Avacopan: Drug-Induced Liver Injury and Vanishing Bile Duct Syndrome

Avacopan is an oral small-molecule antagonist of the complement component 5a receptor (C5aR1, also called CD88). It was developed for adults with severe active anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis, where it was intended to reduce glucocorticoid exposure while being used with other standard immunosuppressive treatment. Its pharmacovigilance history now illustrates two different evidence problems: a clinically important hepatic safety signal, and a separate loss of confidence in the pivotal efficacy data.

This article follows the medicine from its discovery and clinical programme to the safety and regulatory developments reported by 3 October 2026. FDA’s post-marketing case review, FDA’s pending US withdrawal proposal, the European Commission’s completed revocation, and the UK MHRA’s managed suspension have different scopes and legal status. They should not be collapsed into one global decision.

Why avacopan became a pharmacovigilance case study

The liver concern did not appear from nowhere after launch. Hepatotoxicity was already identified in clinical development and described as a serious adverse reaction in product information. Post-marketing reports later added a more specific and severe phenotype: drug-induced liver injury (DILI), often cholestatic or mixed, including cases with biopsy-confirmed vanishing bile duct syndrome (VBDS) and fatal outcomes. The signal therefore concerns both the progression of a known risk and the appearance of a new clinical pattern.

The benefit side also changed. In 2026, regulators questioned the integrity and reliability of the ADVOCATE pivotal-trial data. FDA proposed withdrawal of the US approval; the EU marketing authorisation was subsequently revoked; and the UK suspended use for new patients while arranging a transition for existing patients. These efficacy-data actions are not findings about liver causality. They matter to the integrated benefit–risk assessment because a serious risk is harder to justify when the evidence supporting benefit cannot be relied upon.

Molecule, mechanism and development rationale

ChemoCentryx discovered and developed avacopan under the code CCX168. The scientific rationale was to interrupt C5a signalling through C5aR1, a pathway implicated in neutrophil activation and inflammatory vascular injury in ANCA-associated vasculitis. Avacopan blocks C5aR1; it does not inhibit C5 cleavage or remove the need for background immunosuppressive treatment. The proposed therapeutic idea was targeted control of a disease-relevant inflammatory pathway, with the possibility of reducing glucocorticoid exposure.

The published preclinical and first-in-human pharmacology programme described CCX168 as a potent, selective, orally administered C5aR antagonist. It included receptor-binding and functional assays, pharmacology models and a randomised Phase 1 study in healthy volunteers. This established human exposure and pharmacodynamic activity, not clinical efficacy in vasculitis. The originating company was ChemoCentryx; the public primary literature does not identify a single individual inventor whose attribution is needed to explain the product’s origin. [1]

Clinical development and ownership

The programme advanced from early pharmacology into a Phase 2 proof-of-concept study and then the Phase 3 ADVOCATE trial. The intended clinical role was adjunctive: avacopan was tested with rituximab or a cyclophosphamide-based regimen, not as stand-alone induction therapy. The pivotal programme compared a 52-week avacopan course with a 20-week prednisone taper, alongside standard immunosuppression. These distinctions matter when interpreting the original treatment claim and its later reassessment.

In 2016, ChemoCentryx licensed Vifor (International) exclusive commercialisation rights for Europe and certain other markets; the territory was later expanded. ChemoCentryx retained the US development and commercial role. Amgen completed its acquisition of ChemoCentryx on 20 October 2022, after avacopan had received US approval. Vifor Fresenius Medical Care Renal Pharma France was the EU marketing-authorisation holder. These arrangements explain why the development company, US application holder and EU authorisation holder appear under different names in regulatory records. [2, 3]

From Phase 2 to authorisation

The early CLEAR study provided Phase 2 evidence that supported further development of C5aR blockade in ANCA-associated vasculitis. It was a relatively small exploratory programme and could not establish the full benefit–risk profile expected from a confirmatory trial. ADVOCATE was the pivotal Phase 3 study. It enrolled 331 patients with granulomatosis with polyangiitis (GPA) or microscopic polyangiitis (MPA), and compared avacopan 30 mg twice daily for 52 weeks with a tapering prednisone regimen. Both groups also received background treatment with rituximab or cyclophosphamide followed by azathioprine. [4, 5]

