Exposure-Response describes the quantitative relationship between systemic drug exposure (e.g. concentration in plasma over time) and an observed pharmacological effect – including both desired efficacy and unwanted adverse effects. Exposure-response analyses are a core component of clinical pharmacology because they support dose-finding, benefit-risk assessment and the rational derivation of dosing recommendations.
What does “exposure” mean in practice?
“Exposure” does not refer to the administered dose itself but to the concentration actually achieved in the body, which is determined by pharmacokinetic parameters such as absorption, distribution, metabolism and elimination. Common exposure metrics include, for example, AUC (area under the concentration-time curve), Cmax (maximum concentration), Ctrough (trough level), or concentrations at defined time points.
For study planning and data interpretation, it is crucial that exposure can vary substantially between individuals, for example due to body weight, age, organ function, genetics, concomitant medication or adherence. A purely dose-response view therefore often provides an incomplete picture.
“Response”: efficacy and safety as endpoints
“Response” can be operationalised very differently depending on the indication and development phase: as a biomarker change, clinical endpoint, symptom score, time-to-event variable, or also as the occurrence of an adverse event. In oncology, for example, objective response or a progression event is relevant, whereas in other areas laboratory parameters or clinical scores are more prominent.
It is important that response endpoints are defined in the study protocol and the statistical analysis plan, and that measurement time points and collection methods are consistent. Exposure-response models are only as good as the underlying data quality and the clinical plausibility of the endpoint definition.
Typical models and analytical approaches
In practice, both exploratory and model-based approaches are used. Analyses often start with graphical analyses (e.g. response against AUC) and simple regression models to identify patterns. Depending on the data type, linear or non-linear models are used, such as Emax models for saturating effects, logistic models for binary endpoints, or time-to-event models in suitable designs.
For repeated measurements and heterogeneous patient populations, population PK/PD models are a common framework for accounting for covariates and quantifying variability. On this basis, simulations can be performed (e.g. for alternative dosing regimens) and dosing recommendations for subpopulations can be derived.
Role in dose-finding and lifecycle management
Exposure-response is particularly important in early phases (dose escalation and proof-of-concept), because this is where the “therapeutic range” is defined: what exposure is required for efficacy, and from what point does the risk of clinically relevant adverse effects increase? In later phases, the analyses support the selection of the final dose, the dosing regimen (e.g. once daily vs. twice daily) and, where applicable, the need for therapeutic drug monitoring.
A further practical benefit lies in justifying dose adjustments: if response does not further increase at higher exposure while adverse effects increase, this argues for a lower target exposure or tighter upper limits. Conversely, low exposure (e.g. due to interactions or absorption problems) can explain an apparently missing efficacy. Exposure-response analyses thus also become an instrument for root-cause analysis of unexpected study results.
For benefit-risk assessment, it is also relevant whether there are subgroups with systematically different exposure (e.g. patients with impaired renal function). Such findings often feed into the summary of product characteristics, for example as a concrete dosing recommendation or as an indication of the need for monitoring (e.g. laboratory parameters, clinical signs of toxicity).
Exposure-response analyses also remain relevant after authorisation, for example for dose adjustment in renal or hepatic impairment, for drug-drug interactions, or for paediatric extrapolation. They are thus a typical component of regulatory discussions and can gain importance in the context of Post-Authorisation Safety Studies or variation applications.
Operational implementation in the study context (sponsor/CRO)
Operationally, exposure-response requires robust PK sampling plans, correct sample logistics, validated bioanalysis and integrated data management (e.g. linking concentration data with clinical endpoints and timestamps). Common errors include inconsistent sampling time points, missing dosing times, incomplete concomitant medication data, or undocumented dose interruptions, which distort the exposure estimate.
For sponsor and CRO, it is also important to document the models, assumptions and sensitivity analyses in a traceable manner, as this content is often used in regulatory dossiers (e.g. Module 2/5 of a CTD/eCTD). Clear traceability from raw data to derived exposure metrics supports audit and inspection readiness.
FAQ
Why is exposure-response often more informative than a pure dose-response analysis?
Because the same dose can lead to very different concentrations in different individuals. Exposure-response accounts for this variability and links effect to the exposure actually achieved – making dosing decisions more robust as a result.
Which exposure metrics are most commonly used?
AUC and Cmax are very commonly used, supplemented by Ctrough for chronic therapy. Which metric is appropriate depends on the mechanism of action, the PK, and the expected time course of effect.
When should exposure-response analyses be planned in the development programme?
Ideally early, because the sampling design, bioanalysis and endpoint definitions determine later modelability. In practice, exposure-response questions are already addressed in Phase I/II and further developed in late phases for the final dose justification.
Regulatory references (selection): ICH E6(R3) Good Clinical Practice; Regulation (EU) 536/2014 (Clinical Trials Regulation); EMA Guideline on the investigation of drug interactions (for exposure assessment, depending on context).