Risk assessment models and survival in pulmonary arterial hypertension: a SPAHR analysis.

Background: Multicomponent improvement (MCI) is a novel endpoint for predicting survival in patients with pulmonary arterial hypertension (PAH), included in the sotatercept clinical program. For the first time, we investigated the prognostic value of MCI, ESC/ERS 4-strata risk (4SR) assessment, and the non-invasive French risk stratification score (FRS), for predicting survival in PAH patients in Sweden. All risk prediction models are based on three parameters: WHO-FC (World Health Organization Functional Class), NT-proBNP, and 6MWD (6-minute walk distance).

Methods: Data from the Swedish PAH & CTEPH Registry (SPAHR) collected 2008-2021 were used for the analyses. The association of MCI achievement, 4SR, and FRS calculated at 6 months (6M), with transplant-free (TF) survival was investigated in the whole cohort, as well as categorized by age (<65 and ≥65 years). All risk prediction models are based on three parameters: WHO-FC (World Health Organization Function Class), NT-proBNP, and 6MWD (6-minute walk distance). Kaplan-Meier estimate/Log-Rank test and Cox proportional model were used for survival analyses.

Results: The study included 411 patients (70% women) with a median [IQR] age of 66y.²¹ At 6M, the mean (SD) NT-proBNP decrease was 808 (603) and the mean 6MWD increase was 44 (11) meters. Median survival/follow-up time was 3.5y [1.7, 5.4]. After adjustment for sex and comorbidities, achievement of MCI independently predicted TF-survival; one MCI-criterion met (HR 0.65; CI 0.46-0.92, p=0.015); two MCI-criteria met (HR 0.45; CI 0.31-0.66, p<0.001); all three MCI-criteria met (HR 0.32; CI 0.21-0.52, p<0.001). Likewise, 4SR and FRS demonstrated a strong association with TF-survival with patients achieving lower risk scores exhibiting longer survival compared to those with higher risk scores. Patients ≥65Y more often had connective tissue disease-associated PAH, lower DLCO, more pronounced comorbidity burden, higher risk at baseline, less improvement during follow-up, and worse TF-survival then patients <65Y.

Conclusion: All models were found to have prognostic relevance for TF-survival. Risk prediction was incremental with the number of low-risk criteria met, while improvements in only one of 6MWD, NT-proBNP, or FC showed a modest association with survival. The risk assessment tools predicted outcome in patients across both age categories.