Stephen A. Paget, MD: It is a pleasure today to introduce a good friend, Leonard Calabrese, who is the Director of Clinical Immunology at the Cleveland Clinic Foundation and a Professor of Medicine at Ohio State University. Len is world renowned for his interest and focus for many years relating to immunologic and autoimmune diseases. But one of his particular interests, and one I know he has spoken on quite a bit in the past, are the vasculitides. He has given us some of the most basic information on primary angiitis of the brain, and today he will talk about new treatment strategies for vasculitis. It is a pleasure to welcome you to HSS and New York.
Leonard H. Calabrese, DO: Thank you very much, Steve. It is always a great pleasure to be here, and I am particularly glad to be back in New York City after all these tumultuous events and trying to get back to normal. I can't imagine the emotions and the hardships you all have been through, but we are glad to be back. This morning, I am very happy to talk about treatment strategies for systemic vasculitis.
I am going to start out talking a little bit about the targeted diseases, of all the large -- kind of -- pie of vasculitis, that we will concentrate on. I will talk about some of the history and biases that go into interpreting data on clinical trials and outcomes in vasculitis. I will propose to you what I think is the gold standard of therapy, and then we will discuss some innovations, perhaps improvements of this gold standard, and then finally talk about some more futuristic therapies.
Of all the systemic vasculitides, I am going to focus on a few. We will talk a little bit about classic polyarteritis, which is probably the rarest of the diseases and, for the most part, where most of the data is coming now from controlled clinical trials -- the ANCA associated vasculitides. This includes Wegener's granulomatosis, microscopic polyangiitis and Churg-Strauss disease.
I don't have to tell an audience of this level of sophistication, the historical data on these illnesses and, when we look at the older literature, untreated PAN - it was a formidable disease and is a formidable disease, but perhaps not as formidable as the literature would suggest. We do know that Wegener's granulomatosis is, generally speaking, a progressive illness and some of the other forms of ANCA-associated vasculitis, at least to the point that, without therapy, they are at the minimum highly morbid diseases and generally quite fatal.
However numerous forms of bias taint previous estimates of this degree of morbidity and mortality. You know the older literature that quotes the highest rates of mortality for PA and Wegener's largely derive from post-mortem studies. It is one of those peculiarities about post-mortem studies that all the patient's are dead at the time of diagnosis, and so many of these older studies are clearly skewed to fatal cases.
Over the years, therapy and complications from end-stage renal disease from management and prevention of opportunistic infections have improved; so this should influence our data. Clearly, and well-known to all of us, older studies discussing classic polyarteritis are probably truly an admixture of PAN and Wegener's and microscopic polyangiitis, and older studies of Wegener's are clearly admixtures of microscopic polyangiitis at the minimum. So we have heterogeneity of diagnosis in these older series.
It is inappropriate to compare sequential series in terms of morbidity and mortality because of the influences of these. And then, lastly, you know some of our greatest treasures of data in systemic vasculitis come from single-center studies, which are highly influenced by the nature of their practice, particularly the NIH. And, to be fair to them, this has to influence how we look at these.
When we talk about treatment of systemic vasculitis what are our goals? Well, obviously, we want patients to survive and, we want to preserve major target organ function. It is very fair to look at renal survival in patients with vasculitic renal involvement. In addition, particularly for the ANCA-associated vasculitides, as you know and as we will review, relapse rate is clearly in excess of 50%. And this is a goal; we have to do better with this or we are just going to continuously treat and retreat these patients. And, then, lastly even when we apply our best therapy and minimize relapse and save end-organ function, a sizable percentage of patient's experience permanent toxicity from their therapy, largely from corticoids and alkylating agents, such as cyclophosphamides. So these are things that we need to go into.
I will try to emphasize this morning the best evidence, and this is not an exercise in academics. I think that we are all trying to attempt to practice evidence-based medicine, and we look for the best external evidence for treating all of our diseases, or we should. This really stems from examining high quality, randomized, controlled clinical trials, when they are available and, thankfully, in vasculitis only in the past five years have we had any sizable randomized trials. In the absence of that, we have tried to look at prospectively carried out carefully done observational cohorts of which I will discuss in some detail this morning. And we try to supplement this with clinically relevant and basic science research elements. Taking the evidence is not equitable with evidence-based medicine - just looking at a clinical trial and applying it to our patient. We have to look at it for validity and applicability, and that is where kind of the savvy of the clinician comes in - trying to apply particularly in the treatment of a rare disease, limited amounts of data. And then lastly it should never be forgotten that, in the true practice of evidence-based medicine, the wishes and desires of the patient have to be mixed in with best available evidence and clinical experience. So it is not a cold, calculated ivory tower approach. It is really trying to integrate evidence with experience in our patient's best interest.
