NCHR’s Testimony at FDA Advisory Committee Meeting on Avacopan

May 6, 2021

Thank you for the opportunity to speak today on behalf of the National Center for Health Research. I am Dr. Meg Seymour, a senior fellow at the center. We analyze scientific data to provide objective health information to patients, health professionals, and policy makers. We do not accept funding from drug or medical device companies, so I have no conflicts of interest.

Today, you are asked to assess data from a single phase 3 clinical trial comparing avacopan to prednisone at 26 and 52 weeks and to discuss the clinical meaningfulness of avacopan for patients.

First, let’s talk about the effectiveness of the drug and the limitations of this information due to study design. We know that there was no significant effect of superiority for avacopan at 26 weeks. The applicant would like to claim that there is a difference at week 52, favoring avacopan. However, as stated by FDA in their briefing document, there are issues with the study design that limit whether or not we can interpret a meaningful clinical benefit for avacopan. 

For example, 87% of patients in the avacopan treatment group also received glucocorticoids during the study period. Although it was prespecificed that glucocorticoids above the protocol specified taper must be discontinued by week 4, that did NOT happen. Instead, 86% of patients in the avacopan group received glucocorticoids between weeks 0 and 26. FDA scientists state that this effectively creates a different comparison: avacopan plus lower dose glucocorticoids vs. higher dose glucocorticoids for patients in the prednisone arm. This causes problems for the interpretability and meaningfulness of the comparison. 

FDA’s briefing document also notes that the observed superiority of avacopan at week 52 may be due to treatment differences in the subgroup receiving rituximab instead of cyclophosphamide plus maintenance azathioprine. Subgroup analyses suggest that avacopan was only effective when compared to patients who did not receive standard-of-care maintenance immunosuppression therapy and may be considered undertreated. This obviously raises questions about the adequacy of the comparisons and clinical meaningfulness of data for avacopan at week 52. Moreover, differences in assessments from the Investigator and the Adjudication Committee occurred in 17 patients measured at week 52.

Although the applicant states that there were more adjudicated relapses after remission in the prednisone group compared to the avacopan group, the study was not designed to assess time to relapse or proportion of relapses. Because remission may be achieved in different types of patients in the two treatment arms, differences in relapse cannot clearly be attributed to the treatment, but instead to differences in the characteristics of the subset of patients included in the analysis. FDA scientists note that this eliminates the advantages of randomization, since the treatment arms are no longer balanced with respect to possible confounders, which leads to biased comparisons between treatment arms and limits the interpretability of results. 

Finally, let’s talk about the safety profile of avacopan.  Although safety events such as infections were generally similar between groups, FDA scientists point out that the safety database is limited when it comes to reliable assessment of rare or latent events. However, the data show that more patients in the avacopan treatment group had AEs and SEAs associated with hepatic abnormalities, such as liver enzyme abnormalities.

Although AAV is a serious disease with an unmet need for new treatments, the FDA must only approve products that have a favorable risk-benefit profile for patients. Due to issues with study design, avacopan has not clearly demonstrated that it is more effective than the existing treatments, and it apparently carries more risk for certain adverse events. We respectfully urge you to consider the shortcomings of the scientific evidence when voting today.



The Advisory Committee voted (10 yes, 8 no) that the risk-benefit profile of avacopan supports its approval.