OR WAIT null SECS
Thomas A. Little, PhD, is president, Thomas A. Little Consulting and BioAssay Sciences, 12401 North Wildflower Lane, Highland, Utah 84003, USA, email@example.com.
A 10-step systematic approach to analytical method development and validation can improve the quality of drug development.
Fundamental to all aspects of drug development and manufacturing are the analytical methods. Analytical methods require development, validation, and controls just as all other product and process development activities. Measurement of API characteristics, the factors that influence them, and key impurities are at the heart of product development for efficacy and safety. This article discusses a systematic approach for analytical method development and validation that was developed in line with the International Conference of Harmonization (ICH) Q2 (R1) Validation of Analytical Procedures: Text and Methodology, Q8 (R2) Pharmaceutical Development, and Q9 Quality Risk Management (1-3).
Historically, insufficient attention has been paid to assay development, how it impacts the product, ongoing release testing, and product control. Simple coefficient of variation (CV) calculations for assay precision are a necessary but insufficient measure of assay efficiency and may be misleading because CV has no relationship to product acceptance and release testing limits.
Figure 1: Measurement process elements. FIGURES ARE COURTESY OF AUTHORS.
Assays and measurement systems must be viewed as a process. The measurement process (see Figure 1) is made up of methods, software, materials and chemistry, analysts, sample preparation, environmental conditions, and instrumentation/equipment. Quality risk management techniques and statistical data analysis techniques should be used to examine the process of measurement and identify factors that may influence precision, accuracy, linearity, signal to noise, limits of detection and quantification, or any other assay attributes to achieve optimal assay results (see Figure 2).
Figure 2: Target product profile (TPP), critical quality attributes (CQAs), and associated analytical methods
Based on ICH guidelines and the author's experience, there are 10 steps to analytical development and method validation.
The purpose of any analytical method should be clear. Will it be used for release testing or for product/process characterization? What are the target product profile parameters (ICH Q8[R2]) and critical quality attributes (CQAs) that the analytical method is associated with? Are there any CQAs that have no clearly defined measurement method? What impurities need to be measured, and what is the risk of not measuring them? Is the assay correlated with other analytical methods? How orthogonal is each assay compared to other assays used to evaluate the product. How does the assay minimize or influence risk during drug development and manufacturing?
Select the method used
There are many analytical methods across the industry, and the method used should have appropriate selectivity and high validity. Valid analytical methods measure the condition of interest. It is possible to have good precision with poor measurement validity. It is possible, for example, to measure the quantity of a protein without knowing how active the protein is. Measures of activity and measures of quantity need to be accurately considered and balanced against other objective measures of the product.
Identify the method steps
Lay out the flow used in the analytical method by using Visio or a similar process mapping software to visualize the sequence used in performing the assay. This layout will be used for development, documentation, risk assessment, and training. All steps should be listed and detailed regarding the flow and use of plates, materials, and chemistry. Identify steps in the process that may influence bias or precision.
Determine product specification limits
The specification limits used to control the release should be determined for the analytical methods used for release testing. Limits may be set using historical data and industry standards, based on statistical k sigma limits, and/or tolerence levels or transfer function. Limits need to reflect the risk to the patient, CQA assurance, and control the flow of materials in the production of the drug substance and drug product.
Perform a risk assessment
The analytical method risk assessment (see Figure 3) is used to identify areas and steps in the analytical method that may influence precision, accuracy, linearity, selectivity, and signal to noise. Specifically, the risk question to be asked is, "Where do we need characterization and development for this assay?"
Figure 3: Analytical method risk management example.
Failure mode effects analysis (FMEA) and or other risk assessment methods may be used when performing a risk assessment. In addition to the traditional FMEA approach, severity, probability and detectability, the influence on CQA and uncertainty to the risk ranking should be added. Specific questions of what may influence precision or what may influence bias or accuracy need to be examined. Each step in the analytical method should be looked at from this point of view.
