Why the Assay Definition Matters
In target-mediated drug disposition (TMDD), antibody can circulate as unbound drug or bound in a drug–target complex. An assay does not automatically report “the antibody concentration” without qualification: its result depends on which molecular forms it detects and whether sample preparation changes those forms. A free-drug assay, a total-drug assay, and a complex assay can therefore produce different concentration-time profiles from the same samples.
The distinction is important when fitting a model to data. A model state represents a specified molecular species, whereas an assay reports the signal from species its reagents and procedures detect. Comparing a free-drug measurement directly with a model’s total-drug prediction, for example, can create an apparent model discrepancy that is actually a mismatch in measurement definitions.
Mapping Assay Measurements to TMDD States
Consider the canonical, one-to-one binding system in one well-mixed compartment. Let \(D\) be the concentration of free antibody, \(R\) the concentration of free target, and \(C\) the concentration of antibody–target complex. All three concentrations must refer to the same compartment and use compatible molar units.
These equations describe the molecular forms, not a guarantee that every assay measures them perfectly. A free-drug assay is designed to detect unbound antibody while minimizing detection of target-bound antibody. A complex assay is designed to detect the associated antibody–target species. A total-drug assay is intended to detect antibody whether initially free or complexed, often by disrupting complexes during sample preparation or by using reagents that can detect antibody in both forms.
“Total drug” does not mean total antibody-related material under every possible analytical definition. An assay may or may not detect degraded antibody, metabolites, or antibody whose binding sites are occupied by other molecules. Its specificity and recovery depend on the reagents, sample matrix, and validated procedure. In the one-to-one model, \(D+C\) is the model quantity to compare with an assay that has been shown to quantify free plus complexed intact antibody in antibody-equivalent units.
The simple sum also depends on stoichiometry. If a model includes several complex species, the total drug concentration is the sum of free drug and the concentration of each complex multiplied by the number of antibody molecules in that species. For a one-antibody/one-target complex, that multiplier is one. When assay output is reported in mass units rather than molar units, molecular weights and the assay’s calibration definition must also be considered before comparing it with a model concentration.
Worked Example: Reconciling Free, Complex, and Total Results
Suppose a hypothetical plasma sample gives a validated free-antibody result of 42 nM and a complex assay result of 18 nM. Assume the canonical one-to-one model applies, both assays quantify their stated species in the same sample, and both have adequate recovery.
The model-implied total-drug concentration is
The complexed fraction of the antibody is \(C/D_{\mathrm{tot}}=18/60=0.300\), or 30.0%. The free fraction is \(D/D_{\mathrm{tot}}=42/60=0.700\), or 70.0%. These fractions sum to one because the assumed model includes only free antibody and one-to-one complexed antibody.
If a separate total-drug assay reports 58 nM, the 2 nM difference is not automatically evidence of a biological inconsistency. It is a 3.33% difference relative to the 60 nM sum and could reflect analytical imprecision or incomplete recovery. The assay results should be assessed using their validated performance and sample handling, rather than forcing the model states to match without qualification.
Binding Kinetics Explain Different Profiles
For the canonical binding reaction, free drug and free target associate to form complex, and complex can dissociate. A common kinetic representation also allows complex internalization, which removes complex from the measured compartment. Let \(k_{\mathrm{on}}\) be the association rate constant, \(k_{\mathrm{off}}\) the dissociation rate constant, and \(k_{\mathrm{int}}\) the complex internalization rate constant. The binding and complex terms in the state equations are
These terms isolate the binding-related changes. Other processes, such as drug input, nonspecific drug clearance, target synthesis, or target turnover, can also contribute to the full equations. The association term has units of concentration per time: \(k_{\mathrm{on}}\) has units of inverse concentration per time, so \(k_{\mathrm{on}}DR\) has units of concentration per time. The first equation loses free drug during association and gains it during dissociation; the second gains and loses complex through the corresponding processes.
Adding the two equations shows which changes cancel when total drug is considered:
Thus reversible binding alone redistributes antibody between \(D\) and \(C\), but does not change \(D+C\). Complex internalization does reduce total drug in this compartment. If there is also nonspecific clearance of free antibody, or drug input, those terms must be added to the total-drug derivative. This result is a local mass-balance statement under the stated one-compartment assumptions; it does not imply that total antibody across the entire body is constant when antibody moves between compartments.
Worked Example: Free Drug Can Fall Faster Than Total Drug
At a hypothetical instant, let \(D=20\) nM, \(R=5\) nM, and \(C=10\) nM. Suppose \(k_{\mathrm{on}}=0.020\ \mathrm{nM}^{-1}\mathrm{h}^{-1}\), \(k_{\mathrm{off}}=0.10\ \mathrm{h}^{-1}\), and \(k_{\mathrm{int}}=0.030\ \mathrm{h}^{-1}\). Ignore input and other clearance for this calculation.
The association rate is \(k_{\mathrm{on}}DR=(0.020)(20)(5)=2.00\) nM/h. Dissociation returns antibody at \(k_{\mathrm{off}}C=(0.10)(10)=1.00\) nM/h. Therefore the net binding-related change in free drug is
Complex changes at \(2.00-1.00-(0.030)(10)=0.700\) nM/h. Total drug is \(D+C=30\) nM, and its binding-related change is \(d(D+C)/dt=-(0.030)(10)=-0.300\) nM/h. The free concentration is declining more rapidly than the total because binding transfers some free antibody into complex; only the internalized fraction leaves the measured total-drug pool in this simplified calculation.
