CME Slides Forum - D. Jayne

A joint Congress by ERA-EDTA and ISN

THE ROLE OF RITUXIMAB IN VASCULITIS THERAPY

David Jayne, Cambridge, UK

Chair: R. Kettriz, Berlin, Gemany

Charles Pusey, London, UK

jayne

Dr David Jayne
University of Cambridge
School of Clinical Medicine
Addenbrooke's Hospital
Cambridge, United Kingdom

Charles, Ladies and Gentlemen thank you very much for this invitation. I think the 2 previous speakers have very clearly outlined the randomised trials that have been done in the last 10 years which have really aimed to optimise and standardise conventional therapies that have been around for 20 or 30 years. But very much part of the strategy is to use this data as a launch pad if you like for the evaluation of newer biologic therapies. The first therapy that we’re interested in is with TNF-α blockade that story remains unresolved and I’m going to tell you about the next biologic become involved with which is rituximab.

I’m going to briefly review observational studies that have been published, what’s going on in terms of randomised controlled trials. I’m going to say one or two words about safety and then end with the current role in ANCA vasculitis therapy.

So it may be a little bit of a surprise that the B cell plays an important role in vasculitis, why should we want to remove the B cell with rituximab? Well, we know there are autoantibodies, there are ANCA which we think are pathogenic in certain circumstances. We know that B cell stimulating cytokines such as BlyS are elevated in vasculitis and to a certain extent correlate with disease activity. We know that B cells with an activated phenotype can be found at sites of tissue injury taking part in the immune and inflammatory response.

So this is data from the group -- which has demonstrated that CD20 positive B cells and B cells expressing PR3-ANCA, the autoantigen as well as CD38 positive plasma cells are all present in the nasal biopsies of patients with Wegener’s. Now we would hope that lots of B cells are present in the kidney as well but in fact B cells are relatively sparse in the kidneys of ANCA vasculitis and that may tell us something about humoral versus cellular influences on the renal pathology.

So I’m sure I don’t need to remind you that rituximab marketed as Rituxan in the US or MabThera in the rest of the world is a chimeric antibody, quite an old antibody now designed to kill B cells and it does this through a variety of different mechanisms and some of these mechanisms are dependent on or are influenced by Fc-receptor polymorphisms. We have the advantage in using a drug that has been given to over 1 million patients in Non-Hodgkin’s lymphoma and that is now licensed for another autoimmune condition, rheumatoid arthritis. So that’s much easier than starting with a new agent.

I’m just presenting a case and I apologise for a case but it illustrates important points. This is a young girl with refractory Wegner’s and you can see on the pre-rituximab slide she has a cavitating lesion in her chest film and a pachimeningitis and extensive sinus uptake on her MR scan. She had failed multiple agents with septic and grossly cushingoid.

She was given rituximab shown here by the arrows at the top and this led to reduction in the activity score shown in the middle panel here, reduction in her ANCA level, reduction in her B cells, the symptoms went away and the steroids were reduced. Everybody was happy. But 90 months later early symptoms of nasal bleeding returned, the BVAS reflecting that and the ANCA levels started to rise. She was given rituximab again and had a good response. So this is the typical pattern that you see when you use rituximab in a patient with relapsing or refractory Wegener’s.

This slide summarises the published observational data. You need to use caution when addressing this sort of data because there’s going to be publication bias, people do not like publishing negative results. Also this table includes both some prospective observational trials as well as some retrospective experience. Clearly these types of studies vary in their quality. But I’ve summarised the results according to whether the patients were in complete remission in green, partial response in blue or no response. You can see that the various trials report varying different response rates but with the majority finding a response in 80-90% of patients.

This trial in the middle which was of 8 patients and only found response in 3 suggested that there may be less efficacy in retro orbital manifestations of Wegener’s. There were one or two problems with this trial and they were relatively under dosed but it alerts us to the possibility that there may be some groups who are potentially less rituximab responsive than others.
In the UK a couple of years ago we conducted a multicentre trial including patients from Hammersmith hospital, from Oxford, from Cambridge and from Great Ormond Street and this comprised 63 patients with refractory vasculitis, mainly Wegener’s granulomatosis. I’ve really summarised the headline results in the left hand panel. So patients came into the trial either with actively relapsing disease shown in orange or with what we call persistent grumbling disease shown in green.
6 months after rituximab 75% of patients were in complete remission. This is a good response for this type of patient group and probably superior to what you would see with reintroduction with cyclophosphamide. Only one patient in this cohort completely failed to respond this patient had retro-orbital Wegener’s but there were 5 others with retro-orbital Wegener’s that did respond. These responses are then reflected in the disease extent index on the right. But note that by 12 months we’re beginning to see black relapses occurring reflecting the case report I showed you earlier.

