Blood Proteins That Flag Hidden Lung Pressure Discovered in Landmark Study

Pulmonary hypertension has long been one of medicine’s quiet killers, a disease in which pressure inside the arteries that carry blood from the heart to the lungs creeps upward until the right side of the heart begins to fail. By the time symptoms such as breathlessness and fatigue appear, the damage is often advanced, and the definitive diagnostic test, an invasive catheter threaded through the veins into the pulmonary circulation, is reserved for patients already suspected of disease. What has been missing is a simple blood test that could reveal the earliest stages of pulmonary vascular stress, long before the heart begins to remodel. Now a large-scale proteomic study published in Nature Communications has delivered the most comprehensive picture yet of the circulating proteins that track pulmonary artery pressure and predict who will go on to develop the disease.

The research team, led by investigators at UT Southwestern Medical Center and Brigham and Women’s Hospital, drew on several independent observational cohorts, including the Atherosclerosis Risk in Communities (ARIC) study and the UK Biobank. Using the SomaScan platform, a technology capable of measuring thousands of proteins simultaneously from a small blood sample, they searched for proteins whose abundance tracked with pulmonary artery systolic pressure estimated by echocardiography. The scale of the analysis matters: rather than examining a handful of candidate biomarkers chosen for their known biology, the team let the data speak, screening the circulating proteome in an unbiased fashion and then validating their findings across separate populations and against independent clinical measurements.

The screen yielded 22 proteins significantly associated with echocardiography-derived pulmonary artery systolic pressure. That number alone would be noteworthy, but the investigators went further, testing whether these same proteins were linked to the gold-standard invasive measures of pulmonary vascular disease obtained during right heart catheterization. Several of the 22 passed this stricter test, meaning their blood levels did not merely echo the ultrasound estimate but genuinely reflected the hemodynamic reality of elevated pulmonary vascular resistance. Even more striking, some of the proteins were associated with structural remodeling of the pulmonary vasculature itself, the thickening and stiffening of vessel walls that represents the pathological substrate of the disease.

This distinction between correlation and mechanism is where the study makes its most provocative contribution. Pulmonary vascular remodeling is the process by which the normally thin, compliant walls of the lung arteries accumulate smooth muscle and scar-like tissue, narrowing the lumen through which blood must flow. If circulating proteins mark this process in asymptomatic people, they could serve as early-warning signals, flagging individuals whose vessels are already under stress before any measurable elevation in pressure. The authors report that several of their candidate biomarkers implicate this early-stage remodeling, effectively pushing the detectable window of the disease back toward its origins.

Among the 22 proteins, one gene stood out: NOTCH3. Genetic analyses conducted as part of the study support a potential causal role for NOTCH3 in pulmonary vascular disease, a finding that will resonate with vascular biologists. NOTCH3 encodes a receptor protein concentrated in vascular smooth muscle cells, where it governs how those cells respond to injury and mechanical stress. When NOTCH3 signaling is perturbed, smooth muscle cells can proliferate and take on a contractile, stiff phenotype that narrows vessels. The new genetic evidence, which suggests that inherited variation in NOTCH3 activity may actually drive susceptibility rather than merely accompany it, elevates the protein from biomarker to potential therapeutic target.

The genetic work produced a second, subtler insight. The researchers identified a common splice-regulating genetic variant that may modify susceptibility to pulmonary vascular stressors. Splice-regulating variants influence how a gene’s messenger RNA is assembled, potentially altering the balance of protein isoforms produced. In practical terms, this means that some individuals may carry a common genetic tweak that changes how their vessels respond to environmental or hemodynamic insults, such as chronic low oxygen, left heart disease, or autoimmune inflammation. The discovery hints at a gene-by-environment architecture for pulmonary hypertension, in which the same stressor produces disease in genetically susceptible people but not in others.

Perhaps the most conceptually ambitious part of the study involved protein-based clustering, a machine-learning approach that groups individuals according to the overall pattern of their circulating proteins rather than any single measurement. The analysis revealed two distinct biological subtypes, or endotypes, associated with increased risk of pulmonary hypertension in the community. This is a departure from the traditional clinical classification of pulmonary hypertension, which sorts patients into groups based on the underlying cause, such as left heart disease, lung disease, or blood clots. The new endotypes instead divide people by biology, suggesting that different molecular pathways converge on the same clinical endpoint, and that prevention strategies might one day be tailored to a patient’s proteomic signature rather than their diagnosis code.

The methodological rigor underlying these findings deserves emphasis. Proteomic studies of this kind face well-known pitfalls: protein levels are influenced by kidney function, inflammation, age, and pre-analytical sample handling, and associations detected in one cohort often evaporate in another. By requiring replication across multiple independent cohorts, by validating against invasive hemodynamics rather than relying solely on echocardiographic estimates, and by integrating genetic instruments to probe causality, the team built a layered case in which each line of evidence compensates for the weaknesses of the others. The use of the ARIC study, a community-based cohort with decades of follow-up, allowed the investigators to test whether baseline protein levels predicted subsequent clinical pulmonary hypertension, converting a cross-sectional snapshot into a genuinely prospective risk model.

The clinical implications are substantial. Pulmonary hypertension affects millions worldwide and carries a prognosis that worsens sharply with delay in diagnosis, yet current screening tools are blunt. Echocardiography is imperfect at estimating pulmonary pressures and is not routinely performed in asymptomatic people. A panel of circulating biomarkers could, in principle, be incorporated into routine blood work, identifying individuals who warrant closer surveillance or early referral to specialized pulmonary vascular centers. The two endotypes add a further layer of potential utility, since they might eventually guide which preventive interventions are offered, much as molecular subtyping has transformed oncology. None of this is imminent; the proteins identified here are candidates, and their performance as screening tools must be tested in prospective, outcome-driven studies before any clinical deployment.

Still, the study marks a turning point in how the field thinks about pulmonary vascular disease. It reframes pulmonary hypertension not as a sudden hemodynamic event but as the end stage of a molecular process that leaves fingerprints in the blood years in advance. It nominates NOTCH3 as a causal actor whose signaling pathway could be druggable with agents already in development for other vascular conditions. And it demonstrates that large-scale proteomics, applied across deep clinical cohorts and interrogated with genetic tools, can extract biological meaning from community-dwelling populations rather than only from patients already sick enough to undergo catheterization. As the cost of proteomic profiling continues to fall and biobanks grow richer, the approach pioneered here is likely to be extended to other forms of vascular disease, offering a template for turning the circulating proteome into a window on the body’s hidden pathophysiology. For patients and clinicians confronting a disease that has historically announced itself too late, the prospect of catching it early, with a simple blood draw, is a genuinely hopeful development.

Subject of Research: Proteomic biomarkers of pulmonary artery pressure and pulmonary hypertension risk

Article Title: Large-scale proteomics identifies circulating markers of pulmonary pressure and pulmonary hypertension risk

Article References: Giugni, F. R., Santos, M., Yang, Y., Lamberson, V., Claggett, B., Hoogeveen, R., Ballantyne, C., Systrom, D., Waxman, A., Buckley, L., Yu, B., San José Estépar, R., Hegde, S. M., & Shah, A. M. (2026). Large-scale proteomics identifies circulating markers of pulmonary pressure and pulmonary hypertension risk. Nature Communications. https://doi.org/10.1038/s41467-026-78432-w

Image Credits: AI Generated

DOI: 10.1038/s41467-026-78432-w

Keywords: pulmonary hypertension, proteomics, biomarkers, NOTCH3, pulmonary vascular remodeling, echocardiography, right heart catheterization, SomaScan, UK Biobank, ARIC study, endotypes, vascular biology