Multimodality salvage of recurrent disease after local excision for rectal cancer

BACKGROUND: Local excision, alone or in combination with chemoradiation, is increasingly considered for rectal cancer. Higher risks of disease recurrence have been demonstrated after local excision. OBJECTIVE: The aim of this study was to examine the outcomes of current-era multimodality salvage for recurrent rectal cancer after local excision. DESIGN: This was a single-institutional retrospective study. SETTINGS: This study was conducted at a tertiary-referral cancer center between 1993 and 2011. PATIENTS: Forty-six patients with recurrent rectal cancer after initial local excision were included. INTERVENTION: Multimodality salvage treatment was performed as appropriate. MAIN OUTCOME MEASURES: The primary outcomes measured were the pattern of disease recurrence, salvage treatments, and resultant overall and re-recurrence-free survival. RESULTS: After the initial local excision, recurrent disease was diagnosed after a median interval of 1.9 years: local/regionally in 67%, distantly in 18%, and both in 15%. Four patients (9%) had recurrence that was unsalvageable, 2 (4%) declined treatment, and 40 (87%) underwent surgical salvage. Preoperative chemoradiation was given in 30 (75%) patients. The R0 resection rate was 80%, requiring multivisceral resection (33%), total pelvic exenteration (5%), and metastasectomy (25%). The rate of sphincter preservation was 33%, and perioperative morbidity was 50%. The first site of failure after salvage was distant in 38% and was local only in 10%. The 5-year overall and 3-year re-recurrence-free survival were 63% and 43%. Pathologic stage at initial local excision, receipt of neoadjuvant chemoradiation before local excision, recurrence pattern after local excision, pathologic stage at salvage, and R0 resection at salvage influenced rerecurrence-free survival. LIMITATIONS: This study was limited by the referral and selection biases inherent in a small study cohort. CONCLUSIONS: Failure after local excision for rectal cancer may not be salvageable. When feasible, multimodality treatment, including multivisceral resection, pelvic irradiation, and chemotherapy, was associated with potentially lasting treatment-related morbidities and only modest success in long-term disease control. These findings should be compared with the expected stage-specific outcomes of standard proctectomy for early-stage rectal cancer, when local excision is being considered.