Postoperative surveillance in patients with colorectal cancer who have undergone curative resection: a prospective, multicenter, randomized, controlled trial

Abstract

Purpose: Although systematic postoperative surveillance of patients with colorectal cancer has been demonstrated to improve survival, it remains unknown whether a more intensive strategy provides any significant advantage. This prospective, multicenter, randomized, controlled trial was aimed at comparing the efficacy of two different surveillance strategies in terms of both survival and recurrence resectability.

Patients and methods: Patients with stage II or III colorectal cancer were allocated randomly to either a simple surveillance strategy including clinical evaluation and serum carcinoembryonic antigen monitoring, or an intensive strategy in which abdominal computed tomography or ultrasonography, chest radiograph, and colonoscopy were added.

Results: A total of 259 patients were included: 132 were observed according to the simple strategy and 127 were observed according to the intensive strategy. Both groups were similar with respect to baseline characteristics and rate and type of tumor recurrence. After a median follow-up of 48 months, there was no difference in the probability of overall survival in the whole series (hazard ratio [HR] = 0.87; 95% CI, 0.49 to 1.54; P = .62). However, the intensive strategy was associated with higher overall survival in patients with stage II tumors (HR = 0.34; 95% CI, 0.12 to 0.98; P = .045) and in those with rectal lesions (HR = 0.09; 95% CI, 0.01 to 0.81; P = .03), mainly due to higher rate of resectability for recurrent tumors. Colonoscopy was responsible for the detection of the highest proportion (44%) of resectable tumor recurrence in the intensive arm.

Conclusion: A more intensive surveillance strategy improves the prognosis of patients with stage II colorectal cancer or those with rectal tumors. Inclusion of regular performance of colonoscopy seems justified up to the fifth year of follow-up, at least.