Background: Anti-angiogensis agents were reported to enhance the efficacy of PD-1 inhibitor by inhibiting the infiltration of suppressive immune cell populations. Clinical studies indicated the synergistic anti-tumor effect of the combination of PD-1 inhibitors and anti-angiogenesis agents (PA). We aim to assess the efficacy of PA versus chemo-PA regiment in mGC, as well as the potential predictive biomarkers. Methods: A total of 55 GC tumors were analyzed using the next-generation sequencing (NGS), immunohistochemistry (IHC), and the tumor immune microenvironment (TIME) by multiplex immunofluorescence (mIHC). Clinical outcomes [Disease Control Rate (DCR), Progression-free survival (PFS) and overall survival (OS)] were compared between patients receiving PA treatment (n=32) and chemo-PA (n=23) treatment. Exploratory biomarker analyses were conducted for the prediction for the efficacy of PA/chemo-PA treatment. Results: Patients receiving PA regiment had comparable DCR (69% vs 50%, P =0.25) and PFS (median PFS: 6.53 vs 4.73 months, HR=0.70, 95%CI: 0.27-1.78, P =0.52), compared with those treated with combination of chemo-PA regiment as second or later-line treatment in mGC. Patients with HMCN1 wild-type showed significant improvements in DCR (77% vs 0%, P =0.023) and PFS (median PFS: 6.53 vs 1.17 months, P =0.0017), versus those with HMCN1 mutations in PA treatment cohort. HMCN1 mutations are associated with lower infiltration of endothelial cells and lower score of VEGF ligand-receptor interaction, according to the EPIC and pathway enrichment analysis ( P< 0.05). In addition, analysis of TIME by mIHC revealed that the fraction of CD68+ tumor-associated macrophages in the tumor center was higher in responders than non-responders before treatment (median: 17.59% vs 2.45%, P=0.016). Conclusions: PA regiment and chemo-PA regiment had similar efficacy in second or later line treatment of mGC. HMCN1 mutations maybe a predictive marker for PA regiment in HER2-negative mGC, which should be further validated in a large cohort.