Impact of BRAF and RAS mutations on first-line efficacy of FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab: analysis of the FIRE-3 (AIO KRK-0306) study

Abstract Background RAS and BRAF mutations have been identified as negative prognostic factors in metastatic colorectal cancer. Efficacy of 5-fluorouracil, leucovorin, irinotecan( FOLFIRI) plus bevacizumabin patients with RAS -mutant tumours needs to be further evaluated. Whether to treat patients with BRAF -mutant tumours with either bevacizumabor anti-epidermal growth factor receptor (EGFR) antibodies remains unclear. Methods Patients treated within the FIRE-3 trial were retrospectively tested for BRAF and RAS mutations using formalin fixated paraffin embedded (FFPE) tumour material applying pyrosequencingfor KRAS and NRAS exon 2, 3 and 4 mutations as far as for BRAF mutations. Survival analysis was done using Kaplan–Meier estimationand differences were expressed using the log-rank test. Overall response rate (ORR) was compared using Fisher's exact test. Data from a central independent radiological response evaluation were used to calculate early tumour shrinkage (ETS) and depth of response (DpR). Results Overall, 188 patients with RAS -mutant tumours and 48 with BRAF -mutant tumours were identified. In BRAF -mutant patients, ORR was numerically higher in the cetuximabversus the bevacizumabarm (52% versus 40%), while comparable results were achieved for progression-free survival (PFS; hazard ratio [HR] = 0.84, p = 0.56) and overall survival (OS; HR 0.79, p = 0.45). RAS mutation was associated with a trend towards lower ORR (37% versus 50.5%, p = 0.11) and shorter PFS (7.4 versus 9.7 months; HR 1.25; p = 0.14) in patients receiving FOLFIRIplus cetuximabversus bevacizumab, but OS was comparable (19.1 versus 20.1 months; HR 1.05; p = 0.73), respectively. ETS identified subgroups sensitive to cetuximab-based treatment in both BRAF - (9/17) and RAS -mutant (18/48) patients and was associated with significantly longer OS. DpR was comparable between both treatment arms in RAS - and BRAF -mutant patients, respectively. Conclusions In BRAF - and RAS -mutant patients, cetuximab- and bevacizumab-based treatment had comparable survival times. ETS represents an early parameter associated with the benefit from anti-EGFR, while this was not the case with vascular endothelial growth factor A blockade.