CircRNA signatures to identify candidate biomarkers of treatment resistance in BRCA1 -mutant ovarian cancer.

e17558 Background: Most patients with high grade serous ovarian cancer (HGSOC) ultimately relapse with chemotherapy resistant disease, and resistance to Poly (ADP ribose) polymerase inhibitors (PARPi) is increasingly recognised. The mechanisms underlying inherent and acquired resistance remain poorly understood, and no validated biomarkers exist to guide treatment beyond BRCA status. circularRNAs (circRNAs) are stable non-coding RNAs with regulatory roles in gene expression and may capture transcriptional reprogramming associated with treatment resistance. This study examined circRNA expression in chemotherapy and PARPi resistant ovarian cancer (OC) models to identify candidate biomarkers of resistance. Methods: Chemotherapy (carboplatin and paclitaxel) and PARPi resistant (olaparib) COV362 (BRCA1 mutant) sublines were generated using progressive dose escalation. Resistance was confirmed using CCK-8 and BrdU proliferation assays. This was a preclinical, exploratory biomarker discovery study. The primary endpoint was the identification of circRNAs differentially expressed between parental (drug-sensitive), chemotherapy-resistant and PARPi-resistant cell lines. circRNA expression profiling was performed using the Arraystar circRNA microarray platform. Differential expression was assessed using normalised signal intensity, fold-change thresholds and associated statistical outputs, with adjustment for multiple testing. Candidate circRNAs were prioritised based on magnitude of differential expression and annotation quality, and were validated using divergent primers and quantitative PCR. Results: Chemotherapy and PARPi resistant COV362 sublines were successfully established over 10 and 14 months, respectively. circRNA profiling revealed distinct segregation of parental (drug sensitive), chemotherapy and PARPi resistant cell lines using hierarchical clustering. Although global circRNA patterns remained broadly conserved, each model showed a distinct set of upregulated transcripts. At the individual transcript level, however, a subset of circRNAs demonstrated resistance-associated differences and were prioritised for further evaluation. hsa_circ_0085803 (p <0.05) was upregulated in the chemotherapy resistant model, whereas hsa_circ_0003258 (p <0.05) was predominately expressed in the PARPi resistant mode compared to matched parental cell line. Conclusions: Distinct circRNA signatures were associated with chemotherapy and PARPi resistance in COV362, with hsa_circ_0085803 and hsa_circ_0003258 identified as candidate biomarkers of resistance. These findings support the potential of circRNAs as indicators of treatment resistance in HGSOC.