Why the Reference Administration Matters
A bioavailability comparison is only interpretable after its reference has been specified. Asking how much of a dose reaches the systemic circulation is different from asking whether one formulation delivers more systemic exposure than another. The first question calls for a comparison with intravenous administration; the second calls for a comparison with a named product or administration condition. Changing the reference changes the scientific question, even if the same drug and exposure measure are used.
An estimand is the target quantity a study is intended to estimate. It is defined by the treatments or administrations being compared, the population and conditions of interest, and the outcome contrast. The estimand is not itself an observed concentration, an AUC value, or a calculated sample estimate. Those are data or summaries that may be used to estimate the target.
Relative bioavailability, denoted here by \(F_{\mathrm{rel}}\), is the systemic availability of a test product or administration condition relative to that of a specified reference product or condition. The reference need not be intravenous; commonly, both test and reference are non-intravenous products administered by the same route.
These definitions name different targets. Absolute bioavailability asks how a non-intravenous administration compares with direct entry into the systemic circulation. Relative bioavailability asks how two specified administrations compare with each other. A relative estimate therefore has no absolute interpretation unless the reference administration’s absolute availability is known and the relevant assumptions hold.
What Each Comparison Can Establish
For absolute bioavailability, the test administration is typically oral, while the reference is intravenous. Intravenous administration is the reference because it delivers the dose directly to the systemic circulation, bypassing the absorption process required by an oral dose. The comparison can estimate the systemic availability of the oral administration relative to the IV administration. It does not, by itself, identify which processes caused any loss before the drug reached systemic circulation.
For relative bioavailability, the study identifies a test and a reference, such as a new tablet and an established tablet. If the two have the same active drug, route, and appropriately comparable conditions, the result describes the test product’s systemic availability relative to that reference. It does not say what fraction of either dose reached the systemic circulation in an absolute sense. The reference could itself have low, high, or unknown absolute availability.
A useful way to express the distinction is to imagine a measure of systemic exposure per unit of administered dose. Under suitable conditions, the absolute comparison contrasts that quantity for the non-intravenous test with the IV reference. The relative comparison contrasts it for the test product with the named reference product. This description identifies the target without yet prescribing a particular exposure calculation; the AUC-based calculation and its derivation are treated separately.
| Feature | Absolute bioavailability | Relative bioavailability |
|---|---|---|
| Test administration | Specified non-intravenous route or product | Specified test product or condition |
| Reference administration | Intravenous administration of the same drug | Specified reference product or condition; often a formulation using the same route |
| Main question | How does systemic availability compare with direct IV delivery? | How does the test compare with this particular reference? |
| What the comparison alone does not establish | The mechanism responsible for incomplete systemic availability | The absolute fraction available from either product |
Worked Example: Naming an Absolute Bioavailability Comparison
Hypothetical study. A study gives the same drug orally and intravenously to comparable participants under conditions intended to support a valid comparison. The oral administration is the test, and IV administration is the reference. For illustration, suppose the analysis reports systemic exposure per administered milligram of 63 exposure units/mg after oral administration and 100 exposure units/mg after IV administration.
The relevant ratio is the oral value divided by the IV value:
Thus, in this hypothetical comparison, the oral administration’s exposure per administered milligram is 0.63 times the IV value, or 63% of it. This is an absolute-bioavailability comparison because the reference is IV. The calculation illustrates interpretation of a supplied, dose-normalized exposure summary; it is not a derivation of how that summary is obtained.
The result does not reveal whether the difference arose during absorption, before systemic entry, or through some other process that violates the comparison’s assumptions. It also should not be described as proof that exactly 63% of the oral dose was absorbed: the estimand concerns systemic availability relative to IV delivery, not absorption alone.
Worked Example: Naming a Relative Bioavailability Comparison
Hypothetical study. A new capsule (test) and a marketed tablet (reference) contain the same drug and are administered orally. Suppose their reported exposure-per-dose summaries are 88 and 80 exposure units/mg, respectively.
The relative comparison is:
The test capsule’s exposure per administered milligram is 1.10 times the reference tablet’s value, or 10% higher in this hypothetical result. This is a relative comparison because the reference is the tablet, not an IV administration.
The result does not establish that the capsule provides 110% of the dose systemically, nor does it establish that the products are therapeutically equivalent. It describes the test relative to this particular tablet under the study conditions. A different reference could produce a different relative estimate.
