Heritable (Familial) Pulmonary Arterial Hypertension

Mendelian precapillary pulmonary vasculopathy, predominantly autosomal dominant with incomplete, sex-dependent penetrance; BMPR2 is the principal gene, twelve genes have definitive evidence, and autosomal recessive EIF2AK4-related pulmonary veno-occlusive disease is a form with venous/capillary involvement.

Literature and regulatory information updated: 6 October 2026.

0. Initial Reference Table

1. Introduction

Pulmonary arterial hypertension (PAH) is a rare severe disease characterised by obliteration of remodelled pulmonary microvessels, increased PVR, progressively elevated pulmonary arterial pressure, right ventricular hypertrophy and failure, and death if untreated [14]. The first description of pulmonary vascular disease as a distinct entity is attributed to Ernst von Romberg, who described “pulmonary vascular sclerosis” at autopsy in 1891; from the mid-20th century, right heart catheterisation enabled recognition during life of what was then called primary pulmonary hypertension [14]. A genetic origin was proposed in 1954 by Dresdale and colleagues, observing familial cases; autosomal dominant transmission in 14 families was described in the 1980s, and in the late 1990s PPH1 was mapped to 2q31–32 then 2q33 [14]. OMIM assembles classical family descriptions: Melmon and Braunwald documented two proven and three presumed cases across three generations in 1963; Loyd and colleagues (1984) described six deaths across two generations, a 2:1 female predominance and male-to-male transmission excluding X-linked inheritance, including transmission from an affected man through two generations of unaffected women to a boy dying aged 6; and in the National Institutes of Health registry, 6% of 187 cases were familial [4]. In 1997, Morse and colleagues and Nichols and colleagues mapped the locus to 2q31–32, the latter with maximum multipoint lod score 7.86 [4].

In 2000, two independent groups identified the responsible gene. Deng and colleagues genotyped 35 multiplex families, constructed disease haplotypes and, examining three candidate genes, found five BMPR2 variants predicting premature protein termination and two missense variants, absent from 196 control chromosomes [22]. The International Primary PH Consortium (Lane and colleagues) showed that same year that familial primary PH, present in at least 6% and inherited autosomal dominantly with reduced penetrance, is caused by heterozygous germline BMPR2 variants encoding a TGF-β superfamily type II receptor, with predicted defects in ligand binding, kinase activity and heterodimer formation [23]. The Consortium found BMPR2 variants in 7 of 8 linked families and Deng and colleagues in 9 of 19 families with linkage or shared haplotypes; Thomson and colleagues found variants that year in 13 of 50 sporadic primary PH patients, two de novo among five with parental testing possible [4]. In 2001, Trembath and colleagues studied five families and one isolated patient with hereditary haemorrhagic telangiectasia (HHT) and ten PH cases, finding 12q13 linkage and type I receptor ACVRL1 (ALK1) variants in patients whose disease was clinically and histologically indistinguishable from primary PH [24].

Genetic architecture subsequently expanded: SMAD8/SMAD9 in 2009 [25], CAV1 through exome sequencing in 2012 [26], KCNK3 as the first channelopathy in 2013 [27], TBX4 in children with PAH and small patella syndrome in 2013 [28], and EIF2AK4 as a recessive PVOD cause in 2014 [29]. Genome sequencing of 1,038 index cases and 6,385 controls added ATP13A3, AQP1 and SOX17 and validated GDF2 [30]. At the 4th WSPH in 2008, heritable PAH became a distinct group 1 subcategory, and the 2022 classification retains subgroup 1.2; approximately 3% of all PAH is classified as HPAH and 40% as idiopathic [14].

Three conceptual changes structure this article. The first is haemodynamic: the diagnostic threshold for PH has fallen from mPAP ≥25 mmHg to >20 mmHg, and the PVR defining the pre-capillary component has changed from >3 to >2 WU [2, 3]. The second concerns gene validity: an international panel applied the Clinical Genome Resource (ClinGen) methodology and recognised 12 genes with definitive evidence, 3 with moderate evidence, 6 with limited evidence and 5 disputed, which should not be included in diagnostic panels [12]. The third is therapeutic: sotatercept, an activin signalling pathway ligand trap, was the first drug to act on a new pathway in almost two decades and made four-pathway therapy possible [16]. The non-pulmonary manifestations of ACVRL1, ENG and SMAD4 variants (epistaxis, telangiectasias, arteriovenous malformations) belong to HHT; this article covers PAH associated with these genes, its detection and the specific therapeutic implications [21].

2. Molecular Genetics and Pathophysiology

2.1. The bone morphogenetic protein pathway and vascular injury

BMPR2 (2q33.1–q33.2) encodes bone morphogenetic protein receptor type II (BMPR-II), a transmembrane serine/threonine kinase of the TGF-β superfamily. Following BMP2, BMP4, BMP7 or GDF6 binding, a complex of two type II and two type I receptors forms; type II receptors phosphorylate and activate type I receptors, which activate SMAD transcription factors, with alternative p38 MAPK signalling [31]. In pulmonary endothelium, type I receptor ACVRL1 (ALK1) and coreceptor endoglin (ENG) respond to circulating BMP9 (encoded by GDF2) and BMP10, and intracellular effector SMAD9 (SMAD8) conveys signals to the nucleus; this explains clinical overlap between variants in multiple components of the same axis [31, 32]. In pulmonary circulation, signalling restrains endothelial and smooth muscle proliferation; its loss promotes proliferation and apoptosis resistance, with smooth muscle cell and fibroblast proliferation, neointimal lesions and plexiform nests of proliferating endothelial cells narrowing small pulmonary arteries [14, 22]. Remodelling affects every wall layer — intimal thickening, smooth muscle hypertrophy or hyperplasia, adventitial fibrosis and in situ thrombotic occlusion — and plexiform lesions, present in 20–80% of primary PH, showed monoclonal endothelial proliferation in 17 of 22 lesions (77%), versus polyclonal lesions in secondary PH [4].

Haemodynamic consequences are increased PVR and pulmonary arterial pressure, hypertrophy and ultimately right ventricular failure [14, 22]. Current treatments act on four pathways: endothelin receptor antagonists block endothelin-1's vasoconstrictor and proliferative effects; PDE5 inhibitors and soluble guanylate cyclase (sGC) stimulators increase cyclic guanosine monophosphate (cGMP) in the nitric oxide pathway; prostanoids and prostacyclin receptor (IP receptor) agonists increase cyclic adenosine monophosphate (cAMP); and sotatercept, a fusion protein trapping activins and growth differentiation factors, aims to rebalance growth-promoting and growth-inhibiting TGF-β superfamily signalling [2, 16]. In a nine-patient pilot, a peripheral-blood BMPR-II pathway marker transcriptomic panel suggested sotatercept reduced rather than increased BMPR-II signalling, possibly through circulating BMP9 and BMP10 depletion; this is a preliminary mechanistic finding, not clinical efficacy evidence [33].

2.2. Genetic architecture and gene–disease validity

An international expert panel applied ClinGen's semiquantitative system to genetic and experimental evidence for PAH genes. Twelve were definitive, three moderate, six limited (AQP1, BMP10, FBLN2, KLF2, KLK1, PDGFD), TOPBP1 had no known relationship and five were disputed (BMPR1A, BMPR1B, NOTCH3, SMAD1, SMAD4); the authors recommend including all definitive genes, cautiously interpreting variants in moderate or limited genes and excluding disputed or unrelated genes [12]. OMIM numbers seven primary PH forms: PPH1 (BMPR2), PPH2 (SMAD9), PPH3 (CAV1), PPH4 (KCNK3), PPH5 (ATP13A3), PPH6 (CAPNS1) and PPH7 (SOX17), and associates PAH with HHT types 1 and 2 and ischiocoxopodopatellar syndrome; CAPNS1 is absent from 2023 ClinGen curation and should not be interpreted with the same evidence level [4, 12]. The 7th WSPH adopted this classification, anticipates gene reassessment every 3 years and emphasises that disputed genes originate from the candidate-gene era with predominantly experimental evidence [15].

AD, autosomal dominant; AR, autosomal recessive; DLCO, diffusing capacity for carbon monoxide; AVMs, arteriovenous malformations. Sources: classification, mechanism and year from 7th WSPH and ClinGen; phenotypes and proportions from GeneReviews based on PAH Biobank; cytobands from NCBI Gene [1, 12, 13, 15, 34].

The proportion attributed to each gene depends on the denominator. Among patients with familial PAH, 70–87% have an identifiable cause in the known genes, compared with 12–20% of patients with idiopathic PAH [14]. The French PNDS, based on the experience of the national reference centre, gives figures of the same order: a genetic cause identified in about 20% of sporadic PAH, 85% of familial PAH and all families with PVOD [19]. The figure of about 10% that the PNDS attributes to sporadic PVOD is lower than the primary series from the same centre: biallelic EIF2AK4 variants in 5 of 20 histologically confirmed sporadic cases (25%) in the original study and in 27 of 94 patients with sporadic or heritable PVOD/PCH (29%) in the French registry of 2003–2016 [29, 35]. In the North American PAH Biobank, which included 2,572 group 1 patients (1,211 with a previous clinical diagnosis of familial or idiopathic PAH), BMPR2 was identified in 8% of the whole cohort but accounts for about 75% of carriers of variants in PAH genes [1]. The same study identified KLK1 and GGCX as novel candidate genes for idiopathic PAH, in about 0.4% and 0.9% of cases, with a later mean age at onset and milder phenotypes than BMPR2 carriers [34]. An analysis of 4,241 cases from an international consortium associated idiopathic PAH with BMPR2, GDF2, TBX4, SOX17, KDR and, as novel candidates, FBLN2 and PDGFD, and estimated that about 15% of paediatric idiopathic PAH may be explained by de novo variants [36].

2.3. BMPR2

More than 800 independent pathogenic BMPR2 variants have been described [14]. Most cause loss of function with reduced penetrance, but alleles with a dominant-negative effect also exist; large deletions and duplications and non-coding variants are documented, which makes copy-number variation (CNV) analysis mandatory [1, 14]. In sequencing-negative patients (25 families and 106 sporadic cases), MLPA identified BMPR2 deletions in 7 families and 6 sporadic cases, including null deletions that support haploinsufficiency as the predominant mechanism [4]. Loss-of-function BMPR2 variants have also been associated with congenital heart disease, and the gene is linked to three OMIM entries: familial primary PH 1 with or without HHT, fenfluramine- or dexfenfluramine-associated primary PH (both under 178600) and PVOD 1 (265450) [1, 31]. In a retrospective study of patients with fenfluramine-associated PAH evaluated between 1986 and 2004, with a median exposure of six months and a median interval of 4.5 years between exposure and symptoms, 9 of the 40 patients tested (22.5%) had a pathogenic BMPR2 variant [1]. In another series, of 33 unrelated patients exposed to fenfluramine derivatives, 3 (9%) had BMPR2 variants, as did two affected sisters; carriers became ill after shorter exposure (median 1 month versus 4 months), suggesting an interaction between the variant and the drug [4]. GeneReviews also records eight patients with BMPR2 variants and pulmonary AVMs, five with multiple AVMs, three with epistaxis and three without PAH, so BMPR2 also enters the genetic differential diagnosis of HHT [21]. The first report was of a woman with PAH at 24 years, massive haemoptysis at 35 from multiple pulmonary AVMs, epistaxis and nasal telangiectases, carrying the BMPR2 nonsense variant p.Gln433Ter (OMIM designation, = NM_001204.7:c.1297C>T p.(Gln433Ter)) and without variants in ACVRL1, ENG or SMAD4 [4].

An individual-patient-data meta-analysis of 1,550 idiopathic, heritable or anorexigen-associated PAH patients from eight systematically tested cohorts found BMPR2 variants in 448 (29%). Carriers were younger at diagnosis (35.4 ± 14.8 versus 42.0 ± 17.8 years), with higher mPAP (60.5 ± 13.8 versus 56.4 ± 15.3 mmHg), higher PVR (16.6 ± 8.3 versus 12.9 ± 8.3 WU), lower cardiac index (2.11 ± 0.69 versus 2.51 ± 0.92 L/min/m²) and less acute vasoreactivity response (3% versus 16%) [37]. A meta-analysis of 17 non-randomised studies (2,190 patients) confirmed differences in mPAP (weighted mean difference 6.41 mmHg), PVR (3.66 WU), cardiac index (−0.38 L/min/m²) and cardiac output (−0.60 L/min) at diagnosis between carriers and non-carriers [38].

2.4. Other BMP pathway genes: ACVRL1, ENG, GDF2 and SMAD9

ACVRL1 and ENG variants occur predominantly in HHT, approximately 2% developing PAH [14]. The HHT GeneReviews chapter estimates PAH in approximately 1–2%, indicates that group 2 PH is much more frequent and distinguishes high-output PH from systemic arteriovenous shunting through hepatic AVMs (or less often severe anaemia) from PH clinically and histologically indistinguishable from idiopathic PAH; the latter, without significant shunting, occurs mainly with ACVRL1 variants [21]. ACVRL1-associated PAH has median onset at 20 and may be the first obvious sign of HHT manifesting later [14]. The Oxford Desk Reference already recorded primary PH as a rare complication in some ALK1-variant families and noted that ENG, ACVRL1 and BMPR2 belong to the same receptor superfamily [39].

GDF2 encodes BMP9, synthesised as a proprotein that dimerises and undergoes furin cleavage; rare missense variants from PAH patients disrupt cleavage and reduce mature protein, and carriers have reduced plasma BMP9 [15]. Two paediatric patients homozygous for GDF2 nonsense variants, one with PAH and homozygous NM_016204.4:c.76C>T p.(Gln26Ter), the other with pulmonary AVMs and homozygous p.(Glu279Ter), both with facial telangiectases, had undetectable plasma BMP9 and BMP10; neither met Curaçao criteria [32]. HHT GeneReviews includes GDF2-related vascular anomaly syndrome in the differential: isolated PAH is most frequent, telangiectases when reported are usually atypical and only two patients met HHT clinical criteria [21]. The first SMAD8 (SMAD9) case was nonsense variant NM_001127217.3:c.606C>A p.(Cys202Ter), found in idiopathic PAH and the patient's father; truncated protein was neither phosphorylated by type I receptors nor interacted with SMAD4 [25, 40]. The proband was a boy diagnosed aged 8, with two siblings dying of pulmonary disease aged 13 and before 2, and an unaffected carrier father, indicating reduced penetrance; OMIM also records a missense variant (p.Lys43Glu) in a Japanese girl aged 7 requiring transplantation and a truncating variant (p.Arg294Ter) in a woman whose endothelial cells failed to induce miR-21 in response to BMP9 [5].

