To evaluate the outcome and toxicity of a sphincter-sparing treatment strategy in the management of patients with anal-rectal melanoma. Between 1989 and 2000, 23 patients with invasive anal-rectal melanoma were managed with sphincter-sparing surgical resection and adjuvant radiation. Surgery consisted of primary local excision, as well as dissection for patients with documented regional nodal disease. Adjuvant radiation was delivered using a hypofractionated regimen of 30 Gy in five fractions over 2.5 weeks. Adjuvant systemic therapy was delivered to nine patients: cytotoxic chemotherapy in seven and immunotherapy in two. After a median follow-up of 32 months, 15 patients had relapsed and 15 patients had died. The actuarial 5-year overall, disease-specific, disease-free, and distant metastasis-free survival rates were 31%, 36%, 37%, and 35%, respectively. The actuarial 5-year local and regional nodal control rates were 74% and 84%, respectively. No patient had locoregional failure as the sole site of failure and no patient required salvage abdominoperineal resection (APR). By univariate analysis, patients with nodal disease at presentation had a decreased actuarial 5-year disease-specific (0% v 45%, P =.004), disease-free (0% v 45%, P <.001), and distant metastasis-free survival (0% v 42%, P <.001). The actuarial complication-free survival rate was 71%. Two patients developed mild scrotal edema (grade 1), and four patients developed moderate proctitis requiring prolonged medical management (grade 2). Sphincter-sparing local excision and adjuvant radiation is well tolerated and can effectively control local-regional disease while avoiding the functional morbidity of APR.