Now, I will propose to you that the gold standard therapy for systemic necrotizing vasculitis does not come primarily from a randomized, controlled clinical trial, but comes from the three-decade-plus experience at the National Institutes of Health, which really pioneered our current treatment programs - even in their current forms - starting with Shelley Wolf and Tony Fauci and Gary Hoffman and the current group. This was not a randomized controlled trial, but it was a long-term and prospective study and probably will never be duplicated except for perhaps another institution, which I will mention in a few minutes.
I would like to spend just a couple of minutes talking about that gold standard because I think that, while we all do it, we all vaguely know that this is prednisone and cyclophosphamide (Cytoxan), there are some nuances that are vitally important when you actually apply these to patient care. The treatment model that was developed there was basically combining glucocorticoids in a dose of approximately a milligram per kilogram per day as the initial therapy for life-threatening vasculitis. In addition, it combined oral (p.o.) cyclophosphamide in a dose of 2 mg per kilogram per day. The therapy was initiated, the patient was monitored and, following clinical remission, the patient defervesces - they are less toxic, their acute phase reactants come down. Things are moving in the right direction.
The glucocorticoid was tapered to an every other day regimen until it successfully reached a dose of 0 mg. on an alternate day in about three months.
Cytoxan, as always, is modified according to renal function. In patients with creatinines over 2.5, generally their doses are reduced by a quarter or a third. Now, in addition, for the exceptionally ill patient - the patient with rapidly progressing glomerulonephritis, pulmonary hemorrhage, galloping mononeuritis multiplex - patients are frequently induced with intravenous methylprednisolone pulsing at a gram a day for three days and then, frequently, intravenous Cytoxan in some dose of 3 to 5 mg per day. We are talking 2, 3, 400 mg of Cytoxan for two to three days before converting to oral.
In those patients who still had clinical disease activity, cyclophosphamide could be adjusted upward by 25 mg every two weeks but avoiding, explicitly avoiding, leukopenia and - it was always stated - less than 3,000 total white blood cells. And I really keep my eye on the absolute neutrophil count. Trying to keep it above 1,500 should be emphasized because this has not been always followed when this protocol has been applied in the literature, and leukopenia is not a goal of therapy.
And meticulous follow-up and monitoring is vital for this to work. What is meticulous follow-up and monitoring? Well, it ranges between Gary Hoffman who, for all the time the patients are on alkylators, gets weekly CBCs and urinalysis and gives patients home urinary dipsticks to check for new or recurrent hematuria - to every other week at the outer margin of follow-up. Monthly CBCs being looked at by someone other than the treating physician really are not enough to do the job and, as you are all aware, as the alkylating dose mounts up, that is exactly when the patient is most vulnerable to leukopenia and the complications of opportunistic infections. So the longer you go doesn't mean that things are safer; it just means that we have to watch better.
Now, applying this treatment paradigm, the NIH outcomes are well known: 90% had major responses; 75% would qualify as complete remissions. And today, using Birmingham Vasculitis Activity Score, this would be a BVAS of 0. After a mean follow-up of eight years, the last time this group was reported in detail, only 13% mortality but the expense was very high. After one year of therapy and the alkylating agents were stopped, relapse rate was 50%. Major complications, hospitalizations, parenteral antibiotics were the most common and were experienced in 40%.
Now last year Reinhold Keller, working with Wolfgang Gross in Lubeck, published a very important study that really is not talked about enough because it compared a single cohort of ANCA-associated vasculitis in Germany using the gold standard therapy with a few minor modifications. There were some patients in here treated with pulse therapy. But it kind of served as a modern day comparative study to the NIH experience. And if you look at these, it is pretty bizarre how this disease can just stay the same between continents. The ages and the numbers of patients are very similar, and the ANCA percentage was similar. Those treatments with prednisone and Cytoxan were quite similar. The vast majority were high-grade responses - in 90 to 100%, complete remission somewhat lower in the German series, 50 versus 75%, and relapse rates ranging from 50 to 71%. End-stage renal disease was only noted in about 10% and similar for death. So these were very important comparative data.
Certain things were not reported in this study, but other things were. There were still significant deformities, generally from head and neck disease, and 28% tracheal stenosis in both. Cystitis was significantly less frequent in the German series, probably for two reasons. One, mesna was used in most patients, and there were a considerable number of pulse patients in this group, and it did correlate with reduced total doses of Cytoxan. The infection rate was somewhat lower in the Lubeck group as well, and all of these patients were put on PCP prophylaxis, far more than the earlier reported series. There was a higher incidence of myelodysplastic syndrome in the Lubeck group, and this was really not well explained. We have two studies, single center, looking at a similar number of patients of the same type, and with very similar outcomes.
So our conclusion for the gold standard therapy is that this is an effective but toxic regimen with a very high and unacceptable relapse.
