Introduction: The success of cancer treatments is limited by resistance to chemotherapies, which is difficult to be assessed at the clinical setting. Moreover, molecular drug resistance is multifactorial. For that reason, examination of genetic factors that influence drug metabolism, toxicity, and effectiveness have a great potential to improve treatment. In this study, we present a comprehensive genetic analysis to evaluate the predictive value of a set of polymorphisms on a prospective study of the sensitivity to 5FU/Oxaliplatin based treatments. The purpose is to provide bona fide predictive markers to set the basis to develop an algorithm for treatment optimisation to be implemented at the clinical setting. Methods: Genotyping was performed for known polymorphic variants of genes involved in drug metabolic pathways (TYMS, DPD, MTHFR), drug inactivation (GSTP1) and DNA repair (XPD, XRCC1), by PCR-RFLP method, which consist of an amplification of the sequence including the genetic variant followed by an enzymatic digestion. We prospectively assessed the aforementioned polymorphisms in blood samples from 50 patients undergoing 5-FU/oxaliplatin based treatments, and 50 healthy controls. Clinical outcome was evaluated individually from the beginning of the study. Results: It was found a strong association between the allele with three repeats and variant g (3g) at the promoter of TYMS and a worse response to chemotherapy (95%CI: 2.39-240.63. OR: 24.0; p=0.007) showing more severe side effects, as well as with 6bp insertion allele at the 3’UTR of the same gene (95%CI: 1.17-108.41. OR: 11.25; p=0.03). Moreover, both polymorphisms seem to be at linkage disequilibrium (p-value=0.01), and the haplotype 3R/ins 6bp was significantly associated with a high risk of poorer response to 5FU-based treatment. However, whether it exists a synergistic effect in relation to response of both polymorphisms remains to be determined. The data suggest that the polymorphisms in the promoter and the 3’UTR insertion of the TYMS gene may be associated with resistance to oxaliplatin/5-FU chemotherapy in the adjuvant setting, as previously described. At the short term, there is no statistically significant association between the remaining polymorphisms and response nor clinical outcome to treatment, whereas there seem to be a predisposition or tendency. This points out the relevance of the further evaluation of these patients at the long term along with the resistance. Conclusions: this study demonstrated that the comprehensive genotyping of TYMS might more precisely predict clinical outcome to 5FU/oxaliplatinum-based chemotherapy at the short term. Nevertheless, these markers should be incorporated along with the previous polymorphisms, into long-term prospective clinical trials to confirm its clinical value to individualize treatment for patients with cancer.

[Proc Amer Assoc Cancer Res, Volume 47, 2006]