Otsuka files to convert VOYXACT to full approval in IgA nephropathy, citing a 5.5 mL/min interim eGFR edge at one year

Six months after winning accelerated U.S. approval on a proteinuria surrogate, Otsuka is trying to lock in sibeprenlimab (VOYXACT, sibeprenlimab-szsi) with the number that matters more to nephrologists: preserved kidney function. At the European Renal Association (ERA) Congress in Glasgow on June 4, the company reported interim 12-month estimated glomerular filtration rate (eGFR) data from the Phase 3 VISIONARY trial (NCT05248646) and said it has begun a rolling supplemental Biologics License Application (sBLA) seeking traditional approval.

VISIONARY is a randomized, double-blind, placebo-controlled study of sibeprenlimab 400 mg given subcutaneously every four weeks against placebo in biopsy-confirmed IgA nephropathy, per ClinicalTrials.gov. The trial’s primary endpoint is the relative change in urinary protein-to-creatinine ratio at nine months — the basis for the accelerated approval granted November 25, 2025. The annualized eGFR slope over 24 months is a prespecified secondary endpoint (Otsuka calls it the key secondary); the registry also lists mean eGFR change from baseline over 24 months as a separate secondary. As of this writing, ClinicalTrials.gov shows no posted results for the trial.

What the interim showed

The figures below come from Otsuka’s ERA 2026 presentation as relayed in the company’s press release; they are interim secondary-endpoint data, not a peer-reviewed paper, and the registry has not posted results. In the interim analysis (320 randomized; 152 sibeprenlimab, 168 placebo), two distinct 12-month eGFR endpoints were reported.

On total eGFR change from baseline at 12 months, the treated arm was essentially flat (+0.7 mL/min/1.73 m²; 95% CI −0.9 to 2.3), while placebo fell 4.8 points (−4.8; 95% CI −6.3 to −3.3) — a between-arm difference of 5.5 (95% CI 3.4 to 7.6). Note that the drug arm’s own confidence interval crosses zero, so the result rests on the between-arm comparison, not on an individually significant gain.

On the annualized eGFR slope — a different metric over the same 12 months — both arms were still declining: −3.0 mL/min/1.73 m²/year (95% CI −4.6 to −1.4) on sibeprenlimab versus −7.6 (95% CI −9.1 to −6.1) on placebo, a slope difference of 4.6 per year (95% CI 2.5 to 6.8). In other words, the “flat” characterization applies to the total change-from-baseline endpoint, not to the slope, where the treated arm continued to lose function more slowly than placebo rather than holding steady.

The total-change endpoint looks flat; the slope endpoint still declines. They are not the same number, and only one supports “preserved.”

Sibeprenlimab is a monoclonal antibody targeting APRIL, upstream of the pathogenic IgA that drives the disease. On safety, infections occurred in 49% of sibeprenlimab-treated versus 45% of placebo-treated patients, and injection-site reactions in 24% versus 23% — rates similar to placebo.

According to Otsuka’s release, the rolling sBLA for traditional approval already “includ[es] data from the 24-month eGFR endpoint.” That framing sits somewhat uneasily next to VISIONARY’s design as a roughly 24-month study, and none of the 24-month results have been published or posted to the registry. What remains genuinely outstanding, then, is not whether a 24-month analysis exists but FDA review of the conversion filing and independent, peer-reviewed disclosure of the full dataset.