Bracketing and matrixing are reduced study designs for stability testing. Bracketing examines only the extreme levels of a design factor, while matrixing tests only a pre-selected subset of all possible combinations at each test point; both approaches must nevertheless enable a robust conclusion regarding the retest period or shelf-life.
Bracketing: Testing Justified Extreme Cases
Bracketing assumes that the stability of untested intermediate levels is represented by the tested boundary conditions. For example, for a drug product with three strengths, the lowest and highest strengths can be tested if the composition of the strengths is identical or very similar, and the selected endpoints represent the expected worst case.
The selection of brackets does not automatically follow the numerically smallest and largest strength. Different tablet weights, active substance to excipient ratios, container sizes, or fill volumes can alter the stability-relevant extreme case. Therefore, it is assessed in advance whether degradation, moisture sensitivity, or contact surface could actually lie outside the observed boundary conditions for an intermediate variant.
Matrixing: Distribution of Test Combinations
A full stability design tests every combination of factors at every time point. In matrixing, combinations are distributed across time points so that not all strengths, batches, or container sizes are tested simultaneously. Over the entire study, a pattern emerges that covers the relevant combinations without requiring the full sample size at every single time point.
The matrix must not obscure critical transitions or early signs of degradation. Its planning therefore considers test points, expected stability profiles, number of batches, and the conclusion to be derived from the data. If results show an unexpected trend, an expansion to the full testing scope may be necessary.
The stability units must be clearly defined before reduction. Factors include active substance strength, container size, fill volume, batch, or packaging configuration; however, not every variation is a suitable matrix factor. For example, a bottle with a different headspace size may have a different oxidation risk and should not be treated as equivalent solely due to a similar label.
The matrix plan documents for each time point which combinations are tested and which combinations are deliberately deferred. This makes data collection gaps visible and allows for the evaluation of unusual results against the planned coverage. A plan that omits all early time points for a critical strength would be unsuitable for potentially rapid stability loss.
The evaluation remains tied to the objective. If a retest period is to be extended, the reduced data set must genuinely support the extrapolation or trend assessment. A reduced design is therefore an expert selection decision, not a blanket budget tool.
Demarcation and Limits of Reduced Designs
Bracketing reduces the number of factor levels tested, while matrixing reduces the number of combinations tested at a single time point. The procedures are thus not interchangeable but can be combined in a study if their respective assumptions are separately justified. Both are not instruments to replace unsuitable stability data with fewer measurements.
The reduced design must be documented before the start of the study and consider the risk of a shorter retest period or shelf-life, which might differ with a full design. The material comparability in bracketing is particularly important: large differences in composition or packaging argue against the assumption that one extreme level reliably covers another variant.
The number of batches can also be important for the significance of the results. If batches from different manufacturing periods are included, it is easier to assess whether the reduced design only reflects a single favorable batch.
Significance for Clinical Studies
In clinical programs, multiple strengths, package sizes, and batches may arise in parallel, even though sample material and analytical capacity are limited. A justified reduced design can focus stability planning on the truly critical variants without unnecessarily jeopardizing the study’s labeling and resupply deadlines. However, the justification must already align with the CMC concept for the respective investigational medicinal product configuration.
Full-service CROs like Mediconomics support the coordination of stability protocols, clinical supply plans, and packaging strategies, as well as the consolidation of data for regulatory CMC documentation. They can clarify which strength, batch, and package are covered by bracketing and which samples must be available according to the matrix plan at what time.
Frequently Asked Questions (FAQ)
Are all intermediate strengths never tested in bracketing?
The basic assumption is that the selected extreme strengths cover the intermediate strengths. Whether and when additional intermediate strengths are tested depends on product knowledge, the protocol, and any unusual results.
Does matrixing only save analyses?
It primarily saves tests at individual time points but requires careful statistical and expert planning. The effort thus partly shifts to justification and evaluation.
When must a reduced design be expanded?
An expansion is appropriate if results no longer support the underlying assumptions or if the observed trend is insufficient for deriving the retest period or shelf-life.
Regulatory References
- ICH Q1D “Bracketing and Matrixing Designs for Stability Testing” — explains prerequisites and examples of reduced stability designs.
- ICH Q1A(R2) “Stability Testing of New Drug Substances and Products” — integrates bracketing and matrixing into formal stability studies.
- ICH Q10 “Pharmaceutical Quality System” — describes the management of product knowledge throughout the lifecycle.