Characterize the method
The development/characterization plan for the assay should be defined based on the risk assessment. Determination of sample size and sampling method are key considerations. Assay development can be broken into three steps: system design, parameter design, and tolerance design. System design involves ensuring that the right chemistry, right materials, right technology, and right equipment are being used. Parameter design is typically done by running design of experiments (DOEs) and making sure that the right parameters are selected at their optimal design set point. Characterization of the design space for precision and accuracy is a key assay development outcome. Finally, the allowable variation for key steps in the assay must be defined to assure a consistent outcome. Partition of variation (POV) analysis is recommended to further breakdown precision variation into its influencing factors (4). Plate variation, for example, must be considered when developing analytical methods. Failure to understand plate variation and other sources of assay error will directly mix into the total variation and will be linearly added to product variation and increase limits of quantitation and detection, effectively reducing the assasy range and adding to out-of-specifications rates for product acceptance testing.
Complete method validation and transfer
The method validation requirements should be defined. There are many measures (e.g., amount of API, activity of API, and impurities) of measurement performance that may be used in method validation (see Figure 4). Make sure there is a clear identification of the requirements for each method when organizing the validation plan. Figures 4, 5, and 6 are adapted from Q2 (R1) and identify the requirements to complete the method validation.
Figure 4: Method validation list.
Representative drug substance (DS) and drug product (DP) materials should be in place during validation. Representative materials and standards will assure the limits of detection and quantitation have been correctly calculated and validated and will perform well when measuring and testing actual product.
Figure 5: Method validation quick reference guide.
All method validation tests should be conducted using the correct sample size and sampling method as defined in the method standard operating procedure (SOP). Acceptable results for method validation of all analytical methods should be achieved. Make sure acceptance criteria have been defined for each validation method variable. Aspects of the assay should be modified so that it can pass the validation testing criteria. Finally, it is necessary to determine whether the analytical method is fit for use and ready to transfer to other internal organizations or to external CRO/CMOs. This is determined by meeting all acceptance criteria for precision, bias, and linearity. Equivalence tests are typically used for method transfer.
Figure 6: What is required for method validation and when to use it (Ref. 1).
Define the control strategy
A clear control strategy needs to be put in place once the assay has been developed and validated (5). The following questions should be asked, "What materials will be used for control or reference materials?" "How do you know the standards are stable?" "What will be used for tracking and trending the assay so the true assay/plate variation is known over time?" "What will be used to adjust/correct the assay once drift is detected?" "How will one set of reference materials be transferred to another?"
Train all analysts
Train all analysts using the validated analytical method. If there are concerns that the analyst may have an effect on the results of the analytical method, each analyst should run qualification tests using known reference standards to qualify and certify the analyst on the method. Analyst method errors may include sample selection, sample prep, weighing, mising, diluting, concentrating, location of peak, injection method, and time method.
Impact of the analytical method
Total variation can be expressed in the following equation:
Standard Deviation Total = SQRT (Product Variance + Assay Variance)
Figure 7: Example of accuracy to precision modeling.
As the assay error rises, the total standard deviation also rises. Using the accuracy to precision (ATP) model (see Figure 7), it is possible to visualize the relationship of precision and accuracy on product acceptance rates. The ATP model shows how changes in precision and accuracy impact product acceptance rates and the assay error design space. CV calculation is a good measure of assay error; however, it is not scaled to the acceptance limits, it is scaled to the mean. Rescaling the variation to the release limits helps to clarify if the variation in the assay is fit for use. The number 5.15 is used in the equation to represent 99% of the assay error. Generally, a percent of tolerance of less than 20% is considered an acceptable result; more than 20% will result in a high level of out-of-specification release failures and should be considered for further development:
% Tolerance Measurement Error= (Standard Deviation Measurement Error*5.15)/(USL-LSL)
Where USL is upper specification limit and LSL is lower specification limit. The attention paid to method development, validation, and control will improve the quality of drug development, patient safety, and predictable, consistent outcomes.
Thomas Little, PhD, is president of Thomas A. Little Consulting in Highland, UT. firstname.lastname@example.org.
1. ICH Q2(R1) Validation of Analytical Procedures: Text and Methodology, (2005).
2. ICH, Q8(R2) Pharmaceutical Development (2009).
3.ICH Q9 Quality Risk Management (2006).
4. Little, T.A. Engineering Statistics and Data Analysis-BPM (2012).
5. FDA, PAT—A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance (2004).