Prospective validation of dynamic circulating tumor DNA (ctDNA)–based MRD for predicting progression-free survival and treatment benefit in oral cancer.

e18090 Background: Effective disease control in oral cancer heavily relies on the accurate identification and management of minimal residual disease (MRD). This study investigated the association of ctDNA-based MRD with progression-free survival (PFS), and its potential to predict treatment benefits. Patients and Methods: A prospective cohort study was conducted at Shanghai Ninth People’s Hospital, a tertiary care center involving patients with biopsy-proven squamous cell carcinoma of oral cavity reported between January 2021 and December 2023. Tumor tissues and plasma samples were collected, processed, and sequenced for ctDNA analysis using a 769-gene panel. A tumor-informed strategy, MinerVa (Genecast Biotechnology), was utilized for variants tracking and MRD assessment. ctDNA status was recorded at various time points, including pre-operative, post-operative, and follow-up visits. The primary endpoint of the study was progression-free survival (PFS). Results: As of November 2024, 46 patients were included in the analysis, with a median follow-up period of 297 days. All patients were HPV negative. Eleven (24%) patients experienced disease recurrence, of whom, 3 patients died. Out of 44 patients with post-operative MRD analyzed, 10 (23%) patients were MRD positive. MRD positivity significantly predicted poorer PFS, with a hazard ratio (HR) of 6.07 (95%CI=1.60-22.8, p=0.002) and a negative predictive value (NPV) of 85%. Both MRD and pathological N-stage (pN-stage) were significant predictors of PFS (p=0.002, and p=0.011, respectively). In multivariate analysis, MRD positivity remained an independent predictor of PFS (HR=8.50, p=0.024) even when adjusted for tumor size, pN-stage, and adjuvant treatment regimen. Notably, among post-operative MRD-positive patients, those receiving adjuvant therapy demonstrated significantly better PFS compared to those who did not (HR=5.68, p=0.035). Conversely, adjuvant therapy showed no significant effect in MRD-negative patients (p=0.656). Furthermore, longitudinal analysis of ctDNA in plasma samples collected after surgery but before relapse revealed a positivity rate of 28% (13/46), with the NPV increasing up to 91%. Conclusions : Post-operative ctDNA-based MRD shows promise for identifying disease recurrences and stratifying patients likely to benefit from adjuvant treatment. Continuous MRD assessment during and after adjuvant therapy plays a critical role in disease monitoring and potentially improving clinical outcomes for oral cancers patients. Clinical trial information: NCT05156970 .