SUMMARY

• In the GRIPHON phase 3 study, treatment-emergent adverse events (TEAEs) of pericardial effusion occurred in 1 out of 575 patients (0.2%) in the UPTRAVI group and in 2 out of 577 patients (0.3%) in the placebo group.¹ 
• SPHERE was a US-based, real-world, observational UPTRAVI drug registry study.^2,3 Among the total 829 enrolled patients, 759 were diagnosed with World Health Organization (WHO) Group 1 pulmonary arterial hypertension (PAH).² Four (0.5%) exhibited pericardial effusion on UPTRAVI.⁴ 
• In the TRITON phase 3b study that evaluated initial triple versus initial double oral therapy in newly diagnosed, treatment-naive patients with PAH, pericardial effusion was documented as a serious adverse event (SAE) in 2 of the 119 patients (1.68%) treated with UPTRAVI and 3 of the 127 patients (2.36%) receiving placebo.^5-7
• In the TRACE exploratory phase 4 study, a total of 108 patients were randomized to UPTRAVI (n=53) and to placebo (n=55).^8,9 The incidence of TEAEs related to pericardial effusion was documented in 1 patient (1.9%) within the UPTRAVI arm.¹⁰ 
• EXPOSURE is an ongoing, multicenter, prospective, observational study examining patients with PAH who are initiating PAH-specific therapies in Europe and Canada. At baseline, pericardial effusion was observed in 32 out of 178 patients (18%) within the PAH with Connective Tissue Disease (PAH-CTD) group and in 42 out of 362 patients (12%) within the Idiopathic PAH (IPAH) group.¹¹
• Zhao et al (2024)^12,13 was a retrospective study involving 192 patients with PAH across 10 hospitals in China. Patients with WHO Group 1 PAH, who were on dual therapy with endothelin receptor antagonists (ERA) and phosphodiesterase type 5 inhibitor (PDE5i)/riociguat for at least 3 months, received additional selexipag. Assessments at baseline and at 6 months showed that pericardial effusion was reported across all risk groups, with a decline over time from 10% to 6.7% in low-risk, 21.4% to 15.8% in intermediate-low, 50.6% to 31.9% in intermediate-high, and 43.6% to 23.1% in high-risk patients, respectively. Among different PAH etiologies, idiopathic PAH (IPAH), connective tissue disease-associated PAH (CTD-PAH), and congenital heart disease-associated PAH (CHD-PAH), the percentage of pericardial effusion also decreased at 6 months following selexipag initiation. Cox regression analysis indicated a hazard ratio of 1.353 (95% CI: 0.606–3.021) for the presence of pericardial effusion.
• An exploratory disproportionality analysis of the Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) database from 2004 to 2024 identified 20,057 AE reports potentially related to pericardial effusion involving 19,693 patients across all drugs, with UPTRAVI appearing among the top 50 drugs by report frequency but not by signal strength. Among 19,693 cases of pericardial effusion potentially associated with drug therapy in the FAERS database, the authors reported UPTRAVI was the primary suspect drug in 122 (0.62%) cases and calculated the reported odds ratio (ROR) [95% confidence interval (CI)]: 4.97 (4.16-5.94). The authors’ time-to-onset (TTO) analysis indicated a median onset of 14.5 days, with an interquartile range (IQR) of 4.25 to 157.75 days, based on 24 cases.¹⁴

BACKGROUND

Pericardial effusion can arise from various causes^15-17 and is observed in patients with PAH, with a mean pulmonary artery pressure (mPAP) >35 mmHg.^18,19 Prior to the first global approval of UPTRAVI in December 2015,²⁰ a review by Sahay et al (2013)¹⁹ of 10 published studies reported the prevalence of pericardial effusion in PAH ranging from 15% to 100%. Fenstad et al (2013)²¹ conducted a single-center cohort study from January 1995 to December 2005, involving 577 patients with PAH, found that pericardial effusion was present at diagnosis in 150 patients (26%), as observed on the index echocardiogram. Of the 427 patients with no effusion, pericardial effusion developed in 146 patients upon follow-up echocardiogram with a time to any pericardial effusion of 1107±1078 days.²¹ Despite its clinical significance, the mechanisms underlying pericardial effusion in PAH remain complex and poorly understood, with limited data on its prevalence and etiology.¹⁹ 

The 2022 ESC/ERS guidelines for the diagnosis and treatment of pulmonary hypertension classified pericardial effusion into 3 risk groups to estimate 1-year mortality, distinguishing low risk (no pericardial effusion), intermediate risk (minimal effusion), and high risk (moderate or large effusion). Additionally, pericardial effusion grading further explores right ventricular overload and has prognostic relevance in patients with PAH.²² 

