Diagnostic Strategies in Autoimmune Neurology

Autoimmune neurology is a rapidly evolving subspecialty that focuses on neurologic disorders with atypical immune responses.

In this episode, Aaron Berkowitz, MD, PhD FAAN, speaks with Sean J. Pittock, MD, an author of the article “Overview and Diagnostic Approach in Autoimmune Neurology,” in the Continuum August 2024 Autoimmune Neurology issue.

Dr. Berkowitz is a Continuum® Audio interviewer and a professor of clinical neurology at the University of California, San Francisco

Dr. Pittock is the director for the Center for Multiple Sclerosis and Autoimmune Neurology at Mayo Clinic in Rochester, Minnesota.

Additional Resources

Read the article: Overview and Diagnostic Approach in Autoimmune Neurology

Subscribe to Continuum: shop.lww.com/Continuum

Earn CME (available only to AAN members): continpub.com/AudioCME

Continuum® Aloud (verbatim audio-book style recordings of articles available only to Continuum® subscribers): continpub.com/Aloud

More about the American Academy of Neurology: aan.com

Social Media

facebook.com/continuumcme

@ContinuumAAN

Host: @AaronLBerkowitz

Transcript

Full transcript available here

Dr Jones: This is Dr Lyell Jones, Editor-in-Chief of Continuum, the premier topic-based neurology clinical review and CME journal from the American Academy of Neurology. Thank you for joining us on Continuum Audio, which features conversations with Continuum’s guest editors and authors who are the leading experts in their fields. Subscribers to the Continuum journal can read the full article or listen to verbatim recordings of the article and have access to exclusive interviews not featured on the podcast. Please visit the link in the episode notes for more information on the article, subscribing to the journal, and how to get CME. 

Dr Berkowitz: This is Dr Aaron Berkowitz, and today, I'm interviewing Dr Sean Pittock about his article, “Introduction to Autoimmune Neurology and Diagnostic Approach”, which he wrote with his colleague, Dr Andrew McKeon. This article is a part of the August 2024 Continuum issue on autoimmune neurology. Welcome to the podcast, Dr Pittock. Could you introduce yourself to our audience? 

Dr Pittock: Well, thank you very much, Dr Berkowitz. So, yeah, I'm a neurologist at the Mayo Clinic. I direct the neuroimmunology laboratory with Dr McKeon and Dr Mills here, and I have also been very much involved in the autoimmune neurology section at the American Academy of Neurology. 

Dr Berkowitz: So, many of you probably know Dr Pittock - or if you don't know, you've certainly diagnosed diseases that he has described and written about, and so it's a real honor to get to talk to you today and pick your brain a little bit about some of these complex diseases. So, autoimmune neurology is certainly one of the most exciting subspecialties of our field. I feel like when I talk to students and they ask me to make a case for why they should consider neurology as a career, I tell them, “Of course, I have many reasons I love neurology”, but one thing I mention is that, although many other fields of medicine may have made incredible advances as far as treatments, I can't think of too many other fields outside neurology where entirely new diseases have been described since I've been in training and come out of training - and many of those have been in your field of autoimmune neurology. I can think of cases where I've heard you or one of your colleagues on a neurology podcast describing a new antibody, new disease, and a few weeks later, we see that disease and give a patient a diagnosis that had been elusive from other physicians and hospitals. It's a very exciting, gratifying area. It's also daunting, like, every time I go to the AAN and hear one of your colleagues, there's a new disease, and we realize, “Oops! Was I missing that?” or, “Am I going to see this?” And so, hoping to pick your brain a bit today about some of the key concepts and how to keep them in mind so our listeners can recognize, diagnose, and treat these conditions, even if they can't remember every single antibody in your article and all the new ones you and your colleagues will probably discover between now and when this, um, podcast is released. So, before we get into some of the important clinical aspects of these conditions, could you just lay out sort of the broad breaststrokes, the lay of the land of cell-mediated versus antibody-mediated paraneoplastic versus nonparaneoplastic cell surface versus intracellular - how can we sort of organize this area in our minds? 