What Sample Handling Can Change
An assay result is an observation after collection and processing, not a direct view of the in vivo states at the instant of sampling. If complexes dissociate after collection, an assay intended to measure free drug may detect some antibody that was complexed in vivo. Conversely, if complexes form or become stabilized during handling, the measured free concentration may be lower than the in vivo free concentration. A total-drug assay that fully and consistently recovers both forms can be less sensitive to redistribution between \(D\) and \(C\), but it can still be affected by incomplete recovery or analyte degradation.
The operational meaning of “free” is therefore tied to the validated collection and assay procedure. Relevant checks include whether the assay preserves the binding state during processing, whether its signal is affected by target concentration, whether it detects complexed antibody unintentionally, and whether the total assay recovers both free and complexed drug across the expected range. These are measurement checks, not biological mechanisms to add to the TMDD model unless the data support them.
Worked Example: Ex Vivo Dissociation Biases a Free Assay
A hypothetical sample contains 12 nM free antibody and 8 nM complexed antibody in vivo, so the total is 20 nM. During handling, 25% of the complex dissociates before the free-drug assay is run. Assume the assay then detects all antibody that is free at measurement and that no material is lost.
The amount released from complex is \((0.25)(8)=2\) nM. The post-handling free concentration is therefore \(12+2=14\) nM, while the remaining complex is \(8-2=6\) nM. The total remains \(14+6=20\) nM. A free assay would report 14 nM under these assumptions, although the in vivo free concentration was 12 nM. The result is 2 nM, or 16.7%, higher than the in vivo free value. A complex assay could likewise under-report the original complex concentration. This calculation illustrates why the sample-handling behavior must be characterized before treating an assay value as a model state.
Choosing and Comparing the Measurements
| Measurement | Model quantity in the one-to-one case | Main interpretive use | Key limitation |
|---|---|---|---|
| Free drug | \(D\) | Tracks unbound antibody available to bind target, subject to assay definition | Can be biased by complex cross-detection or ex vivo redistribution |
| Complex | \(C\) | Tracks antibody associated with target | May depend on complex stability, target capture, and the assay’s target specificity |
| Total drug | \(D+C\) | Tracks free plus complexed antibody equivalents | Requires adequate and comparable recovery of both forms; may not measure all antibody-related material |
No one measurement is universally best. Free-drug data are most directly aligned with the unbound state, while total-drug data combine free and complexed antibody. Complex data can help distinguish redistribution into bound form from loss of drug, provided the assay specifically and reliably measures the complex. Measuring more than one form can constrain interpretation, but only if the definitions, units, sampling times, and assay performance are compatible.
When comparing profiles, first determine whether the observed data are free, total, or complexed drug. Then compare them with the corresponding model quantity, using the same compartment and units. If a model predicts \(D+C\), a free-drug observation should not be treated as though it directly measured that sum. Likewise, a plasma total-drug result should not be equated with the sum of drug in plasma and tissue compartments unless the measurement and model explicitly define such a whole-body quantity.
Diagnostics, Failure Modes, and Limits
A useful consistency check in the one-to-one setting is whether free drug plus complex agrees approximately with a separately measured total drug, allowing for analytical uncertainty and differences in assay recovery. Persistent disagreement may signal incompatible assay definitions, units, matrix effects, or an omitted molecular form; it is not, by itself, proof of a model failure. Check that all measurements refer to the same samples and compartment and that concentrations are expressed as antibody equivalents.
- Comparing unlike quantities. A free-drug result is not interchangeable with \(D+C\). Match the assay’s measured species to the model state or state combination.
- Assuming the assay label guarantees selectivity. “Free” and “total” describe intended measurement definitions. Confirm cross-reactivity, recovery, and the effects of sample processing.
- Summing complex concentrations without checking stoichiometry. The identity \(D+C\) assumes one antibody molecule per complex. More complex binding networks require molecule-equivalent accounting for each species.
- Interpreting a changing free profile as total-drug loss. A fall in \(D\) can reflect transfer into \(C\); the mass balance shows that reversible binding cancels from the derivative of \(D+C\).
The simple mapping is limited to the modeled species and compartment. It does not resolve assay-specific cross-reactivity, distinguish intact antibody from fragments unless the assay does so, or account automatically for multiple binding sites and higher-order complexes. Those details must be represented in the model and measurement definition when they materially affect the data.
Check Your Understanding
Answer each question, then compare with the brief solutions below.
- A sample has \(D=15\) nM and \(C=5\) nM. What is total drug in antibody-equivalent units, and what fraction is complexed?
- In the one-to-one model, why does reversible binding alone not change \(D+C\)?
- A free-drug assay detects some antibody released from complexes during handling. What direction of bias in the measured free concentration would this create, assuming no material is lost?
- At a given time, \(C=4\) nM and \(k_{\mathrm{int}}=0.025\ \mathrm{h}^{-1}\). What is the complex-internalization contribution to the rate of change of total drug?
Solutions. (1) Total drug is \(15+5=20\) nM, and the complexed fraction is \(5/20=0.250\), or 25.0%. (2) Association removes drug from \(D\) at rate \(k_{\mathrm{on}}DR\) and adds the same amount to \(C\); dissociation adds drug to \(D\) at rate \(k_{\mathrm{off}}C\) and removes the same amount from \(C\). These terms cancel in the sum. (3) The measured free concentration is biased upward relative to the in vivo free concentration because released antibody is counted as free by the assay. (4) The contribution is \(-k_{\mathrm{int}}C=-(0.025)(4)=-0.100\) nM/h.