Pharmacogenetic Predictors of Severe Chronic Peripheral Neuropathy in Stage II-III Colon Cancer (CC) Patients Treated with Oxaliplatin-Based Adjuvant Chemotherapy (CT). A Gemcad Study

ABSTRACT Background Oxaliplatin-based CT is now the standard adjuvant therapy after resection of stage III and selected high-risk stage II CC. However, this treatment is often associated with cumulative peripheral neuropathy. Predicting individual patient risk for developing severe neuropathy could improve the quality of care by allowing the individualization of treatment. Our aim is to identify single nucleotide polymorphisms (SNPs) in genes involved in oxaliplatin sensitivity to predict severe (grade 2-3) oxaliplatin-induced chronic peripheral neuropathy (OXCPN) among stage II-III CC patients who received adjuvant CT. Methods DNA was extracted from formalin-fixed-paraffin-embedded samples from 202 surgically treated high-risk stage II (29.7%) and stage III (70.3%) CC patients receiving adjuvant oxaliplatin and fluoropyrimidines CT (25.24% FOLFOX, 74.75% CAPOX) from January 2004 to December 2008. Median follow-up was 51.4 months (7-96). Genotyping was performed for 35 SNPs in 18 genes involved in oxaliplatin metabolism (ERCC and XRCC groups), cell cycle and drug transport using MassARRAY (SEQUENOM) technology. A total of 177 stage II-III CC patients also treated with oxaliplatin-based CT were enrolled as a validation set. Results Forty-eight (23.8%) patients experienced grade 2-3 OXCPN. The cyclin H (CCNH) (rs2230641) C/C genotype was associated with a higher risk of severe OXCPN (57.1% C/C, 24.2% C/T, 21.3% T/T; RR: 5.197, p = 0.041). In addition, patients harboring the CCNH (rs2230641) C/C and/or ATP-binding cassette subfamily G member (ABCG2) (rs3114018) A/A haplotype had a higher risk of grade 2-3 OXCPN compared to the CCNH (rs2230641) any T and ABCG2 (rs3114018) any C haplotype (36.5% vs. 19.6%, respectively; RR:2.36; 95% CI, 1.17-4.78; p = 0.022). The ability to predict severe OXCPN of this combined analysis was then independently validated in the second cohort (RR: 2.82; 95% CI, 1.41-5.63; p = 0.003). Conclusions Our data suggest that SNPs in CCNH and ABCG2 can predict the development of severe OXCPN in stage II-III CC patients. The analysis of these SNPs may be useful in personalized adjuvant CT. Disclosure All authors have declared no conflicts of interest.