B106-14 Population Pharmacokinetics Analysis of Treprostinil Using Data From Phase 1 and 2 Studies of Treprostinil Palmitil Inhalation Powder

Abstract Rationale Treprostinil palmitil inhalation powder (TPIP) is a dry powder investigational formulation of treprostinil palmitil, a prodrug of treprostinil, that enables prolonged lung exposure and long-acting vasodilation with once-daily dosing. Phase 2 trials demonstrated safety, tolerability, and efficacy of TPIP in pulmonary arterial hypertension (PAH) and pulmonary hypertension associated with interstitial lung disease (PH-ILD). This population pharmacokinetic (PK) analysis pooled study populations to simulate TPIP PK parameters. Methods Pooled studies (dose range: 80-1280 µg) included: phase 1 single-ascending dose (SAD) study of treprostinil palmitil in suspension (TPIS) in healthy volunteers; phase 1 SAD/multiple-ascending dose study of TPIP in healthy volunteers; two phase 2 studies of TPIP in patients with PAH or PH-ILD; and two ongoing phase 2 open-label extension studies of TPIP in patients with PAH or PH-ILD. Population PK models were assessed based on PK parameter estimates and their precision, effect of covariates, graphical examination of standard diagnostic and population analysis goodness-of-fit plots, graphical examination of agreement between observed and individual post-hoc predicted concentration-time data, and reduction in residual and interindividual variability. To confirm suitability for model-based simulations, prediction-corrected visual predictive checks (PC-VPC) were conducted. Results Data from 265 patients were included; the majority were female (62.3%) and white (65.3%), with mean (SD) age of 48.4 (16.6) years, body surface area of 1.79 (0.215) m2, and estimated glomerular filtration rate of 99.7 (21.2) mL/min/1.73 m2. Model estimated parameters had moderate to pronounced interindividual variability (range: 23% to 107%); residual variability was 31%. Goodness-of-fit plots showed a robust fit to the data, particularly for individual predictions. The PC-VPC plot demonstrated a robust fit of the predicted data to the actual results (Figure); no bias in PC-VPC plots were observed by study population (healthy, PAH, PH-ILD). The only significant covariate relationship was formulation (TPIS vs TPIP) on the slow absorption rate constant for treprostinil; however, the effect was not clinically meaningful. Post hoc analyses demonstrated that treprostinil exposure increased with dose (power model proportionality constants of 1.32 and 1.12 for maximum plasma concentration (Cmax) and area under the curve (AUC), respectively) in healthy, PAH, and PH-ILD study populations. Results of the covariate analysis indicate that clearance of treprostinil was not related to study population, renal function, or hepatic impairment category. Conclusions Population PK model-based simulations captured the observed data adequately, with no systematic differences in the healthy, PAH, and PH-ILD study populations evaluated. Results will inform phase 3 TPIP dose selection. This abstract is funded by: This study was funded by Insmed Incorporated (Bridgewater, NJ, USA). Analyses were conducted by Institute for Clinical Pharmacodynamics (Schenectady, NY, USA) and funded by Insmed Incorporated. All authors contributed to and approved the abstract; writing and editorial assistance were provided by Aarthi Gobinath, PhD, CMPP, of The Lockwood Group (Stamford, CT, USA), funded by Insmed Incorporated.