#NephJC Chat
Tuesday April 20 9 pm Eastern
Wednesday April 21 9pm IST
Wednesday April 21 9pm BST
N Engl J Med. 2021 Feb 18;384(7):599-609
Avacopan for the Treatment of ANCA-Associated Vasculitis
David R W Jayne, Peter A Merkel, Thomas J Schall, Pirow Bekker, ADVOCATE Study Group.
PMID: 33596356
Introduction
In 1982, an antibody against the cytoplasm of neutrophil leukocytes was first identified among 8 patients with pauci-immune necrotizing glomerulonephritis. Although not clear at that time, this short report gave birth to the anti-neutrophil cytoplasmic antibodies (ANCA), which was later used to define a group of systemic inflammatory conditions affecting primarily small-medium sized vessels: ANCA- associated vasculitis (AAV).
Most will remember the recent PEXIVAS trial, but if you do not, you can check out this excellent NephJC summary. PEXIVAS approached the controversial utility of PLEX, and showed the unquestionable safety profile of a lower dose steroid regimen (Walsh M, et al., NEJM, 2020). Wait a minute, did we just say steroids?!
Fun fact: steroids were being pursued with funding acquired because of a World War II rumor that Luftwaffe pilots had access to adrenal extracts which protected them from stress (it was actually the first of three major government funded research priorities, other two being antimalarials and penicillin). They were perceived as a true miracle: treatment of the first patient “Mrs. G” in 1948 was a great success (Hench P. S., Proc Staff Meet Mayo Clin, 1949), which resulted in E. C. Kendall, T. Reichstein and P. S. Hench being the 1950 Nobel Laureates in Physiology of Medicine. But with time their severe side-effects became apparent. Already in 1956, a movie “Bigger than Life” portrayed steroids as a nightmare drug which threatened the safety of both the patient and his family. No wonder why treatments that would decrease our reliance on corticosteroids in AAV are desperately needed.
If you are wondering how we have been debating the optimal treatment of AAV since the 1990’s, you have come to the right place. For an in depth review of the evolution of therapy for ANCA associated vasculitis, we invite you to check out the landmark trials blog post on this topic.
Even though the overall outcomes from this condition have greatly improved with immunosuppressive regimens, 34% of patients progress to ESKD (Corral-Gudino L, et al., Rheumatology, 2011). Also, mortality appears to be higher within the first year of diagnosis (25%), and one third of these patients relapse after 13 months (Booth A et al, AJKD, 2003). It is in this way, that the cumulative dose of immunosuppressive medications, specifically steroids, becomes alarming, particularly because of the side effect profile. This brings another question: Is the high morbidity/mortality related to the disease itself or to therapy associated side effects?
Little M and colleagues, from the European Vasculitis Study Group (EUVAS) showed that more deaths were caused by therapy–associated adverse events (59%) as opposed to active vasculitis (14%) (Little et al, Ann Rheum Dis 2010). Not surprisingly, infection was the most common cause of death within the first 12 months across all trials. And if one has wondered, the risk of infections is higher at least during the first 3 years of AAV diagnosis, with cyclophosphamide and steroids being identified as significant independent risk factors for severe infection (Charlier et al, Annals Rheum Dis 2008).
From Charlier et al, Annals Rheum Dis 2008
But how fast can we start tapering down steroids? Is it safe? We still have unanswered questions. However, the use of steroids beyond 6 months had a significantly lower risk of infections and no difference in time to relapse (McGregor et al, CJASN, 2012).
Now that we have proved that steroids play an important role in the morbidity and mortality of ANCA associated vasculitis, it is time to focus on steroid sparing regimens. In 2007 the stimulation of neutrophils by ANCA leading to complement activation via the alternative pathway was described (Xiao et al Am J Path 2007). Years later, a C5a receptor blocker was proved to protect against MPO-ANCA glomerulonephritis in animal models (Xiao et al, JASN, 2014). This effect was supported by a phase 2 randomized clinical trial, where the C5a receptor inhibitor avacopan was shown to be effective in replacing high-dose steroids in AAV (Jayne et al, JASN, 2017). If you want to know more, check out the NephJC summary about avacopan in AAV.
A few more words on avacopan. An interest in C5a and its receptor has actually been around since the 1990’s because of its potent pro-inflammatory effects (Gerard et al, Nature, 1991). What makes it so special? Targeting C5a avoids the amplification of the inflammatory cascade without affecting the formation of the membrane attack complex C5b-9, our main weapon against encapsulated bacterias (Bekker P, et al, PLOS ONE, 2016). Therefore, using this oral antagonist of the C5a receptor brings a lot of hope.
