Potency is a measure of the ability of a product to confer an effect. As such, potency is considered a critical quality attribute (CQA) to be measured, controlled and monitored as part of a quality target product profile (QTPP).
In vitro methodologies for assessing potency typically employ biological systems such as cells or tissues that partially or completely mimic the mode of action (MoA), or in some cases mechanism of action, leading to a measurable effect when treated with the drug of interest. Biological systems and the potential variables impacting outcomes are complex as compared to typical analytics otherwise employed to measure product CQAs. Care needs to be taken during analytical development to ensure methodologies firstly, reflect the MoA, are also reproducible, accurate, and robust.
Factors potentially impacting reproducibility, accuracy and robustness of a cell-based assay (CBA) might include: cell line, cell bank, cell culture media and supplements, cell passage, growth conditions, confluence, cell density, dose range, dose increment, exposure times, dilution approach, reagent concentration and incubation times, plate layout, edge effect, mixing, cell lysate/supernatant storage time, signal intensity/gain, technique, etc.
Analytical quality by design (AQbD) is a systematic approach to designing quality into test methodology which results in the generation of more robust methodologies, with understood performance characteristics (refer to Figure 1). The first step in AQbD is developing an analytical target profile (ATP). With clear user requirements, a candidate method is selected to identify the most viable analytical strategies to achieve outlined objectives. Having identified chemistries and instrumentation allows for the creation of a desired design space based on known analytical parameters, and known risks related to instrument or chemistry limitations (eg. read reproducibility, incubation times, CO2 ranges, validated temperature controls). This preliminary design space is the first iteration of the method operable design space (MODR) which is further iterated post-development. Finally, having defined the ATP, candidate method, and desired design space, a method development strategy can be defined employing a combination of one-factor-at-a-time (OFAT) and design of experiment approaches (DOE).
While OFAT is generally regarded as less efficient, and rarely discussed as part of AQbD, early proof of concept development of complex multi-stage methodologies often presents many unknowns. When analyzing high effect parameters early on an OFAT approach might be preferred to ensure tight control over all other variables. Once proof of concept is established with basic level of control, DOE approaches offer a variety of advantages over OFAT approaches;
- Cell-based assays are often labor-intensive, time-consuming, and costly to execute, making it impractical to evaluate all factors using an OFAT approach. DOE approaches allow for the consecutive analysis of two or more factors concurrently, reducing time and cost to execute.
- Factor interaction is not measurable using an OFAT approach and are best systematically evaluated using 2-factor or greater DOE designs.
- With proper design and factor combinations, DOE studies often generate additional data sets for core factors, providing increasing precision to affect estimates via larger data sets.
Using the desired MODR, full factorial designs can be employed; however, fractional factorial designs are often more practical. The MODR is then updated to reflect the determined design space. Response optimization includes determination of optimal settings for combinations of bioassay factors to achieve the desired response, often through utilization of center point data generated.
While DOE designs confer certain benefits, the complexity of a cell-based assay and number of parameters can very quickly become difficult or impractical to manage experimentally. Furthermore, while likely manageable statistically, an element of easy-of-understanding might also contribute to practicality of a given approach. As such, an OFAT approach may still be considered a reasonable or potentially preferred approach, at least in part to establish the groundwork and core parameter performance. The OFAT approach can also be used in conjunction with limited parameter DOEs, or of hybrid DOEs for evaluation of low impact parameters to maximize data output while maintaining low data complexity.
The use of AQbD, and DOE approaches in the development of cell-based bioassay provides a strong foundation for downstream activities such as validation and life-cycle management. In the authors opinion, the benefits of an AQbD approach can outweigh the disadvantages; however, there remains a strong argument for traditional OFAT approaches combined with fractional factorial limited parameter DOEs.
References
FDA CFR 21 Part 600 – Biological Products: General
ICH Q8(R2) Pharmaceutical Development.
USP<1032> Design and Development of Biological Assays.