Abstract A046: CLDN18.2 Expression in PDAC: KRAS Independence and Stage IV Prognostic Value

Abstract Background: In pancreatic adenocarcinoma (PDAC) management, KRAS inhibitors are entering into clinical use and claudin 18.2 (CLDN18.2) is an emerging therapeutic target. The expression levels of CLDN18.2 and KRAS mutation status and clinical outcome in PDAC is less studied. We would like to test the hypothesis of the relationship between CLDN18.2 expression levels with KRAS mutation status and CLDN18.2 association with survival. Methods: Retrospective analysis of 135 tumors from PDAC patients with CLDN18.2 tested by immunohistochemistry (IHC, clone 43-14A, Ventana) and KRAS mutation status at MD Anderson were analyzed. CLDN18.2 expression levels by IHC were calculated as % of cells with 2+/3+ membranous staining intensity. The Palantir Foundry system was used to query clinical and molecular information including overall survival (OS) data. Correlations of CLDN18.2 expression, KRAS mutation and OS were tested. Non-parametric tests were used given non-normal distributions of CLDN18.2 levels (Shapiro-Wilk p<0.001). Results: CLDN18.2 staining intensity was mostly 2+/3+ (n=90, 66.7%) and 3+ (n=9, 6.7%). CLDN18.2 prevalence was strongly cutoff-dependent: 40.5% at ≥75%, 59.1% at ≥40%, 71.0% at ≥10%, 73.3% at ≥1%. 122 (90.4%) patients were KRAS mutated and 13 were KRAS wildtype (9.6%). Median CLDN18.2 expression levels were 60% in KRAS-mutated tumors (n=119) versus 10% in KRAS-wildtype tumors (n=13, Mann-Whitney p=0.20). 44 patients were initially diagnosed with stage IV disease and among them, higher CLDN18.2 expression ( ≥75%) was associated with a trend of better OS (HR=0.41, 95%CI 0.10–1.65, p=0.17), with consistent HR<1 at all cutoffs (≥10%–≥90%) with median follow-up of 21.6 months. Conclusions: In a real-world PDAC cohort: (1) CLDN18.2 prevalence is highly threshold-dependent, supporting trial design that accounts for varied therapeutic cutoffs. (2) No statistically significant difference of expression levels of CLDN18.2 between wild type patients and mutated patients which could be due to limited sample size. (3) Stage IV-specific OS signal favoring positive patients (HR=0.41) warrants validation in larger cohorts. Citation Format: Sacha El Khoury, Ryan Lewis, Paul Roy, Saikat Chowdhury, Jarret Feldman, Brandon G. Smaglo, Robert A. Wolff, Shubham Pant, Jason Willis, Maria Pia Morelli, Ryan Huey, Micheal Overman, Huili Zhu, S. Daniel Haldar, Camila Braganca Xavier, Nicole Baranda Balmaceda, Camila Lopez, Anthony Chen, Rebecca Snyder, Mathew H. G. Katz, Huamin Wang, Ethan B. Ludmir, Eugene J Koay, John Paul Shen, Dan Zhao. CLDN18.2 Expression in PDAC: KRAS Independence and Stage IV Prognostic Value [abstract]. In: Proceedings of AACR Drug Discovery and Development (AACR D3) Conference; 2026 Jul 21-24; Boston, MA. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(14_Suppl):Abstract nr A046.