Dissecting clonal evolution to unveil vulnerabilities of chemoresistant ovarian cancer

High grade serous ovarian cancer (HGSOC) is the most lethal gynecologic malignancy, largely due to its asymptomatic onset leading to late-stage diagnosis. Standard treatment combines platinum (Pt)-based chemotherapy and PARP inhibitors (PARPi), which are particularly effective in tumors with homologous recombination (HR) deficiency. Despite initial responses, most patients relapse and ultimately develop resistance to Pt and PARPi, a phenomenon known as Pt/PARPi cross-resistance. Increasing evidence suggest that therapy-driven selective pressure fuels clonal evolution and diversification, fostering tumor heterogeneity and adaptability that enable the emergence of resistant subpopulations. Dissecting the dynamic of clonal selection and adaptation is therefore crucial for identifying vulnerabilities that can be exploited therapeutically. In this PhD project, I investigate the clonal dynamics of HR proficient (HRP) and HR deficient (HRD) HGSOC under Pt/PARPi treatment, aiming to uncover novel molecular mechanisms of cross-resistance and define predictive signatures of treatment response. To this end, I established a panel of HGSOC patient-derived xenograft (HGSOC- PDX) cell lines that maintain histopathological features of HGSOC while preserving high intra-tumor heterogeneity, providing a reliable and robust system to study in-depth clonal evolution in response to treatment. Specifically, low-passage HGSOC-PDX cells were treated with Pt alone or in combination with different generations of PARPi (i.e. niraparib and saruparib) to assess their responses. As expected, HRD models displayed enhanced sensitivity to PARPi and a stronger Pt-PARPi synergy compared to HRP counterparts, while all models remained sensitive to initial Pt treatment. To longitudinally track clonal evolution, we developed clonal replica tumors (CRT) from barcoded HGSOC cells. Using a high-complexity barcoded library at low multiplicity of infection (MOI < 0.3), we generated uniquely tagged cell populations that were stabilized through passaging. Orthotopic implantation of HRP and HRD barcoded models into immunodeficient female mice enabled the generation of large-scale in vivo cohorts with identical clonal composition. Combined Pt/PARPi treatment revealed that HRD CRT displayed prolonged animal lifespan and a significant reduction of metastasis under Pt/PARPi therapy. Strikingly, clonal analyses uncovered the consistent and distinct emergence of Pt/PARPi cross-resistant clones across both HRP and HRD models. In conclusion, this thesis establishes a powerful quantitative and functional platform to track HGSOC clonal evolution with unprecedented resolution. Our findings demonstrate that mechanisms of Pt/PARPi cross-resistance arise independently of HR status, providing a unique framework to dissect the genomic, transcriptomic, and epigenetic features of resistant clones and to identify exploitable vulnerabilities for future therapies.