CME Slides Forum - F. Madore

A joint Congress by ERA-EDTA and ISN

THE USE OF PLASMAPHERESIS

François Madore, Montreal, Canad

Chair: Claudio Pozzi, Milan, Italy

Pierre Ronco, Paris, France

madore

Dr François Madore
University of Montréal
Hôpital du Sacré-Coeur de Montréal
Montréal, Quebec, Canada

Professor Ronco, Professor Pozzi, Ladies and Gentlemen. It’s a pleasure for me to be here this morning with you. In the next 20 minutes or so I will try to review with you the literature regarding the use of plasmapheresis in myeloma associated kidney disease and more specifically try to address some of the controversy surrounding the use of this technique for myeloma associated kidney disease.

This is an overview of my talk. I will first review the rationale for using plasmapheresis in myeloma followed by a quick technical review. But most of my presentation will focus on randomised controlled trials and the results of these trials, as well as the results of recent non-randomised studies that have been published. I will then end my presentation with a couple of words regarding ongoing trials and finally, try to end up with some treatment recommendations.

So the rationale for using plasmapheresis in multiple myeloma is to attempt to reduce as much as possible the exposure to nephrotoxic myeloma proteins.
On one hand, you try to remove as much myeloma protein with plasmapheresis but on the other hand you want to suppress protein synthesis with chemotherapy as much as possible as well. So overall, what you hope for is a faster decrease in myeloma protein with a combination of chemotherapy plus plasmapheresis than with chemotherapy alone.

So what’s the effect of plasmapheresis on protein removal and myeloma protein removal?
Well, this is taken from a paper by Cserti and colleagues published in 2007 and you can see the effect of pheresis on serum light chain level. You can see a sharp decrease in serum light chain level following plasmapheresis but you can also see that there’s a very rapid increase in serum light chain level following therapy and this is explained by two factors. Basically the first is re-equilibration between the intra-vascular and extra-vascular compartments and obviously the second factor is persisting tumour synthesis. So you see that within basically 24 hours you’re back to baseline values in terms of serum light chain levels.

This is again, another slide showing the effect of plasmapheresis on myeloma protein. This graph is taken from a paper by Hutchinson and colleagues published in 2006. This graph actually shows the result of mathematical modelling to predict light chain level and serum light chain levels. These mathematical models are very similar to those used for urea kinetics modelling for instance. So on the first curve here you see the effect of plasmapheresis on myeloma protein and you can see that each plasmapheresis session is associated with a sharp decrease in serum light chain level. However, at some point you end up reaching a plateau so plasmapheresis itself can no longer reduce serum light chain levels at some point.
This is the predicted effect of chemotherapy alone assuming a tumour kill rate of 10%/day. So you can see that with chemotherapy alone if it’s effective and working well, you can have a very nice and progressive decrease in serum light chain level over time.
This is the predicted effect of combined therapy. So by combining chemotherapy plus plasmapheresis you can see a faster decrease in serum light chain levels. So this figure really illustrates very well what is the added effect of plasmapheresis over that of chemotherapy.

If we now move on to a quick technical overview of plasmapheresis, well the basic component of any pheresis treatment is first withdrawal of venous blood. Second, separation of plasma from cellular components and the reinfusion of cellular elements plus either autologous plasma, then the procedure is called plasma exchange or replacement with an alternate replacement solution and then the procedure may be called plasmapheresis.

Separation of plasma from cellular components can be performed by either centrifugation or membrane filtration.

This is a typical centrifugation system where you can see that the blood is pumped out from the patient into a centrifuge and as the centrifuge rotates, obviously the plasma will be separated from the cellular components. The plasma will be pumped out of the centrifuge and discarded and the cellular components are going to be re-infused to the patient along with a replacement solution.

This is what a typical centrifugation device looks like. It’s a fairly complex machine that requires specially trained personnel and it’s also a fairly big machine. So it is a bit cumbersome to move to the bedside especially for instance in the intensive care unit.

This is now a typical membrane filtration system. You see that the blood is pumped out of the patient into a membrane, a membrane which has special characteristics, for instance very large pores that allow for passage of large molecules so that the plasma can be discarded. The cellular elements are re-infused to the patient along with a replacement solution.

This is what a typical membrane filtration device can look like. Actually, this is a machine used for CRRT in the intensive care unit. The great thing about membrane filtration is that it can be performed on standard dialysis machines after some adjustment using specific membranes.
The good thing also about membrane filtration is that it can easily be performed at the bedside of the patient in the intensive care unit, for instance.

Overall centrifugation and filtration have a similar efficacy. The only differences have to do with cost of equipment, mobility of equipment and type of venous access.

A standard of therapy usually lasts for up to four hours and removal up to 4-5 L of plasma. Usually the amount of plasma is replaced by the same amount of either albumin, fresh, frozen plasma or a combination of these solutions.

