Background: Numerous clinical trials show crizotinib has promising efficacy for anaplastic lymphoma kinase (ALK) positive non-small cell lung cancer (NSCLC) patients which trigger the substitution of traditional chemotherapy to be the current standard first-line treatment for these patients. Conversely, few reports systematically analyze toxicity of crizotinib. Hence, we performed a first meta-analysis to determine the risk of crizotinib-related severe adverse events (SAEs) and fatal adverse events (FAEs) in ALK positive NSCLC patients. Materials and Methods: A systematic literature search was conducted through December 2016 to identify clinical trials that reported crizotinib monotherapy in ALK-positive NSCLC patients. Data on crizotinib-related SAEs and FAEs were extracted from each study and pooled to determine the overall incidence and risk. Random-effects or fixed-effects models were conducted to calculate the summary incidence, relative risk (RR), and 95% CIs on basis of the heterogeneity of included studies. Results: 1,924 patients from 11 clinical trials were included. The overall incidence of SAEs and FAEs with crizotinib was 19.9% (95% CI, 14.1% to 23.7%; P < 0.001) and 1.4% (95% CI, 0.9% to 2.1%; P < 0.001), respectively. Meanwhile, Asian patients have lower incidence of SAEs (11.5%, 95% CI: 7.9% to 16.5%). However, significant differences of SAEs (RR: 0.97, 95% CI, 0.79 to 1.18; P = 0.76) and FAEs (RR: 2.24, 95% CI, 0.49 to 10.30; P = 0.30) were not detected between crizotinib monotherapy and chemotherapy. Conclusions: Crizotinib may not increase the risk of SAEs and FAEs in patients with ALK positive NSCLC compared with chemotherapy.