Author's Reply

We appreciate the interesting and thoughtful comments1 on our study.2

Dr. Cha correctly indicates that our results may be limited by a referral bias, selection bias, and recall bias, as the study was based on data obtained through telephone surveys and a retrospective review of medical records from a single tertiary referral center. He also suggests that the poor quality of colonoscopy rather than accelerated change in tumor biology is the major cause of interval cancer.

With regard to the second point, in our study, we aimed to investigate the clinicopathologic features and predictors of interval cancer. It was difficult to evaluate the cause of interval cancer through a retrospective review of medical records. Although colonoscopic images were available in a few cases, endoscopic reports were not available in most cases. The occurrence of interval cancer despite a prior negative colonoscopy result may be due to the poor quality of colonoscopy. We believe that the accelerated change in tumor biology may cause interval cancer as the interval cancer group patients were younger than the sporadic cancer group patients in our study.

Thus, a population-based cohort study with complete follow-up would be needed to avoid referral and selection bias. Moreover, colonoscopic images along with detailed endoscopic reports that include location of the lesion, completeness of endoscopic resection, and quality indicators of the colonoscopy and colonoscopists would also be needed to avoid a recall bias and to confirm the cause of interval cancer in the Korean population.