Qualification, Validation, and Verification 4

"If clearly visible growth of microorganisms is obtained after the incubation, visually comparable to that in the control vessel without product, either the product possesses no antimicrobial activity under the conditions of the test or such activity has been satisfactorily eliminated. The test for sterility may then be carried out without further modification."
It may be advantageous, and more consistent, for the text in Chapter ‹71› to be changed to "Suitability of the Test Method," if not to "Verification of the Test Method." The latter change also may be appropriate for Chapters ‹61› and ‹62›, given that what is being assessed is the verification that the actual test conditions relative to those established during the validation permits the proper functioning of the method. Given the harmonized status of these three chapters, such changes, although possible, would certainly take longer to become official.
The same cautions provided at the end of the section on validation are applicable here. If a method in use previously was derived from a pharmacopeial method but used for a purpose other than satisfying monograph requirements, it is not necessary to adopt a revised method in the pharmacopeia when it becomes official. It is therefore not necessary to reverify the suitability of your test article to the revised method. Likewise, the use of a nonpharmacopeial method for purposes other than satisfying a monograph requirement when a pharmacopeial method exists of potential relevance does not necessitate reverification.
General requirements for validationThere are numerous documents that describe the general approach to a validation process. They describe several characteristics (data elements in Chapter ‹1225›) that may be examined during validation, with specific sets selected based upon the nature of the test method. A brief description of these characteristics is provided herein using the characteristics as outlined in the IC Harmonization Harmonized Tripartite Guideline, Validation of Analytical Procedures: Text and Methodology.
Accuracy is a determination of how close the measured value is (in the case of an analytical method) to the true value. As such, one might define accuracy of method as equal to true value plus error. Error may contain both the systematic error (bias) and imprecision of measurement. With the potential error possible, it is important to include a means of reflecting the "true value" as closely as possible. For many compendial tests, this involves the use of a reference standard. Because a method is expected to be useful over a range of true values, the accuracy should be assessed over the expected range of values to which the method is to be applied. As stated previously, the validation should also state the conditions under which the accuracy was determined. Because it is not possible to determine all possible sets of conditions for which a compendial assay might be applicable, accuracy may need to be verified before use of a validated method. The concept of accuracy is more problematic for microbiological assays.
The precision of a method determined during validation should be representative of the repeatability (reproducibility) of the method. As was the case for the determination of accuracy, it should be determined over the expected range of articles to be measured, and the conditions used during the validation should be clearly stated. As for accuracy, the use of reference standards is common because the goal of the assessment of precision is to determe method repeatability without introducing unknown variance as a result of different test articles or test articles drawn from a heterogeneous source. The latter point also complicates the validation of microbiological assays.
Specificity refers to the ratio of false positives to false negatives. A highly specific method would have a very low ratio, given that it should be able to detect the article of interest present in very low quantities in the presence of much higher quantities of similar but not identical articles. As stated previously, specificity should be determined over the expected range of usage for the method, and conditions used during the validation should be clearly stated.
Linearity, in essence, refers to the existence of a direct relationship between the quantity of article contained in the sample being analyzed and the measured value resulting from the analysis. It is not the purpose of this article to delve into statistical intricacies pertaining to data transformation, the use of linear or nonlinear regression techniques, residual analysis, and so forth. Currently, it is sufficient that an assay purporting to be quantitative in nature must have a demonstrable quantitative relationship between the quantity of material of interest contained in the sample and the measured response.