• Predictive biomarkers
• pharmacogenetics
• clinical pharmacology
• oncology
• treatment toxicity
• treatment resistance

The first part of the study was aimed at retrospectively evaluating the possible association of DPD polymorphisms and treatment-associated toxicities occurred in patients given fluoropyrimidines. The study included 1254 patients divided into two groups. Cohort 1 included 982 patients suffering from gastrointestinal G≥2 and hematological G≥3 ADRs that were examined for DPYD variants c.496A>G, c.1236G>A, c.1601G>A (DPYD*4), c.1627A>G (DPYD*5), c.1679T>G (DPYD*13), c.1896T>C, c.1905+1G>A (DPYD*2A), c.2194G>A (DPYD*6), and c.2846A>T to assess their association with ADRs. Cohort 2 (control group) comprised 272 subjects not requiring dose reduction, delay or discontinuation of treatment that were screened for the same variants of cohort 1. Genetic analysis in the two cohorts were done by Real-Time PCR of DNA extracted from 3 ml of whole blood. The DPYD c.496A>G, c.1601G>A, c.1627A>G, c.1896T>C, and c.2194G>A variants were found in both cohort 1 and 2 while c.1905+1G>A and c.2846A>T were present only in cohort 1. DPYD c.1679T>G and c.1236G>A were not found. Univariate analysis allowed the selection of c.1905+1G>A, c.2194G>A and c.2846A>T alleles as significantly associated with gastrointestinal and hematological ADRs (p<0.05), while the c.496A>G variant showed a positive trend of association with neutropenia (p=0.06). The use of dihydropyrimidine dehydrogenase (DPYD) as a routinary test to predict fluoropyrimidines adverse drug reactions (ADRs) is still debated. The present study associates c.2194G>A to clinically relevant ADRs in addition to the already known c.1905+1G>A and c.2846A>T and suggests the analysis of these three variants to reduce the risk of fluoropyrimidine-associated ADRs.
The second part of the project was aimed at evaluating if circulating tumor nucleic acids may be a good and useful tool to monitor treatment outcome in solid tumors. Patients affected by lung cancer, prostate cancer, pancreatic cancer and melanoma have been enrolled and monitored during treatment with chemo- or target therapy by circulating tumor nucleic acids. Nucleic acids were extracted from plasma samples and analysed by digital droplet PCR for targetable drivers or for acquired mutations selected during therapy. Circulating tumor nucleic acids were find to correlate with tumor burden, disease status and rensponse to treatment, beeing a good candidate to monitor tumor response.