely removed surgically (n = 50) were successfully classified for response to bevacizumab treatment, based on their single sample gene set enrichment scores, when the signature from the seven-gene subset was applied.

[0164] The single sample gene set enrichment score was negatively correlated with the length of survival (r = -0.292, p = 0.019, FIG. 4A), consistent with observations in the discovery cohort that these genes were enriched in patients with reduced response to bevacizumab. Critically, as a negative control, this correlation was not observed in patients who did not receive bevacizumab in the placebo arm of the clinical trial (FIG. 4B).

[0165] Finally, the y-axis intercept of the linear regression model (intercept = 0.0798, FIG. 4A), was used to divide the clinical cohort into RR-high and RR-low groups, based on their single sample gene set enrichment scores. Kaplan-Meier curve survival analysis confirmed the utility of the gene expression signature to identify patients who respond to bevacizumab in the AVAglio clinical trial cohort. Patients with tumors with higher enrichment scores showed poorer disease-free survival compared to patients with lower enrichment score tumors (p = 0.013 , FIG. 4C). Again, this survival benefit was not observed when the single sample gene set enrichment score was applied to the placebo group (FIG. 4D).

[0166] Despite having only a subset of seven genes from the full RR signature represented in this clinical trial cohort, its significant predictive ability was still observed, which was distinct for the bevacizumab arm of the trial. The survival benefit observed is clinically important as it is not currently possible to predict bevacizumab treatment response in patients with non-proneural subtypes of GBM.

[0167] In conclusion, the RR signature accurately predicts clinical response to bevacizumab treatment. Its utility extends beyond a single cancer type, as demonstrated here in NPC and GBM. It also performs consistently across various gene set scoring methods. The gene expression signature disclosed herein may have utility as a clinical-grade test for precision medicine.

[0168] The invention has been described broadly and generically herein. Each of the narrower species and subgeneric groupings falling within the generic disclosure also form part of the invention. This includes the generic description of the invention with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or not the excised material is specifically recitedherein. Other embodiments are within the following claims.

[0169] One skilled in the art would readily appreciate that the present invention is well adapted to carry out the objects and obtain the ends and advantages mentioned, as well as those inherent therein. Further, it will be readily apparent to one skilled in the art that varying substitutions and modifications may be made to the invention disclosed herein without departing from the scope and spirit of the invention. The methods, kits and uses described herein are presently representative of preferred embodiments are exemplary and are not intended as limitations on the scope of the invention. Changes therein and other uses will occur to those skilled in the art which are encompassed within the spirit of the invention are defined by the scope of the claims. The listing or discussion of a previously published document in this specification should not necessarily be taken as an acknowledgement that the document is part of the state of the art or is common general knowledge.

[0170] The invention illustratively described herein may suitably be practiced in the absence of any element or elements, limitation or limitations, not specifically disclosed herein. Thus, it should be understood that although the present invention has been specifically disclosed by exemplary embodiments and optional features, modification and variation of the inventions embodied therein herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention.

[0171] The content of all documents and patent documents cited herein is incorporated by reference in their entirety.References:[1] Urup T, et al. Transcriptional changes induced by bevacizumab combination therapy in responding and non-responding recurrent glioblastoma patients. BMC Cancer. 2017 Apr 18;17(1 ) :278.[2] Sandmann T, et al. Patients With Proneural Glioblastoma May Derive Overall Survival Benefit From the Addition of Bevacizumab to First-Line Radiotherapy and Temozolomide: Retrospective Analysis of the AVAglio Trial. J Clin Oncol. 2015 Sep 1 .[3] Marcel Martin. Cutadapt removes adapter sequences from high-throughput sequencing reads EMBnet.journal 17, 10-12 (201 1 ).[4] Rhie, A., Nurk, S., Cechova, M. et al. The complete sequence of a human Y chromosome. Nature. 621 , 344-354 (2023).[5] Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15-21 (2013).[6] Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30, 923-930 (2014).[7] Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).[8] Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl Acad. Sci. USA 102, 15545-15550 (2005).[9] Hanzelmann, S., Castelo, R. & Guinney, J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 14, 7 (2013).

[0010] Chong WQ, et al. Integration of Antiangiogenic Therapy with Cisplatin and Gemcitabine Chemotherapy in Patients with Nasopharyngeal Carcinoma. Clin Cancer Res. 26, 5320-5328 (2020).

[0011] Foley JW, et al. Gene expression profiling of single cells from archival tissue with laser-capture microdissection and Smart-3SEQ. Genome Res. 29, 1816-1825 (2019).

