Target-Cell Depletion Versus Immune Control of Viral Infection

Two Mechanisms Can Produce Similar Viral Declines

A falling viral load is an outcome, not a mechanism. It can occur because infection has depleted the cells needed for viral production, because immune effectors are removing infected cells, or because both processes are operating. Distinguishing these explanations matters when interpreting treatment or disease trajectories: the mechanisms imply different target-cell preservation and different responses to perturbations.

The target-cell-limited model introduced in earlier material provides a baseline. Here, the comparison is between that model and a model in which an abstract immune effector population increases infected-cell loss. The effector is deliberately not assigned to a particular lineage or molecular pathway. The focus is on model structure and predicted trajectories, not on the detailed biology of a specific immune response.

Constructing the Competing Models

For target-cell-limited infection, use

Here \(s_T\) is target-cell supply (cells/mL/day), \(d_T\) is the uninfected-cell loss rate (day\(^{-1}\)), \(\beta\) is the infection rate constant (mL virion\(^{-1}\) day\(^{-1}\)), \(\delta\) is the infected-cell loss rate (day\(^{-1}\)), \(p\) is viral production per infected cell (virions cell\(^{-1}\) day\(^{-1}\)), and \(c\) is viral clearance (day\(^{-1}\)). In the absence of infection, the target-cell steady state is \(T_0=s_T/d_T\). During infection, the term \(\beta VT\) removes target cells and adds infected cells at the same rate.

An immune-mediated control model retains those processes and adds an effector concentration \(E\), measured in cells/mL:

The parameter \(k\) is an effector-mediated killing rate constant (mL cell\(^{-1}\) day\(^{-1}\)); thus \(kE\) is a rate in day\(^{-1}\). The effector source \(s_E\) has units cells/mL/day, \(a\) and \(d_E\) have units day\(^{-1}\), and \(K_I\) is an infected-cell concentration (cells/mL). This simple expansion term is one possible phenomenological representation of effector dynamics, not a claim that all immune responses follow this equation.

Both models assume that infection is proportional to \(V T\), viral production is proportional to \(I\), and viral clearance is first order in \(V\). The immune model additionally assumes that effector-mediated killing is proportional to \(EI\). These are useful starting assumptions, but they should be questioned if measured data show saturation, delays, or changing rates that the equations cannot reproduce.

Deriving the Local Control Threshold

To see how the mechanisms affect viral growth, hold \(T\) fixed over a short interval and, in the immune model, hold \(E\) fixed as well. The infected-cell and virus equations then form a linear system. Write \(q=\delta\) for the target-cell-limited model and \(q=\delta+kE\) for the immune model. A perturbation proportional to \(e^{\lambda t}\) satisfies

A nonzero solution requires the determinant to equal zero. Therefore,

At the boundary between local growth and decline, \(\lambda=0\), so the threshold target-cell concentration is

If \(T>T_{\mathrm{crit}}\), the dominant eigenvalue is positive and small infected-cell/virus perturbations grow under these fixed-state assumptions; if \(T<T_{\mathrm{crit}}\), they decline. Equivalently, the local reproduction ratio is \(\mathcal{R}_T=\beta pT/(cq)\), with growth when \(\mathcal{R}_T>1\). Immune killing raises \(q\), and therefore raises \(T_{\mathrm{crit}}\): control can occur while more target cells remain. This is a local result, not a guarantee about the full trajectory when \(T\) and \(E\) are changing.

The dimensions provide a useful check. \(\beta pT\) and \(cq\) both have units day\(^{-2}\), so \(\mathcal{R}_T\) is dimensionless. The ratio \(cq/(\beta p)\) has units cells/mL, as required for \(T_{\mathrm{crit}}\).

Predicted Patterns and Their Interpretation

In a target-cell-limited trajectory, infection initially expands while the target-cell pool is abundant. As infection consumes targets, \(T\) may approach or cross \(T_{\mathrm{crit}}\); infected cells and virus then stop growing and can decline. A key prediction is that viral control is associated with substantial target-cell reduction, unless target-cell supply or other processes maintain the pool.

In the immune-mediated model, a rising \(E\) increases the infected-cell loss rate \(\delta+kE\). Viral decline can therefore begin while \(T\) remains above the target-cell-only threshold. Because fewer cells become infected, target-cell loss may also be slower, so target cells can be relatively preserved. This does not mean every immune-controlled trajectory must preserve targets: depletion may have occurred before effector activity increased, or the effector response may be too weak to prevent it.

These are patterns of consistency, not unique fingerprints. The earlier material on biphasic viral decline and long-lived infected cells shows why a change in viral slope need not identify a single process. Here, the same caution applies: observing a viral decline alone cannot distinguish a shrinking target-cell pool from increased infected-cell loss.

Diagnostics, Alternatives, and Limitations

Mechanistic comparison requires more than matching a viral-load curve. A practical diagnostic is to examine whether one parameterization can account for viral load, target-cell abundance, and the timing of any proposed increase in immune activity. If the target-cell-only model predicts control only after a large target-cell decline that is not present in the data, its mechanism is weakened. Conversely, a decline aligned with target loss is compatible with target limitation, but does not rule out concurrent immune killing.

The main alternatives can be compared as follows:

When both models fit viral load similarly, that fit alone cannot choose between them. More informative data may include serial target-cell counts, independent measures of effector abundance or activity, and observations before and during the transition to decline. Perturbations that selectively alter target supply or immune activity can be more discriminating than passive trajectories, provided their other effects are considered.

Several failure modes deserve attention. First, an apparent target-cell decrease may reflect sampling or changes in the measured cell population rather than depletion of the true susceptible pool. Second, the immune model may be weakly identifiable if \(k\) and \(E\) are observed only through their product \(kE\); viral data may constrain the total killing rate without determining its two components. Third, the assumed bilinear infection and killing terms may be inadequate if infection saturates or effector action depends on delays. Finally, parameter uncertainty in \(\beta,p,c,\delta\), or target-cell measurements shifts the inferred threshold. A threshold calculated from point estimates should not be treated as a precisely known biological boundary.

Common Mistake

• Wrong reasoning: “Viral load fell, so the virus must have depleted its target cells.” A decline is compatible with that explanation but also with increased infected-cell loss or other omitted processes.
• Correction: Compare the timing of viral decline with measured target-cell and immune trajectories. Use the model thresholds as conditional predictions, check units and assumptions, and describe the mechanism as consistent with the evidence rather than proven by viral load alone.