How can we get better? Well, there are varying strategies, and I am not going to talk about all of these because there is not enough time. I will say that there has been a very honest effort to limit therapy based on prognosis, and I think that this can be done with certain forms of therapy. Only a few months ago, Loic Guillevin published in Arthritis and Rheumatism, the application of their factor 5 score, which tries to assign prognosis based on renal, cardiac, central nervous system, and gastrointestinal involvement, and took those patients with factor 5 scores of 0 to 1 and treated Churg-Strauss and PAN patients with glucocorticoids alone with some degree of success. There is still, I think, room to treat patients with highly limited Wegener's granulomatosis to the head and neck without alkylating agents, and there will be another rebirth of trimethoprim/sulfa as monotherapy presented in San Francisco next month [the 2001 scientific sessions of the American College of Rheumatology] that I think will be kind of interesting.
I think we are at the area where I'd like to talk a little bit about alternative routes of drug administration - in particular, a few words about pulse therapy and stepdown therapy, which I think is now the new standard of care and has the opportunity not only to limit toxicity but to limit relapse when applied properly.
Pulse therapy clearly has advantages. It is clearly less toxic to the bladder and possibly less so from infection. If it is done properly, there is a lower total dose of alkylator administered. There are numerous open observational studies that have been published claiming that this is highly effective. It is the standard of therapy for many renal physicians treating rapidly progressive glomerulonephritis such as Ron Faulk and colleagues. There are other, carefully done, smaller, prospective but uncontrolled studies that have found that very weak in avoiding relapse, and the data from the NIH were not too dissimilar from the Lubeck group where 70% of their patients were relapsing.
There are two randomized, controlled, clinical trials that deserve to be thought about. The first is the data of Habetts, which was published a couple of years ago. This randomized patients to pulse therapy 0.75 grams per metered square versus oral for ANCA-associated vasculitis; all of them had renal disease, and it was associated with 57% less total drug. There was significantly less toxicity in terms of leukopenia and infection, and in this small study (22 and 25 in both treatment arms), there was an equivalent rate of remission, relapse, and renal outcomes. I think this is really probably the best study that was done of this type, but the caveat here is that this study was powered to look at differences of toxicity, and these small ends really were insufficient to detect small differences in the equivalence in remission and relapse. So I think that this still warrants a large, randomized, controlled clinical trial, which is being done by the European vasculitis group now.
The other study published by Malekea Van, accompanied by an editorial by Gary Hoffman, was a French multi-center, randomized, controlled trial of pulse Cytoxan versus oral Cytoxan, everyone induced by intravenous methylprednisolone pulse. It was stated in this article, this is a very good example of some of the issues, that they are going to compare pulse therapy to the gold standard of therapy - the gold standard being NIH regimen. The little twist here is that everyone received pulse methylprednisolone, no matter if they had mild disease or they had more severe disease. So leukopenia was a goal of therapy and then brought back up again. We don't really know about the monitoring. The bottom line of this study was, while they claimed that earlier remission was superior for pulse but equal at one year, serious infections were more common in oral. There was a higher relapse rate in the pulse - 59 versus 13%. But down here, the overall mortality of this group exceeded 40%. 33% in the pulse group. Almost 60% in the oral group. Whatever they did was unacceptable in terms of standard of care, and I think the editorial from this article really pointed out the issues of trying to apply the gold standard, and it has to be done right when it is applied.
There is a meta-analysis that will be presented in San Francisco by Kirsten DeGroot, which is in press right now, looking at this plus an additional randomized controlled clinical trial, suggesting by and large equivalence of pulse and oral therapy with the caveat that relapse rate is still a problem.
I will tell you that in our practice at the Vasculitis Center at the Cleveland Clinic, one of the greatest pitfalls in the management of vasculitis that we see of patients referred into us, are the patients who are "Cytoxan failures." Virtually all of these patients come in with active disease on pulse therapy. I believe that pulse therapy works for many patients with systemic necrotizing vasculitis. I think it works for, maybe, perhaps the majority of patients, but the caveat is that failing pulse Cytoxan does not make one a Cytoxan failure. We have salvaged many patients who are failing pulse therapy merely by converting to oral therapy, and there are numerous anecdotes that have been reported in the literature, but I say this now to emphasize the common frequency of this. So I think it has a place in mild to moderate disease. I would not think about it in patients with more advanced disease.
I think that this is now where we are moving, and this is step-down therapy. We have a great therapy. Cytoxan is a great drug. To emphasize how great a drug, I'll show you some data. We can get most patients into remission with it. The problem is, if you treat them too long, they get problems. The accumulative rate of bladder cancer, transitional cell cancer of the bladder, you know after 15 years may be in excess of 15%. Myelodysplastic syndrome and combined threats of opportunistic infection. Yet, it does its job. So can we still use Cytoxan and avoid these later complications. The later complications usually come from more prolonged therapy.
