Maureen J. Rawson

Evaluation of the bioequivalence of highly-variable drugs and drug products

AbstractPurpose. To establish procedures for the effective evaluation of bioequivalence (BE) for highly-variable drugs and drug products (HVD/P). Methods. 2- and 4-period crossover BE studies with 24 subjects were simulated which generally assumed within-subject coefficients of variation of 40%. The relationship between the fraction of studies in which BE was accepted (the statistical power) and the ratio of geometric means (GMR) of the two formulations was evaluated for various methods of analysis. These included, primarily, scaled average BE (ABE), the corresponding approach of expanding BE limits (BEL), and, for comparison, unscaled ABE and scaled individual BE (IBE). Results. Scaled ABE and expanding BEL showed very similar properties in both 2- and 4-period studies. They had steeper power curves than scaled IBE. Unscaled ABE had very low statistical power. The acceptance of BE by unscaled and scaled ABE and expanding BEL was almost independent of subject-by-formulation interaction and the ratio of within-subject variations of the two formulations. By contrast, the conclusions reached by scaled IBE were strongly affected by these parameters. Conclusions. Scaled ABE and expanding BEL evaluate BE effectively for HVD/P in both 2- and 4-period investigations. However, additional, useful information can be obtained from 4-period studies.

The Role of Metabolites in Bioequivalence

The role of metabolites in bioequivalence studies has been a contentious issue for many years. Many papers have published recommendations for the use of metabolite data based on anecdotal evidence from the results of bioequivalence studies. Such anecdotal evidence has validity, but the arguments lack weight because the “correct” answers are always unknown. A more promising area of exploration is recommendations based on simulated bioequivalence studies for which the “correct” answers are known, given the assumptions. A review of the literature, however, reveals scant evidence of attempts to apply to real data the pharmacokinetic principles on which the recommendations from simulated studies relied. We therefore applied those principles (based on estimates of intrinsic clearance after oral administration of the parent drug) to four bioequivalence studies from our archives, in which the parent drug and at least one metabolite were monitored. In each case, the outcome is discussed in the context of the complexity of the metabolic processes that impact on the parent drug and the metabolite(s) during the first passage from the intestinal lumen to the systemic circulation. Our observation is that no simple generalization can be made such that each drug/metabolite combination must be examined individually. Our recommendation, however, is that in the interests of safety, bioequivalence decision-making should be based on the parent drug whenever possible.

Bioequivalence: Switchability and scaling

This study demonstrated that the mere fact that two multisource drug formulations are bioequivalent with the same reference formulation does not guarantee that they are bioequivalent with each other. The present unscaled bioequivalence limits (BEL) of 0.80 to 1.25 permitted far greater deviation from unity of the geometric mean ratio (GMR) for multisource formulations with low within-subject variabilities than for drugs with high variabilities. Scaling the BEL drastically reduced the maximum deviation from unity of GMRs for two multisource formulations each bioequivalent with the same reference product while broadening the BEL for highly variable drugs. It was concluded that scaling was consistent with the principle of switchability for toxic drugs with low variability and for safe, highly variable drugs. On the other hand, scaling need not be applied to safe drugs with low variability and should not be applied in the unusual case of a highly variable drug with a narrow therapeutic range.

The application of partial areas in assessment of rate and extent of absorption in bioequivalence studies of conventional release products: experimental evidence

Abstract A bioequivalence study on perphenazine/amitriptyline combination compared three formulations (TST-1, TST-2 and REF) containing the same nominal dose of perphenazine, while TST-2 contained 20% less amitriptyline than the other two formulations. ANOVA-SNK on log-transformed partial areas AUC024 through AUC0∞ showed the intra-subject variabilities of perphenazine to be consistently higher than those of amitriptyline or nortriptyline. There was also a marked increase in intra-subject variability for perphenazine on extrapolating AUC0last to infinity. Thus, whhile the 1–2α confidence intervals (90% CIs) and ratios of geometric means for perphenazine fell within preset limits of 80–125% in all pairwise comparisons, the CIs for AUC0∞ violated the upper limit in comparisons of TST-2 and REF (or TST-2 and TST-1). This outcome could be attributed to high intra-subject variability on AUC0∞. ANOVA-SNK of amitriptyline identified a significant difference in extent of amitriptyline absorbed from TST-2 compared with TST-1 and REF. Thus, in pairwise comparisons between TST-1 and REF, all 90% CIs for amitriptyline partial areas AUC04 through AUC0∞ fell within 80–125% limits. In corresponding comparisons between TST-2 and REF (or TST-2 and TST-1), however, the lower limit of the 90% CIs for AUC01 through AUC0last consistently fell below the 80% cut off. By contrast, 90% CIs associated with the ratios Cmax/AUC04 through Cmax/AUC0∞ all fell within preset limits indicating no differences in absorption rate of amitriptyline, consistent with the theoretical argument that Cmax/AUC is influenced by rate but not extent of absorption. Thus the partial area method is applicable to the evaluation of both relative rate and extent of absorption from conventional release products. Similarly, analyses of nortriptyline data showed that the partial area method is also effective in the evaluation of rate and extent of formation of a metabolite in a bioequivalence study. The partial area method permits a greater number of plasma samples to be devoted to accurate estimation of tmax and Cmax since the necessity to estimate AUC0∞ is obviated.

Effects of food on the pharmacokinetics of methylphenidate.

