Abstract Background: Individuals with polypectomy are still at higher risk of developing colorectal cancer than general population and are thus recommended to undergo colonoscopy surveillance for timely detection and prevention of subsequent colorectal cancer. The current clinical guidelines regarding the interval of surveillance are based on weak evidence. There is substantial overuse of surveillance colonoscopy in low-risk individuals and underuse in high-risk individuals. Effective risk stratification tools for post-polypectomy colorectal cancer (PPCRC) are lacking. Methods: Leveraging the integrated endoscopic, histopathologic and epidemiologic data in 3 population-based cohorts (Nurses’ Health Study (NHS) I, II and Health Professionals Follow-up Study (HPFS)), we developed a risk score to predict incident PPCRC among 26,744 patients with a polypectomy between 1986 and 2017. We validated the risk score in the Mass General Brigham (MGB) colonoscopy cohort of 76,603 individuals with a polypectomy between 2005 and 2017. In the MGB cohort, we used the validated natural language processing algorithms to extract detailed histopathologic data and merged them with the endoscopic and other clinical data from the electronic health record systems. The detailed protocol for the cohort development has been published (PMID: 33590405). We used Cox regression to calculate the hazard ratios (HRs) and assessed the discrimination using C-statistics and reclassification using the Net Reclassification Improvement (NRI). Results: During a median follow-up of 12.8 (IQR: 9.3, 16.7) and 5.1 (IQR: 2.7, 7.8) years in the NHS I, II/HPFS and MGB colonoscopy cohorts, we documented 220 and 241 PPCRC cases, respectively. We identified a PPCRC risk score based on 11 predictors. In the validation cohort, the risk score showed a strong association with higher risk of PPCRC (HR for high vs. low, 3.55, 95% CI, 2.59-4.88) and demonstrated a C-statistic (95% CI) of 0.75 (0.70-0.79), and was discriminatory even within the low- and high-risk polyp groups of patients (C-statistic, 0.73 and 0.71, respectively) defined by the current colonoscopy surveillance guidelines, leading to a NRI of 45% for PPCRC patients. Conclusion: We developed and validated a risk stratification model for PPCRC that can be used to guide tailored colonoscopy surveillance. Citation Format: Markus Dines Knudsen, Liang Wang, Kai Wang, Georgios Polychronidis, Paula Berstad, Kana Wu, Xiaosheng He, Dong Hang, Zhe Fang, Shuji Ogino, Andrew T. Chan, Edward L. Giovannucci, Molin Wang, Mingyang Song. Development and validation of a risk prediction model for post-polypectomy colorectal cancer in 4 prospective US cohorts [abstract]. In: Proceedings of the AACR Special Conference on Colorectal Cancer; 2022 Oct 1-4; Portland, OR. Philadelphia (PA): AACR; Cancer Res 2022;82(23 Suppl_1):Abstract nr PR002.
Read full abstract