Purpose: To assess the pelvic failure among patients with T3 rectal cancer treated with local excision after preoperative chemoradiation.
Methods and materials: Between January 1990 and June 2002, 431 patients with clinically staged T3 rectal cancer were treated with preoperative chemoradiation followed by surgical resection. Full-thickness local excision [Kraske (n = 3) or a transanal excision (n = 23)] was performed in 26 patients because of patient refusal of abdominoperineal resection (APR) (n = 13), medical comorbidity (n = 4), physician preference after a complete clinical response (n = 6), and other reasons (n = 3). All patients were treated with continuous-infusion 5-fluorouracil (5-FU) (300 mg/m(2) Monday to Friday) and concomitant pelvic radiation (45 Gy in 25 fractions with a 3-field belly board technique). Ten local-excision patients received a concomitant boost during the last week of therapy (1.5-Gy second daily fractions) for a total dose of 52.5 Gy. Similar preoperative treatment was followed by total mesorectal excision in 405 patients. Among the local-excision patients, the median tumor size was 3.5 cm (range, 0.5-7 cm). Well-differentiated or moderately-differentiated histology was present in all but 3 cases, and endoscopic ultrasound staging examination was performed in 25 of 26 patients. Based on CT findings, 1 patient was node positive. The median circumference involved by tumor was 33%, (20%-75%). The median distance from the anal verge was 3 cm (range, 1-8 cm).
Results: The mean follow-up was 46 months (range, 5-109 months) in the local-excision group. In the local-excision group, 19 of 26 patients had only residual scarring noted on digital rectal examination and rigid proctoscopy before surgery. Fourteen patients (54%) had a complete histologic response to chemoradiation, 9 patients (35%) had microscopic residual disease, and 3 patients (12%) had gross residual disease. Two intrapelvic recurrences occurred at 76 and 20 months among the 26 patients treated with local excision (6% 5-year actuarial pelvic recurrence rate). This rate compared with an 8% 5-year actuarial pelvic recurrence rate among T3 patients treated with mesorectal excision and a 6% pelvic recurrence rate in the subgroup of mesorectal-excision patients with a complete clinical response to preoperative chemoradiation. One additional local-excision patient recurred in an inguinal lymph node after local excision and subsequently died of metastatic disease. A total of 2 local-excision patients died of metastatic rectal cancer. Actuarial overall survival at 5 years was 86% in the local-excision group compared with 81% among mesorectal-excision patients (p = NS), and 85% in patients with a complete clinical response to chemoradiation followed by mesorectal excision by APR or LAR (p = NS).
Conclusions: In an experience stimulated by patient refusal of APR, highly selected patients who responded well to conventional external-beam radiotherapy (CXRT) were selected to undergo local excision. Most of these patients had pathologic complete response. Local control and survival rates are comparable to those achieved with chemoradiation followed by mesorectal excision. This strategy should be prospectively studied in a group of patients with low rectal cancer who have no clinical evidence of tumor after chemoradiation.