Clustering Pulmonary Hypertension Patients Using the Plasma Proteome.

INTRODUCTION

Patients with pulmonary hypertension are classified according to clinical criteria to inform treatment decisions. Knowledge of the molecular drivers of pulmonary hypertension might better inform treatment choice.

METHODS

Between 2013 and 2021, 470 patients with pulmonary hypertension, 136 disease controls and 59 healthy controls were enrolled as a discovery cohort. Plasma levels of 7288 proteins were assayed (SomaScan 7K platform). Proteins that distinguished pulmonary hypertension from both control groups were selected for unsupervised clustering (k-means clustering of UMAP dimensions). Clinical characteristics and outcomes were compared across clusters. Separate cohorts of serially sampled patients from pulmonary hypertension centers in the United Kingdom (n=229) and France (n=79) provided independent validation.

RESULTS

156 plasma proteins that distinguished pulmonary hypertension from disease and healthy controls formed 4 clusters with diverse 5-year survival rates: 78% (cluster 4), 62% (cluster 2), 44% (cluster 3), and 33% (cluster 1). The distinction and clinical relevance of the clusters were confirmed in validation cohorts by their association with survival. To further characterise the therapeutic relevance of the clusters we investigated 2 experimental drug targets: the Platelet-Derived Growth Factor (PDGF) pathway was up-regulated in cluster 3 compared to other clusters and the Transforming Growth Factor-β (TGF-β) pathway was up-regulated in cluster 1.

CONCLUSION

Plasma proteomic profiling of patients with pulmonary hypertension distinguishes 4 clusters, independent of the clinical classification. These groups, based on differential plasma protein levels, could act as theragnostic biomarkers for new therapies targeting PDGF and TGF-β pathways. This article is open access and distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).