Clinical research on pulmonary hypertension from a 2025 perspective: a narrative review

Abstract

Background and Objective

Pulmonary hypertension (PH) is a clinical syndrome characterized by progressive elevation of pulmonary vascular resistance (PVR) and subsequent right ventricular (RV) failure. The research of PH has undergone unprecedented transformation in 2025, marked by paradigm-shifting therapeutic advances, refined diagnostic criteria, and sophisticated precision medicine approaches. This narrative review aims to synthesize the recent evidence from pivotal clinical trials, epidemiological studies, and translational research that collectively redefine the management of this complex disease spectrum, providing clinicians and researchers with an overview of current developments and future directions.

Methods

The literature search was conducted in PubMed with particular focus on publications from January 2025 to December 2025. Inclusion criteria encompassed clinical trials, observational studies, registry analyses, and research articles reporting novel diagnostic approaches, biomarkers, therapeutic interventions, and management strategies for PH. Studies were selected based on their relevance to clinical practice and scientific significance. The selection process was conducted independently, with consensus reached through discussion for any discrepancies.

Key Content and Findings

The 2025 clinical research on PH evolved towards precision assessment driven by novel biomarkers and artificial intelligence (AI), enhanced imaging [echocardiography, computed tomography (CT), magnetic resonance imaging (MRI)] for improved detection, phenotyping, and risk stratification. Therapeutically, the treatment focus has shifted beyond vasodilation to attenuating pathological remodeling. Most notably, sotatercept, its validation in reduce clinical worsening expanded across diverse patient populations. Alongside, optimized use of classical pathways (endothelin, nitric oxide, prostacyclin) and emerging targeted therapies (e.g., seralutinib, imatinib) may bring more clinical benefits. Interventional therapies have been optimized through large-scale registries and patient-specific factors, while non-pharmacological strategies have gained recognition as essential components of multidisciplinary management. Furthermore, right ventricular targeted therapies have received increasing attention and improved outcomes for patients.

Conclusions

The convergence of sophisticated risk stratification tools and mechanism-based therapeutics in 2025 established a robust foundation for improving survival and quality of life in PH patients.

Introduction

Pulmonary hypertension (PH) comprises a heterogeneous group of disorders characterized by elevated pulmonary arterial pressure, resulting in progressive right ventricular (RV) dysfunction and even death (1,2). The field has evolved dramatically over the past two decades, transitioning from limited therapeutic options to a sophisticated array of targeted interventions addressing diverse pathophysiological pathways. The year 2025 represented a watershed moment in PH research, characterized by the convergence of molecular insights, technological innovations, and rigorous clinical validation that collectively transform patient care.

Traditional therapeutic approaches targeting prostacyclin, endothelin, and nitric oxide (NO) pathways have provided significant symptomatic relief and delayed disease progression (3). However, these therapies primarily address vasomotor dysfunction without directly targeting the underlying vascular remodeling that drives clinical deterioration (4,5). The recognition of this limitation has catalyzed research into novel mechanisms and therapeutic targets, culminating in breakthrough developments that promise to alter the natural history of the disease (6).

This review examines the clinical research landscape of PH from a 2025 perspective, synthesizing evidence across epidemiology, diagnostics, therapeutics, interventional strategies, and special populations. The integration of these advances provides a roadmap for precision medicine approaches that individualize treatment based on molecular phenotypes, genetic profiles, and risk assessment model. We present this article in accordance with the Narrative Review reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0340/rc).

Methods

This narrative review was conducted to synthesize clinical research on PH published in 2025. A literature search of the PubMed database was performed on January 8, 2026, covering publications from January to December 2025. The search strategy combined free text terms—including “Pulmonary Hypertension”, “Clinical Trial”, “Observational Study”, “Treatment”, “Therapy”, “Diagnosis”, “Management”, “Artificial Intelligence”, “Machine Learning”, “Biomarker”, “Surgery”, “Intervention” and “Right ventricle”. Included articles were clinical trials, observational studies, translational research, and reviews reporting novel diagnostic approaches, biomarkers, or therapeutic interventions for PH. Only English-language publications with available full texts were considered. Non-English articles, conference abstracts, editorials, and veterinary studies were excluded. Two authors (Q.B. and H.L.) independently screened and selected studies, with disagreements resolved by a third reviewer (T.W.). Study quality was assessed based on design, methodology, sample size, and clinical relevance. The search strategy is summarized in Table 1.

Table 1. The search strategy summary.

Items	Specification
Date of search	January 8, 2026
Databases searched	PubMed
Search terms used	“Pulmonary Hypertension”, “Clinical Trial”, “Observational Study”, “Treatment”, “Therapy”, “Diagnosis”, “Management”, “Artificial Intelligence”, “Machine Learning”, “Biomarker”, “Surgery”, “Intervention”, “Right ventricle”
Timeframe	January 2025 to December 2025
Inclusion and exclusion criteria	Inclusion: clinical trials, observational studies (cohort studies, registry analyses), translational research, narrative and systematic reviews. Studies reporting novel diagnostic approaches, biomarkers, therapeutic interventions, interventional therapies, and management strategies for PH. English language publications
	Exclusion: non-English language, conference abstracts without full-text availability, veterinary applications, editorials, insufficiently relevant to the review objectives
Selection process	Literature search and initial screening were conducted by all authors. Full-text review and final selection were performed independently by two authors (Q.B. and H.L.). Discrepancies were resolved through discussion with a third author (T.W.) until consensus was reached. Study quality was assessed based on study design, sample size, methodology, and relevance to clinical practice

PH, pulmonary hypertension.

Optimization of diagnostic criteria, risk stratification tools, and precision assessment in PH

Precision medicine demands accurate diagnosis and risk stratification. Emphasizing earlier detection and precise phenotyping, these evolving criteria—complemented by novel biomarkers, multimodal imaging, exercise testing, invasive hemodynamics and artificial intelligence (AI)-driven analytics—collectively enhance detection, classification, and risk-stratification of PH patients.

Innovation in biomarkers

The quest for reliable, non-invasive biomarkers has expanded dramatically beyond traditional natriuretic peptides. Recent studies have identified novel candidates that reflect underlying pathophysiological processes, offering improved diagnostic specificity and prognostic value. Biomarkers serve as non-invasive tools that reflect underlying pathophysiological processes and disease status in the diagnosis and prognostic assessment of PH. Studies in 2025 have not only identified the roles of several novel biomarkers but also observed that certain previously overlooked indicators may be applicable, thereby offering more options for PH diagnosis.

