2997 - Vaginal Melanoma: A Systematic Review of the Literature

E. S. Hollis¹, S. Sud², S. Tewelde³, and E. Lou⁴; ¹University of North Carolina Hospitals, Chapel Hill, NC, ²Department of Radiation Oncology, University of North Carolina, Chapel Hill, NC, ³University of North Carolina at Chapel Hill, Chapel Hill, NC, ⁴North Carolina School of Science and Mathematics, Durham, NC

Purpose/Objective(s): Primary vaginal melanomas are rare malignant mucosal neoplasms for which clinical data and research to guide clinical care are limited. We aim to elucidate the clinical course and outcomes for this disease through a systematic review. Materials/

Methods: Case reports were identified through PubMed and screened using a priori criteria for inclusion per PRISMA guidelines. Initial query for vaginal melanoma identified 682 candidate studies. Following screening by 2 reviewers, 340 studies were excluded, and 47 studies were inaccessible for review. Of the remaining studies, 175 provided individual patient data for 526 patient cases. Clinical data was analyzed with descriptive statistics and overall survival was computed using Kaplan-Meier estimates.
Results: In our preliminary analysis we identified 526 cases with substantial variation in consistency of data elements reported. Median age at diagnosis was 62 years. Postmenopausal bleeding was the most common presenting symptom and the most common disease location was the anterior lower third of the vagina. Median tumor size (maximum dimension) was 3 cm and median depth of invasion was 8 mm. Immunohistochemistry stains most utilized and percent of cases in which these were positive were HMB45 (95%), S100 (97%), and Melan A (100%). Primary treatment was typically surgery (74% of patients), often with wide local excision. 30% of patients received radiotherapy (RT) either as definitive or adjuvant therapy, with 56 cases of external beam RT (EBRT), 23 cases of brachytherapy, and 21 cases of EBRT + brachytherapy reported. Analysis of patterns of RT delivery and association with outcome was limited due to missing data elements: only ~38% of EBRT and ~21% of brachytherapy cases provided both dose and fractionation schedule. In the limited subset that reported this information, the median EBRT dose was 48 Gy delivered via conventional fractionation and median brachytherapy prescription dose was 27 Gy delivered over 3-5 fractions. 29% of patients received systemic therapy with limited reporting of dose and number of cycles. Of the patients with treatment data reported, outcome information was only available for 68% of cases, of which 60% experienced disease recurrence. Median overall survival was 14 months (13.1 - 14.9, 95% confidence interval). Notably, there were challenges with data limitations including aggregate or missing information as well as inaccessible articles for our review.
Conclusion: Our analysis provides a broad scoping overview of vaginal melanoma treatment and outcomes. Surgery with local excision was the most common treatment modality. Overall survival is limited indicating need for optimization of treatment paradigms. Our identification of heterogeneity and missing data elements highlights the critical need for standardized reporting of individual patient data in the setting of rare tumors to facilitate high quality analyses and improvement of patient outcomes.