The 2021 publication reported that avacopan was noninferior to prednisone for remission at week 26 and statistically superior for sustained remission at week 52. Those reported results supported the initial regulatory case, but the published ADVOCATE paper was retracted in June 2026 after concerns about trial conduct and altered outcome data. A retracted article remains important to the historical record, but its efficacy estimates must not be presented as reliable current evidence. FDA, EMA and MHRA describe their regulatory conclusions and the basis for their actions; the company disputes FDA’s proposed withdrawal and submitted additional materials for a hearing. [5–8]

FDA approved avacopan on 7 October 2021 as adjunctive treatment for adults with severe active ANCA-associated vasculitis, including GPA and MPA, in combination with standard therapy. FDA’s prescribing information states that it does not eliminate glucocorticoid use. In the EU, CHMP issued a positive opinion in November 2021; the European Commission granted marketing authorisation on 11 January 2022. The product had entered EMA’s PRIME scheme during development. The EU indication specified use in combination with rituximab or cyclophosphamide. The EU authorisation was revoked in August 2026, as described below. [9, 10]

The hepatic risk before and after launch

Before the post-marketing signal became prominent, hepatic adverse reactions were already recognised. The US label describes serious cases of drug-induced liver injury during clinical trials and includes baseline and follow-up liver testing, laboratory thresholds for interruption or discontinuation, and advice on symptoms such as jaundice and pruritus. Post-marketing VBDS reports led to further explicit warnings and a requirement to stop treatment permanently if VBDS is suspected. Therefore, the 2026 issue is not simply “a new liver risk”: it is a more severe and clinically specific expression of a known hepatic concern. [11]

VBDS is a histopathological syndrome involving loss or disappearance of intrahepatic bile ducts. It can produce prolonged cholestasis and progressive liver dysfunction, and may persist even after the suspected drug is stopped. It is not synonymous with every episode of cholestatic DILI. A case should be classified as VBDS only when the evidence supports that diagnosis, such as biopsy findings interpreted with the clinical course. The mechanism by which avacopan may cause this uncommon phenotype is not established in the cited regulatory assessments; a pharmacovigilance review should describe the observed pattern without presenting a speculative mechanism as fact.

On 31 March 2026, FDA reported that its review of the applicant’s global safety database, medical literature and FAERS data through 9 October 2024 identified 76 DILI cases with reasonable evidence of a causal association with avacopan. Seventy-four had a serious outcome, including 54 hospitalisations and 8 deaths. Of 60 cases with laboratory data to classify the initial injury pattern, 38 were cholestatic or mixed. Median reported time to onset was 46 days among 73 cases with timing information, with a range of 22–140 days. [12]

Seven of the 76 cases described biopsy-confirmed VBDS and were assessed by FDA as having reasonable evidence of association. All seven involved hospitalisation and three were fatal. Four began with cholestatic or mixed laboratory patterns and three with hepatocellular patterns; the median onset was 46 days (range 33–59). Six were reported from Japan and one from Canada. These seven cases are a subset of the 76, not an additional group to add to the DILI total. The geographic distribution is descriptive; by itself it does not establish a population-specific susceptibility.

What the case count does—and does not—show

The 76 reports are a regulatory case-series, not a denominator-based cohort. They do not provide an incidence rate or quantify an individual patient’s absolute risk. Spontaneous reports and sponsor safety-database cases can be affected by incomplete reporting, duplicates, selective reporting, stimulated reporting and missing clinical information. At the same time, a small number of well-documented, severe cases with a coherent temporal pattern and biopsy confirmation can be important for signal detection even when incidence cannot be calculated. The FDA’s wording—“reasonable evidence of a causal association”—should be preserved rather than upgraded into proof for every individual report.