AbstractPurpose. To test the hypothesis that the pharmacokinetics of d-meth- ylphenidate (d-MPH) would be altered by food ingested before administration of an immediate release formulation (dl-MPH- IR) but not when food is ingested before a slow release formulation (dl-MPH-SR). Methods. A randomized, four-phase, open label, crossover design was conducted in 24 healthy men who each received, on separate occasions, dl-MPH-IR and dl-MPH-SR taken after an overnight fast and 15 min after a standardized breakfast (20% protein, 21% fat, 59% carbohydrate). Plasma MPH levels were monitored by a validated, stereoselective, GLC-ECD method. Results. For plasma d-MPH, there were significant differences (ANOVA) between dl-MPH-IR and dl-MPH-SR in tmax, Cmax (peak exposure), and Cmax/AUC (sensitive to rate of absorption). Dl-MPH-SR on average delayed tmax from 2.3 to 3.7 h and lowered Cmax 34%. There was no significant difference between the formulations in AUC (extent of absorption). For dl-MPH-IR, food significantly increased Cmax (23%) and AUC (15%) and for dl-MPH-SR the corresponding increases were Cmax (17%) and AUC (14%). After dl-MPH-IR, food delayed average tmax from 2.0 to 2.5 but had no effect on tmax after dl-MPH-SR. There was no effect of food on Cmax/AUC (rate of absorption). Conclusions. Food caused a significant increase in extent of absorption but had no effect on rate of absorption of d-MPH after either dl-MPHIR or dl-MPH-SR.

Prescribability and switchability of highly variable drugs and drug products

At the AAPS/FDS Workshop (Crystal City, Arlington, VA, March 6-8, 1995), it was agreed that some form of scaling should be permitted for highly variable drugs, although there was no agreement on a method. Currently, much emphasis is focused on developing a practical methodology for individual bioequivalence (IBE) and population bioequivalence (PBE) to replace or complement average bioequivalence (ABE). The latter requires only the mean bioavailabilities of two formulations to be sufficiently similar, whereas PBE also considers their distributions. IBE, on the other hand, is a comparison of the individual responses to the two formulations within subjects and is therefore concerned with switchability (interchangeability) between two multisource formulations. Multisource formulations refer (i) to generic copies of an innovators formulation or (ii) to different formulations used in stages leading up to the final marked formulation. Evaluation of both PBE and IBE require replicate design studies. The FDA Working Group on IBE has been experimenting with methods in which a one-sided 95% confidence interval is computed based on the Bootstrap technique which ensures that the consumer risk is maintained at 5%. The IBE metric can then be scaled according to the within subject variance of the reference formulation. Thus if the variability of the test formulation (T) is higher than that of the reference (R), the formulation may fail IBE but not ABE. Conversely, if R is more variable than T, then the formulations may be considered to be IBE, even with a difference in means of more than 20%. Experimentation with existing data on our files shows that scaling has a considerable effect on the IBE decision for highly variable drugs. Evidence will also be presented to show that scaling makes the conditions more conservative for potent drugs with steep dose response curves reducing the risk of two generic formulations being BE with the same reference product but not BE with each other. On the other hand, broadening the BE limits for safe, highly variable drugs increases statistical power and reduces the number of subjects required. Even with the introduction of scaling, however, it is clearly difficult to obtain a single IBE criterion suitable to be applied to all drug products/studies.

The roles of stereochemistry and partial areas in a parallel design study to assess the bioequivalence of two formulations of hydroxychloroquine: A drug with a very long half life

Abstract A bioequivalence study was carried out to compare the bioavailabilities of two formulations containing racemic hydroxychloroquine. The study was based on a single phase parallel design because the enantiomers of hydroxychloroquine have very long terminal phase half lives. A group of n = 72 healthy young men were randomly assigned to receive a single dose of either test ( n = 36 subjects) or the reference ( n = 36 subjects) formulation after which serial blood samples were harvested over 77 days. The two products were found to be bioequivalent in that 90% confidence intervals fell entirely within bioequivalence limits of 80–125% for the log transformed parameters C max and AUC 4 whether calculated from non-chiral or enantioselective blood concentrations. Similarly, 90% confidence intervals for the partial areas from AUC 4 through AUC Inst all fell within 80–125%. These data suggest that enantioselective assays were not required and that blood samples need only have been harvested up to 12 h.

The impact of stereoisomerism in a bioequivalence study on two formulations of doxepin

Abstract There are few archival data on the impact of stereoisomerism on bioequivalence studies. A bioequivalence study was carried out on two formulations of doxepin containing 15% of the active cis isomer and 85% of the less active trans isomer. Plasma concentrations of cis N-desmethyldoxepin remained at approximately 15% of total in only 2 of 30 healthy volunteers who completed the study. In the remaining 28 subjects, however, the proportion of cis N-desmethyldoxepin steadily increased with time such that by 72 h post dose, cis N-desmethyldoxepin comprised approximately 90% of the metabolite present in plasma. Nevertheless, the 90% confidence intervals (In AUClast, In Cmax and In Cmax/AUClast) for N-desmethyldoxepin fell entirely within bioequivalence li whether stereoselective or non-stereoselective data were analyzed. Thus the results of this study did not support arguments in favor of the use of stereoselective methods in bioequivalence studies.

Retrospective evaluation of relative extent of absorption by the use of partial areas under plasma concentration versus time curves in bioequivalence studies on conventional release products

Abstract Retrospective analyses were carried out on 22 analytes involved in nine bioequivalence studies on conventional release products, in which 90% confidence intervals were calculated for partial areas under the plasma concentration versus time curves (AUCs) from the first sampling time to AUC last , The use of partial areas up to 2× the maximum t max among all subjects in any given study gave the same bioequivalence decisions as AUC last .