Relative efficacy and safety of tacrolimus, mycophenolate mofetil, and cyclophosphamide as induction therapy for lupus nephritis: a Bayesian network meta-analysis of randomized controlled trials

Aims This study aimed to assess the relative efficacy and safety of tacrolimus, mycophenolate mofetil (MMF) and cyclophosphamide (CYC) as induction therapy for lupus nephritis. Methods Randomized controlled trials (RCTs) examining the efficacy and safety of tacrolimus, MMF and CYC for induction therapy in patients with lupus nephritis were included. We performed a Bayesian random-effects network meta-analysis to combine direct and indirect evidence from the RCTs. Results Nine RCTs including 972 patients met the inclusion criteria and pair-wise comparisons were performed, including 11 direct comparisons. Tacrolimus showed a significantly higher overall response rate (complete remission plus partial remission) than CYC (OR 2.35, 95% confidence interval (CI) 1.03-5.45), and was more efficacious than MMF (OR 1.60, 95% CI 0.70-3.57). MMF was superior to CYC in terms of overall response (OR 1.45, 95% CI 0.96-2.42). Ranking probability based on the surface under the cumulative ranking curve (SUCRA) indicated that tacrolimus had the highest probability of being the best treatment for achieving the overall response (SUCRA=0.9321), followed by MMF (SUCRA=0.5385) and CYC (SUCRA=0.0294). In terms of safety, tacrolimus showed the highest probability of decreasing the risk of serious infections (SUCRA=0.9253), followed by MMF (SUCRA=0.4027) and CYC (SUCRA=0.1720). Conclusions Tacrolimus was the most efficacious induction treatment for patients with lupus nephritis, and had the highest probability of decreasing the risk of serious infections. Higher remission rates combined with a more favorable safety profile suggest that MMF is superior to CYC as induction treatment in these patients.