So now if we move on to the review of randomised controlled trials and other non-randomised studies,

well there have been three randomised controlled trials published on the use of plasmapheresis over the years. Two of them were published more than 15-20 years ago and obtained controversial results and actually opposite results. The last recent study was published by Clark in 2005. There was a lot of expectation with this study hoping that it would end the controversy regarding the use of plasmapheresis. Unfortunately, this study was not definitive enough so the controversy is still ongoing.

So in the next couple of slides I’m going to review these three studies with you.
The first study was an Italian one by Zucchelli and included a population of 29 patients with a diagnosis of multiple myeloma made within the last 14 months prior to the study initiation. 80% of these patients had been treated previously with cytotoxic drugs. All patients in this study were treated with hydration and chemotherapy which was melphalan and prednisolone.

Group 1 patients were randomised to plasmapheresis and group 2 were randomised to peritoneal dialysis as needed.

So 14 patients were randomised into the control group and 15 patients were randomised into the pheresis group. Most patients were fairly severe cases as seen here by the percentage of patients requiring dialysis at onset which was close to 80% in that group and 86% in the pheresis group. The very striking result was the number of patients interrupting dialysis: 18% in the control group versus 84% in the plasmapheresis and this was obviously statistically significant. Also the serum creatinine obviously followed the same pattern so it was statistically significantly lower in the pheresis group.

The patient survival was also statistically significant and statistically better in the plasmapheresis group than in the control group.

So the authors concluded that plasmapheresis can provide significant improvement in both renal outcome and patient survival.
Well, there were obviously limitations to this study. First it was a combination, a mixture of newly diagnosed and existing patients with myeloma so it’s very difficult to make a clear interpretation of the results.
Second in group 2, patients were randomised to peritoneal dialysis versus in group 1, they were randomised to hemodialysis. So was the survival effect related to dialysis treatment or to plasmapheresis? The question is open for discussion.

Now if we move on to the second study by Johnson and colleagues in the United States, this study included 21 patients and again patients with a diagnosis being made up to 14 months prior to this study. In addition patients were excluded if they showed prior resistance to melphalan and prednisone. Again, all patients were treated with forced diuresis and chemotherapy.

Patients were randomised to either plasmapheresis or to hemodialysis as needed. 10 patients were randomised into the control group, 11 patients were randomised into the pheresis group.

The number of patients requiring dialysis at onset was 50% versus 63%. So maybe the patients in the pheresis group had slightly more severe disease because of the higher percentage here. If we look at one of their main results, which was the number of patients interrupting dialysis, it was 0 in the control group versus 43% in the pheresis group. However, because of small number size this fairly large difference was not statistically significant.

If we now look at survival, well at 30 months there was basically no difference in survival between the two groups. So the authors concluded that plasmapheresis in their hands does not provide any significant improvement in either renal outcome or patient survival.

Obviously there were limitations to this study, the first one being the small sample size. So, despite a difference of 40% in renal improvement, this was not significantly because of small sample size.
Also more severe cases were randomised to the plasmapheresis group, so this can also explain some of the results.

So the last randomised controlled trial that we have is the one done by Clark and colleagues and published in 2005. This study was performed by the Canadian apheresis group. It was a multicentre study performed in Canada with more than 10 centres recruiting patients and the total number of patients recruited was 97. All patients were new cases of myeloma with exclusion of patients with prior treatment.

The study population was really very different from the two other trials because here we have only new patients. There are no patients with a prior diagnosis. No patients with prior chemotherapy. So this is a totally different population. All subjects were treated with chemotherapy so either melphalan/prednisone or VAD. Patients were obviously randomised to either plasmapheresis or no plasmapheresis and hemodialysis as needed. So 39 patients were randomised to the control group, 58 patients were randomised to the pheresis group. This imbalance can be explained by the randomisation process that the authors used. Actually they used a randomisation scheme by centre, so some centres ended up recruiting very few patients, whereas other centres ended up recruiting a lot of patients. So this explained this imbalance.

The number of patients requiring dialysis at baseline was 36% versus 26%. So we see here that these are less severe cases compared to the prior two studies.
The number of patients interrupting dialysis during the study was 50% in one group and 66% in the other group which was not statistically significant although there was a trend.

If we now look at survival, well, at 6 months during the study there was basically no statistically significant difference in survival.

So Clark and colleagues concluded that plasmapheresis does not provide any significant improvement in either renal outcome or patient survival. There were also some limitations to this study. Despite being the largest study, despite being a multicentre trial, despite being one of the best designed studies up to date, there were limitations and one of them is that there were no biopsy results. A patient did not need to be biopsied to be included in this study. So it’s possible that some of these patients were not myeloma cast nephropathy but rather acute tubular necrosis or other entities. In addition there was no light chain level measurement performed during the study because when the study was done, light chain measurement was not really available throughout all centres.

