Bayesian optimization with amortized search-space reparametrization for precision dosing

Personalizing drug dosing is a problem of sample-efficient optimization: candidate regimens are tried one at a time on a patient, and evaluations are expensive and potentially risky. Here, we use a pharmacokinetic/pharmacodynamic (PK/PD) model of the drug, a mechanistic model calibrated on population data with quantified parameter uncertainty, to reparameterize this search. A dosing network is trained offline against the model, amortizing the map from patient PK/PD parameters to the regimen predicted to be optimal, and online Bayesian optimization then searches over this learned, low-dimensional representation rather than over raw dosages. Because the network's image is a subspace of pharmacologically coherent regimens, the search remains sensible by construction, improving safety and convergence speed. Unlike model-informed precision dosing, our method admits heterogeneous feedback (symptom scores, preferences, wearables), and applies to any drugs for chronic disease with a PK/PD model where no real-time biomarkers can be measured that provide a comprehensive measure of patients’ state, and disease management implies taking into account patient reported outcomes. On an in-silico population of Parkinson's patients under Levodopa treatment, it converges to better regimens with fewer trials than baselines optimization methods.