Risk of second primary colorectal cancer with particular reference to age at diagnosis.

Objective: Patients with a history of colorectal cancer are considered at increased risk of second metachronous colorectal cancer (SM-CRC), for which they frequently receive intensive colonoscopic surveillance. In view of the ambiguous nature of the existing evidence and the growing interest in targeted surveillance, we sought to quantify long-term risk with particular reference to age at diagnosis. Method: The Surveillance Epidemiology and End Results database was used to estimate risk of SM-CRC after first incident colorectal cancer diagnosed between 1975 and 1999. We calculated time-dependent rates using Kaplan-Meier estimates and relative risk compared with the US general population. Results: From 311 689 eligible patients, there were 6387 SM-CRCs. At 15-years following initial diagnosis, the SM-CRC rate was 6.3% (95% CI, 6.1-6.5). For patients with synchronous primary cancers (n = 9936, 3.2%), the 15-year SM-CRC rate increased to 10.5% (95% CI, 9.1-12.2). Younger age predicted for increased relative risks but absolute cumulative rates at 15 years were low (Table). Conclusion: The long-term cumulative risk of SM-CRCs after first colorectal cancer is low, even among younger age patients. These data do not support the routine use of high-frequency colonoscopy surveillance in patients with a history of colorectal cancer. Age groups (years)20-2930-3940-4950-5960-6970-7980+Relative risk - men60.415.14.521.871.341.281.15Relative risk - women18.49.634.082.291.721.461.24Absolute risk - both sexes3.533.744.294.706.157.597.26.