2.5. Genes outside the BMP pathway

CAV1. Caveolin-1 is a caveolar protein abundant in pulmonary endothelium. Exome sequencing identified a frameshift variant in a three-generation family and another de novo variant in a child with idiopathic PAH; endothelial caveolin-1 density in small arteries was reduced [26]. Six relatives had primary PH diagnosed between ages 4 and 67, one died aged 4 and several carriers were unaffected; the de novo variant occurred in a girl diagnosed at 15 months, the only positive among 260 additional patients, making CAV1 a rare cause [6]. Sources differ on mechanism: GeneReviews describes reduced-penetrance loss of function, the ESC table gain of function and dominant-negative effect, and 7th WSPH curation dominant-negative effect [1, 2, 15]. Biallelic loss causes congenital generalised lipodystrophy type 3 and heterozygous de novo variants cause neonatal progeroid lipodystrophy, in which PAH and cataract have been described; the ESC table suggests seeking deficient subcutaneous fat and measuring fasting triglycerides and leptin [2, 31].

KCNK3. This gene encodes the TASK-1 potassium channel maintaining resting membrane potential in pulmonary arterial smooth muscle cells. Ma and colleagues identified familial variant NM_002246.3:c.608G>A p.(Gly203Asp) and five other missense variants in 92 familial and 230 idiopathic PAH patients; all reduced potassium current, restored in vitro by phospholipase A2 inhibitor ONO-RS-082 [27, 40]. According to OMIM, variants occurred in 3 (3.2%) familial probands and 3 (1.3%) of 230 idiopathic patients; diagnostic age ranged from 8 to 44, some had arrhythmias, none responded to acute vasoreactivity testing, three required transplantation and three died in their fourth decade; 2 of 9 carriers were unaffected [7]. De novo gain-of-function variants clustered around the channel's “X gate” cause a distinct disorder, developmental delay with sleep apnoea, without PAH implications; interpreting a KCNK3 variant therefore requires its functional effect [31, 41].

ATP13A3. This encodes a P-type ATPase transporting polyamines from endosomes. In three families, biallelic variants caused treatment-refractory PAH beginning from birth to age 2.5, with four early-childhood deaths and asymptomatic heterozygous parents; the authors propose dose-dependent semidominant inheritance [42].

KDR. Vascular endothelial growth factor receptor 2 (VEGFR2) was associated through Bayesian inference in 13,037 NIHR BioResource participants with loss-of-function variants, reduced carbon monoxide transfer coefficient (KCO), older diagnostic age and computed tomography (CT) parenchymal abnormalities; four further similar cases were identified in PAH Biobank [43]. In two French families, loss-of-function KDR variants were associated with reduced corrected DLCO and interstitial lung disease or emphysema; some carriers had only reduced DLCO [44].

TBX4. T-box 4 regulates lower-limb, pelvic and lung development. Kerstjens-Frederikse and colleagues found TBX4 variants or encompassing 17q23.2 deletions in 6 of 20 children with PAH (30%), all with previously unrecognised small patella syndrome, versus 1 of 49 adults (2%), and found no PAH in 23 small patella syndrome patients [28]; autopsy in one paediatric TBX4 carrier revealed PVOD features [10]. In a cohort of 40 children with idiopathic or familial PAH, 3 (7.5%) had TBX4 variants, all inherited from parents with skeletal syndrome but no PAH; in a series of 19 patients with TBX4 variants or deletions, 10 had neonatal hypoxaemic respiratory failure with persistent PH of the newborn, were discharged and developed childhood PH, and half died or underwent transplantation [10]. Ischiopubic-patellar dysplasia (MIM 147891), including small patella syndrome, is autosomal dominant, caused by TBX4 or SOX9, and characterised by patellar hypoplasia or aplasia with recurrent dislocation, ischial ramus hypoplasia, foot abnormalities and sometimes brachymesophalangy; the skeletal dysplasia atlas describes a benign TBX4 course without PAH mention, reflecting its publication date [45]. OMIM 147891, “ischiocoxopodopatellar syndrome with or without PAH”, adopts Bongers and colleagues' minimum criteria (2001) — patellar aplasia or hypoplasia and absent, delayed or irregular ischiopubic junction ossification or “axe-cut” infra-acetabular notches — and major signs of widened first–second toe spacing and shortened fourth and fifth rays with flat feet; TBX4 causation was demonstrated in six families in 2004 [10]. Biallelic variants cause posterior amelia with pelvic and pulmonary hypoplasia syndrome (OMIM 601360), described in homozygous fetuses of consanguineous heterozygous couples with skeletal syndrome in Iran and India [1, 10]. Luciferase reporter functional studies identified gain-of-function TBX4 variants associated with later PAH onset than loss-of-function variants [15].

SOX17. In 256 congenital-heart-disease-associated PAH patients, rare deleterious variants of this Wnt/β-catenin and Notch transcription factor contributed approximately 3.2%, with missense variants concentrated in its HMG (high mobility group) domain; in 413 idiopathic or familial patients without heart defects, the proportion was 0.7% [46]. Beyond rare variants, common upstream SOX17 enhancer risk alleles confer small but significant risk (odds ratio 1.8) [15].

ABCC8, GGCX and TET2. These have moderate evidence: ABCC8 encodes ATP-sensitive potassium channel SUR1; GGCX, vitamin K-dependent γ-glutamyl carboxylase; and TET2, an epigenetic regulator [1, 12].

2.6. EIF2AK4 and pulmonary veno-occlusive disease

EIF2AK4 encodes GCN2 kinase, phosphorylating eIF-2α during amino acid deprivation, reducing cap-dependent translation and favouring stress-response messenger translation such as ATF4 [31]. Exome sequencing by Eyries and colleagues found biallelic EIF2AK4 variants segregating with disease in all 13 PVOD families and 5 of 20 histologically confirmed sporadic cases; all disrupted gene function [29]. These families, followed at a French reference centre and many of North African origin, had at least two affected siblings with unaffected parents; 22 distinct variants were identified, diagnostic age ranged from 11 to 50 (mostly third decade), almost all underwent transplantation or died, and no genotype–phenotype correlation explained onset variability [9]. Familial pulmonary capillary haemangiomatosis had been described in three French Canadian siblings dying with PH in 1988 [9]. Best and colleagues described EIF2AK4 variants in PCH in the same year [47]. PVOD and PCH share risk factors, genetic substrate and presentation and are considered one group 1 disease: postcapillary lesions of septal veins and preseptal venules, with loose fibrous intimal remodelling potentially occluding the lumen, often associated with capillary proliferation duplicating or triplicating alveolar septal capillary layers [2]. Biallelic EIF2AK4 accounts for at least 25% of all PVOD/PCH; heritable disease begins earlier than PVOD without biallelic variants [14]. How GCN2 loss produces lung-restricted venous and capillary remodelling remains unestablished [31]. A PVOD-like phenotype also occurs with BMPR2 variants (OMIM 265450, predating EIF2AK4 discovery), prompting some to consider PVOD and PAH a spectrum [14]. That entry includes a biopsy-proven woman whose mother died of severe PH, a patient with BMPR2 exon 2 deletion and PVOD histology (the same deletion existed in a family without known PVOD), and BMPR2 variants in 2 of 24 proven PVOD cases versus 4 of 24 without PVOD in a 48-patient series [8].

2.7. Penetrance, modifiers and second hits

Lifetime BMPR2 penetrance is approximately 14% in men and 42% in women [1]. ESC gives approximately 20% overall risk and PAH-ICON approximately 30%, both with female penetrance at least twice male penetrance, allowing disease to “skip” a generation [2, 14]. Analysis of 355 carriers from 53 Vanderbilt registry families with at least 57 observation years showed that apparent genetic anticipation (earlier onset in successive generations) disappears after correction for time-truncation bias [48]. OMIM 178600 still describes anticipation based on 1995 and 2000 studies and gives 10–20% penetrance and female:male ratio 1.7:1; later longitudinal analysis and GeneReviews sex-specific estimates should prevail [1, 4, 48]. TGFB1 polymorphisms modified diagnostic age (means 39.5 and 43.2 with less active genotypes versus 31.6 and 33.1) and penetrance (33%, 72% and 80% according to genotype activity) in BMPR2 carriers, an example of synergistic heterozygosity awaiting confirmation [4]. Rare oligogenic cases include a family with BMPR2 and EIF2AK4 variants, but this inheritance's contribution is unestablished [14].

Genetic burden differs by age. Rare variants explain approximately 42% of paediatric-onset and 12.5% of adult PAH, and de novo variants at least 15% of paediatric cases [49]. In a multinational 124-trio paediatric study, de novo excess was approximately 2.5 times expected [15]. Among environmental factors, anorexigen exposure is best characterised; in 47 HPAH families, 39 had a pathogenic panel variant, and two remaining families reported trichloroethylene, asbestos, tramadol or occupational toxin exposure [1, 50]. Pregnancy was a possible trigger in a BMPR2 carrier with an uncomplicated first pregnancy and severe PAH in the second [51].

2.8. Variants, nomenclature and allelic disorders

MANE conversion and genomic coordinates (GRCh38) were obtained in VariantValidator and confirmed in ClinVar; none was manually calculated [40]. The SMAD9 variant has the same genomic coordinate (NC_000013.11:g.36872722G>T) in historical transcript NM_005905.6 and MANE [40].

Geneticists must recognise several allelic disorders. ACVRL1 and ENG cause HHT types 2 and 1 respectively, and SMAD4 causes combined juvenile polyposis–HHT; SMAD4 is nevertheless disputed for isolated PAH [12, 21]. TBX4 causes small patella syndrome and, homozygously, posterior amelia with pelvic and pulmonary hypoplasia; CAV1 causes lipodystrophies; gain-of-function KCNK3 causes developmental delay with sleep apnoea; and biallelic GDF2 is associated with an HHT-like presentation [1, 31, 32, 41]. Oncological KDR literature on activating somatic mutations and VEGFR2 targeting does not apply to interpreting germline loss-of-function variants [31].

3. Clinical Spectrum and Manifestations by System

3.1. Presentation and natural history

HPAH may manifest from childhood to late adulthood. Initial symptoms are non-specific and develop slowly — dyspnoea, fatigue, chest pain, palpitations — and patients often attribute them to age, deconditioning or excess weight; the mean time from symptom onset to diagnosis is 1.9 years and the mean age at HPAH diagnosis is 34.9 ± 14.9 years [1]. Some patients report no symptoms and suspicion arises on physical examination: an accentuated pulmonary component of the second heart sound; a right ventricular heave, a tricuspid regurgitation murmur from ventricular dilatation or a diastolic pulmonary regurgitation murmur; signs of backward right ventricular failure (peripheral oedema, raised jugular venous pressure, ascites, hepatomegaly, abdominal distension); and signs of forward failure (cyanosis, pallor, cold extremities, delayed capillary refill) [1]. The electrocardiogram may show P pulmonale, right axis deviation, right ventricular hypertrophy, right bundle branch block and, in advanced disease, a strain pattern with ST depression and T-wave inversion in the right precordial and inferior leads; the chest radiograph is abnormal in most patients, but a normal radiograph does not exclude PH [2]. French data, which group all group 1 forms, quantify severity at presentation: exertional dyspnoea is present in more than 95% of patients and about 75% are in New York Heart Association (NYHA) functional class III or IV adapted to PH at the time of diagnosis; dilatation of the right heart chambers favours supraventricular arrhythmias (palpitations) and right ventricular ischaemia (chest pain) [19].

Progressive dilatation of the pulmonary artery gives rise to its own mechanical complications: pulmonary artery aneurysm, compression of the left main coronary artery (with acute coronary syndrome), of the main bronchi (with progressive ventilatory obstruction) or of the recurrent laryngeal nerve, and rupture or dissection, probably responsible for some sudden deaths [19]. Haemoptysis may occur during the course of the disease, favoured by anticoagulation, without necessarily signifying worsening of PAH; the HAS PNDS records it as more frequent in carriers of BMPR2 variants and in Eisenmenger syndrome (section 7.12) [19].

Untreated deterioration is progressive, with mean survival 2.8 years after diagnosis; variability ranges from sudden death to rare decades-long survival [1]. HPAH affects twice as many women as men [2], but in the North American Pulmonary Hypertension Association registry women had 48% lower death risk, attributed to better baseline right ventricular function and ERA/prostanoid response; female predominance is absent in paediatric onset, and children have worse diagnostic haemodynamics but post-diagnostic survival similar to adults [1].

3.2. Genotype–phenotype correlations

Sources: [1, 2, 14, 21, 37, 42–44, 49, 53].

Individual-patient-data meta-analysis showed BMPR2-associated excess risk was similar in men and women but greater with younger diagnosis, and attenuated after adjustment for PVR, cardiac index and vasoreactivity, suggesting haemodynamic severity partly mediates genotype effects [37]. For TBX4-associated PAH, GeneReviews summarises earlier onset, better six-minute walk performance, worse respiratory function, more airway abnormalities and longer event-free survival [1]. In the Spanish PAH registry, eight patients from seven TBX4 families (among 579 adults and 45 children tested) all had moderate-to-severe diffusion reduction in adulthood, with a spectrum including suspected PVOD, interstitial disease, pulmonary vascular abnormalities and congenital heart disease [53]. Classification depends on genetic testing availability: PH with lung developmental abnormalities may be group 1 (HPAH) or group 3 (lung developmental disease), with treatment implications [3]. A substantial proportion initially classified as congenital-heart-disease-associated PAH actually have HPAH, particularly SOX17 variants, found in 3–7% [3, 14].

3.3. Pulmonary veno-occlusive disease and pulmonary capillary haemangiomatosis

Approximately 10% of idiopathic PAH meets PVOD/PCH criteria, corresponding to minimum incidence and prevalence below 1 per million; unlike idiopathic PAH, males predominate and prognosis is worse [2]. Familial disease typically affects young siblings in one generation, with unaffected sometimes consanguineous parents, indicating recessive inheritance [2]. PVOD/PCH may also complicate systemic sclerosis or be associated with alkylating agents (cyclophosphamide, mitomycin C) and solvents such as trichloroethylene [2]; the 7th WSPH moved mitomycin C into the definite PAH-association group [3]. In a PAH-ICON-cited study, none of 13 PVOD patients with intermediate/high trichloroethylene exposure had a pathogenic variant, while 5 of 12 with no/trivial exposure had biallelic EIF2AK4 variants [14]. Most report non-specific exertional dyspnoea and fatigue; clubbing and bibasal crackles may occur; haemodynamics are precapillary because lesions involve small venules and capillaries while left atrial pressure remains normal [2]. Suggestive signs include reduced DLCO, often below 50% predicted, hypoxaemia and, on non-contrast CT, thickened subpleural septal lines, centrilobular ground-glass opacities and mediastinal lymphadenopathy; PAH-drug-induced pulmonary oedema also suggests diagnosis [2]. In 24 biopsy-proven PVOD patients compared with 24 without PVOD, PVOD was associated with more males, greater smoking exposure, lower body mass index, PaO₂, DLCO per alveolar volume and minimum walk-test saturation, and more nodular ground-glass opacities, septal lines and lymphadenopathy; 7 of 16 (44%) receiving PAH drugs developed pulmonary oedema [8]. PVOD may also cause occult alveolar haemorrhage [9]. In the Iberian Roma population, a homozygous founder variant produces aggressive disease with most requiring early transplantation (section 9) [54].

