The Murky Waters of Method Validation
A field seeks clarity on validation in preclinical studies
When drug developers set out to create a generic or a biosimilar they have to prove to regulators that the product is comparable to the reference drug—what is generally referred to as bioequivalence (BE). Bioequivalent drugs must contain the same active ingredient or ingredients, the same dose, the same route of administration and the same strength or concentration as its predecessor.
This is easier said than done. Biologic-based drugs, which are produced in living systems or derived from biologic material produced in living systems, can be particularly tricky to characterize. They are larger than synthetic drugs and have activities that are dependent on specific conformations. The choice of cell line, production process, purification process and the final pharmaceutical formulation all influence the biologic.
The US Food and Drug Administration (FDA) advises an approach that includes "comparisons of structure, function, animal toxicity, human pharmacokinetics/pharmacodynamics, immunogenicity, and safety/effectiveness between the proposed biosimilar and the reference product." Such an approach, the agency says, integrates multiple measures to develop "fingerprints" that can be used to characterize complex products.
Demonstrating bioequivalency of complex biological products requires a multi-stepped approach using analytical assays to demonstrate physiochemical comparability typically followed by preclinical and clinical studies to demonstrate biosimilarity. And because manufacturers never fully disclose exactly how they make their biologics, manufacturing a biosimilar is a bit like trying to bake Julia Child's three-chocolate bombe cake without having the entire recipe or even knowing the best conditions under which to concoct it. The best thing you can do is assess your work at every juncture to see if it passes muster. This level of supporting information is appropriate for submissions of this type, but is it appropriate for all stages of synthetic small molecule development?
Tiered approaches have been used for decades, but to what level CROs, particularly in the US, are using them in the preclinical arena—particularly early discovery—is difficult to say. It's also unclear to what degree regulatory agencies, such as the FDA and European Medicines Agency, are having to adjust their audit criteria to recognize and apply tiered requirements and whether there needs to be a anything other than gold standard for GLP regulated bioanalysis (as outlined in the validation guidelines).
The Global CRO Council (GCC) held an April 14 forum during the 9th Workshop on Recent Issues in Bioanalysis (WRIB) in Miami that touched on a number of the pressing questions facing CROs about the status of method validations in preclinical studies. The discussion was moderated by three leaders in bioanalysis: Elizabeth Groeber, Director of Bioanalytical Chemistry at Charles River's Preclinical Services site in Ohio, Colin Pattison, Manager of Bioanalysis and Immunology at Charles River's Preclinical Services site in Edinburgh, and Richard LeLacheur, Senior Director of Bioanalytical Operations at Agilux Labs in Boston. Eureka Senior Scientific Writer Regina McEnery linked up with Colin following the meeting to talk more about what the forum covered and what they learned from other CROs.
Q: Generally speaking, when are tiered approaches most appropriate in preclinical method validation and why are they necessary?
A: If your study is powered to assess a primary endpoint—whether it's pharmacokinetic (PK) data or a specific biomarker—a tiered approach is almost completely ruled out because you need to be 100% confident in the data that you are generating. For secondary endpoints, such as metabolites used to elucidate clearance pathways (not active) or tissue analysis to give information on distribution throughout the model… which can be additional events of interest, do not carry the same statistical power and do not need to be fully validated, a tiered approach is more relevant. Generally speaking, tiered approaches, which are quicker and cheaper, are used to gather additional information that helps a sponsor justify whether or not to move the compound forward.
Q: What's does the industry say about the use of tiered approaches?
A: The European Bioanalysis Forum (EBF) has developed a scheme describing the use of an analytical method with varying levels of validation. In cases of known or expected toxicity in the metabolite a validation in preclinical following through to clinical is recommended. In cases where the activity is unknown less validated screening and qualified methods may be more appropriate until toxicity of the metabolite becomes known. While this can be a useful tool for industry it provides some challenges for regulators—the FDA has come out repeatedly with a position on this exact discussion—who would more likely differentiate the stage of development and purpose of the study (e.g. Phase I vs. later development studies) for the requirement of greater confidence and subsequently validation. Understanding the activity of the metabolite (irrespective of the 10% rule) will also play a part in this decision making and it may only be later in the development program, perhaps in target population, that this may be fully available. This disparity was highlighted in a survey conducted by the GCC, where one of the respondents noted that they felt comfortable using a tiered method rather than a fully-validated one for one component, only to have an inspector challenge whether the analyte should be treated as a primary endpoint.
Q: What are some typical examples of tiered approaches in early and late-stage drug development?
A: One example early on would be the use of qualified (not fully validated) methods to support GLP large animal telemetry studies. For instance, suppose you have a client who wants to use suitable telemetry behavior to assess the progress of all of its drugs. They might choose to conduct their safety pharmacology studies before the pivotal toxicology studies using qualified methods for quantifying the compound. Another example, further downstream, would be a clinical study designed to identify where a compound has gone due to lower exposure than expected. There are a lot of possible metabolites which have had qualified assays set up to quickly determine if they are present in significant concentrations. If they are and it becomes critical to the program, one would expect to go back and fully validate methodology for future work.
Q: Do CROs generally favor tiered approaches and how many actually use them?
A: It's hard to say industry-wide, but we did pose that question before our meeting to about 30 CROs. Forty-eight percent responded saying they have used a tiered approach either being instructed by or promoting the use to clients. This was less than I was expecting but the balance of the group was weighted towards clinical bioanalysis suppliers, where this may be much less prevalent (not at all in BE) whereas with our extensive work also in the preclinical arena it is something we are much more comfortable implementing. Many CROs have used the tiered approach under a variety of names, and scientific rigor is necessary to ensure all parties are happy with this approach based on the planned use of any data generated.
How to cite:
McEnery, Regina. The Murky Waters of Method Validation. Eureka blog. April 29, 2015. Available: https://eureka.criver.com/the-murky-waters-of-method-validation/