Group 3 Pulmonary Hypertension: From Bench to Bedside

Abstract

Pulmonary hypertension (PH) due to chronic lung disease is categorized as Group 3 PH in the most recent classification system. Prevalence of these diseases is increasing over time, creating a growing need for effective therapeutic options. Recent approval of the first pulmonary arterial hypertension (PAH) therapy for the treatment of Group 3 PH related to interstitial lung disease (ILD) represents an encouraging advancement. This review focuses on molecular mechanisms contributing to pulmonary vasculopathy in chronic hypoxia, the pathology and epidemiology of Group 3 PH, the right ventricular dysfunction observed in this population and clinical trial data that inform the use of pulmonary vasodilators in Group 3 PH.

Introduction

Pulmonary hypertension (PH) due to chronic lung disease is categorized as Group 3 PH in the World Symposium on Pulmonary Hypertension (WSPH) classification system¹. Group 3 PH deserves special attention as chronic lung diseases are increasing in prevalence and there are a lack of sufficient therapeutic options for PH in lung disease. Currently available therapies for Group 1 pulmonary arterial hypertension (PAH) have been tested in Group 3 PH with mixed results². This may reflect a lack of understanding of the mechanisms driving Group 3 PH as well as the heterogeneity of patients and diseases involved. There is disproportionate right ventricular (RV) dysfunction in Group 3 PH patients as compared to PAH despite less severe hemodynamic impairment³, highlighting an additional knowledge gap. As RV function is a critical determinant of outcome in pulmonary vascular disease and chronic lung diseases are a major public health problem, Group 3 PH is an important focus for study and novel therapeutic development. The recent approval of a single PAH therapy for Group 3 PH related to interstitial lung disease (ILD) has re-ignited interest in this topic. In this focused review, we will discuss the available experimental models for studying Group 3 PH, currently known molecular mechanisms that contribute to pulmonary vascular remodeling in chronic hypoxia, review the pathology and epidemiology of Group 3 PH and the RV dysfunction observed in this population and highlight experimental and clinical trial data that inform the use of pulmonary vasodilators in Group 3 PH.

Experimental Models and Molecular Mechanisms of Pulmonary Vascular Remodeling in Chronic Hypoxia

Much of the evidence regarding the molecular mechanisms responsible for pulmonary vasculopathy in chronic lung disease comes from pre-clinical animal models with hypoxia exposure^{4, 5}. Common experimental models of Group 3 PH and their strengths and weaknesses are summarized in Table 1. Although chronic hypoxia may mimic high-altitude PH, it is unclear whether it accurately recapitulates chronic lung diseases such as chronic obstructive pulmonary disease (COPD) and ILD^6–9. As in Group 1 PAH, endothelial cell dysfunction and aberrant cell signaling with unchecked endothelial, fibroblast and smooth muscle cell growth, proliferation and dysregulated angiogenesis are thought to contribute to the pathobiology of Group 3 PH¹⁰. Of particular interest in Group 3 PH is the independent role that cigarette smoke plays in the development of pulmonary vasculopathy (Figure 1A). Emerging themes in chronic hypoxia models of PH include the role of microRNAs (miRs) and progenitor cells in disease pathobiology and altered cellular metabolism and mitochondrial function as novel therapeutic targets (Figure 1B). Experimental models have also elucidated the potential mechanistic benefit of prostacyclins in PH associated with chronic lung disease.

Table 1.

Summary of experimental models of hypoxia in pulmonary hypertension

Model	Species	Methodology	Histopathology	Strengths	Weaknesses
Chronic hypoxia219–222	Rats, mice, Guinea pig, pig, sheep	Chronic exposure to hypoxia	Pulmonary vasoconstriction, muscularization of non-muscular arterioles, increased media thickness and matrix deposition	• Widely used • Simple	• Variable response to hypoxia
SU5416+Hypoxia223–225	Rats, mice	VEGFR-2 blockade coupled with chronic hypoxia	Development of PH with complex plexiform-like lesions	• Mimics human PAH	• Skill-specific • May not adequately model Group 3 PH • More costly
Fawn-hooded rats226, 227	Rats	Rat strain with deficient serotonin uptake into platelets	Rats have immature development of lungs, reduced number of alveoli, and development of PH	• May be useful to study specific pathways in PH	• Mimics limited features of PH • Cost
Broiler chicken228–233	Chicken	Pulmonary diffusion limitation leading to hypoxemia, possibly inherent mitochondrial dysfunction,	Right ventricular hypertrophy, perivascular leakage, vascular medial hypertrophy, perivascular inflammatory infiltration	• May be useful to study specific pathways in PH	• Mimics limited features of PH • Cost
Bleomycin234–236	Rats, mice, rabbits	Administration of bleomycin, commonly intratracheally	Bleomycin leads to pulmonary fibrosis development with increased lung inflammation and muscularization	• Useful to study PH in ILD • Simple • Relatively inexpensive	• Mimics only limited features of human PH
Term and preterm237–243	Mice, rats, rabbits, sheep, baboons	Varies depending on model but commonly preterm animals with mechanical ventilation. Hyperoxia and inflammation by bacterial LPS also used.	Alveolar simplification (rarefication) and loss of surfactant	• Useful to study PH in BPD	• Variability in lung and vascular injury dependent on species, age of animal, and mechanisms of injury • Heterogeneity of lung pathologies in human BPD is not accurately recapitulated in animals

SU5416 = Sugen 5416. VEGFR-2 = vascular endothelial growth factor receptor 2. PH = pulmonary hypertension. PAH = pulmonary arterial hypertension. ILD = interstitial lung disease. LPS = lipopolysaccharide. BPD = bronchopulmonary dysplasia. Adapted from Nogueira-Ferreira, et al. Cardiol and Cardiovasc Med²⁴⁴.

Figure 1.

Molecular mechanisms of pulmonary vasculopathy in Group 3 pulmonary hypertension (PH).

A. Cigarette smoke leads to decreased expression of endothelial nitric oxide (eNOS) along with intimal hyperplasia, increased expression of vascular endothelial growth factor (VEGF), inflammatory infiltrate, and metabolic dysfunction. Poorly differentiated progenitor cells have been observed in the intima of pulmonary arteries; their origin and role in the development of pulmonary vasculopathy remains unclear. B. Hypoxia inducible factor-1 alpha (HIF-1α) and transforming growth factor-beta (TGF-β) signaling are recognized to regulate growth, differentiation, and proliferation of almost all cell types in hypoxia-induced PH. VEGF expression is induced by hypoxia via HIF-1α signaling. HIF-1α is also thought to play a central role in the development of mitochondrial dysfunction that may explain a Warburg-like shift to anaerobic glycolysis. microRNAs (miRs) are small noncoding RNAs that regulate gene expression and in mouse models inhibition of miRs prevent development of pulmonary vasculopathy. Figure created with biorender.com. (Illustration Credit: Ben Smith).