The notion here is to induce with Cytoxan and then switch to a more palatable drug such as azathioprine or methotrexate or others, including leflunomide and microfenelate and other drugs that are coming. The most elegant study, that is still not in press but has been presented in many forms in the last couple of years, is from the European Vasculitis Study Group, called CYCAZAREM [a randomized trial of cyclophosphamide versus azathioprine during remission in ANCA positive systemic vasculitis]; they always have the best names for their studies. Rasheem Lukmani and David Jayne have spearheaded this study and basically it asks this. The hypothesis was that azathioprine would be as effective as Cytoxan for the prevention of mortality and relapse and would probably have fewer side effects. So they designed a study that took ANCA-associated vasculitis, sick patients, these people had threatened vital organ function, renal and/or pulmonary. They could not have a creatinine of greater than 5.6. Those people were put into a fulminant protocol with other drugs, and the study basically looked something like this. Everybody started out on Cytoxan 2 mg per kilogram, prednisone a mg per kilogram, true gold standard therapy. They were treated from 0 to 6 months and, at the point when they went into clinical remission, BVAS's of 0, they were then randomized to continue Cytoxan at a somewhat reduced dose 1.5 mg per kilogram veering off of the gold standard for 12 months, or to go to azathioprine for 12 months, and at that point in time, everyone was put on azathioprine and followed up to the 18 month point.
This was a very large study by vasculitis standards, nearly 140 patients. If you noticed, number one, at the 18 months, the relapse rate is very low. First of all, these patients are all on therapy for a long time. So this is a very favorable thing to begin with. Secondly, the mortality in these patients was quite equal but, thirdly, as one would expect, the greatest number of severe and life-threatening toxicities occurred during the induction/remission phase here when everyone is on Cytoxan but, after that, it was nearly halved in the azathioprine group. You know, as a conclusion, in this study there was a very low relapse. This appeared to be an effective way of reducing relapse at a much lesser cost of toxicity, and I think it's a very, very, very appealing way of getting the most bang from the buck from the alkylator and taking advantage of the safety of an antimetabolite.
There have been several other attempts to do the same thing with different therapies, and Carol Langford at the NIH has reported a very nicely done prospective, open trial of patients with Wegener's disease; 50% had renal disease in this cohort. They were all treated with Cytoxan and glucocorticoids until remission, again about 3 months, and then switched to methotrexate started at 15 mg and escalating up to 25 - very much like the NIH regimen for its use in limited Wegener's and Takayasu. Glucocorticoids were attempted to be withdrawn later as the disease course continued to improve.
Overall, this worked out quite well. After follow-up of 22 months, a mean follow-up, there was a single withdrawal, two for technical reasons, two for toxicity. There were two relapses. There were no deaths in this group. And 23 patients of the 31 patients were totally off of glucocorticoids, and six of these, they tried to get off of methotrexate as well, and their conclusion that stepping down to methotrexate was a viable alternative.
I think the issue here, the caveat for me, is that this is a little tricky, and I always have great trepidation in using methotrexate in people with changing renal function because this is a drug whose PK is directly influenced by GFR, and these patients, they didn't include anybody with creatinines I think greater than 2.2 or 2.3. The issues here of being very vigilant of methotrexate toxicity make this problematic in people with renal involvement and probably a little less appealing to me than azathioprine.
Even smaller yet, another study doing mycophenolate is step-down and ANCA disease, published in the nephrology literature, same principals, induce with Cytoxan. After remission go to mycophenolate. Mycophenolates were are getting doses up to 2 grams a day are fairly well tolerated. I think after 2 grams a day you start losing the benefit of using a better tolerated drug because the incidence of OI's start to mount.
In this study of a small number of patients, 11, there was a high rate of remission. There was a moderate degree of toxicity. Two had leukopenia. Infections were noted in three, including one with CMV, the bane of the long-term mycophenolate group. And they found that this was effective therapy. The Lubeck group has also reported using leflunomide as a similar type step-down drug, but they are now getting to doses of 30 and 40 mg a day and, again, this loses a little appeal when you are starting to use drugs at the upper end of the toxicity threshold. The best data from all of this, in my opinion, is the azathioprine in very modest doses.
All right, moving along from there, what are some other strategies? What about using other drugs from the get-go -- not inducing with Cytoxan but inducing with other agents? And we have a whole bunch that have been reported on: methotrexate, azathioprine, cyclosporine, etoposide, mycophenolate. There are no randomized controlled clinical trials of any of these agents comparing them to gold standard, but they are impressive enough data that, I think, we can chose to use these in certain situations.
