Data from a prospective colonic surveillance program involving 259 patients with chronic extensive Crohn's disease revealed a 25% cumulative risk of developing definite dysplasia or cancer by the 10th surveillance exam after an initial negative screening exam according to the findings of a longitudinal cohort study.
The cumulative risk of an initial finding of flat high-grade dysplasia or cancer after a negative screening colonoscopy was 7% after the ninth surveillance exam, with a median interval of 18 months between exams. The findings suggest that periodic surveillance colonoscopy should be part of routine management of chronic extensive Crohn's disease, said Dr. Sonia Friedman of Brigham and Women's Hospital, Boston, and her associates.
While the increased risk of colonic dysplasia and carcinoma in patients with chronic, extensive ulcerative colitis has been well described, less is known about those risks in patients with long-standing Crohn's disease. Previous studies often lumped together all patients with Crohn's disease and didn't separately examine those with extensive long-standing Crohn's colitis.
In 2001, Dr. Friedman and her associates reported a 22% chance of developing definite dysplasia or cancer by the fourth surveillance exam among 259 patients with chronic Crohn's disease who were followed from 1980 through 1998 (Gastroenterology 2001;120:820-6).
Now they report an update in those same patients, all of whom had at least 7 years of Crohn's colitis affecting at least one-third of the colon. Those in whom the results of screening colonoscopy were negative were contacted for a repeat examination at 2 years. Patients with results classified as indefinite (IND) for dysplasia were contacted for extensive repeat biopsies within 1 year, while those with one area of flat low-grade dysplasia (LGD) were contacted for repeat endoscopy within 1–6 months. Patients with recurrent or multifocal flat low-grade dysplasia (LGD), high-grade dysplasia (HGD), or cancer were referred for surgery. Those with “adenoma-like” polypoid dysplastic lesions that had been removed were contacted for repeat endoscopy within 1–6 months [Epub doi:10.1016/j.cgh.2008.03.019]).
A total of 1,424 examinations (screening and surveillance) was performed, with a median of 5 per patient. In all, 90% of the patients had extensive colitis, and 31% had undergone segmental colon resection. At the initial screen, definite dysplasia was found in 18 patients (7%). Of those 18 patients, 13 had LGD (7 polyps, 6 flat), 2 had HGD (both flat), and 3 had carcinoma (all masses). There were no colonoscopic complications, the investigators said.
The prevalence of definite dysplasia or cancer was significantly higher among patients who were older than 45 years than in those younger, but the prevalence for those older than 45 did not vary with disease duration.
In surveillance exams, a first positive finding of definite dysplasia or cancer was found in 30 patients, including LGD in 22 (14 polyps, 8 flat), HGD in 4 patients (2 polyps, 2 flat), and carcinoma in 4 patients (2 polyps, 2 masses). Analysis of several factors, such as age greater than 45 years or disease duration longer than 20 years at exam, family history of cancer or inflammatory bowel disease, female gender, pancolitis, and prior resection, did not identify any as consistent predictors of risk for dysplasia or cancer over time.
Compared with age- and gender-based cancer registry data from the general population, the 11 cancers detected in this study were significantly more than the 1.13 expected, suggesting that patients with extensive Crohn's disease are indeed at increased risk for developing colon cancer, Dr. Friedman and her associates said.
The calculated cumulative risks—25% for LGD, HGD, or cancer, and 7% for flat HGD or cancer by the 10th surveillance exam—are high, considering that 31% of the screening exams and 4.8% of the surveillance exams were preceded by a partial colon resection. However, these data parallel those of studies of cancer in ulcerative colitis patients with similar extent and duration of disease, they commented.