Severe Renal ANCA Vasculitis: Steroid-Sparing After ADVOCATE's Retraction
A single patient, just started on standard AAV induction. The disagreement isn't whether steroid-sparing is worth pursuing — it's whether the drug built for it still has an evidence base to stand on, and what to do in the meantime.
Deborah N. noticed nothing wrong at all until a routine blood pressure check at a pharmacy flu-shot clinic came back alarmingly high, which is what finally got her into an urgent-care visit that found a creatinine of 3.4 — a number that sent her straight to the hospital rather than home with a prescription. MPO-ANCA came back strongly positive within two days; the kidney biopsy, obtained the same admission, showed crescentic glomerulonephritis involving 60% of glomeruli, severe by any threshold. Rituximab and a standard prednisone taper were started immediately, before the team had fully worked through whether avacopan belonged in the regimen from day one or as a later addition.
ADVOCATE is the trial her case would have turned on a year ago: avacopan, an oral C5a receptor antagonist, tested against a standard prednisone taper on the same background induction, reported as noninferior for remission at week 26 and superior for sustained remission at week 52, with meaningfully less glucocorticoid toxicity along the way. That evidence base has since come apart in a way that reaches the bedside rather than staying in the literature. An FDA investigation found that primary-endpoint assessments in nine of the trial's 331 patients were re-adjudicated after database lock and unblinding — five in the avacopan arm, all moved toward sustained remission, which is what converted a non-superior week-52 result into the superiority finding. The New England Journal of Medicine retracted the paper on June 29, 2026; the FDA proposed withdrawing the drug in April and it remains on the US market only while that hearing runs; the European Commission revoked the EU authorization outright in August. A separate FDA safety communication has described serious drug-induced liver injury, including vanishing bile duct syndrome. What keeps this a live question rather than a closed one is that a blinded independent re-adjudication by the Duke Clinical Research Institute reproduced both noninferiority at week 26 and a sustained-remission difference at week 52 — so the effect may well be real even though the record of it is not. Her creatinine of 3.4 with sixty percent crescents places her in the severe-renal group where post hoc ADVOCATE analyses described the largest eGFR recovery — and equally in the group with the least margin for guessing wrong in either direction.
On rounds, day 3 of induction
The goal is right — she has severe renal involvement and a long taper ahead, and cumulative glucocorticoid toxicity is precisely what we should be trying to avoid. But I'm not starting avacopan today. The New England Journal of Medicine retracted ADVOCATE in June, and the specific finding we'd be leaning on, superiority for sustained remission at week 52, is the finding the post-unblinding re-adjudication produced. That's not a reason to start something new in a patient already three days into a regimen that's working.
I'd push back on treating the retraction as though it settled the pharmacology. The Duke Clinical Research Institute re-adjudication was blinded and independent, and it reproduced both noninferiority at week 26 and a sustained-remission advantage at week 52 — 61.4 percent against 52.4 percent. Same direction, from the analysis done properly. The drug is still approved here, and KDIGO's 2024 AAV guidance still has a place for it.
Saying the evidence “came apart” runs two different failures together — the conduct was indefensible, and the effect estimate survived being redone honestly. Only one of those is a claim about whether the drug works in her.
You're both arguing about avacopan when the decision in front of us doesn't require it. PEXIVAS randomized reduced-dose glucocorticoids against the traditional taper in severe AAV — a population that included patients at her creatinine — and found the reduced-dose regimen noninferior for death or end-stage kidney disease with meaningfully fewer serious infections. That is a steroid-sparing intervention with an evidence base nobody is contesting, and she is three days into a standard taper we can simply convert. I'd also note the renal-severity signal you're citing comes from post hoc analyses, not a prespecified endpoint.
There's a safety asymmetry worth naming too. The FDA's March safety communication described serious drug-induced liver injury with avacopan, including vanishing bile duct syndrome. When we can spare her steroids by a route carrying none of that, I'd rather not spend the uncertainty here. If she relapses, or can't be tapered, avacopan is still there to argue about — by then with a hearing record we don't have yet.
Agreed: convert the standard prednisone taper to the PEXIVAS reduced-dose regimen rather than starting avacopan today, with renal function and urine sediment rechecked weekly for the first month. Rituximab continues unchanged.
Not agreed, and left standing rather than smoothed over: whether the Duke re-adjudication repairs ADVOCATE enough to prescribe from. The nephrologist's position — that a blinded independent redo reproducing the same direction is evidence about the drug, whatever the conduct was — was not overruled, and nobody claimed the retraction proves avacopan does not work. What carried today's decision was narrower: an equivalent steroid-sparing effect is available by a route with an uncontested trial behind it and no liver signal, so the uncertainty does not have to be spent here. If she relapses or cannot be tapered, the question returns — by then with an FDA hearing record that does not yet exist.