Common Structural Errors in Antibody Disposition Models

Why Structural Checks Matter

An antibody PBPK model can produce a smooth plasma concentration-time profile and still be structurally wrong. A duplicated return flux can create antibody, a misplaced degradation term can remove it twice, and a concentration calculated with the wrong tissue volume can misstate the amount available for transport. Adjusting parameters may improve the fit without fixing any of these problems.

The useful starting point is an amount balance. A transport flux leaving one compartment must enter another compartment unless it represents a defined loss from the modeled system. This tutorial uses that principle to check four recurring errors in antibody disposition models. Amounts are denoted by \(A\), concentrations by \(C\), and fluxes by \(J\); amounts are in mg, concentrations in mg/L, and fluxes in mg/day unless stated otherwise.

The checks below assume that compartment amounts are the states, each compartment is well mixed at its defined concentration, and each internal transfer is represented once in each affected balance. These bookkeeping assumptions do not imply that a particular compartment structure is biologically complete. They do make it possible to detect whether the equations conserve antibody as intended.

Use the Whole-System Balance as a Reference

Consider a tissue with vascular amount \(A_V\), interstitial amount \(A_I\), lymph amount \(A_L\), and endosomal amount \(A_E\). Let \(J_{\mathrm{cap}}\) be net transcapillary flux, positive from vascular space to interstitium; \(J_{I,L}\) be interstitial-to-lymph flux; \(J_{L,V}\) be lymph-to-vascular return; and \(J_{\mathrm{up}}\) be uptake into endosomes. Let \(J_{\mathrm{salv}}\) be endosomal salvage returned to vascular space, and \(J_{\mathrm{deg}}\) be irreversible degradation. A simple balance is:

Adding these equations cancels every internal transfer: capillary exchange, lymphatic transport, uptake, and salvage. The remaining whole-system balance is:

Thus, without degradation or another explicitly modeled external loss, the sum of modeled amounts must remain constant after dosing stops. With degradation, the total falls at the degradation rate. This simple sum is a targeted structural check, not a claim that every tissue must use exactly these four compartments.

Check 1: Lymphatic Flux Must Have One Route

Lymphatic drainage can be represented with an explicit lymph compartment or as a direct return from interstitium to vascular space. Either choice can be internally consistent. The error is combining both representations for the same physical flow: for example, transferring antibody from interstitium into lymph and from lymph to vascular space, while also adding a direct interstitial-to-vascular return for that same lymph flow.

A practical diagnostic is to draw the route and label each arrow once. For every arrow, identify the exact source and destination state variables and check that its source term is negative and its destination term is positive with the same magnitude. If a lymph compartment is explicit, a direct return term should not duplicate its return path.

Check 2: Concentrations Must Use the Matching Volume

For a well-mixed compartment, concentration is amount divided by that compartment’s volume: \(C_i=A_i/V_i\). A tissue’s total anatomical volume is not automatically the interstitial fluid volume. If an exchange or uptake flux depends on interstitial concentration, the amount must be divided by the interstitial volume used to define that state. Using a different volume changes the implied concentration and therefore changes the calculated flux.

When a model uses concentration states rather than amount states, the volume basis also enters the differential equations. If \(A_i=V_iC_i\) and \(V_i\) is constant, then \(dA_i/dt=V_i\,dC_i/dt\). Consequently, equal-and-opposite amount fluxes do not generally appear as equal-and-opposite concentration derivatives when compartment volumes differ. The conservation check should be performed on amounts, or on concentration derivatives multiplied by their corresponding volumes.

Check 3: Endosomal Salvage Is Not the Same as Degradation

Endosomal handling separates antibody fates. Uptake moves antibody into the endosomal compartment; salvage returns a portion to the vascular or other specified extracellular space; degradation is an irreversible loss from the modeled antibody system. If degradation is intended but absent, the model retains antibody that should have been removed. Conversely, subtracting degradation from both endosomal and systemic amounts counts the same loss twice.

Check 4: Exchange Terms Must Cancel

For passive transcapillary exchange, a common illustrative form is \(J_{\mathrm{cap}}=PS(C_V-C_I)\), with positive flux defined from vascular space to interstitium. Here \(PS\) has units L/day and the concentrations have units mg/L, giving \(J_{\mathrm{cap}}\) in mg/day. The vascular balance receives \(-J_{\mathrm{cap}}\), and the interstitial balance receives \(+J_{\mathrm{cap}}\). Other models may use more detailed exchange terms, but the internal amount accounting still has to cancel.

A Short Derivation and a Practical Audit

The cancellation rule follows directly from the definition of an internal flux. For one transfer \(J_{i,j}\) from compartment \(i\) to compartment \(j\), its contributions are \(-J_{i,j}\) to \(dA_i/dt\) and \(+J_{i,j}\) to \(dA_j/dt\). Summing the two contributions gives:

Applying this to every internal transfer leaves only inputs and losses that cross the boundary of the full modeled system. This yields a direct audit procedure:

Diagnostics, Alternatives, and Limits

A mass-balance audit is complementary to fitting plasma data. A good fit does not prove that lymph return is represented once, that tissue concentrations use the right volumes, or that endosomal degradation is present. Conversely, a mass-conserving model may still have incorrect parameters, an inappropriate compartment structure, or poor predictions. Conservation is a necessary bookkeeping check, not a sufficient validation criterion.

A useful numerical diagnostic is to evaluate the summed derivative at a chosen state and compare it with the specified external input and loss. A nonzero residual after internal terms are summed points to a structural or coding discrepancy. Unit checks can reveal incompatible flow and concentration units, while route diagrams help expose duplicate pathways that equations alone may obscure. A conservation residual near zero does not rule out two errors that happen to cancel; inspect each transfer individually as well as the total.

An explicit lymph compartment preserves information about lymph transit and its amount. A direct interstitial-to-vascular return can be a simpler approximation when that transit is not represented. These alternatives are not interchangeable within the same balance: the chosen structure must define one route and account for its source and destination. Similarly, amount-based states make conservation especially transparent, whereas concentration-based states can be used reliably if every volume conversion is applied consistently.

The checks do not establish that a chosen endosomal model captures all cellular handling, nor do they determine the biological value of a degradation rate. They identify whether the model equations implement the stated mechanisms without creating or losing antibody unintentionally. Separate evidence and model evaluation are needed to support the mechanisms and parameter values.

Common Mistakes

• “Both lymph terms are small, so keeping both is harmless.” Even a small duplicated flux creates a systematic amount-balance error. Remove the duplicate route rather than assuming it is negligible.
• “The concentration has the right units, so the volume is correct.” Units do not identify the physical compartment represented by a volume. Use the volume matching the amount state.
• “Salvage handles endosomal antibody, so a degradation term is unnecessary.” Salvage is return to another modeled compartment; degradation is irreversible loss. Include each specified fate once.
• “The plasma curve fits, so exchange must be balanced.” Parameter adjustment can conceal structural errors over a limited observation window. Sum the amount equations directly.