Transforming growth factor-beta 1 (TGF-β1) has been recognized to regulate cell growth, differentiation, and proliferation in almost all cell types^{11, 12} and molecular mechanisms integral to the vascular remodeling observed in PH. TGF-β1 expression is closely linked to hypoxia-inducible factor-1 alpha (HIF-1α) and both signaling pathways are well established in the development of pulmonary vascular disease^13–15. In hypoxia exposed rats, TGF-β1 expression was increased as compared to normoxic controls and associated with a decrease in phosphatase and tensin homolog deleted on chromosome 10 (PTEN)¹⁶. Vascular endothelial growth factor A (VEGFA) is also induced by HIF-1α; its central role in the development of PAH is reviewed separately¹⁷. In rats, chronic hypoxia increases VEGFA expression and its receptor, VEGFR2^{18, 19}. Though loss of VEGFB in mice had no effect, its overexpression was beneficial²⁰, suggesting that it plays a protective role alongside VEGFA in hypoxic PH. Platelet-derived growth factor β (PDGFβ) production is stimulated by TGF-β1 and increases expression of VEGFA, augmenting hypoxia-induced endothelial cell proliferation^{21, 22}. Increased expression of angiogenesis mediators and growth factors such as VEGF and receptor-II of TGF-β (TGF-βRII) has been observed in the pulmonary arteries of patients with COPD²³. End-stage COPD patients who underwent lung transplantation and patients with mild to moderate COPD have impaired endothelial-dependent relaxation^{24, 25}. As a result, endothelial dysfunction remains a potential therapeutic target in Group 3 PH. Sildenafil abrogated endothelial reactive oxygen species generation along with improving RV/left-ventricle+septum ratio and decreasing pulmonary vascular and RV fibrosis in a bleomycin-induced ILD mouse model²⁶. Fibroblast growth factors (FGFs) contribute to the proliferation of pulmonary artery endothelial cells (PAECs) and smooth muscle cells. Production of FGF2 is increased alongside HIF-1α in a feedforward manner^{27, 28}. While additional mechanistic work is needed to clarify the role of FGF2 in pulmonary vascular remodeling during hypoxia, FGF2 is elevated in mouse models of chronic hypoxia^{21, 29} and in preterm infants with bronchopulmonary dysplasia (BPD)³⁰.

Cigarette smoke as a toxic insult and cause of pulmonary vascular disease is particularly relevant in Group 3 PH. Independent of airflow obstruction, smoking has been observed to cause intimal hyperplasia³¹, reduced expression of endothelial nitric oxide synthase (eNOS)³², increased VEGF expression²³, infiltration of inflammatory cells³³ and an imbalance of mitochondrial fission and fusion with resultant mitochondrial oxidative stress and dysfunction³⁴. It is believed that cigarette smoking plays a central role in the vascular changes that lead to the development of PH in COPD^35–37. In vitro work has confirmed this hypothesis. In a dose and time-dependent manner, human PAECs exposed to cigarette smoke demonstrate decreased eNOS activity and protein expression³⁸ and decreased prostacyclin synthase mRNA and protein expression³⁹. Vascular remodeling due to cigarette exposure may precede the development of clinically detectable emphysema or airflow obstruction as observed in animal models^40–43.

There is growing interest in the role of miRs in the development of all forms of pulmonary vascular disease. miRs are small noncoding RNAs that regulate gene expression by binding to target mRNAs and promote degradation or inhibit translation^{44, 45}. In mouse models of hypoxia-induced PH, inhibition of miR-21 (A-21) improved vessel muscularization and RV hypertrophy⁴⁶ and inhibition of miR-20a improved bone morphogenic receptor 2 (BMPR2) expression (a known causal pathway in Group 1 PAH) and reduced luminal occlusion of small pulmonary arteries and RV hypertrophy⁴⁷.

Progenitor cells may be important in the development of PH in chronic lung diseases. Poorly differentiated cells have been observed in the intima of pulmonary arteries of patients with Group 3 PH. While their origin and role have not been defined, dedifferentiation of smooth muscle cells, transdifferentiation during endothelial-to-mesenchymal-transition, resident precursor cells, or recruitment from bone marrow-derived progenitor cells are all potential sources.^{35, 48–50}. Bone marrow-derived progenitor cells have been identified in the intima of pulmonary arteries from COPD patients⁵¹ and are associated with endothelial dysfunction and the degree of vascular remodeling^{51, 52}, lending support to this hypothesis.

Metabolic reprogramming and mitochondrial dysfunction are likely critical to the pathogenesis of all subtypes of PH. Increased expression of HIF-1α during hypoxia is theorized to alter cellular metabolism and partially explain the Warburg-like shift to anaerobic glycolysis in PH^{53, 54}. When mitochondrial fission is inhibited in a rat model of hypoxia, pulmonary artery smooth muscle cell proliferation is abrogated and exercise capacity, RV function and hemodynamics improve⁵⁵.

As discussed below, inhaled trepostinil was recently approved for ILD-PH and was associated with improved measures of both pulmonary vascular and lung function in the landmark INCREASE trial⁵⁶. Experimental models provide timely mechanistic insight into these results. Prostacyclin acts largely via the IP receptors to increase cAMP which, in the vasculature, promotes relaxation and limits proliferation and, in platelets, regulates calcium levels to reduce thrombosis^57–59. Prostacyclin appears to also have antifibrotic properties via the IP receptor⁶⁰ and activation of peroxisome proliferator-activated receptors (PPAR)⁶¹, abrogating both parenchymal lung and vascular fibrosis in murine models of bleomycin-induced pulmonary fibrosis and Fra-2 transgenic mice^62–65.

