Colorectal cancer prevention studies: the importance of defining disease risk.

Abstract

Colorectal cancer (CRC) claims 57,000 lives yearly in the United States.1 Although a modest decrease of 1.8% in the incidence rate for this disease occurred from 1985 to 1995, incidence rates since then have stabilized. Currently, CRC is the third leading cause of cancer death in both men and women. These statistics persist despite the understanding that the identification and removal of precursor adenomas during screening colonoscopy can prevent CRC. In the landmark National Polyp Study, reported in 1993,2 patients having colonoscopic screening with polypectomy at intervals of 1 and 3 years after baseline adenoma removal achieved a reduction in CRC incidence of approximately 76% to 90%, compared with three reference populations that did not receive polypectomies. Numerous subsequent studies have validated this observation. A particularly dramatic example of effective cancer prevention by close surveillance is provided by Jarvinen et al.3 These researchers studied asymptomatic members of families with hereditary nonpolyposis colorectal cancer to determine the effect of CRC screening over a 10-year surveillance period. One group of 133 subjects received screening at 3-year intervals, and a second group of 118 subjects was followed without screening. CRC occurred in 4.5% of the screened population versus 15% of the control group, providing a cancer incidence reduction of 62% that likely occurred because of polypectomy. Equally important, tumor stage at cancer diagnosis was much earlier in the screened population, and no deaths caused by CRC occurred in the screened group, versus five deaths in the control group. Although these data make it clear that CRC screening can dramatically decrease CRC incidence, the significant drawbacks of high cost and lack of patient acceptance mean that only 10% to 20% of the target population receives CRC screening. It is clear that, although we must continue to improve access to and utilization of CRC screening, we must also search for less costly and better accepted methods of cancer prevention. In addition to adenoma removal, a wealth of data from human epidemiology suggests that dietary or pharmacologic substances can reduce adenoma formation and, in doing so, prevent CRC.4 Because of the strength of the adenoma-carcinoma link in human observational studies and in polypectomy trials, the reduction of adenoma incidence is used as a surrogate endpoint for CRC in chemoprevention studies. The report in this issue by Chu et al.5 addresses the incidence of synchronous and metachronous adenoma formation in patients with recently diagnosed CRC. In this study, synchronous disease was defined as adenomas discovered at a baseline colonoscopy that took place within 1.5 years after the resection of an early stage CRC. Recurrent adenomas were those occurring at any time following the baseline colonoscopy. These investigators found that 60% of patients had synchronous adenomas, a figure consistent with other studies.6,7 The 3-year cumulative adenoma recurrence in this cohort was 35%. This figure is similar to that found in the placebo arm of an aspirin chemoprevention study reported earlier this year by Sandler et al.8 These patients also had a history of CRC; however, the time from initial cancer diagnosis to baseline colonoscopy was highly variable. Some patients entered the study immediately following cancer diagnosis, and others 5 or more years following successful treatment. Despite these differences in eligibility, the adenoma recurrence rates in the Sandler et al.8 study are comparable. By Kaplan-Meier estimation, the proportion of patients diagnosed with adenomas by 3 years after clearing colonoscopy was approximately 32%.8 Any cancer prevention treatment, even one as seemingly innocuous as calcium supplementation, comes with Received August 14, 2003; accepted August 22, 2003. From Brigham & Women’s Hospital, Boston, Massachusetts. Address correspondence to: Monica M. Bertagnolli, MD, Brigham & Women’s Hospital, 75 Francis St., Boston, MA 02115; Fax: 617-5826177; E-mail: mbertagnolli@partners.org.