Dr Pittock: Yeah. It's complex, and it's really an evolving story. But the importance, really, from the perspective of the reader and the perspective of the clinician is that we're talking about disorders where we can actually do something - we can actually impact patients.  And we think about the concept of stopping and restoring in neurology now. We're talking about disorders where we have the potential to stop these inflammatory immune-mediated disorders and, potentially, by stopping early, we may be able to restore function - so, a really important new and evolving field in neurology, because you don't want to miss these conditions. Trying to get your head around the complexity of these entities is difficult, but what we've done in this chapter is, really, to try and lay the groundwork for the following chapters, but provide somewhat of a simplistic approach, but a practical approach that really, I think, can help clinicians. So, the way I think of it, a lot of autoimmune neurology really has stemmed from the discovery of antibodies that cause neurological disease, and the examples of those would be going back to myasthenia gravis (with antibodies to the acetylcholine receptor), going back to Lambert-Eaton syndrome. And then, you know, even if you go back to the older traditional paraneoplastic disorders (the Hu, the Ri, the Yo), at the end of the day, you really have two essential entities, if you want to be very simple. The first is disorders that are caused by an antibody, and the second are disorders where the antibodies you detect are not causing the disorder, but they're telling you that there's predominantly a cellular or T-cell mediated attack of the nervous system. And I think thinking about the diseases in those kind of simple terms helps us when we think about what would be the best treatment to use in these types of cases.

Dr Berkowitz: Fantastic. I think that's very helpful. And just to make sure it's clear in the minds of our listeners when we're dividing into these sort of causative antibodies versus antibodies that might be, uh (I don’t know if I'm using the word properly), but, sort of epiphenomena (or they're present, but they're not causative) as you said, can you just give some examples of the ones on either side and how making this distinction helps us in practice? 

Dr Pittock: Yes. So, antibodies that are causative of disease - I think, you know, the one that I've done a lot of work on is in neuromyelitis optica, where you have antibodies that are targeting a water channel that sits on an astrocyte, and so it causes NMOSD, or what we consider an autoimmune astrocytopathy. And we know that when the antibody binds to the target, many things can happen. So, when aquaporin-4 antibodies bind to aquaporin-4, they can do a lot of things. They can cause internalization, they can activate complement that results in the killing of the cell - but there can be other situations. For example, when NMDA-receptor antibodies bind to the NMDA receptor, then a variety of different things can occur different to water channel autoimmunity - where, for example, the receptor (the NMDA receptor) is downregulated off the cell surface, and that results, to some extent, in the neuropsychiatric phenomenon that we see in NMDA-receptor autoimmunity. And, obviously, when you have a situation where the antibodies are causing the disease, removal of those antibodies, or the reduction in the production of those antibodies, is going to help patients. Now, on the other side, we have antibodies that we detect in the blood or in the spinal fluid, and those antibodies are targeting proteins that are inside the cell - so those antibodies we don't consider as being pathogenic. Now, remember, there are sometimes situations where proteins that are inside the cell occasionally can be available for antibodies to bind at certain situations. So, for example, in the synapse, amphiphysin or the septins, may at times become available. And so, sometimes, there are targets or antibodies that are somewhat in between those two simplistic concepts. But when we're talking about antibodies that are targeting proteins on the inside of the cell, remember that antibodies don't just suddenly occur. There's a whole process of presentation of target antigen at the lymph node, and then both a T- and a B-cell response. The B-cell response potentially produces the antibodies but also triggers and stimulates T-cells, and those T-cells then go on to cause the disease. And those T-cells are very problematic, because those classical paraneoplastic and the newer ones we've described (and many have described) - these are associated with quite severe neurological disability, and they're very, very difficult to treat. And if you ask me, “Where is the holy grail of autoimmune neurology therapeutic research?” It's in trying to actually figure out ways of treating the predominantly T-cell mediated paraneoplastic and autoimmune neurological disorders. We're making great headway in terms of the treatments of the antibody-mediated neurological disorders.

Dr Berkowitz: That's a helpful overview. So, sticking with this framework, you mentioned as sort of the “causative antibody” category and the antibodies that are predominantly for intracellular antigens, but not believed to be causative - I want to make sure I'm understanding this correctly and we can convey it to our listeners - I believe you said in your paper, then, that the antibodies that are predominantly causative are more likely to be associated with conditions that are very treatable, as compared to the intracellular antibodies that are not thought to be causative, as you just said the disability can be irrecoverable or very hard to treat. And I believe another theme in your paper that you brought out is the antibodies that tend to be causative tend to be cell surface and tend to be less likely to be associated with underlying cancer (although not a perfect rule), and the intracellular antigens more commonly associated with cancer in those cases to look very hard for a cancer before giving up. Are those themes that I understand them from your paper properly, or anything else to add there? 

