Back to 2025 Posters
IN PATIENTS WITH CHRONIC PANCREATITIS METFORMIN USE WAS NOT ASSOCIATED WITH A RISK OF DIAGNOSIS OF DUCTAL ADENOCARCINOMA OF THE PANCREAS
Luis F. Lara
*, Andrew Ofosu, Muhammad H. Dawwas, Moamen Gabr, Rosanne Danielson, Milton Smith, Wei-Wen Hsu, Inuk Zandvakili, Mubarak Sayyar, Davendra Sohal, Jordan Kharofa, Sameer Patel, Gregory C. Wilson, Syed Ahmad
University of Cincinnati, Cincinnati, OH
Introduction
The incidence of pancreatic ductal adenocarcinoma (PDAC) is increasing and will become the second-leading cause of mortality related to cancer. Chronic pancreatitis (CP) is associated with a twofold increased risk of PDAC. In PDAC, precursor lesions progress to the malignant state usually associated with
KRAS and then other successive mutations. PDAC prevention and early diagnosis have been generally unsuccessful. Metformin has been variably associated with a decreased risk of malignancy including PDAC which could be due to its anti-inflammatory properties, decreased hyperinsulinemia and its effect on AMPK, the mTOR complex and the GPL system. We evaluated if the risk of PDAC in patients with CP was affected by taking metformin.
Methods
Patients with a diagnosis of CP who were taking metformin for at least one year were compared to patients with CP not taking metformin and followed for their risk of PDAC at least 3 years after the initial diagnosis of CP. Subjects were obtained from the TriNetX (Cambridge, MA) real-world database of electronic medical. Co-variates included age, sex, race, body weight, smoking and alcohol use status, and the outcome of PDAC was analyzed by excluding 14 ICD codes, 10 medication codes (including insulin use), and 10 CPT codes (including previous pancreatic surgery). The cohorts were 1:1 propensity score matched and a P value <0.001 was considered significant.
Results
There were 161,002 subjects identified. After propensity score matching there were 1,568 patients in the CP and metformin cohort and in the CP not on metformin cohort, respectively. Mean age was 62.1 (±16.3), 60.3% female, 9% Latino, 15.9% Black. The groups were adequately matched for the covariates except weight which was 187.9 ± 55 lbs in the CP and metformin group and 166.5 ± 45 lbs in the control group, P>0.01. There were no cases of PDAC in any of the two groups. We expanded the search for PDAC at any time after the diagnosis of CP and still found no cases.
Conclusions
There were no cases of PDAC in this cohort of CP even after eliminating the 3-year lead time for the diagnosis. It is possible that strict exclusion criteria caused a type II error since the number of evaluable patients decreased from > 161,000 subjects to 1,568 patients in the matched cohorts. It is also possible that PDAC is not as common as described in epidemiological studies which may have a biased population whereas this large database is more representative of real-world data. Metformin may affect PDAC risk and progression through its glucose and insulin metabolic pathway but probably has no effect on PDAC precursors such as
KRAS. The potential antineoplastic effects of metformin could be mediated by AMPK and the mTOR complex which are not a malignant pathway for PDAC, thus metformin may have no effect on PDAC progression in patients with CP.
Back to 2025 Posters
