Share this article on:

Autoimmune Epilepsy: A New Avenue for Treating Medically Intractable Cases

Rukovets, Olga

doi: 10.1097/01.NT.0000449801.86458.a7
Back to Top | Article Outline




Preliminary results from a small, single-center, retrospective study show significant improvement in 62 percent of medically intractable patients with suspected autoimmune epilepsy who initiated a six- to twelve-week course of intravenous methylprednisolone, IVIg, or both.

Within the last year, a patient arrived at the Mayo Clinic epilepsy department with very frequent and uncontrolled seizures, voltage-gated potassium channel complex antibodies, and signal abnormality in the temporal lobe, recalled Sean J. Pittock, MD, professor of neurology and co-director of the Neuroimmunology Laboratory at Mayo Clinic in Rochester, MN. “We knew beyond a reasonable doubt that this patient had autoimmune epilepsy. He continued to have seizures despite being on antiepileptic drugs [AEDs].”

Dr. Pittock recommended that the patient start on intravenous immune globulin (IVIg) because methyprednisolone was contraindicated. The insurance company, however, did not agree.

Shortly after this denial, the patient experienced a seizure that landed him in the intensive care unit with significant head injury. Although he made a substantial recovery, the patient had post-traumatic brain injury syndrome and his uncontrolled seizures continued.

At that point, the insurance company finally agreed to cover IVIg, said Dr. Pittock, and then the patient's seizures stopped completely.

This patient — and others like him — is one of the driving forces behind the research of Dr. Pittock, senior author of a new Neurology study on autoimmune epilepsy. In the April 4 online-first report, Dr. Pittock and colleagues found encouraging, but preliminary results for the use of immunotherapy in medically intractable patients with suspected autoimmune epilepsy. In the relatively small, single-center, retrospective study, the investigators reported significant improvement in 62 percent of patients who initiated a six- to twelve-week course of intravenous methylprednisolone, IVIg, or both.

“About a third of all epilepsy patients have medically intractable epilepsy. If they don't respond to the first or second drug, the likelihood of responding to the third, fourth, or fifth drug becomes less than five percent. So, the fact that in this study we had 34 percent of patients becoming seizure free, and 62 percent having significant improvement, is dramatic relative to antiepileptic medications,” said Dr. Pittock

“These findings are just the start, but they're the start of something very exciting,” he added. “Now, we need a larger, randomized trial in this sub-population of patients.”

Back to Top | Article Outline


This study reports the experience of the Mayo Clinic's auoimmune epilepsy study group consisting of physicians from divisions of epilepsy (led by Jeff Britton, MD) and multiple sclerosis and autoimmune neurology (led by Dr. Pittock). The first author, Michael Toledano, MD, retrospectively reviewed the charts of 110 patients with seizures as the primary complaint seen at the autoimmune neurology clinic by Mayo epileptologist Jeffrey W. Britton, MD, and colleagues. These patients were initially identified using a search in the Mayo Clinic Records Linkage system.

Patients were included if they had: a) autoimmune epilepsy suspected based on the presence of ≥1 neural Ab; personal or family history, or physical stigmata of autoimmunity; and frequent or medically intractable seizures, and b) initiated a six- to 12-week trial of intravenous methylprednisolone (IVMP), IVIg, or both. Based on these criteria, 29 patients were included in the subsequent analysis.

The researchers defined “responders” as those patients who showed at least a 50-percent reduction in seizure frequency after treatment. The choice of treatment was determined by clinician preference.

Overall, 18 patients (62 percent) were identified as responders; 10 (34 percent) of these patients became seizure-free with treatment. Just over half of the patients (52 percent) responded with the first agent, and 43 percent of those who received a second agent after not responding to the first reported improvement.

Among responders, those patients who initiated treatment earlier after symptom onset reported a significantly better response rate (median 9.5 vs 22 months; p=0.048). Adverse events were reported by six people (21 percent), and led to cessation in two of these cases (7 percent): one because of steroid-induced psychosis and another from IVIg-associated aseptic meningitis.

Responders included 14/15 (93 percent) patients with antibodies to plasma membrane antigens, 2/6 (33 percent) of patients seropositive for GAD 65 Abs, and 2/6 (33 percent) of patients without detectable antibodies. Of the responders who were followed for more than 6 months after initiating long-term oral immunosuppression, response was sustained in 85 percent (11/13).

