Co-expressing pattern of multiple biomakers and dynamic change of Claudin18.2 expression after systemic chemotherapy in advanced gastric cancer.

Abstract

4037 Background: Gastric cancer (GC) is a heterogeneous disease classified by Lauren classification and TCGA’s four molecular subtypes. Targeting Claudin (CLDN)18.2 has emerged as a promising therapy. However, a comprehensive understanding of CLDN18.2 in advanced GC, its clinicopathologic correlations, and prognostic significance remains limited. We analyzed the co-expression of multiple biomarkers and dynamic changes in CLDN18.2 following systemic chemotherapy. Methods: This retrospective study included 402 patients with stage IV GC at Yonsei Cancer Center (2015–2023) treated with palliative doublet chemotherapy. Paired pre- and post-treatment tissues were available for 124 patients. CLDN18.2 expression was assessed using the VENTANA CLDN18 (43-14A) RxDx Assay, with positivity defined as ≥75% of tumor cells. HER2, EBV, dMMR, and PD-L1(22C3) were also evaluated by immunohistochemistry. Overall survival (OS) was estimated using the Kaplan-Meier method, and a log-rank test was performed to compare survival according to CLDN18.2 positivity. Results: The CLDN18.2 positivity rate was 30.6%, consistent with previous reports. It was slightly higher in HER2-negative (31.5%) than HER2-positive (27.3%) (p = 0.44). CLDN18.2 positivity was significantly higher in EBV-positive than negative (63.6% vs 29.4%, p < 0.05) and in pMMR versus dMMR (31.7% vs 9.5%, p < 0.05). Regarding PD-L1 status, the positivity rate was higher in the PD-L1 negative compared to the positive by CPS 1(36.1% vs 25.8%, p < 0.05). We observed that CLDN18.2 expression seems to be decreasing with higher PD-L1 expression (28.5 % in patients with CPS ≥5, 24% in CPS ≥10). The OS (median months, 95% CI) based on CLDN 18.2 expression was similar [22.5 (18.6-25.2) vs 23.2 (17.9-27.4) in CLDN18.2 positive group], regardless of HER2 status. Among 124 patients with paired samples, pre-treatment CLDN18.2 positivity was 22.8%, increasing to 36.5% post-treatment. The increase was more pronounced in HER2 negative (39.0%) compared to positive (29.5%). Notably, 79.9% of patients showed consistent CLDN18.2 expression between pre- and post-treatment samples, while 20.1% demonstrated changes in expression. These changes were not associated with other markers such as HER2, EBV, dMMR, or PD-L1. Conclusions: Higher CLDN18.2 positivity rates in specific subgroups, such as pMMR and PD-L1 negative, suggest the potential role of anti-CLDN therapy in these subgroups. This study also highlights the dynamic changes of CLDN18.2 expression in advanced gastric cancer, with an increase observed after first-line systemic treatment in a subset of patients. Further studies should focus on prospective analyses and stratify changes in CLDN18.2 expression by treatment regimen to better understand its role as a therapeutic target and its implications for treatment resistance and disease progression.