PBPK Modeling Investigation of IVIVE–Observed Pharmacokinetic Divergence for VU0467319: A Mechanistic Refinement Workflow for Low-Solubility Compounds

Abstract

Background/Objectives: VU0467319 (VU319), an M1 muscarinic positive allosteric modulator evaluated in Phase 1 studies, exhibits subproportional dose-exposure scaling typical of poorly soluble compounds. This study applies physiologically based pharmacokinetic (PBPK) modeling to investigate the sources of divergence between IVIVE-based predictions and observed single ascending dose (SAD) clinical pharmacokinetics. Methods: A whole-body PBPK model was developed in Simcyp Simulator V24 using in vitro ADME and physicochemical inputs. Clinical pharmacokinetic data were used for sequential, hypothesis-driven model refinement of distribution, clearance, and absorption parameters. Model performance was evaluated at 60–600 mg fasted using virtual populations demographically matched to their corresponding clinical cohort. Results: A Kp scalar of 0.125 recovered model Vss within the observed Vz/F bound; a 10.7-fold downward CLint correction (32.2 to 3 µL/min/mg) reconciled clearance; and a 1 µm particle radius recovered AUC0–∞ within 6% across 60–240 mg. The model met predefined performance thresholds across the four independent verification doses (geometric mean fold error 1.45 for AUC and 1.56 for Cmax; three of four doses within 2-fold), with progressive underprediction at the highest doses (≥400 mg). Supersaturation-ratio analysis indicated that, within the model, sustained supersaturation was unlikely to explain the high-dose underprediction. The refined model reproduced the observed sub-proportional dose-exposure relationship through 240 mg, but underpredicted exposure at 400 and 600 mg. Conclusions: Structured PBPK refinement identified dissolution-limited absorption as the most likely primary contributor to IVIVE–observed divergence for VU319, with secondary contributions from clearance and distribution. These model-derived mechanisms remain hypotheses that require experimental confirmation. The workflow may provide a structured refinement framework for other low-solubility compounds where IVIVE assumptions diverge from observed pharmacokinetics.