3.4. PAH in hereditary haemorrhagic telangiectasia

In families with HHT, PH most often results from high cardiac output due to hepatic AVMs or from left heart disease; pre-capillary PAH indistinguishable from the idiopathic form is rare (about 1–2%) and occurs mainly with ACVRL1 [21]. Therefore, in a patient with HHT and elevated estimated pulmonary systolic pressure, confirmation by RHC and measurement of cardiac output and PAWP are essential before classifying the PH. In patients with ACVRL1-associated PAH, especially young patients, PAH may be the first recognised manifestation, and genetic testing allows the complications of HHT to be anticipated in the patient and relatives [14]. Treatment of epistaxis, gastrointestinal bleeding and AVMs follows the recommendations for HHT; what matters here are the consequences for PAH therapy, namely the bleeding risk of some drugs (section 7) [16].

3.5. Paediatric age

Annual paediatric PH incidence is 64 per million children; 82% of paediatric PAH is transient (persistent PH of newborn or reparable shunts), and idiopathic/heritable PAH and non-transient congenital-heart-disease-associated PAH incidences are 0.7 and 2.2 per million, with prevalences of 4.4 and 15.6 per million [2]. Between 11% and 52% of children with PH have chromosomal, genetic or syndromic abnormalities; HPAH variants occur in 20–30% of sporadic cases with TBX4 and ACVRL1 enrichment, 17% have other PAH-associated disorders such as trisomy 21, and 23% have CNVs not previously associated with PH [2]. At a Chinese tertiary centre, 110 of 170 children (64%) with idiopathic or heritable PAH had pathogenic variants, predominantly BMPR2, ACVRL1 and TBX4; carriers more often had high baseline clinical risk (54% versus 30%) and poorer prognosis, with overall survival 93.4%, 86.7% and 68.6% at 1, 3 and 5 years [55]. These single-centre data are not population estimates.

3.6. Asymptomatic carriers

In DELPHI-2, 55 asymptomatic BMPR2 carriers (26 men; median age 37) were assessed annually for at least 2 years. None had suspected PAH on baseline echocardiography or N-terminal pro-B-type natriuretic peptide (NT-proBNP), but optional RHC accepted by all identified two mild PAH cases (3.6%) and 12 individuals with exercise PH (21.8%); three further cases arose during follow-up (118.8 patient-years), an incidence of 2.3% annually (0.99% in men, 3.5% in women), and all remained low risk on oral therapy at last assessment [56]. ESC emphasises none would have been detected by echocardiography alone [2], while PAH-ICON notes echocardiography and NT-proBNP were useful in already mildly symptomatic cases [14]. Pulmonary systolic pressure above 40 mmHg during mild exercise occurred in 30% of HPAH relatives and 10% of controls, and only relatives with this response and a pathogenic BMPR2 variant later developed overt PAH [14].

4. Clinical Recognition and Warning Signs

Recognising HPAH requires two simultaneous attitudes: suspecting PH in the presence of unexplained exertional dyspnoea and, once PAH is confirmed, asking whether it is heritable. The ESC proposes a three-pronged strategy to bring diagnosis forward: screening asymptomatic high-risk groups (including carriers of BMPR2 variants, with a penetrance of 14–42%, and first-degree relatives of patients with HPAH), early detection of symptomatic patients in at-risk groups and population strategies in dyspnoea clinics or pulmonary embolism follow-up clinics [2]. The HAS PNDS lists the groups in which screening echocardiography should be repeated at regular intervals, even without symptoms: systemic sclerosis, first-degree relatives of a patient with familial PAH, carriers of a variant in a predisposition gene, congenital heart disease and portal hypertension under evaluation for liver transplantation; and it requires the family history to be taken systematically in the initial evaluation of any PAH [19].

Careful observation matters because syndromic diagnoses may go unrecognised: in a major study, TBX4, ACVRL1 and ENG variants were among the six most frequent alterations in adults previously diagnosed with idiopathic PAH [2]. Up to 20% of TBX4-associated PAH lacks obvious skeletal abnormalities, and patellar ossification centres appear only between ages 3 and 6, making knee examination uninformative in early childhood [57].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. HPAH criteria

HPAH diagnosis requires three elements: PAH confirmation by resting RHC (mPAP >20 mmHg with a precapillary component), exclusion of other PH causes, and identification of a heterozygous P/LP HPAH variant (biallelic for recessive genes) and/or confirmed PAH in one or more relatives [1]. A PAH gene variant of uncertain significance (VUS) neither establishes nor excludes diagnosis [1]. Genetic and haemodynamic components complement each other: a variant does not diagnose PAH without RHC, and RHC does not identify the variant defining relatives' risk [31].

Source: ESC/ERS 2022 guidelines, Table 5 and section 7.8.2; definition retained at the 7th WSPH [2, 3].

The French PNDS, which predates the 2022 guidelines, follows the definitions of the 6th WSPH (2018): the same mPAP >20 mmHg, corresponding to the normal resting value (14 ± 3.3 mmHg) plus two standard deviations, but with the precapillary component defined by PVR ≥3 WU rather than >2 WU [19]. The same document stresses that targeted therapies have only been studied, and are only indicated, with mPAP ≥25 mmHg, PAWP ≤15 mmHg and PVR ≥3 WU; patients with mPAP of 21–24 mmHg or PVR between 2 and 3 WU meet the current definition of PAH without trials supporting treatment, a position consistent with that of the 7th WSPH (sections 7.13 and 10.2) [16, 19]. In the 2018 classification used by the PNDS, PAH with a sustained response to CCBs was subgroup 1.5 and PVOD/PCH subgroup 1.6; in 2022 PVOD/PCH became 1.5 [2, 19].

5.2. Diagnostic assessment

Suspected PAH assessment follows international guidelines and should take place at a specialist centre whenever possible [1]. ESC recommends echocardiography as the first non-invasive examination (class I, level B), assigning echocardiographic PH probability from tricuspid regurgitation velocity (threshold >2.8 m/s, class I, level C) and other signs [2]. It also recommends ventilation/perfusion scintigraphy to exclude chronic thromboembolic PH (I, C), CT pulmonary angiography if suspected (I, C), biochemistry, haematology, immunology, human immunodeficiency virus (HIV) serology and thyroid function in all PAH (I, C), abdominal ultrasound for portal hypertension screening (I, C), respiratory function with DLCO (I, C) and chest CT in all PH (IIa, C); open or thoracoscopic lung biopsy is not recommended (III, C) [2]. RHC is recommended to confirm diagnosis and guide treatment (I, B), at experienced centres (I, C) using standardised protocols (I, C) [2].

Indirect signs considered: right ventricle/left ventricle ratio >1, left ventricular eccentricity index >1.1, pulmonary artery diameter >25 mm, right atrial area >18 cm², inferior vena cava >21 mm with inspiratory collapse <50%, end-diastolic pulmonary regurgitation velocity >2.2 m/s and pulmonary acceleration time <105 ms. Source: Table I of the PNDS, adapted from the 2015 ESC/ERS guidelines [19]; the 2022 guidelines keep the three velocity classes and add the TAPSE/pulmonary systolic pressure ratio <0.55 mm/mmHg to the indirect signs [2].

The PNDS organises the aetiological work-up in three steps: first look for left heart disease or respiratory disease (groups 2 and 3, the most frequent causes); then exclude chronic thromboembolic PH by ventilation/perfusion scintigraphy, a normal perfusion scan excluding it; and finally distinguish group 1 from group 5 and define the form of PAH [19]. In addition to those already mentioned, the list of investigations includes HIV and hepatitis B and C serology in all patients, antinuclear antibodies and, if the titre is high, anticentromere, anti-Scl-70, anti-native DNA and anti-RNP antibodies, hepatic ultrasound with Doppler of the portal vein and, when indicated, measurement of the hepatic venous pressure gradient during RHC (a gradient >5 mmHg suggests portal hypertension), and cardiac magnetic resonance imaging when congenital heart disease is suspected and echocardiography is inconclusive; in PAH, lung volumes and flows are usually normal and DLCO may be reduced [19]. On CT, angiography looks for chronic thromboembolic lesions, parenchymal disease, signs of PVOD and vascular malformations such as anomalous pulmonary venous return or patent ductus arteriosus [19].

Vasoreactivity testing is recommended in idiopathic, heritable or drug-associated PAH to detect patients who can be treated with high-dose calcium channel blockers (CCBs) (I, B), using inhaled nitric oxide, inhaled iloprost or intravenous epoprostenol (I, C); a positive response is defined as a fall in mPAP ≥10 mmHg to an absolute value ≤40 mmHg with increased or unchanged cardiac output (I, C) [2]. A positive response is observed in about 12% of patients with idiopathic or drug-associated PAH and in fewer than 5% with HPAH [3]; the French PNDS, which uses inhaled nitric oxide as the test agent, estimates it at about 7% of idiopathic PAH [19]; in the individual-patient-data meta-analysis, 3% of BMPR2 carriers responded compared with 16% of non-carriers [37]. Severity is assessed by World Health Organization (WHO) functional class, six-minute walk distance, cardiopulmonary exercise testing parameters, B-type natriuretic peptide (BNP) or its N-terminal fraction (NT-proBNP), echocardiography, cardiac magnetic resonance imaging and haemodynamics [1].

5.3. Genetic testing: indications

Testing should begin in an affected relative whenever possible; normal testing in an unaffected relative without a known familial variant does not exclude increased risk [15]. In PVOD/PCH, ESC recommends combining clinical, radiological, blood gas, functional and genetic data (I, A) and considers biallelic EIF2AK4 identification confirmatory of heritable PVOD/PCH (I, A), obviating lung biopsy, which is not recommended (III, C) [2].

5.4. Method, coverage and test limitations

PAH-ICON recommends next-generation sequencing of a panel including at least all 12 definitive genes, reviewed with emerging evidence; previously limited testing should be reassessed; rearrangement and CNV detection should use multiplex ligation-dependent probe amplification (MLPA), quantitative PCR, digital PCR, microarrays or read counting, particularly for BMPR2, ACVRL1, ENG and TBX4 [14]. GeneReviews adds that exome/genome deletion/duplication sensitivity may be lower than targeted methods, genome sequencing may detect deep intronic, untranslated-region and CNV alterations missed by exome, and chromosomal microarray detects large BMPR2- or ENG-encompassing deletions [1]. Commercial panel composition varies: in case 10.1, TBX4 had to be added on request to a panel lacking it [57].

Reports should state the gene, versioned reference transcript (e.g. NM_001204.7 for BMPR2), Human Genome Variation Society (HGVS; c. and p.) nomenclature, zygosity, American College of Medical Genetics and Genomics/Association for Molecular Pathology (ACMG/AMP) classification and method limitations; laboratories should submit variants to ClinVar [13, 14]. Classification considers general-population frequency (gnomAD), predicted effect, protein domain, functional studies, cosegregation, de novo occurrence, in silico predictions and inheritance [14]. ACMG/AMP pathogenic and likely pathogenic variants have equivalent clinical decision value [1].

5.5. Practical algorithm

1. Clinical suspicion (dyspnoea, right-heart strain signs, affected relative) → echocardiography and echocardiographic PH probability [2].
2. Intermediate/high probability or at-risk group → PH-centre referral; investigate groups 2 to 5 causes (lung function with DLCO, blood gases, ventilation/perfusion scintigraphy, CT, laboratory testing) [2].
3. RHC with vasoreactivity testing in idiopathic, heritable or drug-associated forms [2].
4. Confirmed group 1 PAH → three-generation history and directed examination (telangiectases, patellae, lipodystrophy, congenital heart disease); non-contrast CT and DLCO for PVOD signs [1, 2].
5. Pre-test genetic counselling → all 12 definitive genes and CNVs; trio exome/genome in children and negative familial or syndromic cases [14, 15].
6. P/LP variant → post-test counselling, cascade testing and carrier surveillance; biallelic EIF2AK4 variant → heritable PVOD/PCH and transplant evaluation [2, 14].
7. VUS or negative result → do not use for predictive testing; periodic reassessment; segregation studies as research [14].

6. Differential Diagnosis

6.1. Other causes of pulmonary hypertension

Cardiopulmonary PH causes are much more frequent than PAH and must be excluded before diagnosis: connective tissue and autoimmune disease, cirrhosis, HIV, anorexigens and groups 2 to 5 diseases [1]. Cardiac disease (group 2) — left ventricular dysfunction, congenital heart disease, valve disease and cardiomyopathies including hypertrophic cardiomyopathy and hereditary transthyretin amyloidosis — is detected by examination, electrocardiography, echocardiography and catheterisation; lung disease and hypoxia (group 3) by examination, lung function and high-resolution CT; pulmonary embolism and large-vessel disease (group 4) by perfusion scintigraphy and angiography, with chronic thromboembolic PH identification important given effective pulmonary endarterectomy [1]. Group 5 includes haematological, systemic and metabolic diseases; ESC mentions glycogen storage disorders, with cases mainly type 1 [2].

6.2. Genetic diseases with pulmonary hypertension or phenotypic overlap

Hereditary cardiomyopathies and valve diseases cause postcapillary PH, and GeneReviews recalls hypertrophic cardiomyopathy and transthyretin amyloidosis in hereditary heart failure with PH [1]. Neuromuscular PH primarily results from hypoventilation, chest wall deformities and chronic hypoxia rather than primary pulmonary vasculopathy [60].

7. Treatments and Risk Management Strategies

HPAH treatment follows idiopathic and drug-associated PAH: guidelines group all three under the same recommendations, and most trials included them together [2, 16]. No authorised therapy specifically targets a genotype; GeneReviews calls FDA-authorised adult PAH therapies “targeted”, with choice depending on administration route and individually acceptable adverse effects [1]. Comparison with the recent past shows the change: a 1998 review summarised in OMIM stated that only a small percentage benefited from CCBs, oral anticoagulation was standard and continuous epoprostenol prolonged life but was complex and costly; this is historical, unlike current practice where routine anticoagulation is not recommended (IIb, C) and initial combination therapy is usual [2, 4]. Genotype nevertheless modifies vasoreactivity probability, early PVOD transplant referral, HHT bleeding risk and management of orthopaedic, cardiac or metabolic comorbidities [2, 15, 16, 37].

