Hello, and welcome to Episode Thirty-Eight of Multiple Sclerosis Discovery, the podcast of the MS Discovery Forum. I’m your host, Dan Keller.
This week’s podcast features part one of a two-part interview with Joseph Berger of the University of Pennsylvania. But to begin, we’d like to tell you about MSDF’s Drug-Development Pipeline.
Twelve drugs are currently approved in the US for the treatment of MS, but there are many more drugs in various stages of clinical and pre-clinical development. We’re keeping daily track of 44 of them in our Drug Development Pipeline.
To visit the pipeline just go to msdiscovery.org and click on Research Resources, and then Drug-Development Pipeline. You’ll find a finely detailed, fully referenced, and easily searchable database of all 44 of those drugs. The database includes details on each drug candidate’s physiology, its progress through pre-clinical and clinical trials, and its regulatory and commercial status.
Science journalist Heather McDonald has managed this database since its inception, and she updates it continuously, whenever new information becomes available. In just the last week, for example, she added one new clinical trial to the database, she updated information on two other clinical trials, and she added 5 other pieces of information. The drugs with important additions and changes were dimethyl fumarate, fingolimod, glatiramer acetate, interferon beta-1a, interferon beta-1b, mitoxantrone, natalizumab, and RPC1063.
Now to the interview. Dr. Joseph P Berger is a professor of neurology and Chief of the MS Division at the University of Pennsylvania in Philadelphia. In part one of our discussion with Dr. Berger, we’re talking about the risk of progressive multifocal leukoencephalopathy (PML), a rare but serious brain infection that occasionally arises in people being treated for multiple sclerosis.
Interviewer – Dan Keller
The topic of quantifying risk and mitigating risk comes up with certain immunosuppressive drugs, notably natalizumab in MS but also with other drugs as well in other conditions. What are some of the confounding factors? Why is this not an easy thing to approach?
Interviewee – Joseph Berger
Well, it’s not easy because it’s so unpredictable. Nobody would have thought that natalizumab would have uniquely predisposed to the development of progressive multifocal leukoencephalopathy. In fact, when natalizumab was introduced, if one would have attempted to predict what would have happened, you might have said, well, we’ll see a wide variety of opportunistic infections of the central nervous system, since this is a drug that prevents the neural immunosurveillance that is necessary to prevent these diseases from occurring. However, that’s not what we see. We don’t see the opportunistic infections of the central nervous system that we see in the AIDS patient; for instance, things like cryptococcal meningitis and toxoplasma and tuberculous meningitis, it simply doesn’t happen. What we see, on the other hand, is this unique increased risk for the development of progressive multifocal leukoencephalopathy. This was an entirely, in my mind, unpredictable event. I suspect that this is true of many of the other drugs that are now coming to market; that our experience with them is limited, they have what we think is a well-defined effect on the immune system – they’re not broadly immunosuppressant – yet our knowledge of the immune system is such that we don’t understand fully the downstream effects they have. And it’s only after we’ve used these drugs for a number of years do we have a comfort level with what sort of risks that are engendered by their use.
But it’s not unique to natalizumab; other drugs can induce this whether in neurologic conditions or even rheumatologic conditions. Is that right?
Yes. In talking about PML, that is true that there are other drugs that carry black box warnings for the development of PML; however, there’s something unique about natalizumab and another drug that is somewhat related to it and now off the market called efalizumab, which was a drug which went by the name of Raptiva and was used for the treatment of psoriasis. So there are drugs that uniquely increase the risk of PML and there are those that marginally increase the risk of PML, and one shouldn’t conflate them. And though a drug carries a black box warning for PML, it doesn’t necessarily mean that the risk is the same as it is with another drug that may also carry such a warning.
And let me explain this a little further. If you look at natalizumab and you look at efalizumab, those are drugs that have been used for conditions that had never previously been associated with progressive multifocal leukoencephalopathy. So despite the fact… And natalizumab, as you know, is used in the treatment of MS and used in the treatment of inflammatory bowel disorders, in particular Crohn’s disease, efalizumab used in the treatment of psoriasis; these are autoimmune diseases. And prior to the availability of these compounds, we did some aggressive immunosuppressive therapies in the treatment of these diseases. We would treat them with drugs like Cytoxan and azathioprine and high-dose steroids; a wide variety of things were employed. Yet until the PML experience with natalizumab and efalizumab, we had never seen PML in the setting of multiple sclerosis, in the setting of inflammatory bowel disease, or in the setting of psoriasis. So that tells you that there’s something unique about the drugs that we’re using and that it’s not necessarily the underlying condition that is responsible.
The second is when you start the drug, you do not see PML develop immediately; it takes some time. So the experience with efalizumab was three or more years, the experience with natalizumab is typically 12 months; actually the vast majority of cases – over 80% - have been on natalizumab for 24 months, so they’re on the drug for a long time. The shortest latency from initiation to the development of PML has been a single case in which it developed within eight months; everything else is 12 or more months. So what is that telling you? That tells you that the drug is doing something fundamentally to overcome the barriers to the development of this disease and that it’s not simply opening up a gate and letting the horses out; it’s doing something to the pathobiology of the disease.
