Separating Absorption From First-Pass Loss

Why the Same Exposure Can Have Different Causes

A low concentration-time exposure after an oral dose can arise because only part of the dose is absorbed, because some absorbed drug is lost before reaching systemic circulation, or because both occur. These explanations matter: they suggest different mechanisms and may lead to different formulation or development decisions. Yet systemic exposure alone usually does not reveal which explanation is responsible.

The key is to distinguish the total fraction reaching systemic circulation from the component fractions that make up that total. Earlier material on fraction absorbed and presystemic availability components defines the oral bioavailability equation. Here, the focus is what that equation permits us to conclude—and what it does not permit us to conclude—from an exposure measurement.

Definition: For an orally administered drug, $F_a$ is the fraction of the administered dose absorbed from the gastrointestinal lumen; $F_g$ is the fraction of absorbed drug that escapes intestinal presystemic loss; and $F_h$ is the fraction reaching the liver that escapes hepatic first-pass loss. The fraction reaching systemic circulation is oral systemic bioavailability, $F$: $$ F = F_a F_g F_h. $$ Each component is a fraction between 0 and 1 under this model. The factors describe successive stages, so their product—not their sum—is the fraction of the administered dose reaching systemic circulation.

This equation is a bookkeeping model of sequential fractions. Of each administered dose, the fraction $F_a$ is absorbed; of that amount, the fraction $F_g$ escapes intestinal loss; and of the amount surviving those stages, the fraction $F_h$ escapes hepatic first-pass loss. Multiplication follows directly from these successive proportions.

What Systemic Exposure Identifies

Under linear pharmacokinetics, with comparable systemic clearance and complete exposure measurements, oral systemic bioavailability can be estimated by comparing dose-normalized oral and intravenous AUC. The calculation identifies $F$, the overall fraction reaching systemic circulation. It does not separately measure $F_a$, $F_g$, or $F_h$.

Key result: An estimate of $F$ supplies one equation, $F_aF_gF_h=F$, for three component fractions. Unless additional information constrains or measures some of those fractions, systemic exposure alone cannot determine their individual values.

The limitation is structural, not a matter of collecting a more precise AUC. A highly precise estimate of the product can still be compatible with many different combinations of its factors. For example, an overall fraction of 0.30 could reflect 30% absorption followed by no further loss, or substantially greater absorption followed by intestinal or hepatic loss.

Worked Example: Estimating Overall Oral Bioavailability

Suppose a hypothetical subject receives 100 mg orally and 10 mg intravenously in a setting where the linear pharmacokinetic assumptions are reasonable. The measured AUCs are 12 mg·h/L after the oral dose and 4 mg·h/L after the intravenous dose. Calculate oral systemic bioavailability.

The dose-normalized AUC ratio is: $$ F = \frac{\mathrm{AUC}_{\mathrm{oral}}/D_{\mathrm{oral}}}{\mathrm{AUC}_{\mathrm{IV}}/D_{\mathrm{IV}}} = \frac{12\ \mathrm{mg\cdot h/L}/100\ \mathrm{mg}}{4\ \mathrm{mg\cdot h/L}/10\ \mathrm{mg}}. $$

The oral dose-normalized AUC is $0.12\ \mathrm{h/L}$, and the intravenous dose-normalized AUC is $0.40\ \mathrm{h/L}$. Therefore: $$ F = \frac{0.12\ \mathrm{h/L}}{0.40\ \mathrm{h/L}} = 0.30. $$

The estimate is 30% systemic bioavailability under the stated assumptions. It does not establish that 70% of the dose was unabsorbed: the portion that did not reach systemic circulation may include both unabsorbed drug and drug lost presystemically.

