Prognostic Heterogeneity of Endosonographic T3 Rectal Cancer

PURPOSE: This study aimed to assess the prognostic implications of uT3 rectal carcinomas according to the tumor thickness and to analyze the correlation between this ultrasound-based parameter and other prognostic factors. METHODS: Seventy-four patients with uT3(pM0) rectal tumors underwent primary surgery from 1996 to 2003. Preoperative endorectal ultrasound was used to assess uN stage, maximum tumor perimeter, and maximum tumor thickness. An ultrasound maximum tumor thickness cutoff point for local recurrence subdividing T3 tumors into uT3a and uT3b was established. RESULTS: Median follow-up was 41 months (range, 24-59). The 5-year actuarial local and overall recurrence rates were 9.82 percent (n = 7) and 42.46 percent (n = 23), respectively. uN stage(P = 0.05), circumferential resection margin involvement (P = 0.002), an ultrasound maximum tumor thickness (P = 0.01), and locally advanced tumors (P = 0.001) were related to a significantly increased risk of local recurrence. An ultrasound maximum tumor thickness (hazard ratio, 1.15; 95 percent confidence interval, 1.0-1.2) and locally advanced tumor (hazard ratio, 17.21; 95 percent confidence interval, 2.99-98.84) were preoperative independent variables for predicting local recurrence. Locally advanced tumor was the only preoperative independent prognostic factor for overall recurrence (P = 0.004; hazard ratio, 1.09; 95 percent confidence interval, 1.0-1.1). An ultrasound maximum tumor thickness with a 19-mm cutoff point, subdividing the T3 tumors into uT3a and uT3b, can be used to predict local recurrence. Locally advanced tumors (P = 0.02) and circumferential resection margin involvement (P = 0.005) showed a significant association with an ultrasound maximum tumor thickness >19mm. CONCLUSIONS: A maximum tumor thickness measured by endorectal ultrasound in pT3 rectal cancer is an independent prognostic factor for local and overall recurrence. An ultrasound maximum tumor thickness cutoff point of 19 mm may be useful to classify patients preoperatively and to select them for primary surgery or neoadjuvant therapy.