Novel biomarkers

To date, only brain natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) remain the biomarkers routinely used in clinical assessment of PH according to the 2022 European Society of Cardiology (ESC)/European Respiratory Society (ERS) risk-assessment model (7). In addition to natriuretic peptides, recent studies have clinically evaluated other biomarkers that have been reported to be involved in the pathological mechanisms of PH. Incorporating new biomarker indicators into existing risk models can enhance their predictive ability. For example, growth differentiation factor-15 (GDF-15), a cytokine associated with stress response, was shown to improve discrimination for mortality prediction and model fit when integrated into the Comparative, Prospective Registry of Newly Initiated Therapies for Pulmonary Hypertension (COMPERA) risk models (8). Novel biomarkers may also distinguish specific PH subtypes and hemodynamic phenotypes. A study demonstrated that plasma GDF-15 levels were specifically elevated in pulmonary veno-occlusive disease (PVOD), a rare and severe subtype of pulmonary arterial hypertension (PAH), enabling differentiation of PVOD from other PAH subtypes via GDF-15 (9). Bone morphogenetic protein 10 (BMP10) was correlated with pulmonary vascular resistance (PVR) but not with pulmonary artery wedge pressure (PAWP), reflecting pre-capillary hemodynamic components, in contrast to NT-proBNP, which was associated with both PVR and PAWP (10).

Non-coding RNAs are merging as noninvasive biomarkers. For idiopathic PAH diagnosis, transfer RNA (tRNA)-derived small RNAs like i-tRF-15:31-Lys-CTT-1 demonstrated diagnostic accuracy with an area under the curve (AUC) of 0.90 and independent prognostic value (11).

Genetic screening

The pathogenesis of PAH is associated with specific genetic variants, and consequently, genetic screening has achieved clinical application. The prevalence of pathogenic variants in Chinese pediatric PAH patients was 64%, predominantly in bone morphogenetic protein receptor 2 (BMPR2), activin A receptor-like type 1 (ACVRL1), and T-Box factor 4 (TBX4) genes. Carriers exhibited higher baseline risk, greater treatment deterioration, and significantly worse survival, establishing genetic profiling as a critical risk stratification tool (12). Notably, rare variants in PAH-related genes such as BMPR2 were identified in 7.4% of patients with systemic lupus erythematosus (SLE)-associated PAH and were associated with a distinct vasculopathy phenotype and a significantly increased risk of mortality (13).

Blood tests and inflammation indices

Blood tests in the PH diagnostic encompass hematologic, metabolic, renal, hepatic, iron status, cardiac and immune markers to identify comorbidities, underlying causes, complications of PH (7). However, the value of some of these markers has been previously overlooked, and their potential is now being increasingly recognized.

Regarding platelet parameters, accumulating evidence showed that platelet count and function were altered in PH. In patients with Group 1, 3, and 4 PH, platelet counts were decreased, whereas platelet volume and D-dimer levels were elevated—changes that were not only attributable to thrombosis (14). Furthermore, another study found that thrombocytopenia (<150,000/µL) portended worse survival (15).

As for metabolic parameters, the metabolic score for insulin resistance (METS-IR), previously established as an effective surrogate for insulin resistance, was first validated by Gao et al. to be elevated in intermediate- to high-risk PH patients and could serve as an independent predictor of clinical worsening (16).

Even for the well-established biomarker NT-proBNP, novel ratio such as the NT-proBNP-to-albumin ratio (NTAR) was shown to outperform NT-proBNP alone in predicting extended hospitalization (17).

Inflammation plays an important role in the pathophysiological processes of PH. While inflammatory cytokines such as C-reactive protein (CRP) and interleukin-6 were associated with mortality in PH (18), simpler and more cost-effective alternative, inflammation-based hematologic indices derived from routine blood tests have gained attention. For instance, Iancu et al. reported that neutrophil-percentage-to-albumin ratio (NPAR) systemic immune-inflammation index (SII) and neutrophil-to-lymphocyte ratio (NLR) could effectively predict in-hospital mortality in PAH or chronic thromboembolic PH (CTEPH) (19).

Emerging technologies: omics and point-of-care innovations

Compared with conventional laboratory tests, omics data such as proteomics and RNA transcriptomics provide high-throughput biological information that helps delineate the systematic pathophysiological profile of patients with PH, and these molecular profiles may also serve as sources of novel biomarkers. For instance, using machine learning (ML) algorithms, researchers identified an immunerelated gene panel from the Gene Expression Omnibus (GEO) database that achieved diagnostic prediction performance with an AUC ≥0.900 in peripheral blood mononuclear cells for diagnosing PAH (20). Furthermore, a plasma proteome study encompassing 470 PH samples revealed four molecular clusters with distinct survival rates, characterized by upregulation of the plateletderived growth factor (PDGF) and transforming growth factor-β (TGF-β) pathways, thereby demonstrating theranostic potential (21). Another metabolomic study showed that fatty acid metabolites, particularly acylcarnitines, correlated with RV-pulmonary arterial (RV-PA) uncoupling and predicted hemodynamic response to imatinib therapy (22).

In addition, emerging clinical laboratory technologies enable physicians to obtain rapidly fluctuating physiological indicators. For example, point-of-care NT‑proBNP testing demonstrated near-equivalence to laboratory standards, with 92% concordance to COMPERA 2.0 thresholds, facilitating realtime risk stratification (23) (Table 2).

Table 2. Biomarkers innovations.

