Integrating Static and Dynamic Models to Predict Cannabidiol-Mediated Metabolic Drug-Drug Interactions.

Al Sahlawi S.; Eltanameli B.; Cicali B.; Cristofoletti R. · Clinical pharmacology and therapeutics · 2026

**Background & Methods** This review implements a stepwise, model-informed framework to predict cannabidiol (CBD)-mediated metabolic drug-drug interactions (DDIs) by integrating basic models, mechanistic static models (MSM), and physiologically based pharmacokinetic (PBPK) models. The analysis evaluates in vitro inhibition parameters for CBD and its primary metabolite 7-hydroxycannabidiol (7-OH-CBD) against multiple cytochrome P450 enzymes, with PBPK model validation against clinical pharmacokinetic data and simulation of DDI scenarios in special populations and common co-prescribed medications. **Key Findings** • **Differential DDI Predictions Across Models**: The mechanistic static model predicted substantial increases in sensitive substrate exposure (3.6-fold for CYP1A2, 4.1-fold for CYP2C9, 2.2-fold for CYP3A4 substrates) and strong CYP2C19-mediated interactions, whereas the validated PBPK model predicted lower, clinically-relevant magnitudes consistent with observed clinical data. • **CYP2C19 as Primary Liability**: The PBPK model identified CYP2C19 inhibition as the most clinically relevant DDI concern for CBD, with moderate interaction risk for sensitive CYP2C19 substrates, while CYP1A2 inhibition produced twofold increases in sensitive substrate exposure. • **Framework Reconciliation**: The integrated approach successfully reconciled disparities between in vitro-derived inhibition parameters and clinical observations, enabling population-specific DDI risk assessment across diverse clinical scenarios and special populations. **Dosage & Administration** Not reported. **Safety & Adverse Effects** Not reported. **Evidence Quality** This review represents a modeling-based mechanistic analysis rather than a prospective clinical trial, limiting direct clinical validation. Strengths include systematic integration of multiple complementary modeling approaches and validation against clinical pharmacokinetic data. Key limitations include: (1) reliance on in vitro inhibition data that may not fully capture complex in vivo pharmacodynamics; (2) potential parameter refinement bias when fitting PBPK models to clinical observations; (3) exclusion of non-metabolic interaction mechanisms (transporter inhibition, induction); and (4) applicability constraints to populations beyond those represented in validation datasets. The framework's predictive utility for untested drug combinations remains unvalidated. Evidence quality is moderate, appropriate for hypothesis-generation and clinical guidance development but requiring prospective clinical DDI studies for specific drug pairs to confirm predictions.

More recent cannabis research