Theory 09/Pharmacology & Drug Action /Pharmacology/3 min
Tolerance Is a Prediction
The same dose in the same person can be safe at home and lethal in a new place. Pharmacokinetics cannot explain that.
Drug tolerance is a learned prediction of the drug's effect rather than receptor desensitisation: it is context-specific, so it should be abolished by changing the cues that predict administration, and overdose risk should track cue novelty rather than dose.
01The evidence
- Environmental specificity of tolerance has been documented as a robust phenomenon, with tolerance developing to a drug in the context in which it is given and failing to transfer to a novel context.Trends in Neurosciences
- Drug anticipation and its implications for the treatment of dependence have been reviewed as a conditioning phenomenon, with the conditioned response opposing the drug effect.Europe PMC
02The tension
Tolerance is taught as downregulation: receptors desensitise, enzymes induce, the dose-response curve shifts. That account is real but it cannot explain context specificity, and context specificity is the part that matters clinically, because it predicts overdose when the context changes.
03The theory
The body learns the contingency between cues and the drug's arrival, and prepares a compensatory response that opposes the drug. The preparation is timed by the cues, so it peaks when the drug is expected.
In the usual context, the compensatory response cancels part of the drug effect, which is what tolerance looks like. In a novel context, the preparation is absent and the full dose arrives unopposed, which is why a habitual user can die from a dose they normally tolerate.
This makes tolerance a prediction rather than a receptor state, and it makes cue novelty the risk factor. It predicts that the largest overdose risk should follow a change of setting rather than an increase in dose, and that re-exposure after a break should be dangerous because the cues are present but the preparation has decayed.
The clinical implication is that tolerance should be treated as a memory that can be extinguished or refreshed, and that dosing advice should include context stability, which no current guideline mentions.
04What this is not
Those mechanisms are real and dose-dependent. This adds a predictive layer that is cue-dependent, and predicts context effects that pharmacokinetics cannot produce.
05Predictions
- P1Tolerance should be abolished or reduced when administration cues change, at constant dose and pharmacokinetics.
- P2Overdose risk should be elevated after a context change more than after a dose increase.
- P3Extinguishing the conditioned cues should reduce tolerance without changing receptor density.
06What would refute it
Tolerance that transfers fully across contexts at fixed dose would leave conditioning as a minor modifier.
07What it would change
It would put context into prescribing guidance and explain a large share of overdose deaths that are currently attributed to potency.
08Evidence base
- Environmental specificity of tolerance — Trends in Neuroscienceshttps://www.sciencedirect.com/science/article/abs/pii/S0166223684801241
- Drug anticipation and the treatment of dependence — Europe PMChttps://europepmc.org/article/MED/3147377