The little girl’s epilepsy was so debilitating that she was virtually nonresponsive. Traditional antiseizure medicines could not reduce the five to 20 seizures she experienced daily when she first came to The Children’s Hospital of Philadelphia.
Trying a new approach, her neurologist, David Bearden, MD, prescribed a drug that targeted a molecular pathway involved in her seizures, and within a month, she was seizure-free for the first time since a few days after her birth.
This success excited hope among epilepsy researchers worldwide that other such successful strategies could soon follow. The case exemplifies the popular concept of precision medicine, which is barreling ahead in cancer but not yet common practice in neurologic disorders such as epilepsy.
“Most drugs for epilepsy work like treating pneumonia with a cough suppressant: It may stop the symptom but doesn’t treat the underlying problem,” said Ethan Goldberg, MD, PhD, a CHOP neurologist and neuroscientist who was senior author of a case report about the little girl’s treatment (of which Dr. Bearden was first author) in Annals of Neurology in 2014. Her treatment, while not yet analogous to an antibiotic, was more precisely targeted to the underlying mechanism of her seizures than most treatments.
The future need for precision medicine is one that epilepsy researchers are approaching with conscious attention to the field’s strengths and unmet challenges. Dr. Goldberg was a presenter at a precision medicine scientific symposium during the American Epilepsy Society annual meeting in December. His colleague, Dennis Dlugos, MD, MSCE, a CHOP pediatric neurologist, was among the major contributors to an international consortium of researchers who authored a roadmap for precision medicine in epilepsy published in The Lancet Neurology this fall.
Finding Genetic Causes of Epilepsy
Several research groups had only recently published findings linking the little girl’s particular genetic form of epilepsy to overactivity of an ion channel, KCNT1 at the time she arrived at CHOP. Recognizing that potential drug therapies blocking KCNT1 might be a useful approach in this form of the disease, Dr. Goldberg searched for existing drugs that could have this effect. Soon, Dr. Bearden developed a treatment plan and prescribed the one drug that fit the bill and already had an established safety profile for use in children, a cardiac drug called quinidine. Quinidine led to her successful, dramatic reduction in seizures, and Dr. Bearden and others have since prescribed it to a number of other children with genetic mutations affecting the KCNT1 channel.
The handful of known KCNT1 mutations are just the tip of the iceberg among epilepsy-related genes — an area where discovery is accelerating progress on the road to precision medicine. Approximately 100 known single-gene causes for epilepsy are known, comprising 10 to 15 percent of epilepsy cases. Half of epilepsy cases have unknown causes that are also presumed to be genetic, with the remaining proportion of cases attributed to known non-genetic causes such as brain injury or infection. Dr. Dlugos and CHOP neurologist Ingo Helbig, MD, were among the leaders of a major international study pinpointing epilepsy mutations in 2014. These findings are all potential targets for precision-medicine therapies — but they are only a first step.
Modeling Mechanisms of Healthy and Epileptic Brains
After identifying epilepsy-causing genes, researchers still need to learn what these genes do, including both their role in the mechanism of disease and in healthy brain functions.
Until researchers better understand those mechanisms, there is a risk that precision treatments, like current antiseizure medicines, could still target the wrong outcome.
“If your output measure is cough, we may be developing better and better treatments for cough, not pneumonia,” Dr. Goldberg said.
Even quinidine, the source of so much hope for precision medicine, could turn out to be little better than an improved seizure suppressant. After treating several more patients with KCNT1 mutations using quinidine, Dr. Bearden noted that most showed improvements in seizures and developmental outcomes, but all remained significantly developmentally delayed. Researchers do not yet know what other mechanisms contribute to the developmental problems in KCNT1-related epilepsy.
A wide array of cell, network, and animal models of epilepsy is key to understanding the mechanisms of the disease and its existing investigational treatments, and to screening for and testing new treatments. Dr. Goldberg and Eric Marsh, MD, PhD, at CHOP, work with epilepsy models including stem cells, fish, mice, and many more. This laboratory study of epilepsy requires a wider range of models than many types of cancer, because neuroscientists find living model systems essential to studying epilepsy as a circuit-based disorder. In contrast, cancer researchers commonly have tumor tissues to directly test treatments on the disease cells, which are more useful as models of that disease’s cell-growth dysregulation.
“All these models together can help us target some of the disease mechanisms, whether testing off-the-shelf medications or going through the process of new drug design,” Dr. Marsh said.
Drug Development and Clinical Testing
Developing new drugs is a challenge because major pharmaceutical companies have not made large investments in developing new epilepsy treatments in recent years, although a few smaller companies are exploring the potential for precision epilepsy treatments. An added challenge with precision treatments is that, by design, this approach targets smaller subgroups of patients, or a smaller commercial market, than do broad-acting drugs.
Preclinical modeling that helps to group different genetic epilepsies into related pathways may help meet that challenge, Dr. Marsh noted. In his modeling, he seeks out signs that many different genetic causes of epilepsy could lead to the same neuronal mechanism of disease, to see if a targeted drug could act on that mechanism — essentially grouping different genetic epilepsies together by function.
Designing clinical trials once new potential precision drugs are identified is difficult for the same reason — dividing a rare disease into even rarer subtypes with tiny patient populations. Making trials even more challenging, the outcome measure in epilepsy studies is patient- or parent-reported incidence of seizures, not a direct biological marker of disease.
“How do we design trials with clinical seizure reporting as the primary outcome measure when you have very few patients and not a lot of options for a big trial?” Dr. Dlugos said. “That’s an open topic for discussion right now between us, the NIH, and the FDA about how that is going to happen.”
Conducting trials for off-the-shelf medicines such as quinidine poses its own set of challenges: Doctors can freely prescribe them off-label before rigorous evidence demonstrates their effectiveness. For that reason, Dr. Bearden has begun a clinical registry for children with KCNT1 mutations to track their physicians’ treatment plans and outcomes, in hopes of gathering useful effectiveness data about various approaches.
The early success treating one little girl with quinidine, while promising, only marked the start of a long road ahead for developing and validating precision treatments for her disease.