Abstract 731: Integrating ctDNA and FDG-PET dynamics to predict pathological response following neoadjuvant systemic therapy in stage III melanoma

Background: Neoadjuvant therapy followed by surgery is the current standard for stage III melanoma, but 60% of patients achieve a major pathological response (MPR) with good long-term outcomes. This raises the potential for modifying treatment strategies to identify patients who may safely omit surgery. Biomarkers such as circulating tumor DNA (ctDNA) and imaging modalities have shown promise in assessing therapeutic response, yet each has limitations when used independently. This study evaluates the relationship between ctDNA clearance and changes in fluorodeoxyglucose(FDG)-positron emission tomography(PET) SUVmax from pre-treatment to pre-surgery, to predict pathological response and the feasibility of non-surgical management in appropriately selected patients. Methods: In the discovery cohort, plasma ctDNA levels and computed tomography (CT) RECIST data were retrospectively analyzed in stage III melanoma patients from the OpACIN-neo trial (N=9), where two doses of neoadjuvant combination immunotherapy was administered. For validation, FDG-PET SUVmax changes and ctDNA dynamics will be assessed in two trial cohorts: NeoTrio (N=45, pembrolizumab ± dabrafenib and trametinib) and NeoPele (N=20, pembrolizumab and Lenvatinib). PET responses will be categorized into complete metabolic response (CMR), near-CMR (SUVmax reduction >90%), partial metabolic response (PMR), stable metabolic disease (SMD), and progressive metabolic disease (PMD). Results: In the OpACIN-neo cohort, 7/9 (78%) patients achieved ctDNA clearance pre-surgery and RECIST partial response, with five achieving MPR and two non-MPR. Two patients with persistent ctDNA pre-surgery, which cleared post-surgery, had RECIST stable disease (one achieved MPR and one non-MPR). None of these nine patients experienced clinical recurrence. Preliminary data from the validation cohort shows that all eight patients (8/33; 24%) with CMR or near CMR achieved MPR, while 5/6 (83%) patients with PMD were pathological non-responders. CT RECIST was useful in evaluating disease burden in the discovery cohort; however, PET imaging is expected to provide greater specificity and stronger correlation with pathological outcomes. Furthermore, ctDNA clearance has demonstrated potential in identifying patients who may avoid surgery and remain recurrence-free, highlighting the complementary value of both modalities. Conclusion: Combining ctDNA dynamics with FDG-PET SUVmax changes can improve the prediction of pathological response to neoadjuvant therapy in stage III melanoma. This integrated approach could help identify patients who will achieve durable responses and can safely avoid surgery, supporting the development of more personalized treatment strategies. Citation Format: Wei Yen Chan, Li Zhou, Edward Hsiao, Jenny H. Lee, Ashleigh Stewart, Russell J. Diefenbach, George Au-Yeung, Maria Gonzalez, Andrew J. Spillane, Robyn P. Saw, Sydney Ch'Ng, Matteo Carlino, Richard A. Scolyer, Alexander M. Menzies, Georgina V. Long, Helen Rizos. Integrating ctDNA and FDG-PET dynamics to predict pathological response following neoadjuvant systemic therapy in stage III melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 731.