Category	Biomarker/omics approach	Key findings/performance	Clinical application
Novel biomarkers	GDF15 (8,9)	COMPERA model C-index increased after incorporating GDF15; distinguished PVOD from PAH (AUC ≥94%)	Improved predictive efficacy; served as diagnostic and prognostic biomarker
Novel biomarkers	BMP10 (10)	Specifically correlated with pre-capillary PH components	Improved diagnostic specificity for pre-capillary PH
Novel biomarkers	tRNA-derived small RNA (11)	i-tRF-15:31-Lys-CTT-1 expression reduced in PAH, with diagnostic AUC 0.90 (discovery) and 0.81 (validation)	Promising non-coding RNA biomarker for PAH diagnosis and prognosis
Genetic screening	BMPR2, ACVRL1, TBX4 variants (12)	64% prevalence in Chinese pediatric PAH; higher baseline risk and worse survival in carriers	Enabled genetic testing for risk stratification and counseling
Genetic screening	Rare PAH-related gene variants (13)	7.2% rare variants in SLE-PAH increased mortality 5.89-fold
Blood tests	Platelet indices (14,15)	Decreased platelet counts with increased size and D-dimer; thrombocytopenia (<150,000/µL) predicted worse survival	The characteristics of PH platelets provided simple prognostic indicators reflecting disease severity
Metabolic marker	METS-IR (16)	Predicted CTEPH clinical worsening; improved COMPERA 2.0	Enhanced risk stratification in CTEPH
Inflammatory indices	Inflammatory hematologic indices (19)	NPAR and SII were associated with in-hospital mortality; NLR correlated with CTEPH survival	Practical, non-invasive tools to predict PH hospitalization duration and mortality, allowing dynamic disease activity monitoring
Omics	Plasma proteome (21)	Identified 4 molecular clusters with distinct survival rates	Enabled molecular classification for targeted therapy; revealed pathological differences
Omics	Metabolomics (22)	Correlated with RV-PA coupling; predicted hemodynamic response to imatinib	Predicted hemodynamic response to Imatinib therapy
Omics	Immune-related gene panel (20)	Achieved AUC ≥0.969 (lung tissue) and 0.900 (peripheral blood)	Machine learning-based immune gene combination for precise PAH diagnosis and molecular classification
Classical biomarkers (enhanced)	NT-proBNP (17,23)	Point-of-care NT-proBNP correctly classified 92% cases; NTAR predicted in-hospital and 3-month mortality with AUC of 0.817	Novel composite markers and point-of-care testing improved PH risk stratification precision over traditional single biomarkers

ACVRL1, A receptor-like type 1; AUC, area under the curve; BMP10, bone morphogenetic protein 10; BMPR2, bone morphogenetic protein receptor 2; COMPERA, Comparative, Prospective Registry of Newly Initiated Therapies for Pulmonary Hypertension; CTEPH, chronic thromboembolic pulmonary hypertension; GDF15, growth differentiation factor-15; METS-IR, Metabolic Score for Insulin Resistance; NLR, neutrophil-to-lymphocyte ratio; NPAR, neutrophil-percentage-to-albumin ratio; NT-proBNP, N-terminal pro-B-type natriuretic peptide; NTAR, NT-proBNP-to-albumin; PAH, pulmonary arterial hypertension; PH, pulmonary hypertension; PVOD, pulmonary veno-occlusive disease; RV-PA, right ventricular-pulmonary artery; SII, systemic immune-inflammation index; SLE, systemic lupus erythematosus; SSc, systemic sclerosis; TBX4, T-Box factor; tRNA, transfer RNA.

Imaging advances and AI applications

Imaging modalities such as echocardiography, computed tomography (CT), and magnetic resonance (MR) are integral to the screening, definitive diagnosis, and prognostic assessment of PH. These techniques are continually evolving toward enhanced precision, non-invasiveness, and intelligent analysis. The imaging advances observed in 2025 encompass not only refined evaluations of conventional metrics but also the emergence of novel technological applications.

Within this context, AI algorithms—particularly ML and deep learning, serve as ideal enablers of this transition. Unlike conventional statistical methods, ML algorithms are capable of identifying nonlinear patterns and associations within large-scale, high-dimensional, and complex datasets that would otherwise remain undetectable (24).

Echocardiography

Echocardiography is used in the diagnostic workup of PH for screening, etiological differentiation, and assessment of RV function. RV-PA uncoupling is recognized as a marker of impending decompensated RV dysfunction and is associated with higher mortality in PH patients (25,26). Assessment of RV-PA coupling by the tricuspid annular plane systolic excursion (TAPSE) to systolic pulmonary artery pressure (sPAP) ratio enhanced the prognostic performance of ESC/ERS risk stratification model and the Registry to Evaluate Early and Long-Term PAH Disease Management (REVEAL) 2.0 risk score (27). Moreover, incorporating echocardiographic parameters of RV function, including RV-PA coupling, into the COMPERA and REVEAL Lite 2.0 risk models yielded a 10% increase in the AUC (28).

However, traditional echocardiography diagnosis relies on operator skill and subjective judgment, presenting significant limitations and time-consuming. In response, AI is transforming echocardiographic assessment by automating measurements. For example, an automated echocardiography evaluation workflow developed using Us2.ai software, attained an intraclass correlation coefficient of 0.94 for measuring tricuspid regurgitation jet velocity (TRJV) (29). In terms of prognostic assessment, clustering analysis based on echocardiographic parameters enables effective risk stratification for all-cause mortality, with prognostic value independent of traditional risk scores and invasive hemodynamic measures (28). Beyond automation, ML further strengthens the association between echocardiographic parameters and the PH diagnosis as well as adverse prognosis. For instance, a multivariable predictive model based on pre-operative echocardiography informs was proved to effectively predict peri-operative cardiorespiratory adverse events in children with idiopathic PAH undergoing cardiac catheterization (30).

CT and positron emission tomography/CT (PET/CT)

CT provides detailed anatomical imaging of pulmonary and cardiac structures, and when combined with AI/ML applications, this imaging data can be transformed into automated, precise analysis that is reshaping PH management (31). Particularly in CTEPH, CT enables detailed assessment of vascular morphology and diagnostic confirmation. For instance, automated chest CT analysis method was developed that partitioned each lung into isovolumetric segments and calculated coefficients of variation across segments to evaluate vascular heterogeneity. Using this approach, patients with CTEPH demonstrated pronounced pulmonary vascular heterogeneity and central redistribution of perfusion (32). Besides, Suchanek et al. developed a fully automated Bayesian analysis for CT pulmonary angiography to quantify the severity and distribution of perfusion change (33). Complementing this, AI-driven segmentation technology was also employed to identify morphological differences in pulmonary artery tortuosity between CTEPH and chronic thromboembolic disease (CTED), demonstrating progressive increases across control, CTED, and CTEPH groups, with positive correlation to mean pulmonary arterial pressure (mPAP) and PVR (34).

Extending beyond CTEPH, these AI-enhanced CT methodologies demonstrate diagnostic utility across diverse PH etiologies. A random forest model was developed using automated CT measurements of central cardiovascular structures to detect PH in idiopathic pulmonary fibrosis (35).

AI technology could also address a challenging diagnostic scenario by reconstructing, segmenting, and visualizing raw chest CT data from tuberculosis destroyed lung patients with PH. By combining the volume of tuberculosis destroyed lung with echocardiographically derived systolic pressure, this approach enabled accurate identification of such patients (36).

