International Harmonization of Regulatory Requirements for Average Bioequivalence and Current Issues in Individual Bioequivalence

Abstract

International harmonization of regulatory requirements for average bioequivalence has made major progress during the last years, particularly with regard to the bioequivalence range and statistical analysis. The multiplicative model and hence the logarithmic transformation of the primary extent and rate characteristics A UC and C,, and-consistent with this-the bioequivalence range of (0.80, 1. 25) are now accepted by all major health authorities including the European Committee for Proprietary Medicinal Products (CPMP), the United States Food and Drug Administration (FDA), and the Canadian Health Protection Branch (HPB). Such a consensus ismissing for the concept of individual bioequivalence. The selection of an appropriate methodology and suitable operative procedures for assessing individual bioequivalence must be addressed in the future by biostatisticians, biopharmaceutical scientists, and regulatory authorities.

South Africa Perspectives for BRICS - ‘Finger Print’ Profiling of the Trace and Heavy Metals in Ginseng and Hypoxis Commercial Available Herbal Products - A Quality Control Tool

“…Consistency was accepted where both duplicate sample determination were ≤20% CV for each trace and heavy metal investigated. The percentage CV criteria were based on the upper limit of acceptable analytical precision, and within the permissible statistical window for bioequivalence assessment [17][18][19]. Table 1 reflects the Ginseng and Hypoxis (African potato) products investigated (Table 1).…”

Clinical cross-over trials in phase I

“…The pro®les are then usually compared using summary measures. Area under the curve (AUC) is generally regarded as being the most important measure and there is international regulatory agreement 22 that trials should be analysed by showing that the 90% con®dence interval (CI) for the ratio of the AUCs should lie between the limits 80% and 125%. (Usually this has been done by analysing log-AUC using a linear model.)…”