Molecular Subtypes of CRC Associated with Differences in Patient Survival

Issue: March 2015

Different molecular subtypes of colorectal cancer were found to be associated with significant differences in patient survival, according to new research data.

“Our results are consistent with the clinical and scientific community’s growing understanding that not all CRCs are created equal,” Amanda I. Phipps, PhD, from the epidemiology department at University of Washington and the public health sciences division at Fred Hutchinson Cancer Research Center in Seattle, told Healio Gastroenterology. “There were dramatic differences in the survival outlook for patients with the different CRC subtypes we explored.”

Amanda I. Phipps

To evaluate the relationship between patient survival and previously proposed CRC molecular subtypes, Phipps and colleagues collected tumor samples from 2,050 patients enrolled in the population-based Seattle Colon Cancer Family Registry who were diagnosed with invasive CRC from 1998 through 2007 and followed for survival through 2012. They classified samples into five subtypes determined by tumor marker combinations:

type 1: microsatellite instability (MSI)-high, positive for CpG island methylator phenotype (CIMP+), BRAF-mutated, KRAS mutation-negative;
type 2: microsatellite stable (MSS) or MSI-low, CIMP+, BRAF-mutated, KRAS mutation-negative;
type 3: MSS or MSI-low, non-CIMP, BRAF-mutation-negative, KRAS-mutated;
type 4: MSS or MSI-low, non-CIMP, BRAF- and KRAS-mutation-negative; and
type 5: MSI-high, non-CIMP, BRAF- and KRAS-mutation-negative.

Among the 65% of participants with complete tumor marker data, 7% had type 1 CRC, 4% had type 2 CRC, 26% had type 3 CRC, 47% had type 4 CRC, 4% had type 5 CRC and 12% had other combinations. Type 2 patients had the lowest likelihood of being diagnosed with stage I CRC, the lowest 5-year survival (46%), the highest CRC-specific mortality (HR=2.2; 95% CI, 1.47-3.31) and the highest overall mortality (HR=1.55; 95% CI, 1.08-2.22). Patients with type 3 tumors also had higher CRC-specific mortality (HR=1.32; 95% CI, 1.07-1.63), and patients with type 5 tumors had the lowest CRC-specific mortality (HR=0.3; 95% CI, 0.14-0.66).

“In particular, patients with type 2 tumors (ie, microsatellite stable, CIMP+, BRAF-mutated) experienced a significantly poorer prognosis than patients with the predominant type 4 tumors (ie, microsatellite stable, non-CIMP, BRAF wild-type),” Phipps said. “We hope these survival differences will motivate and guide the development of more targeted treatment options for patients with CRC.”

This study advances “some new and important associations between molecular alterations and patient survival,” according to an accompanying editorial by Eric R. Fearon, MD, PhD, and John M. Carethers, MD, both from the department of internal medicine and Comprehensive Cancer Center at University of Michigan. Commenting on this and a similar study of “very large numbers of CRC patients,” they wrote that both “found, with remarkably similar percentages for subgroups, that analysis of MSI status (MMR function) and CIMP status along with KRAS and BRAF mutations were informative for CRC-specific mortality. Both groups also reported that MMR-proficient CRCs with KRAS mutations or especially BRAF mutations had poorer outcomes than MMR-proficient tumors that were wild type for either gene. As the authors appropriately note, the findings offer further evidence that studies of the intertumor molecular heterogeneity of CRCs has merit and value.”

For more information:

Fearon ER. Gastroenterology. 2015;148:10-13.

Phipps AI. Gastroenterology. 2015;148:77-87.

Disclosure: Phipps reports no relevant financial disclosures. See the study for a full list of the other researchers’ relevant financial disclosures. Fearon and Carethers report no relevant financial disclosures.