Follow-up of colorectal cancer resected for cure

Abstract

Sixty-four consecutive patients who had undergone curative resection for colorectal carcinoma were studied prospectively to evaluate the roles of sequential CEA determinations and independent instrumental follow-up in the early detection of resectable recurrences. Fifty-two of these patients also were submitted to sequential determinations of other tumor antigens: TPA (tissue polypeptide antigen) and Ca 19-9 (colon cancer antigen detected with a monoclonal antibody), for a retrospective evaluation of their utility as markers of recurrent tumors. Twenty-two recurrences were detected in a period ranging from 12 to 72 months (median, 47 months). CEA was the best predictor of recurrence (sensitivity, 90 percent) when compared with the other two markers (TPA sensitivity, 60 percent; Ca 19-9 sensitivity, 20 percent). When compared with the instrumental or biochemical examinations of the follow-up, CEA was still the most sensitive indicator of relapse although the specificity was quite low (78 percent) if minimal significative increases were considered. History and physical examination were more useful than CEA in detecting local recurrences in rectal cancer where the preoperative CEA level was low. A few second-look explorations based solely on small CEA increases failed to demonstrate recurrence or revealed peritoneal carcinomatosis. Selected second-look surgery based on demonstrated recurrences resulted in a resectability rate of 57 percent. A follow-up program based on frequent CEA assays, history, and physical examinations, including rectal, vaginal, and perineal exploration, is proposed. Extensive instrumental investigations should follow when a minimal significative CEA rise is observed, or when history and physical examinations suggest a possible recurrence. Second-look surgery should be evaluated after confirmed or highly suspected diagnosis of recurrence, on the basis of instrumental or clinical examinations.