Pulmonary hypertension (PH) is defined as a mean pulmonary arterial pressure (mPAP) $\ge$20 mmHg at rest, measured by right heart catheterization. Clinically, PH is classified into five groups based on pathophysiology and treatment strategy [1].

Group 1, or pulmonary arterial hypertension (PAH), is defined as pre-capillary PH that meets specific hemodynamic criteria and occurs in the absence of significant lung or thromboembolic disease. PAH include idiopathic PAH, hereditary PAH, drug- and toxin-induced PAH, disease-associated PAH, and PAH in patients with a long-term response to calcium channel blockers. PAH leads to right ventricular dysfunction, resulting in symptoms such as dyspnea and decreased exercise capacity. It is a life-threatening condition with increased mortality if not treated effectively [2].

The 2022 European Society of Cardiology/European Respiratory Society (ESC/ERS) guidelines lowered the diagnostic threshold for pulmonary hypertension from mPAP $\ge$25 mmHg, as defined in the 2015 ESC/ERS guidelines, to $\ge$20 mmHg and refined risk stratification from a three-tier to a four-tier model, allowing more granular assessment of parameters such as 6-minute walk distance (6MWD), World Health Organization functional class (WHO FC), and N-terminal pro–B-type natriuretic peptide (NT-proBNP) to guide treatment intensity [1, 3].

In South Korea, the 2020 Korean Society of Cardiology/Korean Academy of Tuberculosis and Respiratory Diseases (KSC/KATRD) pulmonary hypertension guideline is being updated to align with the 2022 ESC/ERS recommendations, emphasizing earlier detection and individualized escalation of PAH therapy within both international and domestic practice frameworks [4, 5].

PAH pharmacotherapy recommended in the 2022 ESC/ERS guidelines and the 2020 Korean PH guideline targets three main pathways: endothelin receptor antagonists (ERAs), the nitric oxide–cyclic guanosine monophosphate axis (phosphodiesterase-5 inhibitors [PDE5i] and soluble guanylate cyclase [sGC] stimulators), and prostacyclin (PC) signaling. These agents are used alone or in combination to promote vasodilation and inhibit vascular remodeling. The 2022 ESC/ERS guidelines recommend initial oral combination therapy with ERA + PDE5i for most low- to intermediate-risk patients, with consideration of rapid escalation to triple therapy in higher-risk groups [1].

Despite optimized ERA + PDE5i combination therapy, a substantial proportion of patients remain symptomatic or at intermediate–high risk, prompting consideration of additional pathway modulation.

Riociguat, an sGC stimulator, enhances cyclic guanosine monophosphate production and has demonstrated efficacy in patients with WHO FC II–III PAH and chronic thromboembolic pulmonary hypertension (CTEPH) [6], whereas selexipag, an oral IP prostacyclin receptor agonist, selectively augments prostacyclin signaling and is approved for WHO FC II–III PAH as mono- or combination therapy with ERA and/or PDE5i [7]. In clinical practice, these agents are often positioned differently in the treatment algorithm for patients already receiving ERA + PDE5i. Riociguat is typically considered as a “replacement” strategy for PDE5i non-responders, providing mechanistic substitution within the NO–sGC–cGMP pathway, whereas selexipag is used as an “add-on” to expand pathway coverage by activating the prostacyclin axis in partial responders. Thus, clinicians face a real-world decision: in a patient on ERA + PDE5i who remains inadequately controlled, should they switch from PDE5i to riociguat (replacement) or add selexipag on top of ERA + PDE5i (add-on), recognizing that these approaches differ in pathophysiology, pill burden, tolerability, and health system implications [4, 8].

The Korean Health Insurance system covers ERA drugs (ambrisentan, bosentan, and macitentan), PDE5is (sildenafil), and PCs (selexipag, treprostinil, and iloprost) for PAH treatment. Coverage begins with monotherapy and advances to dual therapy if treatment goals are not met, followed by triple therapy when necessary. Of the oral PCs, only selexipag is reimbursed [9]. Thus, introducing riociguat as an alternative pathway-targeted agent for patients treated with PDE5i + ERA is needed, which requires a comparative analysis of the efficacy and safety of riociguat replacement versus selexipag add-on therapy.

Comparing riociguat replacement and selexipag add-on therapies is essential to optimize therapeutic strategies for patients with PAH receiving PDE5i + ERA combination therapy. Randomized controlled trials (RCTs) have demonstrated the efficacy and safety of each drug independently; however, direct comparative studies are lacking. In this study, we aimed to address this gap by conducting an indirect comparison through a systematic review and meta-analysis, providing clinical evidence to guide optimal treatment strategies and support future updates to Korean guidelines.