What's novel about the study, said Dr. Pittock, is the use of immunotherapy as a diagnostic test. “Even if you have a patient whose epilepsy you suspect is immune-mediated, at the end of the day, the most important test is: does the person actually get better with immunotherapy?”

Dr. Pittock added “our standard practice is currently that patients with suspected autoimmune epilepsy undergo a treatment trial with either intravenous methylprednisolone or IVIg. Unfortunately, over the past few years, there has a significant increase in insurance denials for use of IVIg for suspected autoimmune epilepsy. It is of paramount importance to our patients that we show in a randomized, controlled trial that these immunotherapies do indeed work in the diagnostic and therapeutic management of these patients.”

Moving forward, Dr. Pittock said, there needs to be prospective trials that use specific diagnostic inclusion criteria. “Patients should be randomized to a trial of treatment, or very short duration of placebo, and then crossed over if they are not responding to the treatment arm. They should be followed for a prolonged period of time to investigate whether maintenance immunosuppression is required to prevent relapse and, if so, at what dose and duration. These studies will require support from industry as immunotherapies such as IVIg are very expensive.

One substantial limitation, given the retrospective design of the study, was the way that seizure rates were quantified. “We didn't prospectively count the seizures before we started treatment. We had to look at the chart to find out how many seizures the patient had or how many seizures were documented in the chart,” and the types of measures (daily, monthly) were not always consistent or specified, noted Dr. Pittock.

The study may have been subject to referral bias because these subjects were seen in a specialized autoimmune clinic, the authors wrote. “Also, likely confounders such as AED medication changes during the trial could not be controlled for and non-responders may have been lost to follow-up.”

“If autoimmune epilepsy is proven in these medically intractable patients, then you have a treatable condition, but you are treating the underlying etiology rather than the end result. Therefore, it's important to consider it early because the sooner you get in and modify or reduce the level of attack by the immune system, the more brain you're probably going to protect against scarring or neuronal injury,” said Dr. Pittock.

[Dr. Pittock and colleagues at the neuroimmunology lab at Mayo Clinic are developing a comprehensive autoimmune epilepsy evaluation that includes more than 15 “epilepsy pertinent” neural antibodies which will be offered Mayo Medical Laboratories this summer.]

Back to Top | Article Outline


Epileptologists agreed that this paper presents early, but encouraging data that may improve treatment for the many patients with currently intractable epilepsy.

Orrin Devinsky, MD, FAAN, director of the New York University (NYU) and Saint Barnabas Epilepsy Centers, and a professor of neurology, neurosurgery & psychiatry at the NYU School of Medicine, said that this study builds on existing literature, which suggests the use of immunotherapy for many epilepsies that have a suspected autoimmune etiology. “It provides additional support that these therapies can produce substantial benefits in many patients,” he said.

The main strengths of the paper are the large number of patients seen and the systematic treatment approach and follow-up, he added. However, the lack of randomization, lack of placebo-control, inclusion of multiple patient groups, and use of several forms of therapy (steroids and IVIg) limited the power of the findings, Dr. Devinsky told Neurology Today.

David Spencer, MD, FAAN, professor of neurology at Oregon Health & Science University (OHSU) and director of the OHSU Comprehensive Epilepsy Program, said that the field of autoimmune epilepsy is still in its infancy, and this study provides needed guidance for clinicians on “how to navigate discussions of risk and benefit surrounding immunotherapy in these patients. The findings make a strong argument that the benefits of immunotherapy trials outweigh the risks in many patients.”

The main limitation, he agreed, was the retrospective and uncontrolled design of the study. The generalizability of the findings to the entire population of patients with autoimmune epilepsy, he added, may be limited because “a large proportion of the patients studied had antibodies directed against voltage-gated potassium channel antigens.”

Dr. Spencer noted that he was also “somewhat surprised” that nearly half of patients who did not respond to an initial trial of immunotherapy responded to a second trial. “This suggests we should not give up if there is not a clear response to the first immunotherapy trial in patients in whom there is strong evidence of autoimmune epilepsy.”

Mark Gorman, MD, a neurologist and director of pediatric multiple sclerosis and related disorders at Boston Children's Hospital and Harvard Medical School, who was one of the authors of a population-based study on autoimmune epilepsy in the March 31 online edition of JAMA Neurology, commended Dr. Pittock and colleagues for providing a “practical approach to identifying, treating, and following patients with possible autoimmune epilepsy.”



However, he pointed out that the paper does “not specify how many of the ‘clinical features suggestive of autoimmune epilepsy’ in Figure 1 of the article needed to be present in order for a patient to be included in the study, nor if certain features did or should carry more weight than others.”