Why the Components Are Not Identifiable From the Product

The ambiguity can be shown algebraically. Suppose $F$ is known and lies strictly between 0 and 1. Any proposed values of $F_a$, $F_g$, and $F_h$ that lie between 0 and 1 and satisfy their product equal to $F$ give the same systemic bioavailability. For example, choosing $F_a=0.60$ and $F_g=0.50$ requires $F_h=F/(0.60\times0.50)$, if that value is no greater than 1. The exposure constrains the product, not the choice of factors that produced it.

A simple family of solutions makes the non-uniqueness explicit. Set $F_a=1$ and $F_g=1$; then $F_h=F$ satisfies the equation. Alternatively, set $F_g=1$ and $F_h=1$; then $F_a=F$ also satisfies it. Both combinations are mathematically compatible with the same overall exposure, even though they imply different explanations. There are also combinations in which multiple stages contribute to the reduction.

F_aF_gF_h=F \qquad\Longrightarrow\qquad \text{one observed product does not uniquely specify three factors.}

This is an identifiability issue: different component values produce the same observable quantity. Improving the precision of the AUC comparison reduces uncertainty about $F$, but it does not, by itself, resolve which component values generated $F$.

Worked Example: Three Explanations for the Same Exposure

Suppose a hypothetical oral study estimates $F=0.30$. Consider three possible component combinations.

Absorption-limited combination: $F_a=0.30$, $F_g=1.00$, and $F_h=1.00$. Their product is $0.30\times1.00\times1.00=0.30$.

Intestinal-loss combination: $F_a=0.60$, $F_g=0.50$, and $F_h=1.00$. Their product is $0.60\times0.50\times1.00=0.30$.

Loss at more than one stage: $F_a=0.75$, $F_g=0.80$, and $F_h=0.50$. Their product is $0.75\times0.80\times0.50=0.30$.

All three combinations predict the same overall systemic bioavailability. The exposure estimate alone cannot select among them. In particular, describing the result as “70% first-pass loss” would be unjustified: in the first combination, the reduction is due to incomplete absorption, whereas the other combinations include presystemic losses.

Using Independent Information to Resolve a Component

The component equation becomes useful for separating mechanisms when some component fractions are estimated independently. If $F$, $F_a$, and $F_g$ are known, for instance, the remaining component can be calculated as $F_h=F/(F_aF_g)$. This calculation is only meaningful if the component estimates refer to the same drug, dose conditions, and relevant study setting, and if the assumptions behind each estimate are defensible.

Worked Example: Inferring Hepatic Escape From Independent Estimates

In a hypothetical investigation, systemic exposure implies $F=0.36$. Suppose a separate, validated approach estimates $F_a=0.60$ and $F_g=0.80$. Calculate the implied hepatic escape fraction.

Starting with $F=F_aF_gF_h$, rearrange to obtain: $$ F_h = \frac{F}{F_aF_g} = \frac{0.36}{0.60\times0.80} = \frac{0.36}{0.48} = 0.75. $$

The implied hepatic escape fraction is 0.75, or 75%. This result is conditional on the independent estimates and the component model; it is not obtained from systemic exposure alone. A consistency check is that $0.60\times0.80\times0.75=0.36$, as required.

The same calculation can be understood in terms of dose fractions. Of the administered dose, 0.60 is absorbed. Of that absorbed amount, $0.60\times0.80=0.48$ escapes intestinal loss. Of that amount, 0.75 reaches systemic circulation, giving $0.48\times0.75=0.36$. The fractions of dose lost at the stages are not the same as the conditional component fractions: for example, hepatic loss here is $0.48\times(1-0.75)=0.12$ of the administered dose.

Independent information can come from measurements or designs that bear on a particular stage, such as validated absorption or mass-balance measurements, appropriately interpreted metabolite data, or a mechanistic model supported by data beyond systemic parent-drug AUC. Each approach has its own assumptions. For example, drug recovered in feces need not represent only unabsorbed drug if drug can also enter the gut after absorption. A measurement intended to distinguish mechanisms must therefore account for relevant alternative sources and fates.

