WO2023021330A1

WO2023021330A1 - Compositions and methods for determining a treatment course of action

Info

Publication number: WO2023021330A1
Application number: PCT/IB2022/000463
Authority: WO
Inventors: Anita Sveen; Seyed H. MOOSAVI; Ragnhild A. Lothe
Original assignee: University Of Oslo
Priority date: 2021-08-16
Filing date: 2022-08-15
Publication date: 2023-02-23

Abstract

The present invention relates to compositions, systems, and methods for determining a treatment course of action. In particular, the present invention relates to compositions, systems, and methods for utilizing gene expression profiles to determine drug sensitivity in colorectal cancer.

Description

COMPOSITIONS AND METHODS FOR DETERMINING A

TREATMENT COURSE OF ACTION

CROSS-REFERENCE TO RELATED APPLICATION

The present application claims priority to U.S. Provisional Patent Application No. 63/233,452, filed August 16, 2021, which is hereby incorporated by reference in its entirety.

FIELD OF THE INVENTION

BACKGROUND OF THE INVENTION

Gene expression profiles of colorectal cancers (CRCs) have strong clinical associations. Prognostic value has consistently been shown for signatures of immune and stromal cells infiltrating the tumor microenvironment (1, 2), as well as for different subtyping frameworks incorporating microenvironment-related and cancer cell-intrinsic signals (3, 4). The current consensus framework (the consensus molecular subtypes, CMS) defines four biologically distinct subgroups with associations to clinicopathological factors (cancer stage and tumor localization), molecular markers (microsatellite instability [MSI] and KRAS/BRAF ^600E mutations), and patient survival (3). CMS also reflect therapeutically relevant signaling pathways, such as enrichment with EGFR signaling in CMS2- epithelial/canonical tumors and angiogenic signals in the CMS4-mesenchymal/stromal group, suggesting that CMS could also be used for selection of standard targeted agents (5, 6). However, retrospective analyses of randomized clinical trials comparing combination chemotherapies plus either anti-EGFR or anti-VEGF monoclonal antibodies in the first line treatment of KRAS wild-type metastatic CRCs, showed inconsistent results with respect to the predictive value of CMS (7, 8). These studies have highlighted the unsettled question of the suitability of the CMS framework in the metastatic setting (9).

CMS was originally developed for primary tumors, and metastatic lesions have different expression signals from the tumor microenvironment, as well as a different distribution of the clinicopathological and molecular features associated with CMS (10). Furthermore, CMS4-mesenchymal/stromal tumors are associated with a poor prognosis in the primary setting (3, 6), while patients with CMSl-MSI/immune cancers have a particularly short survival after metastatic dissemination (7, 8, 11). The CRC intrinsic subtypes (CRIS) were identified as a more uniform framework across different sources of CRC samples (4), but the clinical relevance of CRIS has not been equally well addressed. It has been suggested that also metastases can be grouped according to epithelial-like and mesenchymal-like expression signals (12, 13), but only few studies have sampled metastatic lesions.

The liver is the most common site of metastasis from CRC and approximately 30% of the patients develop liver metastasis, commonly with multiple lesions. This is associated with a 5-year overall survival (OS) rate of only approximately 15% (14), although liver resection offers a potential for long-term survival in a subset of the patients (15). A few molecularly- guided systemic treatment options have shown clinical benefit, including anti-EGFR agents in KRASINRAS (RAS) wild-type cancers with a left-sided primary tumor location (16), immune checkpoint inhibitors against MSI cancers (17, is), and targeted combination therapies against BRAF ^600E mutated cancers (19). Molecular pre-screening for therapy selection in the metastatic setting is most commonly based on the primary tumor, justified by the strong concordance between primary and metastatic tumors for the currently “actionable” genetic aberrations (20-22). However, tumor heterogeneity is a major cause of treatment failure, illustrated by the clonal expansion of resistant subclones with acquired RAS mutations during anti-EGFR treatment (23). Gene expression profiles are highly dynamic, and heterogeneity of CMS between matched primary tumors and metastases may be found in as many as 40% of patients (11), further highlighting the need for molecular profiling directly of metastatic lesions.

Customized therapies for CRC are needed.

SUMMARY OF THE INVENTION

The present invention relates to compositions, systems, and methods for determining a treatment course of action. In particular, the present invention relates to compositions, systems, and methods for utilizing gene expression profiles for stratified patient treatment in colorectal cancer, based on subtype-specific drug sensitivities.

Gene expression-based subtyping has the potential to form a new paradigm for stratified treatment of colorectal cancer. However, current frameworks are based on the transcriptomic profiles of primary tumors, and metastatic heterogeneity is a challenge.

Experiments described herein resulted in the development of the de novo liver metastasis subtype (LMS) framework that recapitulated the main distinction between epithelial-like and mesenchymal-like tumors, with a strong immune and stromal component only in the latter. The framework provides identification of biologically distinct epithelial- like subtypes originating from different progenitor cell types. LMS1 metastases had several transcriptomic features of cancer aggressiveness, including secretory progenitor cell origin, oncogenic addictions and microsatellite instability in a microsatellite stable background, as well as frequent RAS/TP53 co-mutations. The poor-prognostic association of LMS 1 metastases was independent of mutation status, clinicopathological variables, and current subtyping frameworks (consensus molecular subtypes and colorectal cancer intrinsic subtypes). LMS1 was also the least heterogeneous subtype in patient- wise multi -metastatic comparisons, and tumor heterogeneity did not confound the prognostic value of LMS 1.

The experiments described herein are the first large study of patient-wise multimetastatic gene expression profiling of colorectal cancer. The new metastasis-oriented subtyping framework provides clinically relevant transcriptomic classification in the context of metastatic heterogeneity, and an LMS1 mini-classifier was constructed that finds use in prognostic stratification, drug development, and customized therapy.

Accordingly, in some embodiments, provided herein is a method for characterizing colorectal cancer (CRC) in a sample from a subject diagnosed with CRC, comprising: a) assaying a sample from said subject for the expression levels of one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or all 9) genes selected from, for example, GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, or MUC 17,' and b) characterizing the CRC as LMS1 CRC when the expression levels of the one or more genes are elevated relative to the levels of the genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5.

Further embodiments provide a method for measuring expression of cancer markers in a sample from a subject diagnosed with CRC, comprising: assaying a sample from said subject for the expression levels of two or more (e.g., 2, 3, 4, 5, 6, 7, 8, or all 9) genes selected from, for example, GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, or MUC 17.

Additional embodiments provide a method for providing a prognosis for CRC, comprising: a) assaying a sample from the subject for the expression levels of one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or all 9) genes selected from, for example, GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, or MUC 17,' b) characterizing the CRC as LMS1 CRC when the expression levels of the one or more genes are elevated relative to the levels of the genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and c) identifying the subject as having poor prognosis when said CRC is characterized as LMS1.

Still other embodiments provide a method for providing a prognosis for CRC, comprising: a) characterizing the CRC as LMS1 CRC when the expression levels of the one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or all 9) genes selected from, for example, GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, or MUC17 in a sample from a subject diagnosed with CRC are elevated relative to the levels of the genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and b) identifying the subject as having poor prognosis when the CRC is characterized as LMS1. In some embodiments, the subject is post hepatic resection.

Some embodiments provide a method for screening compounds, comprising: a) contacting a CRC sample with a test compound; and b) assaying the sample for the expression levels of one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or all 9) genes selected from, for example, GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, or MUC 17. In some embodiments, the method further comprises identifying compounds that inhibit the growth of the CRC sample.

Other embodiments provide a method for screening compounds, comprising: a) characterizing a CRC sample as LMS1-5 based on the expression level of one or more (e.g., 1, 3, 5, 7, 9, 11, 20, 50, 100, or more) genes selected from those in Table 3; b) contacting the sample with a test compound selected from those in Table 4; and c) assaying the ability of said the compound to inhibit growth of the CRC. In some embodiments, the sample is in vitro, ex vivo, or in vivo.

In certain embodiments, provided herein is a method for treating CRC, comprising: a) characterizing the CRC as LMS1 CRC when the expression levels of one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or all 9) genes selected from, for example, GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, oxMUC17 in a sample from a subject diagnosed with CRC are elevated relative to the levels of the genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and b) administering an agent that treats LMS1 CRC.

In further embodiments, provided herein is a method for treating CRC, comprising: a) characterizing said CRC as LMS1 CRC when the expression levels of the one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or all 9) genes selected from, for example, GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, oxMUC17 in a sample from a subject diagnosed with CRC are elevated relative to the levels of the genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and b) administering a CRC treatment that is not 5- fluoruracil or not 5-fluoruracil and folinic acid (FA), Afatinib, Cetuximab, 5-fluoruracil and SN-38 and FA, 5-fluoruracil and Oxaliplatin and FA, oxaliplatin, SN38, Regorafeib or TAS 102 to the subject. In some embodiments, the CRC treatment is, for example, OTS167, ONX- 0914, sepantronium bromide, encorafenib, gedatolisib, doxorubicin, bemcentinib, napabucasin, or LCL161.

In specific embodiments, provided herein is a method for characterizing CRC in a sample from a subject diagnosed with CRC, comprising: a) assaying a sample from said subject for the expression levels of one or more (e.g., 1, 3, 5, 7, 9, 11, 20, 50, 100, or more) genes selected from those listing in Table 3; and b) characterizing the CRC as LMS1-LMS5 based on the expression levels.

In additional embodiments, provided herein is a method for measuring expression of cancer markers in a sample from a subject diagnosed with CRC, comprising: assaying a sample from the subject for the expression levels of two or more (e.g., 2, 3, 5, 7, 9, 11, 20, 50, 100, or more) genes selected from those listed in Table 3.

In yet other embodiments, provided herein is a method for providing a prognosis for CRC, comprising: a) assaying a sample from the subject for the expression levels of one or more (e.g., 1, 3, 5, 7, 9, 11, 20, 50, 100, or more) genes selected from those listed in Table 3; b) characterizing the CRC as LMS1-LMS5 based on the expression levels; and c) providing a prognosis based on the characterization.

In still other embodiments, provided herein is a method for treating CRC, comprising: a) assaying a sample from a subject for the expression levels of one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or all 9) genes selected from, for example, GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, or MUC 17,' b) characterizing the CRC as LMS1 CRC when the expression levels of the one or more genes are elevated relative to the levels of the genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and c) treating the subject with an agent that alters the expression level or one or more activities of the genes.

The present disclosure is not limited to particular test compounds. Examples include but are not limited to, from those listed in Table 4.

The present disclosure is not limited to particular sample types. Examples include but are not limited to, a tissue sample, a biopsy sample, a blood sample and a stool sample.

In some embodiments, the CRC is stage I, II, III or IV. The present disclosure is not limited to the particular prognosis (e.g., poor prognosis) provided. In exemplary embodiments, the prognosis is an increased likelihood of metastasis (e.g., liver metastasis) and/or decreased 5-year survival.

The assaying step utilizes any suitable analysis methods (e.g., including but not limited to, contacting the sample with a reagent selected from a nucleic acid probe or probes that hybridizes to a respective gene product of the one or more genes, nucleic acid primers for the amplification and detection of a respective gene product of the one or more genes, or an antigen binding protein that binds to a respective gene product of the one or more genes).

Additional embodiments will be apparent to persons skilled in the relevant art based on the teachings contained herein.

DESCRIPTION OF THE DRAWINGS

FIG. 1 A-D. Comparison of gene expression profiles of CRLMs with normal liver tissue samples, primary CRCs, CRC cell lines, and PDOs. a PCA showed sample clustering based on sample type and tissue of origin, b PCI versus sample-wise liver scores calculated by GSVA of a set of genes highly expressed in the liver, c Repeated PCA plot of all samples along the PCI and PC2 axes, colored according to the microarray expression levels of ALB and KRT20. d Hierarchical clustering of multiple (2-8) distinct CRLMs from each of 45 patients.

FIG. 2A-F. Unsupervised de novo subtyping of CRLMs based on gene expression, a Quality metrics from NMF classification using input gene sets defined by three different thresholds for the cross-sample SD indicated that the optimal number of sample clusters (K) was either 2 or 5. b The sample clusters at K = 2 factorization were most strongly separated by epithelial-mesenchymal characteristics, as illustrated with a sample-wise epithelial score calculated by GSVA (p-value from t-test). c Heatmap of NMF clustering output at K = 5 factorization, d Pie chart showing the proportion of samples in each of the de novo liver metastasis subtypes (LMS1-5) at K = 5. e PCA plot of samples based on the input gene set for NMF (cross-sample SD > 0.8) and colored according to LMS group, confirms strong separation of the mesenchymal subtype (LMS5) from the four epithelial subtypes (LMS 1-4) along PCI. 1) The proportion of LMS5 samples was higher among CRLMs exposed to neoadjuvant chemotherapy, but there was no significant difference between treatment groups for LMS1, LMS2, and LMS4.

FIG. 3A-B. Molecular characteristics of the de novo LMS framework, a GSEA of selected gene expression signatures shows distinct patterns of activated (red) or down- regulated (blue) pathways, b From top: TP53/KRAS/NRAS/BRAF^N(,00E mutation frequency across patients, with and without subtype stratification (for the latter, calculated per subtype). Bottom: Frequency of RASITP53 co-mutations in each subtype.

FIG. 4A-D. Associations of LMS with clinicopathological factors and patient outcome, a Subtype-wise frequency of clinicopathological variables with a significantly different distribution across the subtypes. Kaplan-Meier plots of 5-year OS stratified b according to the individual LMS groups, c by LMS1 versus LMS2-5 combined and d in combination with translated CMS subtypes as indicated. P-values are calculated by log-rank test and in b FDR corrected by the Benjamini -Hochberg procedure.

FIG. 5A-C. Validation and intra-patient heterogeneity of LMS in additional CRLM samples, a LMS distributions in two publicly available datasets of resected CRLMs, also according to available clinical information, b GSEA results for selected signatures in each external validation series corresponded fairly well with the patterns observed in the in-house series, c Left panel: LMS classifications of multiple CRLM samples from a subset of patients in the in-house series. Right panel: the pie chart summarizes the proportion of overall intra- patient inter-metastatic subtype heterogeneity among the 42 patients with multiple metastatic lesions from the same hepatic resection.

FIG. 6. Overview of the de novo liver metastasis subtypes. The main characteristics of each subtype are summarized. Mut, mutations.

FIG. 7. Overview of study material and analyses.

FIG. 8A-B. Association between selected clinical parameters and gene expression profiles of CRLMs. 8A: Impact of systemic treatment prior to tumor sampling on the gene expression profiles, analyzed as PCI values (from PC A of the 1000 genes with highest SD across patients) and the liver scores. Neoadjuvant chemotherapy was associated with lower PCI values. 8B: Primary tumor sidedness was not associated with any of the gene expression measures. Wilcoxon test p-values are denoted.

FIG. 9. CMS subtyping of CRLMs using the tailored CMS classifier. CRLMs were classified according to CMS using our classifier adapted to the liver setting

FIG. 10. GSEA of the epithelial versus mesenchymal subtype from K=2 factorization.

