Background and aim: Pharmacogenetic studies continue to search for pretreatment predictors of chemotherapeutic efficacy and toxicity in metastatic colorectal cancer. Both genome-wide association studies and candidate gene studies have yielded potential genetic markers for chemosensitivity. We conducted a clinical association study, validating the effect of specific genetic markers cited in recently published papers on the efficacy of the oral 5-fluoro-uracil prodrug capecitabine.
Patients and methods: Germline DNA was collected for 268 metastatic colorectal cancer patients from the CAIRO trial, a multicenter phase III trial, randomizing between combined or sequential first-line treatment with capecitabine, irinotecan, and oxaliplatin. Genotyping was performed for eight single-nucleotide polymorphisms (SNPs), using high-resolution melting curves. Four SNPs are located in the MTRR gene, and another four SNPs showed significant association with 5-fluoro-uracil cytotoxicity in a recent in-vitro genome-wide association study. The primary endpoint was progression-free survival (PFS); secondary endpoints were objective response and overall survival.
Results: In patients receiving capecitabine monotherapy, rs4702484, located in ADCY2 and close to MTRR, was associated with slightly reduced PFS for homozygous wild-type patients (CC 6.2 vs. CT 8.0 months; P=0.018). For the other selected genetic markers, we found no association with PFS, overall survival, or radiologic response upon treatment with capecitabine, either in the total study population or in the capecitabine monotherapy subgroup.
Conclusion: With the exception of rs4702484, we found no evidence of an effect on capecitabine chemosensitivity for any of the studied SNPs. More specifically, variants in methionine synthase reductase (MTRR) are not likely associated with capecitabine efficacy.