Avacopan: enzyme rises uncommon, but 2 of 238 US patients had highly probable liver injury, one severe

Design

Retrospective US registry cohort with RUCAM causality assessment

Population

238 patients with ANCA-associated vasculitis prescribed avacopan

Primary outcome

Liver enzyme elevations and drug-induced liver injury

Effect

AST >3× ULN 4.24 per 100 PY; DILI 2 of 238 (0.8%), both highly probable

This study used the US Rheumatology Informatics System for Effectiveness (RISE) registry to identify 238 patients with ANCA-associated vasculitis prescribed avacopan between October 2021 and March 2025. Liver enzyme elevations were defined by FDA thresholds and assessed for causality using RUCAM.

Rises were uncommon: AST above three times normal at 4.24 per 100 person-years, ALT 2.67, alkaline phosphatase 3.56 and bilirubin 1.80. Two patients (0.8%) had drug-induced liver injury with RUCAM scores of 9 (highly probable), one with biopsy-confirmed severe hepatocellular injury. The first liver test came a median 32 days after starting, with tests about every 125 days thereafter.

The context matters. According to the authors, after 76 worldwide reports of drug-induced liver injury, 66 of them from Japan, the FDA requested market withdrawal of avacopan, which the manufacturer declined, and regulatory reassessment is ongoing. This is the first quantification of the risk in US practice.

The monitoring gap is the actionable finding: testing every four months will miss an injury that develops within weeks.

• Check liver function before starting avacopan and frequently in the first months, not every four months.
• Tell patients to report jaundice, dark urine, itching or abdominal pain immediately.
• Stop avacopan promptly if enzymes rise significantly and assess causality.
• Discuss the unresolved regulatory position when considering avacopan for new patients.
• Review other hepatotoxic drugs, such as co-trimoxazole prophylaxis, at the same time.

Why it matters

Severe liver injury does occur in routine practice, and current monitoring intervals are too long to catch it early.

Don't overread it

With 238 patients and two cases, the true incidence is uncertain, and concurrent drugs in vasculitis make attribution difficult.

The statistics, in plain English

Rates per 100 person-years with confidence intervals spanning roughly tenfold, such as 0.86 to 8.27 for ALT, show how few events there were. Two cases in 238 patients could correspond to a true rate well below or above 1%.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.