Histopathology of Group 3 PH

Most of our knowledge of human histopathologic changes in Group 3 PH comes from explanted lungs from lung transplantation and autopsies. Idiopathic pulmonary fibrosis (IPF)/usual interstitial pneumonia (UIP) and COPD are the most frequently transplanted lung conditions that may be complicated by PH, usually at an advanced stage. Other than chronic hypoxic vasoconstriction, which is an inherent factor in the early evolution of obstructive and restrictive lung disease, other mechanisms that display a morphologic correlate in histology may be the loss of microvessels and capillaries in the progressive destruction of alveolar septa in COPD and IPF, compression of peripheral or central lung vessels due to solid collagen-rich fibrosis in advanced IPF, or direct remodeling of lung vessels with hyperplasia of smooth muscle cells and fibroblasts, leading to constrictive thickening of the media and the intima. IPF is an interstitial process that affects the lung parenchyma very heterogeneously, and interestingly the remodeling of muscular pulmonary arteries and arterioles appears to be more or less restricted to the fibrotic lung areas (Figure 2A). In contrast, it has been shown that the distribution of microvascular changes in IPF lungs is as heterogenous as the fibrotic process, only in an inverse and somehow compensatory fashion, with the highest density of capillaries within non-fibrotic areas, and low microvascular density in fibrotic areas with excessive collagen deposition⁶⁶. Microvascular heterogeneity has been specifically associated with IPF disease evolution by others^{67, 68}.

Fig. 2.

Pulmonary vascular changes in patients with lung disease and Group 3 pulmonary hypertension (PH). A. Pulmonary artery (PA) and its adjacent airway in a patient with idiopathic pulmonary fibrosis (IPF) and PH; note the constrictive intimal fibrosis of the vessel (straight arrows) within the fibrosing area (upper part) and the slender walls of the PA branching out of this area into preserved lung areas (flexuous arrow) with near-normal architecture (lower part). B. Smooth muscle cell hyperplasia within the interstitium (arrows) and within the wall of a small PA, in a patient with IPF-PH; smooth muscle actin-staining. C and D: Capillary density of a patient with chronic obstructive pulmonary disease (COPD) displaying moderate PH (C) and a patient with COPD and severe PH (D): note the loss of ERG-stained endothelial cells within the alveolar septa, corresponding to a decrease of vascular density; photos reprinted with regard to the STM permission guidelines 2022 from Bunel et al. CHEST. 2019⁷². E. PA in a patient with COPD-PH, with severe intimal fibrosis; note the perivascular lymphocytic infiltrate (small blue dots, center). F. Remodeled arterioles in a COPD-PH patient with dense CD5-positive peri-vascular lymphocytic infiltrate (red dots), a feature that can also be observed in pulmonary arterial hypertension.

The partial loss of pulmonary capillaries in Group 3 PH compared to other subtypes of PH has served as an explanation for largely negative clinical trials of pulmonary vasodilators in Group 3 PH. These observations were initially reinforced by animal models demonstrating a reduction in alveolar capillary density with increasing smoke exposure⁶⁹. However these findings have been contested in other experimental settings^{70, 71}. It has been recently shown that lungs from patients with COPD lacking PH and COPD with moderate PH can be discriminated from COPD with severe PH by the degree of microvascular muscularization and the loss of capillaries in the latter group (Figure 2C and 2D)⁷².

Heterogeneous distribution of capillary density with foci of proliferation is known from PAH with overt involvement of pulmonary veins and capillaries, known as pulmonary veno-occlusive disease/pulmonary capillary hemangiomatosis. It has been reported that in lungs from patients with IPF, pulmonary veins show constrictive remodeling in preserved areas as well as in fibrotic areas, in the absence of pulmonary artery remodeling⁷³. Taken together, it appears that all compartments of the lung vasculature may be involved in the natural history of Group 3 PH, even if it is still unclear to what extent and with what hemodynamic consequences.

Genetic Predisposition in Group 3 PH

Recent studies have identified genetic variants that predispose to or are associated with the development of Group 3 PH in chronic lung disease. Variants of kinase insert domain receptor (KDR), which encodes for VEGFR2 has been shown by several groups to correlate with impaired gas transfer, parenchymal lung disease and older age^74–76. In neonates, single nucleotide polymorphisms (SNPs) of dual-specificity phosphatases (DUSPs) can predict PH in patients with BPD⁷⁷ and variants of endothelin-1 (EDN1)⁷⁸, carbamoyl phosphate synthetase I (CPS1), neurogenic locus notch homolog protein 3 (NOTCH3), and SMAD family member 9 (SMAD9)⁷⁹ increase risk for development of persistent pulmonary hypertension of the newborn (PPHN). Taken together, these data suggest that genetic background may explain some of the heterogeneity observed in chronic lung diseases and their variable and patient-specific propensity for pulmonary vascular complications.

Genetic variability may also provide insight into mechanisms of disease development. Patients with IPF-PH have a unique genetic signature including expression of mediators of pulmonary artery smooth muscle cell (PASMC) and endothelial cell proliferation, Wnt signaling, complement system activation and extracellular matrix (ECM) remodeling and apoptosis^{80, 81}. This genetic signature was observed to be similar to IPF patients with and without PH, suggesting that genetic reprogramming may occur prior to the development of PH or that additional mechanisms or “hits” predispose to the development of PH in IPF⁸². Similarly, microarray analysis of laser capture microdissected pulmonary arteries from patients with COPD shows that expression of genes related to retinol metabolism and ECM remodeling is uniquely characteristic of COPD-PH⁸¹.

Epidemiology of Group 3 PH Due to Lung Disease and/or Hypoxia

Following left-sided heart disease, Group 3 PH is the second leading cause of PH^{83, 84} and both the prevalence and incidence are increasing⁸⁴. Patients afflicted with PH as a complication of parenchymal lung disease have the worst long-term survival as compared to all other PH groups^{3, 84, 85}. Group 3 PH patients have higher medical costs⁸⁶ and increased morbidity and mortality^{87, 88} as compared to patients with chronic lung disease without PH.

Obstructive and restrictive lung diseases and chronic hypoxia can all lead to PH (Table 2). COPD is the most common type of obstructive lung disease and is highly prevalent in the general population. Restrictive lung diseases include primarily diffuse parenchymal lung diseases (DPLDs), a heterogeneous group of disorders that includes idiopathic interstitial pneumonias (IIPs), connective tissue disease associated (CTD) ILDs, and unusual ILDs like lymphangioleiomyomatosis (LAM) (Figure 3)^{89, 90}. Not all parenchymal lung diseases with PH are classified as Group 3 PH. For example, pulmonary Langerhans cell histiocytosis and sarcoidosis are in Group 5 PH due to multifactorial mechanisms¹. In the absence of parenchymal lung disease, chronic hypoxia (e.g., altitude) is another known cause of PH. Additional etiologies for Group 3 PH include sleep disordered breathing (SDB) and developmental lung disease, namely BPD.

Table 2.

Clinical classification of pulmonary hypertension (PH) due to lung diseases and/or hypoxia

Group 3 PH due to lung diseases and/or hypoxia

3.1 Obstructive lung disease

3.2 Restrictive lung disease

3.3 Other lung disease with mixed restrictive/obstructive pattern

3.4 Hypoxia without lung disease

3.5 Developmental lung disorders

Group 5 PH with unclear and/or multifactorial mechanisms

5.2 Systemic and metabolic disorders

Pulmonary Langerhans cell histiocytosis

Sarcoidosis

Adapted from Simonneau, et al. ERJ. 2019.¹

Figure 3.

