SAN FRANCISCO – Patients with stage I colorectal cancer should be followed as closely as patients with higher-stage primary tumors after resection, according to a prospective 6-year surveillance study of 1,202 British patients.
The reason is that the incidence of recurrences that can be treated surgically with curative intent is the same in stage I patients as it is in patients with stage II and III primaries, about 6% (J. Am. Coll. Surg. 2014:219;e46-47).
Following resection with clear margins, the patients were randomized about 300 per group to either serial CT surveillance, serial carcinoembryonic antigen (CEA) monitoring, both, or minimum surveillance, and followed for a median of 4.4 years. About a quarter of the subjects had Dukes’ A primaries and a quarter Dukes’ C primaries, and about half had Dukes’ B primaries. The A, B, C designations are similar to the stage I, II, and III designations more common in the United States.
On both sides of the Atlantic, guidelines focus on active surveillance for higher-stage primaries, but are ambivalent about monitoring stage I tumors because they are less likely to recur and the benefit of follow-up has been uncertain.
Dr. Sian Pugh
That needs to change because treatable recurrences are what matters, and they are as likely in low-stage disease as in high-stage disease, lead investigator Dr. Sian Pugh, a colorectal surgeon at the University of Southampton (England), said at the annual clinical congress of the American College of Surgeons.
“Picking up recurrences that are not treatable doesn’t help anyone; it just gives you bad news earlier,” she said. The benefit of follow-up is “finding treatable disease, and that’s independent of the stage of the primary tumor. We recommend equivalent follow-up for all patients with resected Dukes’ A-C colorectal cancer. The guidelines [should] be reconsidered,” she asserted.
Her team also thinks that the most cost-effective way to monitor patients is probably with CEA monitoring every 3 months for the first 2 years, CEA monitoring every 6 months thereafter until year 5, and a single CT at 12-18 months to catch recurrences that don’t express CEA. That strategy was three times more likely than minimal surveillance to find treatable recurrences, about the same as serial CTs and serial CTs with regular CEA monitoring.
“We don’t think patients need to be followed up quite as intensively as the guidelines suggest,” – for instance, CT scans every 6 months – “but we are waiting for more health economic analysis,” Dr. Pugh said.
Although treatable recurrences were equally likely in all the groups, they were more common in patients with lower-stage primary cancers. Among recurrences in patients with Dukes’ A primary tumors, 50% (13/26) were treated surgically with curative intent, compared with 40% (32/81) in patients with Dukes’ B primaries and 24% (20/82) in those with Dukes’ C.
Pulmonary recurrence was most frequent with rectal primary tumors, and multisite recurrence was most common from right-colonic cancers. Median survival following recurrence was 2.28 years and was influenced by stage and site of primary. Following recurrence, survival was highest in those with lower-stage and rectal primaries, and lowest in patients with higher-stage primaries and recurrence from the right colon.
Dr. Pugh had no disclosures. The work was funded by the U.K. National Health Service.