BioPharm

www.biopharminternational.com October 2018 BioPharm International eBook 25 largest range of means has the most impact on method performance. While range analysis can be used to iden- tify the optimal value of different factors, this method can't be used to determine the significance of different factors. ANOVA, using the F-test, is then employed for more quantitative analysis. The F value for each factor is indicative of the ratio of the variance for each factor to that of the experimental error. The percentage con- tribution of each factor, reflecting the factor's influ- ence, is the percentage of the sum of square deviation due to that factor in the total sum of square deviation. Additionally, the determination of relationships among variables is typically performed by regression analysis. Once the method design space has been established, confirmatory studies should be conducted. The method should contain appropriate system suitability requirements to monitor and ensure con- sistent system functionality. Additionally, the method should include assay acceptance criteria in conjunc- tion with the analysis of a suitable product reference standard or assay control. The chosen criteria should be in-line with the intended purpose of the method. Results from the analysis of the reference standard/ assay control should be plotted and trended to moni- tor method performance over time. There should be a well-defined plan in place on how to address any assay drift; similarly, a strategy should be available for tran- sitioning from an old lot of reference standard/assay control to a new one. METHOD QUALIFICATION After completing the development and depending on the stage of clinical development, it is important to perform either a method qualification, recom- mended for methods used for analysis of products in pre-investigational new drug application (IND) evalu- ation up to Phase I, or a method validation, for meth- ods used to support products in Phase II and beyond. Validation work should be performed in accordance to ICH Q2(R1) (4) and the draft guidance issued by FDA (5). Briefly, when validating an analytical method, it is good practice to minimally ensure that: • All systems used in the execution of the validation study be qualified or calibrated, as appropriate. • Analysts are fully trained on the method. • Method validation requirements are clearly defined. • Representative materials are used during the valida- tion. • All method validation tests are conducted by fol- lowing defined protocols. • All prospectively set validation acceptance criteria are met. Development of solid, reliable analytical methods is not only important for the measurement of particular product attributes, it also has a significant impact on defining product specifications. When establishing product specifications, the acceptable range is based on the combination of product variability and variability of the testing methods, as defined by Equation 1. Product Specification = X × √ (σ 2 an + σ 2 pr ) [Eq. 1] Where X ranges between 2 and 3, in most cases σ 2 pr = Lot to lot product variability resulting from manufacturing process variability σ 2 an = Analytical variability σ 2 an = σ 2 rep. + σ 2 int. [Eq. 2] σ 2 rep. = Intra-run variability; σ 2 int. = Inter-run variability Poorly performing analytical methods not only increases the risk of method dependent out-of- specification results but will also lead to wider product specifications, due to a disproportionate contribution of analytical variability (Equation 1). A wider specification range heightens the potential for allowing non-conform- ing product lots to meet specification limits and thus be accepted for release when such lots should be rejected. To summarize, the benef its of using QbD are significant: • Fewer analytical method-related out-of-specifica- tion and failure investigations • Lower failure rates in method transfer • Method changes can occur without re-validation, if continuing to operate within the defined design space • New technologies can be more easily introduced into existing methods when developed through QbD • Ease of adaptability to continuous improvement. It is clear that QbD is the right approach when developing/validating analytical methods for use in testing pharmaceutical/biopharmaceutical products. REFERENCES 1. ICH, Q8(R2), Pharmaceutical Development (ICH, 2009). 2. ICH, Q9, Quality Risk Management (ICH, 2005). 3. C. Ye, et al., J. Pharmaceut. Biomed. 23, 581-589 (2000). 4. ICH, Q2(R1), Validation of Analytical Procedures: Text and Methodology (ICH, 2005). 5. FDA, Analytical Procedures and Methods Validation for Drugs and Biologics (FDA, 2015). BP Biopharma Laboratory Best Practices Analytical Methods

• Cover