A Rapid LC–MS-MS Method for the Quantitation of Antiepileptic Drugs in Urine

Epilepsy is a common neurologic disease that requires treatment with one or more medications. Due to the polypharmaceutical treatments, potential side effects, and drug-drug interactions associated with these medications, therapeutic drug monitoring is important. Therapeutic drug monitoring is typically performed in blood due to established clinical ranges. While blood provides the benefit of determining clinical ranges, urine requires a less invasive collection method, which is attractive for medication monitoring. As urine does not typically have established clinical ranges, it has not become a preferred specimen for monitoring medication adherence. Thus, large urine clinical data sets are rarely published, making method development that addresses reasonable concentration ranges difficult. An initial method developed and validated in-house utilized a universal analytical range of 50–5,000 ng/mL for all antiepileptic drugs and metabolites of interest in this work, namely carbamazepine, carbamazepine-10,11-epoxide, eslicarbazepine, lamotrigine, levetiracetam, oxcarbazepine, phenytoin, 4-hydroxyphenytoin, and topiramate. This upper limit of the analytical range was too low leading to a repeat rate of 11.59% due to concentrations >5,000 ng/mL. Therefore, a new, fast liquid chromatography–tandem mass spectrometry (LC–MS-MS) method with a run time under 4 minutes was developed and validated for the simultaneous quantification of the previously mentioned nine antiepileptic drugs and their metabolites. Urine samples were prepared by solid-phase extraction and analyzed using a Phenomenex Phenyl-Hexyl column with an Agilent 6460 LC–MS-MS instrument system. During method development and validation, the analytical range was optimized for each drug to reduce repeat analysis due to concentrations above the linear range and for carryover. This reduced the average daily repeat rate for antiepileptic testing from 11.59% to 4.82%. After validation, this method was used to test and analyze patient specimens over the course of approximately one year. The resulting concentration data were curated to eliminate specimens that could indicate an individual was noncompliant with their therapy (i.e., positive for illicit drugs) and yielded between 20 and 1,700 concentration points from the patient specimens, depending on the analyte. The resulting raw quantitative urine data set is presented as preliminary reference ranges to assist with interpreting urine drug concentrations for the nine aforementioned antiepileptic medications and metabolites.