Anal sphincter conservation for patients with adenocarcinoma of the distal rectum: long-term results of radiation therapy oncology group protocol 89-02

Purpose: To assess the outcome of a multi-institutional, national cooperative group study attempting functional preservation of the anorectum for patients with limited, distal rectal cancer. Methods and Materials: Between September 21, 1989 and November 1, 1992, a Phase II trial of sphincter-sparing therapy was conducted for patients with clinically mobile rectal cancers located below the pelvic peritoneal reflection. Protocol treatment was designed for patients who were, in the judgement of their attending surgeon, unsuitable for anal sphincter conservation in the context of anterior resection, and would have required abdominoperineal resection (APR) as conventional surgical therapy. Primary cancers were estimated to be 4 cm or less in largest clinical diameter, and occupied 40% or less of the rectal circumference. Chest radiography and computerized axial tomography (CT) of the abdomen and pelvis excluded patients with overt lymphatic or hematogenous metastases. Protocol surgery was intended to remove the primary cancer by en-bloc, transmural excision of an ellipse of rectal wall by transanal, transcoccygeal, or trans-sacral technique, while conserving the anal sphincter. Based on tumor size, T classification, grade, and adequacy of surgical margins, patients were allocated to one of three treatment assignments: observation, or adjuvant treatment with 5-fluorouracil (5-FU) and one of two different dose levels of local-regional radiation. After completion of protocol therapy, patients were observed with follow-up that included periodic general physical and rectal examination, determinations of CEA, abdominopelvic CT, chest radiography, and surveillance endoscopy. Sixty-five eligible and analyzable patients were registered. Results: With minimum follow-up of 5 years and median follow-up of 6.1 years, 11 patients have failed: 3 patients recurred local-regionally only, 3 patients had distant failure alone, and 5 patients manifested local-regional and distant failure. Eight patients died of intercurrent illness. Local-regional failure correlated with T-category revealed: T 1 1/27 (4%), T 2 4/25 (16%), and T 3 3/13 (23%). Local-regional failure escalated with percentage involvement of the rectal circumference: 2/31 (6%) among patients with cancers involving 20% or less of the rectal circumference, and 6/34 (18%) among patients with cancers involving 21–40% of the circumference. Distant dissemination rose with T-category with 1/27 (4%) T 1, 3/25 (12%) T 2, and 4