Functional tumour heterogeneity in advanced ovarian cancer

Abstract

High-grade serous ovarian cancer (HGSOC) is the most lethal gynaecological malignancy. It is characterised by high degrees of genomic instability and heterogeneity, with many patients initially responding to chemotherapy but ultimately relapsing as they develop of resistance to platinum chemotherapy. The diversity of platinum resistance mechanisms and complete lack of predictive biomarkers means that delivering the best treatment for ovarian cancer patients remains challenging. Novel therapies targeting the molecular pathways dysregulated in HGSOC are being explored with the goal of personalised treatment and improved outcome for patients. This study aims to understand the extent of intra-tumour heterogeneity (ITH) in HGSOC and define the relationship between ITH at the phenotypic and molecular levels with clinical outcomes.

Multiple tumour samples were collected from defined anatomical regions during maximal effort primary debulking surgery for HGSOC, and relapse samples were collected for comparative studies. Tumour cells were extracted, placed in short-term cultures, treated with cisplatin and/or DNA-PKcs inhibitor, and levels of apoptosis and viability were then measured. Cisplatin chemotherapy-induced signalling changes were investigated in primary tumour cells using reverse phase protein array (RPPA) proteomics. Phylogenetic trees were constructed, and clonal expansion indices (CEI) were quantified using the Minimum Event Distance for Intra-tumour Copy-number Comparisons (MEDICC) algorithm.

Data from phenotypic apoptosis assays demonstrated spatial and temporal heterogeneity in response to cisplatin across different disseminated tumours. There was a trend of higher variability in response to cisplatin showing a better progression-free survival in our cohort. We observed that the combination of DNA-PKcs inhibitor and cisplatin enhanced apoptosis in primary ovarian tumour cells. Our results also demonstrated vast proteomic heterogeneity in our cohort by examining platinum-induced protein signaling changes in our samples. Linear regression analyses of proteomic data showed significant alterations of HSP27, p38, and Chk1 proteins amongst others following cisplatin treatment. Pathway enrichment analysis showed the PI3K-AKT signaling pathway as a significantly enriched pathway induced by cisplatin. eEF2K and PR were identified as potential biomarkers of prognosis. Inhibition of Chk1 using prexasertib was demonstrated to enhance and restore platinum sensitivity in combination with cisplatin treatment in platinum sensitive and resistant HGSOC cells. Furthermore, CNV data analysis revealed vast heterogeneity across our sample cohort, with a higher clonal expansion index showing association with worse survival outcomes.

Taken together, the results in this thesis illustrates clearly the inter- and intratumour heterogeneity that exists in HGSOC. While it has been demonstrated before at the CNV level, this is the first time showing phenotypic heterogeneity and to some extent proteomic heterogeneity in HGSOC. Our research supports further investigation into the prognostic biomarker selection and clinical development of DNA-PK and Chk1 inhibitors in combination with cisplatin to overcome platinum resistance in ovarian cancer. The association of the clonal expansion index with survival outcomes warrants further investigation for CEI as a predictive factor of platinum-response in HGSOC.