NEW YORK (Reuters Health) – When colonoscopy is not possible, combining sigmoidoscopy and a fecal immunochemical test (FIT) would be superior for colorectal cancer (CRC) screening than either method alone, Japanese researchers say.

“The most sensitive screening option for detecting CRC is colonoscopy,” lead author Dr. Jun Kato acknowledged in email to Reuters Health. But he noted that access to colonoscopy is often limited by financial and manpower restrictions, whereas sigmoidoscopy and FIT are more generally available. “I considered that the combination of these tests may maximize the detectability of CRC and, consequently, benefit populations with lower medical resources.”

Dr. Kato, from the Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, and his associates performed a 1-sample FIT on 21,794 asymptomatic patients who also had colonoscopy the same day. None of the patients actually underwent sigmoidoscopy. Instead, the findings on colonoscopy were taken to “theoretically” represent what would have been found with sigmoidoscopy.

The authors’ definition of neoplasia — adenomas 10 mm or more in diameter or with high-grade dysplasia, or invasive cancer — did not take villous histology into account.

The study population was 72% male, with a mean age of 48 years.

In the December issue of Clinical Gastroenterology and Hepatology, the researchers report that FIT was positive in 5.6% of cases. Nearly 20% of patients had neoplasia, including 3593 (16.5%) with adenomas larger than 10 mm, 721 (3.3%) with advanced neoplasia, and 68 (0.3%) with invasive cancer.

In one analysis, the authors assumed that sigmoidoscopy would assess the rectosigmoid colon. In that circumstance, the sensitivity for detecting advanced proximal neoplasia would be 16.3% with rectosigmoid findings of advanced neoplasia, 37.6% with rectosigmoid findings of any neoplasia, and 22.3% with FIT, they said.

When “sigmoidoscopy” findings of any neoplasia and FIT were combined, the sensitivity for proximal neoplasia rose to 48.6%. The sensitivity of the combined test results was 62.5% for detecting proximal invasive cancer.

A similar pattern was seen when sigmoidoscopy was assumed to extend to the distal portion of the splenic flexure.

Even though sigmoidoscopy does not examine the right side of the colon, “many previous studies have showed predictability of sigmoidoscopy for proximal neoplasia,” Dr. Kato said, perhaps because “subjects who have distal neoplasia…are also likely to have proximal neoplasia.”

Limitations of their study include the homogeneous Japanese cohort, relatively young age, and predominance of males, as well as the testing of only 1 FIT specimen rather than 2 or 3. Moreover, the authors admit, “to know the real power of sigmoidoscopy, true sigmoidoscopy screening should be done.”

Dr. Kato’s team concludes that combined sigmoidoscopy/FIT “should be considered especially in communities that are not capable of colonoscopy screening, but it should be repeated at appropriate intervals to reduce the detection-failure rate of clinically relevant proximal lesions.”

In an editorial, Dr. James E. Allison from the University of California, San Francisco emphasizes that using colonoscopy results as a substitute for sigmoidoscopy weakens the study.

In email to Reuters Health, he pointed out that the “reach” of sigmoidoscopy “is very variable and depends on the anatomy of the patient, the skill of the operator, and the adequacy of the preparation.” More accurate data regarding the effectiveness of flexible sigmoidoscopy should be coming from results a trial that’s presently underway, he added.

In his editorial, Dr. Allison criticizes professional guidelines and funding agencies for their heavy emphasis on optical colonoscopy for colorectal cancer screening. “Warnings and admonitions about screening tests other than colonoscopy are particularly troublesome as evidence mounts that colonoscopy may not prevent as many cancers in the right colon as in the left,” he writes.

One problem, he explained in his email, is that “advanced neoplasms are not found in a significant number of patients who have distal neoplasms, and many patients with advanced neoplasms in the right colon do not have any distal neoplasms that might be found by colonoscopy.”

He concludes: “If important information is to be forthcoming for a national policy on screening for CRC, funding must be available for studies of all tests with established or promising evidence for efficacy.”

Reference:
Clin Gastroenterol Hepatol 2009;7:1269-1271,1341-1346.