A second strategy is to compare conditions expected to change a particular process. Such comparisons can provide useful evidence, but a change in systemic exposure does not itself prove that only one component changed. An intervention could alter absorption, intestinal escape, hepatic escape, or more than one of them. Mechanistic attribution requires independent support for the proposed pathway.

Assumptions, Diagnostics, and Common Pitfalls

The component equation is a simplified representation of sequential fractions. Its interpretation depends on defining the fractions consistently and applying them to the same administered dose and conditions. The exposure-based estimate of $F$ also relies on the assumptions needed for a dose-normalized AUC comparison: dose proportionality over the comparison, comparable systemic clearance, and sufficiently complete and comparable AUC measurements. If those assumptions do not hold, the estimated exposure ratio may not represent the intended systemic availability.

Several checks help identify problems in an analysis:

Check the exposure comparison. Confirm that doses and AUCs are expressed in compatible units and that the dose-normalized ratio is calculated in the correct order.
Check component definitions. Each factor is conditional on reaching the preceding stage. Treating a fraction of the original dose as though it were a conditional escape fraction can produce incorrect multiplication.
Check consistency. If independently estimated factors imply a product that conflicts with measured $F$, investigate measurement uncertainty, mismatched conditions, model assumptions, and whether the factors actually represent the defined stages.
Check theoretical bounds. Under the component model, each fraction is between 0 and 1. Estimates outside these bounds indicate that uncertainty or assumptions need attention; they should not be interpreted as literal component fractions without further evaluation.
Check for unsupported mechanistic claims. A single systemic exposure estimate cannot locate loss at the absorption, intestinal, or hepatic stage.

Common mistake: “The oral bioavailability is 30%, so 70% of the dose was lost in first-pass metabolism.” This confuses the overall fraction that failed to reach systemic circulation with a specific mechanism. The remaining 70% can include unabsorbed drug as well as intestinal and hepatic presystemic loss. The correction is to report the measured total as $F=0.30$ and state that the component contributions are not identified by that exposure result alone.

The main alternative to an exposure-only interpretation is not a different rearrangement of the same equation. It is additional evidence: measure or constrain one or more components independently, then use the product relationship to infer any remaining component. This approach can distinguish mechanisms only to the extent that the added evidence is valid and sufficiently informative. Without it, the scientifically appropriate conclusion is that several component patterns remain compatible with the observed systemic exposure.

Key takeaway: Oral systemic bioavailability is the product $F_aF_gF_h$. Dose-normalized systemic exposure can estimate that product under suitable pharmacokinetic assumptions, but cannot generally separate incomplete absorption from intestinal or hepatic presystemic loss. Component attribution requires additional, independent information.

Check Your Understanding

Answer each question, then compare your reasoning with the solutions below.

If $F=0.24$, give two different valid combinations of $F_a$, $F_g$, and $F_h$ that produce this value.
A hypothetical analysis estimates $F=0.42$, $F_a=0.70$, and $F_g=0.75$. What is the implied $F_h$, and what assumptions qualify this result?
Why does a more precise estimate of oral systemic bioavailability not, by itself, identify the absorption fraction?
A researcher says that an oral bioavailability estimate of 0.40 proves that 60% of the dose was metabolized before reaching systemic circulation. What is wrong with this conclusion?

Solutions. (1) For example, $0.24\times1.00\times1.00=0.24$, while $0.60\times0.80\times0.50=0.24$. Both are valid combinations, but they imply different component patterns. (2) $F_h=0.42/(0.70\times0.75)=0.42/0.525=0.80$. This is an implied fraction, conditional on the component estimates being valid, applying to the same conditions, and matching the model definitions. (3) Exposure identifies the product $F_aF_gF_h$; many combinations can have the same product, regardless of how precisely that product is estimated. (4) The estimate describes the total fraction reaching systemic circulation, not the mechanism of the remaining fraction. Some drug may be unabsorbed, and the exposure alone does not separate that amount from intestinal or hepatic presystemic loss.