Molecular Subtypes and the Evolution of Treatment Decisions in Metastatic Colorectal Cancer

Colorectal cancer (CRC) was one of the first solid tumors to be molecularly characterized, with several genes and pathways implicated in tumor initiation and growth.1 CRC carcinogenesis is the result of a stepwise accumulation of genetic events in oncogenes (e.g., KRAS, NRAS, BRAF, PIK3CA) and tumor suppressor genes (e.g., APC, TP53, SMAD4, PTEN) that deregulate key signaling pathways driving progression, such as WNT/β-catenin, transforming growth factor beta (TGFβ), epidermal growth factor receptor (EGFR), and downstream mitogen-activated protein kinase (MAPK) and phosphoinositide 3-kinase (PI3K). These alterations are found in most advanced tumors, which can be biologically classified as either chromosomally instable or microsatellite instable (MSI), depending on the pattern of additional genomic and epigenomic events. Although tumors that are chromosomally instable harbor increased copy-number alterations linked to TP53 mutations, MSI tumors have hypermutation as a result of defective DNA mismatch repair (MMR) system, with a clear molecular origin (MLH1, MSH2, MSH6, or PMS2 inactivation) arising from germline or somatic events.2 With the recent characterization of the transcriptomic subtypes of the disease (consensus molecular subtypes [CMS]), convergent pathway dependencies were identified: CMS1 (MSI immune) is hypermutated and MSI, with strong immune cell infiltration and cytotoxic signaling activation; CMS2 (canonical) is epithelial, chromosomally instable, immune desert, and shows marked WNT and EGFR signaling dependence; CMS3 (metabolic) is epithelial and has mixed chromosomal and microsatellite instability, evident metabolic dysregulation, and mutations in the MAPK pathway; and CMS4 (mesenchymal) is chromosomally instable, with prominent TGFβ activation linked to microenvironment infiltration with immunosuppressive and stromal cells.3 The CMS groups reflect distinct biology of primary CRC tumors and have a clear prognostic impact in early-stage and advanced disease, with CMS4 mesenchymal tumors having twofold increased chances of relapse after curative therapy and CMS1 MSI immune tumors associated with dismal prognosis in the metastatic setting.3,4 In parallel with our increased understanding of molecular drivers and tumor-microenvironment interactions in CRC, major advances in drug development of targeted agents and immunotherapies have shown the value of tumor profiling in the clinic. Molecular heterogeneity plays an important role in treatment selection in metastatic CRC (mCRC), with KRAS/NRAS (RAS) mutated tumors intrinsically resistant to