Clinical classification of diseases causing Group 3 pulmonary hypertension (PH). Aside from chronic obstructive pulmonary disease (COPD), many diseases responsible for Group 3 PH belong to the diffuse parenchymal lung diseases (DPLDs) and are further classified based on their etiology. Idiopathic interstitial pneumonias (IIPs) are idiopathic diseases of varying prevalence but represent an important group of diseases that contribute to the development of PH. *Belongs to Group 5 PH. IPF = idiopathic pulmonary fibrosis. NSIP = nonspecific interstitial pneumonitis. RB-ILD = respiratory bronchiolitis-interstitial lung disease. DIP = desquamative interstitial pneumonia. COP = cryptogenic organizing pneumonia. AIP = acute interstitial pneumonia. LIP = lymphocytic interstitial pneumonia. PPF = pleuroparenchymal fibroelastosis. UIP = usual interstitial pneumonia. CPFE = combined pulmonary fibrosis and emphysema. LAM = lymphangioleiomyomatosis. Adapted from Nathan, SD. “Inhaled Treprostinil in Interstitial Lung Disease Associated Pulmonary Hypertension: The INCREASE Study.” Breaking News: Clinical Trial Results in Pulmonary Medicine. ATS Meeting. May 28, 2020.

COPD is the most prevalent chronic respiratory disease worldwide and the third leading cause of death in the United States⁹¹. Most patients with COPD will develop mild PH, as characterized by a mild elevation in pulmonary pressures. Approximately 1 – 5% of COPD patients develop severe PH, with a mean pulmonary artery pressure (mPAP) > 35 – 40 mmHg at rest⁹². Though the rate of progression of PH in COPD is recognized to be slow (< 1 mmHg increase in mPAP per year)⁹³, the concomitant presence of PH with COPD is a strong predictor of mortality. There is an inverse relationship between mPAP and pulmonary vascular resistance (PVR) and survival in COPD^94–96. Recent data from the COMPERA (Comparative, Prospective Registry of Newly Initiated Therapies for Pulmonary Hypertension) and GRAPHIC (GRAz Pulmonary Hypertension in COPD) registries demonstrate that both mPAP and PVR are strong predictors of survival in COPD associated PH; a threshold PVR > 5 Woods units (WU) discriminates mortality in these registries^{97, 98}. The presence of PH in COPD patients is more strongly associated with mortality than forced expiratory volume in the first second (FEV₁) or gas exchange⁹⁹. Similar to idiopathic PAH, the morphology of vascular lesions in COPD correlates with the severity of PH¹⁰⁰.

DPLDs include the IIPs, combined pulmonary fibrosis and emphysema (CPFE) and CTD-associated ILDs. Most data about IIP-associated PH comes from studies of IPF, the most common form of IIP. Other common DPLDs include chronic hypersensitivity pneumonitis, nonspecific interstitial pneumonitis (NSIP), unclassifiable ILD and LAM. The IIPs have differing radiographic and pathologic features; however, all of them, especially the ones characterized by elements of fibrosis, may be complicated by PH. Approximately 8 – 15% of IPF patients have a mPAP ≥ 25 mmHg during initial workup¹⁰¹. Limited data are available regarding the prevalence of PH in other fibrotic ILDs. PH was reported in 31% of a cohort of 35 patients with idiopathic NSIP and 44% of a cohort of 50 patients with chronic hypersensitivity pneumonitis^{102, 103}. More advanced parenchymal lung disease is associated with worse hemodynamics^104–106 and even echocardiographic PH¹⁰⁷ or minimal elevations in mPAP (> 17 mmHg)¹⁰⁸ are associated with reduced survival in IPF patients. Interestingly, the degree of parenchymal lung involvement, based on PFTs¹⁰⁹ and high-resolution CT fibrosis score^{107, 110–113}, does not correlate with PH severity. In COMPERA, a PVR > 5 WU outperformed the now accepted definition of chronic lung disease with severe PH (mPAP ≥ 35 mmHg or mPAP ≥ 25 mmHg with CI < 2.0 L/min/m²) in discriminating survival¹¹⁴.

With the advent of wide-scale computed tomography imaging, CPFE has been recognized as a unique phenotype of ILD that is associated with poor outcomes and severe concurrent PH. As many as 30 – 50% of patients with CPFE are estimated to develop PH^{115, 116} and survival is poor^115–117. Incident PH in this population is severe in approximately one-half of patients (68% with mPAP > 35 mmHg, 48% with mPAP > 40 mmHg) at the time of right heart catheterization¹¹⁵. These patients are often characterized by profound diffusion limitation, which may be a marker of pulmonary vascular involvement^{34, 118–120}.

CTD-associated ILD presents a particular challenge, as CTD-related vasculopathy is an important cause of Group 1 PAH, making differentiation between Group 1 and Group 3 disease more difficult. Moreover, the presence of both PH and ILD is associated with significantly higher mortality than either of the processes alone¹²¹.

In the absence of lung disease, chronic hypoxia predisposes individuals to the development of PH, particularly at high altitude (defined as an elevation > 2500 m above sea level¹⁰⁹). Though reliable prevalence data are lacking, high-altitude PH is a significant problem in certain regions of the world. Chronic hypoxia leads to pulmonary vasoconstriction and over time can cause pulmonary vascular remodeling and elevations in PVR^{122, 123}. This process along with erythrocytosis are hallmark features of high-altitude PH¹²⁴. Though there are some data on the use of pulmonary vasodilators in this population, high-altitude PH is entirely reversible on re-exposure to normal inspired oxygen tension¹⁰⁹.

Sleep disordered breathing (SDB) includes obstructive sleep apnea (OSA) and obesity hypoventilation syndrome (OHS). Both OSA and OHS are characterized by nocturnal hypoventilation, however PH in isolated OSA is uncommon and typically mild¹⁰⁹. In contrast, patients with OHS or overlap syndrome (OSA and COPD) are at high risk for pulmonary vascular complications^{125, 126}. Though prevalence estimates of PH in OHS vary, OHS itself is common in the general population (about 0.4% of the US population)¹²⁷ and PH as a complication of OHS is frequently severe and associated with RV failure and poor outcomes¹²⁸. As obesity becomes more prevalent in the general population, pulmonary vascular disease as a complication of OHS or overlap syndrome may become much more common.