7.1. General measures

The HAS PNDS develops the same measures with some differences of detail and one fundamental divergence [19]. It calls for avoidance of tobacco, alcohol and other addictive behaviours; recommends regular physical activity, such as walking, adapted to symptoms, and advises against intense exertion; reserves supervised rehabilitation for stable patients selected by the centre and carried out in specialised centres, because of the risk of exertional syncope; strongly advises against prolonged stays above 1,500 m; and advises avoiding air travel, or travelling with oxygen, in patients in class III or IV or with PaO₂ <60 mmHg, a broader criterion than that of the ESC, which is based on PaO₂ or saturation and prior oxygen use [2, 19]. It prefers elective surgery and locoregional anaesthesia whenever possible, because general anaesthesia is a factor of morbidity and mortality, and requires multidisciplinary management with the PH centre when it cannot be avoided [19]. It recommends influenza and pneumococcal vaccination and the complete vaccination schedule in children; diuretics are combined with a low-sodium diet, with the dose adjusted to weight, oedema and right-sided filling pressures; long-term oxygen is indicated with PaO₂ <60 mmHg at rest, with a mainly symptomatic aim, without expecting saturation to normalise when there is a right-to-left shunt, and ambulatory oxygen may be used if it corrects exertional desaturation with symptomatic benefit [19]. On the social side, it advises recognition of disability by the Maison départementale des personnes handicapées (MDPH), the status of disabled worker (RQTH) and workplace adaptation or therapeutic part-time work [19].

The fundamental divergence is anticoagulation. The PNDS recommends starting effective anticoagulation with a vitamin K antagonist (warfarin first line), with a target international normalised ratio (INR) of 1.5–2.5, in patients without an associated disease, that is, with idiopathic, heritable or anorexigen-associated PAH; it considers the data on direct oral anticoagulants insufficient and advises against anticoagulation in associated forms, especially systemic sclerosis, unless there is another indication or a tunnelled catheter for intravenous prostacyclin [19]. The later 2022 ESC/ERS guidelines note that there are no randomised trials, that registry data are contradictory (the largest suggested benefit in idiopathic PAH, not confirmed by others) and that no meta-analysis has included methodologically robust studies, and therefore do not recommend routine anticoagulation and leave the decision to the individual case (IIb, C) [2]. In HPAH, the choice should weigh the genotype-specific bleeding risk: BMPR2-associated haemoptysis and the telangiectases and AVMs of carriers of ACVRL1, ENG or GDF2 variants are arguments against systematic anticoagulation [19, 21]. The warfarin dose is individual and titrated to the INR; bosentan may reduce warfarin levels and require adjustment [2].

7.2. Vasoreactive patients

Fewer than 10% of patients with idiopathic, heritable or drug-associated PAH respond to vasoreactivity testing, and the proportion is even lower in BMPR2 carriers [2, 3, 37]. In responders, the ESC recommends high-dose CCBs (I, C), full reassessment with RHC at 3–4 months (I, C), continuation if the patient is in functional class I or II with marked haemodynamic improvement (mPAP <30 mmHg and PVR <4 WU) (I, C), and initiation of specific PAH therapy if the patient remains in class III or IV or without marked improvement (I, C); CCBs are not recommended without vasoreactivity testing or in non-responders (III, C) [2]. The French PNDS restricts testing to idiopathic, heritable and anorexigen-associated forms, uses inhaled nitric oxide, requires the CCB to be introduced up to the maximum tolerated dose, adjusted to the clinical and haemodynamic response and possibly higher than that used in systemic hypertension, and defines a sustained response as functional class I or II with maintained haemodynamic improvement, at least equal to that obtained in the acute test, after one year or more of CCB therapy; efficacy is confirmed by RHC, with repeat testing if necessary, and CCBs should not be used in associated forms [19]. In France, this use is outside the marketing authorisation (AMM), under the prescriber's responsibility and with specific information for the patient [19]. The doses used in practice (ESC Table 19) are, for amlodipine and felodipine, a start of 5 mg/day and a target of 15–30 mg/day; for diltiazem, 60 mg twice daily up to 120–360 mg twice daily; for nifedipine, 10 mg three times daily up to 20–60 mg two or three times daily, reached progressively; the most frequent adverse effects are hypotension and peripheral oedema [2]. These CCB doses in PAH constitute off-label use, as the labels refer to systemic arterial hypertension and angina. In these indications and in adults, the US labels set, for amlodipine, 5 mg/day with a maximum of 10 mg/day (2.5–5 mg/day from 6 to 17 years, with no doses above 5 mg/day studied in children) [61]; for extended-release felodipine, 2.5–10 mg/day [62]; for extended-release diltiazem (Cardizem CD®), 120–240 mg/day initially, up to 480 mg/day, with limited experience above 360 mg/day and doses up to 540 mg/day studied in trials [63]; and for extended-release nifedipine, 30–60 mg/day, without titration above 90 mg/day or 120 mg/day, depending on the formulation; the safety and efficacy of felodipine, diltiazem and nifedipine have not been established in children [64, 65]. The doses recommended in PAH are those of the ESC; the targets (amlodipine and felodipine 15–30 mg/day, diltiazem up to 720 mg/day, nifedipine up to 180 mg/day) exceed the maxima of these labels and are therefore outside their conditions, in addition to being an unlabelled indication [2].

7.3. Strategy in non-vasoreactive patients

Initial decisions use multiparametric three-stratum risk assessment; follow-up uses a simplified four-stratum model (functional class, walk distance, BNP or NT-proBNP), with observed 1-year mortalities 0–3%, 2–7%, 9–19% and >20% for low, intermediate-low, intermediate-high and high risk [2]. Without cardiopulmonary comorbidities and at low/intermediate risk, ESC recommends initial PDE5i–ERA combination (I, B; under Grading of Recommendations Assessment, Development and Evaluation [GRADE], low-quality evidence and conditional recommendation), specifically ambrisentan–tadalafil (I, B) or macitentan–tadalafil (I, B), and advises against initial oral macitentan–tadalafil–selexipag triple therapy (III, B); at high risk, initial PDE5i–ERA–intravenous/subcutaneous prostanoid should be considered (IIa, C) [2]. AMBITION reduced clinical failure risk with initial ambrisentan–tadalafil versus pooled monotherapy (HR 0.50; 95% confidence interval [CI] 0.35–0.72; p<0.001) [2, 66].

At follow-up, intermediate-low risk on ERA and PDE5i warrants considering added selexipag (IIa, B) or switching PDE5i to riociguat (IIb, B); intermediate-high/high risk warrants considering intravenous/subcutaneous prostanoid addition and transplant referral (IIa, C) [2]. Among sequential recommendations with morbidity/mortality outcomes, ESC recommends adding macitentan to PDE5i or oral/inhaled prostanoids (I, B), selexipag to ERA and/or PDE5i (I, B), and oral treprostinil to ERA or PDE5i/riociguat monotherapy (I, B), and advises against adding bosentan to sildenafil (III, B) [2]. The 7th WSPH algorithm (2024) retains initial ERA–PDE5i for non-high-risk patients, preferring tadalafil with macitentan or ambrisentan, retains initial parenteral-prostanoid triple therapy for high risk, rejects initial oral triple therapy and introduces sotatercept for intermediate-low/intermediate-high escalation alongside prostacyclin-pathway agents; maximum therapy now comprises four drugs [16]. It suggests preferring selexipag or oral treprostinil over sotatercept with increased bleeding risk such as HHT [16].

The HAS PNDS algorithm (2020) predates sotatercept and the four-strata model. At high risk (class IV, or class III with a low cardiac index or elevated right atrial pressure), it recommends initial combination with intravenous epoprostenol and an ERA and/or PDE5i (double or triple); at intermediate risk, initial dual oral therapy (ERA with PDE5i); at low risk, it leaves the choice between monotherapy and dual oral therapy [19]. Reassessment takes place 3 to 6 months later; if risk remains intermediate or high, sequential combination and referral for pre-transplant assessment follow [19]. In elderly patients with cardiovascular comorbidities or left ventricular diastolic dysfunction, the PNDS prefers initial oral monotherapy, which coincides with the exception provided by the ESC for patients with cardiopulmonary comorbidities [2, 19]. The difference at low risk (monotherapy accepted by the PNDS; dual combination recommended by the ESC, I, B) is recorded in section 7.13 [2, 19]. In HPAH, the PNDS does not provide a specific strategy: the diagnostic and therapeutic approach is that of idiopathic PAH, with the addition of a genetics clinic referral [19].

7.4. Adult drugs, dosing and adverse effects

The doses of ERAs, PDE5i, riociguat and selexipag come from the cited US labels; the doses of prostanoids, which depend on the device and tolerance, are those of ESC Table 19, which gives doses in current clinical use without excluding alternatives, and the titration schedules of the labels (FDA; European SmPC for inhaled iloprost, whose US label was not obtained) are shown for comparison; the 1-year doses of parenteral prostanoids are observed values [2, 76, 77, 80]. In the EU, the sildenafil SmPC sets 20 mg 3×/day with no higher titration; titration up to 80 mg 3×/day is exclusive to the US label [72, 82]. The European bosentan SmPC allows, in the event of late deterioration on 125 mg 2×/day, an increase to 250 mg 2×/day after assessment of hepatic risk [83]. ERAs and riociguat are contraindicated in pregnancy [2].

The French PNDS adds details of monitoring and access. With bosentan, it reports hepatic cytolysis in about 7% of patients, reversible with dose reduction or discontinuation, monthly transaminases, contraindication with transaminases above three times the upper limit of normal and haemoglobin monitoring one month after initiation, because of the risk of anaemia; with ambrisentan, cytolysis is less frequent and there are no relevant interactions, but the contraindication with baseline transaminases above three times normal and haemoglobin monitoring at the first month remain [19]. Sildenafil and tadalafil have no hepatic toxicity, but are contraindicated in severe hepatic impairment (Child–Pugh C), after recent stroke or myocardial infarction and in visual loss due to non-arteritic anterior ischaemic optic neuropathy [19]. Epoprostenol, with a half-life of about 3 minutes, requires continuous infusion by a pump connected to a tunnelled subclavian or jugular catheter, started in hospital, with no defined optimal dose; subcutaneous treprostinil uses a minipump similar to an insulin pump; selexipag, taken twice daily, reduced the first morbidity/mortality event by 40% in the phase 3 placebo-controlled trial (GRIPHON) [19]. Inhaled iloprost requires a nebuliser producing particles of 0.5–3 µm; in 2020 the PNDS recorded that it had ceased to be marketed in France and was not being started in new patients, which shows that the availability of each prostanoid varies between European countries [19]. Prescription of targeted therapies outside the approved indications is only done in France after the opinion of the reference or competence centre [19].

7.5. Comparative regulatory status

Authorisation does not ensure marketing or reimbursement nationally. In Brazil, ANVISA authorised four-pathway drugs (ambrisentan, bosentan, sildenafil, riociguat, iloprost, selexipag and sotatercept), but only ambrisentan, bosentan, iloprost, selexipag and sildenafil are available through the Unified Health System [18]. Consulted sources provided no Portuguese access data.

7.6. Sotatercept: pivotal trials and genetic implications

These trials have interpretative limitations. All added sotatercept to background therapy, excluded PVOD/PCH and included minority HPAH populations, so HPAH efficacy is inferred from subgroups or small analyses; PULSAR genetics included only 25 pathogenic-variant carriers [17, 90]. ZENITH and HYPERION stopped early, tending to overestimate effect magnitude; HYPERION event reduction mainly reflected exercise deterioration, with similar deaths between arms [88, 89]. Hazard ratios across different populations, durations and outcomes cannot rank sotatercept against other therapies indirectly, and 76% relative reduction must be viewed against absolute population events: placebo 54.7% in ZENITH and 36.9% in HYPERION [16, 88, 89]. No randomised paediatric data exist: phase 2 uncontrolled MOONBEAM (NCT05587712) examines pharmacokinetics and safety at ages 1 to <18, with primary completion expected in 2028 [91].

Safety has particular genetic relevance. In STELLAR, epistaxis occurred in 22.1% with sotatercept and 1.9% with placebo, telangiectases in 16.6% and 4.4% (median onset 36.1 weeks), and haemoglobin elevation and thrombocytopenia were managed with dose delays/reductions; serious bleeding occurred in 4% versus 1% (STELLAR), 7% versus 5% (ZENITH) and 4% versus 2% (HYPERION), more often with prostanoids, antithrombotics or low platelets [17]. The September 2026 label adds post-marketing angiodysplasia-associated gastrointestinal bleeding, advising consideration of endoscopy for recurrent or unexplained gastrointestinal bleeding [17]. These effects overlap HHT manifestations, prompting 7th WSPH suggestions of other escalation options for bleeding risk, including ACVRL1 or ENG carriers with HHT [16]. Isolated reports include an immobile, oxygen-dependent TBX4 HPAH patient on the transplant list removed after starting sotatercept, with effect sustained at 1 year [92]; another HPAH patient with a GDF2 VUS had neither positive nor negative effects [93]. Case reports cannot establish genotype-specific efficacy, and a GDF2 VUS cannot be treated as causal; the biomarker study suggesting reduced BMP signalling included nine patients [33].

7.7. Pulmonary veno-occlusive disease and pulmonary capillary haemangiomatosis

There is no established medical therapy for PVOD/PCH. The response to PAH drugs is limited and there is a risk of pulmonary oedema due to venous obstruction; incomplete and transient improvement has been described in individual patients, so therapy should be used with great caution, in experienced centres, with slow titration, diuretics and oxygen, and with full information to the patient [2]. The ESC recommends referral of eligible patients for evaluation at a transplant centre as soon as the diagnosis is established (I, C) and allows drugs approved for PAH with close monitoring of symptoms and gas exchange (IIb, C); lung transplantation is the only curative therapy and examination of the explanted lung confirms the diagnosis [2]. In the form associated with the Iberian EIF2AK4 founder variant, most patients required early transplantation [54]. The HAS PNDS bases suspicion on the presence of at least two of the three CT signs (interlobular septal thickening, centrilobular ground-glass nodules, mediastinal lymphadenopathy), a markedly reduced DLCO (often <50%) and profound hypoxaemia, which may be absent; it recommends that the decision to start monotherapy be taken case by case at the centre, with very close monitoring of tolerance, and that transplant eligibility be assessed at the same time, as it is the only curative therapy [19]. As the heritable form is recessive, the PNDS directs genetic information mainly to the patient's siblings and, where there is consanguinity, to the whole family [19].

