Validation of a novel approach for dose individualization in pharmacotherapy using gabapentin in a proof of principles study

Study objective: To demonstrate the premise of individualized dosing charts (IDCs) as a clinical-bedside decision-support tool to individualize dosage regimens for drugs in which the interpatient variability is controlled by the pharmacokinetic (PK) behavior of the patient, to calculate the optimal sampling schedule (OSS), which minimizes the number of blood samples per patient. The approach is illustrated with available PK data for gabapentin.

Design: Retrospective proof of principles study using gabapentin PK data from a published clinical trial.

Patients: Nineteen subjects in a trial designed to uncover the importance of the genetic contributions to variability in gabapentin absorption, renal elimination, and transport; subjects were monitored for 36 hours after administration of a single dose of gabapentin 400 mg, and plasma concentrations were determined at 14 time points.

Measurements and main results: When the PK profiles were different between subjects, the IDCs are dramatically different from each other and from the IDC for an "average" patient representing the patient population. The dose amount and dosing interval must be adjusted to maximize the probability of staying within the target concentration range. An optimal sampling methodology based on the assumption-free Bayesian approach is used to distinguish the PK profile of an individual patient from the patient population. In the case of gabapentin, only two optimally selected test blood samples, at 1.5 and 6 hours after administration of a single doses, were necessary. The average sensitivity and the average specificity of the OSS was 99% and 96%, respectively.

Conclusion: IDCs display the risk of a patient violating the target concentration range for any dosage regimen. They can be used as a clinical-bedside decision-support tool in a patient-physician partnership to decide on a dose amount and dosing interval that are medically acceptable while practical and convenient to ensure compliance. By using the assumption-free Bayesian approach and the OSS, the number of samples required from a new patient to individualize the dosage regimen can be reduced significantly while preserving high levels of sensitivity and specificity. Prospective studies are being planned to validate the encouraging results. This approach can be extended to any drug if PK data and a target concentration range are available for either therapeutic drug monitoring or target concentration intervention.