So this retrospective study has allowed us to ask one or two other questions and one of the questions we asked is well what happens to ANCA? Now, the problem with following ANCA in refractory vasculitis is that only about half of your patients tend to be ANCA positive to start with but we do see a significant fall of ANCA and this is quite predictable. ANCA starts to fall at about 6 weeks after rituximab and reaches a nadir at about 6 months after rituximab. Unfortunately, following ANCA is not a good enough biomarker to judge when you should retreat because you can see we had 14 relapses without an ANCA rise and only 5 relapses with an ANCA rise but if you have a patient whose ANCA is rising that does put them in a high risk group for relapse. So I’m not saying ignore ANCA I’m just saying you can’t re-dose rituximab on the basis of ANCA.

We compared two different rituximab regimens which just so happened to have been evenly distributed in this cohort. So most of us started with the so-called lymphoma regimen of weekly doses for 4 weeks of 375 mg/ m² but then we moved for the sake of convenience to a 2 dose regimen. Now the older regimen gives you an average dose of between 2.5 and 3 g depending on the size of the patient, so the 2 g dose is clearly giving slightly less rituximab but it’s two visits to hospital not 4. What we found in this retrospective study is no difference in terms of survival of B cell depletion and no difference in relapse free survival.

Just look at right hand graph for a moment and you can see that by about 12 months you’re getting relapses occurring in 40-50% of patient’s and in fact the meantime in Wegener’s to relapse after rituximab is 12 months. We also looked at the question of should you continue immunosuppression after you give rituximab? Most of these patients were coming in on mycophenolate or azathioprine or methotrexate or cyclophosphamide and in the early days when we were not confident about whether rituximab would work, we tended to carry on the immunosuppression. When we got more confidence, we started to stop the immunosuppression and this simply compares those two time periods and it demonstrates no difference in relapse free survival, as to whether you continue the immunosuppression or whether you stop it.
Now I do not think this data is the answer to this question and I think this is a valid question for future studies what should you give after rituximab? But my current practice is to stop immunosuppressives when we give rituximab.
So how are you going to deal with the fact that 50% of your patients are going to flare in 12 months and 70 or 80% of the them are going to flare in 2 years? Well, the approach that we’ve taken in Cambridge is to protocolise repeat dose every 6 months. We looked at following B cells and we found that if you dose rituximab on the basis of B cell return, you’re really going to miss about a third of the relapses because they occur before B cells have returned at least as measured by conventional FACS analysis.

So again this is retrospective data but it simply shows two groups of patients, those in the purple line are given pre-emptive repeat doses of rituximab of 1g every 6 months. Whereas those in the green line are just given the rituximab induction course and then no subsequent rituximab. I think you can see there’s a large difference in time to flare.
So again our current practice, I’m not recommending this but our current practice is to use protocolised retreatment but this is controversial.

So yesterday we presented the data of a small phase II randomised controlled trial. Here I’m moving from using rituximab in refractory disease to using rituximab as an induction agent for new patients. So this trial recruited 44 patients. They all had ANCA-associated vasculitis with nephritis, they were elderly, mean age of 68 and they had severe renal disease with a mean GFR of 20 ml/min coming into the trial. Approximately a 5th were dialysis requiring. We had a 3:1 randomisation so 33 received a rituximab-based regimen which included two initial doses of cyclophosphamide and we compared that to a standard cyclophosphamide regimen and we found no difference in our primary endpoint of achieving remission.

The rituximab-based regimen is shown here in red and you can see that virtually all patients achieved the sustained remission endpoint.
So that’s good news, it appears to work in induction but we were slightly surprised by our second endpoint which was a safety endpoint. Because we thought that if we were not exposing these patients to more than 2 pulses of cyclophosphamide, we would see a safety benefit and we saw no safety benefit.

