Surgical Outcomes of Vaginal or Cervical Melanoma

Objective: To evaluate the effectiveness of radical resection compared with non-radical resection for vaginal or cervical melanoma. Methods: We retrospectively analysed the clinical data of post-operative patients with primary lower genital tract melanoma hospitalised at Peking University Cancer Hospital between Jan 2014 and Dec 2020. The study endpoints were recurrence-free survival (RFS) and overall survival (OS). Kaplan-Meier method-plotted survival curves and univariate and multivariate Cox proportional hazards regression models were used to identify the factors associated with RFS and OS, and to calculate hazard ratios (HRs) and associated 95% confidence intervals (95% CIs). Results: A total of 80 patients were included. Thirty-one patients had received non-radical resection, and 49 patients had received radical resection. The median patient age was 55.5 (IQR 45.3-60.0) years. Sixty-two (77.5%) patients had vaginal melanoma. Sixty-four patients (80.0%) had received post-operative adjuvant therapy. The median follow-up time was 36.0 months (95% CI 10.1-62.1 months). Sixty-four patients developed recurrence, and 44 patients died. The median RFS (mRFS) was 6.0 months (95% CI 3.4-8.6 m), and the RFS for the radical resection group was longer than that for the non-radical resection group (9.5 vs. 5.3 m), with no significant difference (P > 0.05). The median OS (mOS) was 25.9 months (95% CI 14.4-37.4 m). The mOS was 24.6 months (95% CI 10.3-38.9 m) and 25.9 months (95% CI 10.9-40.9 m) in the non-radical resection group and the radical resection group, respectively. Multivariate Cox regression analysis showed that surgical approach, infiltration depth of the tumour, lymph node metastasis, and post-operative adjuvant therapy were independent risk factors for RFS and that post-operative adjuvant therapy was an independent risk factor for OS. Conclusion: By performing multivariate analysis, which corrected for potential confounding factors, we identified surgical procedures that were associated with RFS, and we found that RFS and OS in patients with vaginal melanoma and cervical melanoma benefitted from post-operative adjuvant therapy.