Serum Claudin-18.2 Expression and Gut Microbiota Dysregulation in Hepatocellular Carcinoma: Correlation with Immune Checkpoint Inhibitor Response

Background: Hepatocellular Carcinoma (HCC) remains a therapeutic challenge, with Immune Checkpoint Inhibitors (ICIs) offering limited response rates. This study investigates the prognostic value of serum Claudin-18.2 (CLDN18.2) and gut microbiota in HCC patients treated with anti-PD-1 therapy. Methods: A prospective cohort of 220 HCC patients (BCLC stage B/C) receiving camrelizumab was enrolled. Serum CLDN18.2 levels were measured by ELISA, and stool microbiota was analyzed via 16S rRNA sequencing. Primary endpoints included objective response rate (ORR), overall survival (OS), and progression-free survival (PFS). Results: High serum CLDN18.2 (>150 ng/mL) was observed in 48% of patients and correlated with larger tumor size (≥5 cm: OR=2.31, p=0.002), microvascular invasion (MVI: OR=1.89, p=0.014), and lower ORR to ICIs (18% vs. 39%, p<0.001). Multivariate analysis showed high CLDN18.2 independently predicted worse OS (HR=2.17, 95% CI: 1.38-3.42, p=0.001) and PFS (HR=1.94, 95% CI: 1.22-3.09, p=0.005) (Table 1). Gut microbiota analysis identified Ruminococcus gnavus and Alistipes indistinctus as key predictors of ICI resistance. High R. gnavus abundance was associated with reduced ORR (12% vs. 35%, p<0.001) and shorter OS (HR=2.45, 95% CI: 1.56-3.86, p<0.001) (Table 2). Combined CLDN18.2 and R. gnavus improved prognostic accuracy (C-index=0.81 vs. 0.69 for single markers, p=0.003). Patients with both high CLDN18.2 and R. gnavus had the worst outcomes (median OS=7.8 months vs. 16.5 months in low-risk groups, p<0.001) (Table 3). Mechanistically, R. gnavus promoted PD-L1 expression via TLR4/MyD88 signaling in vitro. Conclusion: Serum CLDN18.2 and R. gnavus are independent predictors of ICI resistance in HCC. Their combination may optimize patient selection for ICIs, paving the way for microbiota-targeted adjuvant therapies.