Although Subpart I has not yet been formally incorporated into GMP regulations, which are currently under review, the Food and Drug Administration (FDA) has issued guidelines that define the agency's expectations in this field. Initially, the FDA published those sections of the International Conference on Harmonisation (ICH) Guidelines dealing with analytical procedures that had been recommended for adoption by the parties to ICH. These were: Q2A Text for Validation of Analytical Procedures, October, 1994 1 and Q2B Validation of Analytical Procedure: Methods, November, 1996.⁷ Document Q2A lays down broad guidelines on the types of tests that require validation, while Guideline 2B gives additional details on validating various test characteristics. The European Union, Japan, Canada, and Australia are among the parties that have also adopted these ICH guidelines.

FDA issued two further documents. The first was "Guidance for Industry: Analytical Procedures Validation," in August 2000.⁸ This details analytical methods and procedures validation required for the chemistry, manufacturing, and controls (CMC) documentation for each investigational new drug (IND), new drug application (NDA), and biologic license application (BLA), with a description of the validation package that should be submitted with each type of application.

The FDA's "Guidance for Industry: Bioanalytical Method Validation" followed in May, 2001.⁹ This discusses the development of methods for chemical, microbiological, and ligand-binding assays of drug products and their metabolites in "biological matrices" such as blood, serum, plasma, or urine. These would be required for pharmacological and bioavailability studies on drugs. The guidance also considers the application of validated methods to routine product analysis. More recent FDA guidelines incorporate recommendations contained in earlier ICH documents. They form the basis for the advice on analytical method validation in this article.

Qualification, Validation and Verification

So what does this mean in real terms? The need to characterize the product more stringently and to use validatable production and testing methods increases with the progress of the drug's development. In pre-clinical testing and Phase I and early Phase II studies, it is possible to use analytical methods that are not fully validated, but "qualified." A qualified method is one for which there is insufficient knowledge of the test's performance to document full validation, but some effort has been made to determine the method's reliability and how to control variability — a performance assessment.

Proper test qualification follows a protocol similar to full validation, but over a shorter time frame and with only a few parameters tested. If the assay is to be used in commercial production, however, it is essential that data developed during test qualification studies can be applied to eventual validation studies. Alternatively, it could be that qualification work will indicate that the method is not unsuitable — another valuable result.

A set of rules must be established for control and use of test qualification. Apart from ensuring that the test method is clearly described in a standard operating procedure (SOP), it must be subject to the same change control rules as a validated test; most importantly, the SOP should specify when the qualified method can be used — for example "for the assay of Phase I clinical lots only." Because there will be only limited experience of the test's performance, it is desirable that it be performed by an experienced analyst capable of identifying potential risks in the procedure and knowledgeable regarding how to ensure that test variability is minimal. Proper calibration of apparatus and standards should be routine in a qualified test.

By the time one reaches Phase III clinical trials, authorities expect that processes and test methods are those that will be used to manufacture the product for sale. In fact, it only makes sense to perform expensive, large-scale definitive trials on a product that truly represents what will eventually be marketed. This is particularly important with biopharmaceutical products, as manufacturing processes are still considered to have a major effect on final product characteristics. Therefore, more reliance must be placed upon fully validated analytical methods, especially for the release of the final product. The general consensus is that the suitable time for a changeover from qualified methods to validated methods for a biopharmaceutical product is at the Phase IIb stage.