Zolbetuximab-based chemotherapy in CLDN18.2-positive gastric cancer: real-world tumor response and early albumin kinetics

BACKGROUND: Zolbetuximab combined with fluoropyrimidine-oxaliplatin chemotherapy has demonstrated clinical benefit in CLDN18.2-positive, HER2-negative advanced gastric and gastroesophageal junction adenocarcinoma (GC/GEJC). In clinical practice, however, gastrointestinal toxicity and serum albumin decline are frequently encountered during treatment, and the relationship between early albumin kinetics and antitumor efficacy remains unclear. METHODS: We conducted a single-center retrospective exploratory study of patients with unresectable or recurrent CLDN18.2-positive GC/GEJC treated with zolbetuximab-based chemotherapy between July 2024 and August 2025. Tumor response was evaluated according to Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 in patients with measurable disease. The objective response rate (ORR) and early tumor shrinkage (ETS ≥20%) were assessed. Serum albumin kinetics were analyzed using baseline albumin, landmark nadir albumin measured at or immediately before the first radiological evaluation, and relative change in albumin from baseline (ΔAlb). Associations between albumin metrics and tumor response were explored using exploratory analyses, including continuous-variable modeling. Relative dose intensity and chemotherapy backbone (CAPOX vs FOLFOX) were also descriptively examined. RESULTS: Among 38 eligible patients, 24 had measurable disease and were evaluable for RECIST-based response. Partial response was observed in 18 patients, yielding an ORR of 75.0% (95% CI, 55.1-88.0). Early tumor shrinkage ≥20% was achieved in a proportion of evaluable cases. Baseline serum albumin showed a moderate positive correlation with landmark nadir albumin (Spearman's ρ = 0.35, P = .030) and with ΔAlb (ρ = 0.52, P = .0007), indicating that patients with higher baseline albumin tended to experience greater absolute albumin decline. Neither landmark nadir albumin nor ΔAlb was significantly associated with the ORR or ETS. Early reductions in serum albumin were frequently observed during initial treatment cycles but did not correspond to impaired tumor response. CONCLUSIONS: In this real-world exploratory cohort, zolbetuximab-based chemotherapy achieved objective tumor responses consistent with prior clinical trials. Early serum albumin decline was not associated with inferior tumor response. These findings suggest that early albumin decline should not be interpreted as a surrogate marker of treatment failure or reduced antitumor efficacy but rather as a treatment-related physiological change. Prospective studies are warranted to clarify the clinical implications of albumin dynamics and optimize supportive care during anti-CLDN18.2 therapy.