Size the BE study before you run it

What is at stake

A failed BE study costs a development cycle. An oversized one costs money and volunteers. Simulating the study first shows the probability of success for each design and sample size.

Question of interest

Will the test product be bioequivalent to the reference, and what study design and size will show it?

Population PK or PBPK models of the reference and test products, informed by in vitro dissolution where relevant, simulate BE trials across designs and sample sizes. Used for design, model risk is low to medium. Used to support a waiver or an alternative BE approach, it is high, and the approach is discussed with the agency first.

What reviewers will ask

Is the within-subject variability realistic?

It comes from published or pilot data on the reference product, and we report how sensitive power is to it.

Is the in vitro to in vivo link validated?

When dissolution informs absorption, the link is checked against observed data before it is used to predict.

Is this a design aid or BE evidence?

The context of use says which. Using a model as BE evidence is discussed with the agency first.

What you file with the agency

BE study design rationale and sample-size justification
Modeling report and M15 assessment table
Where relevant, the model-based approach proposed for agency discussion

What we need from you

Formulation and in vitro dissolution data for test and reference
Any pilot PK data
The relevant product-specific guidance, if one exists

Confidence intervals of the test to reference geometric mean ratio across simulated crossover trials, compared with the 0.80 to 1.25 limits. — Virtual BE trials.

What it may replace

It can prevent an under- or over-sized BE study, and in some cases support discussion of an alternative BE approach. Whether model-integrated evidence can reduce a study is decided case by case with the agency.

Scientific approach

Reference model. Built from published data and labelling, or from your pilot data, with between- and within-subject variability that matches the observed spread.

Test product. Absorption informed by dissolution through an in vitro-in vivo relationship or a PBPK absorption model, when the data support it.

Trial simulation. Hundreds of virtual trials per design and sample size, with the standard average-BE analysis applied to each.

Output. Probability of concluding BE by design and N, and the drivers of failure, so the study is sized to evidence rather than convention.

Relevant guidance: ICH M13A on bioequivalence for immediate-release solid oral dosage forms; FDA product-specific guidances where they exist; ICH M15.

Is this the decision in front of you?

Fixed price, paid in two milestones. A guaranteed date. Agency questions on this analysis answered at no extra cost. Pricing · How we work

Scope this package