And then lastly, the incidence with which we see PML with natalizumab – and presumably with efalizumab, although the numbers were much smaller – is extraordinarily high in the appropriate context. So for natalizumab, the risk of developing PML, provided you’re on the drug for two years, you’ve seen prior immunosuppressive therapy, and you’re JC virus antibody-positive so that you have been exposed to the virus that causes this disease, if you have all three of those, your risk is on the order of 1 in 90 or thereabouts. That is a risk that is commensurate with what we see with HIV-associated PML, so it’s very, very high.
However, if one looks at these other drugs which I have called Class 2 agents in several papers now; drugs like rituximab, also a monoclonal antibody though directed against CD20, drugs like brentuximab vedotin or mycophenolate mofetil – which is CellCept – those drugs, too, carry black box warnings for the development of PML; however, the setting in which PML occurs with their use is almost always with a condition that already predisposes you to the development of PML. So with rituximab, for instance, it’s seen with lymphoproliferative disorders, or with transplantation, or with autoimmune diseases in which PML had already been described long before the use of rituximab for the condition. And the same is true with these other drugs.
The second is there’s no latency to the development of the disease, so this is strictly a stochastic event; you may start rituximab today and in two weeks’ time develop PML. There’s no way that somebody’s developed PML in two weeks’ time. What that indicates is that individual was predisposed to developing PML, that virus was already in their brain, it was percolating there, your immune system was suppressing it adequately so it wasn’t expressing itself. And now you’ve done something, you’ve tweaked it a bit and the PML is now expressing itself because you’ve introduced the drug, but the drug fundamentally is not changing the pathobiology of the disease.
Lastly, although we do not have good figures on this, but the best data that I have is that we’re talking about orders of magnitude lower risk with these other drugs. So rituximab, for instance, the risk is probably on the order of 1 in 30,000, or something to that effect. That compared to 1 in 90 when you have all the risk factors that I described with natalizumab. So we’re talking orders of magnitude difference. So I’d suggest that we avoid conflating these drugs when talking about PML risks, and I think that this is something that is generalizable for other risks; I mean, PML is just one risk, but we see other infections and other things that have occurred with other drugs that we’ve employed tweaking the immune system, and I don’t think that one should necessarily put all these drugs that cause these things in the same boat.
These drugs that are used in other conditions that do in themselves predispose to PML, when they’re used in MS which as a disease does not, on its own, predispose to PML, these same drugs – azathioprine, cyclophosphamide, mycophenolate – add to the risk when you give natalizumab?
That’s what it looks like. So when Biogen looked at the data that they had available from the initial cases of natalizumab-associated PML, one of the risks that they identified was the increased risk of the development of PML in those individuals that had previously received immunosuppressive therapy. And it really didn’t seem to matter which immunosuppressive therapy it was, it was any immunosuppressive therapy. However, it may be different with the different immunotherapies, it’s simply that the numbers weren’t large enough for one to say that this was particularly associated with azathioprine. People in Europe like to use azathioprine often very early in the course of the disease, so they were seeing a bit more PML than we had seen in the United States at least initially. And it wouldn’t surprise me if there aren’t certain immunosuppressive agents that increase the risk significantly compared to others, but we simply don’t have that data. And what is known is that it really doesn’t matter, any of them can do that.
Without really having a firm understanding of the pathogenesis of PML – you know the risks but maybe not exactly why it’s occurring – how do you come up with a framework for mitigating risk; is it purely empiric?
That’s an excellent question. So it turns out that this was a back-of-the-textbook disease; this was the disease that occurred very, very rarely. Between 1958 and 1984 in a review published by Ben Brooks and Deward Walker, there were only 230 cases that they were able to come up with; 1958, of course, is when the disease was first described. So this was a very, very rare disease until the AIDS pandemic where people developed some interest in it, and then it really became interesting when we saw it with natalizumab. And there’s been more resources put into the study of this disease since then, so that we do have a better understanding of the pathogenesis. But in identifying these risks, we’ve worked backwards; you know, we say, alright, what does natalizumab do? So why is it that we see this increased risk?
So in getting back to your question, we know that immunosurveilling the brain is important; so if you have a drug that prevents appropriate immunosurveillance of the central nervosus system, it should not surprise you that the risk of PML is increased. And we do think that the alpha-4 beta-1 integrin inhibition that occurs preventing the entry of JC virus-specific cytotoxic T lymphocytes into the brain is in a large measure – but not completely – contributing to the development of PML. We also know other things. For instance, the virus that we are likely infected with – and the infection occurs very early in our lives; seroepidemiologic studies indicate that most individuals that are infected are infected before the age of 20 – that that virus is a virus that is ubiquitous but incapable of growing effectively in glial tissues; it is a virus found in urine and found in the urinary tract, it’s found in kidney and renal pelvis and bladder, and it’s found in high concentrations in some people’s urine. That virus will not grow in the brain. So something has to happen to the virus.