FIG. 11 A-C. PCA and liver score distribution among LMS groups, a-b The same plots as Fig. la and lb, respectively, with CRLM samples colored according to LMS. c Distribution of the “liver scores” among the LMS groups indicated no influence of the proportion of hepatocyte signals on de novo subtyping. FIG. 12A-C. Selected single-sample GSVA scores across the LMS groups, a MSI-like signature score for each sample (one randomly selected per patient) across the LMS groups. Red asterisk denotes the single sample with confirmed MSI+ status, b Cytotoxic T cell and MSI-like scores plotted by MSI-status in primary CRCs illustrate the relationship between MSI and cytotoxic T cell infiltration in the primary setting, c KRAS-addiction signature scores in CRLMs with confirmed KRAS mutation plotted according to LMS.

FIG. 13. GSEA in CRLMs with RAS/TP53 co-mutations.

FIG. 14A-C. Kaplan-Meier plots of 5-year CSS according to LMS and translated CMS subtypes. Kaplan-Meier plots of 5-year CSS stratified by an individual LMS subtype, b LMS1 versus LMS2-5 grouped, and c according to both LMS1 and translated CMS subtypes.

FIG. 15. Kaplan-Meier plots of 5-year OS and CSS according to epithelial and mesenchymal subtypes.

FIG. 16A-B. Kaplan-Meier plots of 5-year OS and CSS according to LMS and TP53IRAS co-mutations. a LMS1 was associated with a poor patient outcome compared to LMS2-5 when analyzing only patients with R0/R1 resections (excluding both patients with R2 resection in the liver, and patients with extra hepatic disease, totally 42 patients), b Patient stratification according to both LMS1 versus LMS2-5 and TP53IRAS co-mutation versus no co-mutation showed that there was no significant difference for LMS1 tumors with and without co-mutations.

FIG. 17A-E. CRIS classification of the in-house CRLM samples, a Heatmap represents the zscore of gene expression for gene templates (in rows) in each sample (in columns), both grouped according to the five CRIS classes, b GSEA results from comparisons of the CRIS groups using two sets of gene signatures; top: signatures provided in the original CRIS paper; and bottom: in-house compiled gene sets, c TP53 w RAS mutation distribution in each subtype, confirming frequent TP53 wild-type status in CRIS-A and CRIS-D, as well as frequent RAS wild-type status in CRIS-C. d Kaplan-Meier plots of 5- year overall survival (OS) and cancer-specific survival (CSS) according to the CRIS groups, e Combined survival analyses of LMS and CRIS were focused on the LMS1 and CRIS-B groups, and LMS1 and CRIS-B combined was associated with the worst outcome, followed by LMSl/non-CRIS-B, non-LMSl/CRIS-B, and non-LMSl/non-CRIS-B.

FIG. 18. GSEA of CRLMs in two external data sets according to LMS. Enrichment patterns from GSEA were concordant between each of the two independent datasets and the in-house material (plot corresponding to Fig. 3a). FIG. 19A-D. Kaplan-Meier curves of OS and CSS according to LMS1 and tumor heterogeneity. Only patients with R0/R1 resections in the liver were included for analyses. There was no significant difference in the 5-year survival rates a between patients with homogenous versus heterogeneous LMS classifications in inter-metastatic comparisons; or b between patients homogenously classified with LMS1 in all samples versus patients with heterogeneous classifications including at least one LMS1 lesion/sample. LMS1 had poor prognostic associations independent of tumor heterogeneity, as shown by stratification of all patients according to c homogenous LMS 1 classification (LMS1 in all samples analyzed) versus LMS2-5 plus heterogeneous LMS1 classification; and d LMS1 in at least one lesion versus LMS2-5 in all lesions.

FIG. 20. LMS1 mini-classifier is correlated with signatures of LMS 1 characteristics. FIG. 21. LMS1 mini-classifier captures the poor-prognostic value of LMS 1.

FIG. 22A-B. A) There was a large proportion of LMS 1 samples among the cell lines, due to a high frequency of KRAS/NRAS and BRAFV600E mutations. B) Gene set enrichment analyses confirmed that LMS1 cell lines had similar gene expression characteristics to LMS1 liver metastases, including an MSI-like and serrated phenotype with strong oncogenic signaling.

FIG. 23 shows drugs with strong relative activity in LMS1 versus LMS-other.

FIG. 24A-B. A) The frequency of LMS 1 among tumor organoids was similar to liver metastases. LMS1 organoids had frequent /?/ / ^Vfi00F and KRAS/NRAS mutations. B) Gene set enrichment analyses confirmed that LMS1 organoids had similar gene expression characteristics to LMS1 liver metastases.

FIG. 25 shows drugs with strong relative activity in LMS1 versus LMS-other.

DEFINITIONS

To facilitate an understanding of the present invention, a number of terms and phrases are defined below:

As used herein, the term “sensitivity” is defined as a statistical measure of performance of an assay (e.g., method, test), calculated by dividing the number of true positives by the sum of the true positives and the false negatives.

As used herein, the term “specificity” is defined as a statistical measure of performance of an assay (e.g., method, test), calculated by dividing the number of true negatives by the sum of true negatives and false positives. The term “neoplasm” as used herein refers to any new and abnormal growth of tissue. Thus, a neoplasm can be a premalignant neoplasm or a malignant neoplasm. The term “neoplasm-specific marker” or “cancer marker” refers to any biological material that can be used to indicate the presence of a neoplasm. Examples of biological materials include, without limitation, nucleic acids, polypeptides, carbohydrates, fatty acids, cellular components (e.g., cell membranes and mitochondria), and whole cells. The term “colorectal neoplasm-specific marker” refers to any biological material that can be used to indicate the presence of a colorectal neoplasm (e.g., a premalignant colorectal neoplasm, a malignant colorectal neoplasm, a metastatic colorectal neoplasm). Examples of colorectal neoplasmspecific markers include, but are not limited to, the 13 gene signature described herein.

As used herein, the term “amplicon” refers to a nucleic acid generated using primer pairs. The amplicon is typically single-stranded DNA (e.g., the result of asymmetric amplification), however, it may be RNA or dsDNA.

As used herein, the term “metastasis” is meant to refer to the process in which cancer cells originating in one organ or part of the body relocate to another part of the body and continue to replicate. Metastasized cells subsequently form tumors which may further metastasize. Metastasis thus refers to the spread of cancer from the part of the body where it originally occurs to other parts of the body. As used herein, the term “metastasized colorectal cancer cells” is meant to refer to colorectal cancer cells which have metastasized; colorectal cancer cells localized in a part of the body other than the colorectal.

As used herein, “an individual is suspected of being susceptible to metastasized colorectal cancer” is meant to refer to an individual who is at an above-average risk of developing metastasized colorectal cancer (e.g., liver metastasis). Examples of individuals at a particular risk of developing colorectal cancer are those whose family medical history indicates above average incidence of colorectal cancer among family members and/or those who have already developed colorectal cancer and have been effectively treated who therefore face a risk of relapse and recurrence. Other factors which may contribute to an above-average risk of developing metastasized colorectal cancer which would thereby lead to the classification of an individual as being suspected of being susceptible to metastasized colorectal cancer may be based upon an individual's specific genetic, medical and/or behavioral background and characteristics.

As used herein, the term “liver metastasis subtype” or LMS (e.g., LMS1, LMS2, LMS3, LMS4, and LMS5) refer to molecular subtypes of colorectal cancer (CRC). LMS subtypes characterize the ability and/or likelihood of a CRC tumor to metastasize to the liver and/or the prognosis of the patient. In some embodiments, LMS1 tumors are characterized by aggressive metastasis to the liver. In some embodiments, the LMS is determined by the level of expression of a plurality of markers associated with the particular LMS subtype (e.g., markers described herein).

The term "amplifying" or "amplification" in the context of nucleic acids refers to the production of multiple copies of a polynucleotide, or a portion of the polynucleotide, typically starting from a small amount of the polynucleotide (e.g., a single polynucleotide molecule), where the amplification products or amplicons are generally detectable. Amplification of polynucleotides encompasses a variety of chemical and enzymatic processes. The generation of multiple DNA copies from one or a few copies of a target or template DNA molecule during a polymerase chain reaction (PCR) or a ligase chain reaction (LCR; see, e.g., U.S. Patent No. 5,494,810; herein incorporated by reference in its entirety) are forms of amplification. Additional types of amplification include, but are not limited to, allele-specific PCR (see, e.g., U.S. Patent No. 5,639,611; herein incorporated by reference in its entirety), assembly PCR (see, e.g., U.S. Patent No. 5,965,408; herein incorporated by reference in its entirety), helicase-dependent amplification (see, e.g., U.S. Patent No. 7,662,594; herein incorporated by reference in its entirety), hot-start PCR (see, e.g., U.S. Patent Nos. 5,773,258 and 5,338,671; each herein incorporated by reference in their entireties), intersequence-specfic PCR, inverse PCR (see, e.g., Triglia, et al. (1988) Nucleic Acids Res., 16:8186; herein incorporated by reference in its entirety), ligation-mediated PCR (see, e.g., Guilfoyle, R. et al., Nucleic Acids Research, 25:1854-1858 (1997); U.S. Patent No. 5,508,169; each of which are herein incorporated by reference in their entireties), methylation-specific PCR (see, e.g., Herman, et al., (1996) PNAS 93(13) 9821-9826; herein incorporated by reference in its entirety), miniprimer PCR, multiplex ligation-dependent probe amplification (see, e.g., Schouten, et al., (2002) Nucleic Acids Research 30(12): e57; herein incorporated by reference in its entirety), multiplex PCR (see, e.g., Chamberlain, et al., (1988) Nucleic Acids Research 16(23) 11141-11156; Ballabio, et al., (1990) Human Genetics 84(6) 571-573; Hayden, et al., (2008) BMC Genetics 9:80; each of which are herein incorporated by reference in their entireties), nested PCR, overlap-extension PCR (see, e.g., Higuchi, et al., (1988) Nucleic Acids Research 16(15) 7351-7367; herein incorporated by reference in its entirety), real time PCR (see, e.g., Higuchi, etl al., (1992) Biotechnology 10:413-417; Higuchi, et al., (1993) Biotechnology 11:1026-1030; each of which are herein incorporated by reference in their entireties), reverse transcription PCR (see, e.g., Bustin, S.A. (2000) J. Molecular Endocrinology 25:169-193; herein incorporated by reference in its entirety), solid phase PCR, thermal asymmetric interlaced PCR, and Touchdown PCR (see, e.g., Don, et al., Nucleic Acids Research (1991) 19(14) 4008; Roux, K. (1994) Biotechniques 16(5) 812-814; Hecker, et al., (1996) Biotechniques 20(3) 478-485; each of which are herein incorporated by reference in their entireties). Polynucleotide amplification also can be accomplished using digital PCR (see, e.g., Kalinina, et al., Nucleic Acids Research. 25; 1999- 2004, (1997); Vogelstein and Kinzler, Proc Natl Acad Sci USA. 96; 9236-41, (1999);

International Patent Publication No. W005023091 A2; US Patent Application Publication No. 20070202525; each of which are incorporated herein by reference in their entireties).

As used herein, the terms "complementary" or "complementarity" are used in reference to polynucleotides (i.e., a sequence of nucleotides) related by the base-pairing rules. For example, the sequence "5'-A-G-T-3'," is complementary to the sequence "3'-T-C-A-5'." Complementarity may be "partial," in which only some of the nucleic acids' bases are matched according to the base pairing rules. Or, there may be "complete" or "total" complementarity between the nucleic acids. The degree of complementarity between nucleic acid strands has significant effects on the efficiency and strength of hybridization between nucleic acid strands. This is of particular importance in amplification reactions, as well as detection methods that depend upon binding between nucleic acids.

As used herein, the term "primer" refers to an oligonucleotide, whether occurring naturally as in a purified restriction digest or produced synthetically, that is capable of acting as a point of initiation of synthesis when placed under conditions in which synthesis of a primer extension product that is complementary to a nucleic acid strand is induced (e.g., in the presence of nucleotides and an inducing agent such as a biocatalyst (e.g, a DNA polymerase or the like) and at a suitable temperature and pH). The primer is typically single stranded for maximum efficiency in amplification, but may alternatively be double stranded. If double stranded, the primer is generally first treated to separate its strands before being used to prepare extension products. In some embodiments, the primer is an oligodeoxyribonucleotide. The primer is sufficiently long to prime the synthesis of extension products in the presence of the inducing agent. The exact lengths of the primers will depend on many factors, including temperature, source of primer and the use of the method. In certain embodiments, the primer is a capture primer.

As used herein, the term "nucleic acid molecule" refers to any nucleic acid containing molecule, including but not limited to, DNA or RNA. The term encompasses sequences that include any of the known base analogs of DNA and RNA including, but not limited to, 4 acetylcytosine, 8-hydroxy-N6-methyladenosine, aziridinylcytosine, pseudoisocytosine, 5- (carboxyhydroxyl-methyl) uracil, 5 -fluorouracil, 5-bromouracil, 5- carboxymethylaminomethyl-2-thiouracil, 5-carboxymethyl-aminomethyluracil, dihydrouracil, inosine, N6-isopentenyladenine, 1 -methyladenine, 1 -methylpseudo-uracil, 1- methylguanine, 1 -methylinosine, 2,2-dimethyl-guanine, 2-methyladenine, 2-methylguanine, 3-methyl-cytosine, 5 -methylcytosine, N6-methyladenine, 7-methylguanine, 5- methylaminomethyluracil, 5-methoxy-amino-methyl-2-thiouracil, beta-D-mannosylqueosine, 5'-methoxy carbonylmethyluracil, 5-methoxyuracil, 2-methylthio-N- isopentenyladenine, uracil-5-oxyacetic acid methylester, uracil-5-oxyacetic acid, oxybutoxosine, pseudouracil, queosine, 2-thiocytosine, 5-methyl-2-thiouracil, 2-thiouracil, 4-thiouracil, 5 -methyluracil, N- uracil-5-oxyacetic acid methylester, uracil-5-oxyacetic acid, pseudouracil, queosine, 2- thiocytosine, and 2,6-diaminopurine.

As used herein, the term “nucleobase” is synonymous with other terms in use in the art including “nucleotide,” “deoxynucleotide,” “nucleotide residue,” “deoxynucleotide residue,” “nucleotide triphosphate (NTP),” or deoxynucleotide triphosphate (dNTP).

An “oligonucleotide" refers to a nucleic acid that includes at least two nucleic acid monomer units (e.g, nucleotides), typically more than three monomer units, and more typically greater than ten monomer units. The exact size of an oligonucleotide generally depends on various factors, including the ultimate function or use of the oligonucleotide. To further illustrate, oligonucleotides are typically less than 200 residues long (e.g., between 15 and 100), however, as used herein, the term is also intended to encompass longer polynucleotide chains. Oligonucleotides are often referred to by their length. For example a 24 residue oligonucleotide is referred to as a "24-mer". Typically, the nucleoside monomers are linked by phosphodiester bonds or analogs thereof, including phosphorothioate, phosphorodithioate, phosphoroselenoate, phosphorodiselenoate, phosphoroanilothioate, phosphoranilidate, phosphoramidate, and the like, including associated counterions, e.g., H⁺, NH₄ ⁺, Na⁺, and the like, if such counterions are present. Further, oligonucleotides are typically single-stranded. Oligonucleotides are optionally prepared by any suitable method, including, but not limited to, isolation of an existing or natural sequence, DNA replication or amplification, reverse transcription, cloning and restriction digestion of appropriate sequences, or direct chemical synthesis by a method such as the phosphotriester method of Narang et al. (1979) Meth Enzymol. 68: 90-99; the phosphodiester method of Brown et al. (1979) Meth Enzymol. 68: 109-151; the diethylphosphoramidite method of Beaucage et al. (1981) Tetrahedron Lett. 22: 1859-1862; the triester method of Matteucci et al. (1981) J Am Chem Soc. 103:3185-3191; automated synthesis methods; or the solid support method of U.S. Pat. No. 4,458,066, entitled "PROCESS FOR PREPARING POLYNUCLEOTIDES," issued Jul. 3, 1984 to Caruthers et al., or other methods known to those skilled in the art. All of these references are incorporated by reference.

