Statistical Evaluation of Interchangeability of Biosimilar Drug Products
Date Issued
2015
Date
2015
Author(s)
Hsiung, Ya-Hsuan
Abstract
Biological products are defined as therapeutic agents or medical devices manufactured by living systems or living organisms. Because of the complexity of manufacturing process, biological products are costly. A Biosimilar product is approved if it is highly similar to a reference biological product which is under patent protection. So, after the expiration of the patent protects of the reference products, the biosimilar with the same therapeutically effect may get into the market and saves numerous patients who can only afford the less expensive biosimilar product than the reference one. However, unlike the conventional chemical drug with small molecules, the biological products are composed of complex compounds such as proteins, nucleotides. Even a minor difference in clinically inactive components may cause a serious adverse effect or death. Thus, the traditional methods in evaluation for bioequivalence of a generic drug are not appropriate for evaluating biosimilar products. Therefore, stricter statistical criteria are necessary for assessment between the biosimilar drug product and the reference product. One of the requirements for biosimilar drug products stated by the United States Food and Drug Administration is that biosimilar drug products are expected to achieve the same clinical effects in any given patient as the reference biological products. Therefore, biosimilar drug product and the reference biological product are interchangeable. Interchangeability dictates that the marginal distribution of the responses from the biosimialr product and its reference are close within a pre-specified allowable limit. We propose the variance of the subject-by-formulation interaction (SBFI) as a criterion for evaluation of interchangeability. Under the two-period four-sequence (2 x 4) crossover design, we develop a statistical testing procedure based on SBFI. Simulation studies were conducted to empirically investigate the size and power of the proposed method. Sample size determination is also suggested. Numerical examples are used to illustrate the applications of the proposed method.
Subjects
Biological product
Biosimialr product
Interchangeability
Subject-by-Formulation Interaction
Cross-over Design
Modified Large Sample Method
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-104-R02621206-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):0155770e5f4f24e4ce24b613f9a63686