The innovations in molecular imaging with PET/CT represent another recent research direction in 2025 and provide unique functional insights. Using ¹³N-ammonia and ¹⁸F-fluorodeoxyglucose PET/CT to evaluate RV perfusion and glucose metabolism, Goncharova et al. demonstrated that increased RV/left ventricle (RV/LV) perfusion ratios may indicate early coronary flow adaptation in low-risk iPAH (37). ¹⁸F-labeled fibroblast activation protein inhibitor (FAPI) PET/CT visualized fibrotic remodeling in pulmonary arteries and the RV, suggesting utility for treatment monitoring (38). Extending this, FAPI uptake in the target-to-background ratio of RV free wall (TBR_RVFW) independently predicted clinical worsening and enhanced prognostic models, with TBR_RVFW >2.1 indicating poor outcomes (39).

MR imaging (MRI)

Cardiac MR (CMR) provides precise assessment of atrial and ventricular size, morphology and function. Compared to feature-tracking CMR which provides only global strain values, recent advances in imaging technology have improved the sensitivity and accuracy. Specifically, layer-specific strain analysis assisted by a deformable registration algorithm identified low- or intermediate-risk PAH patients with worse prognosis, thereby enabling early risk stratification (40). Further enhancing hemodynamic diagnostic capability, 4-dimensional (4D)-flow MRI quantification of left pulmonary artery vortex energetics effectively distinguished PAH from pulmonary venous hypertension with an AUC of 0.89 (41), while patient-specific computational fluid dynamics modeling based on 4D-flow MRI and CMR showed that time-averaged wall shear stress (TAWSS) linearly correlates with invasive mPAP (42), offering a potential non-invasive assessment method for PH severity. Beyond these approaches, emerging technologies such as hyperpolarized ¹²⁹Xe MRI oscillation mapping was applied to characterize PH patients and may provide more sensitive detection of pulmonary microvascular dysfunction (43).

AI/ML-based automated quantification of MRI is currently being investigated as an alternative to invasive hemodynamic parameters. For instance, a study incorporating 1,646 cardiac MR cine examinations from multiple views developed an explainable convolutional neural network (CNN) model, which not only achieved a Pearson correlation coefficient of 0.80 in predicting mPAP, but also highlighted the RV in the short-axis view as particularly informative through attention weight analysis (44). Additionally, a deep learning model constructed from CMR cine images demonstrated promising predictive efficacy for estimating mPAP and PVR severity in pediatric PAH patients (45). AI can also enhance CMR prognosis analysis through risk assessment modeling. Zhang et al. demonstrated that integrating RV ejection fraction and interventricular septum extracellular volume into COMPERA 2.0 significantly enhanced prediction of clinical deterioration in connective tissue disease (CTD)-associated PAH (46).

Multimodal AI/ML applications

A single data source often provides limited information, constraining predictive performance. To address this, researchers are developing AI/ML-based diagnostic and prognostic models that integrate multimodal data including clinical records, imaging, and biochemical factors, thereby enriching the feature space and enabling more complex disease representations (24). One key application of such multimodal models is the early identification of PH in patients with CTD or other underlying conditions, as PH frequently emerges as a serious complication of these primary diseases, yet the underlying mechanisms of this transition are complex and remain poorly understood. Niu et al. developed the first ML-based nomogram model incorporating age, CRP, anti-double-stranded DNA (anti-dsDNA), pericarditis, and SLE Activity Index (SLEDAI) to predict SLE-interstitial lung disease (ILD)-PAH risk (47). Additionally, random forest applied to Computerized Registry of Patients with Venous Thromboembolism (RIETE) registry data of 5,217 pulmonary embolism (PE) patients effectively predicted CTEPH discrimination, identifying key predictors such as chest pain, PE location, and troponin levels (48).

On the other hand, AI can leverage basic tests like electrocardiogram and chest radiography (CXR) to enable accurate PH screening in resource-limited settings. Studies including thousands of patients showed that integrating these modalities through deep learning improved detection performance, achieving high predictive accuracy and effective early screening for PH and related conditions (49,50).

Meanwhile, ML algorithms continue to advance. On one hand, hybrid models integrating high-precision eXtreme Gradient Boosting (XGBoost) model with transparent white-box model—sure independence screening and sparsifying operator (SISSO) symbolic regression to extract key clinical features, and achieved accurate, non-invasive mPAP prediction in small datasets, while delivering fully transparent mathematical formulas (51). On the other hand, deep learning architectures have evolved toward sophisticated multimodal fusion; Zhao et al. developed a transformer-based multimodal fusion model integrating denoising autoencoders for tabular data, Bidirectional Encoder Representations from Transformers for textual reports, and Vision Transformers for CXR imaging via self-attention mechanisms, providing interpretable modality contribution visualizations (52). These complementary approaches—symbolic regression for mathematical transparency and attention-based multimodal fusion for clinical reasoning mimicry, collectively bridge the gap between black-box performance and real-world clinical utility in PH prediction.

These advances collectively herald a new era of precision PH management, where AI-integrated multimodal imaging enables accurate, non-invasive phenotyping and risk stratification across diverse clinical settings (Figure 1).

Figure 1.

How AI assists different imaging modalities in PH. AI, artificial intelligence; CT, computed tomography; ML, machine learning; MRI, magnetic resonance imaging; PET, positron emission tomography; PH, pulmonary hypertension.

Right heart catheterization

Right heart catheterization remains the gold standard of diagnosing and classifying PH. In 2025, the American Heart Association released standardized protocols for baseline and provocative invasive hemodynamic assessment, including vasodilator challenges, volume loading, and invasive exercise hemodynamic testing to optimize PH diagnosis and phenotyping (53). In this context, proper patient positioning during the preparatory phase of hemodynamic assessment is critical, as it significantly influences measurements. Specifically, upright catheterization reducing mPAP and pulmonary capillary wedge pressure compared to supine assessment, thereby improving specificity for detecting elevated filling pressures and minimizing false-positive post-capillary PH diagnosis (54). Building upon these technical refinements, exercise right heart catheterization has emerged as a critical tool to unmask distinct RV adaptation patterns and characterize subtype-specific PVR-compliance relationships during stress, offering enhanced prognostic stratification beyond resting parameters (55,56). Furthermore, recent advances in automated analysis and interpretation, including generative AI pipelines with built-in guardrails that enable precise structured data extraction from unstructured right heart catheterization notes, and computer vision-based computational pipelines that digitize pressure-time waveforms for single-beat RV pressure-volume analysis, have now enabled efficient, scalable processing of hemodynamic data to calculate load-independent indices and facilitate AI-assisted PH diagnosis in real-world clinical settings (57,58).

