Abstract
OBJECTIVE: We sought to evaluate the effectiveness and safety of utilizing radiotherapy (RT) with standard fractionation, with or without intraoperative RT (IORT), to treat locally recurrent rectal cancer (LRRC). METHODS: Retrospective review of 25 patients with LRRC treated with standard fractionation RT from 2005 to 2011. 15 patients (60%) had prior pelvic RT and 10 (40%) had synchronous metastases. The median equivalent dose in 2-Gy fractions was 30 and 49.6 Gy in patients with and without prior RT, respectively. 23 patients (92%) received concurrent chemotherapy and 16 (64%) underwent surgical resection. Eight patients (33.3%, four with and four without prior RT) received IORT. A competing risks model was developed to estimate the cumulative incidence of local failure with death treated as a competing event. RESULTS: Median follow-up was 36.9 months after the date of local recurrence. 3-year rates of overall survival (OS), local control (LC) and death with LC were 51.6%, 73.3% and 69.2%, respectively. On multivariable analysis, surgical resection was significantly predictive of improved OS (p < 0.05). If surgical resection were removed from the multivariable model, given the collinearity between IORT delivery and surgical resection, then IORT also became a significant predictor of OS (p < 0.05). Systemic disease at the time of local recurrence was not associated with either LC or OS. No patient had grade ≥3 acute or late toxicity. CONCLUSION: RT with standard fractionation is safe and effective in the treatment of patients with LRRC, even in patients with significant risk of systemic disease and/or history of prior RT. Advances in knowledge: The utility of RT with standard fractionation, generally with chemotherapy, in the treatment of LRRC is demonstrated. In this high-risk cohort of patients with a 40% incidence of synchronous metastatic disease, surgical resection of the recurrence was the major predictor of OS, though a benefit to IORT was also suggested. No patients had grade ≥3 acute or late toxicity, though 40% had undergone prior RT, underscoring the tolerability of standard fractionation RT in this setting.