Consensus molecular subgroups (CMS) of colorectal cancer (CRC) and first-line efficacy of FOLFIRI plus cetuximab or bevacizumab in the FIRE3 (AIO KRK-0306) trial

BACKGROUND: FIRE-3 compared first-line therapy with FOLFIRI plus either cetuximab or bevacizumab in 592 KRAS exon 2 wild-type metastatic colorectal cancer (mCRC) patients. The consensus molecular subgroups (CMS) are grouping CRC samples according to their gene-signature in four different subtypes. Relevance of CMS for the treatment of mCRC has yet to be defined. PATIENTS AND METHODS: In this exploratory analysis, patients were grouped according to the previously published tumor CRC-CMSs. Objective response rates (ORR) were compared using chi-square test. Overall survival (OS) and progression-free survival (PFS) times were compared using Kaplan–Meier estimation, log-rank tests. Hazard ratios (HR) were estimated according to the Cox proportional hazard method. RESULTS: CMS classification could be determined in 438 out of 514 specimens available from the intent-to-treat (ITT) population (n = 592). Frequencies for the remaining 438 samples were as follows: CMS1 (14%), CMS2 (37%), CMS3 (15%), CMS4 (34%). For the 315 RAS wild-type tumors, frequencies were as follows: CMS1 (12%), CMS2 (41%), CMS3 (11%), CMS4 (34%). CMS distribution in right- versus (vs) left-sided primary tumors was as follows: CMS1 (27% versus 11%), CMS2 (28% versus 45%), CMS3 (10% versus 12%), CMS4 (35% versus 32%). Independent of the treatment, CMS was a strong prognostic factor for ORR (P = 0.051), PFS (P < 0.001), and OS (P < 0.001). Within the RAS wild-type population, OS observed in CMS4 significantly favored FOLFIRI cetuximab over FOLFIRI bevacizumab. In CMS3, OS showed a trend in favor of the cetuximab arm, while OS was comparable in CMS1 and CMS2, independent of targeted therapy. CONCLUSIONS: CMS classification is prognostic for mCRC. Prolonged OS induced by FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab in the FIRE-3 study appears to be driven by CMS3 and CMS4. CMS classification provides deeper insights into the biology to CRC, but at present time has no direct impact on clinical decision-making. The FIRE-3 (AIO KRK-0306) study had been registered at ClinicalTrials.gov: NCT00433927.