7.8. Lung transplantation and other interventions

ESC recommends transplant evaluation for candidates inadequately responding to combined oral treatment, defined by intermediate-high/high risk or REVEAL >7 (I, C), and listing for high risk or REVEAL ≥10 despite optimised therapy including subcutaneous/intravenous prostanoids (I, C) [2]. In children with insufficient response or no drug access, Potts shunting, balloon atrial septostomy or transplantation may be considered; Potts experience is limited to slightly more than 100 patients, with mortality 12–25% [2].

The HAS PNDS considers bilateral lung transplantation the technique of choice and reserves heart–lung transplantation for PAH associated with uncorrectable congenital heart disease; as the eligibility assessment is lengthy, it asks for it to be brought forward and for the patient to be referred to a transplant centre as soon as the response to first-line therapy is insufficient [19].

Source: Table IX of the HAS PNDS [19]. In France, in the presence of severe and persistent right ventricular failure with a short-term threat to life, national priority access to a graft can be requested from the Agence de la biomédecine, assessed by an expert external to the centre and valid for 8 days, renewable once [19].

7.9. Paediatric age

Paediatric PAH treatment is extrapolated from adults and age-adjusted, using risk stratification and observational combination-treatment evidence (I, C), and should take place at paediatric-experienced centres (I, C) [2, 49].

The HAS PNDS points out that, in children, PH is due mainly to congenital heart disease and, secondly, to idiopathic and heritable forms, with the same diagnostic algorithm and the same distribution of genetic causes as in adults; it distinguishes persistent pulmonary hypertension of the newborn, almost always reversible, treated with inhaled nitric oxide or, in severe and prolonged forms, with extracorporeal membrane oxygenation (ECMO) [19]. The therapeutic strategy overlaps with that of adults, with more frequent use of the subcutaneous route for prostanoids, and the Potts anastomosis is proposed as an alternative or bridge to transplantation in forms that escape triple therapy; referral to paediatric PH competence centres or to congenital heart disease reference centres (M3C network) is considered indispensable [19]. When the PNDS was written, only sildenafil and bosentan had paediatric authorisation among oral drugs; current European information also includes tadalafil, ambrisentan, macitentan and riociguat (table above and section 7.5) [19, 71].

7.10. Genotype-specific therapeutic implications

7.11. Developments in the last 12 months and investigational treatments

Other programmes had unfavourable outcomes or remain early: KER-012, another activin-pathway modulator, had TROPOS terminated after pericardial effusions; phase 3 SERANATA of seralutinib for interstitial-lung-disease-associated PH was suspended by the sponsor; and phase 3 TRANSCEND of extended-release tacrolimus for PAH with functional limitation despite optimised therapy awaited recruitment [91]. Tacrolimus had been assessed in a 23-patient phase 2 trial (TransformPAH, NCT01647945) [91]. Gene therapy remains prospective: the 7th WSPH considers cataloguing genes, mechanisms, cell types and natural history essential, alongside gene-specific registries enabling genetically informed trials [15]. None of these programmes has HPAH-specific results in consulted sources.

7.12. Emergencies and acute complications

Right ventricular failure is the leading cause of death in PAH, and acute decompensation carries a very poor short-term prognosis; the HAS PNDS asks for it to be treated in an intensive care unit located at or near the PH centre [19]. The attending physician should contact the centre promptly in the event of syncope, persistent worsening of dyspnoea, worsening signs of right heart failure or renal failure [19].

7.13. Divergences between the HAS PNDS and the other guidelines

The French PNDS was published in 2020 and predates the 2022 ESC/ERS guidelines, the 7th WSPH (2024) and the authorisation of sotatercept; most divergences result from this chronology, and where the recommendations differ the more recent guidelines are the reference [2, 16, 19].

8. Reproductive Options and Family Counselling

8.1. Recurrence risks in autosomal dominant inheritance

HPAH is usually autosomal dominant [1]. Some patients have an affected parent; others have de novo variants of unknown proportion [1]. A proband may appear uniquely affected because relatives' disease is unrecognised, penetrance reduced, a parent died before symptoms or parental onset is late; de novo occurrence can therefore only be asserted after molecular exclusion in both parents with confirmed biological parentage [1]. If absent in both, consider a de novo variant or parental gonadal (or somatic and gonadal) mosaicism; leucocyte DNA misses some somatic mosaicism and all exclusively gonadal mosaicism [1].

8.2. Autosomal recessive inheritance (EIF2AK4, ATP13A3)

Parents of a biallelic patient are presumed heterozygous, and parental testing confirms phase and permits reliable risk assessment; if only one carries a variant, consider de novo occurrence in the proband or parental postzygotic mosaicism, while apparent homozygosity may represent an undetected deletion or uniparental isodisomy [1]. At conception, siblings have 25% affected risk, 50% carrier risk and 25% probability of inheriting neither variant [1]. EIF2AK4 heterozygotes are not at risk of EIF2AK4-related PVOD; heterozygotes for recessive ATP13A3 variants may be at PAH risk, requiring confirmation [1]. Counselling should address consanguineous union risks in EIF2AK4 families [14]. Similar reasoning applies to TBX4: with both partners heterozygous (reported in consanguineous unions), each pregnancy has 25% homozygous posterior amelia with pelvic/pulmonary hypoplasia risk, alongside dominant skeletal syndrome and PAH risks in heterozygotes [1, 10].

8.3. Genetic counselling and predictive testing

ESC recommends pre-test counselling by appropriately trained PH professionals or geneticists addressing penetrance, at-risk relatives, reproduction, genetic discrimination and psychosocial aspects, and considers geneticist/genetic counsellor counselling critical before testing asymptomatic relatives [2]. PAH-ICON adds that identifying a proband P/LP variant creates a medicolegal responsibility to inform patients of family implications and offer interested relatives counselling; disadvantages of knowing predisposition should be discussed, including life, private health, disability and long-term care insurance not covered by all anti-discrimination laws; carriers may experience worry or guilt and non-carriers “survivor guilt” [14]. At a French reference centre between 2003 and 2014, counselling was offered to 529 PAH and 100 PVOD patients; presymptomatic diagnosis offered to 272 HPAH relatives identified the variant in 36.4% [99].

The HAS PNDS describes the French pathway. The diagnosis is disclosed at the reference or competence centre, at the end of the initial evaluation, by the specialist together with the centre's nurse and psychologist, and the psychological consequences for the patient and family are assessed systematically [19]. A genetics clinic appointment is recommended in idiopathic, familial and anorexigen-associated forms and in PVOD; at the first appointment the pedigree is drawn up and the risk of being a carrier, the mode of transmission (dominant with incomplete penetrance in PAH, recessive in PVOD) and the recurrence risk are explained; the patient who accepts testing signs informed consent; the analysis is performed in a specialised genetics laboratory; and any positive result is communicated at a further counselling appointment [19]. Since Decree No. 2013-527 of 20 June 2013, a patient informed that they carry a causal genetic alteration has a legal obligation in France to pass the information on to the family; the geneticist or genetic counsellor defines which relatives should be informed and of what risk, and the patient is invited to offer them a genetics clinic appointment [19]. In dominant HPAH, the information concerns the whole family; in recessive PVOD, mainly the siblings, and the whole family if there is consanguinity [19]. This is an obligation under French law; the international PAH-ICON consensus treats informing relatives as a medico-legal responsibility of the clinician [14].

Presymptomatic testing in minors should be considered carefully to preserve the right to an open future because prevention is unproven and penetrance incomplete; the most immediate childhood benefit is release from surveillance after negative testing, and not all countries permit testing minors for these conditions [14]. GeneReviews recommends asymptomatic-relative testing through a genetic counsellor or medical geneticist with discussion of surveillance, treatment options and treated prognosis [1]. Brazil's 2026 Brazilian Thoracic Association (SBPT) guideline records that testing and counselling are not routine and the technology is unavailable in national clinical practice [18].

8.4. Reproductive options

Genetic risk determination and prenatal/preimplantation discussion are ideally pre-pregnancy, and counselling is appropriate for affected or at-risk young adults [1]. With a known familial variant, prenatal and preimplantation genetic testing are possible; professional and family perspectives differ [1]. The 7th WSPH lists not testing the pregnancy, prenatal testing (rarely performed), preimplantation diagnosis following in vitro fertilisation, donor gametes, adoption and not having children, emphasising age-/sex-specific risk estimates and treatability [15]. French-centre BMPR2 screening enabled preimplantation diagnosis offers to two couples [99]. VUS should not be used for preimplantation diagnosis [14]. Consider DNA banking in probands without molecular diagnosis [1].

8.5. Pregnancy in affected women and carriers

Historically, pregnancy with PAH had maternal mortality up to 56% and neonatal mortality up to 13%; with current treatment maternal mortality decreased but remains 11–25% [2]. ESC recommends clear contraception counselling (I, C), immediate expert-centre counselling including genetics for women considering or becoming pregnant (I, C), termination at PH centres with psychological support (I, C), and permits adoption or surrogacy with preconception genetic counselling (IIb, C); ERAs and riociguat are not recommended because of teratogenic potential (III, B) [2]. If pregnancy continues, stop ERAs, riociguat and selexipag; CCBs, PDE5i and inhaled/intravenous/subcutaneous prostanoids are considered safe despite limited evidence; bosentan reduces hormonal contraceptive efficacy, and implants or intrauterine devices offer low failure [2]. Sotatercept may harm the fetus and requires effective contraception until 4 months after the last dose; breastfeeding during treatment is not recommended [17]. With controlled low-risk disease and normal or near-normal haemodynamics, decisions should be individual and shared, and preconception genetic counselling considered in HPAH [2]. Pregnancy as an asymptomatic-carrier trigger remains debated from isolated observations [14, 51].

The HAS PNDS maintains the position prior to the 2022 guidelines: pregnancy is formally contraindicated, save in exceptional cases, because of the risk of life-threatening worsening for mother and child; if it proceeds, management is multidisciplinary, with the PH centre, a level 3 maternity unit and a thoracic surgery team available for circulatory support in the event of peripartum right ventricular failure [19]. Its contraception recommendations are more detailed than those of the ESC [19]:

Source: section 4.3.1 of the PNDS [19]. The ESC adds that emergency hormonal contraception is safe in PH and that surgical sterilisation may be considered, with perioperative risk [2].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Frequency and what each figure measures

Estimates vary about tenfold between registries, mainly reflecting differences in case ascertainment and access to genetic testing; the prevalence of diagnosed disease is much lower than the carrier frequency, given incomplete penetrance [31]. Frequencies for a group should not be transferred to an individual gene: the proportion attributed to BMPR2 (about 75% of identified variants) is not a population prevalence [1]. Hospital series, such as the Chinese paediatric cohort with 64% pathogenic variants, measure diagnostic yield at one centre and not prevalence [55]. The French registry figures relate to the whole of group 1 and to diagnosed patients; the mean age of about 55 years reflects the weight of associated forms and elderly patients, and does not apply to HPAH, diagnosed on average at 35 years [1, 19].

9.2. Sex and age

HPAH affects twice as many women as men, while recent US and European general PAH data show increasing diagnoses at ≥65 with cardiovascular comorbidities and more balanced sexes [2]. Before puberty, sex ratio approaches 1:1 [15]. Female BMPR2 penetrance is at least twice male penetrance, and annual DELPHI-2 incidence was 3.5% in women versus 0.99% in men [14, 56].

9.3. Ancestry, founder populations and cohorts

Major genomic cohorts — US PAH Biobank, UK NIHR BioResource, the British idiopathic/heritable cohort and PVDomics — are predominantly European, and 7th WSPH considers ancestry diversity essential to characterise genomic variation and treatment response [15]. PAH Biobank and the 4,241-case international consortium association analyses were restricted to European ancestry to control population structure [34, 36]. A 17-study meta-analysis found no Asian–White differences in BMPR2-associated mPAP and PVR, but greater cardiac index/output reduction in White patients; these non-randomised studies had methodological heterogeneity [38].

The clearest founder effect is the Iberian Roma population. In five Spanish Roma families with severe PAH and often multiple affected members, without variants in then-known genes, homozygosity mapping identified homozygous EIF2AK4 NM_001013703.4:c.3344C>T p.(Pro1115Leu) in all families; most required early transplantation, and authors emphasise its counselling and recurrence-risk importance in this population [40, 54]. In HHT, an ACVRL1/ENG allelic condition, potential founder variants include ACVRL1 in French departments Ain, Jura and Deux-Sèvres; these frequencies concern HHT rather than PAH [21]. In genetically homogeneous Finland, primary PH prevalence was 5.8 per million and annual incidence 0.2–1.3 per million, without a common BMPR2 founder among 26 sporadic and 4 familial cases; the authors suggested pathogenic BMPR2 variants are relatively recent [4]. French-centre EIF2AK4 PVOD families were largely North African [9].

Reference American and European HPAH data derive from Vanderbilt's family registry (355 BMPR2 carriers in 53 families) [48], PAH Biobank [34], the French reference centre [56, 99] and the Spanish registry, offering testing since 2011 to all idiopathic, heritable or associated PAH and PVOD [53]. A panel identified pathogenic variants in 39 of 47 HPAH families [50].

9.4. Brazil and Portugal

Brazil's GRADE-developed 2026 SBPT guideline records that genetic testing and counselling are not routine and BMPR2 detection is unavailable clinically, describing access inequalities: among ANVISA-approved drugs, only ambrisentan, bosentan, iloprost, selexipag and sildenafil are available through the Unified Health System [18]. It positions sotatercept as an alternative to sequential prostanoid triple therapy in non-low-risk patients [18]. Consulted sources provided no Portuguese epidemiological, genetic or access data; this does not demonstrate absence of published studies.

9.5. Special populations

Children have greater genetic burden, with de novo variants in at least 15%, and benefit from trio exome/genome [15, 49]. Women of childbearing age require contraceptive and preconception counselling (section 8.5) [2]. Anorexigen-exposed patients should be tested because approximately one-fifth tested with fenfluramine-associated PAH carried BMPR2 variants [1]. HHT and PAH require HHT-centre coordination and bleeding-risk attention [16, 21].

9.6. Reference networks and patient organisations

The EU's ERN-LUNG, launched in 2017, includes a PH core network of more than 20 full members contributing approximately 1,500 new PAH/chronic thromboembolic PH patients annually and requiring and monitoring centre quality standards [2, 20]. ESC recommends centres inform patients about and collaborate with patient organisations [2]. Resources include PHA Europe [100], PHAware and the US Pulmonary Hypertension Association, maintaining a US centre list [1, 101]. PAH-ICON joins large and small centres to study rare genotypes, oligogenic inheritance and treatment response [14].

