Method Profile
Link pulmonary vascular signaling, resistance and walking-distance outcomes to simulate trial designs in a weighted virtual population.
Explore how disease severity, background therapies, dosing and hypothetical enrichment affect oral treprostinil trial outcomes.
Endothelial/smooth-muscle signaling and contraction/proliferation model; resistance-to-six-minute-walk-distance regression with measurement noise and a placebo term.
5,000 sampled parameter sets filtered to 4,909 plausible virtual patients; prevalence weighting using five trials; repeated sampling of trial-specific populations and dose trajectories.
Calibration and test information
FREEDOM-M, PHIRST, ARIES, FREEDOM-C and FREEDOM-C2 for prevalence weighting; later FREEDOM-EV response comparison. FREEDOM-EV baseline data also informed the resistance-to-walk-distance regression.
Virtual-trial variables
Functional-class composition, background PDE5 inhibitor/endothelin antagonist treatment, trial size, duration, time since diagnosis and oral treprostinil titration.
Optimize separate functional-class prevalence weights, then sample 100 virtual trials per fitted/test comparison. A separate enrichment analysis sampled 1,000 trial realizations per scenario; keep simulated latent biomarkers and significance probabilities distinct from clinical observations.
Five-trial calibration followed by a FREEDOM-EV trial-workflow test and hypothetical enrichment scenarios.
4,909 plausible virtual parameter sets after filtering 5,000; 100 realizations per fitted/test comparison; 1,000 realizations per enrichment scenario. These are not recruited participant counts.
Published trial treatment effects and statistical significance; original versus enriched trial designs.
FREEDOM-M median simulated effect +25 m versus published +23 m at 12 weeks. FREEDOM-EV simulation predicted 99% significance probability at week 24 although observed significance began at week 36. Combined enrichment/reduced-dual-therapy scenarios raised success fractions from 23%/22% to 81%/75% at trial sizes 350/310.
A United Therapeutics systems-pharmacology model linked pulmonary vascular biology to simulated six-minute walk responses. It reproduced several oral treprostinil trial trends, but underestimated later variability and predicted a significant response earlier than the FREEDOM-EV trial observed.