Exercise testing

Exercise testing serves as a critical tool for detecting exercise intolerance arising from pulmonary vascular limitation and RV dysfunction. However, conventional assessments such as the 6-minute walk distance (6MWD) provide limited pathophysiological insight (59). Recent advances demonstrated that cardiopulmonary exercise testing (CPET) parameters such as peak oxygen pulse, exhibited stronger correlations with hemodynamics and cardiac index than 6MWD. Moreover, a prognostic model incorporating a four-strata CPET score yielded superior survival prediction compared with model including 6MWD (60). Furthermore, level 3 CPET was employed to guided treatment decisions by identifying dynamic PVR changes during exertion (61). Beyond conventional CPET, a novel eccentric cycling exercise test revealed distinct cerebral and muscular oxygenation patterns, potentially optimizing exercise rehabilitation for PH patients (62).

Therapeutic advances: from symptom management to disease modification

The therapeutic landscape of PH has experienced a paradigm shift of unprecedented magnitude in 2025, moving beyond traditional vasodilator-focused approaches toward therapies that directly target vascular remodeling and disease progression.

Novel targeted therapeutics

The cornerstone breakthrough: sotatercept and its clinical evidence

Sotatercept, a first-in-class type IIA activin receptor (ActRIIA) fusion protein, functions as an activin signaling inhibitor (63). It acts by sequestering excess ligands like activins and GDFs, thereby rebalancing pro- and anti-proliferative signals and directly targeting vascular remodeling at its mechanistic root (64). Building upon the pivotal Phase III STELLAR trial: 40.8 m 6MWD improvement over placebo and benefits across eight secondary endpoints in World Health Organization (WHO) functional class II/III participants (63), which led to its U.S. Food and Drug Administration (FDA) approval in 2024, Sotatercept’s position was further reinforced in 2025.

The Phase 3 ZENITH trial specifically enrolled high-risk patients with advanced disease (WHO functional class III/IV) and a REVEAL Lite 2 risk score ≥9, revealing that add-on Sotatercept reduced the composite endpoint of death, lung transplantation, or hospitalization by 76% compared to placebo, with only 17.4% of treated patients experiencing events vs. 54.7% in the control group (65). Besides, the HYPERION trial investigated early intervention, randomizing patients within one year of diagnosis and demonstrating similar efficacy, with Sotatercept reducing clinical worsening events from 36.9% to 10.6% (66). These findings establish Sotatercept’s value across the disease timeline. Beyond its application across patients with varying disease severity, the efficacy of Sotatercept has also been investigated in increasingly granular patient subpopulations. A planned PULSAR analysis demonstrated consistent safety and efficacy in PAH-associated genes mutation (such as BMPR2) carriers vs. noncarriers, with no differential changes in PVR or 6MWD at 24 weeks (67). Another real-world evidence documented its benefits even in patients with significant cardiopulmonary comorbidities (68). Complementing these trials, a pooled analysis of PULSAR and STELLAR confirmed consistent benefits across 429 patients, demonstrating improvements in exercise capacity, PVR, and WHO functional class (69). This broad efficacy was further validated in a separate pooled analysis stratified by baseline cardiac index, which showed significant improvements in 6MWD, PVR, and NT-proBNP levels irrespective of cardiac index thresholds (70). Across studies, the adverse event profile remained predictable, dominated by epistaxis, telangiectasia, and increased hemoglobin (65,67,69,70), though multicenter surveillance continued to monitor rare complications such as pericardial effusions (71). Together, these data underscore Sotatercept’s universal applicability, validating its use from diagnosis through advanced disease stages and across diverse genetic and hemodynamic profiles, fundamentally transforming PAH management strategies (Figure 2).

Figure 2.

Sotatercept clinical progression: post-STELLAR evidence and expanding applications in 2025. 6MWD, 6-minute walk distance; FDA, Food and Drug Administration; HR, hazard ratio; PAH, pulmonary arterial hypertension; WHO FC, World Health Organization functional class.

Beyond Sotatercept’s breakthrough: emerging mechanistic innovations

Beyond Sotatercept’s breakthrough, the therapeutic landscape has expanded through distinct mechanistic innovations. The therapeutic potential of Imatinib in PAH is undergoing critical reassessment. Imatinib, a tyrosine kinase inhibitor originally for leukemia, has been investigated for PAH due to its anti-proliferative properties (72). A pivotal 2025 dose-finding Phase II study (PIPAH) repositioned low-dose oral imatinib as a promising candidate, demonstrating that a dose of 200 mg daily was well-tolerated and produced improvements in hemodynamics, including reductions in mPAP and PVR (73). This contrasts with the previous halt in development due to significant adverse events associated with the 400 mg dose used in the IMPRES trial and the recent failure of the inhaled formulation (AV-101) in the IMPAHCT study, which did not meet its primary endpoints (74-76).

Future development hinges on several key strategic considerations. First, the potential direct cardiotoxic effects on the right ventricle, as suggested by preclinical studies indicating aggravation of pressure-overload-induced RV failure via the c-Jun N-terminal kinase (JNK)/RUNX family transcription factor 2 (Runx2) pathway, must be addressed (77). This insight points toward a promising therapeutic strategy: combining Imatinib with RV-protective agents, such as JNK inhibitors, to uncouple its beneficial pulmonary vascular effects from detrimental cardiac consequences. Second, advancing a precision medicine approach is essential. As mentioned above, a distinct circulating metabolomic signature particularly involving acylcarnitines, was associated with RV-PA coupling and response to Imatinib (22). These biomarkers could enable patient stratification to identify individuals most likely to derive a favorable benefit-risk ratio from therapy. Consequently, the future of Imatinib in PAH depends on rigorously evaluating the optimized low-dose regimen within trials that incorporate RV monitoring, explore rational combination therapies, and leverage biomarker-guided patient selection.

Inhaled Seralutinib, another tyrosine kinase inhibitor targeting PDGF receptor α/β, colony stimulating factor 1 receptor, and KIT proto-oncogene receptor tyrosine kinase, demonstrated durable efficacy in the TORREY trial extension, with median PVR decreasing by 143 dyne·s/cm⁵ over 72 weeks, advancing to Phase III PROSERA evaluation (78).

