Abstract
BACKGROUND: Adjuvant therapies are now considered standard care for high-risk melanoma patients. It is urgent to identify reliable biomarkers to select patients at high risk of relapse, who will derive more benefit from adjuvant therapy. METHODS: Targeted next-generation sequencing using a 437 cancer-related gene panel was performed on primary tumor samples from 55 patients in stage I-III melanoma and postsurgical plasma samples from 46 patients. Association analysis of clinico-genomic characteristics with disease-free survival was conducted. RESULTS: The median DFS was 39.2 months (rangement: 1.4-49.7 months). Patients receiving anti-PD-1 adjuvant therapy had a longer DFS than those receiving adjuvant interferon therapy (mDFS, NA vs. 21.3 months, HR 0.36 [95% CI: 0.13-1.03], P = 0.047). No significant correlation of DFS with driver mutations in BRAF, NRAS and KIT was observed. Chromosomal instability score (CIS) was identified as an independent predictive factor of DFS following adjustment for clinical and genetic factors (HR 4.06 [95% CI: 1.28-12.89], P = 0.017), with an inferior DFS observed in patients with high CIS compared with the CIS-low patients (mDFS, 14.3 vs. 49.7 months, HR 3.90 [95%CI: 1.40-11.00], P = 0.005). In addition, patients with a maxVAF > 1% in the postsurgical ctDNA had a worse DFS compared with those with a maxVAF ≤1% (mDFS, 11.0 vs. 49.7 months, HR 3.70 [95% CI: 1.20-11.00], P = 0.012). Finally, CIS and postsurgical ctDNA status were considered as independent factors for recurrence risk in stage I-III melanoma. CONCLUSION: Identification of clinical and genetic biomarkers for recurrence risk prediction in resected melanoma may aid in clinical decision-making for early disease monitoring and optimal design of therapeutic strategies.