Diagnosing Priming-Limited Versus Effector-Limited Resistance

Why Separate Priming From Tumor-Site Effector Failure?

A small tumor-reactive T-cell population in a tumor can arise for different reasons. Few cells may have been primed to recognize tumor antigen in the first place. Alternatively, priming may have generated an adequate pool, but few cells may enter the tumor, or infiltrating cells may kill poorly. These mechanisms can produce similar tumor growth while implying different biological bottlenecks.

The purpose of this diagnosis is to locate the limiting stage in a chain: primed-cell supply, entry or retention in the tumor, and killing by cells at the tumor site. It is a mechanistic inference, not a direct observation of cause. The earlier material on locating checkpoint effects at priming or tumor sites motivates this separation; here the key task is to quantify the pattern across stages.

The saturating factor \(T/(H+T)\) represents reduced total killing at low target abundance when fewer targets are available, and approaches one as \(T\) becomes large relative to \(H\). The model is a compact accounting framework, not a claim that every tumor or T-cell population follows this exact function. Its value for diagnosis is that it keeps the candidate limiting stages visible.

Assumptions Before Interpreting the Pattern

A comparison across patients, time points, or model conditions is informative only if the quantities represent comparable biological states. The estimate of \(P\) should reflect tumor-reactive, primed cells rather than all circulating T cells. \(E_T\) should refer to viable tumor-infiltrating effectors, not simply cells in blood or cells adjacent to the tumor. The estimate of \(\kappa\) should concern killing per effector under conditions relevant to the tumor, not a result from a substantially different activation or target context.

The interpretation also assumes that differences in the saturation term have been accounted for. Tumors with different \(T\) can have different modeled killing rates even when \(P\), \(f\), and \(\kappa\) match. Comparisons should therefore use a common tumor-burden reference or explicitly calculate the target factor for each case. Antigen heterogeneity, target-cell resistance, regulatory suppression, myeloid suppression, and changing cell states can also alter the effective killing estimate or violate the simple model’s assumptions.

In practice, \(E_T/P\) is best treated as an approximate stage ratio, not a universal biological constant. It may be affected by sampling time, local T-cell expansion, survival, retention, and loss from the tumor. If \(P\) and \(E_T\) are measured at different times or with incompatible methods, their ratio should not be presented as a clean measure of tumor entry.

Decomposing the Killing Rate

Since \(E_T=fP\), the model can be written as the product of four factors: primed-cell abundance, the fraction represented among tumor infiltrates, per-cell killing, and target availability. This factorization gives a precise way to ask which stage is associated with a reduction in modeled killing.

For a reference condition, use subscript \(0\). Dividing the two expressions gives the relative killing rate:

This identity follows by substituting \(K=\kappa fP S(T)\) and \(K_0=\kappa_0 f_0P_0 S(T_0)\), then grouping each numerator with its corresponding reference value. If tumor burden and \(H\) are the same in both conditions, the saturation ratio is one. A reduction in \(K\) can then be attributed, within this model, to reductions in \(P\), \(f\), \(\kappa\), or some combination of them. When the tumor burdens differ, the saturation ratio must be considered before assigning the difference to T-cell biology.

A priming-limited pattern is low \(P\) relative to an appropriate reference, with no comparably large deficit in infiltration fraction or per-cell killing. An effector-limited pattern has a reasonably preserved primed pool but a low \(f\), a low \(\kappa\), or both. Low \(f\) points to an effector-side access, retention, or expansion limitation; low \(\kappa\) points to weak function among the cells that are present. These labels describe the dominant measured bottleneck, not mutually exclusive disease categories.

How to Make and Check the Diagnosis

A useful analysis first defines a reference condition, then estimates the stage quantities at comparable times. The reference might be a pretreatment state, a model control, or a biologically justified response condition. It should not be chosen simply because it makes one proposed mechanism appear large. The following sequence makes the logic explicit.

Diagnostics should include checking whether the inferred stage deficit persists across reasonable sampling times and whether it is robust to uncertainty in \(P\), \(E_T\), \(\kappa\), and \(H\). If a small change in an uncertain parameter reverses the classification, the result is not a firm mechanistic diagnosis. A large difference between predicted and observed tumor change is also a warning: growth, antigen escape, other immune populations, and spatial heterogeneity may be important outside this simplified killing expression.

Alternatives, Limitations, and a Common Pitfall

The factorized calculation is useful when stage-specific estimates are available. If only tumor size or tumor growth is observed, priming-limited and effector-limited explanations are generally not identifiable from that outcome alone: several combinations of \(P\), \(f\), and \(\kappa\) can yield the same \(K\). A richer mechanistic model may represent priming, trafficking, local expansion, exhaustion, suppression, antigen loss, and tumor growth as separate processes. Such a model can express more hypotheses, but it also needs more data and may have parameters that cannot be separately estimated. A targeted perturbation that changes priming or tumor-site function can strengthen localization, but a perturbation may affect multiple stages and must be interpreted accordingly.

The framework has further limits. The ratio \(f=E_T/P\) can be misleading when cells proliferate or die after entering the tumor, and an average \(\kappa\) can conceal a mixture of strongly and weakly functional cells. Measurements from one tumor region may not represent the whole lesion. In addition, tumors may change their antigen expression or susceptibility, altering killing without changing the number of effectors. The diagnosis is therefore best reported as “most consistent with” a limiting stage, together with the evidence and unresolved alternatives.

Common mistake: “The tumor contains few T cells, so priming must have failed.” This reasoning skips the distinction between the size of the primed pool and the fraction that reaches or persists in the tumor. If \(P\) is adequate but \(E_T/P\) is low, the evidence instead supports an effector-side infiltration limitation. Conversely, a low \(P\) does not by itself prove a priming defect if the measurement misses relevant cells or is taken at an uninformative time. Measure and compare the stages before naming the bottleneck.