Why Unequal Doses Need Careful Interpretation
An AUC comparison describes exposure after the doses that were actually administered. If the test formulation is given at a larger dose than the reference, its AUC may be larger simply because more drug was given. The raw ratio of AUCs therefore does not, by itself, isolate the difference between formulations. To compare formulations on a dose-normalized basis, adjust the raw ratio by the doses.
The key distinction is between exposure under the administered regimens and exposure per unit dose. The first is described by the raw AUC ratio. The second is described by the dose-adjusted ratio, under assumptions that make dose normalization meaningful. The calculation is simple; interpreting it correctly requires keeping track of which dose belongs to which formulation.
The dose factor is reference dose divided by test dose, \(D_R/D_T\), not the other way around. This ordering follows directly from dividing the test AUC per dose by the reference AUC per dose:
When AUC is measured in, for example, concentration-times-hours and dose in milligrams, each dose-normalized AUC has units of concentration-times-hours per milligram. Their ratio is dimensionless. A value of \(F_{\mathrm{rel}}=0.80\) means that the test formulation produced 80% of the dose-normalized AUC of the reference in the comparison; equivalently, its dose-normalized AUC was 20% lower. It does not mean that the test regimen produced 20% less total exposure unless the administered doses were equal.
What Makes Dose Adjustment Meaningful?
Dose normalization is interpretable when AUC is proportional to dose over the dose range being compared. In a standard linear pharmacokinetic model for the same drug and route, AUC after an extravascular dose can be represented as \(AUC=F D/CL\), where \(F\) is systemic availability and \(CL\) is systemic clearance. Dividing by dose gives \(AUC/D=F/CL\). If clearance and other relevant conditions are comparable, a ratio of dose-normalized AUCs can therefore compare formulation-related systemic availability.
The calculation does not itself prove that these conditions hold. Before interpreting a dose-adjusted ratio, check that the compared AUCs refer to the same analyte and the same AUC definition, such as AUC over the same specified interval. Confirm that the dose units and formulation labels are correct, and that the conditions support dose-proportional exposure over the doses used. Comparisons across different populations or markedly different study conditions may also reflect differences other than formulation.
This tutorial focuses on the point calculation. In an actual study, the observed AUCs may vary across participants, and an inferential analysis is needed to quantify uncertainty. A ratio calculated from summary values is not, on its own, evidence that a population-level formulation difference is statistically established.
Worked Example: A Higher Test Dose Produces a Larger Raw AUC
In a hypothetical comparison, the test formulation is administered at 200 mg and has an AUC of 18 mg·h/L. The reference formulation is administered at 100 mg and has an AUC of 12 mg·h/L. Assume the AUCs are comparable measures and exposure is dose-proportional over these doses.
Raw exposure ratio. The observed AUC ratio is \(18/12=1.50\). Thus the test regimen produced 1.50 times the AUC of the reference regimen. This comparison includes the fact that the test dose was twice as large.
Dose adjustment. Use the reference dose over the test dose:
The same result follows from comparing AUC per milligram: \(18/200=0.0900\) and \(12/100=0.1200\), in units of h/L. Their ratio is \(0.0900/0.1200=0.7500\). Under the stated assumptions, the test formulation's dose-normalized AUC is 75.0% of the reference's, or 25.0% lower. The raw AUC is higher, but that does not mean the test formulation has higher dose-normalized exposure.
Worked Example: A Large Raw Ratio Can Become a Smaller Adjusted Ratio
In a second hypothetical comparison, the test formulation is administered at 300 mg and produces an AUC of 60 mg·h/L. The reference formulation is administered at 100 mg and produces an AUC of 30 mg·h/L. Assume the same AUC definition is used and dose proportionality is reasonable over the compared range.
The raw ratio is \(60/30=2.00\): the test regimen produced twice the AUC. But the test dose was three times the reference dose, so the dose factor is \(100/300\):
The multiplication uses the unrounded factor \(100/300\), so \(2.00\times(100/300)=0.6666\ldots\), which rounds to 0.6667. Equivalently, the dose-normalized AUCs are \(60/300=0.2000\) and \(30/100=0.3000\) mg·h/L per mg; \(0.2000/0.3000=0.6667\). The adjusted comparison is therefore about 66.7% of the reference, not twice the reference. The raw ratio answers a different question: how the two administered regimens' total AUCs compare.