In France, the 2020 PNDS describes a network organisation: a PH reference centre (PulmoTension), with its coordinating site at Bicêtre Hospital (AP-HP, Le Kremlin-Bicêtre), collaborating sites at Cochin Hospital and the Georges-Pompidou European Hospital, a constituent site at Marie Lannelongue Hospital (Le Plessis-Robinson; thoracic surgery and heart–lung transplantation) and regional competence centres, including paediatric centres (for example, Necker–Enfants malades Hospital) and overseas centres (Réunion and Martinique), a total of 27 centres [19]. The coordinating centre includes a genetic counsellor [19]. The initial evaluation, disclosure of the diagnosis, and initiation and renewal of targeted therapies take place in these centres, which reassess efficacy at least once a year [19]. The centre is part of the RespiFiL rare respiratory diseases health filière, created in 2014 and funded by the Ministry of Health, and of the Réseau HTAP; the patient association HTaPFrance, founded in 1996, represents patients with PH of all causes and takes part in therapeutic education programmes; France Assos Santé and the Maladies Rares Info Services portal (health and life pathway with a rare disease) complete the resources [19]. PAH is recognised as an “off-list” long-term condition (affection de longue durée), with exemption from patient co-payment under a care protocol drawn up by the attending physician with the medical adviser [19].

10. Illustrative Clinical Case

10.1. TBX4-associated PAH in an adult with absent patellae

Shrivastava and colleagues at Mayo Clinic described in 2021 a White man with Bavarian paternal ancestry presenting with cough and exertional dyspnoea and diagnosed with group 1 PAH aged 54 [57]. His pedigree (original article Figure 1: https://pmc.ncbi.nlm.nih.gov/articles/PMC8446047/figure/fig1/) showed a father with congenitally absent patellae and PAH who died aged 64; among nine siblings, a sister had PAH and died aged 3, a brother aged 58 had absent patellae and dyspnoea, another aged 55 had absent patellae without cardiovascular complaints, and in the next generation his daughter aged 36, son aged 32 and a nephew had absent patellae and were asymptomatic [57]. There was no lung disease, connective tissue disease, ischaemic heart disease, HIV or smoking. Examination showed a central venous catheter, accentuated pulmonary second sound, soft left sternal systolic murmur and bilaterally absent patellae [57].

Assessment found haemoglobin 17 g/dL, negative antinuclear antibodies, CT without underlying lung disease, mild obstruction with normal volumes and slightly reduced diffusion, and echocardiography with mild right ventricular enlargement/thickening and estimated systolic pressure 43 mmHg [57]. RHC on epoprostenol and sitaxentan showed pulmonary arterial pressure 48/24 mmHg (mean 37), mean right atrial pressure 2 mmHg, PVR 4.58 WU, pulmonary capillary pressure 7 mmHg and cardiac index 3.51 L/min/m²; knee radiography confirmed absent patellae (Figure 4: https://pmc.ncbi.nlm.nih.gov/articles/PMC8446047/figure/fig4/) [57]. BMPR2 sequencing at 58 was negative; at 71, a PAH panel with TBX4 added identified heterozygous splice donor variant NM_001321120.2:c.1021+1G>A (reported c.1021+1G>A, IVS7+1G>A), classified pathogenic; his asymptomatic daughter and two grandchildren aged 5 and 3 were carriers [40, 57]. Epoprostenol and sitaxentan were stopped for abdominal discomfort and persistent dyspnoea; sildenafil–iloprost–ambrisentan was well tolerated, iloprost later stopped following improvement to class II and ambrisentan replaced by macitentan for pruritus; aged 72 on macitentan and sildenafil, he remained class II with estimated systolic pressure 42 mmHg and no hospitalisation [57].

The case illustrates several points of this article. TBX4-associated PAH, more typical of childhood, may manifest in adulthood, and skeletal abnormalities, when present, are recognised only if the examination is directed [28, 57]. In the family, small patella syndrome and PAH segregated independently — carriers with absent patellae without PAH and, in the sister who died aged 3, PAH at an age when the patellae would not yet have ossified — demonstrating incomplete penetrance and variable expressivity and justifying family screening based on genetic testing rather than on the skeletal phenotype [57]. The case also shows the importance of checking panel composition and repeating testing in patients previously studied only for BMPR2, as the PAH-ICON consensus and the ESC recommend [2, 14]. Sitaxentan (Thelin®), used at the start of treatment, had its European authorisation withdrawn in January 2011, and the management in the report reflects approximately two decades of evolution, not a recommended regimen [71]. Surveillance of the carrier grandchildren and the daughter should follow the principles of section 11, with counselling on testing of minors as described in section 8.3 [14].

10.2. Asymptomatic KCNK3 carriers with haemodynamics below PAH criteria

Rodighiero and colleagues described in 2026 a family whose female infant proband presented at 5 months with symptomatic PAH and an ostium secundum atrial septal defect, subsequently closed, and received sildenafil; the panel found KCNK3 NM_002246.3:c.289G>A p.(Gly97Arg), initially VUS and later laboratory-reclassified as likely pathogenic from loss-of-function studies [40, 102]. A paternal aunt died of PAH aged 2 without testing. Segregation identified the variant in her father and uncle, both asymptomatic (class I) [102].

Her father, aged 32, had an accentuated pulmonary second sound and mild hypokalaemia; echocardiography could not estimate pulmonary systolic pressure without tricuspid regurgitation; RHC showed mPAP 25 mmHg, PVR 1.66 WU and PAWP 15 mmHg, without inhaled nitric oxide response. Her uncle, aged 28, had estimated systolic pressure 28 mmHg but shortened pulmonary acceleration with midsystolic notching; RHC showed mPAP 25 mmHg and PVR 0.92 WU [102]. The authors acknowledged PAH treatment haemodynamic criteria (mPAP >20 mmHg with PVR >2 WU) were unmet and, based on severe family history, began sildenafil 20 mg three times daily in both brothers (alongside diltiazem and amlodipine prescribed for systemic hypertension); both remained asymptomatic 2 years later, one after stopping sildenafil for financial reasons [102].

The case must be read alongside guidelines. ESC defines mPAP >20 mmHg with PVR ≤2 WU and PAWP ≤15 mmHg as “unclassified PH”, warranting clinical follow-up rather than PAH [2]. The 7th WSPH recalls no approved therapies for mPAP 21–24 mmHg and, if treating without robust data, prefers initial monotherapy and close surveillance [16]. ClinVar has conflicting variant classifications, reinforcing consolidated classification for predictive decisions [40]. The authors' management is an individual option they acknowledge as absent from guidelines, not routine recommendation [102].

11. Prognosis and Long-Term Follow-up

11.1. Prognosis

Untreated mean HPAH survival after diagnosis is 2.8 years [1]. Individual-patient meta-analysis associated BMPR2 with increased death/transplantation risk (age-/sex-adjusted HR 1.42; 95% CI 1.15–1.75) and death (HR 1.27; 95% CI 1.00–1.60), greater with younger diagnosis [37]. Chinese paediatric idiopathic/heritable carriers had poorer treatment response and survival than non-carriers [55], while in a paediatric cohort TBX4 carriers survived better than BMPR2 carriers [49]. PVOD/PCH prognosis is poorer than idiopathic PAH [2]. All DELPHI-2 screening-detected PAH remained low risk on oral treatment at last assessment, suggesting without proving early-detection benefit [56].

Risk stratification guides prognosis and decisions. The simplified four-stratum model assigns WHO class points (I–II, 1; III, 3; IV, 4), six-minute walk points (>440 m, 1; 320–440 m, 2; 165–319 m, 3; <165 m, 4) and BNP (<50, 50–199, 200–800 and >800 ng/L) or NT-proBNP (<300, 300–649, 650–1 100 and >1,100 ng/L), classifying by rounded mean [2]. Achieving and maintaining low risk is the adult goal and should be considered adequate paediatric response (IIa, C) [2]. However, combined AMBITION, GRIPHON and SERAPHIN analysis found clinical worsening in more than 20% of low-risk patients within 3 years, prompting 7th WSPH to consider additional, unvalidated goals such as PVR <5 WU, >50–60% reduction or normal right ventricular size/function [16].

Source: Table III of the HAS PNDS, which reproduces the 2015 ESC/ERS model [19]. The 2022 guidelines kept three strata for the initial assessment and revised the BNP and NT-proBNP thresholds, which in the four-strata follow-up model are those shown above [2]. The PNDS identifies functional class, walk test, cardiac index, right atrial pressure and BNP/NT-proBNP as the most discriminating parameters, and sets as treatment goals functional class I–II, walk test ≥440 m, cardiac index ≥2.5 L/min/m² and right atrial pressure <8 mmHg; a patient in class I–II, with a walk test ≥440 m and BNP <50 ng/L or NT-proBNP <300 ng/L, can be classified as low risk without invasive assessment [19]. If risk remains intermediate or high, therapy is intensified; on maximal therapy, listing for transplantation is considered [19].

11.2. Follow-up of patients with PAH

Adapted from ESC/ERS 2022 Table 17 (visual reading; “indicated”, “should be considered” and “may be considered” correspond to green, yellow and orange), with intervals adjusted to needs, aetiology, risk, demographics and comorbidities; no evidence shows routine follow-up RHC improves outcomes compared with clinically indicated RHC [2]. Full reassessment at 3–4 months after starting/changing treatment is the escalation decision point [16]. Bosentan requires monthly transaminases; sotatercept requires haemoglobin and platelets before each first five doses and periodically thereafter [2, 17].

The HAS PNDS sets its own schedule, with more frequent invasive assessment [19]:

The intervals are adapted to the case; cardiac magnetic resonance imaging may add prognostic information at diagnosis and during follow-up [19]. The first assessment after a targeted therapy, at 3–6 months, includes at least functional class, walk test if possible, BNP or NT-proBNP and RHC, and each visit records events, quality of life, nutritional and weight status, adherence to treatment and diet, and school, professional and social life [19]. The more recent ESC considers follow-up RHC optional in stable patients and found no evidence that its systematic repetition improves outcomes (section 7.13) [2]. Drug-specific laboratory monitoring (monthly transaminases and haemoglobin at the first month with bosentan; haemoglobin at the first month with ambrisentan) appears in section 7.4 [19].

11.3. Surveillance of carriers and at-risk relatives

Sources: [1, 2, 14, 19, 56].

These recommendations are not identical, and the difference is well founded. GeneReviews bases annual surveillance on consensus documents and not on trials [1]. The ESC recommends annual screening (I, B) but stresses that, in DELPHI-2, none of the cases would have been detected by echocardiography alone and that the best strategy and interval remain to be defined [2]. The PAH-ICON consensus recommends a multimodal assessment and accepts shorter intervals in carriers with a hypertensive response to exercise, who appeared to be at higher risk in DELPHI-2 [14]. In practice, carriers and relatives should know the early symptoms and, in the presence of dyspnoea, undergo a complete assessment immediately [14]. Heterozygotes for an EIF2AK4 variant are not at risk of EIF2AK4-related PVOD and do not need this surveillance, unlike siblings with the same biallelic genotype as the proband [1]. In carriers of ACVRL1 or ENG variants, HHT surveillance follows the recommendations specific to that disease; the GeneReviews chapter on HHT includes transthoracic echocardiography with agitated saline contrast, with estimation of pulmonary systolic pressure, in the initial evaluation of adults [21]. Carriers of TBX4 variants benefit from orthopaedic assessment, and those of SOX17 from assessment for congenital heart disease [15].

11.4. Transition and long-term care

To facilitate paediatric–adult transition, ESC recommends the same PH definition in children (mPAP >20 mmHg) [2]. Long-term care includes contraception, family planning and young-adult genetic counselling, periodic VUS reassessment (approximately every 3 years) and repeat genetic testing after historically limited panels [1, 2, 14]. Specialist palliative care should be offered when appropriate and multidisciplinary teams should periodically monitor adherence [2].

The HAS PNDS requires close collaboration between paediatric centres and adult PAH centres to organise transition [19]. It assigns to the attending physician the monitoring of clinical and laboratory tolerance of treatments, adjustment of anticoagulation and symptomatic therapies, detection of signs of worsening and regular assessment of the psychological, social and family impact, and to the community nurse a central role in patients on continuous infusion of epoprostenol or treprostinil (preparation of infusions and catheter care) [19]. Therapeutic education follows the four steps defined by the HAS — shared educational diagnosis, personalised skills programme, sessions and assessment of acquired skills — and covers the disease, medicines, diet, physical activity and, where relevant, genetics, with a team that may include the genetic counsellor and in liaison with the patient association or expert patients [19]. This education gains importance because primary care professionals know the disease poorly and because the medicines are dispensed through two channels, hospital and community pharmacy; the patient and carers become the bearers of safety information to these professionals [19].

12. Summary and Final Considerations

HPAH is a precapillary pulmonary vasculopathy defined by mPAP >20 mmHg, PAWP ≤15 mmHg and PVR >2 WU, encompassing familial and apparently sporadic PAH with an associated P/LP variant [1, 2]. Since BMPR2 identification in 2000, ClinGen established 12 definitive genes — BMPR2, ACVRL1, ATP13A3, CAV1, EIF2AK4, ENG, GDF2, KCNK3, KDR, SMAD9, SOX17, TBX4 — and 3 moderate genes, advising against disputed diagnostic-panel genes [12, 22, 23]. BMPR2 explains approximately three-quarters of identified variants, with sex-dependent incomplete penetrance (approximately 14% male, 42% female), earlier severe disease, rare vasoreactivity and increased death/transplantation risk [1, 37].

Genotype adds information unavailable from phenotype alone: ACVRL1, ENG and GDF2 link PAH to HHT and bleeding risk; TBX4 links paediatric PAH to small patella syndrome and lung developmental abnormalities; SOX17 to congenital heart disease; KDR to low DLCO/parenchymal disease; CAV1 to lipodystrophy; and biallelic EIF2AK4 reclassifies PVOD/PCH, requiring vasodilator caution and early transplant referral [2, 14, 15]. Testing should be offered after counselling for idiopathic, familial, anorexigen-associated, congenital-heart-disease-associated or PVOD/PCH, using definitive-gene panels with CNVs and paediatric trio exome/genome [14, 15].

Treatment follows risk-stratified idiopathic PAH algorithms, usually initial ERA–PDE5i, parenteral prostanoids at high risk and, since 2024, fourth-pathway sotatercept, reducing clinical events on background therapy but with few HPAH patients and no PVOD [2, 16, 17]. The past twelve months brought US/EU sotatercept indication extensions, HYPERION label updates with angiodysplasia-associated gastrointestinal bleeding and severe-hypersensitivity contraindication, US paediatric selexipag approval and preliminary ralinepag (positive) and seralutinib (non-significant primary outcome) findings [75, 81, 85, 97, 98].

