Multiomic characterisation of high grade serous ovarian carcinoma enables high resolution patient stratification

ABSTRACT Background High grade serous ovarian carcinoma (HGSOC) is the most common type of ovarian cancer; most patients experience disease recurrence which accumulates chemoresistance, leading to treatment failure. Previous investigations have characterised HGSOC at the genomic and transcriptomic level, identifying subtypes of patients with differential outcome and treatment response. However, the relationship between molecular events identified at the gene sequence, gene copy number and gene expression levels remains poorly defined. Methods We perform multi-layer molecular characterisation of a large retrospective HGSOC cohort (n=362) with detailed clinical annotation to interrogate the relationship between patient groups defined by gene mutation, copy number events, gene expression patterns and infiltrating immune cell burden. We construct a high resolution picture of the molecular landscape in HGSOC and identify features of tumours associated with distinct clinical behaviour in patients. Results BRCA2 -mutant ( BRCA2 m) and EMSY -overexpressing cases demonstrated prolonged survival (multivariable hazard ratio 0.40 and 0.53) and higher chemotherapy response rates at first- and second-line treatment. CCNE1 -gained ( CCNE1 g) cases demonstrated shorter survival (multivariable hazard ratio 1.52, 95% CI 1.10-2.10), under-representation of FIGO stage IV cases (P=0.017) and no significant difference in treatment response. We demonstrate marked overlap between the TCGA- and derived subtypes: the TCGA DIF, IMR, PRO and MES subtypes correlated with the Tothill C4, C2, C5 and C1 subtypes (P<0.001). IMR/C2 cases displayed higher BRCA1/2 m frequency (25.5% and 32.5%) and significantly greater infiltration of immune cells (P<0.001), while PRO/C5 cases had the highest CCNE1 g rate (23.9% and 22.2%) and were uniformly low in immune cell infiltration. The survival benefit for cases with aberrations in homologous recombination repair (HRR) genes was apparent across all transcriptomic subtypes (hazard ratio range 0.48-0.68). There was significant co-occurrence of RB loss and HRR gene aberrations (P=0.005); RB loss was further associated with favourable survival within cases harbouring HRR aberrations (multivariable hazard ratio 0.50, 95% CI 0.30-0.84). Conclusions These data paint a high resolution picture of the molecular landscape in HGSOC, better defining patients who may benefit most from specific molecular therapeutics and highlighting those for whom novel treatment strategies are needed to improve outcomes.