ANORECTAL MELANOMA - A 64-YEAR EXPERIENCE AT MEMORIAL-SLOAN-KETTERING-CANCER-CENTER

PURPOSE: Operative management of patients with anorectal melanoma is controversial. To formulate a rational approach to patients with this disease, we reviewed our experience from 1929 to 1993. METHODS: Records of all patients treated at our center with anorectal melanoma from 1929 to the present were reviewed. survival analyses were graphically displayed using the Kapian-Meier product-limit method, and distributions were compared using the log-rank test. Fisher's exact test was used to compare groups with small sample sizes. RESULTS: Survival for the entire group (n = 85) was poor, 17 percent at 5 years (median, 19 months). Among the 71 patients with resectable disease, the five-year, disease-free survival distribution of patients who underwent abdominoperineal resection (APR) was more favorable than that of patients who underwent local procedures only, although this was not statistically significant (27 percent vs. 5 percent, APR vs. local procedures, respectively; P = 0.11). However, those who had an APR were more likely to survive long term than those who did not (P < 0.05). ALL ten long-term survivors were women. Nine had undergone APR, and one had a wide local excision. Of the nine survivors following APR, eight had negative and one had positive mesenteric nodes. Median size of the primary tumor in survivors following APR was 2.5 cm, compared with 4.0 cm for patients who did not survive long term following APR. CONCLUSIONS: APR should be considered in patients with localized anorectal melanoma, particularly those with smaller tumors and no evidence of nodal metastases.