We conducted an indirect treatment comparison (ITC) analysis through a systematic literature review and meta-analysis of RCTs to evaluate the efficacy of riociguat replacement versus selexipag add-on therapy in patients with PAH receiving PDE5i + ERA combination therapy. This analysis followed the recommendations of the Cochrane Handbook and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines (Supplementary Table 1). The systematic review protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO; CRD42024524391).

PAH is a rare disease with relatively low prevalence in South Korea. However, recent demographic changes and increased diversity in underlying conditions have contributed to a rising number of diagnoses. Between 2002 and 2018, PAH prevalence increased approximately 75-fold, and incidence increased 12-fold, highlighting improved diagnostic capabilities and a growing disease burden. Associated healthcare costs increase with the number of treated patients, underscoring the need for evidence-based, cost-effective therapeutic decisions [19].

In this study, we conducted an indirect comparison to evaluate the treatment effects of riociguat replacement and selexipag add-on therapy in patients with PAH receiving PDE5i + ERA combination therapy. Our indirect comparison relied on a single common comparator (ERA + PDE5i $\pm$ placebo) linking two treatment pairs (riociguat $\pm$ ERA vs. ERA $\pm$ PDE5i vs. selexipag $\pm$ ERA $\pm$ PDE5i). Although a network meta-analysis (NMA) could technically be applied, the Bucher’s method was considered more appropriate in this context because it is simpler, more transparent, and provides stable estimates for a small network with only a single common comparator [20]. We also evaluated the assumptions of similarity and exchangeability by comparing baseline characteristics across REPLACE and GRIPHON, including age, sex, PAH etiology, WHO functional class, 6MWD, NT-proBNP, background ERA + PDE5i use, and follow-up periods (24 weeks vs. 36 months, using 24–26-week data for GRIPHON where possible).

Although REPLACE enrolled only WHO FC III patients while GRIPHON included WHO FC II–III, the ranges of 6MWD and NT-proBNP were broadly comparable. The Coghlan post-hoc analysis, reporting outcomes in patients receiving ERA + PDE5i at baseline (100%), allowed an anchored comparison more closely aligned with REPLACE.

The phase II selexipag trial (Simonneau 2012) differed in having a small sample size, markedly higher NT-proBNP levels (likely influenced by outliers) [10], and a 17-week assessment. It was included in the primary analysis to preserve information but excluded in pre-specified sensitivity analyses.

The primary analysis showed that riociguat replacement and selexipag add-on therapies did not differ in terms of all assessed outcomes, including the composite endpoint of clinical worsening, 6MWD, NT-proBNP levels, WHO FC improvement rates, and overall and serious adverse event rates.

Despite the lack of statistical significance, riociguat replacement therapy showed a numerical trend toward greater efficacy in preventing clinical worsening than selexipag add-on therapy. However, the CIs were wide, reflecting the limited number of included studies, low event rates, and relatively small sample sizes. The HR for clinical worsening (0.167; 95% CI: 0.019–1.495) did not meet the non-inferiority margin of 1.185, which was calculated using the FDA fixed-margin method [21], suggesting that firm conclusions regarding non-inferiority cannot be drawn.

Our findings are consistent with those of Ornstová et al. [22], who reported no significant differences between riociguat and selexipag in an indirect comparison. However, their analysis was limited to WHO FC improvement and 6MWD as outcome measures, whereas we employed a composite endpoint of clinical worsening as the primary outcome. Traditionally, PAH clinical trials have relied on single clinical measures, such as 6MWD, as primary endpoints. Recently, however, there has been a shift toward composite endpoints reflecting clinical worsening or improvement [23]. Accordingly, we adopted a composite endpoint in our ITC to align with current trends and capture the overall clinical impact.

Our study included a more comprehensive set of RCTs and conducted sensitivity analyses focusing on patients receiving background ERA + PDE5i therapy and those classified as WHO FC III. These analyses are aligned with current PAH treatment guidelines, which suggest that patients receiving ERA + PDE5i therapy who experience clinical worsening may be considered for riociguat switching or selexipag add-on therapy [1]. We acknowledge that neither therapy demonstrated statistical superiority. However, the numerical trend toward a greater benefit with riociguat switching provides supportive information for clinical decision-making. In practice, this trend, combined with the mechanistic rationale (riociguat as a PDE5i replacement vs. selexipag as an add-on to augment the prostacyclin pathway) and patient-specific factors such as tolerability, pill burden, comorbidities, and cost, may help guide individualized therapeutic sequencing, particularly in patients who remain inadequately controlled on background ERA + PDE5i therapy.