The FDA communication recommended liver-panel tests every two weeks during the first month, monthly for the following five months, and then as clinically indicated. It advised prompt discontinuation, evaluation and consideration of alternatives if ALT or AST exceeded three times the upper limit of normal (ULN), ALP exceeded two times ULN, or symptomatic cholestasis such as jaundice or pruritus occurred. The current US label separately specifies baseline testing and testing every four weeks for six months; for patients of Japanese descent it says to consider testing every two weeks for three months, then every four weeks for three months. The label advises prompt evaluation and consideration of pausing therapy for ALT or AST above three times ULN; it directs discontinuation until drug-induced injury is ruled out if AST or ALT is above five times ULN, ALT or AST is above three times ULN with total bilirubin above two times ULN, ALP is at least two times ULN, or jaundice or pruritus occurs. It directs immediate, permanent discontinuation if VBDS is suspected. These are distinct documents with different wording. Clinicians and safety teams should follow current local product information and applicable safety communications, and should not silently combine thresholds or intervals into a fictitious single rule. [11, 12]

Assessing a suspected avacopan-related liver case

A useful case assessment begins with a dated clinical timeline rather than a single laboratory value. Capture treatment start and stop dates, dose, interruptions, dose changes, symptom onset, each liver-test result and the date it was collected. Include baseline liver function and the patient’s underlying disease activity. Record the full medication history, including recent antibiotics, immunosuppressants, supplements and over-the-counter products, because concomitant exposures may be competing causes or contributors.

For each relevant result, preserve the laboratory’s reference range and calculate the R ratio at initial presentation, then track how the pattern changes over time [16]:

R = (ALT ÷ ALT ULN) / (ALP ÷ ALP ULN)

An R ratio of at least 5 is generally hepatocellular, 2 or less cholestatic, and between 2 and 5 mixed. The ratio helps describe the biochemical pattern; it does not establish cause, severity or VBDS. Track bilirubin, INR, albumin and clinical status as well as ALT, AST and ALP. A predominantly cholestatic presentation may emerge with pruritus, jaundice, dark urine or pale stools, while aminotransferases alone can understate the evolving clinical problem.

Follow-up should seek information that can change causality or outcome assessment: imaging to exclude obstruction, viral and autoimmune investigations as clinically appropriate, competing diagnoses such as sepsis or disease-related injury, hepatology opinion, biopsy and pathology interpretation if performed, management, dechallenge course, recurrence on re-exposure if it occurred, transplant and vital status. Rechallenge should never be undertaken simply to establish causality where serious DILI or VBDS is suspected. Document unavailable information and the attempts to obtain it; do not convert “not reported” into “absent.”

A hypothetical example illustrates the distinction. A patient develops pruritus and jaundice about six weeks after starting avacopan, with raised ALP and bilirubin. The report is medically significant and needs prompt clinical evaluation even if ALT is only modestly elevated. The initial assessment should record a cholestatic or mixed pattern using the available reference ranges, check for obstruction and competing causes, and continue follow-up after discontinuation. Without biopsy evidence, code and describe suspected cholestatic DILI; do not call it confirmed VBDS. This is an illustrative scenario, not one of the FDA cases.

For narrative and data-entry quality, keep source facts separate from the reporter’s and company’s assessments. An accurate narrative should reconcile onset dates, laboratory units, seriousness, treatment action and outcome with structured fields. See How to Write an ICSR Case Narrative and Drug-Induced Liver Injury (DILI) for broader case-processing principles.

The 2026 regulatory actions: distinct questions, distinct outcomes

The hepatic signal prompted stronger warnings and monitoring. Separately, regulators reassessed whether ADVOCATE could be relied upon to demonstrate efficacy. That second issue concerns data integrity and evidence of benefit, not a new finding that avacopan caused liver injury.

