Abstract Background Pulmonary arterial hypertension (PAH) is a progressive and fatal disease and as such, it is imperative that patients remain on therapy to delay progression and maintain functional status. Previous research has demonstrated differences in overall survival in the real-world by endothelin receptor antagonists (ERA) treatments with a lower risk of death with macitentan than ambrisentan and bosentan. However, real-world data on persistence to ERAs are limited. Purpose To compare persistence to macitentan versus ambrisentan and bosentan) in patients with PAH. Methods This retrospective, non-interventional, observational study identified patients between 10/21/2013–3/31/2021 (study period) from the Decision Resources Group (DRG) Real World Data Repository. The repository links medical and prescription claims from commercial, Medicaid, and Medicare plans with electronic health records (EHRs) to provide longitudinal patient-level data representative of over 300 million patients in the US. Patients selected must have been dispensed an ERA prescription from 4/14/14–3/31/20 AND 1) either an inpatient diagnosis of pulmonary hypertension (PH), or 2) a right heart catheterization with any PH diagnosis, or 3) were taking a phosphodiesterase 5 inhibitor (PDE5i). The primary outcome was persistence defined by time from index to drug discontinuation (90-day gap after the end of the last prescription's days of supply). Descriptive analysis comprised means, standard deviations, medians, and interquartile ranges for continuous variables and frequencies and percentages for categorical variables. Three-sample proportion tests were conducted at baseline to detect differences between the three cohorts and identify potentially confounding variables. Hazard ratios (HR) for pairwise comparisons (macitentan vs ambrisentan; macitentan vs bosentan) were calculated using an inverse probability of treatment weighting (IPTW)-adjusted Cox proportional regression model, to adjust for potential confounders (see Table 1) of the probability of receiving a specific ERA. Data were analyzed by Kaplan-Meier method. Results A total of 4,526 patients were included, of whom 1,205 received macitentan, 2,592 received ambrisentan and 729 received bosentan. Study population characteristics were representative of a general PAH population (Table 1). Unadjusted Kaplan-Meier analysis showed that median persistence was shortest for ambrisentan (12.6 months) followed by bosentan (13.6 months) and greatest for macitentan (15.8 months; overall log-rank p-value: 0.061). Compared to macitentan, patients on other ERAs had a greater risk of discontinuation: ambrisentan (10%) and bosentan (18%) (Table 2). Conclusions Persistence to therapy was significantly greater for macitentan-treated patients than in those receiving ambrisentan or bosentan. Further research examining variability in key sub-groups as well as the impact of improved persistence on long-term outcomes is warranted. Funding Acknowledgement Type of funding sources: Private company. Main funding source(s): Janssen Scientific Affairs, LLC