Analysis of Colorectal Cancer Occurrence During Surveillance Colonoscopy in the Dietary Polyp Prevention Trial

Abstract

Colonoscopy can reduce mortality and incidence of colorectal cancer (CRC). However, interval CRC is occasionally detected in subjects with recent colonoscopy. Systematic evaluation of the reasons for interval cancer may be helpful in improving the quality of colonoscopy and in developing guidelines for surveillance. Methods: The Polyp Prevention Trial (PPT) was a randomized, controlled study evaluating the effect of a dietary intervention on adenomatous polyp recurrence. Subjects underwent a baseline colonoscopy (T0 exam) for eligibility, and were scheduled to undergo a repeat exam at 1 and 4 years. We examined the circumstances surrounding the diagnosis of CRC in the PPT population including demographic information, T0 colonoscopy findings, time interval to cancer diagnosis, location, size, and staging. An algorithm was developed to analyze and classify each cancer occurrence into one of four etiologies: 1) Incomplete polyp removal (with subsequent cancer at that site), 2) False negative biopsy (where a suspicious area was biopsied but cancer was not detected), 3) Missed cancer (cancer that occurred in a different location from the site of previous polyp removal and was diagnosed in a relatively short time frame from the most recent colonoscopy), and 4) New cancer. Results: Of 2079 patients, 15 cancers developed in 14 subjects (0.67%) over 5810 person years of observation (PYO)(2.4 cases/1000 PYO). 71.4%(10/14) had an advanced adenoma and 50% had ≥2 adenomas detected at the baseline examination. The cancers were found throughout the colon; 53.3% (8/15) were proximal to the splenic flexure and 33.3% (5/15) were in the cecum or at one of the flexures. Based on our analysis, 7/14 or 50% of patients had a potentially “avoidable” cancer, with diagnosis due to incomplete removal of an advanced adenoma with cancer developing at that site subsequently (4/14) or an apparent missed cancer (3/14). The “incomplete removal” category highlights the importance of assuring the complete excision of advanced adenomas. Missed cancers were relatively large and were all located proximal to the hepatic flexure, emphasizing the importance of assuring cecal intubation. Conclusion: Interval cancers occur with some frequency despite colonoscopy. Improved colonoscopy quality and follow up of potential incompletely removed high risk adenomas may have reduced cancer incidence, or resulted in earlier cancer detection in up to 50% of incident cancers in the PPT.