In parallel, hymecromone emerged for ILD-PH through hyaluronan synthesis inhibition, achieving an unadjusted 66 m 6MWD improvement in a phase IIa proof-of-concept trial despite non-significant PVR changes (79).

For PH associated with left heart disease (LHD-PH), the relaxin mimetic AZD3427 entered Phase IIB Re-PHIRE trials as the first agent leveraging this hormone’s vasodilatory and anti-fibrotic properties (80).

Immunomodulation has surfaced as a novel principle, with a retrospective study showing over one-third of SLE-PAH patients achieved complete hemodynamic normalization after induction of immunosuppressive therapy, correlating with improved survival (81), supporting targeted immunosuppression in autoimmune-associated PH.

Collectively, these advances demonstrate PAH treatment evolution toward precision medicine across diverse mechanisms—kinase inhibition, matrix remodeling, hormonal modulation, immunomodulation, and nanotechnology—addressing critical gaps in ILD-PH, LHD-PH, and other PH subtypes.

Innovative applications of classical PH therapeutic pathways

Parallel to novel drug development, research is refining the application of established PH-targeted therapies, expanding their use to new populations and optimizing treatment strategies. Recent evidence has demonstrated that dose optimization, strategic combination regimens, and phenotype-specific applications can unlock substantial additional value from classical pathways, generating meaningful improvements across diverse patient populations.

Endothelin receptor antagonist (ERA) pathway: dose refinement and evidence expansion

The ERA class has undergone critical reappraisal through several pivotal studies, with recent research focusing on dose optimization, combination strategies, and exploration in new indications. Macitentan (usual dose of 10 mg), the most extensively studied ERA, has demonstrated divergent outcomes based on dose and patient selection (82), underscoring the importance of precision dosing.

Long-term extension data from the MERIT studies indicated that the established 10 mg dose maintained hemodynamic benefits with a manageable safety profile in inoperable CTEPH (83). In contrast, to investigate the efficacy of higher dose macitentan, a phase III MACiTEPH study using 75 mg of macitentan was conducted in CTEPH patients already on background therapy, but failed to improve exercise capacity in most participants, leading to its discontinuation for futility (84).

Furthermore, real-world evidence supported the utility of ERA-based combination therapy. Data from the OPUS/OrPHeUS registries suggested that macitentan combined with tadalafil was effective and well-tolerated in patients with PAH and cardiovascular comorbidities including diabetes mellitus hypertension, and obesity, showing comparable outcomes to those without such comorbidities (85). In summary, while the ERA pathway remains central to PH treatment, evidence underscores the importance of precise dosing and careful patient selection, with ongoing research aimed at refining its therapeutic application.

NO signaling pathway: expanding real-world validation and novel applications

The NO pathway constitutes a fundamental therapeutic target in PH, with two principal drug classes enhancing downstream cyclic guanosine monophosphate (cGMP) signaling: phosphodiesterase-5 inhibitors (PDE5is) and soluble guanylate cyclase (sGC) stimulators (86). Accumulating real-world evidence has underscored the therapeutic value of PDE5is across various PH etiologies. In PH associated with chronic obstructive pulmonary disease (COPD), PDE5i treatment administered to 418 patients in the Pulmonary Vascular Research Institute (PVRI) GoDeep meta-registry, was linked to a significant reduction in mortality, a benefit robustly observed across different disease severities and comorbidities (87). Based on the same database, this survival advantage extended to patients with ILD-associated PH, where PDE5i therapy was associated with improved outcomes specifically in severe cases characterized by elevated PVR, including subtypes like idiopathic pulmonary fibrosis (88). Furthermore, in combined post- and pre-capillary PH, PDE5i use was connected to a meaningful reduction in PVR and a positive trend in survival (89).

Riociguat, an sGC stimulator, offers a distinct mechanism by directly enhancing cGMP production. As the first drug approved for inoperable or persistent CTEPH (90), a 2025 COMPERA registry analysis of 1,451 medically treated CTEPH patients further validated that riociguat was associated with significantly improved 5-year survival compared to other PH medications (91). Similarly, studies demonstrated that continuing Riociguat rather than other drugs improved 3-year survival and helped prevent the worsening of exercise intolerance in CTEPH patients after balloon pulmonary angioplasty (BPA) (92). Besides, real-world meta-analyses confirmed its capacity to significantly improve exercise capacity and key hemodynamic parameters, such as mPAP and PVR, in CTEPH (93). For PAH, riociguat improved risk stratification scores and 6MWD in patients with different risk stratification, and was also effective and safe even in patients with cardiometabolic comorbidities (94,95).

Combination strategies incorporating NO-pathway agents show promise. Upfront use of sildenafil (a PDE5i) with ambrisentan improved clinical and hemodynamic parameters in CTEPH (96), while combining sildenafil with statins may benefit patients with COPD-PH (97). In severe ILD-PH, combining a PDE5i with an inhaled prostacyclin analog appeared superior to PDE5i monotherapy (88).

Additionally, early-phase investigational approaches, such as angiogenic cell therapy with endothelial nitric oxide synthase (eNOS)-enhanced endothelial progenitor cells, hinted at novel therapeutic avenues (98). Collectively, these findings highlight the expanding and evolving role of NO pathway modulation in PH management, though further controlled trials are warranted to confirm the efficacy of combination and emerging biologic strategies.

Prostacyclin pathway: optimizing delivery methods and treatment sequencing

Research into the prostacyclin pathway remains central to PAH management, supporting its role. A large real-world study (EXPOSURE) including 2014 patients reported a higher survival rate and less functional impairment for patients initiated on prostacyclin receptor agonist Selexipag, compared to those on other PAH-specific therapies (99). Current efforts focused on optimizing drug delivery and treatment sequencing. Regarding efficacy across patient profiles, a post-hoc analysis of the TRIUMPH and FREEDOM-EV trials indicated that the benefits of inhaled or oral Treprostinil on exercise capacity and clinical worsening risk were consistent regardless of the presence of cardiovascular comorbidities (100). The long-term BREEZE Optional Extension Phase demonstrated that transitioning from Treprostinil inhalation solution to a dry powder inhaler (DPI) was safe, maintained efficacy, and improved patient satisfaction (101), suggesting that development of more convenient administration methods is necessary. Beyond delivery, sequencing strategies are being refined. A small study showed that intensive parenteral treprostinil could help high-risk patients achieve predefined hemodynamic targets, enabling a successful transition to oral selexipag (102).

