• Study Seeks Better Classification System for Pulmonary Vascular Disease

    A new precision medicine initiative to intensively characterize pulmonary hypertension (PH) and pulmonary vascular disease (PVD) through the use of deep clinical phenotyping and comprehensive “omics” analysis – known as PVDOMICS – will drive a new understanding of PVD and refine the current classification methods, say researchers, based on initial findings.

    The study, published online Monday and in the Aug. 16 issue of the Journal of the American College of Cardiology, noted that the diagnosis and clinical classification of PH requires chest and cardiac imaging, pulmonary function testing and cardiopulmonary hemodynamics, according to the World Symposium on Pulmonary Hypertension (WSPH).

    Within this framework, the results of clinical testing assign patients to one of five groups based on presumed common disease characteristics, similar pathological features and predicted response to treatment, said the authors of the study – led by Anna R. Hemnes, MD, from Vanderbilt University Medical Center, Nashville, Tennessee, on behalf of the PVDOMICS Study Group.

    However, the team noted that although the traditional WSPH classification is useful for defining treatment and prognosis at some level, the entire spectrum of PVD and PH may not be fully captured by such classifications.

    Indeed, they added that the potential to improve understanding and clinical care of PVD, coupled with advances in translational science, has led the National Institutes of Health and National Heart, Lung and Blood Institute to support PVDOMICS (Redefining Pulmonary Hypertension through Pulmonary Vascular Disease Phenomics).

    “This approach may refine the traditional WSPH groups 1-5 classification by potentially identifying new, more precise phenotypes within PVD that will enhance scientific understanding and lead to novel targeted therapeutics,” said Hemnes and colleagues, adding that PVDOMICS tests the general hypothesis that integration of clinical metrics with detailed “omic” measures in the blood will enhance understanding of PVD, identify cohorts for novel therapies and potentially facilitate an updated classification scheme for PVD.

    Study setup

    Via the PVDOMICS Study Group, Hemnes and colleagues set out to describe clinical characteristics and transplant-free survival in 1,193 people recruited into the PVDOMICS cohort, of whom 750 had PH, dominated by idiopathic pulmonary arterial hypertension in WSPH Group 1 patients and heart failure with preserved ejection fraction in Group 2 patients.

    By WSPH groups, there were: 353 patients from Group 1; 136 Group 2 patients; 172 from Group 3; 57 from Group 4; and 32 patients from Group 5.

    Of those recruited, just under 27% of recruits were new patients, said the authors, noting that the median duration of PH in group 1 was 4.6 years (with 84% already on PH-directed treatment) and was 1.3-3.2 years in the other groups.

    All participants in PVDOMICS underwent a deep phenotyping protocol that included cardiopulmonary exercise testing and sleep studies, with the intention of recording >2,000 clinical variables on each participant. Also collected were 6-minute walk testing, pulmonary function testing, home overnight sleep study, chest computed tomography (CT), ventilation/perfusion lung scan, electrocardiogram, echocardiogram, cardiac magnetic resonance imaging, routine clinical laboratory measurements and cardiopulmonary exercise testing.

    “This comprehensive phenotyping protocol, performed across the full spectrum of PVD, disease comparator groups, and healthy control subjects, uniquely positions the PVDOMICS cohort to better understand PH as a spectrum and appreciate heterogeneity within a WSPH group and similarities across WSPH groups,” noted the authors.

    Key findings

    “Early observations of this cohort include not only a description of the spectrum of pulmonary vascular disease and precursors in the United States, but also offer novel observations of similarities and differences across PH groups and comparators,” said the authors.

    “A surprising observation is the nearly 40% prevalence of mixed etiology PH,” they revealed, noting that multiple WSPH groups were identified in 38.9% of PH subjects.

    Hemnes and colleagues noted that novel findings from the analysis included: lower diffusing capacity for carbon monoxide (DLCO) in Groups 1-3; enlarged right atrial (RA) volume index shared between Groups 1-4; unexpected, frequent presence of ground glass opacities on CT; sleep alterations in Group 1; and poorest survival in Group 3 PH.

    Indeed, they added that 50.2% of Group 1 subjects were found to have ground glass opacities on chest computed tomography, while nocturnal desaturation was frequently observed in Groups 1, 3 and 4.

    The team concluded that PVDOMICS achieved its goal of enrolling subjects across a spectrum of PVD – including a subset with mild disease and mixed etiology PH – both of which are understudied groups.

    “Early findings among PH patients identified low DLCO as a common feature in Groups 1-3 PH, the unexpected presence of ground glass opacities in one-half of Group 1 enrollees, and a common finding of enlarged RA in Groups 1-4 PH even without RV changes, suggesting that this may be a sensitive indicator of PH,” said the team.

    “This rich data set is facilitating new understanding of the full spectrum of PVD and will allow refinement of the current PVD classification,” they said.

    Patients do not always fit neatly into groups

    Writing in an accompanying editorial comment, Martin R. Wilkins, MD, from the National Heart and Lung Institute at Imperial College London, noted that PH is a more prevalent comorbidity than commonly understood – adding that a lack of specific symptoms means that PH can go unsuspected for many years.

    Wilkins added that while the WSPH clinical classification is used to inform treatment decisions, “it is acknowledged that the major groups are lumpy and harbor heterogeneity.”

    For example, he noted that Group 1 includes congenital heart disease and connective tissue disease as well as idiopathic and heritable pulmonary arterial hypertension, whereas Group 3 hosts chronic obstructive pulmonary disease, idiopathic lung fibrosis, sleep apnea and chronic hypoxia.

    “Patients do not always fit neatly into one group, and judgments based on binary decisions (whether the pulmonary artery wedge pressure is above or below 15 mm Hg) and semiquantitative assessments (the degree of parenchymal lung shadowing on chest computed tomography) can lead to patients being ‘misclassified’,” he said.

    The editorialist noted that the new study begins to draw attention to clinical features shared by the major groupings, but added that the “real potential” of the PVDOMICS study data lies with a deeper dive into the data.

    “A quick win would be to use machine learning to interrogate just the clinical data collected so far,” he said, noting that the few forays with this approach into patients with idiopathic pulmonary arterial hypertension – utilizing up to 30 available parameters – have identified three to four clusters.

    “It will be fascinating to see what machine learning makes of the richer PVDOMICS data set from all PH presentations,” he said. “The planned genetic, transcriptomic, proteomic, and metabolomic analyses will further enhance the vocabulary available to describe each patient and likely offer insights into the main molecular drivers underpinning each patient cluster.”

    “Identifying druggable targets among the molecular pathology for new or repurposed medicines coupled with robust biomarkers defining each patient will advance personalized treatment.”


    Hemnes AR, Leopold JA, Radeva MK, et al. Clinical Characteristics and Transplant Free Survival Across the Spectrum of Pulmonary Vascular Disease. J Am Coll Cardiol 2022;80:697-718.

    Wilkins MR. Pulmonary Hypertension: Dissecting a Complex Phenotype. J Am Coll Cardiol 2022;80:719-721.

    Image Credit: ibreakstock – stock.adobe.com

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Review our Privacy Policy for more details