Negative influence of delayed surgery on survival after preoperative radiotherapy in rectal cancer.

ABSTRACT

OBJECTIVE: In Europe, until recently the standard treatment for locally advanced rectal cancer was preoperative radiotherapy (RT). The objective of this study was to evaluate the influence on survival of intervals between diagnosis and treatment. PATIENTS AND METHODS: The influence on survival of intervals between diagnosis and surgery (Dg-Surg), diagnosis and initiation of RT (Dg-Rad), and completion of RT and surgery (Rad-Surg) was evaluated in a retrospective series of patients treated with preoperative RT. Between 1991 and 1998, 102 patients received treatment with preoperative RT without concomitant chemotherapy at the René Gauducheau Cancer Center. Patients generally received 45 Gy (80%) in 25 fractions over 35 days for T2-T3-T4 N0-N1 M0 rectal adenocarcinoma located mainly (62.7%) in the lower third of the rectum (< or = 5 cm from anal margin). Thirty-five pN1 patients were treated with postoperative chemotherapy. Differences between survival were assessed by the log-rank test, and prognostic factors by the Cox test. RESULTS: Median time was 14.7 weeks for Dg-Surg, 4.6 weeks for Dg-Rad and 5.1 weeks for Rad-Surg. Median follow-up from diagnosis was 57.4 months. Five-year local relapse-free survival was 83.9%, metastasis-free survival 64% and overall survival 60.8%. No factor was predictive of tumour response to RT. Log-rank and multivariate analysis showed that overall survival was significantly influenced by lower-third tumours, pT, pN and Dg-Surg (poorer survival when > or = 16 weeks OR = 2.59, P = 0.005). Metastasis-free survival correlated significantly with Dg-Surg (> or = 16 weeks OR = 2.05, P = 0.05). CONCLUSION: An interval of more than 16 weeks between diagnosis and surgery may reduce overall survival of patients treated with preoperative RT for locally advanced rectal cancer. Surgery should be performed shortly after completion of RT for patients with no possibility of sphincter preservation, or a minimal risk of morbidity from an abdominoperineal excision.