Association of peripheral basophils with tumor M2 macrophage infiltration and outcomes of the anti-PD-1 inhibitor plus chemotherapy combination in advanced gastric cancer

Background Although the anti-programmed death-1 (PD-1) inhibitor plus chemotherapy combination has been approved as the standard first-line treatment for advanced gastric cancer, a proportion of patients do not significantly benefit from this therapy. Who would respond poorly to this treatment and the underlying mechanisms of treatment failure are far from clear. Methods We retrospectively analyzed the associations between the peripheral basophils at baseline and clinical outcomes in 63 advanced gastric cancer patients treated with anti-PD-1 plus chemotherapy and 54 patients treated with chemotherapy alone. Immunohistochemistry and immunofluorescence staining in gastric cancer samples were utilized to investigate the basophil-related immunophenotype. Results The optimal cutoff of basophil count to distinguish responders to anti-PD-1 plus chemotherapy from non-responders was 20.0/μL. Compared with the low basophil group (≤ 20.0/μL, n = 40), the high basophil group (> 20.0/μL, n = 23) had a significantly lower objective response rate (ORR 17.4% vs. 67.5%, p = 0.0001), worse progression-free survival (median PFS 4.0 vs. 15.0 months, p = 0.0003), and worse overall survival (median OS not reached, p = 0.027). Multivariate analyses identified a basophil count of > 20.0/μL as an independent risk factor for a worse ORR (OR 0.040, 95% CI 0.007–0.241, p = 0.0004), worse PFS (HR 3.720, 95% CI 1.823–7.594, p = 0.0003) and worse OS (HR 3.427, 95% CI 1.698–6.917, p = 0.001). In contrast, there was no significant association between peripheral basophil counts and tumor response or survival in the chemotherapy-alone group (p > 0.05). In primary gastric cancer samples, we observed a correlation between higher peripheral basophil counts and the accumulation of tumor-infiltrating basophils (r = 0.6833, p = 0.005). Tumor-infiltrating basophils were found to be spatially proximate to M2 macrophages within TME and positively correlated with tumor M2 macrophage infiltration (r = 0.7234, p = 0.0023). The peripheral basophil counts also had a significant positive correlation with tumor-infiltrating M2 macrophage counts (r = 0.6584, p = 0.003). Further validation in tumor samples treated with the neoadjuvant anti-PD-1 inhibitor plus chemotherapy combination suggests that the peripheral basophils, tumor infiltration of basophils, and M2 macrophages were significantly more abundant in non-responders than in responders (p = 0.0333, p = 0.0007, and p = 0.0066, respectively). Conclusions The peripheral basophil count was observed to be a potential biomarker of anti-PD-1 efficacy for advanced gastric cancer. Moreover, basophils may induce an immune-evasive tumor microenvironment by increasing M2 macrophage infiltration, which could be a potential immunotherapeutic target for advanced gastric cancer.