There are several developmental lung disorders associated with PH, including alveolar capillary dysplasia (ACD) with or without misalignments of the veins (MPV), congenital diaphragmatic hernia (CDH), lung hypoplasia, pulmonary alveolar proteinosis (PAP) and abnormalities in surfactant protein^{129, 130}. BPD is a clinical syndrome of injury to preterm lungs characterized by disruption of alveolarization and microvascular development and caused by multiple antenatal and postnatal exposures, including early gestational age, low birth weight, maternal infections and the use of mechanical ventilation¹³¹. The disorder causes impaired proximal airway and bronchoalveolar development and often abnormal vascular remodeling and vascular growth arrest (or rarefication of the pulmonary vasculature)¹³². BPD is the most frequent complication of extreme preterm birth (defined as infants born < 28 weeks gestational age)^{131, 133} and the strongest risk factors for BPD remain prematurity and low birth weight^134–140. The prevalence of BPD is increasing, likely due to the increased survival of extremely low gestational age newborns^{141, 142}. Pulmonary vascular disease can result from extrapulmonary right-to-left shunting across a patent foramen ovale and/or ductus arteriosus causing hypoxia and a sustained elevation in PVR (PPHN)¹⁴³ to later and more persistent PH after hospital discharge and in young adults. Although BPD is a syndrome of lung injury, its relationship to PH is not sequential – antenatal stresses like oligohydramnios and prolonged premature rupture of membranes cause primary vascular abnormalities and lead to elevations in PVR and subsequent development of BPD^143–146.

Right Ventricular Dysfunction in Group 3 PH

Though the term “cor pulmonale” has no consensus definition, it is commonly used to describe RV dysfunction as a result of chronic lung disease associated PH^{147, 148}. RV dysfunction is a major determinant of outcome in all forms of pulmonary vascular disease^149–151. A recent study utilizing echocardiographic RV fractional area change (FAC) assessment in a cohort of 147 patients with Group 3 PH showed that RVFAC < 28% discriminated between long term outcomes¹⁵². Because of the typically slow increases over time in PVR in Group 3 PH¹⁵³, the RV has time to compensate, often resulting in RV hypertrophy without systolic dysfunction^{154, 155}. The prevalence of RV hypertrophy in chronic lung disease varies from 50% on echocardiogram of patients with restrictive lung disease¹⁵⁶ to 76% at autopsy of COPD patients¹⁵⁷. Diastolic dysfunction is the predominant form of RV dysfunction in patients with chronic lung disease as has been demonstrated both in COPD¹⁵⁸ and in healthy individuals exposed to acute hypoxia¹⁵⁹. This increase in RV thickness is accompanied by cardiomyocyte hypertrophy, cardiac ECM remodeling, alterations in cellular metabolism, and in some models, an increase in RV capillary density^160–163.

Whether by impaired ventilation/perfusion matching from alveolar destruction in COPD, loss of pulmonary capillary beds from fibrosis in ILD, hypoventilation in OHS, or hypoxic vasoconstriction due to low inspired oxygen tension at high-altitude, a common pathway linking chronic lung diseases and hypoxia to RV dysfunction in Group 3 PH is the elevation in RV afterload. In addition, pathologic changes in the lungs directly affect RV mechanics. Lung hyperinflation in COPD leads to decreased venous return and RV preload¹⁶⁴ along with mechanical compression of the ventricles¹⁶⁵. Similarly, the increased intrapleural pressure required for inspiration in COPD and during apneic episodes of sleep disordered breathing likely increase left atrial pressure and RV afterload¹⁶⁶.

A persistently elevated RV afterload will inevitably cause systolic dysfunction. Systolic dysfunction may be limited to the most severe cases of PH and underscores the tremendous ability of the cardiovascular system to compensate to stress¹⁶⁷. As in Group 1 PAH, there is sexual dimorphism in RV systolic function in Group 3 PH. Males had worse RV systolic function assessed by transthoracic echocardiography as compared to females in one study¹⁵², despite females having a significantly higher PVR and a trend toward lower pulmonary artery compliance. As PVR increased in males, RV contractility worsened, whereas it was preserved in females, suggesting a sex-based differential RV response to pulmonary afterload in Group 3 PH¹⁵².

Diagnosis of Group 3 PH

Echocardiography is the initial test of choice in the diagnosis of PH¹⁰⁹ as it is noninvasive and can identify other etiologies of PH including left-heart disease, which is highly prevalent in patients with chronic lung disease^{168, 169}. However, pulmonary artery pressure estimates from echocardiography alone are often inaccurate and may overdiagnose PH in patients with chronic lung disease¹⁷⁰. Plasma levels of brain natriuretic peptide (BNP) or NT-proBNP may also be useful and were strong predictors of mortality in a mixed ILD cohort¹⁷¹. However, BNP levels lack sensitivity and can be confounded by the presence of left-heart disease¹⁷².

Pulmonary function tests establish obstructive, restrictive and mixed ventilatory defects in patients with lung disease. The diffusing capacity of carbon monoxide (D_LCO) may be particularly revealing in patients with Group 3 PH and is associated with lower survival¹⁷³. In COPD and ILD, PH is associated with a lower D_LCO than expected based on ventilatory impairments^{107, 110, 112, 174, 175}. The six-minute walk test (6MWT) has several features which may help detect PH complicating primary lung disease. A reduction in six-minute walk distance (6MWD) and pronounced desaturation during the walk are associated with the presence of PH and reduced survival in chronic lung disease¹⁷⁶. Reduced heart rate recovery (maximal heart rate achieved minus heart rate at one minute post exercise < 13 beats/min) has also been shown to predict PH in patients with IPF¹⁷⁷. Computed tomography (CT) is widely used in patients with lung disease to assess the burden of parenchymal lung abnormalities and/or to screen for lung cancer. A ratio > 1 (range 0.9 – 1.1) of the main pulmonary artery to ascending aorta diameter on CT imaging may predict PH in both COPD and IPF^178–180 and combined with other non-invasive measures improves the accuracy of predicting hemodynamic PH in this population^{174, 179}. Finally, an RV to LV ratio > 1 measured on CT is associated with increased PVR and predicts mortality in ILD patients¹⁸¹. Table 3 lists findings that when present increase the likelihood of PH in patients with chronic lung diseases.

Table 3.

