By Brian Glass, senior analytical consultant, Pharmatech Associates
The International Council for Harmonization’s (ICH) new draft guidance (Q14) for Analytical Procedure Development describes an enhanced approach for method development, validation, and life cycle management. ICH Q2 was also updated to reflect the proposal in Q14: both are now available for public consideration and feedback, and both draft guidelines agree with the proposed USP <1220> chapter on analytical life cycle management. What’s new is that the enhanced approach described in Q14 employs elements of quality by design (QbD) and risk management tools and applies these to analytical procedures. While the enhanced approach is not mandatory, there are benefits to the methodology.
Traditional Vs. Enhanced Approaches To Analytical Development
The traditional approach taken for analytical development is to develop a procedure that may work, followed by validation of the method without a thorough understanding of all factors that influence the method's performance. Once validated, the method remains unchanged until issues inevitably arise, and the method is modified and revalidated.
ICH Q8, Q9, and Q10 guidelines describe a paradigm shift from traditional manufacturing and quality systems management to an approach primarily focused on understanding the process, maintaining a state of control, and continuous improvement to build quality into the product from the beginning. With the new draft guidance, ICH Q14 applies this enhanced approach to analytical procedure development and validation. It requires some adjustment to the validation landscape. The ICH Guidance draft update for validating analytical procedures (ICH Q2) reflects the validation activities necessary to implement the enhanced approach.
The traditional approach remains a valid model, but the enhanced approach has benefits. From a scientific viewpoint, the primary advantage is gaining a deeper insight into critical method properties, providing more regulatory flexibility when changes are necessary. The enhanced approach allows the development and validation of an analytical control space that permits the adjustment of some variables (over a range of values) if necessary. A more thorough understanding of these critical parameters provides a more robust methodology.
Analytical Target Profile Development
The first step is the development of the analytical target profile (ATP). The ATP is analogous to the quality target product profile (QTPP) described in ICH Q8. This document defines the goal of the development and the desired performance characteristics of the procedure along with its associated uncertainty. Once the ATP has been developed, the control strategy links critical quality attributes (CQAs) to critical process parameters (CPPs). The ATP is used to guide the development of the procedure and is the first step in assuring that the final method will be fit for its intended purpose. Knowledge management is a critical aspect of the method development process — knowledge regarding the physical and chemical properties of the drug substance and the drug product is necessary. Additional knowledge will also be gained during the development process and should be preserved. Risk management is assessed at all stages of the procedure’s life cycle to ensure that the method is in control and meets the ATP criteria. Risk assessments help determine the required control and replication strategies.
Robustness And The Enhanced Approach
The enhanced approach discussed in Q14 revolves around the complete understanding of the process and the parameters used to produce the reportable results. The thrust of the robustness study is to develop this understanding. Properly designed, the robustness study provides information regarding how adjustments to individual variables affect the result and how variables interact with other variables. In the traditional approach, robustness is usually determined by varying individual factors one at a time (OFAT) and evaluating the impact of the change.
This approach is still valid, but crucial variable/variable interaction information is lost.
A chemometric approach using design of experiments (DoE) and statistical analysis has the potential to identify critical variables that impact the method and variable/variable interactions. Robustness is approached differently for every technique, and a strategy is needed to maximize the amount of information obtained while minimizing the experimental activities.
For reversed phase HPLC screening strategies, select variables that have the most significant effect on the alpha term of the general resolution equation, since these have the greatest impact on the resolution. Factors affecting selectivity (alpha) are column type/chemistry, mobile phase (pH, buffers, ion-pair reagent, etc.), organic solvent (MeOH, ACN, etc.), and gradient slope or initial buffer/solvent concentration.
After screening, the method is optimized by manipulating factors that primarily affect the efficiency term (N) results in a set of conditions that will meet the requirements of the ATP. Based on the DoE data, a prediction model describes how changes in method parameters affect the method's performance. Statistical techniques such as ANOVA (for least-squares multiple regression models) verify the model's predictive ability and provide data for justification and validation of the method operational design region (MODR). Adjustments to the method within the MODR do not require regulatory approval before implementation.
Validation And Life Cycle Management
The revision of ICH Q2 allows for use of the enhanced approach and describes the validation activities required. Validation is the last step in the traditional approach and is often performed in the same manner for every method and treated as a formality. When justified, development data can be used to supplement the validation report resulting in less validation activities. The analytical life cycle management approach focuses primarily on the method's bias (accuracy) and precision as part of validation to verify conformance to the ATP. During routine use and monitoring of the procedure, additional data becomes available that is used to assess the performance of the method and compliance with the ATP. Monitoring the method performance is also critical in the life cycle management process.
The new guidelines also discuss multivariate procedures (MVP) and real-time release testing (RTRT). These are important aspects of continuous pharmaceutical manufacturing involving process analytical technology (PAT), but PAT can apply to traditional manufacturing (batch processing). PAT sensors measure a CQA's physical and chemical properties during the manufacturing process. Input from multiple sensors coupled with a multivariate model can also predict CQAs that may not be directly measurable.
The Q14 guideline and the Q2 revision discuss model calibration, verification, and validation steps. Models may be constructed and calibrated using verified samples and a validated reference method, but model validation is always determined using an independent data set. The model's performance is evaluated during use and often requires updating and revalidation, which again feeds into the life cycle approach for change management. In some cases, the data generated can alleviate the requirement for post-processing release testing.
Sometimes, changes to established procedures or models are necessary to reflect the knowledge gathered over time. These changes can occur at any stage in the life cycle. In the traditional approach, changes to the analytical methodology are reported to the regulatory authorities before implementation. Depending on the severity of the change, the method may require full or partial revalidation followed by approval before implementing the change. An approved post-approval change management protocol (PACMP) and product life cycle change management plan (PCLM) ensure that possible changes will be acceptable. This principle reflects the ICH Q12 guidance regarding the path to changes to the CMC section of marketing applications.
Analytical data is extensively used in marketing applications to demonstrate the safety and quality of pharmaceutical products. Since decisions regarding the acceptability of a batch are only as good as the data provided, well-developed analytical methodology replete with a knowledge of the total error involved with the measurement reduces the risk of making a wrong decision. Armed with the new guidelines, manufacturers reduce the risk of releasing a batch that does not conform to specifications.
About The Author:
Brian Glass, senior analytical consultant at Pharmatech Associates, has over 30 years’ experience in the fields of analytical research and development, quality control, validation, technical transfer services, and process development. He has held senior analytical positions in private startups, established pharmaceutical manufacturing facilities, and large CMO/CDMO organizations. His career spans all facets of the drug development process for small- and large-molecule therapeutics, from preclinical to commercial products, utilizing different modes of drug delivery systems (solid dosage, injectable, inhalants, oral solutions/suspensions, and soft gel capsules). He holds a B.S. in zoology from Louisiana Tech University.