Extent of mesorectal invasion is a prognostic indicator in T3 rectal carcinoma.

Abstract

The aim of this study was to determine if local recurrence (LR) rates in patients with minimally invasive and advanced T3 rectal cancer are different. This may influence the use of adjuvant therapy. Consecutive patients with T3 rectal cancer undergoing curative surgery were classified into minimally invasive or advanced groups. Minimally invasive T3 was defined as a tumour that had invaded beyond the muscularis propria on microscopic examination only, whereas advanced T3 tumours had invasion beyond the muscularis propria that was obvious on macroscopic examination and confirmed histologically. Local recurrence rates of the two groups were compared by construction of Kaplan-Meier curves. The log-rank test was used to determine equivalence, and Cox regression to estimate the hazard ratio. The Grambsch- Therneau test and graphical comparison of predicted and observed Kaplan-Meier curves was used to test the proportional hazards assumption. There were 222 patients in total, 74 in the minimally invasive group and 148 in the advanced. The overall LR rate was 11.2%. The LR rates in the minimally invasive and advanced groups were 5.4% and 14.2%, respectively. The log-rank test gives a P value of 0.042 for equivalence, with the minimally invasive patients doing significantly better. The hazard ratio estimated by Cox regression was 0.35 (early relative to advanced), 95% confidence intervals (0.12, 1.0). There was no evidence of confounding by age at surgery, pathology type, gender or postoperative adjuvant therapy. The extent of invasion into the mesorectum appears to be an independent prognostic variable. If oncologically sound surgical techniques are employed, the LR rate of patients with minimal invasion is low. Adjuvant therapy may not confer additional benefit in this group.