Sensitivity of circulating tumor DNA (ctDNA) for disease recurrence or relapse in melanoma patients.

9584 Background: ctDNA has emerged as a biomarker for minimal residual disease (MRD) detection in colorectal cancer and other cancer types. Its role in MRD monitoring and disease detection for melanoma patients merits further investigation. We aimed to assess the sensitivity of a commercially available ctDNA test to detect recurrence or relapse in patients with melanoma. Methods: A retrospective cohort analysis was performed using a personalized, tumor-informed, ctDNA assay (Natera) on prospectively collected plasma from patients with a diagnosis of melanoma from December 2021 to January 2025 with longitudinal follow up. Inclusion criteria were patients with no evidence of disease (NED) following definitive surgery or clinical remission with systemic therapy; and an undetectable ctDNA level. We evaluated the sensitivity of ctDNA detection of patients who had biopsy and/or radiographic confirmation of recurrence or relapse; and further stratified this based on anatomic sites of recurrence or metastasis (mets). Sensitivity was computed and logistic regression models were conducted to assess predictors of ctDNA detection. Results: 116 patients met the inclusion criteria. 48% (n=56) had confirmed recurrence or relapse. Among these patients, 82% (n=46) had NED following surgery and 18% (n=10) had clinical remission with systemic therapy. 75% (n=42) were resected stage II/III, 7% (n=4) were resected stage IV, 4% (n=2) were unresectable stage III in remission, and 14% (n=8) were unresectable stage IV in remission. Melanoma primaries included: 82% (n=46) cutaneous, 7% (n=4) mucosal, 5% (n=3) uveal, 5% (n=3) unknown. ctDNA was detected in 30 out of 56 patients with confirmed recurrence/relapse, with an overall sensitivity of 53.6%. Sensitivity varied by site of recurrence/relapse. Sensitivity for ctDNA detection of brain mets was 33.3% (4/12); skin/mucosa/muscle recurrence or mets was 37.5% (6/16); lung mets was 61.5% (8/13); bone mets was 75% (3/4); liver mets was 87.5% (7/8), and lymph node mets was 87.5% (14/16). The odds of detecting ctDNA at relapse or recurrence were significant amongst patients with lymph node mets (OR=10.5, 95% CI 2.10-52.58, p=0.004) and patients with multiple anatomic sites (2+) of mets (OR=7.61, 95% CI 1.27-66.67, p=0.032). Conclusions: ctDNA monitoring shows potential in detecting recurrence or relapse in patients with melanoma. We found that location of metastatic disease impacts the shedding of ctDNA and its detection of macroscopic disease. Larger studies are needed to refine the role of ctDNA monitoring and its clinical application in melanoma surveillance.