The complement cascade showing the point of intervention of CCX168, aka avacopan (from Becker, et al)
The ADVOCATE trial, as its name tells us, seeks to advocate for patients with AAV, decrease the cumulative dose of steroids and improve the side effect profile of this condition. This phase 3 randomized clinical trial compared avacopan with a glucocorticoid tapering regimen for the treatment of AAV.
The Study
Methods
Funding
This study was designed and supported by ChemoCentryx, the biopharmaceutical company that holds the US commercial rights to Avacopan. Two academic authors were consultants for ChemoCentryx and two authors were employees and shareholders of ChemoCentryx. All authors participated in the design and analysis of the trial as well as writing of the manuscript. A contract research organization (CRO) named Medpace conducted the trial.
Objective
The primary objective of the study was to evaluate the efficacy of avacopan to induce and sustain remission, compared to tapering doses of prednisone therapy, for children and adults with active ANCA vasculitis, who also received cyclophosphamide or rituximab followed by azathioprine.
Design
This was a global randomized, double-blind, placebo-controlled, phase 3 clinical trial.
Setting
Patients were enrolled from 143 centers across North America, Europe, and Asia. The trial was conducted between March 15, 2017 until November 1, 2019.
Inclusion Criteria
Clinical diagnosis of granulomatosis with polyangiitis or microscopic polyangiitis, consistent with Chapel-Hill Consensus Conference definitions,
Newly diagnosed or relapsed where cyclophosphamide or rituximab is needed,
At least one major item, or at least 3 minor items, or at least the 2 renal items of proteinuria and hematuria in the Birmingham Vasculitis Activity Score (BVAS),
eGFR ≥15 mL/minute/1.73 m2,
Positive test for anti-PR3 or anti-MPO (current or historic) antibodies,
Aged ≥ 12 years,
Female patients with adequate contraception,
Male patients with partners of childbearing potential with vasectomy or partner using contraception.
Exclusion Criteria
Pregnant or breastfeeding,
Alveolar hemorrhage requiring invasive pulmonary ventilation support anticipated to last beyond 14 day screening period,
Any other multisystem autoimmune disease,
Required dialysis or plasma exchange in the prior 12 weeks,
History of kidney transplant,
Received cyclophosphamide within 12 weeks prior to screening,
Received > 3grams of IV methylprednisolone (or equivalent) in the prior 4 weeks,
Received >10mg of oral prednisone (or equivalent) in the prior 6 weeks,
Received rituximab or other B-cell antibody within 52 weeks of screening or 26 weeks provided B cell reconstitution has occurred (i.e., CD19 count > 0.01x109/L),
Received anti-TNF treatment within 12 weeks prior,
Currently receiving a strong cytochrome P450 3A4 enzyme inducer,
Symptomatic/unstable heart disease within 12 weeks prior,
History of cancer within 5 years,
Evidence of chronic or active Tuberculosis, HBV, HIV, HCV within prior 6 weeks,
WBC count < 3500/μL, neutrophil count < 1500/μL, or lymphocyte count < 500/µL,
Evidence of hepatic disease.
Randomization and Treatment
Patients were stratified prior to randomization based upon each of the following factors
After stratification, patients were randomized, using an interactive response technology system and a minimization algorithm, in a 1:1 ratio to one of two treatment groups: 30mg of Avacopan or Placebo.
For those curious, the standard of care group steroid taper consisted of the following regimen:
Endpoints
The two primary efficacy endpoints were:
Clinical remission at 26 weeks defined by BVAS of 0 and no steroid exposure for 4 weeks before week 26,
Sustained remission at week 26 and at week 52 defined by BVAS of 0 and no steroid exposure for 4 weeks before week 52.
Need a reminder of BVAS?
The secondary endpoints included:
Steroid toxic effects per the Glucocorticoid Toxicity Index (GTI) during the first 26 weeks,
Early remission or BVAS of 0 at week 4t
Change from baseline in health-related quality of life scores
Relapse rate
Change from baseline eGFR
Change from baseline urinary albumin : creatinine ratio
Statistical Analysis
A sample size of 300 was calculated to provide the trial with 90% power to detect noninferiority of avacopan to prednisone. Noninferiority was defined as a margin of 20% difference between the groups and an incidence of remission in the prednisone group of 60%.
