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

Anita Sveen1,2, Jarle Bruun1,2,3, Peter W Eide1,2

  • 1Department of Molecular Oncology, Institute for Cancer Research, Oslo University Hospital, Oslo, Norway.

Insights

Translating consensus molecular subtypes (CMS) of colorectal cancer to preclinical models enables targeted drug discovery. This study identified subtype-specific drug sensitivities and potential repurposing strategies for chemoresistant CMS4 tumors.

Area of Science:

  • Oncology
  • Genomics
  • Translational Research

Background:

  • Standard colorectal cancer treatments have limited efficacy.
  • Consensus molecular subtypes (CMS) offer a framework for personalized treatment but lack translation to preclinical models.
  • Drug discovery is hindered by the absence of reliable preclinical models for CMS stratification.

Purpose of the Study:

  • To translate CMS classification into preclinical colorectal cancer models.
  • To identify subtype-specific drug sensitivities through high-throughput screening.
  • To explore targeted drug repurposing opportunities for distinct CMS groups.

Main Methods:

  • Analysis of CMS in primary colorectal cancers, cell lines, and patient-derived xenografts (PDX).
  • Development of an optimized classifier for preclinical model CMS stratification.
  • High-throughput in vitro drug screening of 459 drugs across 33 cell lines.

Main Results:

  • Validated CMS molecular and clinicopathologic characteristics in 409 primary colorectal cancers.
  • Developed and validated a cancer cell-adapted classifier for 148 cell lines and 32 PDXs, successfully recapitulating CMS biology.
  • Identified subtype-dependent drug responses, including sensitivity to EGFR/HER2 inhibitors in CMS2 and HSP90 inhibitors in CMS1 and CMS4.
  • Demonstrated potential of 5-fluorouracil and luminespib combination to overcome chemoresistance in a CMS4 PDX model.

Conclusions:

  • Successfully translated CMS classification to preclinical colorectal cancer models.
  • Uncovered subtype-specific drug sensitivities and potential therapeutic targets.
  • Identified promising avenues for targeted treatment repurposing, particularly for the chemoresistant CMS4 group.