The utility of prostacyclins was also being evaluated beyond Group 1 PAH. While inhaled Treprostinil proved effective for PH associated with ILD, its trial in COPD-PH was halted due to safety concerns, highlighting the heterogeneity of Group 3 PH (103). A relevant case report further illustrated the potential of adding inhaled treprostinil to intravenous therapy for a patient with systemic sclerosis-PH and ILD refractory to epoprostenol (104). Overall, contemporary research underscores a trend towards personalized, accessible prostacyclin therapies and strategic treatment escalation.

Beyond classic pathways: strategic therapeutics innovation

Beyond the three classic therapeutic pathways, contemporary research on PH increasingly focuses on combination therapies and tailored, integrated strategies. For those with cardiovascular comorbidities, recent registry analyses challenge the guideline preference for initial monotherapy. Studies indicated that initial dual oral combination therapy improved symptoms, exercise capacity, and risk status with good tolerability, despite no clear long-term survival difference compared to monotherapy (105,106). At the same time, safety monitoring remains a consistent priority. A large pharmacovigilance analysis of the FDA Adverse Event Reporting System (FAERS) database reported thousands of adverse drug events such as dyspnea, nasal congestion for all medications (107), underscoring the need for vigilant monitoring of adverse events in clinical practice. Collectively, these findings underscore a shift towards nuanced, combination-based management for PH patients (Table 3).

Table 3. Innovative applications of classical PH therapeutic pathways.

Category	Drug name	Target population	Main efficacy
ERA	Macitentan	Inoperable or persistent/recurrent CTEPH	10 mg long-term treatment maintained hemodynamic improvements with controllable safety (83); 75 mg had no improvement (84)
ERA	Macitentan + tadalafil	PAH patients with cardiovascular comorbidities	Combination therapy showed good efficacy and tolerability in real-world evidence (85)
PDE5i	Sildenafil + ambrisentan	CTEPH patients	Upfront combination improved clinical function and hemodynamics (96)
PDE5i	Sildenafil + statins	COPD-PH patients	Combination therapy improved 6MWD, hemodynamics and overall quality of life (97)
PDE5i	PDE5i (general)	CpcPH	Treatment associated with survival improvement and reduced PVR (89)
PDE5i	PDE5i (general)	GoDeep metaregistry	Survival benefits in severe ILD-PH patients (88); PDE5i reduced mortality of COPD-PH patients (87)
sGC stimulator	Riociguat	CTEPH patients	Real-world evidence: improved 6MWD by 35.86 m, reduced mPAP by 9.23 mmHg (93); survival benefit vs. other medications (91); continuation of riociguat post-BPA improved exercise capacity (92)
sGC stimulator	Riociguat	PAH patients	COMPERA 2.0 analysis confirmed risk-reduction benefits in both low- and high-risk patients (95); safe and effective regardless of cardiometabolic comorbidities number (96)
Prostacyclin receptor agonist	Selexipag	PAH patients	Higher survival rate vs. other PAH therapies (99)
Prostacyclin analogue	Treprostinil	PAH patients	Inhaled treprostinil improved 6MWD across all comorbidity subgroups, while oral Treprostinil reduced clinical worsening risk by 36–45% (100); treprostinil DPI had long-term safety and continued 6MWD improvement (101); intensive parenteral treprostinil improved high-risk PAH patients 6MWD and NT-proBNP (102)
Prostacyclin analogue	Inhaled treprostinil	Group 3 PH patients	Increased 31.12 m 6MWD in ILD-PH but caused safety concerns in COPD-PH (103); treprostinil plus IV epoprostenol improved hemodynamics in refractory SSc-PH with ILD (104)

6MWD, 6-minute walk distance; BPA, balloon pulmonary angioplasty; COMPERA, Comparative, Prospective Registry of Newly Initiated Therapies for Pulmonary Hypertension; COPD, chronic obstructive pulmonary disease; CpcPH, combined post- and precapillary PH; CTEPH, chronic thromboembolic pulmonary hypertension; DPI, dry powder inhaler; ERA, endothelin receptor antagonist; ILD, interstitial lung disease; mPAP, mean pulmonary arterial pressure; NT-proBNP, N-terminal pro-B-type natriuretic peptide; PAH, pulmonary arterial hypertension; PDE5i, phosphodiesterase-5 inhibitors; PH, pulmonary hypertension; PVR, pulmonary vascular resistance; sGC, soluble guanylate cyclase; SSc, systemic sclerosis.

Interventional and device-based therapies

Interventional approaches and device-based therapies have revolutionized the management of PH, particularly for CTEPH, while emerging data illuminate their role across PH subtypes.

Interventions: PEA, BPA, PADN and hybrid approaches

Pulmonary endarterectomy remains the gold standard for operable CTEPH, while BPA provides an alternative for inoperable patients. A narrative review highlighted hybrid approaches combining PEA and BPA, where multidisciplinary teams tailor therapy based on lesion location and patient characteristics (108). Comparative studies revealed that while both modalities improve outcomes, PEA demonstrated superior hemodynamic benefits, including reduced resistance-compliance time and improved cardiac index, whereas BPA offered advantages in distal lesions and functional class improvement (109,110). Recent data from a nationwide Japanese registry covering 5,207 procedures confirmed BPA’s real-world effectiveness, achieving a 55.6% reduction in PVR with low mortality (0.2%) (111). Optimal BPA targets remained under investigation; achieving mPAP <30 mmHg post-procedure correlated with favorable outcomes, with ≤20 mmHg associated with superior functional capacity and reduced therapy dependence (112). Emerging catheter-based interventions show promise. As an emerging catheter-based intervention, pulmonary artery denervation (PADN) reduced PVR by 3.0±0.3 WU and improved exercise capacity (113), while investigations into its long-term safety and efficacy are being broadened to encompass populations with medication-refractory PH (114).

Patient-specific factors influencing interventional outcomes

Age and genetic factors influence interventional outcomes. In patients ≥70 years old, BPA provided comparable hemodynamic improvements to PEA but with greater functional class enhancement (115). However, carriers of the RNF213 p.Arg4810Lys variant exhibited poorer responses to BPA and higher complication rates (25% vs. 8%), warranting genetic screening (116). While patient-specific factors shaped outcomes, procedural expertise is equally critical. The learning curve significantly impacts BPA outcomes: a single-center study demonstrated improved oxygenation and reduced ventilation requirements as operator experience increased (117). However, even with optimal technique, complications such as reperfusion lung injury overlap with pneumonia in 48% of cases post-PEA, emphasizing the need for vigilant postoperative surveillance (118).

