Lun Xin, MSc, Senior Scientist, Drug Product Development
Todd Stone, PhD, Director, Analytical Development
Mourad Mellal, MSc, Director, Statistics
Yunsong Li, PhD, Senior Director, Product Development
- Catalent Biologics
Pharmaceutical Quality by Design (QbD) is a systematic approach to drug product development that begins with predefined objectives, and emphasizes both product and process understanding and control based on sound science and quality risk management.1
The principles of the QbD approach that are applied to analytical method development are termed Analytical QbD (AQbD). The primary objective of analytical method development is to provide an optimized analytical procedure ready for method validation and the evaluation of fi t-for-purpose metrics.2 The Design of Experiment (DoE) approach is used during analytical method development to design three types of studies - optimization, robustness, and ruggedness - each providing unique information that is important for method development, and ultimately, method performance. This approach is not limited to a specific modality and can be used to develop a variety of methods.
Analytical Quality by Design
The AQbD methodology is well-suited to help ensure the development of high-quality analytical methods in the shortest time, using the fewest resources necessary. Often, a sponsor company engages a development and manufacturing partner with expertise using QbD to complete method development. Following technical transfer, a preliminary protocol is established, and a risk analysis is conducted to identify the key risk variables in the method. The variables are broadly grouped into the categories below:
- Experimental: where variables such as column temperature are studied in method optimization and/or robustness DoE studies;
- Control: where variables such as the type of instrumentation are fixed; and
- Noise: where variables such as different lots of critical reagents are studied in ruggedness DoE studies.
Optimization, Robustness, and Ruggedness in Design
The three types of studies applied to analytical method development each have different objectives, and can be set up using full factorial or fractional factorial designs.
Method Optimization
The objective of method optimization is to identify the critical analytical parameters of the method (usually controlled continuous variables), establish set points or targets for those parameters, determine the operating ranges necessary to achieve acceptable performance, and reduce bias while increasing precision.3
The best way to accomplish these goals is by DoE methodology. This includes conducting multi-factorial (multi-test parameter) experiments to optimize the method to give results with good accuracy and precision. It is assumed that set points need to be established as part of the study. At this stage, it would also be worthwhile to consider the economics of test parameters, as learnings from the DoE may help reduce costs associated with the utilization of the test method.
Robustness
The objective of a robustness study is to understand the impact of variation in test parameter settings on the performance of the test method.4
It is assumed with this type of study that the set points or targets of the critical test parameters have already been established. It is of interest to learn how robust the method is to variation around these controlled set points, which are usually continuous variables.
A full factorial robustness DoE would be appropriate with fewer than four factors, while a fractional factorial or Plackett-Burman design is recommended when four or more test parameters are to be studied. Careful consideration should be given to the ranges utilized in the experiment, as the goal is to understand the robustness, not optimization.
Ruggedness
The objective of a ruggedness study is to evaluate uncontrolled, normal variation form typically discrete variables, such as different analysts, laboratories, equipment, reagent lots, and days.5
Ruggedness studies are used to understand the magnitude of variation due to noise factors. This type of study is typically focused on estimating the magnitude of variation for each source (i.e., absolute variances or standard deviations), as opposed to DoE experiments that explore optimization and robustness where averages and factor effects are of primary interest. The factors under study are often referred to in statistics literature as “random factors”. This suggests that the levels chosen to be studied are meant to represent a bigger population of potential levels. For example, three analysts involved in a ruggedness study are meant to represent a population of analysts for whom inferences are to be made.
Ruggedness studies estimate repeatability and reproducibility, obtain a more accurate and precise estimate of intermediate precision, and provide a source of data to help evaluate specifications.
Finally, a ruggedness study is not only used to identify special causes of variation, but also to assess sources of common variation that might be encountered in normal operating conditions.
Summary
The quality of analytical methods is key to the overall quality profile of a pharmaceutical product. In an industry that strives to rapidly attain regulatory approval with the highest product quality possible, the DoE methodology is uniquely suited to building quality into the product and process design. The proposed strategies are intended to quickly assess the readiness of analytical methods, and then assemble the appropriate DoE studies. The lessons learned using DoE methodology can be readily used to support future method transfer, validation, and investigational studies.
The Catalent Biologics analytical development team has extensive experience in AQbD and the application of DoE to method development, including the execution of rigorous statistical design and analyses. Offering one of the broadest ranges of biologics and other large molecule analytical services in the industry under full CGMP compliance, Catalent Biologics analytical services can be leveraged as a standalone service or integrated into development and manufacturing.
To learn more about how our partners use AQbD to help accelerate drugs to market, visit www.catalent.com/biologics/biologics -analytical-services
References
- ICH, Q8(R2), Pharmaceutical Development. 2009
- Burgess, C., et al. Fitness for use: Decision rules and target measurement uncertainty. Pharmacop. Forum, 2016. 42(2)
- Politis, S.N., et al. Design of experiments (DoE) in pharmaceutical development. Drug Dev. Ind. Pharm., 2017. 43(6): p. 889-901
- Burns, D.T., et al. A tutorial discussion of the use of the terms “Robust” and “Rugged” and the associated characteristics of “Robustness” and “Ruggedness” as used in descriptions of analytical procedures. J. Assoc. Publ. Analysts, 2009. 37: p. 40-60
- Borman, P.J., et al. Method ruggedness studies incorporating a risk-based approach: A tutorial. Anal. Chim. Acta, 2011. 703(2): p. 101-113
Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special
offers from American Pharmaceutical Review delivered to your inbox!
Sign up now!