A "sequence" of a biopolymer refers to the order and identity of monomer units (e.g., nucleotides, etc.) in the biopolymer. The sequence (e.g., base sequence) of a nucleic acid is typically read in the 5' to 3' direction.

As used herein, the term "subject" refers to any animal (e.g., a mammal), including, but not limited to, humans, non-human primates, rodents, and the like, which is to be the recipient of a particular treatment. Typically, the terms "subject" and "patient" are used interchangeably herein in reference to a human subject.

As used herein, the term "non-human animals" refers to all non-human animals including, but are not limited to, vertebrates such as rodents, non-human primates, ovines, bovines, ruminants, lagomorphs, porcines, caprines, equines, canines, felines, aves, etc.

The term "gene" refers to a nucleic acid (e.g, DNA) sequence that comprises coding sequences necessary for the production of a polypeptide, RNA (e.g, including but not limited to, mRNA, tRNA and rRNA) or precursor. The polypeptide, RNA, or precursor can be encoded by a full length coding sequence or by any portion of the coding sequence so long as the desired activity or functional properties (e.g, enzymatic activity, ligand binding, signal transduction, etc.) of the full-length or fragment are retained. The term also encompasses the coding region of a structural gene and the including sequences located adjacent to the coding region on both the 5' and 3' ends for a distance of about 1 kb on either end such that the gene corresponds to the length of the full-length mRNA. The sequences that are located 5' of the coding region and which are present on the mRNA are referred to as 5' untranslated sequences. The sequences that are located 3' or downstream of the coding region and that are present on the mRNA are referred to as 3' untranslated sequences. The term "gene" encompasses both cDNA and genomic forms of a gene. A genomic form or clone of a gene contains the coding region interrupted with non-coding sequences termed "introns" or "intervening regions" or "intervening sequences". Introns are segments of a gene that are transcribed into nuclear RNA (hnRNA); introns may contain regulatory elements such as enhancers. Introns are removed or "spliced out" from the nuclear or primary transcript; introns therefore are absent in the messenger RNA (mRNA) processed transcript. The mRNA functions during translation to specify the sequence or order of amino acids in a nascent polypeptide. The term “locus” as used herein refers to a nucleic acid sequence on a chromosome or on a linkage map and includes the coding sequence as well as 5’ and 3’ sequences involved in regulation of the gene.

DETAILED DESCRIPTION OF THE INVENTION

For example, in some embodiments, the present invention provides a method for characterizing CRC, determining a treatment course of action in a subject diagnosed with CRC, screening compounds for use in treating CRC, providing a prognosis to a subject with CRC, and/or treating CRC comprising: a) identifying a LMS classification for a colorectal cancer sample; and b) characterizing CRC, determining a treatment course of action in a subject diagnosed with CRC, screening compounds for use in treating CRC, providing a prognosis to a subject with CRC, and/or treating CRC comprising based on the LMS classification.

I. Identification of LMS classification

In some embodiments, the LMS classification is determined by assaying a sample for the level of expression in one or more genes listed in Table 3 (e.g., 1, 3, 5, 7, 9, 11, 15, 25, 50, 100 or more). In some embodiments, the genes are one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or all 9) genes selected from, for example, GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, or MUC17.

Any patient sample suspected of containing the genes may be tested according to methods of embodiments of the present invention. By way of non-limiting examples, the sample may be tissue (e.g., a colorectal biopsy sample or other tissue sample), blood, stool or a fraction thereof (e.g, plasma, serum, etc.).

In some embodiments, the patient sample is subjected to preliminary processing designed to isolate or enrich the sample for the pseudogenes or cells that contain the pseudogenes. A variety of techniques known to those of ordinary skill in the art may be used for this purpose, including but not limited to: centrifugation; immunocapture; cell lysis; and, nucleic acid target capture (See, e.g., EP Pat. No. 1 409 727, herein incorporated by reference in its entirety). While the present invention exemplifies several markers specific LMS classification, any marker that is correlated with the LMS classification or drug sensitivity may be used, alone or in combination with the markers described herein. A marker, as used herein, includes, for example, nucleic acid(s) whose production or mutation or lack of production is characteristic of a colorectal neoplasm or a prognosis or treatment thereof. Depending on the particular set of markers employed in a given analysis, the statistical analysis will vary. For example, where a particular combination of markers is highly specific for sensitivity of colorectal cancer to a particular treatment, the statistical significance of a positive result will be high. It may be, however, that such specificity is achieved at the cost of sensitivity (e.g., a negative result may occur even in the presence of colorectal cancer). By the same token, a different combination may be very sensitive (e.g., few false negatives), but has a lower specificity.

Particular combinations of markers may be used that show optimal function with different ethnic groups or sex, different geographic distributions, different stages of disease, different degrees of specificity or different degrees of sensitivity. Particular combinations may also be developed which are particularly sensitive to the effect of therapeutic regimens on disease progression. Subjects may be monitored after a therapy and/or course of action to determine the effectiveness of that specific therapy and/or course of action. Markers for other cancers, diseases, infections, and metabolic conditions are also contemplated for inclusion in a multiplex or panel format.

The methods are not limited to a particular type of mammal. In some embodiments, the mammal is a human. In some embodiments, the colorectal neoplasm is premalignant. In some embodiments, the colorectal neoplasm is malignant. In some embodiments, the colorectal neoplasm is colorectal cancer without regard to stage of the cancer (e.g., stage I, II, III, or IV). In some embodiments, the colorectal cancer is stage IV.

A. DNA and RNA Detection

Expression of the cancer marker genes of the present invention are detected using a variety of nucleic acid techniques known to those of ordinary skill in the art, including but not limited to: nucleic acid sequencing; nucleic acid hybridization; and nucleic acid amplification. These techniques utilize colorectal informative reagents such as nucleic acid probes and primers that hybridize to or can be used to amplify gene products of the cancer marker genes so that the level of expression of the respective cancer marker gene can be determined. 1. Sequencing

Illustrative non-limiting examples of nucleic acid sequencing techniques include, but are not limited to, chain terminator (Sanger) sequencing and dye terminator sequencing. Those of ordinary skill in the art will recognize that because RNA is less stable in the cell and more prone to nuclease attack experimentally RNA is usually reverse transcribed to DNA before sequencing.

Chain terminator sequencing uses sequence-specific termination of a DNA synthesis reaction using modified nucleotide substrates. Extension is initiated at a specific site on the template DNA by using a short radioactive, or other labeled, oligonucleotide primer complementary to the template at that region. The oligonucleotide primer is extended using a DNA polymerase, standard four deoxynucleotide bases, and a low concentration of one chain terminating nucleotide, most commonly a di-deoxynucleotide. This reaction is repeated in four separate tubes with each of the bases taking turns as the di-deoxynucleotide. Limited incorporation of the chain terminating nucleotide by the DNA polymerase results in a series of related DNA fragments that are terminated only at positions where that particular di- deoxynucleotide is used. For each reaction tube, the fragments are size-separated by electrophoresis in a slab polyacrylamide gel or a capillary tube filled with a viscous polymer. The sequence is determined by reading which lane produces a visualized mark from the labeled primer as you scan from the top of the gel to the bottom.

Dye terminator sequencing alternatively labels the terminators. Complete sequencing can be performed in a single reaction by labeling each of the di-deoxynucleotide chainterminators with a separate fluorescent dye, which fluoresces at a different wavelength.

A variety of nucleic acid sequencing methods are contemplated for use in the methods of the present disclosure including, for example, chain terminator (Sanger) sequencing, dye terminator sequencing, and high-throughput sequencing methods. Many of these sequencing methods are well known in the art, See, e.g., Sanger et al., Proc. Natl. Acad. Sci. USA 74:5463-5467 (1997); Maxam et al., Proc. Natl. Acad. Sci. USA 74:560-564 (1977); Drmanac, et al., Nat. Biotechnol. 16:54-58 (1998); Kato, Int. J. Clin. Exp. Med. 2:193-202 (2009); Ronaghi et al., Anal. Biochem. 242:84-89 (1996); Margulies et al., Nature 437:376- 380 (2005); Ruparel et al., Proc. Natl. Acad. Sci. USA 102:5932-5937 (2005), and Harris et al., Science 320:106-109 (2008); Levene et al., Science 299:682-686 (2003); Korlach et al., Proc. Natl. Acad. Sci. USA 105:1176-1181 (2008); Branton et al., Nat. Biotechnol. 26(10): 1146-53 (2008); Eid et al., Science 323:133-138 (2009); each of which is herein incorporated by reference in its entirety.

In some embodiments, deep sequencing is utilized to provide an analysis of the sequence and frequency of RNA molecules in the samples. Suitable deep sequencing techniques include, but are not limited to, next generation sequencing techniques such as single molecule real time sequencing (Pacific Biosciences), sequencing by synthesis (Illumina, Inc.), 454 pyrosequencing (Roche Diagnostics, Inc.), SOLiD sequencing (Life Technologies, Inc.), and ion semiconductor sequencing (Life Technologies, Inc.).

2. Hybridization

Illustrative non-limiting examples of nucleic acid hybridization techniques include, but are not limited to, in situ hybridization (ISH), microarray, nuclease protection assay, and Southern or Northern blot.

In situ hybridization (ISH) is a type of hybridization that uses a labeled complementary DNA or RNA strand as a probe to localize a specific DNA or RNA sequence in a portion or section of tissue in situ), or, if the tissue is small enough, the entire tissue (whole mount ISH). DNA ISH can be used to determine the structure of chromosomes. RNA ISH is used to measure and localize mRNAs and other transcripts (e.g., pseudogenes) within tissue sections or whole mounts. Sample cells and tissues are usually treated to fix the target transcripts in place and to increase access of the probe. The probe hybridizes to the target sequence at elevated temperature, and then the excess probe is washed away. The probe that was labeled with either radio-, fluorescent- or antigen-labeled bases is localized and quantitated in the tissue using either autoradiography, fluorescence microscopy or immunohistochemistry, respectively. ISH can also use two or more probes, labeled with radioactivity or the other non-radioactive labels, to simultaneously detect two or more transcripts.

In some embodiments, gene expression is detected using fluorescence in situ hybridization (FISH). In some embodiments, FISH assays utilize bacterial artificial chromosomes (BACs). These have been used extensively in the human genome sequencing project see Nature 409: 953-958 (2001)) and clones containing specific BACs are available through distributors that can be located through many sources, e.g., NCBI. Each BAC clone from the human genome has been given a reference name that unambiguously identifies it. These names can be used to find a corresponding GenBank sequence and to order copies of the clone from a distributor. The present invention further provides a method of performing a FISH assay on human colorectal cells, human colorectal tissue or on the fluid surrounding the human colorectal cells or tissue. Specific protocols are well known in the art and can be readily adapted for the present invention. Guidance regarding methodology may be obtained from many references including: In situ Hybridization: Medical Applications (eds. G. R. Coulton and J. de Belleroche), Kluwer Academic Publishers, Boston (1992); In situ Hybridization: In Neurobiology; Advances in Methodology (eds. J. H. Eberwine, K. L. Valentino, and J. D. Barchas), Oxford University Press Inc., England (1994); In situ Hybridization: A Practical Approach (ed. D. G. Wilkinson), Oxford University Press Inc., England (1992)); Kuo, et al., Am. J. Hum. Genet. 49:112-119 (1991); Klinger, et al., Am. J. Hum. Genet. 51:55-65 (1992); and Ward, et al., Am. J. Hum. Genet. 52:854-865 (1993)). There are also kits that are commercially available and that provide protocols for performing FISH assays (available from e.g., Oncor, Inc., Gaithersburg, MD). Patents providing guidance on methodology include U.S. 5,225,326; 5,545,524; 6,121,489 and 6,573,043. All of these references are hereby incorporated by reference in their entirety and may be used along with similar references in the art and with the information provided in the Examples section herein to establish procedural steps convenient for a particular laboratory.

In some embodiments, the present invention utilizes nuclease protection assays. Nuclease protection assays are useful for identification of one or more RNA molecules of known sequence even at low total concentration. The extracted RNA is first mixed with antisense RNA or DNA probes that are complementary to the sequence or sequences of interest and the complementary strands are hybridized to form double-stranded RNA (or a DNA-RNA hybrid). The mixture is then exposed to ribonucleases that specifically cleave only single-stranded RNA but have no activity against double-stranded RNA. When the reaction runs to completion, susceptible RNA regions are degraded to very short oligomers or to individual nucleotides; the surviving RNA fragments are those that were complementary to the added antisense strand and thus contained the sequence of interest. Suitable nuclease protection assays, include, but are not limited to those described in US 5,770,370; EP 2290101A3; US 20080076121; US 20110104693; each of which is incorporated herein by reference in its entirety. In some embodiments, the present invention utilizes the quantitative nuclease protection assay provided by HTG Molecular Diagnostics, Inc. (Tuscon, AZ).

3. Microarrays

Different kinds of biological assays are called microarrays including, but not limited to: DNA microarrays (e.g., cDNA microarrays and oligonucleotide microarrays); protein microarrays; tissue microarrays; transfection or cell microarrays; chemical compound microarrays; and, antibody microarrays. A DNA microarray, commonly known as gene chip, DNA chip, or biochip, is a collection of microscopic DNA spots attached to a solid surface (e.g, glass, plastic or silicon chip) forming an array for the purpose of expression profiling or monitoring expression levels for thousands of genes simultaneously. The affixed DNA segments are known as probes, thousands of which can be used in a single DNA microarray. Microarrays can be used to identify disease genes or transcripts (e.g., genes described herein) by comparing gene expression in disease and normal cells. Microarrays can be fabricated using a variety of technologies, including but not limiting: printing with fine-pointed pins onto glass slides; photolithography using pre-made masks; photolithography using dynamic micromirror devices; ink-jet printing; or, electrochemistry on microelectrode arrays.

Southern and Northern blotting is used to detect specific DNA or RNA sequences, respectively. DNA or RNA extracted from a sample is fragmented, electrophoretically separated on a matrix gel, and transferred to a membrane filter. The filter bound DNA or RNA is subject to hybridization with a labeled probe complementary to the sequence of interest. Hybridized probe bound to the filter is detected. A variant of the procedure is the reverse Northern blot, in which the substrate nucleic acid that is affixed to the membrane is a collection of isolated DNA fragments and the probe is RNA extracted from a tissue and labeled.

