Whole exome sequencing reveals diverse genomic relatedness between paired concurrent endometrial and ovarian carcinomas.

INTRODUCTION: Concurrent non-serous endometrial and ovarian tumours are often managed clinically as two separate primary tumours, but almost all exhibit evidence of a genomic relationship. METHODOLOGY: This study investigates the extent of relatedness using whole exome sequencing, which was performed on paired non-serous endometrial and ovarian carcinomas from 27 patients with concurrent tumours in a cohort with detailed clinicopathological annotation. Four whole exome sequencing-derived parameters (mutation, mutational burden, mutational signatures and mutant allele tumour heterogeneity scores) were used to develop a novel unsupervised model for the assessment of genomic similarity to infer genomic relatedness of paired tumours. RESULTS: This novel model demonstrated genomic relatedness across all four parameters in all tumour pairs. Mutations in PTEN, ARID1A, CTNNB1, KMT2D and PIK3CA occurred most frequently and 24 of 27 (89 %) tumour pairs shared identical mutations in at least one of these genes, with all pairs sharing mutations in a number of other genes. Ovarian endometriosis, CTNNB1 exon 3 mutation, and progression and death from disease were present across the similarity ranking. Mismatch repair deficiency was associated with less genomically similar pairs. DISCUSSION: Although there was diversity across the cohort, the presence of genomic similarity in all paired tumours supports the hypothesis that concurrent non-serous endometrial and ovarian carcinomas are genomically related and may have arisen from a common origin.

Compartment-specific multiomic profiling identifies SRC and GNAS as candidate drivers of epithelial-to-mesenchymal transition in ovarian carcinosarcoma.

BACKGROUND: Ovarian carcinosarcoma (OCS) is an exceptionally aggressive and understudied ovarian cancer type harbouring distinct carcinomatous and sarcomatous compartments. Here, we seek to identify shared and compartment-specific events that may represent potential therapeutic targets and candidate drivers of sarcomatous compartment formation through epithelial-to-mesenchymal transition (EMT). METHODS: We performed multiomic profiling (exome sequencing, RNA-sequencing, microRNA profiling) of paired carcinomatous and sarcomatous components in 12 OCS cases. RESULTS: While paired sarcomatous and carcinomatous compartments demonstrate substantial genomic similarities, multiple loci are recurrently copy number-altered between components; regions containing GNAS and SRC are recurrently gained within the sarcomatous compartment. CCNE1 gain is a common event in OCS, occurring more frequently than in high grade serous ovarian carcinoma (HGSOC). Transcriptomic analysis suggests increased MAPK activity and subtype switching toward poor prognosis HGSOC-derived transcriptomic subtypes within the sarcomatous component. The two compartments show global differences in microRNA profiles, with differentially expressed microRNAs targeting EMT-related genes (SIRT1, ZEB2) and regulators of pro-tumourigenic pathways (TGFß, NOTCH); chrX is a highly enriched target of these microRNAs and is also frequently deleted across samples. The sarcomatous component harbours significantly fewer CD8-positive cells, suggesting poorer immune engagement. CONCLUSION: CCNE1 gain and chrX loss are frequent in OCS. SRC gain, increased GNAS expression and microRNA dysregulation represent potential mechanisms driving sarcomatous compartment formation.