ANALYTICAL METHOD VALIDATION AND POSTVALIDATION
Acceptance criteria must be set for validation in accordance with the ICH Q2 guideline, preferentially as a deliverable of
method qualification. Therefore, all information gathered during method development and qualification is crucial for assessing
validation readiness and establishing acceptance criteria in the validation protocol in accordance with process capability
and product profile (see Figure 1). This compilation exercise is important in verifying that the method is ready to validate to avoid the burden of validation
Figure 1: Points to consider for performing validation readiness assessment.
Once the method is ready to validate, it is strongly recommended that the ICH Q2 referential for analytical method validation
is used (see Table III). The analytical validation exercise should ideally occur before pivotal studies and after clinical proof-of-concept is established
for the candidate.
Table III: Analytical method validation factsheet.
Although good validation practices are described in ICH Q2, this document does not detail the practical implications for validation;
for example, only a few specifics are included regarding experimental design and statistical data treatment. A clear policy
is required for cGMP compliance in data acquisition and treatment, which includes developing good statistical practices. Different
guidelines from the US Pharmacopeial Convention such as USP <1010> or <1033>, or from the industry such as SFSTP's (Société Française des Science et Techniques Pharmaceutiques or French
Society of Pharmaceutical Science and Technology) total error describe good approaches on how to use inferential statistics
and statistical intervals as well as present graphical results (e.g., using accuracy profiles) (10). Potential language gaps
with the ICH, namely in the ISO definition of trueness versus accuracy, have been appropriately described to avoid misinterpretations
of the validation reports. The experimental design should address all parameters from ICH Q2 following method categorization
(i.e., identity, limit or quantitative impurity, potency, or active moiety). The minimal number of runs for studying accuracy
and precision is best defined based on statistical t-test considerations from initial performance assessment (intermediate
precision σ2) and acceptance criteria (σ) (11, 12):
These strategies meet regulatory expectations in terms of risk management of making type I/II errors as well as helping the
sponsor to understand the risk-benefit of extensive experimental designs used in method validation.
A validation report is issued after the completion of the experimental plan where results are compared to acceptance criteria
set in the protocol. Any nonconformity towards acceptance criteria has to be properly captured in the quality system and thoroughly
investigated, preferentially using the laboratory policy for out-of-specification (OOS) investigation as background. It is
intended that no broadening of acceptance criteria be decided at this stage and that a validation failure recovery plan be
established. The recovery plan is typically composed of method (re)improvement and validation amendment(s). These undesirable
events are, however, best prevented with sufficient prior method qualification level and adequate validation readiness assessment.
Finally, method validation cannot be seen as a discrete activity. The regulatory expectation is that the project sponsor has
its own policy on postvalidation activities including method transfer and maintenance, historical trending of analytical capability,
and risk assessment of changes carried out in validated methods. Good statistical practices should ensure that postvalidation
activities do not alter the validated status of the method through equivalence demonstration, such as using the two one-sided
t-Test (TOST), and that method performance be continuously monitored using control charts (1, 12). Postvalidation activities
should be appropriately captured in the annual product quality review in accordance to ICH Q7A to provide continuous assurance
that the method remains suitable for its intended use.
THE ANALYTICAL LIFECYCLE ROADMAP
With all the requirements identified and understood, a comprehensive analytical lifecycle roadmap is incorporated in the project
sponsor's policy that is capable of managing the practical implications of the project (see Table IV) and staging these events across the development plan (see Figure 2).
Table IV: Practical implications of analytical method lifecycle.
Practical implications related to each step of the analytical lifecycle are then translated into defined analytical packages
with regulatory-compliant deliverables staged throughout the clinical strategy (see Figure 2). These analytical packages can be used for driving the project in terms of budget and resource allocation from a phase-dependent
perspective and act as yes-no decision points with respect to the general project roadmap.
Figure 2: The analytical lifecycle roadmap.
In conclusion, it is incumbent of the project sponsor to build a comprehensive roadmap that would drive the project through
the different stages of clinical development in a manner that fits the economic realities of the business of developing new
biologic drug candidates without compromising on regulatory compliance.