Indirect comparisons have evolved from conservative approaches, which recommend their use only when direct RCTs are absent or compromised in internal/external validity [24, 25], to more flexible approaches that allow indirect comparisons when direct evidence is unavailable, as advocated by the National Institute for Health and Care Excellence (England) in 2013 [26] and the Canadian Agency for Drugs and Technologies in Health (Canada) in 2006 [27]. In 2016, the Pharmaceutical Benefits Advisory Committee (Australia) [28] emphasized rigorous verification of the exchangeability assumption and detailed reporting procedures. In Korea, the 2014 Health Insurance Review and Assessment Service guidelines similarly recommend indirect comparisons when direct RCT evidence is unavailable. This study met the fundamental prerequisites for indirect comparison—a systematic literature review and meta-analysis—to ensure similarity and consistency among the included studies [29]. Nevertheless, these results should be interpreted with caution due to the inability to directly validate consistency between indirect and direct comparisons.

This study has some limitations. REPLACE was an open-label study, which may have influenced symptom reporting or motivation; however, independent assessors and multiple objective endpoints were used to mitigate this risk. Furthermore, the RCTs included in this study were relatively few, resulting in wide CIs due to low event rates and varying follow-up durations. Differences in patient severity levels also present a limitation. Further direct comparative RCTs and long-term follow-up studies are needed to clarify the relative efficacy and safety of the two therapies, providing more robust evidence to inform optimal treatment strategies for patients with PAH.

Additional comparisons restricted to patients on baseline ERA + PDE5i and to those in WHO FC III likewise showed similar HR directions but wide confidence intervals that crossed unity, underscoring that no firm conclusion on superiority or inferiority can be drawn. Moreover, GRIPHON estimated HRs for clinical worsening using a proportional hazards model over 36 months, and our indirect comparison implicitly assumes that this HR is approximately constant over time when contrasted with the 24-week HR from REPLACE; if hazards were in fact time-dependent (e.g., early event clustering or delayed treatment effects), time-scale differences could bias the comparison. Finally, the absence of individual patient-level data prevented adjustment for important prognostic factors, dosing and adherence to background therapy, and potential confounding by indication. For example, patients switched to riociguat may have had more severe disease at baseline than those who remained on PDE5i and received selexipag add-on. Therefore, these limitations reinforce the need for adequately powered head-to-head RCTs.

In conclusion, this indirect comparison suggests that riociguat replacement and selexipag add-on therapy provide comparable clinical benefits for patients with PAH who remain inadequately managed with PDE5i+ERA therapy. Given the current limited evidence, therapeutic decisions should be individualized, considering patient characteristics, risk profiles, drug mechanisms, and reimbursement policies. Continued research is essential to build a more robust evidence base and optimize PAH management strategies.

ERA, Endothelin receptor antagonist; HR, Hazard ratio; ITC, Indirect treatment comparison; MD, Mean differences; PAH, Pulmonary arterial hypertension; PH, Pulmonary hypertension; PICO, Participants/Intervention/Comparator/Outcome; RCT, Randomized controlled trial; RR, Risk ratio.

All data used in this systematic review were extracted from publicly available, peer-reviewed articles. No new data were generated for this study. All included publications are cited in the manuscript, and the data are available within the cited articles.

J-EA: Methodology, Validation, Formal analysis, Data Curation, Writing - Original Draft, Visualization, Project administration; JHC: Conceptualization, Methodology, Software, Formal analysis, Writing - Original Draft, Visualization; MJK: Validation, Writing - Review & Editing; A-YL: Investigation, Writing - Review & Editing; GUH: Investigation, Writing - Review & Editing; W-JS: Investigation, Writing - Review & Editing; SHL: Data Curation, Supervision, Writing - Review & Editing; D-SK: Data Curation, Supervision, Writing - Review & Editing; SunYY: Conceptualization, Resources, Writing - Review & Editing; K-ML: Methodology, Resources, Writing - Review & Editing; Su-YY: Conceptualization, Resources, Data Curation, Writing - Review & Editing, Funding acquisition. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

Not applicable.

Not applicable.

This study received funding by Bayer. The funding source had no role in the study design, data collection and analysis, decision to publish, or manuscript preparation.

The authors declare no conflicts of interest. Sun-Young Yi and Kyung-Min Lee are employees of Bayer or have a financial relationship with the company. However, the company had no role in the handling or conduct of the study. The author(s) had full access to all data in the study and take full responsibility for the integrity of the data and the accuracy of the data analysis.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/RCM46524.