Abstract 2236: A novel integrated predictive model for pancreatic cancer

Abstract Background: Pancreatic cancer is the 7th leading cause of cancer-related mortality and the 12 th most common cancer globally. Many studies have identified risk factors for pancreatic cancer, nevertheless, there have been few combined risk factor models developed for the UK specific population. Due to the high incidence with a poor survival rate together with the potential high preventable rate, it is important to establish an accessible predictive risk factor model for identifying potential high-risk populations. Study aim: To establish a pancreatic cancer risk prediction model by integrating the current risk factors together with established genomic biomarkers. Method:Using data from the UK biobank we have investigated the role of established and suggested risk factors for pancreatic cancer. The established risk factors include age, BMI, smoking, alcohol consumption, diabetes and pancreatitis. We also calculate the added prediction from the current polygenic risk score (PRS) to evaluate the risk for genetic predisposition. Last, we utilise the multivariate logistic regression models to establish a novel risk model by integrating the well-known risk factors such as cigarette smoking, heavy alcohol consumption, increased BMI, chronic pancreatitis, diabetics, some germline mutation, and our new evaluated PRS and biological clocks. Results: We will present the performance of the novel integrated model which will be useful for use in primary care to identify the at-risk population. This will assist in the promotion of prevention recommendations and for future health check follow-up. Citation Format: Temin Ke, Artitaya Lophatananon, Kenneth Muir, Xinzhu Yu. A novel integrated predictive model for pancreatic cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2236.