4. Amplification

Nucleic acids (e.g., cancer marker genes) may be amplified prior to or simultaneous with detection. Illustrative non-limiting examples of nucleic acid amplification techniques include, but are not limited to, polymerase chain reaction (PCR), reverse transcription polymerase chain reaction (RT-PCR), transcription-mediated amplification (TMA), ligase chain reaction (LCR), strand displacement amplification (SDA), and nucleic acid sequence based amplification (NASBA). Those of ordinary skill in the art will recognize that certain amplification techniques (e.g., PCR) require that RNA be reversed transcribed to DNA prior to amplification (e.g, RT-PCR), whereas other amplification techniques directly amplify RNA (e.g, TMA and NASBA).

The polymerase chain reaction (U.S. Pat. Nos. 4,683,195, 4,683,202, 4,800,159 and 4,965,188, each of which is herein incorporated by reference in its entirety), commonly referred to as PCR, uses multiple cycles of denaturation, annealing of primer pairs to opposite strands, and primer extension to exponentially increase copy numbers of a target nucleic acid sequence. In a variation called RT-PCR, reverse transcriptase (RT) is used to make a complementary DNA (cDNA) from mRNA, and the cDNA is then amplified by PCR to produce multiple copies of DNA. For other various permutations of PCR see, e.g., U.S. Pat. Nos. 4,683,195, 4,683,202 and 4,800,159; Mullis et al., Meth. Enzymol. 155: 335 (1987); and, Murakawa et al., DNA 7: 287 (1988), each of which is herein incorporated by reference in its entirety.

Transcription mediated amplification (U.S. Pat. Nos. 5,480,784 and 5,399,491, each of which is herein incorporated by reference in its entirety), commonly referred to as TMA, synthesizes multiple copies of a target nucleic acid sequence autocatalytically under conditions of substantially constant temperature, ionic strength, and pH in which multiple RNA copies of the target sequence autocatalytically generate additional copies. See, e.g., U.S. Pat. Nos. 5,399,491 and 5,824,518, each of which is herein incorporated by reference in its entirety. In a variation described in U.S. Publ. No. 20060046265 (herein incorporated by reference in its entirety), TMA optionally incorporates the use of blocking moieties, terminating moieties, and other modifying moieties to improve TMA process sensitivity and accuracy.

The ligase chain reaction (Weiss, R., Science 254: 1292 (1991), herein incorporated by reference in its entirety), commonly referred to as LCR, uses two sets of complementary DNA oligonucleotides that hybridize to adjacent regions of the target nucleic acid. The DNA oligonucleotides are covalently linked by a DNA ligase in repeated cycles of thermal denaturation, hybridization and ligation to produce a detectable double-stranded ligated oligonucleotide product.

Strand displacement amplification (Walker, G. et al., Proc. Natl. Acad. Sci. USA 89: 392-396 (1992); U.S. Pat. Nos. 5,270,184 and 5,455,166, each of which is herein incorporated by reference in its entirety), commonly referred to as SDA, uses cycles of annealing pairs of primer sequences to opposite strands of a target sequence, primer extension in the presence of a dNTPaS to produce a duplex hemiphosphorothioated primer extension product, endonuclease-mediated nicking of a hemimodified restriction endonuclease recognition site, and polymerase-mediated primer extension from the 3' end of the nick to displace an existing strand and produce a strand for the next round of primer annealing, nicking and strand displacement, resulting in geometric amplification of product. Thermophilic SDA (tSDA) uses thermophilic endonucleases and polymerases at higher temperatures in essentially the same method (EP Pat. No. 0 684 315).

Other amplification methods include, for example: nucleic acid sequence based amplification (U.S. Pat. No. 5,130,238, herein incorporated by reference in its entirety), commonly referred to as NASBA; one that uses an RNA replicase to amplify the probe molecule itself (Lizardi et al., BioTechnol. 6: 1197 (1988), herein incorporated by reference in its entirety), commonly referred to as QP replicase; a transcription based amplification method (Kwoh et al., Proc. Natl. Acad. Sci. USA 86:1173 (1989)); and, self-sustained sequence replication (Guatelli et al., Proc. Natl. Acad. Sci. USA 87: 1874 (1990), each of which is herein incorporated by reference in its entirety). For further discussion of known amplification methods see Persing, David H., “In Vitro Nucleic Acid Amplification Techniques” in Diagnostic Medical Microbiology: Principles and Applications (Persing et al., Eds.), pp. 51-87 (American Society for Microbiology, Washington, DC (1993)).

5. Detection Methods

Non-amplified or amplified nucleic acids can be detected by any conventional means. For example, the cancer marker genes described herein can be detected by hybridization with a detectably labeled probe and measurement of the resulting hybrids. Illustrative nonlimiting examples of detection methods are described below.

One illustrative detection method provides for quantitative evaluation of the amplification process in real-time. Evaluation of an amplification process in “real-time” involves determining the amount of amplicon in the reaction mixture either continuously or periodically during the amplification reaction, and using the determined values to calculate the amount of target sequence initially present in the sample. A variety of methods for determining the amount of initial target sequence present in a sample based on real-time amplification are well known in the art. These include methods disclosed in U.S. Pat. Nos. 6,303,305 and 6,541,205, each of which is herein incorporated by reference in its entirety. Another method for determining the quantity of target sequence initially present in a sample, but which is not based on a real-time amplification, is disclosed in U.S. Pat. No. 5,710,029, herein incorporated by reference in its entirety.

Amplification products may be detected in real-time through the use of various selfhybridizing probes, most of which have a stem-loop structure. Such self-hybridizing probes are labeled so that they emit differently detectable signals, depending on whether the probes are in a self-hybridized state or an altered state through hybridization to a target sequence. By way of non-limiting example, “molecular torches” are a type of self-hybridizing probe that includes distinct regions of self-complementarity (referred to as “the target binding domain” and “the target closing domain”) which are connected by a joining region (e.g., nonnucleotide linker) and which hybridize to each other under predetermined hybridization assay conditions. In a preferred embodiment, molecular torches contain single-stranded base regions in the target binding domain that are from 1 to about 20 bases in length and are accessible for hybridization to a target sequence present in an amplification reaction under strand displacement conditions. Under strand displacement conditions, hybridization of the two complementary regions, which may be fully or partially complementary, of the molecular torch is favored, except in the presence of the target sequence, which will bind to the singlestranded region present in the target binding domain and displace all or a portion of the target closing domain. The target binding domain and the target closing domain of a molecular torch include a detectable label or a pair of interacting labels (e.g., luminescent/quencher) positioned so that a different signal is produced when the molecular torch is self-hybridized than when the molecular torch is hybridized to the target sequence, thereby permitting detection of probe:target duplexes in a test sample in the presence of unhybridized molecular torches. Molecular torches and a variety of types of interacting label pairs are disclosed in U.S. Pat. No. 6,534,274, herein incorporated by reference in its entirety.

In some embodiments, a TaqMan™ detection system is utilized to detect and quantify expression of the cancer marker genes. The TaqMan probe system relies on the 5 '-3' exonuclease activity of Taq polymerase to cleave a dual-labeled probe during hybridization to the complementary target sequence and fluorophore-based detection. As in other real-time PCR methods, the resulting fluorescence signal permits quantitative measurements of the accumulation of the product during the exponential stages of the PCR; however, the TaqMan probe significantly increases the specificity of the detection. TaqMan probes consist of a fluorophore covalently attached to the 5 ’-end of the oligonucleotide probe and a quencher at the 3’-end. Several different fluorophores (e.g. 6-carboxyfluorescein, acronym: FAM, or tetrachlorofluorescein, acronym: TET) and quenchers (e.g. tetramethylrhodamine, acronym: TAMRA, or dihydrocyclopyrroloindole tripeptide minor groove binder, acronym: MGB) are available. The quencher molecule quenches the fluorescence emitted by the fluorophore when excited by the cycler’s light source via FRET (Fluorescence Resonance Energy Transfer). As long as the fluorophore and the quencher are in proximity, quenching inhibits any fluorescence signals. TaqMan probes are designed such that they anneal within a DNA region amplified by a specific set of primers. As the Taq polymerase extends the primer and synthesizes the nascent strand (again, on a single-strand template, but in the direction opposite to that shown in the diagram, i.e. from 3' to 5' of the complementary strand), the 5' to 3' exonuclease activity of the polymerase degrades the probe that has annealed to the template. Degradation of the probe releases the fluorophore from it and breaks the close proximity to the quencher, thus relieving the quenching effect and allowing fluorescence of the fluorophore. Hence, fluorescence detected in the real-time PCR thermal cycler is directly proportional to the fluorophore released and the amount of DNA template present in the PCR.

Another example of a detection probe having self-complementarity is a “molecular beacon.” Molecular beacons include nucleic acid molecules having a target complementary sequence, an affinity pair (or nucleic acid arms) holding the probe in a closed conformation in the absence of a target sequence present in an amplification reaction, and a label pair that interacts when the probe is in a closed conformation. Hybridization of the target sequence and the target complementary sequence separates the members of the affinity pair, thereby shifting the probe to an open conformation. The shift to the open conformation is detectable due to reduced interaction of the label pair, which may be, for example, a fluorophore and a quencher (e.g., DABCYL and EDANS). Molecular beacons are disclosed in U.S. Pat. Nos. 5,925,517 and 6,150,097, herein incorporated by reference in its entirety.

Other self-hybridizing probes are well known to those of ordinary skill in the art. By way of non-limiting example, probe binding pairs having interacting labels, such as those disclosed in U.S. Pat. No. 5,928,862 (herein incorporated by reference in its entirety) might be adapted for use in the present invention. Probe systems used to detect single nucleotide polymorphisms (SNPs) might also be utilized in the present invention. Additional detection systems include “molecular switches,” as disclosed in U.S. Publ. No. 20050042638, herein incorporated by reference in its entirety. Other probes, such as those comprising intercalating dyes and/or fluorochromes, are also useful for detection of amplification products in the present invention. See, e.g., U.S. Pat. No. 5,814,447 (herein incorporated by reference in its entirety).

Another illustrative detection method, the Hybridization Protection Assay (HP A) involves hybridizing a chemiluminescent oligonucleotide probe (e.g, an acridinium ester- labeled (AE) probe) to the target sequence, selectively hydrolyzing the chemiluminescent label present on unhybridized probe, and measuring the chemiluminescence produced from the remaining probe in a luminometer. See, e.g., U.S. Pat. No. 5,283,174 and Norman C. Nelson et al., Nonisotopic Probing, Blotting, and Sequencing, ch. 17 (Larry J. Kricka ed., 2d ed. 1995, each of which is herein incorporated by reference in its entirety).

B. Protein Detection The cancer marker genes described herein may be detected as proteins using a variety of protein techniques known to those of ordinary skill in the art, including but not limited to: protein sequencing; and, immunoassays.

1. Sequencing

Illustrative non-limiting examples of protein sequencing techniques include, but are not limited to, mass spectrometry and Edman degradation.

Mass spectrometry can, in principle, sequence any size protein but becomes computationally more difficult as size increases. A protein is digested by an endoprotease, and the resulting solution is passed through a high pressure liquid chromatography column. At the end of this column, the solution is sprayed out of a narrow nozzle charged to a high positive potential into the mass spectrometer. The charge on the droplets causes them to fragment until only single ions remain. The peptides are then fragmented and the masscharge ratios of the fragments measured. The mass spectrum is analyzed by computer and often compared against a database of previously sequenced proteins in order to determine the sequences of the fragments. The process is then repeated with a different digestion enzyme, and the overlaps in sequences are used to construct a sequence for the protein.

In the Edman degradation reaction, the peptide to be sequenced is adsorbed onto a solid surface (e.g, a glass fiber coated with polybrene). The Edman reagent, phenylisothiocyanate (PTC), is added to the adsorbed peptide, together with a mildly basic buffer solution of 12% trimethylamine, and reacts with the amine group of the N-terminal amino acid. The terminal amino acid derivative can then be selectively detached by the addition of anhydrous acid. The derivative isomerizes to give a substituted phenylthiohydantoin, which can be washed off and identified by chromatography, and the cycle can be repeated. The efficiency of each step is about 98%, which allows about 50 amino acids to be reliably determined.

2. Immunoassays

Illustrative non-limiting examples of immunoassays include, but are not limited to: immunoprecipitation; Western blot; ELISA; immunohistochemistry; immunocytochemistry; flow cytometry; and, immuno-PCR. Polyclonal or monoclonal antibodies detectably labeled using various techniques known to those of ordinary skill in the art (e.g, colorimetric, fluorescent, chemiluminescent or radioactive) are suitable for use in the immunoassays.

Immunoprecipitation is the technique of precipitating an antigen out of solution using an antibody specific to that antigen. The process can be used to identify protein complexes present in cell extracts by targeting a protein believed to be in the complex. The complexes are brought out of solution by insoluble antibody -binding proteins isolated initially from bacteria, such as Protein A and Protein G. The antibodies can also be coupled to sepharose beads that can easily be isolated out of solution. After washing, the precipitate can be analyzed using mass spectrometry, Western blotting, or any number of other methods for identifying constituents in the complex.

A Western blot, or immunoblot, is a method to detect protein in a given sample of tissue homogenate or extract. It uses gel electrophoresis to separate denatured proteins by mass. The proteins are then transferred out of the gel and onto a membrane, typically polyvinyldiflroride or nitrocellulose, where they are probed using antibodies specific to the protein of interest. As a result, researchers can examine the amount of protein in a given sample and compare levels between several groups.

An ELISA, short for Enzyme-Linked ImmunoSorbent Assay, is a biochemical technique to detect the presence of an antibody or an antigen in a sample. It utilizes a minimum of two antibodies, one of which is specific to the antigen and the other of which is coupled to an enzyme. The second antibody will cause a chromogenic or fluorogenic substrate to produce a signal. Variations of ELISA include sandwich ELISA, competitive ELISA, and ELISPOT. Because the ELISA can be performed to evaluate either the presence of antigen or the presence of antibody in a sample, it is a useful tool both for determining serum antibody concentrations and also for detecting the presence of antigen.

Immunohistochemistry and immunocytochemistry refer to the process of localizing proteins in a tissue section or cell, respectively, via the principle of antigens in tissue or cells binding to their respective antibodies. Visualization is enabled by tagging the antibody with color producing or fluorescent tags. Typical examples of color tags include, but are not limited to, horseradish peroxidase and alkaline phosphatase. Typical examples of fluorophore tags include, but are not limited to, fluorescein isothiocyanate (FITC) or phycoerythrin (PE).

Flow cytometry is a technique for counting, examining and sorting microscopic particles suspended in a stream of fluid. It allows simultaneous multiparametric analysis of the physical and/or chemical characteristics of single cells flowing through an optical/electronic detection apparatus. A beam of light (e.g, a laser) of a single frequency or color is directed onto a hydrodynamically focused stream of fluid. A number of detectors are aimed at the point where the stream passes through the light beam; one in line with the light beam (Forward Scatter or FSC) and several perpendicular to it (Side Scatter (SSC) and one or more fluorescent detectors). Each suspended particle passing through the beam scatters the light in some way, and fluorescent chemicals in the particle may be excited into emitting light at a lower frequency than the light source. The combination of scattered and fluorescent light is picked up by the detectors, and by analyzing fluctuations in brightness at each detector, one for each fluorescent emission peak, it is possible to deduce various facts about the physical and chemical structure of each individual particle. FSC correlates with the cell volume and SSC correlates with the density or inner complexity of the particle (e.g, shape of the nucleus, the amount and type of cytoplasmic granules or the membrane roughness).