Findings suggestive of pulmonary hypertension in chronic lung disease

Pulmonary function testing	• DLCO < 30% or worsening DLCO in the setting of preserved lung volumes
6-Minute Walk Test	• Marked or worsening exertional desaturation • Severely reduced or worsening 6MWT distance • Impaired (decreased) heart rate recovery following exercise (< 13 beats per minute)
CT scan	• Increased pulmonary artery to aorta ratio (> 0.9) • RV to LV ratio > 1
Echocardiogram	• RVSP elevation > 45 • RV dilation • TAPSE < 1.8cm • RVOT diameter > 3.4cm • Reduced RV fractional area change • Reduced RV ejection fraction on 3D ECHO
Laboratory studies	• Elevated BNP

ILD = interstitial lung disease. D_LCO = diffusion capacity of carbon monoxide. 6MWT = 6-minute walk test. RV = right ventricle. LV = left ventricle. RVSP = right ventricular systolic pressure. TAPSE = tricuspid annular plane systolic exertion. RVOT = right ventricular outflow tract. ECHO = echocardiogram. BNP = brain natriuretic peptide. Adapted from King, et al. CHEST. 2020.²⁴⁵

Right heart catheterization (RHC) remains the gold standard for the diagnosis of pulmonary vascular disease including Group 3 PH. RHC is required in the evaluation of Group 3 PH to assess for left ventricular fluid and/or pressure overload – i.e., to distinguish pre- vs. postcapillary PH – and to define PH severity. Hemodynamics are needed and often repeated to evaluate patients with clinical worsening or progressive gas exchange abnormalities disproportionate to obstructive or ventilatory impairments, to evaluate for lung transplant and for inclusion in clinical trials or registries¹⁸². In cases where off-label therapy is being considered to treat non-ILD forms of Group 3 PH, or to initiate inhaled treprostinil for ILD-associated PH, RHC is required.

Patients with chronic lung disease are at risk for both precapillary and postcapillary PH due to concurrent left heart disease. As patients with chronic lung disease tend to have significant respiratory variation, several breath cycles should be captured to measure hemodynamics including the pulmonary capillary wedge pressure (PCWP) which should be measured at end-expiration¹⁰⁹. In the setting of underlying parenchymal lung disease, Group 3 PH is diagnosed when the mPAP is > 20 mmHg, the PCWP is ≤ 15 mmHg, and the PVR is ≥ 3 WU¹. Patients with risk factors for post-capillary PH, such as obesity, systemic hypertension, atrial fibrillation, diabetes mellitus and OSA, and echocardiographic evidence of diastolic dysfunction, should be considered for provocative maneuvers such as fluid loading and exercise, to determine a dominant driver of pulmonary pressures. The 2011 German Cologne PH Conference and the most recent WSPH statement introduced a concept of chronic lung disease with severe PH, defined as mPAP ≥ 35 mmHg or mPAP ≥ 25 mmHg with a low cardiac index (< 2.0 L/min/m²)^{109, 183}. This distinction is supported by evidence that patients with COPD and PH with a mPAP ≥ 40 mmHg demonstrate functional impairment due to circulatory-limitation as compared to COPD patients with moderate PH (mean mPAP of 31 mmHg)¹⁸⁴. These observations have been recapitulated in IPF patients with mPAP ≥ 35 mmHg^{87, 105, 185}. As discussed above, recent registry data suggests that PVR thresholds may provide improved prognostication over mPAP values^{97, 114} which may reflect the degree that combined assessment of pulmonary pressures and the RV response are more prognostic in this population. There are no data to support routine vasodilator testing in Group 3 PH⁹⁹, although in patients with chronic hypoxia it may be important to assess for a hemodynamic response to oxygen at the time of catheterization.

Differentiating Group 1 PAH from Group 3 PH can be challenging in many patients, particularly in those with underlying CTD. The degree of pulmonary vascular disease and parenchymal lung involvement exists on a continuum for many patients, however there are distinguishing features that may help to define the predominant pathophenotype (Table 4)¹⁰⁹.

Table 4.

Distinguishing features of Group 1 pulmonary arterial hypertension and Group 3 pulmonary hypertension

Criteria Favoring Group 1 PAH	Parameter	Criteria Favoring Group 3 PH
Extent of Lung Disease
• Normal or mildly impaired spirometry (FEV1 >60% predicted in COPD; FVC >70% predicted in IPF) • Low DLCO in relation to obstructive/restrictive changes (i.e. DLCO “out of proportion” to spirometry/lung volumes)	Pulmonary function testing	• Moderate to severely impaired spirometry (FEV1 <60% predicted in COPD; FVC <70% predicted in IPF) • DLCO corresponds to obstructive/restrictive changes (i.e. DLCO “in proportion” to spirometry/lung volumes)
Absence of or only moderate parenchymal abnormalities	High resolution CT scan	Characteristic parenchymal abnormalities
Hemodynamic Profile
Moderate-to-severe PH	Right Heart Catheterization or Echocardiogram	Mild-to-moderate PH
Ancillary Testing
Present	Additional PAH risk factors (e.g. family history or identified mutations associated with PAH, connective tissue disease, HIV, history of drug/toxin exposure, etc)	Absent
Features of circulatory limitation to exercise: • Preserved breathing reserve • Reduced O2 pulse • Low CO/VO2 slope • Mixed venous oxygen saturation at lower limit • No change or decrease in PaCO2 during exercise	Cardiopulmonary exercise test	Features of ventilatory limitation to exercise: • Reduced breathing reserve • Normal O2 pulse • Normal CO/VO2 slope • Mixed venous oxygen saturation above lower limit • Increase in PaCO2 during exercise

PAH = pulmonary arterial hypertension. PH = pulmonary hypertension. FEV₁ = forced expiratory volume in the first second. COPD = chronic obstructive pulmonary disease. FVC = forced vital capacity. IPF = idiopathic pulmonary fibrosis. D_LCO = diffusion capacity of carbon monoxide. CT = computed tomography. HIV = human immunodeficiency virus. Adapted from Nathan, et al. ERJ. 2019.¹⁰⁹

Therapeutic Strategies

Treatment of the underlying lung disease remains the initial therapeutic strategy in Group 3 PH including bronchodilators in COPD, immunosuppression and antifibrotics in ILDs, and nocturnal positive pressure therapy in sleep disordered breathing^{125, 126, 128}. Appropriate and timely lung transplant referral is essential in patients with advanced lung disease, given the lack of effective therapies and that the development of PH is associated with worse outcomes. As such, the presence of moderate to severe PH in COPD and evidence of PH by echocardiography or RHC in ILD is a specific indication for lung transplantation listing¹⁸⁶. Long-term oxygen therapy is frequently prescribed for PH patients who are hypoxic, however this has only been prospectively evaluated in COPD where it was noted to prevent an increase in mPAP and slightly decrease mPAP when used for >18 hours per day^{187, 188}. There are no data examining the benefit of long-term oxygen on clinical outcomes or survival in PH and this remains a significant knowledge gap. RV dysfunction and pulmonary artery pressures (as measured by transthoracic echocardiography) appear to transiently increase during COPD exacerbations, physiology that should be taken into account when initiating therapy in this patient population¹⁸⁹.

