Pathway level subtyping identifies a slow-cycling and transcriptionally lethargic biological phenotype associated with poor clinical outcomes in colon cancer independent of genetics

Abstract

Molecular stratification, across many tumour types, has used gene-level transcriptional data to identify subtypes associated with distinct genotypes and biological traits, as exemplified by the consensus molecular subtypes (CMS), and more recently the intrinsic CMS (iCMS), in colorectal cancer. In an attempt to develop molecular subtypes that more closely align to cancer-relevant phenotypic traits in KRAS mutant tumours, here we present an approach that uses gene ontology and biological activation state information, rather than gene-level data, for the initial stages of class discovery. In doing so, we define three unique pathway-derived subtypes (PDS); where PDS1 tumours are highly proliferative and display good prognosis, PDS2 tumours are stroma/immune-rich with intermediate prognosis. The final subtype, PDS3, represent a previously overlooked subset of tumours within CMS2, which display a ‘lethargic’ biological phenotype with neural-like traits and the worst prognosis. Remarkably, these biological and clinical features remain consistent across tumour samples independent of KRAS mutational status, supporting the use of PDS for defining cancer-relevant phenotypes regardless of genetics.

Image-Based Consensus Molecular Subtyping in Rectal Cancer Biopsies and Response to Neoadjuvant Chemoradiotherapy

“…The results of this study therefore support the development of an inexpensive clinical tool to assign patients to subtype targeted biological interventions in future clinical trials. Beyond CMS in CRC, the on-going development morpho-molecular classification models across cancer types and molecular signatures offer a new type of cost-effective computational tools to support clinical decision making [18,36]. This surrogate for transcriptional analysis can also be of use for research studies with restricted funding, or to revisit existing trial cohorts by studying associations between image-based CMS classification and clinical variables of interest without the need for extra tissue material.…”