Metabolites In Bioequivalence Assessment

In the majority of cases assessment of BE relies on the plasma concentrations of the parent drug since this is either the only reported therapeutic moiety or it is not metabolized. Concern is raised, however, when the parent drug is metabolized and the metabolite(s) exhibit comparable therapeutic activity with the parent drug. On the other hand, obvious reasons for measuring the metabolite(s) are (i) whenever an inactive prodrug is metabolized to an active metabolite and (if) the parent drug concentrations are too low, while metabolite(s) plasma levels are quantifiable. The reader can find several examples in the literature, whereas the target species for measurement is either the metabolite(s) or the parent drug and the metabolite(s) (68-76).

Computer-simulated BE studies are a powerful tool in this field of research since the modeling assumptions along with the values of the parameters are specified and the results can be contrasted with the assumptions used. The simulations are based on classical PK models with the formation of metabolite taking place during the presystemic absorption and/or during subsequent recirculation through the liver. The simulations try to explore which of the species is the most appropriate for BE decision making on the basis of statistical criteria such as the width of the relevant CIs. One should recall, however, that all these approaches are approximations of the reality because the complexity-variability in hepatic clearance can be also a function of the magnitude of alternative elimination processes for the drug and/or the metabolite. During the last 15 years or so, several simulation studies on the role of metabolites in BE have been published (77-82). Many of these studies have been reviewed by Midha and colleagues (83), and the use of metabolites in BE studies has been the subject of a recent Bio-International congress (28).

The first study (77) in this topic published in 1991 was based on a simple firstorder one-compartment PK model, with exclusive formation of a metabolite during recirculation through the liver. The authors focused on the rate of metabolite elimination being either limited by its formation or its excretion. Simulated BE studies were carried out with random error added to the absorption rate constant values of the R and T formulation. The statistical analysis based on the comparison of variability (using 90% CIs) associated with the Cmax values of parent drug and the metabolite revealed that the former was greater than the latter. Although their simulation results were contrasted with experimental BE studies of four drugs, caution should be exercised whether the drugs fulfill the modeling assumptions relevant to the metabolism of drug (83).

The second study by the same authors four years later (78) utilized a two-compartment model, with formation of the metabolite taking place either presystemically or during recirculation through the liver. Again, comparisons were based on the variability of Cmax values for the parent drug and the metabolite as a function of the variabilities used for the absorption rate constant of the parent drug, k.d, as well as the first-pass formation of the metabolite, kf. The variability of Cmax values of the parent drug and the metabolite was found to follow the magnitude of variability associated with k.d and kf, respectively.

The work of Tucker and colleagues (79) has been based on a model in which the formation of metabolite in the liver takes place both on first passage and on subsequent recirculation through the organ. The analysis was focused on AUC values derived from simulation studies of drug and metabolite kinetics. The PK parameters considered were intrinsic, CLint and renal clearance, CLr as well as the hepatic blood flow, <2h- According to the authors, metabolite data have to be used for high extraction ratio drugs, namely, CLint > Qh■ For low extraction ratio drugs (CLint < Qh), the parent drug data are preferred; however, when CLr is low, one has to use metabolite data. The basic conclusion of the study is that the within-subject variabilities of metabolic and renal clearances are the basic determinants for the use of drug or metabolite data since they determine the sensitivity of AUC to the differences of fraction of dose reaching the general circulation.

In similar work, Rosenbaum and Lam (80) studied the sensitivities of the parameters AUC and Cmax of the parent drug and the metabolite to variabilities associated with the intrinsic and hepatic clearance. A simple PK model was utilized with the formation of a single metabolite taking place during first passage. The statistical analysis of data revealed that the parent drug had wider 90% CIs around the point estimates for the ratio (T/R) of geometric means of AUC and Cmax than the corresponding one for the single metabolite. In a similar vein, Rosenbaum (81) used a semiphysiological pharmacostatistical model to study the manner in which intraindividual variability in hepatic clearance is transferred to AUC of a drug and its metabolite. The model assumes the formation of metabolite in the liver both on first passage and on subsequent recirculation through the organ. The results indicated that as the drug's hepatic extraction ratio increased, the variability of the drug's AUC was increased, whereas that of the metabolite decreased.

Jackson (82) carried out simulations, focusing on the response of parent drug and metabolite 90% CIs for AUC and Cmax to equivalent and inequivalent immediate release formulations. A linear first-pass model with random error added to the model parameters: renal clearance, hepatic clearance, systemic clearance, and liver blood flow. Specific values were assigned to the absorption rate constant and fraction absorbed to investigate problems associated with equivalent and nonequivalent immediate release formulations. According to Jackson (82), the Cmax for the parent drug provided the most accurate assessment of BE. On the contrary, the metabolite Cmax was found to be insensitive to changes related to rate of absorption. In addition, when the value of the intrinsic clearance is higher than the liver blood flow, the use of the metabolite Cmax data can lead to a conclusion of BE for truly bioinequivalent products.

In parallel, the use of prodrugs in therapy is pertinent to the matter since most of them are rapidly absorbed from the gastrointestinal tract and rapidly biotransformed to the active metabolite. Prodrug blood levels tend to be very low and much more variable when compared with the active metabolite. It should be noted that many prodrugs (ACE inhibitors, some statins, valacyclovir, fenofibrate) were not quantified with analytical methods of high sensitivity in PK studies by the innovator because of their short residence time and low blood levels. However, the continuous evolution in mass spectrometry allows today for the reliable measurement of prodrugs for a reasonable period of time. Thus, the measurement of both the prodrug and the active metabolite for the assessment of BE remains to be further evaluated. To emphasize the contradictory approaches as well as the incoherence of the description of the current guidelines (22,24) for the role of metabolites in BE assessment, we quote below two characteristic extracts. The FDA guideline (22) states, "The moieties to be measured in biological fluids collected in bioavailability and bioequivalence studies are either the active dug ingredient or its active moiety in the administered dosage form (parent drug) and, when appropriate its active metabolite. . . . Measurement of a metabolite may be preferred when parent drug levels are too low to allow reliable analytical measurement in blood, plasma or serum for an adequate length of time. ... If the metabolite contributes meaningfully to safety and/or efficacy, we also recommend that the metabolite and the parent drug be measured." The EMEA guideline (24) states, "In most cases evaluation of bioavailability and bioequivalence will be based upon the measured concentrations of the parent compound. In some situations, however, measurements of an active or inactive metabolite may be necessary instead of the parent compound. . . . Bioequivalence determinations based on metabolites should be justified in each case bearing in mind that the aim of a bioequivalence study is intended to compare the in vivo performance of T and R products. In particular if metabolites significantly contribute to the net activity of an active substance and the pharmacokinetic system is nonlinear, it is necessary to measure both parent drug and active metabolite plasma concentrations and evaluate them separately."

0 0

Post a comment