CLINICAL DATA

GRIPHON

The GRIPHON phase 3 study was a multicenter, double-blind, parallel-group, placebo-controlled, event-driven study evaluating the efficacy and safety of oral UPTRAVI in adult patients with WHO Group 1 PAH. Patients with symptomatic PAH (N=1156) were randomized 1:1 to receive UPTRAVI (n=574) or placebo (n=582) twice daily. The median duration of study treatment was 70.7 weeks for patients who received UPTRAVI and 63.7 weeks for those who received placebo. Safety was assessed throughout the treatment period and for up to 7 or 30 days after the last intake of UPTRAVI or placebo.^23,24

Pericardial effusion was noted at baseline in 10 of 574 patients (1.7%) randomized to receive UPTRAVI and in 10 of 582 patients (1.7%) randomized to receive placebo.¹ Of the patients randomized to placebo, 4 did not receive study drug and were excluded from the safety analysis, and 1 received a single dose of 8 tablets of UPTRAVI and was assigned to the UPTRAVI group for the safety analysis.²⁴ During the double-blind (DB) period of GRIPHON study, TEAEs of pericardial effusion occurred in 1 out of 575 patients (0.2%) in the UPTRAVI group and in 2 out of 577 patients (0.3%) in the placebo group.^1,25

GRIPHON open label (OL) was a multicenter study to assess the long-term safety and tolerability of UPTRAVI in patients with PAH.²⁶ The OL safety analysis set included all randomized participants who received at least 1 dose of UPTRAVI or placebo in main GRIPHON study (AC-065A302; NCT01106014) and entered the GRIPHON OL study (AC-065A303; NCT01112306).²⁷ Among the 953 patients receiving UPTRAVI during the DB and/or OL periods, pericardial effusion was observed in 11 patients (1.2%). In the OL safety analysis set of 709 patients, 10 (1.4%) experienced pericardial effusion, including 1 (0.14%) case reported as a SAE.^27-29

SPHERE

SPHERE was a US-based, multicenter, prospective, real-world, observational UPTRAVI drug registry study that enrolled adult patients (n=829) aged ≥18 years from November 2016 to March 2020. The study aimed to provide real-world evidence regarding the use of UPTRAVI in patients with PAH over an 18-month follow-up period.^2,3 Among the total 829 patients, 759 were diagnosed with WHO Group 1 PAH.² Notably, 4 (0.5%) patients in the previously initiated group, defined as patients who enrolled more than 60 days after UPTRAVI initiation, exhibited pericardial effusion while on UPTRAVI.⁴ However, it remains unclear whether these pericardial effusions were present at baseline prior to initiating UPTRAVI or developed during treatment.

TRITON

TRITON was a multicenter, double-blind, randomized, placebo-controlled phase 3b study, that evaluated initial triple (macitentan, tadalafil, and UPTRAVI) versus initial double (macitentan, tadalafil, and placebo) oral therapy in newly diagnosed, treatment-naive patients with PAH. Patients were divided into 2 groups, randomized to either initial triple therapy (n=123) or initial double therapy (n=124) for 26 weeks.⁷ Among the 247 patients randomized, pericardial effusion was present at baseline (prior to the start of study drug therapy) in 3 patients (2.4%) from the triple therapy group and in 4 patients (3.2%) from the double therapy group.³⁰ Among the patients in the triple therapy group, 119 received UPTRAVI, while the 4 patients who did not receive UPTRAVI were allocated to the initial double therapy group for the safety analysis.^7,31 One patient randomized to initial double therapy did not receive any treatment and was excluded from the safety analysis set.⁷ Pericardial effusion was documented as a SAE in 2 of the 119 patients (1.68%) who received at least 1 dose of UPTRAVI and 3 of the 127 patients (2.36%) who received placebo.^5,6

TRACE

TRACE was a prospective, multicenter, randomized, placebo-controlled, double-blind exploratory phase 4 study with 2 parallel treatment groups: UPTRAVI (with stable background PAH therapy at enrollment) and placebo (with stable background PAH therapy at enrollment) to explore daily life physical activity level and self-reported symptoms and their impacts for 24 weeks. A total of 108 patients were randomized in a 1:1 ratio: 53 to UPTRAVI and 55 to placebo in both the full analysis set and safety analysis set.^8,9 The incidence of TEAEs related to pericardial effusion was documented in 1 patient (1.9%) within the UPTRAVI arm.¹⁰ 

EXPOSURE (Ongoing Study)

EXPOSURE is an ongoing, multicenter, prospective, observational study examining patients with PAH who are initiating PAH-specific therapies in Europe and Canada. This study began in September 2017¹¹ and is anticipated to conclude in June 2028.³² Eligible patients include individuals aged ≥18 years with WHO Group 1 PAH, who are initiating a new PAH-specific therapy either within 1 month of enrollment or at the time of enrollment.¹¹