Pulmonary Vasodilators

Pulmonary vasodilators have long been theorized to worsen gas exchange in patients with chronic lung disease due to worsening of hypoxic vasoconstriction and ventilation/perfusion mismatch, although this has not been substantiated in clinical studies⁹⁹. When delivered via an inhaled route, pulmonary vasodilators may access better ventilated areas of the lung, a feasible paradigm for the treatment of Group 3 PH^190–192. On balance, PAH therapies have modest and inconsistent benefits in Group 3 PH patients¹⁹³. Because of the complexities of adequately phenotyping patients, who may have overlapping features from multiple WSPH groups, and the potential for harm with pulmonary vasodilators, consideration of PH-specific therapy should be done at expert centers.

Endothelin receptor antagonists (ERAs) have no evidence of benefit and may be harmful in ILD-PH patients. Bosentan was evaluated in a randomized clinical trial (RCT) of 60 participants with fibrotic ILD and PH confirmed by invasive hemodynamics. Neither the primary endpoint (reduction in PVR index by 20% at 16 weeks) nor secondary endpoints including functional capacity, symptoms, adverse events or death were significantly different between the bosentan group and placebo¹⁹⁴. The ARTEMIS-IPF trial enrolled 492 participants with IPF to ambrisentan or placebo. The trial was terminated early due to an increase in IPF disease progression and hospitalizations and a trend toward increased mortality and decreased lung function in the ambrisentan group¹⁹⁵.

There have been numerous RCTs of phosphodiesterase-type 5 (PDE5) inhibitors in patients with chronic lung disease. A study of tadalafil in 120 COPD participants failed to improve 6MWD or quality of life¹⁹⁶. A pilot double-blind, placebo-controlled RCT of sildenafil in COPD-PH (SPHERIC-1) demonstrated a decrease in PVR and improved quality of life metrics with no effect on 6MWD¹⁹⁷. STEP-IPF examined the role of sildenafil in 180 IPF participants regardless of the presence of PH. Although there was no improvement in the primary endpoint of 6MWD, a post-hoc subgroup analysis demonstrated that IPF patients with RV dysfunction randomized to sildenafil had a lower decline in 6MWD and improvements in two quality of life measurements^{198, 199}. A more recent controlled trial examined the addition of sildenafil to pirfenidone in severe IPF patients (D_LCO ≤ 40%) at risk of developing PH, but when compared to placebo, found no difference in prevention of disease progression²⁰⁰.

Riociguat, a soluble guanylate cyclase stimulator, was studied in RISE-IIP, a randomized, double-blind placebo-controlled trial of 147 ILD-associated PH participants with a primary endpoint of 6MWD. This trial was stopped early due to an increase in adverse events and mortality in the riociguat group with no improvement in the 6MWD²⁰¹.

Parenteral prostanoid use has only been reported in case reports or small series in Group 3 PH. In an open-label study of parenteral treprostinil in severe PH associated with ILD (mPAP > 35 mmHg), there was a modest improvement in 6MWD, RV function, and hemodynamics²⁰².

Results of the recently published INCREASE trial⁵⁶ of inhaled treprostinil support the hypothesis that inhaled pulmonary vasodilators could cause less ventilation-perfusion mismatch in Group 3 PH compared to systemic therapies. This trial was a randomized, double-blind, placebo-controlled multicenter study that evaluated inhaled treprostinil administered by an iNEB (adaptive aerosol delivery system) device in 326 participants with ILD-associated PH with change in 6MWD at week 16 as the primary endpoint. The least-squares mean difference in the change from baseline 6MWD with treprostinil compared to placebo was 31 m (95% CI, 17 – 45; p < 0.0001) and treprostinil reduced clinical worsening (hazard ratio 0.61, 95% CI 0.40 – 0.92; p < 0.04). The reduction in clinical worsening was driven by the improvement in 6MWD and treprostinil did not improve quality of life. There tended to be more women and CTD patients in the treprostinil group, which may have biased the results as these groups tend to respond more favorably to PAH therapy. A post-hoc analysis demonstrated a placebo-corrected 169 mL improvement in forced vital capacity (FVC) in a subgroup of participants with IPF and a 108 mL improvement in FVC in a subgroup of participants with IIPs at week 16. Participants with IIPs on inhaled treprostinil had numerically fewer exacerbations²⁰³. These results, along with preclinical data suggesting that treprostinil has direct antifibrotic effects on the lung^{204, 205}, have led to the ongoing TETON trial of IIP without PH (NCT04708782). In another post hoc analysis of INCREASE, treatment with inhaled treprostinil resulted in decreased likelihood of disease progression defined as either ≥ 5% decline in 6MWD, ≥ 10% decline in FVC, acute exacerbation, cardiopulmonary hospitalization, lung transplantation, or death (22 vs 36% multiple progression events in treatment vs placebo groups, respectively; p = 005)²⁰⁶. The United States Food and Drug Administration approved inhaled treprostinil for Group 3 PH due to ILD on April 1, 2021, marking the first and only approved PAH therapy for Group 3 PH due to ILD in the United States.

Registry data mirrors what has been learned from clinical trials to-date in Group 3 PH. Hoeper and colleagues compared the safety and efficacy of off-label PAH therapy in 151 ILD-associated PH patients with severe PH and RV dysfunction (mPAP 37 ± 9 mmHg and cardiac index 2.1 ± 0.61 L/min/m²) in COMPERA²⁰⁷. The majority of this group was treated with PDE5 inhibitors and showed a modest improvement in 6MWD and functional class. A subsequent analysis of COPD-associated PH in COMPERA demonstrated a 30 m increase in 6MWD²⁰⁸. Despite these improvements in intermediate end points, both analyses confirmed that these groups had a significantly worse survival as compared to idiopathic PAH patients. Group 3 PH patients in the Scottish Pulmonary Vascular Unit treated with pulmonary vasodilators (primarily PDE5 inhibitors) demonstrated no improvement in 6MWD, but a significant decrease in serum NT-proBNP levels²⁰⁹. Patients with CPFE had a significant reduction in serum NT-proBNP levels and an improvement in 6MWD of 41 m (above the minimally important difference of 33 m²¹⁰), whereas this observation was not seen in patients with ILD or emphysema. Thenappan and colleagues performed a meta-analysis of five RCTS in COPD-PH, two RCTs in ILD-PH, and four single-arm clinical trials in ILD-PH². Although there was a signal toward improved hemodynamics, PAH therapy (the majority with PDE5 inhibitors in COPD-PH and PDE5 inhibitor, ERA, riociguat, and parenteral treprostinil in ILD-PH) showed no symptomatic benefit in COPD-PH or ILD-PH though treatment did not worsen hypoxemia. The single-arm ILD-PH studies demonstrated improvement in 6MWD, however these findings were not recapitulated in either the COPD-PH or ILD-PH RCTs.

