Tacrolimus matched mycophenolate for lupus nephritis over 96 weeks, with a safety twist

Treating active lupus nephritis with tacrolimus plus a steroid worked about as well over nearly two years as the standard combination of mycophenolate plus a steroid, a randomised trial has found — but the two regimens carried different risks, and all three deaths in the study occurred among patients taking mycophenolate [s1]. The result adds a head-to-head comparison to a field where the choice between these drugs has rested largely on indirect evidence [s1].

Lupus nephritis is inflammation of the kidneys driven by systemic lupus erythematosus, and the more severe forms can progress to kidney failure if the immune attack is not controlled. Mycophenolate is a widely used standard of care; tacrolimus, a calcineurin inhibitor, is an alternative, and both are given alongside glucocorticoids [s1]. Guidelines increasingly describe the two as interchangeable first-line options, or combine them, but the evidence for that has leaned heavily on short induction studies and indirect comparisons rather than a sustained head-to-head over the two years it takes to judge whether a kidney response holds [s1].

Most lupus-nephritis trials have also been dominated by Western populations, even though the disease is more common and often more aggressive in people of Asian and other ancestries — one reason a trial run across Asian centres adds something the earlier literature lacks [s1].

What the trial did

The study was a prospective, multicentre randomised controlled trial conducted across Asian centres, with the University of Hong Kong as sponsor [s1][s2]. It enrolled 130 patients with biopsy-proven Class III or IV lupus nephritis, with or without additional Class V disease, and randomly assigned them to one of two regimens as continuous induction-and-maintenance therapy for 96 weeks: 65 to tacrolimus (targeting a trough level of 6–8 ng/mL) and 65 to mycophenolate (1 g twice daily) [s1]. Both groups received the same glucocorticoid backbone — intravenous methylprednisolone for three days followed by oral prednisolone [s1].

The primary endpoint was a sustained renal response at week 96, defined strictly as a reduction in proteinuria of more than 50%, an estimated glomerular filtration rate of at least 60 mL/min per 1.73 m², and no disease flare or need for rescue therapy over the period [s1].

What it found

At week 96 there was no statistically significant difference in the primary endpoint: a sustained renal response was achieved by 58.5% of the tacrolimus group and 69.2% of the mycophenolate group [s1]. Complete renal remission followed the same pattern, at 55.4% versus 63.1%, as did the earlier read at week 48, where sustained response rates were 63.1% and 66.2% respectively [s1]. In each case the numerically higher figure sat with mycophenolate, but the trial — with 65 patients per arm — was not large enough for those gaps to reach significance [s1].

The safety picture was where the two drugs diverged. Overall adverse-event rates were identical at 80.0% in both groups, but the pattern differed: acute kidney injury and tremor were more common with tacrolimus, while low white-cell counts were more common with mycophenolate [s1]. More consequentially, treatment-related serious adverse events and serious infections were significantly more frequent in the mycophenolate group, and all three deaths in the trial occurred in that group [s1].

How to read it

This is a null result on efficacy — the trial did not show one regimen cured more kidneys than the other — paired with a safety signal that cuts the other way than the efficacy numbers [s1]. Both findings need caution. With 130 patients, the study was underpowered to exclude a real efficacy difference, so the slightly higher response rates with mycophenolate cannot be dismissed, and three deaths is too few to draw firm conclusions about mortality [s1]. The open-label elements and the single-region, predominantly Asian population also limit how widely the result generalises, since responses to lupus-nephritis therapy vary by ancestry [s1].

What the trial does establish is that tacrolimus plus glucocorticoid is a reasonable comparator to the mycophenolate standard over a sustained period, with a different — not uniformly better or worse — balance of harms [s1]. For a clinician, that reframes the choice as one about which risks a particular patient can best tolerate: the acute kidney injury and tremor that came more often with tacrolimus, or the low white-cell counts and the serious infections that were significantly more frequent with mycophenolate [s1].

What to watch

Larger or pooled analyses will be needed to settle whether the efficacy gap is real and to confirm the infection and mortality signal seen here [s1]. This article describes research and is not medical advice; treatment of lupus nephritis is a decision for treating clinicians.

Sources

• Tacrolimus versus mycophenolate as induction-maintenance for lupus nephritis over 96 weeks — Kidney International, 29 September 2026
• Trial registration, NCT02630628 — ClinicalTrials.gov