So what are the conclusions from the randomised controlled trials that have been published? Well, we still have two small trials that reached opposite results and these raised at the time of publication a lot of controversy. We also have one larger negative trial by Clark and colleagues and it was hoped at the time that this study was published that it would end the controversy. Unfortunately, it did not end the controversy. The controversy is still ongoing.

Just to fuel the controversy I would say Leung and colleagues published in KI in 2007 a non-randomised study, an observational study which was a retrospective analysis of 40 of their cases and they showed that patients with good response to plasmapheresis and chemotherapy have an excellent recovery rate of 75%. So patients with free light chain reduction with pheresis greater than 50% had a 75% recovery rate. So this is excellent and it suggests that there is probably a subset of patients who may respond very well to plasmapheresis and chemotherapy. So this is something to remember when interpreting the literature.

There are some ongoing trials that will hopefully provide more data to end the controversy and provide more solid recommendation. So the first ongoing trial is the MERIT trial which is the Myeloma Renal Impairment Trial. It’s a trial led by Doctor Gaskin in the UK and it plans to enroll 280 patients over the next couple of years. All patients will receive dexamethasone and cytotoxic chemotherapy and will be randomised to either plasmapheresis or not. The primary endpoint is renal recovery.

There is also another trial ongoing which is the European Trial of FLC removal by extended hemodialysis in cast nephropathy. This trial is not about plasmapheresis itself, it’s rather about a new technique called extended hemodialysis with high cut-off membrane. So for those of you not familiar with this technique you’ll hear much more about it in the next session by Doctor Hutchinson. But just to say a few words. This new technique is actually hemodialysis with special membranes that allow for passage of large molecules and hence passage of myeloma protein. With this novel technique you can reach excellent elimination of myeloma protein, actually you can get even greater elimination of myeloma protein than with plasmapheresis in some circumstances. So this is a very interesting technique. They’re planning to evaluate this novel technique with a randomised controlled trial and they hope to enroll 90 patients with a primary endpoint of independence from dialysis. Although this is different from plasmapheresis it is still similar in its intent in the sense that it wishes to eliminate rapidly excess myeloma protein while chemotherapy kicks in and provides its effect.

So if I can end up with some treatment recommendations, well, the benefits of plasmapheresis still remain to be demonstrated unfortunately. We have 3 randomised controlled trials and we have several non-randomised studies but despite these studies the benefits of plasmapheresis have not been convincingly demonstrated up to now. One thing that we can conclude though is that plasmapheresis should not be performed in all new patients with myeloma. We need to be selective. Plasmapheresis is a technique associated with a lot of side-effects, so it’s something that should not be done in every incoming patient with a new diagnosis of myeloma. However, for patients in whom aggressive immunotherapy is contemplated, plasmapheresis may be an option. There is certainly a subset of patients who may benefit from plasmapheresis.

But in these patients if you choose to go ahead with plasmapheresis, I suggest that renal biopsy should be obtained first to make sure that you’re treating a myeloma cast nephropathy and that you’re not treating a tubular acute necrosis or another entity. So renal biopsy should probably be obtained first in all patients in which plasmapheresis is projected or attempted. Second, light chain measurements should be obtained to monitor the effect of plasmapheresis, to monitor whether or not you’re effective in reducing serum light chain level with plasmapheresis and to adapt the intensity of your plasmapheresis treatment. Hopefully, within the next couple of years, we’ll get the results of the ongoing clinical trials and we will be able to have a definite answer regarding the effects and the benefits of plasmapheresis. Then, hopefully, we’ll be able to provide more precise and better treatment recommendations.

Chairman: Thank you very much Doctor Madore for this very comprehensive review of plasmapheresis trials we have time for one or two questions. Yes?

Question: Thank you for an excellent presentation. Do you have any absolute recommendation for starting plasmapheresis in myeloma patients? For example, our level of light chain or IgA? We have something in their references that we have some points but for patients for starting plasmapheresis what are the absolute indications and/or relating indications? Thank you.

Dr. Madore: Well, there are no strict recommendations and I could not give you a level of serum light chain at which for instance that you would say, ok we need absolutely to perform plasmapheresis in this patient. However, the higher the serum light chain level, the more effective or at least the better should be the effect of plasmapheresis but I cannot provide you with let’s say a specific number.

Dr. Madore: Yes, we don’t have any data and maybe with the ongoing trials, we’ll get some data. We would need some data to say ok with such a level of light chain in the plasma you can go ahead with plasmapheresis and you’ll have some benefits. But we don’t have this yet.

Chairman: Thank you very much if there are no other questions we’ll move on to the last presentation.