Additionally, the way patients were identified for this study from a highly specialized clinic may have only captured “the tip of the iceberg,” he said, in terms of the potential contribution of autoimmunity in patients with epilepsy. “This likelihood is highlighted by the fact that 79 percent of patients had one or more neural autoantibodies.” It is possible, Dr. Gorman continued, that many more patients have autoimmune epilepsy without these antibodies, but the autoimmune cause is not suspected and thus they are not referred to a specialized clinic.

In their JAMA Neurology paper, Dr. Gorman and colleagues took a different approach to examining autoimmune epilepsy by including “patients with epilepsy who were not pre-selected by the treating clinicians as being suspected of having an autoimmune cause.” They found that nearly 18 percent of about 10,000 epilepsy patients had one or more of twelve autoimmune diseases. “While it remains to be seen whether such patients have an autoimmune basis for their epilepsy, our finding suggests that autoimmune causes of epilepsy may be more common than implied by the Mayo case series of 29 patients,” he added.

Based on both studies, Dr. Gorman said, the “key next steps will be to improve the identification of patients with suspected autoimmune epilepsy in general neurology and epilepsy clinics; to identify the underlying mechanisms in patients with autoimmune epilepsy, especially in those who do not harbor a known neural autoantibody; to prospectively assess the efficacy and side effects of the proposed treatment algorithm or a variation of it; to determine the best approach to patients with suspected autoimmune epilepsy who do not respond to ‘first line’ immunotherapies such as methylprednisolone or intravenous immunoglobulin; and to identify the best long-term treatments for patients with autoimmune epilepsy.”

Dr. Spencer added that in “the patients [in the current study] with neural plasma membrane antigen antibodies — most of them directed at components of the voltage-gated potassium channel — the retrospective findings strongly argue that a trial of immunotherapy is warranted and likely highly effective. For patients with antibodies directed at intracellular antigens or ‘antibody-negative’ patients, the message is less clear.”

The biggest continuing challenge for epilepsy care, said Dr. Devinsky, is understanding how far autoimmune epilepsy extends beyond the recognizable cases, such as those presented in the current Neurology paper.

Furthermore, he said, “Would other patients with treatment resistant epilepsy who are not suspected of having an autoimmune epilepsy also benefit from immunomodulatory therapies? I suspect some would, but the challenge is to better understand how the immune system and inflammation contribute to the broader epilepsies and how our therapies could be better tailored in these patients.”



Sean J. Pittock, MD, professor of neurology and co-director of the Neuroimmunology Laboratory at Mayo Clinic, discusses the findings from his recent Neurology study on autoimmune epilepsy and how they may help diagnose and treat patients with intractable epilepsy in the future:

Back to Top | Article Outline


  • Acute to subacute onset (maximal seizure frequency less than or equal to three months)
  • Multiple seizure types of faciobrachial dystonic seizures
  • AED resistance
  • Personal or family history (1st degree relative) of autoimmunity
  • History of recent or past neoplasia
  • Viral prodrome
  • Evidence of CNS inflammation
    • CSF (elevated protein, pleocytosis, oligoclonal bands, +CSF index)
    • MRI (Mesial temporal or parnechymal T2 hyperintensity)
    • Hypermetabolism on functional imaging (PET)
  • Detection of neural autoantibody

Source: Toledano M, et al. Utility of an immunotherapy trial in evaluating patients with presumed autoimmune epilepsy. Neurology 2014. E-pub: April 4.

Back to Top | Article Outline


•. Kotsenas AL, Watsona RE, Pittock SJ, et al. MRI findings in autoimmune voltage-gated potassium channel complex encephalitis with seizures: One potential etiology for mesial temporal sclerosis. Amer J Neuroradiol 2013. E-pub 2013 July 18.
    •. Rüegg S, Panzer JA. Editorial: Immune therapy for pharmacoresistant epilepsy: Ready to go. Neurology 2014. E-pub 2014 April 4.
      •. Toledano M, Britton JW, McKeon A, et al. Utility of an immunotherapy trial in evaluating patients with presumed autoimmune epilepsy. Neurology 2014. E-pub 2014 April 4.
        •. Quek AM, Britton JW, McKeon A, et al. Autoimmune epilepsy: clinical characteristics and response to immunotherapy. Arch Neurol 2012;69(5):582–593.
          © 2014 American Academy of Neurology