Abstracts

Rationale: Generic modified-release (MR) formulations of the antiepileptic drugs (AEDs) carbamazepine, phenytoin sodium, divalproex sodium, and levetiracetam are currently marketed in the US. MR formulations offer the advantages of less frequent dosing, improved maintenance of therapeutic drug plasma concentrations, and better patient compliance. In a previous study, we found that the smoother drug release provided by a MR AED formulation decreased serum fluctuations, leading to fewer breakthrough seizures. Thus, the objective of the present retrospective study is to evaluate the robustness of the bioequivalence (BE) measures determined in the pivotal clinical studies submitted to the FDA to support marketing approval of generic MR AEDs. BE studies, which determine whether a generic and corresponding brand drug provide the same rate and extent of absorption, are generally performed in healthy normal subjects given single doses of the drugs. The BE measures compared statistically are Cmax (peak drug plasma concentrations, to assess absorption rate), and AUC (total area under the drug plasma concentration versus sampling time profile, to assess absorption extent). Methods: We collected Cmax and AUC data from the pivotal BE studies used to support the approvals of 25 generic AED MR formulations. Data were from BE studies under both fasting and fed conditions. We determined distributions of 90% confidence interval (CI) limits for generic/brand ratios of AUC and Cmax in the BE studies. We also obtained from the FDA-approved labeling information about the inactive ingredients in the generic and brand formulations. Results: We evaluated 53 BE studies, including data from 1570 subjects. The 90% CIs of the generic/brand AUC and Cmax ratios differed by <15% in 86.8% and 77.4% of the BE studies, respectively. The 90% CIs of the generic/brand AUC and Cmax ratios differed by 15 to 25% in 13.2% and 22.6% of the studies, respectively. In general, BE measures were more variable for MR AEDs, compared to our previous observations of immediate release approved generic and brand AEDs. However, all MR AED products surveyed were well within the BE acceptance limits of 80 to 125% (90% CIs). Intersubject variability in Cmax and AUC was moderate and did not differ significantly between generic and brand. MR AED formulations of poorly soluble drugs demonstrated higher rates of variability irrespective of whether the formulation was generic or brand. Conclusions: Generic MR AED formulations provided total and peak drug delivery similar to that of brand products, ensuring therapeutic equivalence. Generic formulations were bioequivalent to the corresponding brand formulations despite differences in excipients and release mechanisms. This comparison of BE measures obtained in the pivotal studies supporting the approvals of generic MR AEDs suggests that patients can switch from brand to generic formulations with little or no changes in drug plasma concentrations and thus no apparent compromise in the safety and efficacy.