In summary, data supporting the use of PAH therapy in Group 3 PH are limited and further study is needed. Vasoactive medications may have a benefit in parenchymal lung disease patients with severe hemodynamic impairment with depressed cardiac output, which may identify a target population for future studies. Riociguat and ambrisentan are now contraindicated in PH associated with ILD, other ERAs should be avoided and evidence is conflicting or limited for the use of PDE5 inhibitors. Inhaled treprostinil is now approved for use in patients with PH complicating ILDs in the United States.

Emerging Therapies

Inhaled treprostinil is currently being studied in the PERFECT trial in patients with COPD-associated PH (NCT03496623). Inhaled treprostinil requires at least four nebulized inhalations each day and its major side effect is cough. A treprostinil-prodrug based nanoparticle with slower time release is in development and may have less cough associated with its use^211–214. PULSE, a clinical trial of inhaled nitric oxide in PH-ILD, has demonstrated favorable patient reported outcomes²¹⁵ and increased physical activity in a Phase II trial²¹⁶ and is currently being studied in a Phase III trial (NCT03267108). Bardoxylone methyl, a novel agent with broad ranging anti-inflammatory effects, is also being studied in PH-ILD (NCT02036970).

Since Group 3 PH patients have worse RV dysfunction than other subgroups for a given RV afterload, therapies that target RV remodeling and dysfunction may be of particular benefit to Group 3 PH patients. Early work examined digoxin as one such agent, without evidence of benefit²¹⁷. As in left ventricular dysfunction, mechanical assist devices for the RV may one day provide symptomatic relief in end-stage patients and serve as destination therapy or a bridge to transplant. Initial data is promising²¹⁸ and a feasibility trial is currently underway to examine the safety of a device targeting pulmonary arterial compliance in Group 3 PH patients (NCT05001711).

Conclusion

Group 3 PH encompasses a group of patients with highly prevalent lung diseases and poor outcomes. Discrete pathobiological mechanisms in this population remain an active area of study as does the development of novel therapeutic targets. Recent advances in therapeutic options include more robust approaches to clinical phenotyping and the benefit of inhaled treprostinil in ILD-associated PH, which will hopefully stimulate further research for effective therapies in all patients suffering from Group 3 PH.

Table 5.

Clinical trials evaluating pulmonary vasodilators in Group 3 pulmonary hypertension

Drug Class	Author (year)	Number of Participants	PH Definition	Lung Disease	Treatment	Duration	Primary Outcome	Result
ERA	Stolz et al246 (2008)	30	Echo PASP ≥ 30mmHg	COPD	Bosentan 125mg BID	12 weeks	Change in 6MWD	No difference
	Valerio et al247 (2009)	32	RHC mPAP ≥ 25mmHg	COPD	Bosentan 125mg BID	18 months	PFTs, hemodynamics, 6MWD, dyspnea ratings/quality of life	Improvement in mPAP, PVR, 6MWD, dyspnea ratings
	Raghu et al195 (2013)	492	RHC mPAP ≥ 25mmHg	ILD	Ambrisentan 10mg daily	16 weeks	Time to ILD progression	No difference; stopped early due to increased adverse events
	Corte et al194 (2014)	60	RHC mPAP ≥ 25mmHg	ILD	Bosentan 125mg BID	16 weeks	Decrease in PVRi from baseline ≥20%	No difference
PDE5 inhibitor	Collard et al248 (2007)	11	Echo PASP > 35mmHg or RHC mPAP ≥ 25mmHg	ILD	Sildenafil 20 – 50mg TID	12 weeks	Change in 6MWD	Significant improvement
	Han et al198, 199 (2010)	180	N/A	ILD	Sildenafil 20mg TID	12 weeks	20% improvement in 6MWD	No difference except in subgroup with RV systolic dysfunction
	Rao et al249 (2011)	33	Echo PASP > 40mmHg	COPD	Sildenafil 20mg TID	12 weeks	Change in 6MWD	Significant improvement
	Goudie et al196 (2014)	120	Echo PASP > 30mmHg or PA-AT < 120ms	COPD	Tadalafil 10mg daily	12 weeks	Change in 6MWD	No difference
	Vitulo et al197 (2017)	28	RHC mPAP ≥ 35mmHg and post-bronchodilator FEV1 < 30% or mPAP ≥ 30mmHg and post-bronchodilator FEV1 > 30%	COPD	Sildenafil 20mg TID	16 weeks	Reduction in PVR	Significant improvement
	Behr et al200 (2021)	177	RHC mPAP ≥ 20mmHg	ILD	Sildenafil 20mg TID + Pirfenidone 801mg TID	52 weeks	Disease progression	No difference
Prostacyclin analogue	Saggar et al202 (2014)	15	RHC mPAP ≥ 35mmHg	ILD	Treprostinil 34 ± 21ng/kg/min	12 weeks	Hemodynamics, 6MWD, PFTs, ECHO, dyspnea ratings/quality of life	Improvement in hemodynamics, 6MWD, dyspnea ratings
	Waxman et al56, 203 (2021)	326	RHC mPAP ≥ 25mmHg	ILD	Treprostinil 72 μg QID	16 weeks	Change in 6MWD	Significant improvement; Improvement in FVC in IPF subgroup
sGC stimulator	Nathan et al201 (2019)	147	RHC mPAP ≥ 25mmHg	ILD	Riociguat 0.5 – 2.5mg TID	26 weeks	Change in 6MWD	No difference; stopped early due to increased adverse events

PH = pulmonary hypertension. ERA = endothelin receptor antagonist. RHC = right heart catheterization. mPAP = mean pulmonary artery pressure. ILD = interstitial lung disease. BID = two times per day. PVRi = pulmonary vascular resistance index. PVR = pulmonary vascular resistance. PFT = pulmonary function test. 6MWD = 6 minute walk distance. ECHO = echocardiogram. PDE5 = phosphodiesterase type-5. PASP = pulmonary artery systolic pressure. TID = three times per day. PA-AT = pulmonary artery acceleration time. FEV₁ = forced expiratory volume in the first second. QID = four times per day. sGC = soluble guanyl cyclase.