Abstract
<title>Abstract</title> <p>Background Lurasidone hydrochloride, a Biopharmaceutics Classification System (BCS) Class II drug, exhibits dissolution-limited absorption and a pronounced food effect, making prediction of its in vivo performance challenging. This study developed a dissolution-informed physiologically based pharmacokinetic (PBPK) model integrated with biorelevant dissolution testing to establish a PBPK-assisted in vitro–in silico correlation (IVISc). Methods Dissolution studies of a 40 mg lurasidone tablet were performed in fasted- and fed-state simulated gastric and intestinal media (FaSSGF, FeSSGF, FaSSIF, and FeSSIF). Experimentally generated dissolution profiles were incorporated into a PBPK model developed using PK-Sim® to predict plasma concentration–time profiles, pharmacokinetic parameters, and food effects. Model performance was evaluated using fold error (FE), average absolute fold error (AAFE), goodness-of-fit analysis, and visual predictive checks. Results Dissolution was significantly influenced by gastrointestinal conditions (p < 0.001), with the highest drug release observed in FeSSGF (85.1%), followed by FeSSIF (78.8%), FaSSGF (61.8%), and FaSSIF (57.7%). PBPK simulations demonstrated the highest predicted systemic exposure in FeSSGF with a Cmax of 64.26 µmol/L and an AUCtend of 5138.05 µmol·min/L. Compared with FaSSGF, FeSSGF produced a 37.7% increase in Cmax and a 59.9% increase in AUCtend, while FeSSIF increased Cmax and AUCtend by 36.3% and 40.2%, respectively, relative to FaSSIF. The PBPK-assisted IVISc showed a strong linear relationship between dissolution extent and predicted exposure (R² = 0.9545). Model validation demonstrated excellent predictive performance with a mean FE of 1.04, AAFE of 1.04, and strong agreement between predicted and observed pharmacokinetic parameters (R² = 0.97 for Cmax and 0.98 for AUCtend). Sensitivity analysis identified solubility and CYP3A4-mediated clearance as the principal determinants of systemic exposure. In simulations of hepatic impairment, predicted Cmax increased from 64.2 to 77.0 µmol/L, while AUCtend increased from 5138.05 to 5395.20 µmol·min/L. Conclusions The dissolution-informed PBPK framework successfully translated biorelevant dissolution behavior into pharmacokinetic outcomes with high predictive accuracy. The integrated PBPK-assisted IVISc demonstrated robust prediction of food effects and systemic exposure, supporting its application in formulation optimization, special population assessment, virtual bioequivalence, and model-informed drug development for poorly water-soluble drugs.</p>