Supplementary materials: Matching-adjusted indirect comparison of kidney function in patients with immunoglobulin A nephropathy treated with nefecon or sparsentan

These are peer-reviewed supplementary materials for the article Matching-adjusted indirect comparison of kidney function in patients with immunoglobulin A nephropathy treated with nefecon or sparsentan' published in the Journal of Comparative Effectiveness Research.Supplementary methodsSupplementary Table S1: Patient baseline characteristics used for the MAICs.Supplementary Table S2: Input data for the network meta-analysis.Supplementary Table S3: Results from matching-adjustment for the unanchored MAIC.Supplementary Table S4: Results of the fixed-effects network meta-analysis.Supplementary Figure S1: Distribution of weights derived for the anchored MAIC.Supplementary Figure S1: Distribution of weights derived for the anchored MAIC.Supplementary Figure S2: Distribution of weights with nefecon (A) and placebo (B) in NefIgArd, derived for the unanchored MAIC.Supplementary Figure S3: Forest plot for the unanchored MAIC for eGFR (A), UPCR (B), and UACR (C) at different time points; forest plot for the unanchored MAIC for time to confirmed 40% eGFR reduction, ESKD, or all-cause mortality (D).Supplementary ReferencesAim: We compared the effects of nefecon, an oral targeted-release budesonide formulation, and sparsentan, an oral, dual endothelin-angiotensin receptor antagonist, on estimated glomerular filtration rate (eGFR) in patients with immunoglobulin A nephropathy, a leading cause of chronic kidney disease. Materials & methods: We conducted an anchored matching-adjusted indirect comparison (MAIC) using patient-level data from NefIgArd (NCT03643965; n = 364), a randomized (1:1) trial of nefecon plus optimized renin–angiotensin system inhibitor (RASi) therapy versus placebo plus RASi; and aggregate data from PROTECT (NCT03762850; n = 404), a randomized (1:1) trial of sparsentan versus irbesartan, an angiotensin receptor blocker. Mean absolute eGFR change and mean relative urine protein-to-creatinine and urine albumin-to-creatinine ratio changes from baseline at 9, 12 and 24 months (NefIgArd) or 36, 48 and 106 weeks (PROTECT) were analyzed using amixed-effects model for repeated measures. A composite outcome (time to confirmed 40% eGFR reduction, end-stage kidney disease or all-cause mortality) was also included. An unanchored MAIC and network meta-analysis were used as sensitivity analyses. Results: The matching process reduced the effective sample for the NefIgArd trial from 364 to 208. Absolute eGFR change significantly favored nefecon over sparsentan at 9 months (mean difference,ml/min/1.73m2 [95% credible interval]: 5.7 [3.1–8.2]), 12 months (3.5 [1.0–6.0]) and 24 months (3.3 [0.0–6.5]). Differences in other outcomes were generally not statistically significant. Sensitivity analysis results were consistent with the main findings. Conclusion: In patients with immunoglobulin A nephropathy, nefecon plus optimized RASi may preserve kidney function to a greater extent than sparsentan.