So in the left hand graph I’m showing survival to first severe adverse event and you can see there is no difference and if you look at the proportion with severe adverse events, there is a proportion with infections or the mortality there is no difference between the rituximab-based regimen and the conventional cyclophosphamide regimen.
So our sort of interpretation, our take on this if you like is that most of the adverse events happen early in the first 4-6 weeks when essentially the protocols were the same because they were receiving two pulses of cyclophosphamide but it’s probably the age and the general sickness of the patients that is an important predictor and it’s the high dose steroids as well. Maybe we’ve now optimised cyclophosphamide.

So what else is going on in terms of randomised trials? Well I’ve mentioned the RITUXVAS trial and next month again at the ANCA workshop in Lund where everything will be happening we’ll also anticipate the results of the RAVE trial. The RAVE trial is a larger 200 patient randomised trial and in contrast to RITUXVAS the rituximab group received no cyclophosphamide at all it’s just the same dose of rituximab and oral and IV steroids. So we’re looking forward to that with interest. Unfortunately, right now nothing else is happening in the randomised trial arena with rituximab but clearly there’s a lot of interest in doing further studies.

So what about safety? Well, we have not seen any clear evidence of an increased infective risk. We have seen hypogammaglobuminemia in a small number of patients.

Around a third of patients develop HACA which are antibodies to rituximab and occasionally these do impair the ability of rituximab to re-induce remission.

Then we see infusion reactions and these have been mild to moderate in about a quarter of patients. Severe infusion reactions are rare and with repeat dosing this is 1st, 2nd and 3rd we don’t see any increase in infusion reactions.

So moving on to the summary. I think we have sufficient data now to suggest that rituximab can be considered for patients with refractory ANCA vasculitis.
I don’t feel right now we can recommend rituximab as an induction agent routinely, we need more trials.
I think there are several safety concerns. We’ve got some early safety concerns about possible contribution to infection and then we’ve got late safety concerns about the safety of repeat dosing with rituximab.
The other thing that’s important to mention is that there are many alternative B cell depleting and modulating agents that are available or becoming available.

Chairman: David thank you very much indeed for an excellent talk. This paper’s open for questions and comments from the audience. I can’t see any yet. Yes, there’s one down towards the back if someone could take the microphone down there thank you.

Question: Thank you very much for the presentation. I would like to ask, you mentioned about the refractory ANCA positive vasculitis to give after 6 months another dose of rituximab. In comparison with cyclophosphamide you know because these cases are responding to cyclophosphamide so it is not that resistant or refractory so still did you advise also to give a post dose after 6 months or not? Thank you.

Dr. Jayne: Is your question did we give more cyclophosphamide? Because we did not.

Question: I’m talking about cases which are not refractory, usual cases or induction.

Question: Induction you know you advise also to give after 6 months -- refractory or not?

Dr. Jayne: Well, firstly I’m not recommending rituximab for routine induction because I don’t think the studies have been performed yet. Secondly your question really is what relapse rate will we anticipate after a single course of rituximab? Well, I think we’re going to anticipate a lower relapse rate because induction patients will have a lower relapse risk. They do not have a history of relapse and they’re going to include more microscopic polyangitis patients that have a lower relapse risk. So we do not know. Next year we will year 2 year data on our RITUXVAS trial and we will have some information about the relapse risk.

Question: In the induction trial was it possible that the severity of the disease gave negative results? The severity of disease was less, this is why we couldn’t get a positive response with rituximab versus cyclophosphamide?

Dr. Jayne: I think the randomised trial has told us 2 things at least in a preliminary way. I think the rituximab regimen was effective in inducing remission although the patients were very sick. It was effective but it did not appear to confer a safety benefit. I think you’re correct that is because we’re treating a very sick patient cohort.

Question: Quite a few patients do end up with ESRD. What is the optimal time to think about a transplant in these patients?

Dr. Jayne: This is not a transplant talk I’m going to duck the question, I apologise.

Question: Do you think there’s potential for other anti-B cell antibodies to takes over, perhaps be more effective or have less adverse effects?

Dr. Jayne: Of course, one of the problems we have when using a B cell depleting therapy is what we measure is peripheral b CELLS AND we know that peripheral B cells are killed very easily by rituximab but what we really want to deplete are tissue resident B cells particularly at sites of inflammation. We know they are relatively protected from rituximab induced killing. So yes I think there is interest in producing more effective B cell depleting agents. There are several case reports I’m sure you’re familiar with a case from Birmingham where there is a patient who did not get a good result with rituximab although they had peripheral B cell depletion. They had ongoing B cells at the sites of injury.

Chairman: Great I think in the interest of time we better move on and thank you very much again.