Dr Pittock: Yes, I think that that's exactly the message that we were trying to get across, so that's good news that you’ve picked up on the themes. I think, yeah, in simple terms, remember that when a cytotoxic T-cell identifies the peptide that its T-cell receptor will target, the ultimate outcome is poor, all right? T-cells are like the marines - they don't mess around. Once they find their target, they eliminate that target, and so, it's really difficult to treat those types of diseases if you get them late. And most patients with cytotoxic T-cell mediated paraneoplastic neurological disorders, oftentimes, by the time they get to a center of excellence, the boat has left the dock in many respects - in other words, it's too late. So, you know, I will often see patients, for example, with progressive cerebellar degeneration (say, in the context of Purkinje cell autoantibody type 1 antibodies and a breast cancer), and if those patients are in a wheelchair at the time that I see them, there's very, very little that we can do. So, you really want to try and get that patient into the office, you know, when they're using a cane (or not), and then, potentially, you have the opportunity - using very aggressive immunosuppressive medications - to make a difference. And that is quite different to other scenarios, where, for example, if you have NMDA-receptor encephalitis - as many of the readers will know, this is a condition that is very treatable, and most patients do very well, because the antibodies, they're disrupting function, but they're not killing the neuron, as we see in those more aggressive, paraneoplastic cytotoxic T-cell mediated diseases.

Dr Berkowitz: Also, in terms of searching for an underlying cancer, another theme in your paper as I understood (but want to make sure I'm understanding and conveying to our listeners and hear your thoughts), that the cell surface and treatable antibody-mediated syndromes, as you mentioned (NMO, NMDA) tend to be less associated with underlying cancers (although can be), whereas the intracellular antigens, um, a much higher percentage of those patients are going to end up having underlying cancers. Is that correct, or any notable exceptions to be aware of in that framework? 

Dr Pittock: Yeah, I think the major exception to the rule for the antibodies that are targeting intracellular antigens is the GAD65 antibody story. We generally don't consider the stiff person syndrome, cerebellar ataxia, or other autoimmune neurological disorders associated with very high levels of GAD65 antibodies - those are generally not paraneoplastic. And then there are always exceptions on both sides. You know, one of the benefits of understanding the implications of certain antibodies is trying to understand, you know, what is the likelihood of identifying a malignancy, which antibodies are high-risk antibodies (in other words, high-risk paraneoplastological disorders), and which are low risk in terms of cancer? And, you know, age and the demographic of the individual is often important, because we know, for example, with NMDA-receptor antibodies, the frequency of ovarian teratoma varies with the age of the patient.

Dr Berkowitz: Fantastic. And we encourage our listeners to read your articles – certainly, some very helpful tables and figures that help to elucidate some of these broad distinctions Dr Pittock is making - but just to summarize for the antibody-related part of autoimmune neurology, we have one category of cell-surface antibodies and another of intracellular antibodies. Both can cause very severe and varied neurologic presentations, but the cell surface tend to be more treatable, less likely to be associated with the underlying cancer, and the intracellular less treatable, more likely to be associated with the underlying cancer - but, as with everything in neurology and medicine, exceptions on both sides. Is that a fair aerial view of some of the details we've discussed so far, Dr Pittock? 

Dr Pittock: Yeah, I think so. I mean, I also think that, you know, not only, at least, for the antibody-mediated disorders (you know, as we discussed) we have drugs that will reduce the production of those antibodies, but we're also learning a lot more about the cytokine and chemokine signatures of these disorders. For example, NMO, water-channel antibody-mediated diseases are associated with elevated levels of IL-6. We know, for example, in LGI1 encephalitis and other encephalitides, that IL-6 also is elevated at the time of that encephalitic process. And so, the potential to target IL-6 with, you know, drugs that inhibit IL-6 and the IL-6 receptor, these potentially have, you know, a role to play in the management of these types of patients - whereas in the T-cell mediated disorders, you know, no advance has been made in the treatment of those conditions, I would say, in over 50 years. So, for example, the standard of treatment is steroids and then drugs that impact the bone marrow, and so we really haven't moved forward in that respect. And that, I think, is an area that really needs drive and enthusiastic out-of-the-box thinking so that we can try to get better treatments for those patients.

