Toxicogenomic profiling of fluorouracil/leucovorin/oxaliplatin/docetaxel (FLOT) in advanced gastric cancer: A translational study of the German Arbeitsgemeinschaft Internistische Onkologie (AIO).

Authors:

Abstract
E14624 Background: To develop toxicogenomic profiles as predictors of clinically relevant grade 3/4 toxicities of chemotherapy in advanced gastric cancer (AGC). Methods: This randomized phase II study compared fluorouracil, leucovorin, oxaliplatin ± docetaxel (FLO vs. FLOT) in pts with AGC. As part of the study protocol blood samples of 143 pts were prospectively collected for DNA extraction and genotyping. 24 polymorphisms in 17 genes as part of the metabolism or transmembrane transport of the applied drugs, the folate metabolism or DNA repair were analyzed. Toxicity data were analysed using aggregated data and univariate Poisson regression to take into account varying therapy cycles. Multiple regression analysis was performed using non-aggregated data with repeated measures and generalized estimation equations in SAS Proc Genmod. Thereby, it was possible to include maximum information of each cycle per individual. The modified variance structure was modelled as exchangeable. Results: For clinical study data see abstract of Al-Batran et al. Univariate analyses revealed significant associations between various covariates (genotypes, clinical factors) and toxicities. Generalized estimation equations confirmed most of them in the multivariate setting. Significant associations were observed between neutropenia and treatment arm (p = 0.002), TS1394del6 (p = 0.016), GSTT1 (p = 0.028), GSTM1 (p = 0.023), ABCG2-C421A (p < 0.001), XRCC1-R399Q (p < 0.001), XPD-D312N (p = 0.004), XPD-K751Q (p = 0.007), SDF1-1'A (p = 0.033); between diarrhea and treatment arm (p = 0.001), ABCB1-G2677T (p = 0.032), XPD-K751Q (p = 0.013), MAD1-G558A (p = 0.015); between sensory neuropathy and performace status (p = 0.001), oxaliplatin dose (p = 0.009), XPD-K751Q (p = 0.029), OPRT-G213A (p = 0.003), SDF1-1'A (p = 0.035), MAD1-G558A (p = 0.017). Conclusions: Toxicogenomic profiling represents a powerful tool to identify potential relevant polymorphisms to predict chemotherapy induced toxicity in AGC pts. Validation of these markers is warranted and ongoing within the next clinical study of FLOT in AGC that has recently closed accruel with a total of 250 pts. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration sanofi-aventis sanofi-aventis