The aim of this publication is to describe the characteristics, treatment patterns, tolerability, and outcomes of patients with PAH-CTD who are treated with UPTRAVI in clinical practice. Patients with IPAH were included for contextual comparisons. Pericardial effusion was present at baseline in 32 out of 178 patients (18%) within the PAH-CTD group and in 42 out of 362 patients (12%) within the IPAH group.¹¹

Analysis of a retrospective study

Zhao et al (2024)¹² was a multicenter, retrospective study involving 192 patients (mean age 43±14.7 years) with PAH across 10 hospitals in China. Patients with WHO Group 1 PAH, who were on dual therapy with endothelin receptor antagonists (ERA) and phosphodiesterase type 5 inhibitor (PDE5i)/riociguat for at least 3 months, received additional selexipag. In addition to other clinical assessments, echocardiography parameters were measured at the time of selexipag initiation (baseline) and after 6 months of treatment. Echocardiographic data (including presence of pericardial effusion) were obtained using standard views with commercially available devices.

Patients were stratified into four risk groups: low, intermediate-low, intermediate-high, and high. Pericardial effusion was reported in all groups. In the low-risk group, pericardial effusion was reported in 10% (n=22) of patients at baseline, decreasing to 6.7% at 6 months follow-up. The intermediate-low group reported pericardial effusion in 21.4% (n=35) at baseline, which declined to 15.8% at 6 months. In the intermediate-high group, pericardial effusion was reported in 50.6% (n=91) of patients at baseline, reducing to 31.9% at 6 months. The high-risk group reported pericardial effusion in 43.6% (n=44) at baseline, which decreased to 23.1% at 6 months.¹² 

Among patients with IPAH, pericardial effusion was reported at baseline in 35 out of 97 patients (36.1%) and 14 patients (14.4%) at 6 months follow-up. In CTD-PAH, pericardial effusion was reported in 18 out of 46 patients (39.1%) at baseline and 3 patients (6.5%) at 6 months. Among patients with CHD-PAH, pericardial effusion was reported in 11 out of 46 patients (23.9%) at baseline and 4 patients (8.7%) at 6 months.¹³ 

In the Cox regression analysis, the hazard ratio for the presence of pericardial effusion was reported as 1.353 (95% CI: 0.606–3.021).¹² 

Analyses of the FAERS database

An exploratory disproportionality analysis of the FAERS database from the first quarter (Q1) of 2004 to the second quarter (Q2) of 2024 identified a total of 20,057 AE reports potentially related to pericardial effusion involving 19,693 patients across all drugs. The FAERS analysis process involved initially identifying 1017 different primary suspect drugs associated with these cases. This list was then narrowed to the top 50 drugs by report frequency. Disproportionality analysis was subsequently performed to identify drugs with positive signals, independently of report volume. When examining these results, UPTRAVI was among the top 50 drugs based on report frequency; however, it did not rank within the top 50 when ranked by signal strength.¹⁴ 

Among 19,693 cases of pericardial effusion potentially associated with drug therapy in the FAERS database, the authors reported UPTRAVI was the primary suspect drug in 122 (0.62%) cases and calculated the reported odds ratio (ROR) [95% confidence interval (CI)]: 4.97 (4.16-5.94). The authors’ TTO analysis indicated a median onset of 14.5 days, with an IQR of 4.25 to 157.75 days, based on 24 cases. Supplementary data showed that for UPTRAVI, the TTO analysis ranged from 0 day to 640 days across 29 patients; however, 5 patients with pericardial effusion on day 0 were excluded, resulting in 24 patients with a TTO range of 1 to 640 days included in the analysis.¹⁴

Another real-world analysis of the FAERS database from Q1 2016 to Q1 2023 found that among patients taking UPTRAVI, 113 cases of pericardial effusion were reported out of a total of 9598 cases of AEs (ROR [95% CI]: 5.43 [4.51–6.54]).³³ 

Both articles list several limitations. Despite the body of literature (published prior to the approval of UPTRAVI in December 2015²⁰) describing the presence of pericardial effusion in the setting of PAH^19,21,34,35 the authors of these two FAERS analyses failed to account for potential confounding by indication.^36-38

Disclaimer: Results must be interpreted with caution as the FAERS database may receive reports containing incomplete, inaccurate, untimely, or unverified information. The FDA acknowledges limitations in analyzing data from the FAERS database. The incidence or prevalence of events cannot be determined solely from this reporting system, as it may be subject to under-reporting and lacks comprehensive data on the frequency of use. While consumers and healthcare professionals are encouraged to report AEs, the event may have been related to the underlying disease being treated, or caused by some other drug being taken concurrently, or occurred for other reasons. FAERS data comprise only one part of the FDA’s post-market surveillance data, and the information presented within the database does not establish a causal relationship between the drug product and the reported AEs.³⁹

Literature Search

A literature search of MEDLINE®, EMBASE®, BIOSIS Previews®, DERWENT® (and/or other resources, including internal/external databases) was conducted on 7 June 2025.