Design and Reporting Quality of Noninferiority Trials in Kidney Transplantation: A Systematic Review.

Abstract

Key points: Many kidney transplant noninferiority trials used unjustified or overly generous margins, risking adoption of inferior therapies. Inadequate analyses, poor missing data handling, and weak Consolidated Standards of Reporting Trials (CONSORT) adherence limit trial interpretation. Stronger noninferiority trial design and reporting are needed to improve evidence quality in kidney transplantation.

Background: Noninferiority trials are essential in kidney transplantation, where established standard-of-care treatments make placebo-controlled designs infeasible, and their use is increasing. We aimed to assess the design and reporting quality of noninferiority trials in kidney transplantation, specifically assessing adherence to existing guidelines.

Methods: We searched medical literature analysis and retrieval system online, Embase, and Cochrane Central Register of Controlled Trials for randomized controlled noninferiority trials in kidney transplantation published up to February 2025. Two investigators reviewed and extracted data. Study quality was assessed regarding design and reporting quality using key noninferiority design criteria and the Consolidated Standards of Reporting Trials 2012 extension, respectively.

Results: We identified 43 unique trials: 36 (84%) compared a novel immunosuppressive regimen to standard of care and 26 (60%) were industry sponsored. The noninferiority margin (the worst-case loss in efficacy considered clinically acceptable) was not justified in 28 (65%) studies and poorly justified in seven (16%). Of 32 studies with a categorical primary outcome, 30 (94%) assessed absolute risk reduction rather than relative risk. Most studies had high withdrawal and dropout rates, yet 33 (77%) did not specify methods for handling missing data. Intention-to-treat analysis alone was used to determine primary efficacy in 24 (56%) studies. Twelve studies (28%) provided an incorrect conclusion, most claiming equivalence when noninferiority had not been adequately demonstrated. Both design and reporting quality have improved over time ( P = 0.02, P = 0.002, respectively). No association was evident between design or reporting quality and journal impact factor or sponsor type.

Conclusions: We found major deficiencies in the design, conduct, and reporting of noninferiority trials in kidney transplantation.