AbstractAntidiabetic medication may modify the incidence of hepatocellular carcinoma (HCC). We aimed to compare the use of different antidiabetic strategies and the incidence of HCC. PubMed, Embase.com and Cochrane Library databases were searched up to 31 October 2015 and randomized controlled trials (RCTs), cohort studies or case-control studies were included for our analyses. A total of thirteen studies enrolling 481358 participants with 240678 HCC cases who received at least two different strategies were retrieved in this analysis. Direct comparisons showed that use of metformin (risk ratio [RR] 0.49, 95% CI 0.25–0.97) was associated with a significant risk reduction of HCC, while insulin (RR = 2.44, 95% CI 1.10- 5.56) may significantly increase the risk. Indirect evidence also suggested that insulin (RR = 2.37, 95% CI 1.21–4.75) was associated with a significantly increased risk of HCC. Additionally, metformin was effective in reducing the risk of HCC when compared with sulphonylurea (RR = 0.45, 95% CI 0.27–0.74) and insulin (RR = 0.28, 95% CI 0.17–0.47). Notably, metformin was hierarchically the best when compared with other antidiabetic therapies for the prevention of HCC. In summary, available evidence suggests that metformin was the most effective strategy to reduce HCC risk when compared with other antidiabetic interventions.