Calculation Checks and Interpretation
A reliable calculation can be done in two equivalent ways: apply the dose factor to the raw AUC ratio, or first calculate each AUC per dose and then take their ratio. Agreement between the two is a useful arithmetic check. The order of doses provides another check: if the test dose is larger than the reference dose, \(D_R/D_T\) is below 1, so the dose adjustment reduces the raw ratio. If the test dose is smaller, the factor is above 1 and increases it.
Worked Example: Test Dose Smaller Than Reference Dose
In a hypothetical study, the test dose is 100 mg with an AUC of 40 mg·h/L, while the reference dose is 150 mg with an AUC of 36 mg·h/L. Assume the AUC measures are comparable and exposure is proportional to dose across this range.
The raw AUC ratio is \(40/36=1.1111\). The dose-adjustment factor is \(150/100=1.5000\), because the reference dose is larger:
Checking by dose-normalized AUC gives \(40/100=0.4000\) and \(36/150=0.2400\) mg·h/L per mg. Their ratio is \(0.4000/0.2400=1.6667\). Under the assumptions, the test formulation's dose-normalized AUC is about 166.7% of the reference's, or 66.7% higher. Although the raw AUCs are fairly close, the test formulation reached that exposure with a smaller dose.
These calculations are point comparisons, not a complete study conclusion. In data from multiple participants, the chosen summary and statistical analysis must reflect the study design and variability. A ratio of group means and a mean of individual ratios are not generally the same quantity. The calculation here establishes how the dose factor enters; it does not prescribe an inferential method or establish a regulatory decision.
Common Mistakes, Alternatives, and Limitations
- Using the dose factor upside down. Multiplying by \(D_T/D_R\) reverses the adjustment. Start from \((AUC_T/D_T)/(AUC_R/D_R)\) to verify that the factor is \(D_R/D_T\).
- Calling the raw AUC ratio relative bioavailability. With unequal doses, the raw ratio compares the administered regimens. It does not remove the dose difference.
- Interpreting a dose-adjusted result without checking dose proportionality. If AUC does not scale proportionally with dose, dividing by dose need not produce a valid formulation comparison across unequal doses. This calculation cannot repair that problem.
- Mixing AUC definitions, units, or formulation labels. Use the same AUC endpoint, align units, and verify which dose belongs to test and reference before calculating.
- Treating a point ratio as proof of a difference or equivalence. A point estimate has no uncertainty measure by itself. Statistical conclusions require a suitable analysis of the study data.
The main alternative to dose adjustment is to compare formulations at the same dose, which makes the raw AUC ratio easier to interpret as a formulation comparison when other conditions are comparable. If unequal doses are used, the dose-normalized ratio is the direct arithmetic alternative, but it relies on dose proportionality. Reporting both the raw AUC ratio and the adjusted ratio can be useful, provided their meanings are stated separately. Neither arithmetic approach removes confounding from differences in study conditions or substitutes for an analysis of uncertainty.
Check Your Understanding
For each question, identify what the ratio means and show the dose adjustment where needed.
- A test dose of 250 mg gives an AUC of 50 mg·h/L, and a reference dose of 100 mg gives an AUC of 30 mg·h/L. Find the raw ratio and the dose-adjusted ratio.
- The raw test-to-reference AUC ratio is 0.80. The test dose is 100 mg and the reference dose is 200 mg. Calculate the dose-adjusted ratio.
- Why is it incorrect to conclude that a test formulation has greater dose-normalized exposure just because its raw AUC is greater when the test dose is also greater?
- Name two checks that should be made before interpreting a dose-adjusted AUC comparison.
Solutions. For question 1, the raw ratio is \(50/30=1.6667\). The dose factor is \(100/250=0.4000\), so \(F_{\mathrm{rel}}=(50/30)(100/250)=0.6667\). The test regimen has the larger raw AUC, but the test dose-normalized AUC is 66.7% of the reference's under the assumptions. For question 2, \(F_{\mathrm{rel}}=0.80(200/100)=1.60\); the adjusted test-to-reference ratio is 1.60. For question 3, the raw ratio includes the dose difference as well as any formulation difference, so it does not isolate exposure per unit dose. For question 4, suitable checks include confirming the same AUC definition and analyte, verifying dose units and formulation labels, and assessing whether dose proportionality and comparable study conditions are reasonable.