Immuno-polymerase chain reaction (IPCR) utilizes nucleic acid amplification techniques to increase signal generation in antibody-based immunoassays. Because no protein equivalence of PCR exists, that is, proteins cannot be replicated in the same manner that nucleic acid is replicated during PCR, the only way to increase detection sensitivity is by signal amplification. The target proteins are bound to antibodies which are directly or indirectly conjugated to oligonucleotides. Unbound antibodies are washed away and the remaining bound antibodies have their oligonucleotides amplified. Protein detection occurs via detection of amplified oligonucleotides using standard nucleic acid detection methods, including real-time methods.

II. Data Analysis

In some embodiments, a computer-based analysis program is used to translate the raw data generated by the detection assay (e.g, the expression level a given marker or markers) into data of predictive value for a clinician. The clinician can access the predictive data using any suitable means. Thus, in some preferred embodiments, the present invention provides the further benefit that the clinician, who is not likely to be trained in genetics or molecular biology, need not understand the raw data. The data is presented directly to the clinician in its most useful form. The clinician is then able to immediately utilize the information in order to optimize the care of the subject.

The present invention contemplates any method capable of receiving, processing, and transmitting the information to and from laboratories conducting the assays, information provides, medical personal, and subjects. For example, in some embodiments of the present invention, a sample (e.g, a biopsy or a serum or stool sample) is obtained from a subject and submitted to a profiling service (e.g, clinical lab at a medical facility, genomic profiling business, etc.), located in any part of the world (e.g, in a country different than the country where the subject resides or where the information is ultimately used) to generate raw data. Where the sample comprises a tissue or other biological sample, the subject may visit a medical center to have the sample obtained and sent to the profiling center, or subjects may collect the sample themselves (e.g, a stool sample) and directly send it to a profiling center. Where the sample comprises previously determined biological information, the information may be directly sent to the profiling service by the subject (e.g, an information card containing the information may be scanned by a computer and the data transmitted to a computer of the profiling center using an electronic communication systems). Once received by the profiling service, the sample is processed and a profile is produced (i.e., expression data), specific for the diagnostic or prognostic information desired for the subject.

The profile data is then prepared in a format suitable for interpretation by a treating clinician. For example, rather than providing raw expression data, the prepared format may represent a diagnosis or risk assessment for the subject, along with recommendations for particular treatment options. The data may be displayed to the clinician by any suitable method. For example, in some embodiments, the profiling service generates a report that can be printed for the clinician (e.g, at the point of care) or displayed to the clinician on a computer monitor.

In some embodiments, the information is first analyzed at the point of care or at a regional facility. The raw data is then sent to a central processing facility for further analysis and/or to convert the raw data to information useful for a clinician or patient. The central processing facility provides the advantage of privacy (all data is stored in a central facility with uniform security protocols), speed, and uniformity of data analysis. The central processing facility can then control the fate of the data following treatment of the subject. For example, using an electronic communication system, the central facility can provide data to the clinician, the subject, or researchers.

In some embodiments, the subject is able to directly access the data using the electronic communication system. The subject may chose further intervention or counseling based on the results. In some embodiments, the data is used for research use. For example, the data may be used to further optimize the inclusion or elimination of markers as useful indicators of a particular condition or stage of disease or as a companion diagnostic to determine a treatment course of action.

III. Compositions & Kits

Compositions for use in the diagnostic methods described herein include, but are not limited to, kits comprising one or more colorectal cancer informative reagents as described above. In some embodiments, the kits comprise one or more reagents for detecting altered gene expression (e.g., of the genes described herein) in a sample from a subject having or suspected of having colorectal cancer. In some embodiments, the kits contain reagents specific for a cancer gene marker, in addition to detection reagents and buffers.

In preferred embodiments, the reagent is a probe(s) that specifically hybridizes to a respective gene product(s) of the one or more genes, a set(s) of primers that amplify a respective gene product(s) of the one or more genes, an antigen binding protein(s) that binds to a respective gene product(s) of the one or more genes, or a sequencing primer(s) that hybridizes to and allows sequencing of a respective gene product(s) of the one or more genes. The probe and antibody compositions of the present invention may also be provided in the form of an array. In preferred embodiments, the kits contain all of the components necessary to perform a detection assay, including all controls, directions for performing assays, and any necessary software for analysis and presentation of results.

III. Methods of Use

As disclosed herein, the present invention provides a method for characterizing CRC, determining a treatment course of action in a subject diagnosed with CRC, screening compounds for use in treating CRC, providing a prognosis to a subject with CRC, and/or treating CRC comprising based on a LMS classification.

For example, in some embodiments, the levels of expression of the genes described herein are used to determine an LMS classification. In some embodiments, the LMS classification provides a prognosis (e.g., likelihood of long term survival and/or likelihood of liver metastasis).

In some embodiments, customized therapies are developed based on LMS classification. For example, in some embodiments, the agents in Table 4 are screened across LMS classes to determine specific therapies for each class.

In some embodiments, subjects identified as LMS1 are not given 5 -fluoruracil and are instead offered alternative treatments. In some embodiments, the treatment is not 5- fluoruracil or not 5-fluoruracil and folinic acid (FA), Afatinib, Cetuximab, 5-fluoruracil and SN-38 and FA, 5-fluoruracil and Oxaliplatin and FA, oxaliplatin, SN38, Regorafeib or TAS 102 to the subject.

In some embodiments, the CRC treatment is, for example, OTS167, ONX-0914, sepantronium bromide, encorafenib, gedatolisib, doxorubicin, bemcentinib, napabucasin, or LCL161. In some embodiments, samples are assayed for LMS classification at one or more timepoints (e.g., before, during, or after treatment) and the treatment is altered or continued based on the LMS classification.

Table 4

In some embodiments, treatments described herein are administered with one or more conventional treatments for CRC or in combination with surgical or radiation therapies. EXPERIMENTAL

The following examples are provided in order to demonstrate and further illustrate certain preferred embodiments and aspects of the present invention and are not to be construed as limiting the scope thereof.

Example 1

MATERIALS AND METHODS

Patient material

A total of 829 samples from CRLMs, non-malignant liver tissue, primary CRCs and pre-clinical CRC models have been analyzed in the study. The in-house series of metastatic CRC included 171 patients treated by hepatic resection at Oslo University Hospital between October 2013 and March 2018 (Table 1). The median age at surgery was 65 years (range: 24- 85) and the median follow-up time was 41 months. The patients had a median of 4 liver metastases (range: 1-23) on radiological imaging before treatment, and fresh-frozen samples were collected from distinct metastatic lesions larger than 5 mm and from adjacent, macroscopically non-malignant tissue in the resected liver specimens. From these patients, 283 CRLM samples were analyzed. The dataset for intra-patient tumor heterogeneity analyses (totally 158 samples from 47 patients) included multiple metastatic lesions (from the same resection) from 42 patients (mean of 3 and median of 2 lesions per patient, range 2-7), 2-4 multiregional samples from each of 15 lesions, and 1-3 lesions from hepatic re-resections of 7 patients. Adjacent non-malignant liver tissue samples from 19 patients were also analyzed. All patients provided written informed consent.

Table 1. Clinicopathological characteristics of patients with resected CRLM in the Oslo series

Gender, male 106 62

Primary tumor location

Proximal colon 36 21

Distal colon 135 79

Primary tumor differentiation (unknown, n = 19)

Well 25 15 Moderate 107 63

Poor 20 12

Nodal status primary tumor (unknown, n = 8)

NO 50 29

N1 62 36

N2 51 30

Synchronous (within 6 months) liver metastasis 133 78

Previous resection/radiofrequency ablation of

39 23

CRLM

Systemic oncological treatment prior to tumor

156 91 sampling

Neoadjuvant chemotherapy for this metastatic

situation

Previous chemotherapy before this metastatic

52 30 situation

Molecularly targeted treatment, previous or

30 neoadjuvant

Radiofrequency ablation 22 13

R-status liver

RO-resection 71 42

R1 -resection³ 91 53

R2-resection^b 9 5

Extrahepatic disease 40 23

Multiple CRLM analyzed

47/141/158

(patients, tumors, samples) ^al mm resectional margin or lesions treated with radiofrequency ablation; ^b Macroscopic residual tumor in liver (visible at surgery or by radiological examination).

RNA and DNA were extracted using the Allprep DNA/RNA/miRNA Universal kit (Qiagen

GmBH, Hilden, Germany), and nucleic acid concentrations were measured with Nanodrop spectrophotometry (Thermo Fisher Scientific, Waltham, MA, USA). RNA quality was assessed by the RNA integrity number (RIN) using the Bioanalyzer 2100 system (RNA 6000

Nano kit; Agilent Technologies, Santa Clara, CA, USA), and all samples had RIN > 6 (median 9.4). Previously published in-house data from the primary tumor of 170 patients treated surgically for stage I-IV CRC at Oslo University Hospital (6), 34 CRC cell lines (24), and 15 patient-derived organoids (PDOs) grown from resected CRLMs (25) were included for comparison. Additionally, two publicly available datasets including 141 and 167 resected CRLMs were analyzed for independent, external validation (4, 12).

An overview of the study material and analyses is shown in Figure 7.

Gene expression analyses

All CRLM samples (n = 283) and adjacent non-malignant liver tissue samples from 19 patients were analyzed for gene expression at exon-resolution on the GeneChip Human Transcriptome Array 2.0 (HTA 2.0; Thermo Fisher Scientific) using 100 ng of total RNA as input, and following the manufacturer’s protocol. All the primary tumors and pre-clinical CRC models have previously been analyzed on the same array (GSE96528, GSE79959). The raw intensity CEL files were background corrected, normalized, summarized at the genelevel, and log2 transformed using the robust multi-array average (RMA) method implemented in the justRMA function in the affy package (26) in R, using the custom Entrez CDF file (v22) from Brainarray (27). Pre-processing was performed across sample sets as defined by the downstream analyses (all 521 in-house samples, CRLMs and non-malignant liver tissue, or CRLMs only). Entrez IDs were converted to HGNC gene symbols using the org.Hs.eg.db package (v 3.7.0) from Bioconductor (28).

Principal component analysis (PCA) was performed in R by the prcomp function based on genes (n = 1000) with the highest standard deviation (SD) across samples, and hierarchical clustering was similarly performed using Manhattan distance and ward.D2 linkage in the R package stats. Differential gene expression analysis was performed by Empirical Bayes estimation in the R package limma (29), with Benjamini -Hochberg correction for the false discovery rate (FDR). Gene set enrichment analysis (GSEA) with FDR correction was performed using the camera function in limma and a collection of 57 CRC-related gene sets. Sample-wise liver scores were calculated by the gsva function implemented in the R package GSVA (30), based on a set of 157 genes with expression enrichment in the liver, retrieved from The Human Protein Atlas MSI and mutation analyses

MSI status was determined by PCR-based analysis of either the BAT25/26 mononucleotide markers or with the MSI Analysis System, version 1.2 (Promega, Madison, WI, USA). The CRLMs have previously been sequenced for hotspot mutations in KRAS and NRAS exons 2-4, BRAF exon 15, and for mutations in all coding regions of TP53 (exon 2-11) (22).

CMS and CRIS classification of CRLMs

One randomly selected CRLM sample from each of 169 patients in the in-house series was classified according to both the CMS and CRIS transcriptomic frameworks. For CMS classification, we have recently developed an algorithm tailored to CRLMs, taking into consideration the different distribution of clinicopathological factors and molecular subgroups in the primary and metastatic settings, as well as the different tumor microenvironment in the liver (P.W.E., S.H.M., I.A.E., T.H.B., Langerud J., K.C.G.B., B.I.R., B.A.B., A.N., R.A.L., A.S., submitted manuscript). The tailored classifier is available in an updated version of the R package CMScaller (https://github.com/Lothelab/CMScaller). Using this classifier, 129 of the CRLMs (76%) were confidently classified. CRIS classification was performed using the cris classifier function in the CRISclassifier R package (4) with default settings. This resulted in confident classification of 139 (82%) CRLMs.

Unsupervised de novo transcriptomic classification of CRLMs

Unsupervised classification of CRLMs based on gene expression was performed using nonnegative matrix factorization (NMF) with the Brunet method, as implemented in the NMF R package (31, 32), with 100 repetitions and a pre-defined rank of 2 to 6. The classification was performed for single metastases from each patient to ensure sample independence (n = 169; the same tumor samples used for CMS and CRIS classification). Features for NMF included genes annotated as protein-coding, lincRNA and miRNA genes (n = 25969) in the BioMart database (retrieved April 2019) after a two-step filtering approach: (i) only genes that were up-regulated in the CRLMs compared with the non- malignant liver tissue samples in unpaired differential expression analysis by limma were considered (n = 6247; log2 fold change > 0, FDR corrected p-values < 0.05); (ii) only genes with largest expression variation among the CRLMs (SD > 0.8; n = 514) were retained. The rationale to include only over-expressed genes in CRLMs was to reduce the influence of normal cell contamination in bulk tumor gene expression data (70% [110/157] of the liver- enriched genes retrieved from The Human Protein Atlas were among the 313 genes down- regulated in CRLMs compared to non-malignant liver tissue). Gene expression values (log2- scale) were exponentially transformed (linear scale) prior to NMF. Classification at K = 5 was largely concordant when comparing thresholds for the input features of SD > 0.7 (n =

763 genes) and SD > 0.8 (Cohen’s K = 0.87, 95% confidence interval [CI] = 0.84-0.94).

LMS prediction model for classification of independent samples

A supervised random forest classifier for subtyping of independent samples into the five de novo LMS groups was trained on the CRLMs with a positive silhouette value in the initial LMS discovery analysis (n = 163 CRLMs, here called the training set). Template features were the same as the input for NMF. The best performing subset of genes (largest prediction accuracy) was identified by recursive feature elimination implemented with the rfe function in the R package caret (33), and with initial NMF class labels from the training set as outcomes. Function options were set to “parRF” method, 3 times 7-fold repeated cross- validation, “random” search for tuning parameter, “multiclass” summary function, and “Accuracy” metric. The weight of each gene included in the final model (n = 180, Table 3) was calculated by the varlmp function in the caret package. The trained model was then applied to all in-house CRLMs using the predict function in the R stats package. The performance of the prediction model was estimated for the training set used for LMS discovery (class labels from subtype discovery were considered “true”) using the confusionMatrix function in the caret package.

Validation of the LMS framework in external datasets

Gene expression datasets with accession numbers GSE131418 (12) and GSE73255 (4), generated on Rosetta/Merck Human RSTA Custom Affymetrix 2.0 and Illumina HumanHT-12 V3.0 bead chips platforms, respectively, were retrieved from the GEO. For GSE131418, raw CEL files from 141 resected CRLMs in the MCC dataset were processed using the justRMA function in the affy package and the provided CDF file (HuRSTA_2a520709.cdl). Entrez IDs were mapped to HGNC symbols using the org.Hs.eg.db package and expression values for non-unique symbols were median aggregated. GSE73255 included 167 unique CRLMs retrieved using the getGEO function in the R package GEOquery. Probe IDs were converted to Entrez IDs and HGNC symbols using the illuminaHumanv4.db package from Bioconductor (28) and the org.Hs.eg.db package, respectively. Genes with the highest cross-sample variance were selected for entries with non-unique symbols, and expression values were log2 transformed. Both gene expression matrices were centered by the column/sample-wise mean and scaled by the column/sample- wise SD. Classification according to LMS was performed for the two data sets separately, and according to the approach described for the independent in-house samples above, with the exception that new prediction models were trained with template features represented in each of the two external datasets (GSE131418: n = 480/514 genes, 93%; GSE73255: 462/514, 90%). In brief, a supervised random forest classifier was trained in the in-house training set (samples with known LMS labels), using recursive feature elimination to select the subset of the respective template gene sets with largest prediction accuracy (estimated by cross- validation in the training set; GSE131418: n = 230 genes in the final model; GSE73255: n = 390). The trained models were applied to the corresponding external dataset using the predict function in R.