For the family, identification of a P/LP variant enables cascade testing, carrier surveillance — annual according to the ESC and GeneReviews, multimodal according to the PAH-ICON consensus — and informed reproductive decisions, including preimplantation genetic testing [1, 2, 14]. DELPHI-2 showed a PAH incidence of 2.3% per year in BMPR2 carriers and detected cases at a low-risk stage, but the optimal screening strategy and the threshold for treating carriers with mild haemodynamic abnormalities remain to be defined [16, 56]. Genomic cohorts are mostly European and North American; in Brazil, genetic testing is not routine, and no data from Portugal were found in the sources consulted [15, 18]. In France, the 2020 PNDS organises the pathway in reference and competence centres, makes informing relatives legally mandatory after a molecular diagnosis and diverges from later guidelines mainly on anticoagulation, the PVR threshold, initial therapy at low risk and the frequency of follow-up RHC [2, 19].

13. Bibliographic References

1. Geiger H, Tinker RJ, Talbert J, Phillips JA 3rd, Austin ED. Heritable Pulmonary Arterial Hypertension Overview. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle: University of Washington; 1993–2026. Published 18 Jul 2002; comprehensively updated 9 Apr 2026. NBK1485. PMID:20301658.
2. Humbert M, Kovacs G, Hoeper MM, Badagliacca R, Berger RMF, Brida M, et al. 2022 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension. Eur Heart J. 2022;43(38):3618–731. doi:10.1093/eurheartj/ehac237. PMID:36017548. Simultaneous publication: Eur Respir J. 2023;61(1), doi:10.1183/13993003.00879-2022.
3. Kovacs G, Bartolome S, Denton CP, Gatzoulis MA, Gu S, Khanna D, et al. Definition, classification and diagnosis of pulmonary hypertension. Eur Respir J. 2024;64(4):2401324. doi:10.1183/13993003.01324-2024. PMID:39209475.
4. Online Mendelian Inheritance in Man (OMIM). #178600 Pulmonary hypertension, primary, 1; PPH1. Johns Hopkins University; created 2 Jun 1986, last updated 25 Jun 2025. OMIM 178600. Accessed 7 Oct 2026.
5. Online Mendelian Inheritance in Man (OMIM). #615342 Pulmonary hypertension, primary, 2; PPH2. Johns Hopkins University. OMIM 615342. Accessed 7 Oct 2026.
6. Online Mendelian Inheritance in Man (OMIM). #615343 Pulmonary hypertension, primary, 3; PPH3. Johns Hopkins University. OMIM 615343. Accessed 7 Oct 2026.
7. Online Mendelian Inheritance in Man (OMIM). #615344 Pulmonary hypertension, primary, 4; PPH4. Johns Hopkins University. OMIM 615344. Accessed 7 Oct 2026.
8. Online Mendelian Inheritance in Man (OMIM). #265450 Pulmonary venoocclusive disease 1, autosomal dominant; PVOD1. Johns Hopkins University. OMIM 265450. Accessed 7 Oct 2026.
9. Online Mendelian Inheritance in Man (OMIM). #234810 Pulmonary venoocclusive disease 2, autosomal recessive; PVOD2. Johns Hopkins University. OMIM 234810. Accessed 7 Oct 2026.
10. Online Mendelian Inheritance in Man (OMIM). #147891 Ischiocoxopodopatellar syndrome with or without pulmonary arterial hypertension; ICPPS. Johns Hopkins University. OMIM 147891. Accessed 7 Oct 2026.
11. Orphanet. Orphadata: epidemiological data (en_product9_prev.xml) — ORPHA:275777, ORPHA:422, ORPHA:182090, ORPHA:31837, ORPHA:431353 and ORPHA:199241 [Internet]. Prevalence file. Accessed 7 Oct 2026.
12. Welch CL, Aldred MA, Balachandar S, Dooijes D, Eichstaedt CA, Gräf S, et al. Defining the clinical validity of genes reported to cause pulmonary arterial hypertension. Genet Med. 2023;25(11):100925. doi:10.1016/j.gim.2023.100925. PMID:37422716.
13. National Center for Biotechnology Information. NCBI Gene and NCBI Datasets: product reports (MANE Select transcripts) and cytogenetic locations of BMPR2, ACVRL1, ENG, SMAD9, CAV1, KCNK3, TBX4, SOX17, GDF2, EIF2AK4, ATP13A3, KDR, ABCC8, GGCX and TET2 [Internet]. NCBI Datasets. Accessed 7 Oct 2026.
14. Eichstaedt CA, Belge C, Chung WK, Gräf S, Grünig E, Montani D, et al. Genetic counselling and testing in pulmonary arterial hypertension: a consensus statement on behalf of the International Consortium for Genetic Studies in PAH. Eur Respir J. 2023;61(2):2201471. doi:10.1183/13993003.01471-2022. PMID:36302552.
15. Austin ED, Aldred MA, Alotaibi M, Gräf S, Nichols WC, Trembath RC, et al. Genetics and precision genomics approaches to pulmonary hypertension. Eur Respir J. 2024;64(4):2401370. doi:10.1183/13993003.01370-2024. PMID:39209481.
16. Chin KM, Gaine SP, Gerges C, Jing ZC, Mathai SC, Tamura Y, et al. Treatment algorithm for pulmonary arterial hypertension. Eur Respir J. 2024;64(4):2401325. doi:10.1183/13993003.01325-2024. PMID:39209476.
17. Merck Sharp & Dohme LLC. WINREVAIR (sotatercept-csrk) for injection, for subcutaneous use. Prescribing Information. BLA 761363, revision of September 2026 (S-012, S-016, S-017). FDA label; earlier structured version (revision of October 2025) in DailyMed. Accessed 7 Oct 2026.
18. Amado VM, Pereira MC, Correa RA, Campos FAFT, Fernandes C, Ferreira EVM, et al. Brazilian guidelines for the pharmacological treatment of pulmonary hypertension. Official document of the Brazilian Thoracic Society based on the Grading of Recommendations Assessment, Development, and Evaluation methodology. Respir Res Clin Pract. 2026;52(3):e20250528. doi:10.36416/1806-3756/e20250528. PMID:42740758.
19. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Hypertension artérielle pulmonaire. Centre de référence de l’hypertension pulmonaire PulmoTension; filière RespiFiL. Saint-Denis La Plaine: HAS; 7 Apr 2020 (text revised in December 2020). HAS p_3167170.
20. ERN-LUNG. European Reference Network on rare respiratory diseases [Internet]. ern-lung.eu. Accessed 7 Oct 2026.
21. McDonald J, Gossage JR, Stevenson DA. Hereditary Hemorrhagic Telangiectasia. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle: University of Washington; 1993–2026. Published 26 Jun 2000; updated 19 Feb 2026. NBK1351. PMID:20301525.
22. Deng Z, Morse JH, Slager SL, Cuervo N, Moore KJ, Venetos G, et al. Familial primary pulmonary hypertension (gene PPH1) is caused by mutations in the bone morphogenetic protein receptor-II gene. Am J Hum Genet. 2000;67(3):737–44. doi:10.1086/303059. PMID:10903931.
23. International PPH Consortium; Lane KB, Machado RD, Pauciulo MW, Thomson JR, Phillips JA 3rd, et al. Heterozygous germline mutations in BMPR2, encoding a TGF-beta receptor, cause familial primary pulmonary hypertension. Nat Genet. 2000;26(1):81–4. doi:10.1038/79226. PMID:10973254.
24. Trembath RC, Thomson JR, Machado RD, Morgan NV, Atkinson C, Winship I, et al. Clinical and molecular genetic features of pulmonary hypertension in patients with hereditary hemorrhagic telangiectasia. N Engl J Med. 2001;345(5):325–34. doi:10.1056/NEJM200108023450503. PMID:11484689.
25. Shintani M, Yagi H, Nakayama T, Saji T, Matsuoka R. A new nonsense mutation of SMAD8 associated with pulmonary arterial hypertension. J Med Genet. 2009;46(5):331–7. doi:10.1136/jmg.2008.062703. PMID:19211612.
26. Austin ED, Ma L, LeDuc C, Berman Rosenzweig E, Borczuk A, Phillips JA 3rd, et al. Whole exome sequencing to identify a novel gene (caveolin-1) associated with human pulmonary arterial hypertension. Circ Cardiovasc Genet. 2012;5(3):336–43. doi:10.1161/CIRCGENETICS.111.961888. PMID:22474227.
27. Ma L, Roman-Campos D, Austin ED, Eyries M, Sampson KS, Soubrier F, et al. A novel channelopathy in pulmonary arterial hypertension. N Engl J Med. 2013;369(4):351–61. doi:10.1056/NEJMoa1211097. PMID:23883380.
28. Kerstjens-Frederikse WS, Bongers EM, Roofthooft MT, Leter EM, Douwes JM, Van Dijk A, et al. TBX4 mutations (small patella syndrome) are associated with childhood-onset pulmonary arterial hypertension. J Med Genet. 2013;50(8):500–6. doi:10.1136/jmedgenet-2012-101152. PMID:23592887.
29. Eyries M, Montani D, Girerd B, Perret C, Leroy A, Lonjou C, et al. EIF2AK4 mutations cause pulmonary veno-occlusive disease, a recessive form of pulmonary hypertension. Nat Genet. 2014;46(1):65–9. doi:10.1038/ng.2844. PMID:24292273.
30. Gräf S, Haimel M, Bleda M, Hadinnapola C, Southgate L, Li W, et al. Identification of rare sequence variation underlying heritable pulmonary arterial hypertension. Nat Commun. 2018;9(1):1416. doi:10.1038/s41467-018-03672-4. PMID:29650961.
31. GeneGuidance (user database). Gene records for BMPR2, ACVRL1, ENG, SMAD9, CAV1, KCNK3, TBX4, SOX17, GDF2, EIF2AK4, ATP13A3, KDR, ABCC8, GGCX, TET2, SMAD4, SMAD1 and BMP10: identifiers, function, clinical curation and linked conditions. Export of 26 Sep 2026, extracted using ggdump.py on 7 Oct 2026.
32. Hodgson J, Ruiz-Llorente L, McDonald J, Quarrell O, Ugonna K, Bentham J, et al. Homozygous GDF2 nonsense mutations result in a loss of circulating BMP9 and BMP10 and are associated with either PAH or an "HHT-like" syndrome in children. Mol Genet Genomic Med. 2021;9(12):e1685. doi:10.1002/mgg3.1685. PMID:33834622.
33. Jones RJ, De Bie EMDD, Deliu N, Ng AYKC, Dunmore BJ, Gräf S, et al. Sotatercept reduces bone morphogenetic protein signaling in patients with pulmonary arterial hypertension. Sci Transl Med. 2026;18(850):eads5175. doi:10.1126/scitranslmed.ads5175. PMID:42160453.
34. Zhu N, Pauciulo MW, Welch CL, Lutz KA, Coleman AW, Gonzaga-Jauregui C, et al. Novel risk genes and mechanisms implicated by exome sequencing of 2572 individuals with pulmonary arterial hypertension. Genome Med. 2019;11(1):69. doi:10.1186/s13073-019-0685-z. PMID:31727138.
35. Montani D, Girerd B, Jaïs X, Levy M, Amar D, Savale L, et al. Clinical phenotypes and outcomes of heritable and sporadic pulmonary veno-occlusive disease: a population-based study. Lancet Respir Med. 2017;5(2):125–34. doi:10.1016/S2213-2600(16)30438-6. PMID:28087362.
36. Zhu N, Swietlik EM, Welch CL, Pauciulo MW, Hagen JJ, Zhou X, et al. Rare variant analysis of 4241 pulmonary arterial hypertension cases from an international consortium implicates FBLN2, PDGFD, and rare de novo variants in PAH. Genome Med. 2021;13(1):80. doi:10.1186/s13073-021-00891-1. PMID:33971972.
37. Evans JD, Girerd B, Montani D, Wang XJ, Galiè N, Austin ED, et al. BMPR2 mutations and survival in pulmonary arterial hypertension: an individual participant data meta-analysis. Lancet Respir Med. 2016;4(2):129–37. doi:10.1016/S2213-2600(15)00544-5. PMID:26795434.
38. Wu J, Huang Q, Zhang Y, De Z, Fu H, Zhan Y, et al. Impact of BMPR2 mutation on the severity of pulmonary arterial hypertension: a systematic review and meta-analysis. Respir Res. 2025;26(1):74. doi:10.1186/s12931-025-03145-5. PMID:40022182.
39. Firth HV, Hurst JA. Hereditary haemorrhagic telangiectasia (HHT). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 446–447.
40. VariantValidator (GRCh38) and National Center for Biotechnology Information. ClinVar: records 6515 (SMAD9 c.606C>A), 60479 (KCNK3 c.608G>A), 60480 (KCNK3 c.289G>A), 1327321 (TBX4 c.1021+1G>A) and 426060 (EIF2AK4 c.3344C>T) [Internet]. VariantValidator; ClinVar. Accessed 7 Oct 2026.
41. Sörmann J, Schewe M, Proks P, Jouen-Tachoire T, Rao S, Riel EB, et al. Gain-of-function mutations in KCNK3 cause a developmental disorder with sleep apnea. Nat Genet. 2022;54(10):1534–43. doi:10.1038/s41588-022-01185-x. PMID:36195757.
42. Machado RD, Welch CL, Haimel M, Bleda M, Colglazier E, Coulson JD, et al. Biallelic variants of ATP13A3 cause dose-dependent childhood-onset pulmonary arterial hypertension characterised by extreme morbidity and mortality. J Med Genet. 2022;59(9):906–11. doi:10.1136/jmedgenet-2021-107831. PMID:34493544.
43. Swietlik EM, Greene D, Zhu N, Megy K, Cogliano M, Rajaram S, et al. Bayesian inference associates rare KDR variants with specific phenotypes in pulmonary arterial hypertension. Circ Genom Precis Med. 2020;14(1):e003155. doi:10.1161/CIRCGEN.120.003155. PMID:33320693.
44. Eyries M, Montani D, Girerd B, Favrolt N, Riou M, Faivre L, et al. Familial pulmonary arterial hypertension by KDR heterozygous loss of function. Eur Respir J. 2020;55(4). doi:10.1183/13993003.02165-2019. PMID:31980491.
45. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Ischio-pubic-patellar dysplasia (MIM 147891). In: Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford: Oxford University Press; 2018. p. 782.
