PI left the institution.

In up to 1 out of 3 patients with epilepsy, seizures continue to occur despite the use of one or more antiepileptic medications. Patients also have significant problems with side-effects of these medications as doses are increased.

Our body naturally generates miniature pumps located on the surfaces of many organs to get rid of toxic substances, and antiepileptic medications can be considered by the cells of the body to be a toxin. Research with epileptic brain regions have shown an increase in the amount of drug pumps, therefore getting rid of antiepileptic drugs. One of these pumps is called p-glycoprotein (P-gp for short). Medications may be unable to penetrate and stay within the parts of the brain that need them them most. This may mean that the amount of drug is actually lower in the parts of the brain that cause seizures, and higher in the rest of the brain, which may be why patients may still feel side-effects when seizures are still occurring.

Research in animals has shown that blocking the P-gp pumps can improve how bad, and how many seizures occur as well as the length of seizures. Blockage of the pumps can be done using a different type of medication. Some medications that are used for common problems have been discovered to also block P-gp pumps. One of these, carvedilol, is used to treat heart failure and high blood pressure. It has been found to be very safe in these patients, and does not have a lot of side-effects. We plan to add this medication in addition to patient's anti-seizure medications to see if it will improve epileptic seizures.

The reason why some patients have high amounts of P-gp pumps and others do not may be related to their genetics. A simple blood test can be used to determine a person's potential to produce high quantities of the pumps. This study will also attempt to show that the genetics will affect how well the P-gp blocking will work.

The Center for Disease Control reports that epilepsy afflicts 2.7 million Americans with annual costs of $15.5 billion. They estimate that 3% of Americans will have a diagnosis of epilepsy by age 80, and decided in 1997 to focus on treatment, with a motto of "no seizures, no side effects".

Antiepileptic drugs (AED) can fail, despite being structurally unrelated and acting on different parts of the nervous system. This refractory state constitutes up to 35% of the epilepsy population, and may be due to pharmacoresistance. Efflux transporters, such as P-glycoprotein (Pgp), are present at the bloodbrain barrier and serve to pump out structurally unrelated compounds, likely serving as a method for the removal of toxins (and drugs). Upregulation of efflux transporters such as Pgp by tumor cells are thought to contribute to chemotherapy resistant cancer tumors, but Pgp has also been found focally at seizure foci. Its overexpression was also noted in blood vessel endothelial cells following temporal lobe resection for intractable epilepsy. Case series have shown mRNA for MDR1, the gene encoding Pgp, to be 10x greater in the medial temporal lobes of patients with temporal lobe epilepsy, as compared to those without epilepsy. Pathological examination following surgical resections have found that epilepsy causing lesions such as cortical dysplasias, encephalitis, tuberculous leptomeningitis, tuberous sclerosis and astrocytomas express Pgp in neurons and/or glia, whereas normal brain parenchyma does not. In animal and cell research, upregulation has been seen following seizure induction and status epilepticus. Many AEDs are validated substrates to Pgp in animal studies. Delivery of these medications to the brain is likely associated with Pgp and in some cases, presence of the substrate may upregulate Pgp.

When Pgp inhibitors were added to animal models of drug resistant epilepsy, there were significant improvements in seizure frequency, duration and severity, providing proof-of-concept at the animal level. Carvedilol and verapamil, among other medications, have been found to be potent Pgp inhibitors. Verapamil and dexverapamil, either oral or intravenous, has been used as Pgp-inhibitors in clinical trials, with success as an adjuvant in malignant lymphoma and a phase III study as an adjunct in chemorefractory, metastatic breast carcinoma. There have been no clinical trials published using Pgp-inhibition in epilepsy.

Inclusion Criteria:

• probable or definite localization-related, primary generalized or symptomatic generalized epilepsy that is medically-refractory, as defined by treatment failure of at least 2 anti-epilepsy drugs at standard doses, despite medication compliance as determined by the treating neurologist
• at least 3 seizures/month in the 3-month period prior to randomization. Seizures that will be considered include generalized tonic clonic, complex partial, myoclonic and absence seizures. Simple partial seizures must have an observable motor component or have been otherwise been documented by videoEEG to be a definite seizure.
• Patients with prior epilepsy brain surgery or vagal nerve stimulator implantation will be allowed if medication and seizure frequency has been stable for the prior 3 months.
• Ages between 10 and 75 years will be eligible for inclusion. Elderly patients without a history or symptoms of cardiovascular disease may be eligible on a case-by-case basis. No patients older than 75 will be included due to the possible cardiovascular side-effects.
• Pre-menopausal women must be utilizing two reliable forms of birth control or abstinence
• ability of the patient to understand the concept of a clinical trial by answering the following questions appropriately: o will your seizures get better, worse or stay the same? Response in the spirit of: Any of the 3 could happen.

Exclusion Criteria:

• pregnancy or breast-feeding
• systolic blood pressure <100mmHg
• resting heart rate < 55 bpm
• concurrent calcium channel, beta-blocker or digoxin therapy
• Known hypersensitivity to carvedilol or any component of the formulation
• Decompensated cardiac failure requiring intravenous inotropic therapy
• Coronary artery disease with history of angina or Any cause of unstable angina
• Second- or third-degree AV block or sick sinus syndrome
• Bronchial asthma or related bronchospastic conditions
• Severe hepatic or renal impairment
• Active drug or alcohol dependence, that, in the opinion of a study investigator, would interfere with adherence to study requirements
• Any acute medical or psychiatric illness requiring inpatient admission; exceptions are elective epilepsy monitoring or elective procedures

The principal investigator has left the institution. Attempts to contact the PI have been unsuccessful. Columbia will never have access to the data. Thus, data will not be analyzed. The only information available is the number of participants who started and completed the study, which was last reported to and approved by the IRB in October 2009.