Chapter 2 focuses on design and analysis of short-term stability studies such as accelerated testing. Secondary structure was evaluated using circular dichroism. Lin indicated that a matrixing design should not be applied to sampling times at two endpoints i. Based on the appropriate choice of a subset of T1 , Nordbrock 1992 provided some useful reduced stability designs for long-term stability studies at room temperature: Design 1 complete : Every batch-by-strength-by-package combination is chosen. Several methods for determination of drug shelf-life have been proposed, as discussed in detail in the previous chapter.
In addition, it is intended to give a well balanced summary of current and recently developed statistical methods in stability analysis. However, it may be of interest for the pharmaceutical companies to establish an interval estimate rather than a point estimate for the drug shelf-life. This book provides the material necessary for you to perform stability designs and analyses in pharmaceutical research and development. It is acceptable to use a matrix when there are multiple sources of raw materials e. As a result, one of the following models can be selected for determining the degradation pattern of the drug product by testing the above hypotheses. In this case harmonization of the definition of room temperature may not be useful. There are, in general, two methods for determining the strength i.
For strength, we only test the lowest and highest strengths i. Minimum Approach for Multiple Batches If the preliminary tests for the equality of slopes and the equality of intercepts are not rejected at the 0. In the first stage S1 of disintegration testing, six dosage units are tested. These methods include methods proposed by Chow and Shao 1991 , Shao and Chow 1994 , and Ho, Liu, and Chow 1993. There are also reports in which the drug has been decomposed by exposing it to one, two, three, four, or more conditions among acidic, neutral, or alkaline hydrolysis, photolysis, oxidation, and thermal stress. The expiration dating period is determined based on a statistical confidence or prediction interval analysis. Two of the three batches should be at least pilot scale.
Some were stable for as long as 10 years beyond their expiration dates. In practice, it is of interest to consider a matrixing or bracketing design. Pharmaceutical companies often receive queries regarding the safety of newly expired drug products. These baisc design considerations include background information, regulatory considerations, design factors, sampling time considerations, sample size, statistical analysis, and other issues, which are described below. In addition, shelf life projection is usually based on confidence band of a regression line. If each combination is to be tested at the time intervals of 0, 3, 6, 9, 12, 18, 24, 36, and 48 months for a 4-year stability study, there are a total of 270 3 × 3 × 3 × 10 assays. In addition, the guidelines require that all estimated expiration dating periods be applicable to all future batches.
It then compares some commonly employed study designs and discusses both fixed and random batch statistical analyses. As a result, the design of the matrix will be dictated by the factors needing to be covered and evaluated. Formulated micro pellets showed delayed in vitro dissolution behavior, probably due to optimized concentration of polymer. Different intercepts or slopes may indicate that there is batch-to-batch variation in intercepts or slopes. Hence, the expected value of M S β in the analysis of covariance table presented in Table 5. Further, some adverse reactions in patients have been traced to impurities in the active ingredient. Hence, v t is a positive convex function.
Metformin, a guanidine derivative is commonly used as a drug of choice for the treatment of non insulin dependent diabetes mellitus. It is not a common practice to increase test power by increasing the level of significance. Levothyroxine sodium is the sodium salt of the levo isomer of the thyroid hormone thyroxine T4. Also included in this section are issues that often occur in the design of accelerated stability testing. Where a special procedure is specified in the test for content uniformity in the individual monograph, the results should be adjusted. It should be noted, however, that the above comparison of methods did not take this regulatory practice into consideration.
The other method is to simply fit a nonlinear regression model directly to the original data of the reaction rates. It has been demonstrated that impurities in a finished drug product can cause degradation and lead to stability problems. The response surface methodology and multiple response optimizations utilizing a polynomial equation were used to search for the optimal formulation with a specific release rate at different time intervals. Such scale-ups to commercial size production may produce significant problems in consistency among batches. The design assumes that the stability of each subset of samples tested represents the stability of all samples at a given time point. The lower confidence limit shelflife estimate is 55. These design considerations include: r Batch sampling considerations r Container-closure and drug product sampling r Sampling-time considerations 76 Stability Designs These designs considerations focus primarily on how data are to be collected.
Hoepli 5, 20121 Milano - Italy Tel. The first method provides an initial strength by definition of exactly 100 without error. Half the combinations chosen are tested at 3, 9, 18, 36, and 48 months, and the other half are tested at 6, 12, 24, and 48 months. In addition, this book is intended to give a well-balanced summary of current regulatory perspectives and recently developed statistical methods in the area. These criteria mainly focus on the power of detecting factor effect, which are briefly described below.