Accurate identification of interval colorectal cancers: results from a single Australian jurisdiction

Introduction: Interval colorectal cancers (CRCs) are those diagnosed within 5 years of a negative colonoscopy. We investigated the proportion of interval CRCs that develop within 6–60 months of colonoscopy, and their clinical and molecular characteristics. Methods: We performed a cohort study of all individuals who underwent colonoscopy between 2001 and 2008 at a single large teaching hospital. These records were linked with (i) the regional cancer registry using probabilistic record linkage and (ii) the hospital medical record using deterministic linkage. Matches were verified, and pathology, tumor stage and mismatch repair status were extracted. Results: A total of 10 854 individuals had a colonoscopy at our centre in the period. Linkage to ACT and NSW Cancer Registries detected 384 and 98 CRCs for notification dates of 2001–2013 (ACT) and 2001–2010 (NSW). The second match process identified a further 55 CRCs from a search of the hospital electronic medical records using relevant ICD‐10 diagnosis codes. A total of 404/537 CRCs (58% male) were included in this study after excluding cases that did not have a colonoscopy between 2001–2008, diagnosis of carcinoid tumor, metastatic malignancy of unknown primary and cases that could not be verified by pathological or clinical records. There were seventeen (4.2%) interval cancers. Individuals with interval cancers were more likely to have diverticulosis at the index colonoscopy (p = 0.019), and the index colonoscopy was more likely to be performed by a trainee (p = 0.044). Interval cancers were somewhat more likely to have mucinous histology (p = 0.059). There were no differences with respect to CRC site, stage, mismatch repair or BRAF mutation status. Conclusions: In our population‐based cohort study in Australia, 4.2% of patients with CRC had interval cancers. The data show that improvement is possible in colonoscopy training and surveillance in diverticulosis. These data are comparable to rates of interval cancers from other population studies and establish a benchmark for Australian jurisdictions