Nowadays, it has become clear that extracellular vesicles (EVs) are not a cellular waste disposal vesicle but are an essential part of an intercellular communication system. Besides the use of EVs in biomarker studies and diagnostics, the potential of EV‐therapeutics has been seen by many. They provide unique properties for disease therapy, including strong immune‐modulatory actions, the possibility of engineering, low immunogenicity, and the capability of crossing biological barriers. Proof‐of‐concept of EV‐therapeutics for various pathologies has been achieved in preclinical studies. However, clinical trials with EVs have only been emerging slowly. Here, we aim to provide a comprehensive overview of the current state‐of‐the‐art concerning clinical studies using EVs in human therapy. By approaching the current knowledge in a systematic manner, we were able to include 21 reports for meta‐analysis of safety and evaluation of efficacy outcomes. Overall, we have shown that EV‐based therapy is safe with a low incidence of serious adverse events (SAE; 0.7% (95%‐CI: 0.1–5.2%), and adverse events (AE; 4.4% (95%‐CI: 0.7–22.2%). Subgroup analysis showed no significant difference in SAE when comparing autologous versus allogeneic administration, as well as engineered versus non‐engineered EV products. A significantly higher number of AE was seen in autologous versus allogeneic administration. However, the clinical relevance remains questionable. Evaluation of the clinical outcomes of immunostimulatory, immunosuppressive or regenerative EV‐therapies indicated improvement in the majority of treated patients. Despite these promising results, data need to be approached with caution due to a high heterogeneity in the EVs manufacturing methods, study design, and reporting of (S)AE. Overall, we conclude that EV‐based therapy is safe and presents a promising opportunity in therapy. More efforts are needed in the standardization and harmonization of reporting of EV isolation and characterization data as well as in the reporting of (S)AE to allow inter‐study comparison.