APrognostic and Carboplatin Response Predictive Model in Ovarian CancerA Mono-Institutional Retrospective Study Based on Clinics and Pharmacogenomics

Carboplatin is the cornerstone of ovarian cancer (OC) treatment, while platinumresponse,dependent on interindividual variability, is the major prognostic factor for long-termoutcomes. This retrospective study was focused on explorative search of genetic polymorphisms inthe Absorption, Distribution, Metabolism, Excretion (ADME) genes for the identification ofbiomarkers prognostic/predictive of platinum-response in OC patients. Ninety-two advanced OCpatients treated with carboplatin-based therapy were enrolled at our institution. Of these, weshowed that 72% of patients were platinum-sensitive, with a significant benefit in terms of OS (p =0.001). We identified an inflammatory-score with a longer OS in patients with lower scores ascompared to patients with the maximum score (p = 0.001). Thirty-two patients were genotyped for1931 single nucleotide polymorphisms (SNPs) and five copy number variations (CNVs) by theDMET Plus array platform. Among prognostic polymorphisms, we found a potential role ofUGT2A1 both as a predictor of platinum-response (p = 0.01) and as prognostic of survival (p = 0.05).Finally, we identified 24 SNPs related to OS. UGT2A1 correlates to an "inflammatory-score" andretains a potential prognostic role in advanced OC. These data provide a proof of concept thatwarrants further validation in follow-up studies for the definition of novel biomarkers in thisaggressive disease.