Use of Matrix Designs in Stability Studies: A Simulation Study
- PDF / 572,593 Bytes
- 10 Pages / 504 x 720 pts Page_size
- 68 Downloads / 229 Views
Dmg I n f o m r i u n Jounial. Vol. 33. pp. 859--868.1999 Pnnted in the USA. All rights reserved.
USE OF MATRIX DESIGNS IN STABILITY STUDIES: A SIMULATION STUDY KIMBERLY L. DEWOODY,PHD Riostatistics Department. Centocor Inc., Malvern. Pennsylvania
DAMARAJU RAGHAVARAO, PHD Department of Statistics. Temple University, Philadelphia, Pennsylvania
A stability study is designed to study the degrudution profile of a drug product over time. It is required that ut leust three batches of the drug product in its,finished form be studied over time. Often it is desired to manufacture the product in various forms, using a variety
of packaging options. If three batches of product are studied for each combination of these experimentul conditions at each o j the recommended sampling times, the cost of the study cun be substuntid. Simulation studies ure used to examine the ejyect of matrix designs; it is shown that the perjormonce of these reduced testing plans is similar to the full sampling plan in terms of poner to detect slope dqferences and estimution of the espiration dating period. Key Words: Drug stability study; Matrix designs; Uniform matrix designs; Expiration dating period
90 = a + b xED- 4.s [ n
INTRODUCTION AS PART OF AN APPLICATION for licensure of a new drug or biologic agent, it is necessary to provide data on the stability of the product over time. The Food and Drug Administration (FDA) has put forth guidelines (1) with recommendations on the design and analysis of stability studies. These guidelines are currently under revision (2). The 1987 guidelines focus on the requirements for establishing an expiration dating period based on multiple batches of a single drug product in a single packaging. For a drug product that has a linear degradation profile, the expiration dating period, xED,is defined as the point of intersection between the lower 9 5 8 confidence bound for the mean degradation curve and the lower product specification limit and may be found by solving Reprint address: Kimberly L. DeWoody, Centocor Inc.. 200 Great Valley Parkway. Malvem. PA 19355.
’ + (X - xEn)’/SS,]’’’
for xED,where 90 is the lower specification limit; n, x, and SS, denote the number of sampling times, the mean of the sampling times, and the sum of squared deviations of the sampling times. respectively; s denotes the sample estimate for the standard deviation of the random error term in the regression model related primarily to variability of the assay; a and b denote the sample estimates for the intercept and slope of the regression line; and t, is the appropriate critical value from the t-distribution with n-2 degrees of freedom. The recommended sampling times are at three-month intervals during the first year, six-month intervals during the second year, and yearly thereafter (1,2). If it is desired to manufacture the product in various forms using a variety of packaging options, the cost of sampling each combination of factors at each of the FDA recommended sampling times can be substantial. As a re-
859
Downloaded from di
Data Loading...
