CME Slides Forum - R. Glassock

A joint Congress by ERA-EDTA and ISN

EVIDENCE-BASED THERAPY FOR IDIOPATHIC MEMBRANOUS NEPHROPATHY-2009

Richard Glassock, Los Angeles, USA

Chair: Dontscho Kerjaschki, Vienna, Austria

Claudio Ponticelli, Milan, Italy

glassock

Prof. Richard Glassock
The David Geffen School of Medicine
at UCLA
Los Angeles, CA, USA

Thank you very much Claudio and Doncho and thank you Pierre for asking me to come to this meeting and discuss with this audience the 2009 status on evidence-based therapy of idiopathic membranous nephropathy which I’m going to try to do in about15 minutes.

Now it would be very appropriate to begin in a talk regarding evidence to understand a little bit about how evidence is judged. Beginning about 6 or 7 years ago the group at McMaster University headed by Professor Guyatt began to systematise the ways of assessing the quality of evidence and they came up with what’s called the GRADE criteria, which divided evidence into 4 basic groups A, B, C and D or high quality, moderate quality, low quality and very low quality.

They assigned to each of these kinds of descriptive terms such as for grade A that further research was unlikely to change the estimate of an effect, such as might be seen in a very large well powered and extremely well designed randomised controlled trial. At the opposite end of the spectrum is evidence in which the estimate of the effect is very uncertain, for example case reports and case series. I’m going to use this framework to explain the current status of evidence in membranous nephropathy, evidence for management.

Now it’s appropriate to begin with 2 caveats. First the Parachute Paradigm. One does not need a randomised controlled trial to prove the efficiency of a parachute to overcome the effects of gravity. By that I mean there are some trials that are not randomised that are just so convincing that you don’t need grade A randomised controlled trial evidence. On the other hand, we have to understand that a personalised medicine and comparative effectiveness are intention with each other. That is to say that evidence assessment, the kind of evidence that I’m going to be discussing approaches groups of patients in which the effect of treatment is measured on an average patient meeting the entry criteria of the trial. This kind of evidence does not fully acknowledge the variability observed in individual patients which are the kind of patients that you take care of on a day to day basis.

Now, Claudio has introduced this talk with a consideration of the natural history of membranous nephropathy and from that flows logically the goals of treatment. First of all, we want to delay in some measurable way the progression to end stage renal failure and of course, ideally to prevent it entirely. We also have to take into account the complications of the nephrotic state itself taking measures to avoid them, such as thrombosis infection and perhaps even atherosclerosis. We also, of course, the principal of -- avoid serious adverse events from therapy itself, such as infection or drug-dependent nephrotoxicity.

Now the remainder of my talk is going to be divided into 2 components; the old and the new therapies. By the old therapy I’m talking about cyclical alkylating agents, the so-called Milan or Ponticelli Regimen and calcineurin inhibitors. The new therapies are ACTH, mycophenolate mofetil and rituximab.

But before I discuss the evidence surrounding these 5 forms of therapy, I do want to comment that as of now glucocorticoids alone, monotherapy with glucocorticoids for at least 6 months or less are ineffective in the management of idiopathic membranous nephropathy and the evidence underlying that statement is Grade A.

well the cyclical or alternating steroids and cytotoxic regimen first popularised by my good friend and colleague Professor Ponticelli and his colleagues involves 2 cycles. The first cycle given at months 1, 3 and 5 are intravenous methylprednisolone or oral prednisone and the second cycle given at months 2, 4 and 6 are either chlorambucil or cyclophosphamide given in the doses shown. Most people nowadays I think use cyclophosphamide in preference to chlorambucil. Over the years the group here in Milan have produced a number of papers and conducted several randomised controlled trials.

This slide summarises the cumulative evidence base for the efficacy of this regimen. The first event response to this form of therapy in idiopathic membranous was about 83% with the complete and partial remissions about equal. The last follow-up, which for many patients was more than 10 years up to 20 years, the response was 74% indicating obviously that relapses do occur.

