Liquid biopsy to monitor genetic alterations in plasma as non-invasive pharmacogenetic-based approach to evaluate predictive biomarkers during treatment in cancer patients.

At present, in oncology the therapeutic strategies are defined according to the molecular landscape assessed with tissue biopsies and, thus, using DNA and/or RNA obtained from a fragment of the primary tumour or a single metastatic lesion. Anyway, mutational analysis in a single-site biopsy is encumbered by several limitations. A huge step forward in oncology to characterize tumor heterogeneity and plasticity at diagnosis and during the course of treatment was done with the use of a promising minimally invasive tool., i.e. via a blood draw. All cells can release in human blood their materials, including cell-free DNA (cfDNA), RNA, proteins and vesicles (such as exosomes). This approach is the expression of the overall spatial heterogeneity from the overall sites of disease and it is able to identify emerging sub-clones responsible for treatment resistance. This thesis was aimed at evaluating if a pharmacogenetic approach based on circulating predictive biomarkers (i.e. circulating cell-free tumor DNA and RNA extracted from exosomes) may be a good and useful tool to monitor treatment outcome in solid tumors. Patients affected by lung, breast, prostate, pancreatic cancer and melanoma have been enrolled and monitored during different treatment, i.e. targeted therapy, hormonal therapy, chemotherapy and immunotherapy by circulating tumor nucleic acids. Nucleic acids were analysed by digital droplet PCR for targetable drivers, for acquired mutations and gene expression selected during therapy. Circulating tumor nucleic acids were find to correlate with tumor burden, disease status and response to treatment, beeing a good candidate to monitor tumor response and support routine clinical practice.