Lung transplantation

For end-stage disease, lung transplantation remains critical, yet registry data reveal alarming under-referral patterns—only 12% of PAH patients were referred, often too late despite guideline recommendations (119). Given this scarcity of interventions, optimizing transplant strategies becomes paramount. Bilateral lung transplantation achieved comparable 1- and 5-year survival to heart-lung transplantation, supporting its use in appropriately selected patients (120).

Collectively, these advances underscore that interventional and surgical PH therapy requires individualized, multidisciplinary strategies balancing efficacy, safety, and patient-specific factors.

Non-pharmacological interventions and multidisciplinary management

Non-pharmacological strategies are increasingly recognized as essential components in multidisciplinary PH management, addressing symptoms, functional capacity, and long-term outcomes through diverse mechanisms.

Exercise and rehabilitation

Exercise training demonstrated safety and efficacy in improving exercise capacity and quality of life for CTEPH and PAH patients, with no associated adverse events (121). Home-based exercise programs produced comparable improvements to center-based interventions, enhancing 6MWD by 54.85 meters and peak oxygen uptake without safety concerns (122). High-intensity inspiratory muscle training significantly improved respiratory muscle strength, exercise capacity, and dyspnea perception in symptomatic patients (123).

Respiratory support & oxygen therapy

Oxygen therapy provides benefits in PH by alleviating hypoxia-induced pulmonary vasoconstriction and reducing RV workload (124). Positive airway pressure therapy substantially reduced pulmonary artery pressures in obstructive sleep apnea/obesity hypoventilation syndrome, with a mean reduction of 5.96 mmHg overall and 11.41 mmHg in patients with baseline PH (125). High-altitude exposure exacerbated nocturnal hypoxemia in pulmonary vascular disease, decreasing mean nocturnal SpO₂ from 91% to 83% and requiring supplemental oxygen therapy in 37% of patients (126). The prevalence of long-term oxygen therapy (LTOT) in PAH and CTEPH is approximated 20%, with functional class, hemodynamic severity, and diffusing capacity serving as LTOT prescription predictors (127).

Patient management & adherence

Nurse-led, multidisciplinary, guideline-directed intensive management significantly improved overall and event-free survival in PH patients, particularly benefiting high-risk patients and those with advanced RV dysfunction (128). However, these survival gains were jeopardized by medication nonadherence, which affected 6.1–22% of patients and correlates with 50% more emergency department visits, 13.3% more hospital admissions, and worse quality of life (129,130). The roots of this adherence gap are multifaceted: predictors included older age and higher risk classification, while barriers encompassed poverty, unmarried status, and inadequate insurance coverage (129,131).

Gut microbiota-targeted interventions

Increasing evidence suggests that the gut microbiota is involved in the development and progression of PH, providing new therapeutic targets. A pioneering study, for the first time, examined the safety and feasibility of microbiota transplant therapy in patients with PAH, achieving modest donor engraftment accompanied by transient reductions in circulating pro-inflammatory cytokines (132). Another research showed antibiotic-driven microbiota modulation decreased mPAP in PH patients, particularly hypoxic lung disease-associated cases, while enriching potentially protective Lactobacillus species (133). Notably, dietary nitrate supplementation enhanced exercise capacity in WHO Group 3 PH, improving endurance shuttle walk time by a median of 30 seconds and endothelial function (134). These findings support microbiome-targeted strategies as novel therapeutic avenues.

Other interventions

As a modern improved version of traditional Chinese medicine acupuncture therapy, electroacupuncture safely and effectively alleviated dyspnea in COPD-associated PH, improving modified Medical Research Council scores and 6MWD performance (135).

These multifaceted non-pharmacological approaches, integrated with pharmacotherapy, optimize PH management across disease severities and phenotypes.

RV targeted therapy

Pressure overload induced RV failure remains the primary cause of mortality in PH. Paradigm shift toward therapies have emphasized direct targeting of RV function and RV-PA coupling, guided by prognostic tools like the TAPSE/sPAP ratio, three-dimensional (3D) echocardiography, and metabolomic profiling for personalized treatment selection (136-138). The negative famotidine trial underscores the limitations of single-pathway metabolic interventions in complex RV remodeling (139). According to multiple molecular research, emerging therapeutic directions include transient receptor potential vanilloid type 2 (TRPV2) inhibition (140), mitochondrial protection via SUL-150 (141), long noncoding RNA (lncRNA)-modulated micropeptides (142) and gut microbiome modulation (143). Future success depends on integrating advanced imaging, metabolomics, and biomarkers such as cellular communication network factor 2 (CCN2) to enable personalized combination therapies addressing both RV dysfunction and pulmonary vascular disease simultaneously (144).

Limitations

This review provides a synthesis of the 2025 advances in PH clinical research, encompassing diverse domains from molecular biomarkers to interventional therapies. The inclusion of large-scale clinical trials, real-world registry data, and cutting-edge technological innovations strengthens the evidence base presented. However, several limitations should be acknowledged. First, this is a narrative review rather than a systematic review. Consequently, no formal grading of evidence was performed and the literature search was limited to the PubMed database. These methodological constraints indicate that the analysis may not capture all relevant studies or fully account for publication bias. Second, the rapidly evolving nature of PH research means that some findings may be superseded by newer evidence. Third, the heterogeneity of PH subtypes limits the generalizability of certain therapeutic approaches across all patient populations. Fourth, long-term outcome data for novel therapies such as Sotatercept remain limited, and the sustainability of treatment effects requires further validation. Additionally, the integration of AI tools into clinical practice faces challenges related to algorithmic bias, interpretability, and regulatory approval. Finally, disparities in healthcare access and resource availability may limit the implementation of precision medicine approaches in certain regions.

Conclusions

The year 2025 represents a transformative period in PH clinical research, characterized by paradigm-shifting advances that fundamentally alter disease management. ML algorithms applied to multimodal data will likely guide therapy selection, monitor treatment response, and predict clinical outcomes with unprecedented accuracy. The emergence of Sotatercept and other disease-modifying therapies targeting vascular remodeling represents a departure from traditional vasodilator-focused approaches, offering the potential to alter the natural history of the disease rather than merely palliate symptoms.

The future of PH clinical research lies in precision medicine approaches integrating advanced diagnostics, disease-modifying therapies, and personalized management strategies to transform patient outcomes.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.