Recently, the FDA updated a long-standing, decades old guidance on analytical test method validation based on revisions of the ICH Q2(R2) guidelines. Traditional test method validation requirements have been streamlined and altered to both provide flexibility on new types of multivariate and non-linear analytical test methods, and to focus on the validation-critical parameters that show that the method is reliable during routine use.
Overview of FDA’s Updated ICH Q2(R2) Guidelines
What has not changed is the expectation that a test method is thoroughly developed and suitable for routine use. Test methods should have a defined purpose and measure some critical characteristic(s) of the drug substance or product. Validation protocols and pre-defined acceptance criteria should be written prior to validation. Analytical test method validations must be completed prior to NDA submission and prior to characterization and release of the pivotal clinical trial materials used in Phase III clinical studies.
Changes from the guidance include:
- Refocusing on the critical validation parameters of:
- Specificity/Selectivity
- Range
- Accuracy/Precision
- Test method robustness and sample/reagent stability (formerly, validation criteria) should instead be emphasized and demonstrated during method development on a case -by-case basis or should be made available upon request. Robustness testing should show the reliability of an analytical procedure in response to deliberate variations in analytical procedure parameters as well as the stability of the sample preparations and reagents for the duration of the procedure.
- Stability indicating properties of the test are demonstrated in the specificity/selectivity section by analysis of degraded samples, spiked samples, stress studies, and/or aged products.
- The reportable range of the test method should encompass the upper and lower end of the specification limits (see Table 2, below).
- Analytical test method transfer now requires partial or full revalidation at the receiving site; co-validation can still be used.
- Allows for validation of multivariate analytical procedures (e.g. spectra with many wavelength variables)
- Calibration data are used to create the calibration model
- Test data are then used to estimate the model’s performance
- Model validation uses a set of independent samples to validate the model
- For example, on a spectral ID method (FTIR) a well characterized reference standard is used to establish a calibration spectrum. During validation test samples of both the reference material, a representative sample of the material, and one or more materials that differ from the analyte are analyzed to demonstrate the discriminative ability of the library model and changes in the spectrum.
FDA’s Focus on Critical Validation Parameters
Selectivity/Specificity
Tests should show an absence of interference from other substances and be specific to the target analyte. A lack of specificity can be compensated in some cases by utilizing a combination of two or more procedures. For example, denaturing gel electrophoresis may separate a protein monomer from a covalently linked dimer. However, non-covalently linked aggregates may need a secondary assay to quantitate. In some cases where the analytical technology is specific (NMR, MS, etc.), additional experimental studies may not be required if justified.
Required for: ID tests, Assay, Purity, and Impurity tests
Range
The range of the assays must cover both the upper end and lower end of the specification limits and typically represent the highest and lowest sample concentrations where the procedure produced reliable results. For linear responses, the guidance is unchanged with predefined correlation coefficients, y-intercept, and slope of the regression lines still being required.
One significant change to the new guidance document is the incorporation of nonlinear responses. In these cases, a model or function describes the activity/concentration present and correlates it to the response. For example, many immunoassays show S-shaped responses rather than a linear response.
If the measure of the analyte is close to the lower range limit of the procedure (e.g., Impurities) the limit of quantitation and limit of detection should be established.
Required for: ID tests, Assay, Purity, and Impurity (both limit and quantitative) tests
Accuracy and Precision
Accuracy and precision of the assay can be evaluated independently or in a single study. Accuracy should be established across the range for the analytical procedure. Recovery studies of a known quantity of the analyte in the sample matrix are typically used. Triplicate samples at three different concentrations are analyzed at covering the range of the test method.
One change to the accuracy validation criteria is for multivariate analytical procedures. The test method should be evaluated for the root mean square error of prediction (RMSEP). If RMSEP is found to be comparable to acceptable root mean square error of calibration, then this indicates that the model is sufficiently accurate when tested with an independent test set. Qualitative applications such as classification, misclassification rate, or positive prediction rate can be used to characterize accuracy.
Precision, with its evaluation of repeatability, intermediate precision, and reproducibility (if greater than one laboratory) are primarily unchanged. Multivariate analysis precision is evaluated with the routine metrics of RMSEP above.
Required for: ID tests, Assay, Purity, and Impurity (quantitative) tests
Where ProPharma's Regulatory Sciences Experts Can Assist
ProPharma’s team of chemistry, manufacturing, and controls (CMC) experts have over 50 years of cumulative experience and can assist our clients in both reviewing the content of their analytical test method validations per the new expectations as well as authoring the relevant FDA submissions. Our reviews are extremely thorough, going in-depth to make sure that the conclusions listed in the report are evidence-based, data-driven, and will hold up to scientific and regulatory scrutiny.
Interested in learning how we can help with your regulatory and CMC-related regulatory needs? Our team of expert regulatory consultants is here to help.
