Response to Liu and Zhang

We appreciate the opportunity to respond to the comments by Liu and Zhang [1] regarding our study.In their comments, the authors discussed several limitations in our study including unmeasured confounding factors.We used a target trial emulation framework to examine the associations of GLP-1RAs with pancreatic cancer incidence in patients with type 2 diabetes (T2D) who had no prior diagnosis of pancreatic cancer.We acknowledged our study has the limitations inherent in observation studies using patient electronic health records, including uncontrolled and unmeasured confounding factors and bias that preclude causal inferences and need to be examined in randomized clinical trials.The following are our responses to specific comments raised by the authors: (1) in this study, the GLP-1RA and comparison groups were propensity-score matched for family history of cancer and genetic susceptibility to cancer.However, patient electronic health records (EHRs) capture limited information on specific cancer gene biomarkers.While the genetic and family history data are important for subgroup analysis, they are not confounders because the choice of GLP-1RAs or other anti-diabetic medications for treating T2D is unlikely to be determined by patient cancer genetic biomarkers; (2) the EHR data we used did not capture data on medication