Dr Berkowitz: This has been a helpful overview. I look to dive into some of the scenarios that frontline practitioners will be facing thinking about these diseases. An important point you make in your article is that autoimmune and antibody-mediated neurologic syndromes can affect any level of the neuraxis. Even just our discussion so far, you've talked about anti-NMDA receptor encephalitis, you've talked about myasthenia gravis (that's at the neuromuscular junction), you've talked about paraneoplastic cerebellar degeneration - there can be an “itis” of any of our neurologic structures and that “itis” can be antibody-mediated. So, one of the key messages you give us is, one, that these are sort of in the differential diagnosis for any presenting neurologic syndrome, and, two, sort of one of the key features of the history, really, to keep in mind (since we could be anywhere along the neuraxis) is the subacute presentation when this should really sort of be top of mind in our differential diagnosis - so, many of these patients are going to be mystery cases at the outset. And one striking element you bring out in the paper is that, sometimes, the MRI, CSF, electrophysiology studies may be normal or nonspecifically abnormal, and although it's very helpful when we can send these antibody panels out, in some cases, resources are limited or institutions have certain thresholds before you can send these out (because neurologists love to send them in). Sometimes, they are not necessarily appropriate. So, love to hear your thoughts on when we should be sending these panels. What are some clues? Um we have a subacute neurologic presentation at any level of the neuraxis, and when it's not anti-NMDA receptor encephalitis, that is sort of a clear phenotype in many cases. How you would approach a patient, maybe, where the MRI is either normal or borderline abnormal (or people are squinting at the medial temporal lobe and saying, “Maybe they're a little brighter than normal”), CSF is maybe normal or nonspecifically, um, and the protein is a little high, but no cells? What clues do you use to say, you know, “These are the patients where we should be digging deep into antibody panels and making sure these are sent and not miss this diagnosis?” 

Dr Pittock: Well, thank you. That's a good question. So, I think, you know, first of all, these are complex cases. So, the patient is sitting in front of you and you're trying to figure out, first of all, Is this a hardware or a software problem? Are we definitively dealing with an encephalitis or an organic neurological entity that's immune-mediated? And, you know, the way I think of it is, for me, you see a patient, it's a twenty-five-piece jigsaw puzzle and you've got two pieces, and you're trying to say, “Well, if I step back and look at those two pieces, do I have any sense of where we're going with this patient?” So, the first thing you need to do is to collect data, both the clinical story that the patient tells you (and I think you make the good point that that subacute onset is really a big clue), but subacute onset, also fluctuating course, sometimes, can be important. The history of the patient - you know, Is the patient somebody who has a known history of autoimmune disease? Because we know that patients that have thyroid autoimmunity are more likely to have diabetes, they're more likely to have gastrointestinal motility or dysmotility, they're more likely to have a variety of different immune-mediated conditions. So, is there a family history or a personal history of autoimmunity? Is the patient at high risk for malignancy? Are there clues that this potentially could be a tumor-initiated immune process affecting the nervous system? The neurological exam also is extremely important because, again, that helps you, first of all, kind of define and get some objectivity around what you're dealing with. So, does the patient have hyperreflexia? Are there signs that there is neurological involvement? And then, really, what I think we need to do is to try and frame the predominant neurological presentation. So, what is the major issue? Because a lot of these patients will have multiple complaints, multiple symptoms, and it's very important to try and identify the major presentation. And that's important, because the neural autoantibody tests are now presentation-defined - in other words, they're built around the neurological presentation, because the old approach of just doing, apparently, a plastic evaluation is gone, because we've got to a stage where we have now so many neural antibodies, you can't test every single neural antibody. So, if you're suspecting that there may be an autoimmune neurological component, then you really need to think about what would be the most appropriate comprehensive evaluation I need to do for this patient. So, for example, if a patient comes in with a subacute-onset encephalopathy, you're probably going to want the autoimmune encephalitis evaluation, and then you have to pick whether it's going to be serum or spinal fluid - and as we outlined in the paper, there are certain antibodies that are better detected in serum versus spinal fluid. So, for example, in adults over the age of 50, LGI1 is much more accurately detected in serum than spinal fluid, and the absolute opposite is true for NMDA-receptor antibody detection. One of the most important components of the neurological evaluation is the spinal fluid, but actually looking at the white cell count - and in fact, sometimes, it's quite interesting to me that I'll often see patients referred with a diagnosis of encephalitis and autoimmune encephalitis, and yet they haven't had a spinal fluid examination. So, the presence of a white cell count, you know, greater than five is hugely helpful - it's like two pieces of that twenty-five-piece jigsaw, because that really tells you that there is something inflammatory going on. And now, in terms of imaging, you're right - some patients will have normal MRI. And if you really do think that there's evidence of - you know, for example, you do an MRI, but you're getting a good sense that there's a temporal lobe seizure occurring, MRI looks normal, the EEG shows some abnormalities in the mesial temporal area - you know, considering additional imaging modalities (like PET scan of the brain), I think, is reasonable. We know that in NMDA-receptor encephalitis cases, 30% of patients will have normal MRI but they'll often have abnormalities on the PET scans. So, I think, what we do is we try to gather data and gather information that allows us to add in pieces of that jigsaw so that, eventually, after we've done this evaluation, we can see now we have ten pieces. If we step back, we say, “Yes, now we know what this condition is”, and then we essentially plan out the therapeutic approach dependent on what we've found. In terms of identification of underlying malignancy, you know, different people have different approaches. Our approach generally has been to try to get a PET-CT scan of the body as our first go-to test, because, actually, we found that CT chest abdomen and pelvis really actually delivers the same amount of radiation - and from a cost perspective, it's about the same - and we have found that PET-CTs really do provide a higher sensitivity for cancer detection.