[0012] Barbie DA, et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1 . Nature. 462, 108-12 (2009).

[0013] Tomfohr J, Lu J, Kepler TB. Pathway level analysis of gene expression using singular value decomposition. BMC Bioinformatics. 6, 225 (2005).

[0014] Korotkevich G, et al. Fast gene set enrichment analysis. bioRxiv 060012 (2021 ).

Claims

CLAIMSWhat is claimed is:1 . A method for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, the method comprising: a) determining gene expression levels, in a sample obtained from the subject, to obtain a gene expression profile of the sample; b) analysing the gene expression profile based on a reduced response (RR) gene signature, and / or an improved response (I R) gene signature, wherein the RR gene signature comprises a RR gene set comprising at least two genes selected from the group consisting of: FAM110C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL11A1 , ID2, ABCA2, HM0X1 , HLA-C, FTL, GJA5, COL5A2, MAN1 A1 , CD248, LUM, FM03, MTRNR2L1 , GUCY1A1 , and MCL1 , and wherein the IR gene signature comprises a IR gene set comprising at least two genes selected from the group consisting of: ODC1 , HES6, MUC4, MTND3P19, LMO4, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 ; and c) predicting the responsiveness of the subject to treatment with a VEGF inhibitor based on the analysis of the gene expression profile.

2. The method of claim 1 , wherein the RR gene set comprises fibroblast cell-specific genes that define a fibroblast cell gene signature, preferably the fibroblast cell-specific genes comprise a set of genes selected from one of the following: i) SPARC, GPC6, TIMP2, ISLR, COL5A2, and LUM; ii) SPARC, SFRP4, ISLR, COL1 1 A1 , COL5A2, and LUM; or iii) SPARC, HLA-A, CCL2, MYL9, HM0X1 , HLA-C, and LUM.

3. The method of claim 1 or 2, wherein the RR gene set consists of FAM110C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL11 A1 , ID2, ABCA2, HM0X1 , HLA-C, FTL, GJA5, COL5A2, MAN 1 A1 , CD248, LUM, FM03, MTRNR2L1 , GUCY1 A1 , and MCL1 ; and the IR gene set consists of ODC1 , HES6, MUC4, MTND3P19, LM04, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 .

4. The method of any one of claims 1 -3, wherein the cancer is ovarian cancer, lung cancer, brain cancer, head and neck cancer or colorectal cancer, preferably wherein the sample is a head and neck tumour, preferably a nasopharyngeal tumour, or a brain tumour, preferably a glioblastoma tumour, morepreferably a proneural or non-proneural glioblastoma tumour, preferably a mesenchymal, or proliferative glioblastoma tumour, or is an unclassified glioblastoma tumour.

5. The method of any one of claims 1 -4, wherein the sample is a tumour sample comprising tumour cells from the tumour epithelial compartment, which includes epithelial cells, as well as microenvironment cells admixed with epithelial cells within the same compartment.

6. The method of any one of claims 1 -5, wherein the sample is a tissue sample, preferably a fixed tissue sample, more preferably a formalin fixed paraffin embedded (FFPE) tissue sample, or a fresh or frozen tissue sample, optionally the tissue sample is microdissected by laser capture micro-dissection (LCM), preferably the microdissected tissue sample comprises epithelial cells.

7. The method of any one of claims 1 -6, wherein the VEGF inhibitor is an anti-VEGF antibody, preferably bevacizumab.

8. The method of any one of claims 1 -7, wherein the subject is a human who has been diagnosed with the cancer, and has either received treatment or is treatment-naive.

9. The method of any one of claims 1 -8, wherein the step of analysing comprises differential gene expression (DGE) analysis and / or gene set enrichment analysis, and / or gene set scoring analysis of the gene expression profile.

10. The method of any one of claims 1 -9, wherein the step of analysing comprises gene set enrichment analysis by one or more gene set enrichment methods, wherein the one or more gene set enrichment methods provide a gene set enrichment score that is predictive of the responsiveness of the subject to treatment with the VEGF inhibitor, optionally the significance of the gene set enrichment score of the RR and / or IR gene signature to predict the responsiveness of the subject to treatment with the VEGF inhibitor is based on a threshold enrichment score.11 . The method of claim 10, wherein the one or more gene set enrichment methods comprise ssGSEA and GSVA.

12. The method of any one of claims 1 -11 , wherein the subject is predicted to have a reduced response (RR) to treatment with the VEGF inhibitor if a positive gene set enrichment score for the RR gene signature is obtained; or the subject is predicted to have an improved response (IR) to treatment with the VEGF inhibitor if a positive gene set enrichment score for the IR gene signature is obtained.