LMS1 mini-classifier

A two-class random forest model (LMS1 versus rest of the subtypes) was trained using the train function in the caret package on differentially expressed genes identified from limma analysis comparing LMS1 to all other subtypes in the in-house training series (FDR corrected p < 0.05, log fold-change > 1.6, n = 9 genes). The prediction model was trained using 7-fold leave-one-out cross-validation. The optimal value of the mtry parameter was identified using the tuneLength option in the train function. Class labels were predicted using the predict function and were compared with original class labels in the complete in-house dataset.

For independent validation of this 9-gene LMS1 mini-classifier, CRLM samples from the in-house training series, GSE131418, and GSE73255 were merged based on common genes and batch corrected using the ComBat function in the R package sva. Missing values for UCA1 in GSE73255 were imputed by its median expression across the batch-corrected dataset. The random forest model for the 9-gene signature was re-trained on the batch- corrected in-house training series and applied to the full dataset using the predict function. The prediction accuracy of the trained model (LMS1 versus LMS2-5 distinction) was 100% for re-classification of the in-house training series.

Statistical analysis

All statistical tests were two-sided and performed in R (v3.5). Fisher’s exact, Pearson’s chi-squared, t-test, and Wilcoxon tests were performed using fisher.test, chisq.test, t.test, and wilcox.test functions in R package stats, respectively. Spearman’s correlation was calculated using stat cor function in R package ggpubr. Cohen’s Kappa was calculated using the confusionMatrix function in R package caret. 5-year OS and cancer-specific survival (CSS) curves were estimated with the Kaplan-Meier method using the survfit function in the R package survival. Pairwise log-rank tests were performed to compare survival curves using the pairwise survdiff function in the R package survminer, with the method for p-value adjustment set to the Benjamini -Hochberg procedure. The time to event or censoring was calculated from initiation of treatment for the CRLMs, either neoadjuvant treatment or hepatic resection. All deaths were registered as events for OS, and death from CRC was defined as an event for CSS, with censoring of patients who died from other causes. Patients without events the first 5 years of follow-up were censored. Hazard ratios were calculated in univariable and multivariable Cox proportional hazards analyses using the coxph function in R package survival and p-values were calculated using Wald test. The proportional hazards assumption was checked using the cox.zph function and was supported for all Cox models.

Results

Variable impact of the tumor microenvironment on gene expression profiles of CRC liver metastases

To assess the primary -to-metastasis transcriptomic landscape in CRC, gene expression profiles of 283 resected liver metastasis samples and 19 non-malignant liver tissue samples from 171 patients (Table 1) were initially compared with primary CRCs (n = 170) and pre-clinical CRC models derived from primary tumors (n = 34 cell lines) or resected CRLMs (n = 15 PDOs). In PCA, the CRLMs ranged from the primary CRCs to the non- malignant liver samples along the first principal component (PCI), although the closer vicinity to the primary CRCs indicated resemblance of the metastasized cancer cells to the tumors of origin (Fig. la). The CRLMs had a larger spread along PCI than the primary CRCs (10-90th percentile range of PCI values of 29 and 6.3, respectively), indicating a highly variable degree of influence from the liver tumor microenvironment in CRLMs. This was confirmed by calculation of a sample-wise "liver score" based on genes with high expression in the liver (see Methods), which correlated strongly with PCI of the CRLMs (Fig. lb). The liver scores of the CRLMs spanned from the non-malignant liver samples (range 0.40 to 0.89) to the primary CRCs (range -0.3 to -0.58) and cell lines (range -0.27 to -0.54). The large variation in the degree of tumor microenvironment infiltration in the CRLMs was further illustrated by the gene expression levels of the hepatocyte differentiation marker ALB, which was highest in the non-malignant liver samples and decreased gradually in the CRLMs along PCI (Fig. 1c). The opposite expression pattern was found for the intestinal differentiation marker KRT20. Notably, 27% of the CRLMs (75/283) had liver scores within the range of the primary CRCs (liver score < -0.3; Fig. lb), indicating negligible influence from the liver tumor microenvironment in these samples. Three CRLM samples from three patients clustered close to the non-malignant liver samples in PCA (Fig. la) and were excluded from further analyses.

The majority of patients had received chemotherapy prior to sampling of the CRLMs (Table 1). PCI values were slightly lower for CRLMs treated in a neoadjuvant setting compared to chemo-naive and/or previously treated tumors (one randomly selected sample per patient; Fig. 8), indicating an impact of neoadjuvant chemotherapy on the gene expression profiles. Among other clinicopathological characteristics, only R2 resections in the liver and extra-hepatic disease were associated with PCI values and the liver scores.

Exploratory analyses indicated pronounced intra-patient transcriptomic heterogeneity among metastatic lesions, illustrated by hierarchical clustering of 2-8 CRLMs from each of 45 patients (total n = 139 lesions; Fig. Id). Only 13 patients (29%) had multiple CRLMs that clustered together, while 62% of the patients (28/45) had CRLMs that separated on at least two of the five main branches. The remaining 9% of the patients (4/45) had metastases that clustered on the same main branch, although not adjacent to each other. Patient- wise clustering versus separation of samples was not associated with exposure to neoadjuvant chemotherapy (Fisher’s exact p = 0.3). However, a comparison of CRLM liver scores showed that hepatocyte infiltration was higher in samples from patients with separation of metastases into different clusters, compared to patients with adjacent sample clustering, indicating an association with inter-metastatic heterogeneity (although not statistically significant; Wilcoxon p = 0.07).

De novo transcriptomic subtypes of CRLM

By adapting CMS classification to liver metastases and developing a new version of the R package CMScaller (34) (v2.0.1), we have shown that CMS has limited discriminatory power in CRLM. Most metastatic lesions were classified into one of only two subtypes, based on epithelial-mesenchymal characteristics (Fig. 9). In addition, CMS classification was strongly influenced by systemic treatment prior to sampling, showing strong enrichment with CMS4-mesechymal/stromal tumors and concomitant depletion of CMS2-epithelial/canonical among tumors exposed to neoadjuvant chemotherapy (Fig. 9). We therefore investigated a new intrinsic classification framework for CRLM that captures additional biological information. Unsupervised classification of single CRLMs from each patient (n = 169 samples, patient-wise random selection) was performed by NMF of a filtered set of 514 genes, selected both for having upregulated expression in CRLMs compared to non- malignant liver tissue samples, and for high expression variation among the CRLMs (see Methods). Quality metrics from NMF classification, including the cophenetic correlation coefficient and silhouette width, were highest at K = 2 and K = 5 across different input gene sets defined by the expression variation threshold (Fig. 2a). GSEA of a custom collection of CRC-related gene sets (n = 57; Fig. 10) indicated that sample classification at K = 2 resulted in subtypes with predominantly epithelial (cluster 1: 76% of tumors) or mesenchymal (cluster 2: 24% of tumors) characteristics (Fig. 2b). Classification at K = 5 resulted in four additional sub-classes within the epithelial subtype, with a 98% concordance between epithelial and mesenchymal-like subtypes at the two factorization levels (Cohen’s K = 0.98, 95% CI = 0.95- l; Fig. 2c).

The five de novo sample clusters, hereafter called liver metastasis subtypes (LMS), each represented 18% (LMS1), 10% (LMS2), 19% (LMS3), 30% (LMS4), and 24% (LMS5) of the tumors (Fig. 2d, Fig. lla-b). PC A confirmed that epithelial (LMS 1-4) versus mesenchymal (LMS5) characteristics represented the primary distinction of samples along PCI (Fig. 2e). There was little difference in the distribution of liver scores among the subtypes, indicating that the LMS framework was not confounded by hepatocyte infiltration (Fig. 11c). However, LMS5-mesenchymal was significantly enriched among CRLMs exposed to neoadjuvant chemotherapy (Fig. 21). Among the four epithelial subtypes, only LMS3 showed signs of depletion in the chemotherapy-exposed group.

Enrichments with specific cell types and RASITP53 co-mutations in the LMS framework

Distinct patterns of biological processes among the LMS groups were found by GSEA (Fig. 3a). LMS5-mesenchymal CRLMs were enriched with tumor microenvironment signals, including a strong stromal component and a high relative expression of immune-related gene signatures. LMS1 had strong gene expression-based MSI characteristics, and included the single MSI-high CRLM (CRLMs from all other patients [168/169] were confirmed MSS). The MSI-high sample had the third highest MSI-like score, and most MSS tumors in LMS1 had stronger MSI-like characteristics than MSS tumors in LMS2-5 (Fig. 12a). Notably, the MSI-like score had only a weak correlation with cytotoxic T cell signals among the CRLMs (Spearman’s p = 0.2; Fig. 12b), consistent with the predominantly weak immune response signals in LMS1. LMS1 was further characterized by several oncogenic signatures in the MAPK and MET signaling pathways (including KRAS and BRAF signatures), as well as cancer aggressiveness (cell migration, hypoxia) and a signature of resistance to the standard chemotherapeutic agent 5-fluorouracil. LMS2-4 all had a transit amplifying-like phenotype. LMS2-3 showed enrichments with few other signaling pathways, while LMS4 presented with strong metabolic signals (partly shared with LMS1 and LMS2), TP53 transcriptional activity, and cell cycle-associated signatures (cell cycle checkpoints and DNA repair mechanism; Fig. 3a).

Cell type-specific gene markers extracted from published single-cell RNA sequencing studies indicated clear differences in the most dominating cell type of origin of each subtype. LMS1 CRLMs were highly enriched with genes related to secretory progenitor cells, mucussecreting goblet cells (for example, MUC2 and MUC4), and liver cholangiocytes (for example, KRT7, KRT19, EPCAM, SOX9). LMS2 strongly expressed core gene markers of LGR5+ intestinal stem cells (LGR5, OLFM4, ASCL2, SMOC1 anAMSIl). No inference of the cell type of origin could be made for LMS3 CRLMs. LMS4 showed marked expression of absorptive enterocyte markers, and LMS5-mesenchymal tumors showed strong expression of markers of quiescent stem cells (DLCK1+, PROCR+).

The five LMS groups were further analyzed for enrichment with key genomic markers of CRC beyond MSI status, including mutations of TP53, KRAS, NRAS, and /?/ / ^Vfi00F- (Fig. 3b). TP53 mutations were common across all subtypes, but with a significantly lower mutation frequency in LMS5-mesenchymal tumors (LMS5 versus LMS1- 4: Fisher’s exact p = 5*10'⁶, OR = 0.2;). KRAS mutations were enriched in LMS1- secretory/MSI-like tumors (LMS1 versus LMS2-5: p = 0.002, OR = 3.9), although the mutation frequency was not significantly higher in LMS1 than LMS3 separately (p = 0.4). Notably, there was enrichment with the gene expression-based KRAS addiction signature in LMS1 also when analyzing KRAS mutated CRLMs only, further supporting preferential KRAS signaling in LMS1 (Fig. 12c). NRAS and BRAF^N600E had low mutation frequencies in all subtypes, and there were no significant enrichments. RAS (KRAS NRAS) and TP 53 were co-mutated in 31% (52/169) of the patients, and the co-mutations were enriched in LMS1- secretory/MSI-like CRLMs (LMS1 versus LMS2-5: p = 0.005, OR = 3.2; no significant difference between LMS1 and LMS3 separately: p = 0.4; Fig. 3b). GSEA of RAS/TP53 comutated tumors only showed similar results to the analyses across all tumors, supporting enrichment with several oncogenic signatures in the MAPK and MET signaling pathways in LMS1 (Fig. 13). TP53 transcriptional activity was enriched in LMS1 and LMS4, and LMS4 (together with LMS2) showed significant enrichment with TP 53 mutations in RAS wildtype background (LMS2/4 versus LMS1/3/5: p = 8*10'⁴, OR = 3.8, 95% CI = 1.6-9.7).

Poor prognosis associated with LMSl-secretory/MSI-like CRLMs

Several clinicopathological variables were differently distributed across the LMS groups. LMSl-secretory/MSI-like and LMS5-mesenchymal were enriched with CRLMs originating from poorly differentiated and proximal (right-sided) primary tumors compared to LMS2-4 (tumor differentiation: OR = 8.4, 95% CI = 2.5-36.4, p = 9*10'⁵; tumor location: OR = 2.6, 95% CI = 1.1-6.0, p = 0.02; Fisher’s exact test; Fig. 4a). Synchronous liver metastases were most frequently found in the LMS5-mesenchymal group (OR = 4.6, 95% CI = 1.3-24.6, p = 0.009). Furthermore, analyses of the 160 patients with R0/R1 resections in the liver showed prognostic associations to 5-year OS and CSS. Patients in the LMSl-secretory/MSI- like group had a 5-year OS rate of 15%, which was lower than for patients with LMS2-5 CRLMs, analyzed both individually (significantly different for each of LMS3-5; Fig. 4b) and collectively (hazard ratio [HR] = 2.2, 95% CI = 1.4-3.6, Wald test p = 9xl0’⁴; Fig. 4c). A similar association was found with 5-year CSS as the endpoint (LMS1 versus LMS2-5: HR = 1.9, 95% CI = 1.2-3.3, Wald test p = 0.01; Fig. 14). Notably, patient stratification based on epithelial or mesenchymal characteristics (from NMF classification at K = 2) had no prognostic associations (Fig. 14). Multivariable Cox proportional hazards analyses including the clinicopathological parameters with univariable prognostic associations (patient gender, primary tumor differentiation grade, systemic oncological treatment prior to tumor sampling, R2 resection in the liver, and extra-hepatic disease) showed that the LMS framework (LMS1 versus LMS2-5) was an independent prognostic factor for both 5-year OS and CSS (adjusted HR = 2.4, 95% CI = 1.4-4.0, Wald test p = lx IO’⁴, and adjusted HR = 2.1, 95% CI = 1.2-3.7, Wald test p = 0.008, respectively). Furthermore, exclusion of patients with extra-hepatic disease and/or R2 resections in the liver (the clinicopathological factors with the strongest prognostic association) did not preclude the prognostic value of the LMSl-secretory/MSI-like group (Fig. 16a).

RASITP53 co-mutations were also associated with a worse 5-year OS (HR = 1.6, 95% CI = 1.1-2.5, Wald test p = 0.02 among patients with R0/R1 resection in the liver), and to investigate whether enrichment with co-mutations was the underlying factor for the prognostic value of LMSl-secretory/MSI-like CRLMs, we compared patients with LMS1 and LMS2-5 tumors with/without co-mutations. This indicated that LMS1 was associated

De noov with a poor patient survival independent of co-mutation status in a bivariable analysis (adjusted HR = 2.0, 95% CI = 1.2-3.3, Wald test p = 0.004), with worst prognosis for comutated LMS1, but no significant difference between co-mutated LMS2-5 and LMS1 without co-mutations (Fig. 16b).

