The Predictive Role Of MTHFR And ABCG2Single Nucleotide Polymorphisms On The Outcome Of Advanced Colorectal Cancer Treated With First-Line Chemotherapy

[Objective] Chemotherapy is one of the main treatments for advanced colorectal cancer, with the same efficacy of the regimen contained with L-OHP (FOLFOX or XELOX) or with CPT-11(FOLFIRI) in first-line treatment of advanced colorectal cancer patients, but not all patients can benefit from the chemotherapy they received. Single nucleotide polymorphism(SNP) is a hot spot in the present study as predictive or prognostic markers, and studies have showed that the SNPs of MTHFR C677T and A1298C are associated with the efficacy of chemotherapy for colorectal cancer, but the conclusions are not yet uniform; It was reported that ABCG234G>A was associated with relative susceptibility to FOLFOX and resistance to FOLFIRI in Caucasians; Vitro studies have showed that the SNP of ABCG2C421A can influence the expression of ABCG2, thus affecting the removal of certain chemotherapy drugs. Therefore, the purpose of this study is to explore (1)the predictive factors for the efficacy of first-line chemotherapy in advanced colorectal cancer, and (2) the predictive factors for the different efficacies of different chemotherapy regimens, combining with the SNPs of MTHFR C677C, A1298C, ABCG2G34A and C421A and the clinical features of patients, which may provide basis for the individual treatment in the advanced colorectal cancer patients.[Methods] We retrospectively collected advanced colorectal cancer patients treated with first-line standard FOLFOX/XELOX or FOLFIRI regimen between January2009and May2011at Fudan University Cancer Hospital, closely following up the efficacy evaluation, time to disease progression and time to death. The gene polymorphisms of MTHFR and ABCG2were detected using gene sequencing method, after DNA extraction from peripheral blood karyocytes and PCR amplification. All statistical analysis was performed using SPSS software package17.0with P≤0.05as statistically significant. Univariate analysis by chi-square test or Fisher exact test and multivariate analysis by logistic regression model were used to analyse early response rate. Kaplan-Meier curve with Log-rank test was used in PFS and OS univariate analysis, and Cox regression model was used in PFS and OS multivariate analysis.[Results] Of154patients,92received first-line FOLFOX or XELOX chemotherapy, while62received FOLFIRI. The median PFS was8. lm (95CI:6.9～9.3m), median OS was30.7m (95%CI:18.2～43.2m), with the early response rate of31.8%. The univariate analysis showed that PFS was associated with the number of favorable genotypes of MTHFR677C/C, MTHFR1298A/A, ABCG2G/A or A/A, and ABCG2421C/A or A/A, radical resection of primary lesion and the number of metastatic organs. Patients with3-4favorable genotypes had a longer PFS than those with0-2favorable genotypes (9.8m vs.7.5m, P=0.013), patients with radical resection of primary lesion had a longer PFS (8.8m vs.6.3m, P=0.001), and patients with a single metastatic organ had a longer PFS(8.6m vs.7.1m, P=0.034). The multivariate analysis showed that the number of favorable genotypes (HR=0.627,95%CI:0.427～0.920, P=0.017) and radical resection of primary lesion (HR=0.524,95%CI:0.320～0.859, P=0.010) were independent factors for PFS. Both univariate and multivariate analysis showed that OS was associated with radical resection of primary lesion. For response rate, on univariate analysis, radical resection of primary lesion and the time of metastasis were influencing factors, but there was no significant difference on the multivariate analysis.The stratified univariate analysis showed that, on univariate analysis, in FOLFIRI favorable group with0～1of MTHFR677C/C、MTHFR1298A/C or C/C、ABCG234G/G and ABCG2421C/A or A/A genotypes, PFS was associated with radical resection of primary lesion, the number of metastatic organs and the time of metastasis, and in FOLFXO/XELOX favorable group with2～4above-mentioned genotypes, PFS was associated with radical resection of primary lesion, the response rate of patients receiving FOLFOX or XELOX was higher than those receiving FOLFIRI. The multivariate analysis showed that in FOLFIRI favorable group, the first-line chemotherapy regimen(P=0.019), the time of metastasis(P=0.002) and radical resection of primary lesion (P=0.001) were independent factors for PFS, with reduced progression risks of57.8%,23.8%and82.9%when treated with FOLFIRI, with synchronism and radical resection of primary lesion respectively; while in FOLFOX/XELOX favorable group, the first-line chemotherapy regimen (P=0.040) and the time of metastasis (P=0.039) were independent factors for PFS, with increased progression risks of1.722and1.757times when treated with FOLFIRI and with synchronism respectively. However, for response rate, it only showed that in FOLFIRI favorable group, sex was an independent factor (P=0.027), and the risk for disease progression in women was3.468times to that in men.[Conclusions] SNPs detection may play a predictive role in selecting first-line chemotherapy for advanced colorectal cancer patients and may predict the efficacy. Radical resection of primary lesion is an independent factor for both PFS and OS.