In the final analysis complete remissions lasting for a decade or more could be achieved in around 35-40% of the patients. Renal survival was outstanding in the patients treated with this regimen. 98% at 10 years with censoring by death of the patients with renal survival not censored for death was still excellent at 92%.

The side effects of this cyclical regimen are rather mild and occurred only in 9% of the patients, no case of hemorrhagic cystitis and you’ll note no cases of malignancy. The cumulative dose of cyclophosphamide in this 6-month cyclical regimen is only 180 mg/kg, which is far below the usual recommended dose of cyclophosphamide to avoid the common side effects of gonadal toxicity, alopecia and alike.

Now doctor Ponticelli’s work in Italy has been confirmed in India by Doctor Jha who reported on a 10-year randomised prospective trial comparing the methylprednisolone-cyclophosphamide regimen to control and as you can see here, like Professor Ponticelli’s work remissions were very common in the treated group but, of course, did occur spontaneously in up to 35% of the patients in the control group.

So the evidence in 2009 for cyclical cyclophosphamide and steroids are that they are effective and relatively safe for the treatment and we have long-term follow up the evidence here is Grade A.

Now calcineurin inhibitors have also been studied. I’ll show you two trials only the first by Cattran and colleagues reported in 2001 which involved the use of cyclosporine for 26 weeks in a group of patients with idiopathic membranous nephropathy and at the conclusion of the 26 week treatment period about 72% had gone into a complete or partial remission, whereas only about 22% had entered into remissions in the control arm. However, as soon as the cyclosporine was stopped, a fraction of patients underwent relapses and they continued to relapse such that by the end of the last follow up at 78 weeks those remaining in remission were only about 40% compared to about 14% in the controls. Still a statistically significant difference but the difference between the two groups was narrowing with time due to relapses.

Now doctor Praga in Spain has studied another approach using calcineurin inhibitors that is tacrolimus monotherapy in which tacrolimus alone without steroids was use. This was a rather short term trial, 18 months in duration but a clear cut statistical difference in complete and partial remissions was seen when tacrolimus was administered compared to a control group. Importantly even though the follow up was rather short in this trial a 50% decline in eGFR was seen very uncommonly in the tacrolimus group, whereas it was found in about 28% of the control group.

So calcineurin inhibitor, either cyclosporine or tacrolimus, is effective and probably safe for the treatment of membranous nephropathy but the follow ups are short, less than 5 years, relapses are common. So I’ve graded the evidence here as grade B, not quite as good as that for cyclical therapy.

This is not the natural ACTH that I’m going to be reporting on but rather the synthetic form the ACTH (1-24AA) in idiopathic membranous nephropathy. The first report of this was by Berg and colleagues, a small short-term trial of only 5 patients treated with parenteral Synacthen-depot which is available only in Europe, it’s not available in the United Sates. You will note in this study that there was a progressive increase in serum albumin, a progressive decline in albuminuria over this short-term trial with twice weekly injections of the synthetic ACTH.

Doctor Ponticelli and his colleagues conducted a randomised comparator trial in which cyclical methylprednisolone and cytotoxic agents for 6 months were compared to synthetic ACTH twice a week for 1 year.

They found in this trial that complete remissions were strikingly more frequent with ACTH. Partial remissions were less common. ACTH shown here in red such that the overall response rate was roughly equivalent although the ratio of complete and partial remissions was different. Relapses, although the follow up is relatively short in these patients, seem to be less frequent in the ACTH group.

An additional randomised controlled trial with rather short-term follow up was also reported at the ASN meeting a few years ago. A small trial, 15 patients in each group, not a comparator trial, a trial comparing untreated and ACTH treated patients with striking results 14 out of 15 patients undergoing complete or partial remission and the ACTH treated group with only 2 complete or partial remissions in the untreated patients.

Thus I think the evidence in 2009 is that ACTH is of benefit in producing remissions in idiopathic membranous nephropathy but its long-term effects are not yet know and for that reason I’ve graded it evidence Grade B.