Comparison of LMS with established trans criptomic frameworks

A direct comparison of LMS with the CMS (adapted to the liver metastatic setting) and CRIS frameworks showed only moderate subtype concordances, although LMS did not represent a statistically independent subtype distribution (Table 2). Notably, only 66% of CMS4-mesenchymal CRLMs (38/58) were included in LMS5-mesenchymal. Furthermore, 91% (10/11) of CMSl-MSI/immune CRLMs were found in LMS1, but CMS1 constituted only 45% (10/22) of the total number of tumors in this de novo subtype. Combined survival analyses of the LMS and CMS frameworks in patients with R0/R1 resections in the liver, focusing on the poor-prognostic subtypes LMS1 and CMS1, indicated that LMS provided the strongest prognostic stratification. The worst prognosis was found for patients classified as LMSl/non-CMSl (5-year OS rate of 11%), followed by LMS1/CMS1 (22%) and non- LMSl/non-CMSl (45%), respectively (log-rank p < 0.004 for both 5-year OS and CSS; Fig. 4d).

Table 2. Correspondence of the de novo subtypes with CMS and CRIS in resected CRLMs

With respect to the LMS and CRIS (Fig. 17) frameworks, the best subtype concordance was found between CRIS-C and LMS4 (69% [31/45] of samples in LMS4 were also CRIS-C), while 86% of samples in LMS1 were either CRIS-A or CRIS-B (Table 2). Survival analysis focused on LMS1 and CRIS-B showed a survival rank with worst outcome for LMS1/CRIS- B > LMSl/non-CRIS-B > non-LMSl/CRIS-B > non-LMSl/non-CRIS-B (log-rank p < 0.006 for both OS and CSS; Fig. 17e).

LMSl-secretory/MSI-like defines a distinct subtype of CRLMs across independent datasets

For validation of LMS in independent samples, a random forest LMS prediction model was developed (see Methods) and initially applied to two external gene expression datasets of resected CRLMs. These included 141 samples from Gene Expression Omnibus (GEO) accession number GSE131418 and 167 samples from GSE73255 (4, 12), analyzed on two separate microarray platforms. In comparison with the in-house dataset, there was a skewed distribution of LMS2-4 in both external datasets (Fig. 5a). The LMS4 group encompassed a relatively large proportion of samples, at the apparent cost of samples classified as LMS2 (missing from both datasets) or LMS3. The remaining subtype distributions were largely proportional to the in-house material, and GSEA indicated that several LMS characteristics were recapitulated in both independent datasets (Fig. 5b, Fig. 18). LMS1 was found to have an epithelial and secretory phenotype with strong MSI-like and BRAF-like expression signals. LMS3 and LMS4 both had a transit amplifying phenotype, and LMS4 tumors additionally had strong signaling of MYC targets. LMS5 was identified as the only mesenchymal-like subtype and presented with a strong stromal and immune component. Investigation of the available clinicopathological information (in the GSE131418 dataset) supported that the subtype distribution was not associated with exposure to neoadjuvant treatment (Fisher’s exact p = 0.3), and that LMS1 CRLMs were more likely to originate from proximal primary tumors (OR = 2.9, 95% CI = 0.9-9.2, p = 0.04; Fig. 5a).

Frequent intra-patient inter-metastatic subtype heterogeneity does not confound the prognostic value of LMS 1

The random forest LMS prediction model was also applied to multiple additional CRLM samples from each of 47 patients in the in-house series (total n = 158 samples) to analyze intra-patient tumor heterogeneity. The prediction model had an overall balanced classification accuracy of 98% among the samples also included for initial subtype discovery (95% CI = 94-99). Intra-patient inter-metastatic subtype heterogeneity was observed in 21 (50%) of the 42 patients with multiple distinct lesions from the same hepatic resection, and intra-tumor heterogeneity was observed in 5 (33%) of the 15 lesions with multiregional samples (Fig. 5c). LMS1 was the most homogeneous subtype, with inter-metastatic heterogeneity in 43% of the patients (6 of 14) with at least one LMS1 CRLM/sample, while LMS2 and LMS3 were most heterogeneous (in 100% and 93% of the patients, respectively). Intra-patient inter-metastatic subtype heterogeneity was not associated with patient survival (log-rank p > 0.2 for 5-year OS and CSS; Fig. 19a). However, considering the high frequency of subtype heterogeneity, we investigated its possible influence on the prognostic associations of LMS1. There was no statistical survival difference between patients with homogeneous LMS1 CRLMs and patients with inter-metastatic LMS1 heterogeneity (Fig. 19b). We further analyzed the impact of LMS1 heterogeneity in the complete patient series (n = 160 patients with R0/R1 resections of the liver) by switching the inclusion of patients with inter-metastatic LMS1 heterogeneity between the LMS1 group and the LMS2-5 group. This indicated no impact of tumor sampling or tumor subtype heterogeneity on the prognostic value of LMS1 (Fig. 19c-d).

Development of LMS1 mini-classifier

We explored a method to identify the clinically relevant subgroup of LMS1 CRLMs using a simpler test based on a small number of genes. A two-class mini-classifier containing genes with high relative expression in the LMS1 group (n = 9 genes; GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUC 17) was constructed in the in-house training series (single samples from each of the 169 patients). When applied to the complete in-house set of 280 CRLM samples, the classifier provided largely concordant classifications (LMS1 versus LMS2-5; Cohen’s k = 0.86). Single-sample gene set scores of the LMS1 miniclassifier genes, calculated by GSVA (30), correlated strongly with corresponding scores for the gene signatures enriched in LMS1 tumors, including the secretory progenitor signature, the MSI-like signature and the KRAS addiction score (Spearman’s p > 0.6 across samples, p < 2* IO ⁶; Fig. 20). Furthermore, the mini-classifier accurately captured the prognostic association of the LMS1 group (5-year OS: HR = 2.2, 95% CI = 1.4-3.6, Wald test p = 0.001; Fig. 21), indicating that this 9-gene mini-classifier can be used for independent prognostic stratification of resected CRLMs. Validation analyses in the two external datasets (GSE131418 and GSE73255 combined) showed a prediction accuracy of 0.89 (95% CI 0.85- 0.92) for the LMS1 versus LMS2-5 distinction.

LMS is a metastasis-oriented gene expression-based subtyping framework of CRC that identifies clinically relevant biological traits also in the context of inter-metastatic heterogeneity. Clinical relevance was illustrated by an independent poor-prognostic association of one of the five subtypes, for which a mini-classifier was developed to facilitate prognostic stratification and further clinical testing.

Table 3:Predictive RF model specification

The comparison between the class labels generated by trained model with initial NMF results ("true classes)

LMS1 LMS2 LMS3 LMS4 LMS5

Sensitivity 0.97 1.00 0.91 1.00 1.00

Specificity 1.00 1.00 1.00 0.97 0.99

Pos Pred Value 1.00 1.00 1.00 0.94 0.98

Neg Pred Value 0.99 1.00 0.98 1.00 1.00

Balanced Accuracy 0.98 1.00 0.95 0.99 1.00

Overall Statistics

Accuracy, 95% CI : 0.9763(0.9405, 0.9935)

Genes (n=180) selected in final RF model

HGNC SYMBOL Variable Importance

ANXA1 4.028922

MYO5A 3.820845 CLIP4 3.531559 DOCK2 3.474523 OLFM4 3.377483

SATB2 3.111687

GCNT3 2.647251 PIK3CG 2.617658 PLXNC1 2.547657 NCKAP1L 2.400621

KCTD12 2.257365

MS4A4A 2.108349

DOCK8 2.086296

CD84 2.007672

TM6SF1 1.970468

CCDC88A 1.966743

DOCK10 1.895612

GRM8 1.881159

IRAK3 1.842757

LY6G6D 1.839537

RUBCNL 1.826001

CYP2S1 1.705941

CD53 1.694796

SLC1A3 1.693602

TFEC 1.686233 CLEC7A 1.682343 FAM129A 1.659564 LIPH 1.658395 HAVCR2 1.644755 LRRC19 1.622117 PLCB4 1.610988 LYZ 1.603399 C10orf99 1.592357 CLDN3 1.588768 PTPN13 1.580874 PTPRC 1.575673 MMP2 1.546425 S0AT1 1.486325 KLHL6 1.475388 SLC44A4 1.465186 FYB1 1.444504 PTPN22 1.441309 CTSK 1.433805 ERBB3 1.430605 MIR3064 1.39103 CXCR4 1.384604 RAB25 1.363964 LY86 1.362637 RNF43 1.348234 PRSS8 1.344449 KRT19 1.342967 CDCA7 1.30295 CKMT1A 1.299945 SLC26A3 1.298774

GPR160 1.294991 EPCAM 1.266816 ANXA3 1.26413 SERPINB5 1.260218 SLC26A2 1.250455 FXYD3 1.250274 ITGAX 1.242068 TMEM236 1.240576 MAL2 1.229092 NEXN 1.215866 LINC01559 1.209408 CEACAM5 1.208992 HLA-DMB 1.202914 MIR3606 1.199324 PPP1R1B 1.196714 GNG4 1.192038 MSR1 1.187966 C3AR1 1.187466 CLDN7 1.186266 RPS6KA6 1.184784 LGALS4 1.179312 HACD4 1.175523 MIR1245A 1.169185 CTSE 1.167827 TRIM31 1.163603 GPR34 1.162415 SLC7A11 1.156094 SLC6A20 1.150969 TMPRSS4 1.148368 PLXDC2 1.14761 GUCY2C 1.138988 PRR15L 1.13786 NCR3LG1 1.137512 CYTIP 1.136516 MYO1A 1.132288 RAB31 1.127623 ACSL5 1.124359 ABHD17C 1.120578 AHNAK2 1.119498 CD 109 1.119227 BCAT1 1.11486 AXIN2 1.112681 S100A14 1.11237 CD86 1.111294 GPR171 1.108812 TMEM45B 1.106099 DACH1 1.094417 GPNMB 1.088468 PERP 1.086342 HHLA2 1.083737 BAIAP2L1 1.081115 NXPE4 1.077667 PLA2G4A 1.066132 CDHR1 1.063087 PARM1 1.06119 KCNH8 1.050701 LINC02418 1.043968 MUC13 1.04131 GPA33 1.041307 ARHGEF38 1.037298 PRELID 3B 1.036084 MNDA 1.028474 NFE2L3 1.027185 LINC01006 1.022638 AKR1B1 1.015158 LINC01748 1.005264

TRPM6 1.005217

MIR5047 1.003117

CLDN4 1.00228

TSP AN 1 0.995382

CEACAM1 0.99335

KLK6 0.974171

RNASE6 0.972042

RGS1 0.967222

UCA1 0.964426

DUOX2 0.964162

MIR4774 0.960555

MIR421 0.955301

CAB39L 0.952603

VIL1 0.93116

AEBP1 0.927887

EVI2B 0.920389

ANKRD40CL 0.910785

CD48 0.900947

DSG3 0.88464

NKD1 0.880343

MANSC1 0.877133

LAIR1 0.871913

PL TP 0.870614

OXGR1 0.856237

LAMC2 0.851361

GPRC5A 0.849822

LCN2 0.840073

ITGBL1 0.833335

GLRA2 0.813295

DPEP1 0.812086

FRZB 0.802694

AGR2 0.798932

PLS1 0.797142

CEACAM7 0.790504

LGR5 0.779722

MEP1A 0.770133

KRT20 0.761736

CD37 0.754149

FAP 0.752675

TYROBP 0.741951

BCL2L15 0.74073

CPA6 0.736865

HTR1D 0.734167

MIR555 0.732893

SLAMF8 0.726914 GALNT3 0.716639

SLC6A8 0.70102

MYO5C 0.668367

ITGB6 0.666069

HUNK 0.664041

SCEL 0.646821

TAS2R4 0.601437

CD96 0.547906

COL17A1 0.540569

MOXD1 0.536921

PTPRR 0.524766

TFF1 0.507233

GPSM2 0.460679

NCF2 0.323671

MUC20 0.125731

Example 2

Material and Methods:

Colorectal cancer cell lines (n = 104) were screened for sensitivity to libraries of 461 or 528 drugs (largely overlapping). All samples were analyzed for gene expression on Human Transcriptome 2.0 Arrays and classified according to LMS using the prediction model developed for colorectal cancer liver metastases in Moosavi et al., Genome Med 2021;13,143. Welch’s t-test for differential drug sensitivity between LMS1 and the remaining LMS groups (LMS other) were performed.

Results:

There was a large proportion of LMS 1 samples among the cell lines, due to a high frequency of KRAS/NRAS and FF4F^V600E mutations (FIG. 22A). Gene set enrichment analyses confirmed that LMS1 cell lines had similar gene expression characteristics to LMS1 liver metastases, including an MSI-like and serrated phenotype with strong oncogenic signaling (FIG. 22B). Several drugs showed strong relative activity in LMS1 versus LMS- other (FIG. 23) are shown. P-values are included in Table 5.

Table 5: The drugs with strongest differential activity between LMS1 and LMS-other are listed. LMS1 has frequent BRAF ^600E mutations and are more sensitive to BRAF^V600E inhibitors. Few LMS1 samples are wild-type for KRAS/NRAS and FF4F^V600E, and EGFR inhibitors are not active in LMS1. DSS, drug sensitivity score

Example 3

Material and Methods: Tumor organoids (n = 98) were cultured from liver lesions of patients (w=45) treated by hepatic resection for colorectal liver metastases at Oslo University Hospital, following the protocol described in Bruun et al., Clin Cancer Res 2020;26:4107-19. Multiple lesions (median 2, range 2-7) were grown from each of 29 patients. All organoids were screened for sensitivity to 24 drugs, and half of the organoids were screened for 56 drugs. All organoids were analyzed for gene expression on Human Transcriptome 2.0 Arrays and classified according to LMS using the prediction model developed for colorectal cancer liver metastases in Moosavi et al., Genome Med 2021;13,143. Wilcoxon’s rank sum tests were performed to analyze differential drug activity between LMS1 and the remaining LMS groups (LMS other). Results:

The frequency of LMS1 among tumor organoids was similar to liver metastases. LMS1 organoids had frequent BRAF ^600E and KRAS/NRAS mutations (FIG. 24A). Gene set enrichment analyses confirmed that LMS1 organoids had similar gene expression characteristics to LMS1 liver metastases (FIG. 24 B). Drugs with strong relative activity in LMS1 versus LMS-other are shown in FIG. 25. P-values are included in Table 6. Higher sensitivity to BRAF inhibitors in LMS1 is due to the high frequency of FF4F^V600E mutations in this subtype. Samples are colored according to the patient the organoids were derived from, and grey indicates patients with only one organoid.

Table 6: The drugs with strongest differential activity between LMS1 and LMS-other are listed. Few LMS1 samples are wild-type for KRAS/NRAS and FF4F^V600E, and EGFR inhibitors are not active in LMS1. LMS1 also has low sensitivity to standard chemotherapies.