46. Zhu N, Welch CL, Wang J, Allen PM, Gonzaga-Jauregui C, Ma L, et al. Rare variants in SOX17 are associated with pulmonary arterial hypertension with congenital heart disease. Genome Med. 2018;10(1):56. doi:10.1186/s13073-018-0566-x. PMID:30029678.
47. Best DH, Sumner KL, Austin ED, Chung WK, Brown LM, Borczuk AC, et al. EIF2AK4 mutations in pulmonary capillary hemangiomatosis. Chest. 2014;145(2):231–6. doi:10.1378/chest.13-2366. PMID:24135949.
48. Larkin EK, Newman JH, Austin ED, Hemnes AR, Wheeler L, Robbins IM, et al. Longitudinal analysis casts doubt on the presence of genetic anticipation in heritable pulmonary arterial hypertension. Am J Respir Crit Care Med. 2012;186(9):892–6. doi:10.1164/rccm.201205-0886OC. PMID:22923661.
49. Ivy D, Rosenzweig EB, Abman SH, Beghetti M, Bonnet D, Douwes JM, et al. Embracing the challenges of neonatal and paediatric pulmonary hypertension. Eur Respir J. 2024;64(4):2401345. doi:10.1183/13993003.01345-2024. PMID:39209483.
50. Shaukat M, Grünig E, Haas S, Haas J, Panahi M, Granzow M, et al. Heritable pulmonary arterial hypertension: new genetic findings and environmental triggers. Sci Rep. 2026;16(1):4121. doi:10.1038/s41598-025-34167-0. PMID:41611798.
51. Limoges M, Langleben D, Fox BD, Shear R, Wieczorek P, Rudski LG, et al. Pregnancy as a possible trigger for heritable pulmonary arterial hypertension. Pulm Circ. 2016;6(3):381–3. doi:10.1086/686993. PMID:27683615.
52. Girerd B, Montani D, Coulet F, Sztrymf B, Yaici A, Jaïs X, et al. Clinical outcomes of pulmonary arterial hypertension in patients carrying an ACVRL1 (ALK1) mutation. Am J Respir Crit Care Med. 2010;181(8):851–61. doi:10.1164/rccm.200908-1284OC. PMID:20056902.
53. Hernandez-Gonzalez I, Tenorio J, Palomino-Doza J, Martinez Meñaca A, Morales Ruiz R, Lago-Docampo M, et al. Clinical heterogeneity of pulmonary arterial hypertension associated with variants in TBX4. PLoS One. 2020;15(4):e0232216. doi:10.1371/journal.pone.0232216. PMID:32348326.
54. Tenorio J, Navas P, Barrios E, Fernández L, Nevado J, Quezada CA, et al. A founder EIF2AK4 mutation causes an aggressive form of pulmonary arterial hypertension in Iberian Gypsies. Clin Genet. 2015;88(6):579–83. doi:10.1111/cge.12549. PMID:25512148.
55. He Y, Li Q, Zhang C, Keller BB, Shen Y, Gu H. The genetic epidemiology and genotype-phenotype correlations among Chinese children with idiopathic and heritable pulmonary arterial hypertension. Respir Res. 2025;26(1):231. doi:10.1186/s12931-025-03249-y. PMID:40604975.
56. Montani D, Girerd B, Jaïs X, Laveneziana P, Lau EMT, Bouchachi A, et al. Screening for pulmonary arterial hypertension in adults carrying a BMPR2 mutation. Eur Respir J. 2021;58(1):2004229. doi:10.1183/13993003.04229-2020. PMID:33380512.
57. Shrivastava S, Kruisselbrink TM, Mohananey A, Thomas BC, Kushwaha SS, Pereira NL. Rare TBX4 variant causing pulmonary arterial hypertension with small patella syndrome in an adult man. JACC Case Rep. 2021;3(12):1447–52. doi:10.1016/j.jaccas.2021.05.019. PMID:34557690. Full text (PMC8446047).
58. Mulder BJM, van de Laar IMBH, De Backer J. Heritable thoracic aortic diseases: syndromal and isolated (F)TAAD. In: Baars HF, van der Smagt JJ, Doevendans PAFM, van Tintelen JP, editors. Clinical Cardiogenetics. 3rd ed. Cham: Springer; 2020. p. 309–43 (p. 321, 324 and 329). doi:10.1007/978-3-030-45457-9_19.
59. Jones KL, Jones MC, del Campo M. Myhre syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 692–697.
60. Schade van Westrum SM, Wahbi K, Bonne G, van der Kooi AJ. Hereditary neuromuscular diseases and cardiac involvement. In: Baars HF, van der Smagt JJ, Doevendans PAFM, van Tintelen JP, editors. Clinical Cardiogenetics. 3rd ed. Cham: Springer; 2020. p. 371–85 (p. 371). doi:10.1007/978-3-030-45457-9_22.
61. Viatris Specialty LLC. NORVASC (amlodipine besylate) tablets. Prescribing Information. NDA 019787; version of 15 Feb 2023. DailyMed. Accessed 26 Oct 2026.
62. Par Health USA, LLC. Felodipine extended-release tablets, USP. Prescribing Information. ANDA 200815; version of 30 Sep 2024 (SPL v15). DailyMed. Accessed 1 Nov 2026.
63. Bausch Health US LLC. CARDIZEM CD (diltiazem hydrochloride) extended-release capsules. Prescribing Information. NDA 020062; version of 10 Dec 2025. DailyMed. Accessed 26 Oct 2026.
64. Nifedipine extended-release tablets. Prescribing Information. ANDA 075269; version of 19 Mar 2026. DailyMed. Accessed 26 Oct 2026.
65. Chartwell RX, LLC. Nifedipine extended-release tablets. Prescribing Information. ANDA 077127; version of 17 Dec 2025. DailyMed. Accessed 26 Oct 2026.
66. Galiè N, Barberà JA, Frost AE, Ghofrani HA, Hoeper MM, McLaughlin VV, et al. Initial use of ambrisentan plus tadalafil in pulmonary arterial hypertension. N Engl J Med. 2015;373(9):834–44. doi:10.1056/NEJMoa1413687. PMID:26308684.
67. Gilead Sciences, Inc. LETAIRIS (ambrisentan) tablets. Prescribing Information. NDA 022081; version of 22 Jun 2026. DailyMed. Accessed 7 Oct 2026.
68. Actelion Pharmaceuticals US, Inc. TRACLEER (bosentan) tablets and tablets for oral suspension. Prescribing Information. NDA 021290/209279; version of 17 Nov 2025. DailyMed. Accessed 7 Oct 2026.
69. Actelion Pharmaceuticals US, Inc. OPSUMIT (macitentan) tablets. Prescribing Information. NDA 204410; version of 11 Apr 2025. DailyMed. Accessed 7 Oct 2026.
70. Actelion Pharmaceuticals US, Inc. OPSYNVI (macitentan and tadalafil) tablets. Prescribing Information. NDA 218490; version of 11 Apr 2025. DailyMed. Accessed 7 Oct 2026.
71. European Medicines Agency. Medicines data (medicines output, JSON report): Winrevair, Opsumit, Volibris, Adcirca, Revatio, Adempas, Uptravi, Tracleer, Yuvanci, Trepulmix, Ventavis, Thelin [Internet]. EMA report. Version of 7 Oct 2026.
72. Viatris Specialty LLC. REVATIO (sildenafil) tablets, oral suspension and injection. Prescribing Information. NDA 021845/022473/203109; version of 18 Dec 2024. DailyMed. Accessed 7 Oct 2026.
73. United Therapeutics Corporation. ADCIRCA (tadalafil) tablets. Prescribing Information. NDA 022332; version of 15 Dec 2025. DailyMed. Accessed 7 Oct 2026.
74. Bayer HealthCare Pharmaceuticals Inc. ADEMPAS (riociguat) tablets. Prescribing Information. NDA 204819; version of 13 Aug 2026. DailyMed. Accessed 7 Oct 2026.
75. Actelion Pharmaceuticals US, Inc. UPTRAVI (selexipag) tablets and for injection. Prescribing Information. NDA 207947/214275; version of 28 May 2026. DailyMed. Accessed 7 Oct 2026.
76. Actelion Pharmaceuticals US, Inc. VELETRI (epoprostenol) for injection. Prescribing Information. NDA 022260; version of 28 Nov 2022. DailyMed. Accessed 26 Oct 2026.
77. United Therapeutics Corporation. REMODULIN (treprostinil) injection. Prescribing Information. NDA 021272; version of 2 Jul 2026. DailyMed. Accessed 26 Oct 2026.
78. United Therapeutics Corporation. TYVASO (treprostinil) inhalation solution. Prescribing Information. NDA 022387; version of 14 Apr 2026. DailyMed. Accessed 26 Oct 2026.
79. United Therapeutics Corporation. ORENITRAM (treprostinil) extended-release tablets. Prescribing Information. NDA 203496; version of 20 Nov 2024. DailyMed. Accessed 26 Oct 2026.
80. Bayer AG. Ventavis (iloprost) nebuliser solution: EPAR — Summary of Product Characteristics (EU/1/03/255). European Medicines Agency. EPAR. Accessed 26 Oct 2026.
81. European Medicines Agency. Winrevair (sotatercept): EPAR — Summary of Product Characteristics; Procedural steps taken and scientific information after the authorisation (EMEA/H/C/005647) [Internet]. EPAR. Accessed 7 Oct 2026.
82. Viatris EESV. Revatio (sildenafil): EPAR — Summary of Product Characteristics (EU/1/05/318). European Medicines Agency. EPAR. Accessed 26 Oct 2026.
83. Janssen-Cilag International NV. Tracleer (bosentan): EPAR — Summary of Product Characteristics (EU/1/02/220). European Medicines Agency. EPAR. Accessed 26 Oct 2026.
84. U.S. Food and Drug Administration. Drugs@FDA (openFDA drugsfda): approval history for BLA 761363, NDA 021290, 022081, 204410, 218490, 021845, 022332, 204819, 207947, 020444, 022260, 021272, 022387, 203496 and 021779 [Internet]. openFDA. Accessed 7 Oct 2026.
85. U.S. Food and Drug Administration. Supplement approval letters for Winrevair (sotatercept-csrk), BLA 761363/S-007 (24 Oct 2025) and BLA 761363/S-012, S-016, S-017 (21 Sep 2026). Letter S-007; Letter S-012/S-016/S-017. Accessed 7 Oct 2026.
86. Bayer AG. Adempas (riociguat) film-coated tablets and granules for oral suspension: EPAR — Summary of Product Characteristics (EU/1/13/907). European Medicines Agency. EPAR. Accessed 26 Oct 2026.
87. Hoeper MM, Badesch DB, Ghofrani HA, Gibbs JSR, Gomberg-Maitland M, McLaughlin VV, et al. Phase 3 trial of sotatercept for treatment of pulmonary arterial hypertension. N Engl J Med. 2023;388(16):1478–90. doi:10.1056/NEJMoa2213558. PMID:36877098.
88. Humbert M, McLaughlin VV, Badesch DB, Ghofrani HA, Gibbs JSR, Gomberg-Maitland M, et al. Sotatercept in patients with pulmonary arterial hypertension at high risk for death. N Engl J Med. 2025;392(20):1987–2000. doi:10.1056/NEJMoa2415160. PMID:40167274.
89. McLaughlin VV, Hoeper MM, Badesch DB, Ghofrani HA, Gibbs JSR, Gomberg-Maitland M, et al. Sotatercept for pulmonary arterial hypertension within the first year after diagnosis. N Engl J Med. 2025;393(16):1599–611. doi:10.1056/NEJMoa2508170. PMID:41025556.
90. Montani D, McLaughlin VV, Gibbs JSR, Gomberg-Maitland M, Hoeper MM, Preston IR, et al. Consistent safety and efficacy of sotatercept for pulmonary arterial hypertension in BMPR2 mutation carriers and noncarriers: a planned analysis of a phase II, double-blind, placebo-controlled clinical trial (PULSAR). Am J Respir Crit Care Med. 2025;211(6):1028–37. doi:10.1164/rccm.202409-1698OC. PMID:40035659.
91. ClinicalTrials.gov. Records NCT05587712 (MOONBEAM), NCT05975905 (TROPOS), NCT07181382 (SERANATA), NCT07612657 (TRANSCEND), NCT01647945 (TransformPAH), NCT05934526 (PROSERA) and NCT03626688 (ADVANCE OUTCOMES) [Internet]. API v2. Accessed 7 Oct 2026.
92. Schneijdenberg JW, van den Bosch AE, Kap ME, Kauling RM, Chandoesing PP, van den Toorn LM, et al. Sotatercept in TBX4 associated heritable pulmonary arterial hypertension: a case report of a super responder. Pulm Circ. 2026;16(2):e70319. doi:10.1002/pul2.70319. PMID:42146787.
93. Langleben D, Lesenko L, Fox BD, Eintracht S, Foulkes WD, Rosenblatt DS. Ineffectiveness of sotatercept therapy in a patient with heritable pulmonary arterial hypertension associated with a previously unreported missense variant in GDF2, the gene for bone morphogenic protein-9. Chest. 2025;167(2):e37–9. doi:10.1016/j.chest.2024.09.002. PMID:39939060.
94. GlaxoSmithKline (Ireland) Limited. Volibris (ambrisentan): EPAR — Summary of Product Characteristics (EU/1/08/451). European Medicines Agency. EPAR. Accessed 26 Oct 2026.
95. Eli Lilly Nederland B.V. ADCIRCA (tadalafil): EPAR — Summary of Product Characteristics (EU/1/08/476). European Medicines Agency. EPAR. Accessed 26 Oct 2026.
96. Janssen-Cilag International NV. Opsumit (macitentan) film-coated and dispersible tablets: EPAR — Summary of Product Characteristics (EU/1/13/893). European Medicines Agency. EPAR. Accessed 26 Oct 2026.
97. Gossamer Bio, Inc. Form 8-K: topline results from the Phase 3 PROSERA study of seralutinib (23 Feb 2026); Exhibit 99.1: Second quarter 2026 financial results and business update (13 Aug 2026). U.S. Securities and Exchange Commission. 8-K of 23 Feb 2026; Exhibit 99.1 of 13 Aug 2026. Accessed 7 Oct 2026.
98. United Therapeutics Corporation. Exhibit 99.1 to Form 8-K: ADVANCE OUTCOMES study of ralinepag (2 Mar 2026); Form 10-Q for quarter ended 30 Jun 2026. U.S. Securities and Exchange Commission. 8-K of 2 Mar 2026; 10-Q. Accessed 7 Oct 2026.
99. Girerd B, Montani D, Jaïs X, Eyries M, Yaici A, Sztrymf B, et al. Genetic counselling in a national referral centre for pulmonary hypertension. Eur Respir J. 2016;47(2):541–52. doi:10.1183/13993003.00717-2015. PMID:26699722.
100. PHA Europe — Pulmonary Hypertension Association Europe [Internet]. phaeurope.org. Accessed 7 Oct 2026.
101. PHAware Global Association [Internet]. phaware.global. Accessed 7 Oct 2026.
102. Rodighiero J, Durbin J, Badalato L, Archer SL. Early detection and therapy of asymptomatic group 1 pulmonary hypertension following genetic testing: a case involving familial KCNK3 variants. CJC Open. 2026;8(8):1060–3. doi:10.1016/j.cjco.2026.04.013. PMID:42610021.