FDA’s April proposal is not a final withdrawal order. It stated that the medicine could remain on the US market unless the application holder removed it or FDA’s Commissioner ultimately ordered withdrawal after the hearing process. Amgen’s July submission asserted that additional review supported a favorable benefit–risk profile; that is the company’s position, not an FDA finding. In the EU, the Commission has already acted on the CHMP recommendation. The UK has taken an interim managed-transition approach. These jurisdiction-specific statuses must be checked again before reuse because the proceedings can change. [7, 8, 13–15]

EMA’s referral also set liver-monitoring advice for patients stopping treatment: those treated for less than three months should be monitored at least every two weeks until three months from treatment start; those treated for more than three months should be monitored every four weeks up to six months, then as needed. EMA states that avacopan should be immediately discontinued if VBDS is suspected. This advice concerns management around cessation and complements the authority’s instruction to transition current patients; it should not be presented as a universal global label interval. [13]

Applying the signal in pharmacovigilance practice

For a marketing-authorisation holder, the operational response to a serious hepatic signal is broader than changing a case term or updating one periodic report. The safety system should preserve the evidence chain from source report to medical assessment, aggregate review and regulatory action.

1. Intake and triage: route suspected liver injury promptly for medical review. Capture symptoms, dates, all liver tests and reference ranges, treatment status, concomitant medicines, seriousness and outcome. Escalate possible VBDS, jaundice, hospitalisation, transplantation or death.
2. Case quality and follow-up: seek the missing data that affect diagnosis, pattern, seriousness and competing explanations. Maintain the distinction between confirmed biopsy findings and clinical suspicion. Keep follow-up active after discontinuation because cholestatic injury may resolve slowly.
3. Signal evaluation: review individual cases and the aggregate pattern together. Examine latency, biochemical pattern, dechallenge, pathology, alternative causes, dose and exposure, concomitant therapies, geography and reporting source. Use spontaneous-report databases to identify and characterise a signal, not to estimate incidence without an appropriate denominator and design.
4. Risk-management review: reassess whether the product information, risk-management plan, targeted follow-up, healthcare-professional communication and patient materials adequately describe the phenotype and action thresholds. Evaluate whether monitoring is feasible and whether users understand when to stop and refer.
5. Governance and benefit–risk: record what evidence was considered, uncertainties, dissenting views, actions, accountable owners and effectiveness checks. Update the integrated benefit–risk assessment when clinical benefit evidence, safety evidence or both change. For avacopan, keep the liver signal analysis distinct from the contested efficacy evidence, then bring both into the overall assessment.
6. Regulatory and quality controls: verify regional label versions and authorisation status, assess reporting obligations by jurisdiction, reconcile safety information across affiliates and partners, and retain traceable records of decisions and communications. Data-integrity concerns in a pivotal trial require a parallel quality and governance investigation; they should not be managed as ordinary adverse-event signal work alone.

This approach connects to Signal Validation, Detection and Confirmation, Signal Outcomes and Regulatory Action, and Benefit–Risk Evaluation.

Practical review checklist

When reviewing a suspected avacopan-related hepatic event, ask:

• Is the exposure chronology complete, including stop date and time to symptom onset?
• Are baseline and serial ALT, AST, ALP, bilirubin and relevant reference ranges available?
• Is the pattern classified from the earliest informative sample, and are later changes tracked?
• Are jaundice, pruritus, fatigue, pale stools, dark urine, abdominal symptoms, INR changes and hospitalisation captured?
• Has the reporter been asked about imaging, alternative causes, biopsy and pathology, treatment, dechallenge, transplant and final outcome?
• Is VBDS described only when the clinical and pathological evidence supports it?
• Are duplicate reports, missing follow-up and spontaneous-report limitations accounted for in the aggregate review?
• Are the applicable local label and current regulatory status checked for the patient’s jurisdiction?
• Are the safety concern and the status of efficacy evidence kept distinct in the case assessment and integrated benefit–risk record?
• Can the decision, rationale, source data, follow-up and communication history be reconstructed from the safety file?

The checklist is operational guidance, not a substitute for clinical judgement, local product information or jurisdiction-specific reporting rules. For broader signal governance, see Signal Management Documentation, Records and Inspection Evidence.