References

1. Cal on A, Lonardo E, Berenguer-Llergo A, Espinet E, Hemando-Momblona X, Iglesias M, et al. Stromal gene expression defines poor-prognosis subtypes in colorectal cancer. Nat Genet. 2015;47:320-9. 2. Dienstmann R, Villacampa G, Sveen A, Mason MJ, Niedzwiecki D, Nesbakken A, et al. Relative contribution of clinicopathological variables, genomic markers, transcriptomic subtyping and microenvironment features for outcome prediction in stage II/III colorectal cancer. Ann Oncol. 2019;30:1622-9.

3. Guinney J, Dienstmann R, Wang X, de Reynies A, Schlicker A, Soneson C, et al. The consensus molecular subtypes of colorectal cancer. Nat Med. 2015;21:1350-6.

4. Isella C, Brundu F, Bellomo SE, Galimi F, Zanella E, Porporato R, et al. Selective analysis of cancer-cell intrinsic transcriptional traits defines novel clinically relevant subtypes of colorectal cancer. Nat Commun. 2017;8:15107.

5. Dienstmann R, Vermeulen L, Guinney J, Kopetz S, Tejpar S, Tabemero J. Consensus molecular subtypes and the evolution of precision medicine in colorectal cancer. Nat Rev Cancer. 2017;17:268.

6. Sveen A, Bruun J, Eide PW, Eilertsen IA, Ramirez L, Murumagi A, et al. Colorectal Cancer Consensus Molecular Subtypes Translated to Preclinical Models Uncover Potentially Targetable Cancer Cell Dependencies. Clin Cancer Res. 2018;24:794-806.

7. Lenz HJ, Ou FS, Venook AP, Hochster HS, Niedzwiecki D, Goldberg RM, et al. Impact of Consensus Molecular Subtype on Survival in Patients With Metastatic Colorectal Cancer: Results From CALGB/SWOG 80405 (Alliance). J Clin Oncol. 2019;37: 1876-85.

8. Stintzing S, Wirapati P, Lenz HJ, Neureiter D, Fischer von Weikersthal L, Decker T, et al. Consensus molecular subgroups (CMS) of colorectal cancer (CRC) and first-line efficacy of FOLFIRI plus cetuximab or bevacizumab in the FIRE3 (AIO KRK-0306) trial. Ann Oncol. 2019;30:1796-803.

9. Sveen A, Cremolini C, Dienstmann R. Predictive modeling in colorectal cancer: time to move beyond consensus molecular subtypes. Ann Oncol. 2019;30: 1682-5.

10. Fontana E, Eason K, Cervantes A, Salazar R, Sadanandam A. Context matters- consensus molecular subtypes of colorectal cancer as biomarkers for clinical trials. Ann Oncol. 2019;30:520-7.

11. Piskol R, Huw L, Sergin I, Kljin C, Modrusan Z, Kim D, et al. A Clinically Applicable Gene-Expression Classifier Reveals Intrinsic and Extrinsic Contributions to Consensus Molecular Subtypes in Primary and Metastatic Colon Cancer. Clin Cancer Res. 2019;25:4431-42.

12. Kamal Y, Schmit SL, Hoehn HJ, Amos CI, Frost HR. Transcriptomic Differences between Primary Colorectal Adenocarcinomas and Distant Metastases Reveal Metastatic Colorectal Cancer Subtypes. Cancer Res. 2019;79:4227-41. 13. Pitroda SP, Khodarev NN, Huang L, Uppal A, Wightman SC, Ganai S, et al. Integrated molecular subtyping defines a curable oligometastatic state in colorectal liver metastasis. Nat Commun. 2018;9:1793.

14. Engstrand J, Nilsson H, Stromberg C, Jonas E, Freedman J. Colorectal cancer liver metastases - a population-based study on incidence, management and survival. BMC Cancer. 2018;18:78.

15. Angelsen JH, Hom A, Sorbye H, Eide GE, Loes IM, Viste A. Population-based study on resection rates and survival in patients with colorectal liver metastasis in Norway. Br J Surg. 2017;104:580-9.

16. Arnold D, Lueza B, Douillard JY, Peeters M, Lenz HJ, Venook A, et al. Prognostic and predictive value of primary tumour side in patients with RAS wild-type metastatic colorectal cancer treated with chemotherapy and EGFR directed antibodies in six randomized trials. Ann Oncol. 2017;28: 1713-29.

17. Le DT, Durham JN, Smith KN, Wang H, Bartlett BR, Aulakh LK, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. 2017;357:409- 13.

18. Overman MJ, McDermott R, Leach JL, Lonardi S, Lenz HJ, Morse MA, et al. Nivolumab in patients with metastatic DNA mismatch repair-deficient or microsatellite instability-high colorectal cancer (CheckMate 142): an open-label, multicentre, phase 2 study. Lancet Oncol. 2017;18:1182-91.

19. Kopetz S, Grothey A, Yaeger R, Van Cutsem E, Desai J, Yoshino T, et al. Encorafenib, Binimetinib, and Cetuximab in BRAF V600E-Mutated Colorectal Cancer. N Engl J Med. 2019;381:1632-43.

20. Vakiani E, Janakiraman M, Shen R, Sinha R, Zeng Z, Shia J, et al. Comparative genomic analysis of primary versus metastatic colorectal carcinomas. J Clin Oncol. 2012;30:2956-62.

21. Brannon AR, Vakiani E, Sylvester BE, Scott SN, McDermott G, Shah RH, et al. Comparative sequencing analysis reveals high genomic concordance between matched primary and metastatic colorectal cancer lesions. Genome Biol. 2014;15:454.

22. Brunsell TH, Sveen A, Bjombeth BA, Rosok BI, Danielsen SA, Brudvik KW, et al. High Concordance and Negative Prognostic Impact of RAS/BRAF/PIK3CA Mutations in Multiple Resected Colorectal Liver Metastases. Clin Colorectal Cancer. 2020;19:e26-e47. 23. Siravegna G, Mussolin B, Buscarino M, Corti G, Cassingena A, Crisafulli G, et al. Clonal evolution and resistance to EGFR blockade in the blood of colorectal cancer patients. Nat Med. 2015;21:795-801.

24. Berg KCG, Eide PW, Eilertsen IA, Johannessen B, Bruun J, Danielsen SA, et al. Multi-omics of 34 colorectal cancer cell lines - a resource for biomedical studies. Mol Cancer. 2017; 16:116.

25. Bruun J, Kryeziu K, Eide PW, Moosavi SH, Eilertsen IA, Langerud J, et al. Patient- Derived Organoids from Multiple Colorectal Cancer Liver Metastases Reveal Moderate Intra-patient Pharmacotranscriptomic Heterogeneity. Clin Cancer Res. 2020;26:4107-19.

26. Gautier L, Cope L, Bolstad BM, Irizarry RA. affy— analysis of Affymetrix GeneChip data at the probe level. Bioinformatics. 2004;20:307-15.

27. Sandberg R, Larsson O. Improved precision and accuracy for microarrays using updated probe set definitions. BMC Bioinformatics. 2007;8:48.

28. Huber W, Carey VJ, Gentleman R, Anders S, Carlson M, Carvalho BS, et al. Orchestrating high-throughput genomic analysis with Bioconductor. Nat Methods. 2015;12:115-21.

29. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47.

30. Hanzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics. 2013;14:7.

31. Brunet JP, Tamayo P, Golub TR, Mesirov JP. Metagenes and molecular pattern discovery using matrix factorization. Proc Natl Acad Sci U S A. 2004;101:4164-9.

32. Gaujoux R, Seoighe C. A flexible R package for nonnegative matrix factorization. BMC Bioinformatics. 2010;! 1:367.

33. Kuhn M. Building Predictive Models in R Using the caret Package. Journal of Statistical Software. 2008;28: 1-26. Epub 2008-09-23. English.

34. Eide PW, Bruun J, Lothe RA, Sveen A. CMScaller: an R package for consensus molecular subtyping of colorectal cancer pre-clinical models. Sci Rep. 2017;7:16618.

35. Sveen A, Kopetz S, Lothe RA. Biomarker-guided therapy for colorectal cancer: strength in complexity. Nat Rev Clin Oncol. 2020;17:11-32.

36. Kim CG, Ahn JB, Jung M, Beom SH, Kim C, Kim JH, et al. Effects of microsatellite instability on recurrence patterns and outcomes in colorectal cancers. Br J Cancer.

2016;115:25-33. 37. Chun YS, Passot G, Yamashita S, Nusrat M, Katsonis P, Loree JM, et al. Deleterious Effect of RAS and Evolutionary High-risk TP53 Double Mutation in Colorectal Liver Metastases. Ann Surg. 2019;269:917-23.

38. Datta J, Smith JJ, Chatila WK, McAuliffe JC, Kandoth C, Vakiani E, et al. Coaltered Ras/B-raf and TP53 Is Associated with Extremes of Survivorship and Distinct Patterns of Metastasis in Patients with Metastatic Colorectal Cancer. Clin Cancer Res. 2020;26:1077-85.

39. Iorio F, Knijnenburg TA, Vis DJ, Bignell GR, Menden MP, Schubert M, et al. A Landscape of Pharmacogenomic Interactions in Cancer. Cell. 2016;166:740-54.

40. Van Cutsem E, Cervantes A, Adam R, Sobrero A, Van Krieken JH, Aderka D, et al. ESMO consensus guidelines for the management of patients with metastatic colorectal cancer. Ann Oncol. 2016;27:1386-422.

41. Mlecnik B, Van den Eynde M, Bindea G, Church SE, Vasaturo A, Fredriksen T, et al. Comprehensive Intrametastatic Immune Quantification and Major Impact of Immunoscore on Survival. J Natl Cancer Inst. 2018;! 10.

All publications and patents mentioned in the above specification are herein incorporated by reference. Various modifications and variations of the described method and system of the invention will be apparent to those skilled in the art without departing from the scope and spirit of the invention. Although the invention has been described in connection with specific preferred embodiments, it should be understood that the invention as claimed should not be unduly limited to such specific embodiments. Indeed, various modifications of the described modes for carrying out the invention that are obvious to those skilled in the medical sciences are intended to be within the scope of the following claims.

Claims

1. A method for characterizing colorectal cancer (CRC) in a sample from a subject diagnosed with CRC, comprising: a) assaying a sample from said subject for the expression levels of one or more genes selected from the group consisting of GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUG 17 and b) characterizing said CRC as LMS1 CRC when the expression levels of said one or more genes are elevated relative to the levels of said genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5.

2. A method for measuring expression of cancer markers in a sample from a subject diagnosed with CRC, comprising: assaying a sample from said subject for the expression levels of two or more genes selected from the group consisting of GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUC 17.

3. A method for providing a prognosis for CRC, comprising: a) assaying a sample from said subject for the expression levels of one or more genes selected from the group consisting of GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUC 17,' b) characterizing said CRC as LMS1 CRC when the expression levels of said one or more genes are elevated relative to the levels of said genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and c) identifying said subject as having poor prognosis when said CRC is characterized as LMS1.

4. A method for providing a prognosis for CRC, comprising: a) characterizing said CRC as LMS1 CRC when the expression levels of said one or more genes selected from the group consisting of GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUC17 in a sample from a subject diagnosed with CRC are elevated relative to the levels of said genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and

57 b) identifying said subject as having poor prognosis when said CRC is characterized as LMS1.

5. The method of claim 4, wherein said subject is post hepatic resection.

6. The method of claim 4 or 5, wherein said poor prognosis is an increased likelihood of liver metastasis and/or decreased 5-year survival.

7. A method for screening compounds, comprising: a) contacting a CRC sample with a test compound; and b) assaying said sample for the expression levels of one or more genes selected from the group consisting of GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUC 17 to characterize said sample as LMS1 CRC.

8. The method of claim 7, comprising identifying compounds that inhibit growth of said CRC sample.

9. The method of claim 7 or 8, wherein said test compound is selected from those listed in Table 4.

10. A method for screening compounds, comprising: a) characterizing a CRC sample as LMS1-5 based on the expression level of one or more genes selected from those in Table 3; b) contacting said sample with a test compound selected from those in Table 4; and c) assaying the ability of said test compound in inhibit growth of said CRC.

11. The method of claim 10, wherein said sample is in vitro, ex vivo, or in vivo.

12. A method for treating CRC, comprising: a) characterizing said CRC as LMS1 CRC when the expression levels of said one or more genes selected from the group consisting of GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUC17 in a sample from a subject diagnosed with CRC are

58 elevated relative to the levels of said genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and b) administering an agent treats LMS1 CRC.

13. A method for treating CRC, comprising: a) characterizing said CRC as LMS1 CRC when the expression levels of said one or more genes selected from the group consisting of GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUC17 in a sample from a subject diagnosed with CRC are elevated relative to the levels of said genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and b) administering a CRC treatment that is not 5-fluoruracil to said subject.

14. The method of claim 13, wherein said treatment is not 5-fluoruracil and folinic acid (FA), Afatinib, Cetuximab, 5-fluoruracil and SN-38 and FA, 5-fluoruracil and Oxaliplatin and FA, oxaliplatin, SN38, Regorafeib or TAS 102.

15. The method of any one of claims 13 to 14, wherein said treatment is selected from the group consisting of OTS167, ONX-0914, sepantronium bromide, encorafenib, gedatolisib, doxorubicin, bemcentinib, napabucasin, and LCL161.

16. A method for characterizing CRC in a sample from a subject diagnosed with CRC, comprising: a) assaying a sample from said subject for the expression levels of one or more genes selected from those listing in Table 3; and b) characterizing said CRC as LMS1-LMS5 based on said expression levels.

17. A method for measuring expression of cancer markers in a sample from a subject diagnosed with CRC, comprising: assaying a sample from said subject for the expression levels of two or more genes selected from the group consisting of those listed in Table 3.

18. A method for providing a prognosis for CRC, comprising: a) assaying a sample from said subject for the expression levels of one or more genes selected from those listing in Table 3;

59 b) characterizing said CRC as LMS1-LMS5 based on said expression levels; and c) providing a prognosis based on said characterization.

19. A method for treating CRC, comprising: a) assaying a sample from said subject for the expression levels of one or more genes selected from the group consisting of GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUG 17 b) characterizing said CRC as LMS1 CRC when the expression levels of said one or more genes are elevated relative to the levels of said genes in a sample from a subject not diagnosed with CRC or in subjects with LMS2-5; and c) treating said subject with an agent that alters the expression level or one or more activities of said genes.

20. The method of any of the preceding claims, wherein said one or more genes is two or more.

21. The method of any of the preceding claims, wherein said one or more genes is 4 or more.

22. The method of any of the preceding claims, wherein said one or more genes is 9 or more.

23. The method of any of the preceding claims, wherein said genes are all of GCNT3, CTSE, REG4, TCN1, LCN2, DSG3, UCA1, SERPINB5, and MUC 17.

24. The method of any of the preceding claims, wherein the sample is selected from the group consisting of a tissue sample, a biopsy sample, a blood sample and a stool sample.

25. The method of any of the preceding claims, wherein the CRC is stage I, II, III or IV.

26. The method of any of the preceding claims, wherein the assaying comprises contacting the sample with a reagent selected from the group consisting of a nucleic acid probe or probes that hybridizes to a respective gene product of the one or more genes, nucleic acid primers for the amplification and detection of a respective gene product of the one or

60 more genes, and an antigen binding protein that binds to a respective gene product of the one or more genes.

61