John Stone had a very nicely done, small, observational study of 19 patients with life-threatening Wegener's granulomatosis; 50% of them had renal involvement, but rather smoldering renal involvement, but their creatinines were less than 1.3 - much more restrictive than the Carol Langford. Treatment was initiated with prednisone 40 mg a day and escalating methotrexate up to 22.5 mg, mean 18.7. The outcome of this observational study was, you know, the vast majority improved; three-quarters went into remission. Prednisone was successfully tapered in the vast majority of patients. One single patient went on to require alkylators. So here is a regimen that is fairly well tolerated; we are pretty good with it. But you know when this treatment is stopped, 50% of these people relapse. I will talk a little bit at the end about my casual interpretation of this data - just based on experience at the present time, particularly treating ANCA-associated vasculitis. You know if you have a patient that has gone through this, and they've been induced and now they are hanging out at 15 mg of methotrexate doing great, why take the risk of a 50% relapse rate if you can maintain this with a high degree of safety and the patient wants to do it? So it's a thought - long-term suppression like we are doing in many other diseases.
Carol Langford - here's her most recent compilation of the NIH data, again taking patients with ANCA-associated vasculitis with creatinines of less than 2.5 and treating them with glucocorticoids and methotrexate for a prolonged period of time and, again, trying to taper them off of it. So there were 11 relapses but, again, the data in both of these observational studies are almost identical. Only two patients had a rise in creatinine of greater than 0.2 mg. There were 11 relapses, all occurring as they were trying to withdraw all this therapy, and their conclusion was that for patients with limited disease, methotrexate is a good alternative drug. In the largest Wegener's trial done to date, the Wegener's etanercept trial, the WEGET trial, there are actually two limbs of therapy - one for patient's with limited disease who are induced with methotrexate, the other with more progressive disease who are induced with Cytoxan. So this has established itself actually as a standard of care and, when the etanercept comes later, I'll mention that at the end.
Other drugs to do the same thing, yeah why not. There are limited numbers of patients that have been treated with the chemotherapeutic etoposide. It has somewhat less marrow toxicity. So for those patients that may be totally wiped out that still need to be induced, there are some thoughts. Cyclosporine is not a very appealing drug for these patients because of the high prevalence of renal disease. Yet, it has been used in relatively low doses. And other agents, such as interferon alfa, which are gaining popularity in the treatment of refractory Baccates have been used in limited number of cages of particularly Churg-Strauss disease and occasional patients with Takayasu. These are anecdotes, and I don't usually choose to discuss them very much.
Biologic therapies - I would like to say a few things about a couple, including IVIG and apheresis and a cytokine therapy as we wind up. IVIG is kind of an appealing therapy. Everybody uses it. It has kind of gone through the phases, just like plasmapheresis. And one of my mentors, Alan MacKenzie, used to say that all of these drugs go through the stage of having a honeymoon phase, you know where it is effective for everything, and then it ultimately becomes the whipping boy when people realize that this is not working. We certainly saw that for plasmapheresis. We certainly have seen it for IVIG. And anti-TNF is the next; it's now in its honeymoon phase. It works for everything from chronic fatigue to endometriosis. Those are all clinical trials by the way. There will be an abstract in San Francisco using it for silicone implant-associated "whatever" disease. I just saw that one yesterday.
IVIG needed to have some type of randomized, controlled clinical trial, and David Jayne did it. And while it is not a perfect trial, it is data, and I think there is something to be said here. He took patients with ANCA disease that had some degree of activity and who were on stable doses of their standard therapy, either Cytoxan or azathioprine plus glucocorticoids. Their only intervention after several weeks of stabilization was IVIG 2 grams per kilogram or placebo. And then the patients were followed and assessed for disease activity. The primary outcome of this study was the Birmingham Vasculitis Activity Score, which for those of you who are not well familiar with, it is basically an enumeration on a target organ basis of new clinical events that are ascribed by the investigator to be due to vasculitis, not therapy or intercurrent illness. Primary outcome was 50% reduction of BVAS for three months.
The graphs here basically show that here are patients with placebo, and they got a nice response in their BVAS, and these are patients treated with IVIG and basically up through three months there was a statistical improvement of 50% and a fall in BVAS of the IVIG group versus the placebo group. There was also a fall in CRP but not in ANCA. After three months, this all washed away in terms of differences.
What does this mean to us? I think that no one including David Jayne has said that IVIG is a hot knife cutting through butter in treating vasculitis, but there is biologic activity. Question?
Yes, I have one for you. All of these are ANCA positive, that is the term that is being used. Has anyone looked to see what difference it makes whether it's PR3 or MPO?
You mean whether it's PR3 or MPO?
Yes. In other words suppose is ANCA negative. Or suppose you have a different antigen that you might have suspected. What difference does it make to the therapy?
