Fall colors. Just a brief summary of how this post originated. Eckernförde is a small city north of Kiel and the weekly Sunday destination of my daughter and me because of the wave pool. This past Sunday, daylight saving and the fact that she didn’t like her dinner had confused the little girl, and we had been awake since 4AM. As a consequence, she fell asleep on the way, and I kept driving to let her sleep. We made it as far as Haddeby, and I used this time to mentally put a post together that I had been planning for some time. These are the three things that are often misunderstood with regards to epilepsy and genes.
Misunderstanding 1: There are genetic and non-genetic epilepsies. At first glance, it looks easy and convincing. A seizure disorder is either genetic with a single causative gene or it is non-genetic due to an external trigger. The truth is that this is probably true only for a small minority of epilepsies. For most epilepsies, there is an interaction of innate and acquired factors even though the relative contribution may vary. In addition, the particular contribution can often only be understood on a group level, i.e. the statement that the epilepsy in an individual patient is 70% genetic and 30% non-genetic is meaningless. These thoughts may seem theoretical at first glance, but there is a clear practical consequence. Patients with “genetic epilepsies” are usually not considered for epilepsy surgery as the mutation is thought to be “everywhere in the brain”. The surgical option is usually only considered for epilepsies with a clear focus, a concept that is hard to reconcile with a generalized disease due to an abundant mutation. However, this dichotomy might not necessarily be true. Some patients with Temporal Lobe Epilepsy and mutations in SCN1B became seizure-free with epilepsy surgery, and at least one patient with epileptic encephalopathy and a mutation in STXBP1 had significant improvement after epilepsy surgery. Likewise, a study on microdeletions suggested that the outcome of epilepsy surgery is independent of whether the patient carried a microdeletion or not. Also, epilepsy patients with a clear non-genetic lesion such as cysts, tumors, or unilateral malformations may have additional genetic risk factors. In summary, it might make sense to ask the question whether an identified lesion sufficiently explains the entire phenotype and whether identifiable genetic factors might be at play as well.
Misunderstanding 2: Why look for genes? It’s not going to change treatment anyway. We are not yet at the stage where identification of an underlying, possibly causative gene does automatically impact on patient treatment, and it is not yet clear whether we will ever get to this stage. This lack of direct therapeutic consequences is sometimes used against putting efforts into searching for genetic factors that may play a significant role. There are two important counterarguments to this misunderstanding. First, the impact on treatment is already tangible and is likely to increase in the future. From withholding lamotrigine in patients with Dravet Syndrome to treating previously unclassified epilepsies with the ketogenic diet after discovery of a GLUT1 deficiency, genetic results help align treatment strategies with empirical observations in these syndromes. In addition, genetic epilepsies might benefit from novel technologies, which investigate genetic epilepsies in animals model and use these model systems for drug screening. Secondly, the underlying tone of this misunderstanding implies that genetic testing is only valuable when resulting in treatment. It is often ignored that finding the cause for a person’s epilepsy takes a huge burden off the caretakers’ shoulder. It may help diffuse issues of guilt for the disease and assist in accepting the disease. Also, gene identification gives parents the power to share their experience with others. In that sense, a causative gene may actually help create identify for a disease.
Misunderstanding 3: If I only knew the entire genome, I would know the cause. Also this might sound enticing at first glance. Use the modern technologies to decipher the entire genome, and you will eventually stumble upon the culprit gene. Nothing could be further from the truth. With the large number of rare and unique variants, we will soon get lost and swept away in the flood of rare variants. Even though some studies report on a diagnostic rate of 30% and higher for exome or genome sequencing, the yield is probably much lower in the epilepsies. Why is this? Genetic information needs context. This context might either be phenotypic or genetic. For a patient with a Dravet-like phenotype, I would focus on the described candidate genes in the genome data. With regards to genetic findings, parental genotype will help reducing the genomic noise. A single genome/exome is filled with rare variants and is so rich with information that we need additional context to make sense of this. Also, the genetic information derived from exome or genome sequencing is only of limited value for predicting disease. While there are a few genetic alterations that invariably lead to disease as in the case of Huntingtin alterations in Huntington’s disease, the predictive power of genetic findings for epilepsies is not clear. The phenotypic range may be very high and the frequency in the population is not known in most cases, even for more common causative genes such as SCN1A. Therefore prediction is difficult and even dangerous given the genetic and phenotypic heterogeneity of most epilepsies.
Summary. These three misunderstandings may cloud our understanding of genetics in epilepsies and create unrealistic expectations or fears. I hope that I could address some of these concerns in this post. For me, the purpose of genetics is very humble – I think that we can be happy if we can explain anything at all given all the pitfalls and ambiguities that the human genome has to offer. Genetics never claims to represent the entire picture, but only offers to help find some of the answers that we are looking for.