Abstract
PURPOSE: The incidence of young-onset pancreatic cancer (YO-PC) has risen over the past two decades, yet its molecular characteristics and long-term outcomes remain poorly defined. METHODS: We retrospectively evaluated patients with PC treated at a tertiary referral center from 2016 to 2022 who had available molecular data. Patients were classified as YO-PC (≤50 years) or average-onset PC (AO-PC, >50 years). A subset analysis examined outcomes in those who underwent curative-intent pancreatectomy. Primary end points included overall survival (OS) and recurrence-free survival (RFS). RESULTS: Among 511 patients, 10.9% had YO-PC (n = 56; median age 44 years). Patients with YO-PC more commonly self-identified as non-White compared with patients with AO-PC (41.1% v 26.4%, P = .03). BMI, anatomic stage, and CA19-9 level at presentation were similar between groups. KRAS mutations were the most prevalent somatic alterations in both the YO-PC and AO-PC cohorts (87.0% v 88.0%, P = .83), followed by TP53 (73.5% v 74.1%, P = .93), CDKN2A (14.6% v. 23.6%, P = .20), and SMAD4 (18.2% v 12.9%, P = .34). KRAS-specific allele subtypes were also similar (P = .38). A subset analysis in the surgical cohort (n = 167) yielded similar results. OS was similar for YO-PC and AO-PC (18.7 v 21.7 months; P = .29). In the surgical cohort, OS and RFS for YO-PC and AO-PC were also similar (OS, 31.2 v 47.1 months, P = .34; RFS, 10.3 v 14.7 months, P = .10). CONCLUSION: In this single-institution study, patients with YO-PC and AO-PC demonstrated similar clinicopathologic and molecular profiles; however, the study population to date may be underpowered. Routine molecular testing on all patients diagnosed with PC will be critical to better understand its clinical and molecular heterogeneity and to inform design of practice-changing clinical trials.