Contemporary risk scores predict clinical worsening in pulmonary arterial hypertension - An analysis of FREEDOM-EV

Document Type

Journal Article

Publication Date



The Journal of heart and lung transplantation : the official publication of the International Society for Heart Transplantation








combination therapy; long-term outcomes; oral treprostinil; pulmonary arterial hypertension; risk profile


BACKGROUND: Risk scores integrate clinical variables emphasizing symptoms, exercise capacity, and measures of cardiac strain to predict clinical outcome better than any single value in pulmonary arterial hypertension (PAH). Risk scores have demonstrated prognostic utility for outcomes in registries, and recent studies have suggested that they are also therapy-responsive in controlled trials. METHODS: FREEDOM-EV, a global, placebo-controlled, event-driven study, randomized 690 PAH participants 1:1 to oral treprostinil (TRE) or placebo. Clinical assessments were performed every 12 weeks to calculate the non-invasive French risk assessment (FRA), 4-strata COMPERA, REVEAL 2.0, and REVEAL Lite 2; median follow-up was 58 weeks. The Week 12 risk scores were used to predict time to clinical worsening (from Week 12) with Kaplan-Meier product-limit estimates. Log-rank test was used to calculate the statistical difference among risk categories, and mediation analysis tested the hypothesis that improvements in risk score contributed to reduced likelihood for clinical worsening. We assessed the previously proposed "net clinical benefit" (achievement of FRA low-risk status and absence of clinical worsening). RESULTS: Both REVEAL scores, COMPERA, and FRA at Week 12 predicted subsequent clinical worsening better than baseline risk. Mediation analysis demonstrated that Week 12 risk score reduction explained part of TRE's effect on clinical worsening, especially for those with higher baseline risk. TRE assigned participants were more likely to achieve the previously proposed "net clinical benefit" at Weeks 24 and beyond. Few participants who achieved 'net clinical benefit' had subsequent clinical worsening. CONCLUSIONS: Contemporary risk scores were therapy responsive in FREEDOM-EV and early improvements predicted subsequent outcomes. This post hoc analysis suggests that risk scores may be a surrogate for clinical worsening.