The advent of biologically targeted agents and increased understanding of thyroid carcinogenesis have generated much interest in the development of biologically targeted therapeutic agents for thyroid cancer. Among them, sorafenib is the most commonly studied drug. The current meta-analysis was carried out to estimate the efficacy and safety of sorafenib administered in radioiodine-refractory differentiated thyroid cancer patients. An electronic search was conducted using PubMed/MEDLINE and EMBASE. Statistical analyses were carried out using either random-effects or fixed-effects models according to heterogeneity. All the statistical analyses were carried out using the Stata version 12.0 software. Seven eligible studies were identified. The final results indicated that 22% of the patients (95% CI: 15-28) achieved a partial response. Hand-foot syndrome, diarrhea, fatigue, rash, weight loss, and hypertension were the most frequently observed adverse effects (AEs) associated with sorafenib use and the incidence of these AEs (all grades) was 80% (95% CI: 68-91), 68% (95% CI: 59-77), 67% (95% CI: 57-78), 66% (95% CI: 50-82), 52%(95% CI: 33-72), and 31% (95% CI: 21-42) respectively. Sixty-two percent (95% CI: 36-89) patients required dose reductions due to toxicity of sorafenib. As far as PR and AEs are concerned, the results of this meta-analysis indicate that sorafenib has a modest effect in patients with radioiodine-refractory differentiated thyroid cancer and the high incidence of AEs associated with this agent may affect the quality of patients' lives. Though the use of sorafenib in the treatment of radioiodinerefractory differentiated thyroid cancer is considered promising by most physicians working in this field, more effective agents with less toxicity and cost are still needed.