Scientific and Regulatory Standards for Dissolution Similarity: a Statistical Game of Matchmaking Neither Force to Pass or Nor Fail to Pass
摘要
The principal of generic product development is to match the critical quality attributes. Most of the time during complex product development, life cycle management; biowaiver, pre- and post-change approvals the significant efforts are made by scientist to match the drug release profile. In order to get the vivo bioequivalence testing waived based on in vitro performance of drug product; the dissolution testing is mostly act as a surrogate or performance indicator. Hence, assessment of similarity or equivalence of release profile is most critical aspect with respect to regulatory decision making. Available guideline defines the methodologies and acceptance criteria for same based on data structure e.g., application of mathematical and statistical model like similarity factor (F2), bootstrapped F2, model independent and model dependent approach etc. However, during regulatory review lot of discrepancies usually raise by regulators with respect to similarity demonstration like selection of proper methodology, define suitable acceptance criteria in case of high variability. Current article emphases on the visions behind regulatory expectations, with respect to dissolution profile comparison and highlights the prerequisites and answer the common question like how to choose the correct methodology, what are the limitations, way forward and regulatory expectations and alternative methodologies in order to evaluate the dissolution data statistically to make wise decision on in vitro equivalence. Overall, various approaches are available for dissolution similarity analysis. However, the intension of statistical comparability should be like; neither force the dissimilar product to pass the criteria, nor fail the product which are similar. This comprehensive review will enhance the overall understanding and help the formulation and biopharmaceutics scientists; how to ensure regulatory compliance during similarity evaluation.
Graphical Abstract