The analysis was completed using a modified intention-to-treat approach which included all randomly assigned patients who received at least one dose of trial medication.
Missing data at week 26 and 52 were imputed as “no remission.”
No interim analyses were performed.
The two primary endpoints were evaluated using a gatekeeping procedure (in other words an analysis in sequence to preserve the type 1 error and prevent false positives). The sequence of analysis was as follows: noninferiority at week 26 , noninferiority at week 52, superiority at week 52, and superiority at week 26.
Results
386 patients were assessed for eligibility, 55 were excluded, 331 were randomized. 166 patients were assigned to receive avacopan and 165 were assigned to receive prednisone (see figure 1)
Fig 1. Screening, Randomization, and Follow-up
Baseline characteristics are shown here (Table 1). Key points here are that two thirds of patients received rituximab and one third received cyclophosphamide, and 80% of patients in both groups had renal vasculitis.
Table 1. Demographic and Clinical Characteristics of the Patients at Baseline.
Table 2. Primary and Key Secondary End Points.
Remission at week 26 was seen in 72% of patients receiving avacopan and 70% in the placebo-control (prednisone) group.
Remission at week 52 was seen in 66% of patients receiving avacopan and 55% in the placebo-control (prednisone) group.
Avacopan noninferiority was met at week 26 and 52 (p<0.001) and superiority was met at week 52 only (p=0.007)
To evaluate steroid induced toxicity, the change in glucocorticoid toxicity index at week 26 and 52 was demonstrated (Figure S2).
Fig S2. Glucocorticoid Toxicity Index
At week 52, the least-squares mean change from baseline eGFR was 7.3ml/min/1.73m2 in the avacopan group and 4.1ml/min/1.73m2 in the placebo (prednisone) group
Fig S3. Estimate Glomerular Filtration Rate
Change in health-related quality of life measurements were sustained or improved in patients over 52 weeks. Avacopan patients reported better health related quality of life in several measures (physical health, health perception, emotional health, and vitality)
Fig S4. Short Form 36
The hazard ratio for relapse after remission (Avacopan vs. prednisone) was 0.46 (95% CI, 0.25 to 0.84).
Fig 2. Kaplan–Meier Plot of Time to Relapse.
The mean total prednisone dose was 4mg per day in the avacopan group and 12mg per day in the placebo control (prednisone) group.
Fig S7. Mean Daily Total Prednisone-Equivalent Glucocorticoid Dose (in mg) by Study Week by Treatment Group
Serious adverse events and death are shown in Table 3. The most common adverse event was worsening vasculitis in both groups (10% avacopan and 14% in prednisone group). Infections occurred in 68% of patients receiving avacopan and 75.6% of patients receiving placebo control (prednisone group). 15% of patients in the prednisone group and 22% of patients in the avacopan group had serious infections with the most common serious infection being pneumonia. Two deaths reported in avacopan and four deaths in the placebo control (prednisone) group.
Table 3. Safety Results
Discussion
We know that glucocorticoid side effects remain important for patients. A study focusing on ANCA associated vasculitis identified three most important themes of steroid regimens:
Their effectiveness at the time of diagnosis and relapse, with concerns that withdrawal may cause flares
The emotional, physical and social side effects
The problems of balancing risks and benefits of steroids (Robson J. C. Rheumatol Int, 2018).
The breakthrough of steroid-free regimens
Glucocorticoids were the first drugs to work in ANCA associated vasculitis (Beidleman B., JAMA, 1963) and unfortunately for a long period of time vasculitis treatment required high doses of steroids. We have known that duration of steroid use is associated with long-term damage as assessed by Vasculitis Damage Index (Robson J., Rheumatology, 2015) – which is a predictor of mortality (Exley A., Br J Rheumathol, 1998) but had no other alternative. Recently the PEXIVAS trial (Walsh M., NEJM, 2020) brought the exciting news that quick steroid taper is noninferior to a full steroid dose (although we did focus mainly on plasma exchange issue) and another small study (Pepper R., Rheumatology, 2019) suggested that therapy with both rituximab and cyclophosphamide makes early glucocorticoid withdrawal possible. Avacopan finally brings the long-awaited possibility of steroid-free or steroid-sparing disease remission.
