Statistical Analysis Plan for the Comparative Effectiveness of an Individualized Model of Hemodialysis vs Conventional Hemodialysis: the TwoPlus Trial

<title>Abstract</title> <p> <bold>Background:</bold> At least 25% of patients who start chronic hemodialysis (HD) have appreciable residual kidney function, yet are prescribed conventional thrice-weekly treatment, as those who have minimal or no residual kidney function. We hypothesize that an incremental approach—starting with less frequent treatment and guided by residual function and clinical status—is as safe as conventional HD. <bold>Methods:</bold> This unblinded, parallel-group, individually-randomized trial will test the non-inferiority of clinically matched incremental HD (CMIHD)—twice-weekly HD with medications, progressing to thrice-weekly based on clinical manifestations and/or declining residual kidney function—versus conventional thrice-weekly HD (CHD). Patients are randomized 1:1 to CMIHD or CHD. The primary outcome is a composite of all-cause mortality, emergency department visits, and hospitalizations. Secondary outcomes include quality of life, residual kidney function, functional status, volume management, hematologic, and biochemical parameters. <bold>Results:</bold> We present the <bold/> statistical analysis plan detailing the estimand framework, analyses to be performed for both primary and secondary outcomes, rationale for the modeling framework, and planned sensitivity analyses. We specify intention-to-treat, per-protocol, and as-treated analyses to address adherence and transition from twice- to thrice-weekly HD; approaches to evaluate the impact of non-adherence to recommended dialysis prescription; and sensitivity analyses using alternative models. We outline strategies for handling missing data and multiplicity adjustment. We discuss potential post-hoc conditional power analyses to quantify the impact of differently-achieved final sample size and deviations from design assumptions on non-inferiority power. <bold>Discussion:</bold> The TwoPlus trial will determine whether CMIHD is non-inferior to CHD on safety outcomes in patients with residual kidney function. Simulations confirmed adequate power for the primary composite outcome and robustness of the non-inferiority margin. The plan was developed before the trial database was locked. </p>