Background: Immune checkpoint inhibitors (ICIs) have transformed the management of solid tumors, yet only a subset of patients achieve durable benefit. The gut microbiota is a key modulator of antitumor immunity, and systemic antibiotic therapy (ABT), frequently prescribed—and sometimes overused—in oncology, can profoundly disrupt microbial homeostasis. Observational studies suggest that ABT may impair ICI efficacy, but results remain heterogeneous, warranting an updated synthesis. Methods: We conducted a systematic review and meta-analysis in accordance with the PRISMA 2020 guidelines. PubMed, Scopus, and EMBASE were searched for studies published between 2018 and 2025 evaluating the association between ABT exposure and time-to-event outcomes in patients with solid tumors treated with ICIs. Studies were required to report explicit definition of the ABT exposure window. Random-effects models were considered primary. A sensitivity analysis was performed in non-small cell lung cancer (NSCLC). Results: Fifteen studies encompassing 52,489 patients were included. ABT exposure was associated with significantly worse OS (random-effects HR 1.16, 95% CI 1.03–1.29) and PFS (random-effects HR 1.11, 95% CI 0.95–1.27), indicating an increased risk of death and disease progression compared with no ABT exposure. In the NSCLC sensitivity analysis, ABT was consistently associated with inferior PFS and, when accounting for heterogeneity, with significantly reduced OS, supporting the robustness of the association. Conclusions: ABT administered in temporal proximity to ICIs is associated with clinically meaningful worsening of survival outcomes across solid tumors, consistent with microbiome-mediated impairment of immunotherapy efficacy. These findings support cautious ABT stewardship in patients receiving ICIs and highlight the need for prospective studies integrating microbiome profiling and standardized ABT exposure assessment.

Real-world evidence on the influence of sex, age, and tumour type on pembroli-zumab-associated adverse reaction reporting remains limited. We retrospectively ana-lysed individual case safety reports (ICSRs) submitted to EudraVigilance between 1 January and 31 December 2025 involving pembrolizumab monotherapy. Adverse re-actions were classified according to MedDRA System Organ Class (SOC), and toxicity burden was defined as single- versus multiple-SOC involvement. Associations were assessed using Pearson’s chi-square test and Cramér’s V; sex comparisons across nine SOCs used a Bonferroni-adjusted threshold of p< 0.0056. Overall, 8,488 ICSRs were an-alysed; sex was available in 7,991 reports, age in 6,941, and 6,482 were evaluable across 14 tumour types. Overall toxicity burden did not differ by sex (p=0.110). Endocrine (18.4% vs 12.4%) and blood and lymphatic disorders (13.4% vs 9.5%) were more frequent in women, whereas respiratory (12.4% vs 7.0%) and cardiac disorders (13.9% vs 11.5%) were more frequent in men, all remaining significant after Bonferroni correction. However, none persisted after adjustment for tumour histotype and age (adjusted ORs: 1.06, 0.85, 1.22 and 1.02, respectively). Multiple-SOC reporting differed modestly by age, being lower in individuals >85 years (39.9% vs 47.7% and 48.1%; p=0.026), and by tumour type (p=0.016). Effect sizes were small. Observed sex-related reporting patterns may therefore largely reflect tumour-type composition rather than an independent sex effect. Findings are exploratory and require confirmation in denominator-based prospective studies.