Structured use of the median in the analytical measurement process.

In the pharmaceutical industry, the process of measuring a product's attributes can be very complicated and the potential for an analytical mistake can be quite high. Often, an unexpected result leads to an investigation to assess the possibility that a mistake was made in the laboratory. Traditionally, the data generated in these investigations has been used, along with various outlier tests, to attempt to negate the original data. Sometimes, historical estimates of the S.D. of the analytical method are not available for use in outlier testing and the power of the outlier tests to detect true mistakes without such historical estimates is often very low due to the small amount of data available. This leads to a great deal of inconsistency in the amount of data that is further generated and how the data is ultimately handled in making a decision. Recently, FDA demands for consistent and objective laboratory investigations have raised concerns about these practices. An alternative approach, involving a systematic investigation strategy and data handling via the structured use of the median, is proposed in this paper. The operating characteristics of the traditional and proposed approaches are compared to show their similarity and the advantages of the proposed approach. It is strongly believed by the authors that the structured use of the median will lead to more consistent investigations and data handling, which will benefit industry, the FDA and ultimately, the consumer, by allowing more accurate decisions to be made more efficiently.

Evaluation of key sources of variability in the measurement of pharmaceutical drug products by near infrared reflectance spectroscopy.

Potential sources of variability in the measurement of solid oral drug products by near infrared reflectance spectroscopy were evaluated with statistical experimental design. Spectra were collected for two different tablet types according to the data collection and treatment parameters defined by the experimental design. Each tablet had three different dose-levels. Libraries were constructed using second-derivative spectra. Key figures-of-merit generated during internal and external library validation were used to calculate which parameters most strongly influence the library performance for dose-level discrimination. These responses and their corresponding experimental conditions were evaluated with the screening model in the JMP program. Segment value used for the second-derivative calculation was an an influential factor and had a complex effect. Orientation on the sampling platform also had an influential effect for embossed tablets. Collection of spectra over fewer days decreased variability within the library. More frequent reference spectrum collection improved the performance of libraries to a small degree. A larger sample population increased the range of spectral variability within a dose-level but apparently not the overall performance of the library. The number of scans averaged per spectrum was not an influential factor in this study. These results are summarized and used to recommend an approach to dose-level discrimination.

Statistical approaches to determine analytical variability and specifications: application of experimental design and variance component analysis.

Assessment of analytical variability is recognized as an important factor for the establishment of specifications. Estimation of the variance for an analytical procedure can be accomplished using a variety of approaches. The approach of variance component analysis was applied retrospectively, as well as prospectively, to estimate analytical variance. The prospective approach also included the use of experimental design. Recent new drug substance examples illustrating these approaches are presented. In these examples, the analytical property of potency was evaluated. Factors examined in the experimental design include laboratory, day, analyst, instrument and column. Process variability can also be determined by variance component analysis. For a stable drug substance, combining the analytical and process variances provides an estimate on the total variance for the analytical property of potency. With the total variability statistically derived, an appropriate specification that is consistent with process and analytical capability can be established.

Blend uniformity analysis: validation and in-process testing. Technical Report No. 25. PDA (Parenteral Drug association).

One of the most challenging problems confronting oral solid dosage form manufacturers today is the difficulty in applying scientifically valid methods to blend uniformity validation. The Wolin decision in U.S. v. Barr Laboratories caused the FDA to reexamine and modify its policies on blend uniformity and sampling techniques. The resulting policies are predicated on the assumption that current technology provides a means to consistently collect minute representative samples from much larger static powder blends. A comprehensive review of the scientific literature indicates that current sampling technology is plagued by a propensity for sampling bias. This limitation is particularly troublesome to the pharmaceutical industry which must then hold these samples to very high standards. In response to these concerns the PDA Solid Dosage Process Validation Committee has reviewed approaches to blend uniformity analysis. This technical report provides: a method to determine the appropriate sample size for each product, a holistic approach for establishing meaningful acceptance criteria, a discussion on the use of proper analytical techniques and recommendations for conducting investigations for out-of-specification results.