Why Dose-Normalized Exposure Can Support These Targets
The logic behind exposure comparisons can be summarized without deriving a particular AUC estimator. Suppose drug disposition is linear over the relevant dose range, clearance is the same across the compared administrations, and the measured exposure is proportional to the amount reaching systemic circulation. Then exposure per administered dose is proportional to systemic availability divided by clearance. For two administrations in the same comparison, the shared clearance factor cancels in their ratio. The remaining ratio reflects the relative systemic availability of those administrations.
For an absolute comparison, the IV reference has systemic availability defined as 1 under the reference convention, so the non-IV administration’s ratio targets its availability relative to IV. For a relative comparison, neither administration is fixed at 1: the ratio targets test availability divided by reference availability. This cancellation argument explains why exposure comparisons can address bioavailability, but only when the assumptions are credible. The later derivation of the AUC-based estimator makes the calculation explicit.
Worked Example: Understanding What a Relative Result Does Not Tell You
Hypothetical comparison. A test formulation and a reference formulation have exposure-per-dose summaries of 72 and 60 arbitrary exposure units/mg. The relative estimate is:
The test has 20% greater exposure per administered milligram than the reference in this hypothetical comparison. Now suppose, solely to illustrate interpretation, that the reference has an absolute availability of 0.50 and that the same linear, comparable-clearance assumptions apply. The test’s implied absolute availability would then be:
The arithmetic shows why a relative estimate alone is insufficient to determine absolute availability: the reference’s absolute availability is also needed. Without that additional information, the supported conclusion is only that the test is 1.20 times the reference on the stated exposure-per-dose scale.
The assumptions matter. If clearance differs between the conditions, or exposure is not proportional to dose, a ratio of exposure summaries need not isolate a ratio of bioavailability. The calculation is therefore an interpretation under stated assumptions, not a guarantee that the relative estimate can always be converted into an absolute one.
Assumptions, Diagnostics, and Alternatives
A comparison is most informative when the administrations differ in the intended way and other factors that affect exposure are controlled or appropriately addressed. Important conditions include use of the same active drug and analyte definition, comparable measurement and sampling, and a design that permits the treatment conditions to be compared fairly. For interpretations based on exposure per dose, linear pharmacokinetics over the compared doses and comparable clearance are especially important. These are assumptions to assess, not automatic consequences of calculating a ratio.
Potential warning signs include visibly different dose ranges where dose proportionality is uncertain, inconsistent sampling or assay procedures, different study populations or conditions, and evidence that clearance or other disposition processes differ between treatment periods. In crossover studies, period or carryover effects may also undermine a simple within-person contrast if not addressed by the design and analysis. Such issues can make an apparent availability difference reflect something other than the intended product or route comparison.
The main alternative is not a different name for the same estimand, but a different reference question. Use an IV reference when the aim is absolute availability; use a specified non-IV reference when the aim is relative availability between products or conditions. Comparing raw exposure without accounting for dose can answer a different question when administered doses differ. Neither an absolute nor a relative comparison, by itself, establishes the biological mechanism behind an exposure difference.
A common mistake is to call any comparison of two formulations “absolute bioavailability,” or to interpret a relative estimate of 1.10 as meaning that 110% of the dose reached systemic circulation. The correction is to name the reference first. A test-versus-IV comparison targets absolute bioavailability; a test-versus-product comparison targets relative bioavailability. A relative ratio is meaningful only in relation to its stated reference and assumptions.
Check Your Understanding
Answer each question by identifying the comparison target and stating what the result can and cannot establish.
- A nasal formulation is compared with IV administration of the same drug. Is the target absolute or relative bioavailability? What is the reference?
- A new oral solution is compared with an established oral tablet. The test-to-reference exposure-per-dose ratio is 0.92. State the interpretation and one conclusion that does not follow.
- Two oral products have a relative estimate of 1.15. Can this number alone establish the absolute availability of the test product? Explain.
- Why can differing clearance between two compared conditions complicate interpreting an exposure-per-dose ratio as a bioavailability ratio?
Solutions. (1) It is an absolute bioavailability comparison; IV administration is the reference. (2) The solution has 0.92 times the reference tablet’s exposure per administered dose under the comparison’s assumptions. The result alone does not establish the fraction of the dose reaching systemic circulation. (3) No. The ratio gives test availability relative to the reference; the reference’s absolute availability and suitable assumptions would also be needed. (4) Exposure per dose depends on clearance as well as systemic availability. If clearance differs, it may not cancel in the ratio, so the ratio may not isolate bioavailability.