A Simple Three-Step Procedure for Parametric and Nonparametric Assessment of Bioequivalence
- PDF / 719,654 Bytes
- 14 Pages / 504 x 719.759 pts Page_size
- 96 Downloads / 166 Views
Drug Information Journal. Vol. 31. pp. 167-180, 1997 Printed in the USA. All rights reserved.
A SIMPLE THREE-STEP PROCEDURE FOR PARAMETRIC AND NONPARAMETRIC ASSESSMENT OF BIOEQUIVALENCE JOUNIVUORINEN,
M S C , AND
JARITURUNEN, MSC
Department of Biomemcs, Farrnos Research, Orion Corporation, Turku. Finland
A simple three-step approach for the assessment of bioequivalence is proposed. According to the decision rule of this approach, average bioequivalence is assessed first (Step 1). lf approved, then population bioequivalence will be assessed next (Step 2). Again, if approved, then individual bioequivalence can be assessed as thefinal stage (Step 3). For the evaluation of these bioequivalence criteria, parametric and nonparametric testing procedures are suggested on the original and logarithmic scale of measurement. A clear advantage of these testing procedures is that they can be set up using elementary test statistics and, therefore, they are readily accessible to the practitioners. An example is also presented to illustrate the method. Key Words: Crossover design; Bioequivalence; Drug interchangeability
INTRODUCTION THE CURRENT REGULATORY guidelines only require evidence of equivalence in average bioavailabilities for the assessment of bioequivalence (1,2). There has been a long-standing concern, however, regarding the interchangeability of bioequivalent products. Anderson and Hauck (3) identified two types of interchangeability, what they termed prescribability and switchability, and two corresponding types of bioequivalence, what they called population and individual. A population bioequivalence criterion is for sufficient similarity of the distributions of the two formulations. This will mean variances, not just averages. They also noted that population bioequivalence is sufficient for prescribability, that is, the selection between two products for a drug naive patient. Individual bioequivalence, on the other hand, is needed for a patient who has been taking a product and is to be switched to another formulation. In other words, it must be assured that a subject taking one formulation can switch to the other and retain essentially the same toxicity and efficacy profiles. There seems to be widespread acknowledgment of the importance of ensuring interchangeability of two formulations declared bioequivalent (4-7). There seems to be much less acceptance, however, of the need to change the design of the studies and the statistical methods. These types of
Reprint address: Jari Turunen, Departmenr of Biomemcs, Farmos Research, Orion Corporation, P.O.Box 425, 20101 Turku, Finland.
167 Downloaded from dij.sagepub.com at DALHOUSIE UNIV on April 13, 2015
Jouni Vuorinen and Jari Turunen
168
concerns have led to the development of a simple three-step decision rule for bioequivalence assessment under the two-period crossover model. According to this decision rule, average bioequivalence is assessed with regard to 8,a measure of relative average bioavailability. The two possible decisions about 8 are defined by: N
Data Loading...