Well as you know, here in Bergamo are the groups of Remuzzi and Rugginenti reported in a small uncontrolled trial about 6 years ago that rituximab would reduce the magnitude of proteinuria in patients with membranous nephropathy. It did not commonly result in complete remissions as you can see, because of persistent low grade proteinuria but at least over the period of observation there was no evidence of progression.

My good friend Fernando Fervenza from the Mayo Clinic in the US sitting in the front row here has allowed me to show you this slide which is representative of their pilot trial, still underway, not a randomised controlled trial a proof of concept pilot trial again showing that rituximab given in doses of 1 g times two and the proteinuria shown in a log scale here results in complete remissions and some partial remissions and others and has no effect in other patients.

The summary of the 14 patients they had treated, they were able to get a 65% complete or partial remission rate with 3 out of the 14 patients achieving a complete remission. 2 patients had progression to ESRD. There are no controlled trials of rituximab to date although there are many efforts to achieve a randomised controlled trial and maybe one will be done in the foreseeable future.

So at the present time we would have to say that rituximab may be of benefit in producing remissions in idiopathic membranous nephropathy but its long term effects are not yet know and I would give it an evidence grade of C using the Grade criteria.

We have several small controlled trials. This is one from Chan reported a few years ago comparing MMF plus prednisone versus cyclical cyclophosphamide and methylprednisolone and you will see that there was no difference. The two forms of treatment were roughly equivalent over a short-term follow up of only 15 months with relapse rates approximately equal in the two groups.

Another study was reported by Nayagam comparing again MMF plus methylprednisolone pulses and steroids to the Ponticelli cyclical regimen showing a steady decline in urinary protein excretion as shown here by the urinary protein/ creatinine ratio over the 12 months of follow up.
Overall complete and partial remissions in this study were 64% for MMF and 80% for the Ponticelli regimen, not a statistical difference.

Now another trial was reported just last year from France by Dussol in which MMF monotherapy without steroids, MMF alone was compared to a control or untreated population of patients and you can see here rather dismal results, a complete remission in only one of 18 patients. Partial remissions in 6, not different from the control. No evidence of efficacy for MMF given without steroids.

In an uncontrolled or a quasi controlled study conducted by Branten in the Netherlands compared cyclophosphamide given orally for 1 year, not in a cyclical way with MMF orally both given with steroids. This is a historical comparison and you’ll note in this trial at the end of 12 months there was no difference between the MMF and cyclophosphamide treated group but subsequently to the relapses there was an increasing divergence of the protein/creatinine ratio.

In fact, the relapses after an initial partial or even complete remission were very high in the MMF treated group compared to the cyclophosphamide treated group. So one of the disadvantages of MMF although it may be initially affected is like calcineurin inhibitors has a high relapse rate.

So MMF with the understanding that it must be administered with steroids may have benefit in idiopathic membranous but due to frequent relapse upon discontinuing therapy there may be a limitation and its long-term effects are not yet known and for this reason I’ve given it an evidence grade B-C.

So in 2009 we have 5 evidence-based treatment options. Cyclical oral cyclophosphamide and steroids, perhaps daily cyclophosphamide and steroids although that latter approach has not been studied in a randomised fashion. Calcineurin inhibitors, ACTH, rituximab and MMF plus steroids.

The problem for the future is how to resolve the issues that I’ve listed here.

What is the ideal initial therapy and how should individualised treatment decisions be made? Given the fact that we have 5 forms of treatment that may be effective.

Can initial therapy be safely postponed to permit the observations that Doctor Ponticelli indicated may be needed for spontaneous improvement? If so for how long should we observe the patients? In the case of relapsing a refractory disease what’s the ideal sequence of therapeutic regimens? The answers to these questions are not available.

Now you will note in my discussion I’ve only given you the evidence and the quality of the evidence. I’ve not made any recommendations. The reason for that is that KDIGO clinical practice guidelines for the management of GN which will translate this evidence into recommendations is now underway and will report some time between 2010 and 2011. Although I have opinions regarding this and I’ve written about it for many years, I think it would be better to withhold recommendations until this group has examined this evidence very critically and very thoroughly and provide you with a solid evidence-based recommendation list for the treatment of membranous nephropathy.

