Epilepsy Benchmark IIIB

2005 Report submitted by Benchmark Steward(s):
Tracy Glauser, M.D. (Cincinnati Children’s Hospital)

Background of the benchmark goal: 
Response to anticonvulsant therapy demonstrates marked inter-individual variation both in drug efficacy and adverse effects.  A significant portion of this inter-individual variation is under genetic control.  Recent advances in genetic technology and analysis have permitted investigations into the relationship between the response to anticonvulsant therapy and polymorphisms in drug metabolizing enzymes, drug transporters, and drug receptors.

Current status of the field:
Significant fundamental research is still needed to elucidate the clinically significant associations that will form the basis for developing a genetic fingerprint diagnostic test for anticonvulsant efficacy, treatment resistance or adverse events. Currently, the two main methodologies used in this research are 1) association studies between response to therapy (efficacy, adverse effects) and single nucleotide polymorphisms (SNPs) of drug metabolizing enzymes, drug transporters, and drug receptors and 2) gene expression studies using microarrays. Previously, association studies focused on known common SNPs but since SNP databases are being found to be often incomplete, SNP discovery is becoming more popular. In general, SNPs in drug metabolizing enzymes alter anticonvulsant pharmacokinetics potentially leading to unexpectedly high levels with subsequent adverse effects.  Similarly, SNPs in genes coding for drug efflux transporters alter the anticonvulsant drug’s ability to remain inside target cells resulting in reduced efficacy and subsequent treatment resistance.  Lastly, SNPs in genes coding for the receptor drug binding site are potential causes for reduced receptor binding and subsequent decreased drug action.  A newer approach has been to examine the relationship between response to therapy and changes in gene expression following exposure to medication.               

Activities update: 
Published reports: During the past year, there have been four human pharmacogenetic association studies and one human gene expression study. Two studies (both published in Neurology September 2004) examined the proposed relationship between a polymorphism in the drug efflux transporter gene ABCB1 and treatment resistant epilepsy. The studies reached contradictory conclusions about the potential association. In an editorial in the same issue, these conflicting studies were used to illustrate some of the methodological problems evident in current epilepsy pharmacogenetic research. Another pharmacogenetic epilepsy study utilized population pharmacokinetic modeling approaches to provide phenytoin dosing recommendation based on the patient’s CYP2C9/CYP2C19 genotype. In a separate study, a Japanese team did not find a relationship between gingival overgrowth in phenytoin patients and CYP2C polymorphisms. Lastly, a pilot trial reported distinct blood expression profiles for patients on carbamazepine monotherapy compared to patients on valproic acid monotherapy. This study also described a characteristic blood expression profile for seizure free patients taking valproic acid compared to patients with uncontrolled seizures.

Ongoing studies: NINDS has funded a 20 center 436 patient clinical trial that is focused, in part, on elucidating the genetic contribution to the interindividual variation in anticonvulsant efficacy and adverse events in children with childhood absence epilepsy. For example the relationship between anticonvulsant efficacy and polymorphisms in three different T-type calcium channel genes is being investigated. The Epilepsy Foundation is funding a study examining the relationship between polymorphisms in carbamazepine and valproic acid drug metabolizing enzymes and altered drug pharmacokinetics and adverse events (as judged by a validated side effects questionnaire).

A NINDS funded project is utilizing microarray technology to investigate different patterns of gene expression preliminarily associated with valproic acid and carbamazepine efficacy in children with epilepsy.  No final results are yet available from this study but the pilot gene expression publication described above resulted from this NINDS funded research.   

During the past year, there have been no workshops or symposiums focused solely on the pharmacogenetics of anticonvulsant efficacy or toxicity.            

Top priorities for next 5-10 years:

• Develop a consortium of experienced epileptologists with an interest in pharmacogenetics to conduct coordinated studies combining extensive drug response phenotyping and with analysis of known SNP, SNP discovery studies, and microarray gene expression studies.
• Through these large scale coordinated studies, identify which polymorphisms (or haplotypes) have clinical relevance for anticonvulsant efficacy or adverse events, and develop methodologies (e.g. specialized gene chips) for the rapid (< 4-5 hour) outpatient assessment of these polymorphisms or haplotypes in patients with epilepsy.
• Develop physician and nurse education programs (for all levels of experience) focused on incorporating pharmacogenetic testing into the routine clinical care of patients with epilepsy.

Roadblocks to progress:

• Lack of a coordinated national effort to prioritize specific clinical questions to be answered by the proposed consortium. Efficiency is critical due to the time consuming nature and expense for each association or gene expression study.
• Developing the bioinformatics computational approach and software to easily examine more complex relationships than 1 or 2 SNPs at a time, for example identification of complex haplotypes.
• Reluctance by patient and families to enroll in genetic based research because of concern over future insurability depending on the test results.