Does natalizumab change that in some way? Well, we think it does. We think it does that by causing the release of immature B cells; these immature B cells can harbor JC virus, and that virus as these B cells mature there is an upregulation of transcriptional factors that transactivate the virus, it is occurring in a milieu that may uniquely predispose to a genetic rearrangement of the virus enabling it to become a form of the virus referred to as a prototype strain or a neurotropic strain that can grow in the brain. So we think that there are at least two very large barriers that prevent PML that are affected by natalizumab; there may be others. And as we investigate this disease further, our understanding may improve and these explanations I’m telling you will likely be expanded. And, in fact, it may be that some of the thoughts that we have are wrong, because the story with the B cells is actually somewhat hypothetical; there’s some preliminary evidence supportive of it, and that’s why I tell people that I’m fond of Ralph Waldo Emerson’s comment, that consistency is the hobgoblin of small minds. If I stand before you a year from today and tell you something different, it’s only because we’ve learned more about it.
So how do you mitigate risk and how do you get the point across to patients to let them make informed decisions with you?
So this is difficult, but we lay the facts out to them. And the facts are that there is this risk of PML, the risk in individuals that are JC virus antibody-negative are very small, the product label puts the risk at less than 1 in a thousand; the belief is that it’s significantly less than that. The fact is that we do see individuals who are JC virus antibody-negative months before the development of PML – it’s rare but it occurs – and there have been studies, including my own, that have indicated that the antibody study doesn’t necessarily mean you’ve never been infected by the virus. So one shouldn’t conflate JC virus antibody negativity with never having been infected, but it is a very good marker for the development of PML, and it is one that needs to be monitored carefully at six month intervals.
So we lay out to the patients that if you’re antibody-negative your risk is infinitesimal and an acceptable risk, and that we monitor you carefully. The second is this is a disease that can be life-threatening, and if not life-threatening certainly severely debilitating – multiple sclerosis I’m talking about – and there are people that have very aggressive disease and will accept the risk of developing PML, knowing that the risk may be 1 in 100, but their risk of developing something that was going to leave them wheelchair-bound and totally disabled is very, very high, and they say I’d rather take the risk of the development of PML.
So we do know what the numbers are. There’s a table that’s been published and gets revised periodically that puts into it JC virus antibody positivity, duration of therapy broken up into 24-month epochs, and prior immunosuppressive use. So we know what the risks are; the highest are in individuals that are treated more than 24 months with the drug, are JC virus antibody-positive, and had previously received immunosuppressive therapies.
What about monitoring for signs and symptoms of PML if someone does choose to go on some of these drugs?
Yes, so that’s very, very important. And we do have patients that even in the face of JC virus antibody positivity, we and the patient elect to continue them on the drug. And patients on natalizumab need to be monitored very carefully for the development of PML, and that is a combination of both clinical screening – there is a TOUCH program that queries for the development of symptoms that may be concordant with PML – unfortunately, you also see symptoms of that nature develop in MS patients, as well, so it’s sometimes difficult to tell whether it’s MS or PML – and at that point in time you definitely want to survey them with an MRI. And, in fact, many of us do MRIs at regular intervals in patients on Tysabri attempting to identify the disease – PML – when it is asymptomatic. And those people seem to do best; the prognosis both in terms of disability and in terms of survival is better when you pick the disease up while they’re still asymptomatic, and they have what one would refer to as radiographically isolated PML.
So it’s a combination of vigilance, asking the right questions, performing your physical examination, and obtaining period MRIs, and in those that are JC virus antibody-negative – or even the antibody-positive – repeating that study periodically. And the reason I say that repeating it periodically even in the positive patients is because what’s been demonstrated is that the higher your antibody titer, the greater the risk of developing PML. So there’s a threshold now that’s been identified and individuals above that range have a significantly higher risk of developing PML than individuals who are seropositive but below that level.
And just to clarify; the TOUCH program is the Tysabri Outreach Unified Commitment to Health? This is what you’re referring to in terms of monitoring for signs and symptoms of PML?
That is correct. That is the risk mitigation strategy that is FDA-approved and that Biogen has implemented in an effort to decrease the likelihood of PML developing.
Very good, thank you.
Thank you for listening to Episode Thirty-Eight of Multiple Sclerosis Discovery. This podcast was produced by the MS Discovery Forum, MSDF, the premier source of independent news and information on MS research. MSDF’s executive editor is Robert Finn. Msdiscovery.org is part of the non-profit Accelerated Cure Project for Multiple Sclerosis. Robert McBurney is our President and CEO, and Hollie Schmidt is vice president of scientific operations.
Msdiscovery.org aims to focus attention on what is known and not yet known about the causes of MS and related conditions, their pathological mechanisms, and potential ways to intervene. By communicating this information in a way that builds bridges among different disciplines, we hope to open new routes toward significant clinical advances.
We’re interested in your opinions. Please join the discussion on one of our online forums or send comments, criticisms, and suggestions to firstname.lastname@example.org.