13. The method of claim 12, wherein the positive gene set enrichment score for the RR gene signature is correlated with a predicted length of survival or overall survival of the subject from the cancer, preferably wherein the length of survival or overall survival is provided in months, morepreferably is at least 3 months, at least 4 months, at least 5 months, at least 6 months, at least 10 months, at least 15 months, at least 20 months, at least 25 months, at least 30 months, at least 35 months, or at least 40 months, or at least 50 months, or at least 60 months.

14. The method of any one of claims 1 -13, wherein the step of analysing comprises comparing the gene expression profile of the sample to a reference gene expression profile, wherein a differential gene expression profile in respect of the RR and / or IR gene signature compared to the reference gene expression profile in respect of the RR and / or IR gene signature may be used to predict the responsiveness of the subject to treatment with the VEGF inhibitor, preferably the step of analysing comprises differential gene expression analysis using one or more Differential Gene Expression (DGE) analysis methods.

15. The method of claim 14, wherein the subject is predicted to have a reduced response (RR) to treatment with the VEGF inhibitor if an increased gene expression profile of the RR gene signature compared to the reference gene expression profile is obtained; or the subject is predicted to have a improved response (IR) to treatment with the VEGF inhibitor if an increased gene expression profile of the IR gene signature compared to the reference gene expression profile is obtained.

16. The method of any one of claims 1 -15, wherein the step of analysing comprises gene set scoring analysis by one or more gene set scoring methods, wherein the one or more gene set scoring methods provide a gene set score that is predictive of the responsiveness of the subject to treatment with the VEGF inhibitor, optionally the significance of the gene set score of the RR and / or IR gene signature is determined based on a threshold score to predict the responsiveness of the subject to treatment with the VEGF inhibitor.

17. A method of treating, ameliorating, delaying or preventing cancer in a subject in need of such treatment, the method comprising: a) predicting the responsiveness of the subject to treatment with a VEGF inhibitor by the method according to any one of claims 1 -16; and b) treating the cancer with the VEGF inhibitor if, a decreased gene expression profile of the RR gene signature and / or a disenriched RR gene signature, and / or an absence of an enriched RR gene signature is obtained from the sample; and / or an increased gene expression profile of the IR gene signature and / or an enriched IR gene signature is obtained from the sample.

18. A method for selecting a suitable cancer treatment regimen in a subject with cancer, the method comprising: a) predicting the responsiveness of the subject to treatment with a VEGF inhibitor by the method according to any one of claims 1 -16; andb) selecting a cancer treatment regimen for the subject comprising administering a VEGF inhibitor to the subject if: a decreased gene expression profile of the RR gene signature and / or a disenriched RR gene signature, and / or an absence of an enriched RR gene signature is obtained from the sample; and / or an increased gene expression profile of the IR gene signature and / or an enriched IR gene signature is obtained from the sample.

19. A kit for predicting the responsiveness of a subject with cancer to treatment with a VEGF inhibitor, the kit comprising: i) reagents for determining a gene expression profile in a sample obtained from the subject: ii) reagents for analysing the gene expression profile based on a reduced response (RR) gene signature, and / or an improved response (IR) gene signature, wherein the RR gene signature comprises a RR gene set comprising at least two genes selected from the group consisting of: FAM110C, SPARC, PAM, AEBP1 , UBE2L6, APOL3, HTRA1 , B3GNT9, DHRS12, JTB, MARVELD1 , TMEM176B, GPC6, CFH, EPDR1 , NUP93, RNF130, TIMP2, SFRP4, PLAAT4, HLA-A, PMEPA1 , CCL2, C1 S, RNF146, KDELR3, ISLR, ITM2C, MYL9, COL11A1 , ID2, ABCA2, HM0X1 , HLA-C, FTL, GJA5, COL5A2, MAN1 A1 , CD248, LUM, FM03, MTRNR2L1 , GUCY1A1 , and MCL1 , and wherein the IR gene signature comprises a IR gene set comprising at least two genes selected from the group consisting of: 0DC1 , HES6, MUC4, MTND3P19, LM04, SLC1 A3, RAPGEF5, PKN2, CALB1 , SRSF11 , NHERF1 , H1 -2, CTNNBIP1 , and SRPK1 ; and iii) instructions for using the reagents and interpreting the analysis to predict the responsiveness of the subject to treatment with a VEGF inhibitor.

Images