Dr Berkowitz: Perfect. A lot of very helpful clinical pearls there. So, in closing, Dr Pittock, I've learned a lot from you today. I'm sure our listeners will as well. What does the future hold in this field? What's coming down the pipeline? What are we going to be learning from you and your colleagues that are going to help us take care of patients with these diseases going forward? 

Dr Pittock: Well, thank you, Dr Berkowitz, for that question. I think the future is very bright and very exciting, and, hopefully, some of the more junior members will be enthused by this Continuum series, and, hopefully, we'll go into this area. So, let's talk about the future. The future, I think, is going to be of great interest. Firstly, there's going to be continued discovery of novel biomarkers, and the reasons for that is because of the technical and technological advances we've seen. So, for example, there have been many, many antibodies discovered by us and others that have been discovered on the basis of, for example, phage technology. In fact, the Kelch 11 biomarker discovery in collaboration with UCSF and our group was done on the basis of Joe DeRisi and Michael Wilson's phage approach. And we're actually using that now at Mayo Clinic, and we've discovered about three or four new antibodies just in the last couple of years using this technology (and that here is led by John Mills and Div Dubey). And then, we're also going to see, I think, the evolution of protoarrays much more in biomarker discovery, so, we'll have more antibodies, and again, I think, generally, those antibodies will fall into the two categories we kind of described - so, you know, in terms of the approach to those conditions, maybe not so much change. I do think, though, that the introduction and the utility of comprehensive cytokine and chemokine analysis in the future will assist us in making diagnoses of seronegative encephalitis, but also potentially will direct therapy. So, for example, cytokine A is elevated - maybe that would be a potential target for therapy that's available for these patients with rare and potentially very disabling disorders. Then, when we look at the cytotoxic T-cell mediated disorders, I think the major areas of advance are going to be in better understanding the immunophenotype of cytotoxic T-cell mediated diseases, and then the potential development of tolerization strategies using the specific targets, those specific epitope targets that are involved in paraneoplastic and nonparaneoplastic diseases, and seeing if we can vaccinate patients, but move that immune response into more of a tolerogenic immune response rather than a cytotoxic killing response. And then I think, lastly, we're going to see a dramatic revolution in CAR-T therapeutic approaches to these types of disorders moving forward - and not just, you know, CAR-T therapies that are targeting, you know, CD19 or CD20, but CAR-Ts that are actually personalized and developed so that they can target the specific B- and T-cells in an individual patient and actually do a very fine removal of that autoimmune pathologic process that I think would have significant benefit for patients not only in stopping progression, but also in significantly reducing the potential of side effects - so, a much more targeted approach. So, that's where I think the next ten years is going to be. I think it's very exciting. It's going to require the collaboration of neurologists with, you know, immunologists, hematologists, you know, across the board. So, a very exciting future, I think, for this field. 

Dr Berkowitz: Exciting, indeed. And we have learned so much from you and your colleagues at the Mayo Clinic about these conditions, and I definitely encourage our listeners to read your article on this phenomenal issue that really gives us a modern, up-to-date overview of this field and what's coming down the pipeline. So, a real honor to get to speak with you, pick your brain about some of the clinical elements, pitfalls and challenges, and also hear about some of the exciting signs. Thank you so much, Dr Pittock, for joining me today on Continuum Audio.

Dr Pittock: Thank you very much. 

Dr Berkowitz: Again, today, I've been interviewing Dr Sean Pittock, whose article with Dr Andrew McKeon on an introduction to autoimmune neurology and diagnostic approach appears in the most recent issue of Continuum on autoimmune neurology. Be sure to check out Continuum Audio episodes from this and other issues. And thank you so much to our listeners for joining us today. 

Dr Monteith: This is Dr Teshamae Monteith, Associate Editor of Continuum Audio. If you've enjoyed this episode, you'll love the journal, which is full of in-depth and clinically relevant information important for neurology practitioners. Use this link in the episode notes to learn more and subscribe. AAN members, you can get CME for listening to this interview by completing the evaluation at Continpub.com/audioCME. Thank you for listening to Continuum Audio.