The question is "Has this type of analysis done?" Mostly, the studies have been done on ANCA positive patients. Does it make a difference if you are ANCA positive or ANCA negative? I think they are included for the most part to insure clinical homogeneity for the study. They have been sub-stratified for whether they are reactive to PR3 or myeloperoxidase and, by and large, there have been very minimal differences reported with any of these therapeutics. While there is no control data suggesting that ANCA negatives behave any different, that is the general impression. That is all I can tell you.
So at any rate, my take home from this is that IVIG has some biologic activity. It is of relatively brief duration, and it lasts less than three months. I still will bring this out of the hat to treat patients when disease activity persists but either there is an intercurrent opportunistic infection, or there is significant leukopenia or breaches of the host defenses that would preclude safe escalation of glucocorticoids or cytotoxic drugs. That is the kind of the niche with this. Remember that reversible rises in creatinine can be seen with high does IVIG, and these creatinines may go up to 3, 4, or more, particularly in those with underlying glomerulonephritis, but they did come down in all of them.
Apheresis, you know it is has gone from golden therapy to whipping boy and now we're coming back again to take another little look at it. There have been several randomized controlled clinical trials, all small, looking at apheresis in varying ways and varying diseases. Although the homogeneity of these studies - they don't have a high degree of fit for meta-analysis. I think that when compiled together there is a suggestion from several of these studies, particularly Loic Guillevin and Charles Pusey, that apheresis may be a benefit in certain instances, particularly oliguric renal failure and perhaps in the patients with life threatening pulmonary hemorrhage. We certainly know that it is effective, at least from large observational studies and anti-GBM disease.
And there will be a meta-analysis reported this year of these data. I have not seen the manuscript, and I have a lot of reservations about fitting these studies together. And that is my kind of small take home about apheresis. It is used infrequently but may have a place here. Now, in treating other infection-associated vasculitides, there may be far more of a role. These are immune complex diseases, particularly HBV and HCV, and the data of Guillevin again has suggested that there may be a role in treating those very rare cases of HIV-associated necrotizing vasculitis.
Finally, lets wind up and talk about anti-TNF therapy. There is no doubt that TNF is a major cytokine involved in granuloma formation. We know this from diseases such as leprosy, where we can switch from the tuberculous to the lepromatous form. In the tuberculous form, with well-formed granulomas that are very much of TH1 type completely with tumor necrosis factor, and when the switch occurs, we see a TH2 response with minimal TNF.
In animal models of microbacterial disease, you can neutralize TNF and disrupt granuloma formation at some cost. But there is enough rationale, combined with the fact that in Wegener's granulomatosis there is enough regulation of TH1 type of immune responsiveness - TNF can be seen within the glomerular lesions and within the granuloma - and thus form at least enough of a rationale for limited therapeutic trials of anti-TNF therapies in patients with this disease.
You will see that in San Francisco there will be several open trials of reported anti-TNF therapy in all types of vasculitis. They are all positive, and they are all uncontrolled. There are two randomized, controlled, clinical trials which are underway right now, and these will be extremely valuable to us. The only manuscript in press is the pilot data for the WEGET trial, published by John Stone and Gary Hoffman just a couple of months ago in Arthritis and Rheumatism.
I will review this with you briefly, and I will stop for questions. First of all, this is a pilot study. It was done to assess safety. Always be cautious, as you well know, in trying to interpret efficacy and outcome from a safety trial. We always do it, and there are probably a few little things we can take away from this without being zealots in interpreting the data. Basically, these patients had to have documented Wegener's. They had to have activity of BVAS greater than 1, couldn't have any of the usual types of problems including OI, pregnancy, etc. And, basically, they took patients on stable therapy and added to that etanercept 25 mg twice a week and followed them for six months. They all got PCP prophylaxis - the usual stuff. BVAS, globals and ANCA were looked at.
The study cohort was 20 patients, split for sex, mean age 47. These patients had established disease, as one would expect nearly three-quarters were ANCA positive. A third of them had been in prior remissions and were relapsing. The BVAS at entry was 3.61 to 8. Moderate disease activity. If one looks at these patients, you just get kind of a sense of what they look like. Most of them had active upper airway disease - a small amount of active lung involvement. A fifth of them had active glomerulonephritis. There were a couple of patients with bad, nagging, orbital disease that plagues us all. How do we deal with this terrible problem? And then kind of a mixture of other types of problems.
Previous therapy - if one looks at the percentage of the patients who are on adjunctive medications, the vast majority, virtually everyone, was on prednisone except for two people, and virtually everyone except for a couple of people were on a second agent ranging from cyclosporine to cyclophosphamide. Most of these patients were treated by the gold standard.
These are the patients that I think we can say something about, maybe, efficacy-wise: 14 out of 20 only had the etanercept added to their stable protocol. The other 6 had numerous manipulations of medications in and around starting the etanercept, and I don't think you can say a whole lot from them. And they were followed for six months.
