Colorectal Cancer Consensus Molecular Subtypes Translated to Preclinical Models Uncover Potentially Targetable Cancer Cell Dependencies

Abstract

Purpose: Response to standard oncologic treatment is limited in colorectal cancer. The gene expression-based consensus molecular subtypes (CMS) provide a new paradigm for stratified treatment and drug repurposing; however, drug discovery is currently limited by the lack of translation of CMS to preclinical models.Experimental Design: We analyzed CMS in primary colorectal cancers, cell lines, and patient-derived xenografts (PDX). For classification of preclinical models, we developed an optimized classifier enriched for cancer cell-intrinsic gene expression signals, and performed high-throughput in vitro drug screening (n ¼ 459 drugs) to analyze subtype-specific drug sensitivities.Results: The distinct molecular and clinicopathologic characteristics of each CMS group were validated in a single-hospital series of 409 primary colorectal cancers. The new, cancer cell-adapted classifier was found to perform well in primary tumors, and applied to a panel of 148 cell lines and 32 PDXs, these colorectal cancer models were shown to recapitulate the biology of the CMS groups. Drug screening of 33 cell lines demonstrated subtype-dependent response profiles, confirming strong response to EGFR and HER2 inhibitors in the CMS2 epithelial/canonical group, and revealing strong sensitivity to HSP90 inhibitors in cells with the CMS1 microsatellite instability/immune and CMS4 mesenchymal phenotypes. This association was validated in vitro in additional CMS-predicted cell lines. Combination treatment with 5-fluorouracil and luminespib showed potential to alleviate chemoresistance in a CMS4 PDX model, an effect not seen in a chemosensitive CMS2 PDX model.Conclusions: We provide translation of CMS classification to preclinical models and uncover a potential for targeted treatment repurposing in the chemoresistant CMS4 group.

Long noncoding RNA MIR31HG is a bona fide prognostic marker with colorectal cancer cell‐intrinsic properties

“…( a ) Barplot depicts MIR31HG expression across the Cancer Cell Line Encyclopedia with colorectal cancer samples highlighted in black. ( b ) Beanplot illustrates distributions in median centered MIR31HG expression estimates across normal colonic mucosa, primary and metastatic lesions as well as cell line and PDX models. Numbers below labels indicate the number of samples.…”

“…NCI‐60 cancer cell line ( n = 59) qRT‐PCR miRNA data were downloaded from https://wiki.nci.nih.gov/display/NCIDTPdata (accessed 2015‐Mar‐30) . In‐house pCRC ( n = 409) and normal colonic mucosa ( n = 11) Affymetrix Human Exon 1.0 ST and HTA2.0 microarray gene expression data have previously been published and are available from GEO with accession identifiers GSE24550, GSE29638, GSE69182, GSE79959, GSE97023 and GSE96528 . Normal colonic mucosa ( n = 31) HTA2.0 gene expression will be submitted to GEO.…”

“…For survival analysis, three cohorts of pCRCs were merged (total n = 1,265). The in‐house Norwegian Oslo series is a consecutive, population‐based collection of biopsies from patients that received surgery for stage I–IV pCRCs between 2005 and 2013 ( n = 409) . The French multi‐center Cartes d'Identité des Tumeurs (CIT) is a cohort of stage I–IV colon cancers that underwent surgery between 1987 and 2007 ( n = 566) .…”

“…Colorectal cancer (CRC) patient classification according to the biologically distinct and gene expression‐based consensus molecular subtypes (CMS) identifies a poor‐prognostic CMS4‐mesenchymal subgroup . The value of CMS as a potential framework for stratified treatment is further supported by emerging evidence of subtype‐associated treatment responses . Bulk tumor gene expression profiles do however bear strong imprints of the sampled tumor microenvironment, and both immune infiltration and stromal abundance are correlated to CMS and provide interdependent prognostic information.…”

The Winding Roadmap of Biomarkers Toward Clinic: Lessons from Predictors of Resistance to Anti-EGFRs in Metastatic Colorectal Cancer

“…Although the CMS classification is manly trained on complex transcriptomic patterns and developed as a stratification tool with a clear prognostic impact across all stages of CRC disease [14], these transcriptomic subtypes have been tested also for a potential predictive value with regard to targeted agents. Indeed, based on data from preclinical models, the CMS2 subtype (i.e., “canonical subtype”, characterized by epithelial differentiation and marked expression of a number of oncogenes, among them EGFR, HER2, insulin-like growth factor 2 (IGF2), insulin receptor substrate 2 (IRS2) and transcription factor hepatocyte nuclear factor 4α (HNF4A)) seemed to have a stronger sensitivity to anti-EGFR agents compared to other CMSs [69]. However, the preclinically suggested ability of CMS subtypes to predict the benefit from targeted agents has not been confirmed in post-hoc analyses of randomized phase III trials comparing bevacizumab- versus cetuximab-based first-line therapy for mCRC patients [70,71].…”