Chairman: Thank you Dick for this very important and balanced presentation. Any questions? May I ask a question Dick about the role of evidence-based medicine? Because evidence-based medicine actually is based on randomised controlled trials but randomised controlled trials are often bad trials. I mean, if I compare a useless treatment because the dosage of the drug is inappropriate, the drug is condemned as ineffective and this should be taken into account by our colleagues when they do recommendations. What is your opinion?

Prof. Glassock: Professor Ponticelli has asked a crucial question. What is evidence-based medicine? What is the practice of evidence-based medicine? Well from my perspective, as a consultant who only sees patients in 3rd and 4th opinions I never ever see an average patient with membranous nephropathy. So much of the collection of evidence which is based on randomised trials in large groups of patients in which entry criteria were designed to maximise the opportunity for a clear and unequivocal result have very limited value because every patient has to be treated in a highly individualistic manner. There are many factors that are not included in randomised controlled trials that come to bear on your decisions. But having said that it’s extremely important that we try our goal ultimately is to do our best for our patients and certainly the evidence that comes form these trials, as defective as they may be, as limited as they may provide information for the practice of medicine I think they need to be taken into consideration. Hopefully groups like KDIGO who are looking to try to provide you with helpful ideas on how to manage patients will take that into account.

Question: What is your advice about duration of therapy with calcineurin inhibitors?

Prof. Glassock: Well the time was limited. In my opinion if there is no evidence of a complete or partial remission of proteinuria after 4 to a maximum of 6 months of treatment with a calcineurin inhibitor, you’re better off to discontinue it because the likelihood of seeing any benefit beyond 4-6 months is minimal and the likelihood of toxicity is much greater. Now if you have a patient who develops a complete or partial remission on a calcineurin antagonist do not stop the drug. Reduce the dose gradually to the minimum effective dose. Now that’s a very difficult task because you often don’t know what it is but reduce it to a level where toxicity is minimised. That be 1 or 2 mg/kg/day for cyclosporine, much less for tacrolimus. Follow the patients carefully for features of nephrotoxicity and perhaps then discontinue the drug after 4 or 5 years of remission.

Prof. Glassock: Well I told you that monotherapy with steroids is ineffective but if you’re going to use steroids with calcineurin inhibitors or with MMF, a low dose alternate day is probably adequate 20-40 mg every other day. Professor Ponticelli has outlined exactly what the steroid dose is in the cyclical regimen.

Prof. Glassock: I don’t want to prevent the free communications from getting their turn at the podium.

Question: What’s your recommendation for young female patients or paediatric female patients?

Prof. Glassock: As Claudio indicated, the female patients have a better prognosis, a higher likelihood of spontaneous remission. Unless they are extremely symptomatic with nephrotic syndrome, just disabling oedema marked hypoalbuminemia at high risk for thrombosis I’m perfectly content to observe them for rather prolonged periods. Up to 1-2 years. Now these are generally women with protein excretions between 3-6 g/day. I would not treat a young woman like that. I’d also work them up extensively for lupus. Now knowing that they have lupus but its presenting in only a renal fashion doesn’t help you all that much but it may help you in the future in anticipating lupus in a systemic form. So I’m very conservative in women with membranous unless they’re very asymptomatic.

Question: What about a patient who has a proteinuria less than 4 g? do we have evidence that whether you treat them or not they have the same outcome membranous with a proteinuria like a 3g or 4g?

Prof. Glassock: Well I think Doctor Ponticelli gave you the evidence and I believe that the evidence suggests that there’s no need to consider treatment in membranous unless the urinary protein excretion time average is greater than about 5 g/day. 3-4 g/day with mild hypoalbuminemia, easy to manage oedema and a serum albumin above 2.5 does not require treatment in my opinion and I think Doctor Ponticelli agrees with that. The evidence does not support active specific management of that kind of patient.