What a Bioequivalence Study Actually Measures

Answer first: rate and extent of absorption, and nothing wider

The definition sits at 21 CFR 320.23(b)(1). Two drug products will be considered bioequivalent drug products if they are pharmaceutical equivalents or pharmaceutical alternatives whose rate and extent of absorption do not show a significant difference when administered at the same molar dose of the active moiety under similar experimental conditions, either single dose or multiple dose.

Every element of that sentence narrows the claim. Rate and extent of absorption — not clinical outcome. The same molar dose — not a comparable dose. Similar experimental conditions — not ordinary use.

The companion definition for bioavailability at 320.23(a)(1) says the same thing about a single product: in vivo bioavailability is measured if the product's rate and extent of absorption, as determined by comparison of measured parameters such as concentration of the active ingredient in blood, urinary excretion rates, or pharmacological effects, do not indicate a significant difference from the reference material's. For products not intended to be absorbed into the bloodstream, bioavailability may be assessed by scientifically valid measurements intended to reflect the rate and extent to which the active ingredient becomes available at the site of action.

There is also a statistical requirement stated plainly: statistical techniques used must be of sufficient sensitivity to detect differences in rate and extent of absorption that are not attributable to subject variability.

A difference in rate can be acceptable, and the regulation says when

This is the part that surprises people, and it appears twice in the same section.

A drug product that differs from the reference material in its rate of absorption, but not in its extent of absorption, may still be considered bioavailable — if the difference in rate is intentional, is appropriately reflected in the labeling, is not essential to the attainment of effective body drug concentrations on chronic use, and is considered medically insignificant for that product.

The bioequivalence paragraph carries the same four conditions. Some pharmaceutical equivalents or alternatives may be equivalent in the extent of their absorption but not in their rate, and yet may be considered bioequivalent, because such differences in rate are intentional, are reflected in the labeling, are not essential to attaining effective concentrations on chronic use, and are considered medically insignificant for the particular product studied.

Read that list carefully and you can see what it is guarding. The difference must be deliberate, disclosed on the label, irrelevant to steady-state concentrations, and judged clinically unimportant. An unexplained difference meets none of those conditions.

Six ways to find out, ranked

Section 320.24 does something unusual for a regulation: it puts its accepted methods in an explicit order. The following approaches, it says, in descending order of accuracy, sensitivity, and reproducibility, are acceptable.

First, an in vivo test in humans in which the concentration of the active ingredient or active moiety, and when appropriate its active metabolites, in whole blood, plasma, serum or other appropriate biological fluid, is measured as a function of time — or an in vitro test that has been correlated with and is predictive of human in vivo bioavailability data.

Second, an in vivo test measuring urinary excretion of the active moiety as a function of time, with measurement intervals as short as possible so the rate of elimination is as accurate as possible. The regulation notes this method is not appropriate where urinary excretion is not a significant mechanism of elimination.

Third, an in vivo test measuring an appropriate acute pharmacological effect as a function of time, if that effect can be measured with sufficient accuracy, sensitivity and reproducibility — applicable only when concentration methods are not available.

Fourth, and here is the striking one: well-controlled clinical trials that establish safety and effectiveness, or appropriately designed comparative clinical trials for demonstrating bioequivalence. The regulation states outright that this approach is the least accurate, sensitive, and reproducible of the general approaches for measuring bioavailability or demonstrating bioequivalence.

Fifth, a currently available in vitro test acceptable to FDA, usually a dissolution rate test, that ensures human in vivo bioavailability. Sixth, any other approach deemed adequate by FDA.

The fourth entry is worth sitting with. For this specific question — did the same amount arrive at the same speed — a clinical trial is the weakest instrument on the list, because a clinical outcome is a noisier reading of absorption than a blood concentration curve is. That is not a general statement about clinical trials. It is a statement about matching an instrument to a question.

What such a study looks like when it is run

Section 320.26 describes the standard single-dose design, and the details show how much of the noise is engineered out.

The basic principles: the study should be a single-dose comparison of the test product and the appropriate reference material conducted in normal adults, and both should be administered in the fasting state unless another approach is more appropriate for valid scientific reasons.

The design should be crossover unless a parallel or other design is more appropriate — meaning the same person receives both products in sequence, so each participant is their own comparison. Between the two administrations there must be a drug elimination period, and the regulation specifies its length: at least three times the half-life of the active ingredient or therapeutic moiety or its metabolites, measured in blood or urine, or at least three times the half-life of decay of the acute pharmacological effect.

Blood sampling must be frequent enough to permit an estimate of both the peak concentration and the total area under the curve for a period at least three times the relevant half-life. In a study comparing oral dosage forms, sampling times should be identical.

A crossover design in fasting normal adults with a washout period is about as tightly controlled as human research gets. That is deliberate: the question is narrow, so the design can afford to be narrow too.

When bioequivalence becomes a live concern

Section 320.33 lists the factors the Commissioner considers, when supported by well-documented evidence, to identify products that are not or may not be bioequivalent. Reading them tells you which product characteristics make this question consequential rather than routine.

Among them: evidence from well-controlled clinical trials or controlled observations in patients that the products do not give comparable therapeutic effects. Evidence from well-controlled bioequivalence studies that they are not bioequivalent. Evidence that the products exhibit a narrow therapeutic ratio, which the regulation illustrates as less than a two-fold difference between median lethal dose and median effective dose values, or less than a two-fold difference between minimum toxic and minimum effective blood concentrations, where safe and effective use requires careful dosage titration and patient monitoring. And a competent medical determination that a lack of bioequivalence would have a serious adverse effect in treating or preventing a serious disease.

The section continues into physicochemical and pharmacokinetic factors — low water solubility, slow dissolution, particle size or surface area being critical to bioavailability, poorly dissolving polymorphic forms, a high ratio of excipients to active ingredient, rapid first-pass metabolism, instability in parts of the gastrointestinal tract requiring special coatings, and dose-dependent kinetics in or near the therapeutic range.

The regulatory hook for all of this is 320.21, which requires bioavailability evidence in a full new drug application and bioequivalence evidence against the reference listed drug in an abbreviated application, with routes to waive in vivo evidence in defined circumstances. It also requires new evidence by supplement where a manufacturing site or process changes beyond the variations provided for in the approved application, including a change in product formulation or dosage strength.

One boundary is worth stating explicitly, because it is where consumer confusion concentrates. This entire framework operates on products submitted in new drug applications and abbreviated new drug applications. A compounded preparation is not submitted in either, so there is no bioequivalence determination attached to it — not a favorable one and not an unfavorable one. The absence is structural, not an oversight.