Value of ctDNA-based molecular residual disease (MRD) in evaluating the adjuvant therapy effect in ovarian cancer.

5525 Background: Ovarian carcinoma (OC) is the fifth leading cause of cancer-related mortality in women worldwide. Circulating tumor DNA (ctDNA) is a promising biomarker for ovarian cancer and can be detected using a noninvasive liquid biopsy. It reflects the status of molecular residual disease and post-surgery therapy effect in real-time. In this study, we examined the concordance of mutations detected in ctDNA samples with those identified in the corresponding tumor specimens of different types of OC. Additionally, investigated whether mutations in plasma ctDNA can provide predictive insight into OC and provide a new perspective on the efficacy of postoperative adjuvant therapy. Methods: Stage I-IV ovarian cancer patients were recruited from 2022.11 to 2023.12. Tumor tissues were collected at excision and then subjected to whole exome sequencing(WES) for comprehensive somatic variant detection in 1000x depth. Personalized tumor-informed molecular residual disease (MRD) panel was designed for up to 40 loci including SNVs and InDELS. Plasma samples were collected at baseline and post-operation adjuvant therapy. Plasma samples performed on ctDNA-based MRD assay underwent 100,000x ultra-deep sequencing. Results: The cohort included 20 patients with high-grade serous carcinomas (HGSC), 5 with borderline tumors, 2 with clear cell carcinoma, 2 with mucinous carcinomas, 2 with endometrioid carcinoma, 1 with low-grade serous carcinomas, and 2 with other types. WES identified 2542 variants in 1906 genes, encompassing 2297 single nucleotide variants (SNVs) and 245 insertions/deletions (InDels), with missense mutations accounting for 84.5% of the major mutations. The most frequently mutated gene was TP53 (76.47%), followed by ZNF285 (44.12%), TTN (29.41%), TOPAZ1 (29.41%), THADA (23.53%), and UBE4A (20.59%). The mutation landscape showed no significant differences among subtypes. The baseline ctDNA detection rate is 88.23% for the entire cohort and 90% for HGSC patients. The consistency between plasma and tissue was verified by Pearson (p=2.7e-6) and Spearman (p=3.7e-7) correlation analyses. All HGSC patients underwent adjuvant therapy post-surgery (up to 8 cycles). 75% (15/20) of HGSC patients achieved complete response (CR), 20% (4/20) achieved partial response (PR), and 5% (1/20) showed progressive disease (PD). MRD monitoring showed a continuous decrease in postoperative ctDNA concentration, with 60% (9/15) of patients achieving MRD negativity. Among the patients with CR, 4 became MRD negative for the first time after surgery and maintained negativity after up to 6 cycles of adjuvant therapy. Conclusions: The results of this real-world study indicate that tumor-informed MRD can serve as a prognostic biomarker for OC, and would provide information about postoperative MRD status and long-term monitoring of adjuvant therapy for further precise treatment.