Clinical implications of quantification of mesorectal tumor invasion by endoscopic ultrasound: All T3 rectal cancers are not equal

BACKGROUND: Depth of invasion beyond the muscularis propria (MP) by T3 rectal cancer can vary. The purpose of the present paper was to determine if depth of invasion beyond MP, as assessed by preoperative endoscopic ultrasound (EUS), can predict tumor recurrence in patients with T3 rectal tumors. METHODS: Patients with T3NxM0 rectal cancer, as determined by EUS, who underwent surgical resection (without preoperative neoadjuvant therapy) were reviewed by two blinded endosonographers. Tumors were classified as minimally invasive T3 (invasion 2 mm). RESULTS: Forty-two patients with T3 rectal tumors underwent surgical resection without receiving preoperative neoadjuvant therapy, of whom 14 had minimally invasive T3 and 28 had advanced T3 disease, as determined by preoperative EUS. Median follow up was 19 months. Tumor recurrence rates in minimally invasive and advanced T3 tumors were 14.3% and 39.3%, respectively, P = 0.02 (log-rank test). Adjusting for nodal status and postoperative adjuvant therapy administration, Cox proportional hazards model demonstrated advanced T3 disease (by EUS) to predict tumor recurrence, hazard ratio, 2.28 (95% confidence interval: 1.17-5.81), P = 0.01. CONCLUSIONS: All T3 rectal tumors are not equal, with minimally invasive disease carrying a more favorable prognosis. By discriminating minimally invasive from advanced T3 disease, preoperative EUS provides important prognostic information. Further subclassification of T3 tumors, based on preoperative EUS staging, should be considered to enhance selection of patients for neoadjuvant therapy.