A pharmacogenetic interaction analysis of second-line chemotherapy with bevacizumab in metastatic colorectal cancer patients: Results from the randomized BEBYP trial

Metastatic colorectal cancer (mCRC) remains a major therapeutic challenge, and although bevacizumab combined with chemotherapy has demonstrated clinical benefit, validated predictive biomarkers of its activity are still missing. Moreover, pharmacogenetic analyses of angiogenesis-related genes in mCRC patients have failed to predict treatment efficacy and survival outcomes. Based on this rationale, we conducted a pharmacogenetic study on a different approach, investigating the potential combined effect of EPAS-1, IL-8, VEGF-A, and VEGFR-2 single nucleotide polymorphisms (SNPs) on progression-free survival (PFS) and overall survival (OS) in mCRC patients treated with second-line mFOLFOX-6 or FOLFIRI (CHT) with or without bevacizumab. Germline DNA was obtained from peripheral blood samples, and SNPs were genotyped by Real-Time PCR. Pharmacogenetic interaction analysis was performed using the Multifactor Dimensionality Reduction (MDR) approach to explore potential statistical interactions between SNPs associated with treatment outcomes. This analysis included mCRC patients enrolled in the randomized phase III BEBYP trial (clinicaltrials.gov: NCT00720512) at the University Hospital of Pisa, who received second-line mFOLFOX-6 or FOLFIRI with (n=66) or without (n=66) bevacizumab. In the bevacizumab+CHT group, the MDR analysis identified two pharmacogenetic interaction profiles based on specific combinations of VEGF-A rs1570360 and VEGFR-2 rs11133360 genotypes. Patients carrying the favorable genetic profile showed a median PFS of 8.89 months compared with 5.23 months for those with the unfavorable profile (p=0.001), with a multivariable Cox hazard ratio (HR) of 0.49 (95% CI, 0.25-0.96; p=0.039). Median OS was 20.69 vs. 9.28 months (p=0.04) for favorable vs. unfavorable profiles, respectively, with an adjusted HR of 0.54 (95% CI, 0.3-0.98; p=0.042). In contrast, among patients treated with CHT alone, no significant differences were observed in PFS (5.52 vs. 4.85 months, p=0.642) or OS (16.31 vs. 16.34 months, p=0.969) between the two genetic profiles. Overall, the interaction between VEGF-A rs1570360 and VEGFR-2 rs11133360 genotypes identified a favorable pharmacogenetic profile associated with improved PFS and OS in patients receiving bevacizumab, whereas no association was observed in the CHT-alone group. These findings suggest that MDR analysis of angiogenesis-related genes identifies patients who benefit from bevacizumab-based therapy and strengthens the rationale for a pharmacogenetic selection of candidate patients for anti-angiogenic strategies in mCRC.