Background: Precise understanding of recurrence patterns permits efficient surveillance and effective treatment strategies. The aim of this study was to evaluate recurrence patterns after treatment of rectal cancers, specifically with respect to tumor location and chemoradiotherapy (CRT).
Methods: A single-institution, retrospective cohort of 2,086 consecutive rectal cancer patients, was enrolled between January 2000 and December 2007. All the patients underwent curative operations (R0). Tumor location was classified into lower (≤5 cm), middle (>5 to ≤8 cm), and upper (>8 cm) groups based on the distance of the inferior tumor border from the anal verge; the patients were also characterized according to whether they received preoperative/postoperative CRT.
Results: The lung was the most common recurrence site in the lower group (lower vs. middle/upper; 14.6 vs. 8.9%/8.0%, P = 0.001/0.001). Recurrence patterns were not associated with receipt of preoperative/postoperative CRT. Additionally, RT and CRT did not reduce the rate of pulmonary recurrence (no-RT/preoperative CRT/postoperative CRT, 37.5/37.9/42.6%; P = 0.13). In a multivariate analysis, preoperative level of serum carcinoembryonic antigen, abdominoperineal resection, advanced T category, N category, and circumferential resection margin were identified as independent risk factors for pulmonary recurrence in all groups. Otherwise, low rectal cancer was associated with unresectable pulmonary recurrence (RR = 2.19; 95% CI 1.012-3.072; P = 0.04).
Conclusions: Neither RT nor CRT affects the pattern and rate of recurrence. Tumor location specifically affects recurrence in rectal cancer patients, such that the lower group is a risk factor for unresectable pulmonary recurrences.