Colonoscopic polypectomy reduced the risk of death from colorectal cancer by more than half in a study that followed patients for as long as 23 years, investigators reported in the Feb. 23 issue of the New England Journal of Medicine.
The findings demonstrate that adenomas identified and removed at colonoscopy are clinically important because they have the potential to progress to cancer and cause death, said Ann G. Zauber, Ph.D., of the department of epidemiology and biostatistics at Memorial Sloan-Kettering Cancer Center, New York, and her associates.
The investigators performed long-term follow-up of subjects who had participated in the National Polyp Study, a randomized clinical trial of patients prospectively referred to seven clinical centers for colonoscopy during 1980-1990. The researchers analyzed mortality data from 2,602 of these subjects who had been found to have adenomatous polyps at that initial colonoscopy so they could determine whether removing the lesions had actually saved lives.
The study subjects had been referred for colonoscopy because of positive findings on a barium enema examination (27%), sigmoidoscopy (15%), fecal occult blood testing (11%), or other tests (10%), or because they had symptoms (32%) or a family history of colorectal cancer (5%). The median follow-up was approximately 16 years, with a maximum follow-up of 23 years.
A search of the National Death Index, a registry of all deaths in the United States, was used to identify the 1,246 study subjects (48%) who had died during follow-up, including 12 who died from colorectal cancer.
The authors then determined the expected number of deaths from colorectal cancer in the general population among people of comparable age, sex, and race, and found that 25.4 such deaths would be expected. This figure was determined by using data from the Surveillance, Epidemiology and End Results (SEER) program, and sensitivity analyses confirmed these findings.
This indicated that colonoscopy with polyp removal reduced mortality from colorectal cancer by 53%, compared with the expected rate of this cancer in the general population. "The cumulative mortality rate in the adenoma cohort at 20 years was 0.8%, as compared with an estimated 1.5% in the general population, on the basis of SEER data," Dr. Zauber and her colleagues wrote (N. Engl. J. Med. 2012;366:687-96).
The investigators also compared colorectal cancer mortality in the study cohort against that in a group of 773 National Polyp Study participants who had been found to have nonadenomatous polyps on their initial colonoscopy. During the first 10 years after that colonoscopy, mortality in the adenoma cohort (0.19%) was not significantly different from mortality in this internal control group (0.15%).
Thus, the risk of death from colorectal cancer was comparable between patients whose adenomas were removed at initial colonoscopy and those who only had nonadenomatous polyps, which were also removed, they said.
Finally, Dr. Zauber and her associates used a microsimulation model to estimate what the mortality would have been if the adenomas had not been removed "and the natural history of the adenoma-carcinoma sequence had proceeded without intervention." This model "showed an even larger reduction in mortality [after] polypectomy than the comparison with the SEER incidence-based mortality rates," they noted.
At present, there are three prospective, long-term, randomized controlled trials taking place in northern Europe, Spain, and the United States in which mortality end points after screening colonoscopy will be measured directly. Those data, however, will not be available for at least another decade. In the meantime, the findings of this study indicate that identifying and removing adenomas via colonoscopy significantly cuts the rate of death from colorectal cancer, the authors concluded.
This study was supported by the National Cancer Institute, the Society of Memorial Sloan-Kettering Cancer Center, the Tavel-Reznik Fund, and the Cantor Colon Cancer Fund. The investigators reported no financial conflicts of interest.