The possibilities
Not only was avacopan non-inferior to steroids at week 26, it was also superior to prednisone at week 52. Non-inferiority would have been enough (thinking as we avoid glucocorticoid side effects) - although a margin of 20% is quite high - superiority is simply stunning. One question is the really quick steroid taper in the compared group – much quicker than in PEXIVAS even (see table comparing steroid doses for patients weighing 50-75 kg). This might have caused less sustained remission in prednisone arm of the trial: we are still not sure if steroids can be completely tapered so quickly, and many physicians believe 5 mg of prednisone is safe (Pincus T., Clin Exp Rheumatol, 2011), so the protocol used in ADVOCATE was not really the current standard of care.
Another important possibility is the observed reduction of serious adverse events, there were also less infections in the avacopan group. No Neisseria meningitidis infections were observed (as avacopan does not block the formation of C5b and the membrane attack complex). Since we know mortality from infectious causes is high in ANCA associated vasculitis (Little M., Ann Rheum Dis, 2010) and might be related to glucocorticoid use it is an interesting aspect to pursue.
What’s in it additionally for patients?
The acronym ADVOCATE is totally apt, as this trial truly advocates for patient reported outcome measures (remember how they won NephMadness in 2018?). We have already mentioned how steroid-free regimens are important for both physicians and patients, this trial did try to focus on that and assessed not only the Glucocorticoid Toxicity Index (capturing most common side effects of these drugs but unfortunately developed by physicians only without any patients involved) (Miloslavsky E., Ann Rheum Dis, 2016) but also health-related quality of life. Not surprisingly the group receiving avacopan had less glucocorticoid-induced toxic effects. Quality of life improved in both treatment groups, at 52 weeks both EQ-5D-5L and SF-36 scores were higher in the avacopan-treated patients. It would however seem prudent to finally develop a truly patient-reported outcome (long postulated by the OMERACT group – Cheah J., J Rheum, 2019) to measure the effects of glucocorticoids and test it in further avacopan studies – maybe than, the differences would be much bigger.
What’s in it additionally for nephrologists?
Over 80% of the studied group had renal vasculitis, so it is definitely “our” population of patients. An exciting signal is that at week 52 the eGFR seemed higher in the group treated with avacopan, and the difference was most pronounced in patients with baseline eGFR<30 ml/min/1.73m2. We know from EUVAS (European Vasculitis Study Group) data (Robson J., Annals of Rheum Dis, 2015) that at 6 months 10.7% of patients with MPA and 3,5% with GPA develop end-stage kidney disease and the numbers reach 20% and 9% respectively at long term follow-up (up to seven years), so those few milliliters per minute gained do seem promising.
The limitations
There are of course some limitations to the study:
The non-inferiority margin of 20%, although this concern was minimized by the proved superiority at week 52,
The main treatment was chosen at physician discretion (oral/intravenous cyclophosphamide, rituximab), although it does reflect real-world practice,
The avacopan arm was not really steroid-free, let’s say it was a steroid-sparing group (the total steroid dose was equivalent to 12 mg of prednisone in the steroid group and 4 mg of prednisone in the avacopan group),
Patients had to be tapered to 20 mg or less of prednisone equivalent during the screening period – that could have excluded an important group with worse outcomes,
The really quick taper of steroids, as described above,
The trial population was heterogeneous – both patients with newly diagnosed vasculitis and relapses were included, although on the other hand that’s what our real-world practice looks like.
Another important question is the cost of avacopan. The company states that since it is a small molecule it’s manufacturing cost is anticipated to be lower than for example eculizumab (which is one of the most expensive therapies, so it’s not a great reassurance) but analysts put expectations for the drug sales in the blockbuster category. Prednisone, on the other hand, is cheap, so for some will probably still remain the only option.
Conclusion
As ADVOCATE leaves us full of hope, we still think it pertinent to conclude with a few questions.
Is avacopan the Anca-VAsculitis COngenial PANacea? The answer to that is of course negative.
Is it a breakthrough in our current treatment strategies? Seeing as it won the inaugural edition of RheumMadness– we certainly perceive it that way. It seems that the results of this trial have also convinced the FDA – it has accepted ChemoCentryx’s New Drug Application (NDA) for avacopan and has set July 7, 2021 as the Prescription Drug User Fee Act goal date.
Will it change the future of our patients? We can certainly hope so, but ultimately, time will tell.
Summary written by
Nephrology Fellow,
The Johns Hopkins University,
Nephrology Attending,
Medical University of Białystok, Poland,
and Keia Sanderson,
Assistant Professor Pediatric Nephrologist,
University of North Carolina