Cytidine deaminase single-nucleotide polymorphism is predictive of toxicity from gemcitabine in patients with pancreatic cancer: RTOG 9704

The aim of this study is to validate the prognostic and predictive value of the non-synonymous cytidine deaminase (CDA) Lys(27)Gln polymorphism for hematological toxicity and survival using a randomized phase III adjuvant trial (Radiation Therapy Oncology Group (RTOG) 9704) in pancreatic cancer in which one treatment arm received gemcitabine. CDA is involved in gemcitabine inactivation, and there is conflicting data on the role of the non-synonymous CDA Lys(27)Gln polymorphism in predicting toxicity and survival in cancer patients treated with gemcitabine. RTOG 9704 randomized 538 patients after pancreatic resection to receive radiotherapy with either 5-fluorouracil (5-FU) or gemcitabine. CDA Lys(27)Gln polymorphism genotype was analyzed. We tested an association between CDA single-nucleotide polymorphism genotype and the survival outcome by the Cox proportional hazard model adjusting for other covariates, as well as toxicity by the logistic regression model. There is statistically significant more severe hematological toxicity in patients treated with gemcitabine with either the homozygote wild-type genotype (Lys/Lys) alone (odds ratio (OR) = 0.06, P = 0.01), or in combination with the heterozygote (Lys/Gln; OR = 0.14, P = 0.03) when compared with homozygote variant genotype (Gln/Gln) when adjusted for other covariates. This was not seen in the non-gemcitabine treated arm. There are no genotype differences with respect to survival outcome. In conclusion, in this prospective randomized adjuvant study of patients with pancreatic cancer, the CDA Lys(27)Gln polymorphism is validated as a predictive marker of gemcitabine hematological toxicity, but not with treatment response or survival. The Pharmacogenomics Journal (2012) 12, 395-403; doi:10.1038/tpj.2011.22;published online 31 May 2011