Abstract 30: Impact of race, sex and age on the risk of pancreatic cancer in new onset diabetics in real-world data

• Published in: Cancer research (Chicago, Ill.) Ill.), 2021-07, Vol.81 (13_Supplement), p.30-30
• Main Authors: Porter, Nancy R., Afghani, Elham, Lau, Bryan, Goggins, Michael, Klein, Alison P.
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2021-30

Background: Pancreatic cancer (PC) is the 3rd leading cause of cancer death in the US with a 5 year survival rate of 10%. Diabetes (DM) is both a risk factor and early symptom of PC. Individuals with longstanding DM have a 2-fold increased risk of PC, and up to 1% of older new-onset diabetics (NOD) develop PC within 3 years of their NOD diagnosis. We examined if individuals with NOD could be identified in real world data and estimate the risk of PC for subgroups defined by age, sex and race. Methods: Johns Hopkins School of Medicine conducted this study using the OptumLabs® Data Warehouse (OLDW). The OLDW contains de-identified retrospective administrative claims data, including medical and pharmacy claims and eligibility information. Enrollees from 1/2008-9/2018 with 2 ICD-9 or ICD-10 DM codes after a 1 year wash-out period were considered NOD. Enrollees with no claims evidence of DM for at least a 1-year prior to a randomly selected index date were non-DM. Those with prevalent DM were excluded. Time-to-event analysis was conducted using a flexible Weibull survival model. Results: Our cohort included 5,845,240 individuals, >25% of which are non-White. There were 424,210 (7.3%) cases of NOD of which 1,594 (0.38%) were diagnosed with PC within 2 years of their NOD diagnosis: In the non-NOD cohort 4,865 (0.09%) developed PC. The risk of PC among NOD patients varied by age, sex and race (p