This is the bottom line in terms of BVAS. These are all patients in the light sea green here, and the blue are the 14 patients that only had etanercept added. And one can see that, from beginning to end, there was a significant and impressive drop in BVAS in the whole group and, in particular, those who had only etanercept added. A BVAS of less than 0.5 is not much disease activity.
Now the devil is in the details here. If one looks at remission, 5 of 14 of these patients on the etanercept only did achieve complete remission at six months with a BVAS at 0. Two of them got off of prednisone. If one looks at the prednisone therapy in these 14 patients, it went from 12.9 to 6.4 mean dose at the end of six months, but there was really no effect on acute phase reactants or ANCA. Now, I do not want you to think that patients just went from top to bottom in six months and just whipped through this disease. It was not like that. Really three-quarters of these patients had kind of this see-saw staccato fall in disease activity where there were intermittent grumblings of disease on their way from 3 to 0.5 BVAS.
Ten of these 15 patients who had this kind of see-saw reduction had only minor flares. Five of them had BVAS's greater than two, which is a significant flare. Three of these 15 patients had major flare-ups. Two of them were this terrible orbital disease. One was this patient with worsening and persistent GI vascular insufficiency and glomerulonephritis. Heartening to us, because as we are concerned about opportunistic infections and the reaches of host defenses induced by anti-cytokine therapy, we didn't see much in this open trial. Injection site reactions were as you would expect. Two infections and one patient was heavily immunosuppressed. The cytopenias were all in patients with cytotoxics on board and were ascribed to that. And there were five hospitalizations in two patients, and these were patients with really grumbling diseases. In this early trial, no neurologic complications or pancytopenia. So that was our thought.
The last agent I will discuss is trimethoprim/sulfa. There was a randomized, clinical trial of trimethoprim/sulfa in ANCA-associated vasculitis that deserves to be mentioned. It's the Stegeman trial published in the New England Journal five years ago. We sometimes forget about this. They took all patients with ANCA-associated vasculitis who made it to remission. Then they took half of them and put them on placebo, and half of them and put them on Bactrim (trimethoprim/sulfamethoxazole), one double-strength twice a day, and followed them. There was a statistically significant protective effect of Bactrim for relapse in this group, which is impressive to know and it's good. But when we look at the study again, the only relapses that it was effective at reducing were upper airway relapses. It did not protect the renal or pulmonary relapses and, thus, the flavor of Bactrim again comes out - that you know it has a peculiar predisposition to be effective in patients with upper airway disease. And this is still in the process of a large randomized trial from the Europeans. There will be a follow-up study of this from Stegeman again in San Francisco, looking at a moderate size cohort with very limited disease, again claiming efficacy particularly with those who are staph carriers in their nose, which you know is one of the peculiar and perplexing findings that have been reproducibly reported in this disease.
So I think there is a place for it, upper airway, particularly long-term maintenance of patients with long-term upper airway. Can you transplant patients with these vasculitides? Yes. Can you predict who is going to relapse? No. Is ANCA persistence a contraindication to transplantation? No. This is the study of Knockman pooled analysis of 127 patients giving these conclusions. You can pull the reference out for yourselves, but I wanted to give you the bottom line. I think that is my last thought.
You know what will we see in the next wave. We will see the continuing of the honeymoon effect of stem cell transplants. We will probably see more cytokine-based therapy. We will certainly see more anti-cytokine based therapy as well, and I guess I will stop and open this up for questions.
Questions and Answers
Len, thank you for a wonderful presentation. It is my experience that the relapse is never as bad as the initial or often not. So with the greater good in mind, where you can pulse them, you can go in again with Cytoxan for a month or two and get out with the step-down therapy. Doesn't that make the most sense in winning the battle and winning the long- term war?
I think that that is a strategy that many patients chose, and I have certainly done myself. I think it depends on the individual patient and what they have been through. For the patient that has had grumbling disease in the upper airways and the constitutional symptoms, I think that that is often quite clearly preferred; for the patient who has bought themselves a ventilator and has had pulmonary hemorrhage, those memories linger quite largely, and so I suppose influenced by your experience in the patients. I think you are right about the relapses. Maybe we are more vigilant and we pick it up earlier, and that is why it's better, but there is an impression that it is usually a little less.
One of the differences between a place like this and NIH, which I know, and possibly between Cleveland Clinic, which I don't know, is that we often meet our patients when they come in with bowel perforations, coronaries, sort of committed lethality before we even see them. One thing that struck me as you were talking about IVIG - we have had a couple of good experiences in the past year of patients with bowel perforations that we were afraid to hit with high immunosuppressants whom we treated only with IVIG until they recovered, and I guess I was stunned when they did recover. And at that point, we could treat them more vigorously once they healed. I would like your thoughts about the early committed lethality and how you handle it.