Key takeaways

• Avacopan’s hepatic risk was recognised before authorisation; post-marketing evidence made serious DILI, fatal outcomes and biopsy-confirmed VBDS a more specific safety concern.
• FDA’s 76-case review (including 7 VBDS cases) is a serious case-series signal, not an incidence estimate. Three of the seven VBDS cases were fatal; those deaths are within the total of eight DILI deaths.
• DILI and VBDS require careful chronology, laboratory-pattern assessment, alternative-cause evaluation and outcome follow-up. A suspected cholestatic injury is not automatically confirmed VBDS.
• The 2026 US proposal, EU revocation and UK managed suspension have different procedural status. The FDA’s US proposal was still pending on 3 October 2026.
• Hepatic safety and trial-data integrity are separate lines of evidence. Together, they affect the benefit–risk assessment because risk must be considered against reliable evidence of benefit.

References

1. Jayne DRW, et al. Characterization of Pharmacologic and Pharmacokinetic Properties of CCX168, a Potent and Selective Orally Administered Complement 5a Receptor Inhibitor. PLOS ONE. 2016;11(10):e0164646. doi:10.1371/journal.pone.0164646.
2. ChemoCentryx, Inc. Form 10-K for year ended 31 December 2019. SEC filing.
3. Amgen. Amgen Successfully Completes Acquisition of ChemoCentryx. 20 October 2022. Company announcement.
4. Jayne DRW, et al. Randomized Trial of C5a Receptor Inhibitor Avacopan in ANCA-Associated Vasculitis. J Am Soc Nephrol. 2017;28(9):2756–2767. doi:10.1681/ASN.2016111176.
5. Jayne DRW, et al. Avacopan for the Treatment of ANCA-Associated Vasculitis. N Engl J Med. 2021;384:599–609. Article retracted in June 2026; consult the journal record for the notice. doi:10.1056/NEJMoa2023386.
6. ClinicalTrials.gov. ADVOCATE study, NCT02994927. Trial record.
7. US Food and Drug Administration. CDER proposes to withdraw approval of TAVNEOS. 27 April 2026. FDA notice; Notice of Opportunity for a Hearing and proposal to withdraw approval.
8. Amgen. Second Quarter 2026 Financial Results, including the company’s account of its requested hearing and July submission. Company report.
9. US Food and Drug Administration. NDA 214487 approval letter and multidisciplinary review. Drugs@FDA application 214487.
10. European Medicines Agency. Tavneos EPAR. Authorisation history and product information.
11. US National Library of Medicine. TAVNEOS (avacopan) Prescribing Information, revised May 2026. DailyMed label.
12. US Food and Drug Administration. FDA Identifies Cases of Serious Liver Injury in Patients Taking Tavneos (avacopan). 31 March 2026. Drug Safety Communication.
13. European Medicines Agency. Tavneos referral: review of data integrity, efficacy and liver safety. Referral and recommendations.
14. Medicines and Healthcare products Regulatory Agency. MHRA concludes review of Avacopan Vifor following reassessment of benefit–risk balance. 1 September 2026. GOV.UK.
15. European Commission. Commission Implementing Decision of 4 August 2026 revoking the marketing authorisation for Tavneos; see the decision record linked from the EMA EPAR.
16. Fontana RJ, et al. AASLD practice guidance on drug, herbal, and dietary supplement–induced liver injury. Hepatology. 2023;77(3):1036–1065. doi:10.1002/hep.32689.

Regulatory Note

This article summarises public regulatory and scientific information available on 3 October 2026. It is an educational pharmacovigilance resource, not clinical advice or a substitute for the current locally authorised product information. Regulatory status and safety recommendations can change. At the stated date, the EU authorisation had been revoked; the UK had suspended use, sale and supply for new patients with a managed transition for existing patients; and FDA’s proposed US withdrawal remained pending. Readers should confirm the current position with the relevant competent authority and use jurisdiction-specific product information.