I think that our experience is very similar to yours and very unlike the NIH. The bias there is you've got to be able to get on the airplane and back and forth for your visits. So that kind of separates the perforated bowels and the ICU patients. I think that the caveats of care of those types of patients are, first of all, it's exactly where I said IVIG plays a role - you know - when the risks of immunosuppression outweigh the benefits of controlling the disease. And it is always a terrible situation. The second point, I would say, is that the systemic necrotizing vasculitis patient who has a high degree of functional disability and who has overwhelming disease activity. I do two things somewhat differently than this stylistically. One, I never let up from administering all immunosuppressives, including glucocorticoids and the alkylator by the intravenous route, for the entire hospitalization until they are ready to go. I leave no margin of error to bile availability of slow guts and things like that. So if you've got an IV line, give this intravenously.
Secondly, for those patients who have the highest degree of disease activity and the highest degree of functional disability, I use mega-prophylaxis for opportunistic infection. All of them are on trimethoprim/sulfa. I use fluconazole 100 mg every other day. I use acyclovir 800 mg twice a day with normal renal function. Through the duration of this often blistering immunosuppression and often giving higher doses of Solu-Medrol on a q.8 or q.6 hours. This is where I think we get into trouble. Are there any control data for that? Absolutely not, you know, but we should learn a lot from our oncology colleagues who come in blistering immunosuppressants very frequently and this, stylistically, is very much like they do. I've used this as kind of my standard approach to that type of patient. For the most part, without the bowel perforations and infections, it works okay.
Len, let me take you back to ANCA. When that was first discovered as a very important hallmark of vasculitis, people started to think maybe we'll start to understand how vasculitis occurs, maybe these are antigens and so on. What have we learned from all the ANCA information? Where are we with ANCA? In the work you showed here, it didn't matter really whether they were ANCA positive or ANCA negative in terms of which therapy apparently was given. Can you give us some ideas of what you think is going on with the contribution of ANCA?
Well the first thing I'll say is that April 22 to April 25 is the International ANCA and Vasculitis Conference at the Cleveland Clinic.
I promise to be there.
This will be a great collection of vasculitis scholars, and it is going to be a great meeting. So I will condense what I think I'll find. ANCA, you know, has been a dramatic kind of glue to the field. It has brought people together. This is the tenth one of these conferences. At this late date, there are still camps you know, and there are still dramatic differences of opinion. There are differences of opinion where I work. The fellows who are working with Gary, you know he uses ANCA in the diagnosis and, then, I don't think ever orders another one for the duration, doesn't look at it ever again; he could care less about it. If they are working on my side of the clinic, I want to know what the ANCA is during the course of the disease because I feel that there are data to suggest that it is an independent risk factor for relapse.
In terms of pathogenesis, there are so many reasons to suspect that ANCA can be involved in endothelial injury with its ability to induce and release inflammatory cytokines. Yet your point is well taken. The patients that never had ANCA detected seem to have the same ugly disease as the other ones, and strategies that are designed to simply neutralize ANCA have not been overwhelmingly positive. So I think it is still a question that is unanswered, but you know people are continuing to feel the elephant with this.
It would be very nice to be able to read all these abstracts. The only trouble is that you have to have another set of eyes. Whatever was put out by the American College of Rheumatology is a disgrace. I found out just today that all of us have the same eye problem.
Any new observations on the use of thalidomide?
Thalidomide in vasculitis? Well, the model of using thalidomide in vasculitis is you know ENL, which pathologically is a necrotizing vasculitis in the deep dermis, and it makes it an appealing target. We have been trying to recruit an open study for the use of thalidomide in patients with pure cutaneous vasculitis for a long time right now. We have used it in two patients without any obvious clinical improvement. My concerns about thalidomide in most of these conditions are, if all the vasculitides are prone to it have peripheral nerve involvement and as you are well aware this is a toxicity, we must really be vigilant in the use of thalidomide. I think my enthusiasm for the use of thalidomide in vasculitis is pretty low right now, but I look forward to future findings.
How does trimethoprim/sulfa work with these conditions?
Well you know that is a great question. How does trimethoprim/sulfa work in this? It has some anti-metabolite capacities. It is an antimicrobial. The U-Mass group has a very nice study. Some people says that it works. Actually in the Stegeman trial that will be reported at the ACR, if you are a staph carrier, you are more likely to respond that if you're not. Now it wasn't perfect, but it was statistical saying, "Wow, maybe the staph is actually doing something." Maybe it's priming neutrophils and making them ready to spew their inflammatory contents. There is a randomized, controlled, clinical trial of Bactroban (muciprocin) in limited Wegener's, which should be able to answer this, you know, if it is a positive study. Taking people who are staph carriers and randomizing them to the topical treatment of their infection versus not. It's the only way that I think it can be separated out because the mechanism of action is not known. It will be an interesting study.