Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Nov 20;5(1):505.
doi: 10.1038/s43856-025-01213-x.

Proteomics-based clustering outperforms clinical clustering in identifying people with heart failure with distinct outcomes

Marion van Vugt #  1   2   3   4 , Ruicong She #  5 , Isabella Kardys #  6 , Teun B Petersen  6   7 , Marie de Bakker  6 , K Martijn Akkerhuis  6 , Kadir Caliskan  6 , Olivier C Manintveld  6 , Alicia Uijl  8   9   10 , Jan van Ramshorst  11 , Dimitris Rizopoulos  7   12 , Victor Awm Umans  11 , Eric Boersma  6 , David E Lanfear  13   14 , Folkert W Asselbergs  8   9   15   16 , Jessica van Setten  17 , A Floriaan Schmidt  18   19   20   21   22
Affiliations

Proteomics-based clustering outperforms clinical clustering in identifying people with heart failure with distinct outcomes

Marion van Vugt et al. Commun Med (Lond). .

Abstract

Background: Heart failure (HF) clustering typically relies on clinical characteristics which may not reflect underlying pathophysiology relevant for personalized medicine. We aimed to identify plasma protein profiles of HF patients with reduced ejection fraction (HFrEF).

Methods: Using latent class analysis, we derived clusters based on 1) clinical characteristics, and 2) proteomics (SomaScan) from 379 HFrEF patients (median age 64 years [Q1 56; Q3 72], 73% male). Survival analysis assessed associations with major cardiovascular (CV) events (HF hospitalization, CV death, or advanced therapy), HF hospitalization, CV death, and all-cause mortality. Associations were validated in 511 external patients (median age 72 years [Q1 63; Q3 79], 70% male). We identified differentially expressed proteins and explored whether proteins are targets of developmental or approved drugs.

Results: We show that clinical clustering identifies three patient clusters without distinct disease progression. Contrary to this, clustering based on plasma proteomics identifies three patient clusters with clear differences in disease, which are validated in the external cohort. The slowly progressing cluster 1 includes younger patients with fewer comorbidities, while the rapidly progressing cluster 3 consists of older patients with more atrial fibrillation and renal failure, and the hospitalization cluster 2 is intermediate in many characteristics. Medication use is similar across clusters. Relative to cluster 1, patients in cluster 2 have an increased risk of major CV events (HR 2.31, 95%CI 1.23; 4.36) and HF hospitalization (HR 2.30, 95%CI 1.10; 4.78). Patients in cluster 3 experienced increased event rates of major CV events (HR 5.84), HF hospitalization (6.50), CV death (8.58), and all-cause mortality (5.07). Twelve proteins are differentially expressed across the identified clusters, including druggable CD2, GDF-15, ABO, IGFBP-1, IGFBP-2, and RNase1.

Conclusions: Proteomics-based clustering identifies three HFrEF clusters associated with distinct outcomes that remain undetected using only clinical characteristics.

Plain language summary

Heart failure affects millions of people worldwide, but symptoms and disease course varies greatly. People are often grouped based on basic clinical characteristics, which may miss important biological differences. In this study, we analyse blood proteins from people with heart failure and compare grouping based on these to a grouping based on clinical characteristics. We identify three biological groups of people with heart failure, and each group has a different future risk of hospitalization and death. The results are confirmed in an independent patient group. Our findings suggest that protein profiling can reveal hidden disease subtypes, which could help tailor treatments and improve outcomes for heart failure patients. We also identify proteins that could provide promising drug targets for specific patient groups.

PubMed Disclaimer

Conflict of interest statement

Competing interests: O.C.M. has served on advisory boards of Abbott, Astra Zeneca, Boehringer Ingelheim, and Novartis. A.F.S. has received funding from New Amsterdam and Servier for unrelated work and is an Editorial Board Member for Communications Medicine but was not involved in the editorial review or peer review, nor in the decision to publish this article. I.K. has received travel reimbursement from SomaLogic and Olink. D.E.L. is a consultant for Astra Zeneca, Bayer, Cytokinetics, Illumina, RyCarma, has participated in research with Akros, AstraZeneca, Cytokinetics, Lilly, Kardigan, Novartis, Pfizer, Somalogic, and has a patent (held by Henry Ford Health) for a beta‐blocker response polygenic score. All other authors declare no competing interest.

Figures

Fig. 1
Fig. 1. Proportion of clinical characteristics per cluster.
Bars indicate the distribution of clinical characteristic for the clinical (left) and proteomic clustering (right). Numerical data underlying the figure are presented in Supplementary Data 4-5. AF atrial fibrillation, CAD coronary artery disease, MI myocardial infarction, NYHA New York Heart Association.
Fig. 2
Fig. 2. HFrEF patient flow between the different clustering models.
The flow represents the proportion of HFrEF patients (n = 379) that were assigned to different clusters in the various approaches. For example, this figure shows that the clinical clusters (i.e., ischemic cluster 1 in blue, hypertensive cluster 2 in yellow, and cardiomyopathy cluster 3 in red) are very different from the proteomic clusters, whereas the proteomic slowly progressing cluster 1 and combined cluster 1 are almost identical.
Fig. 3
Fig. 3. Event-free survival is significantly different across proteomics-based clusters.
Kaplan–Meier curve for the clinical outcomes stratified by proteomic clusters. Differences were assessed using the two-sided log-rank test without adjustment for multiple testing, resulting in significant differences for all outcomes: p = 7.57 × 10–11 for major cardiovascular event, p = 2.91 × 10–10 for HF hospitalization, p = 1.53 × 10–8 for cardiovascular death, and p = 2.00 × 10–8 for all-cause mortality. HF heart failure.
Fig. 4
Fig. 4. The association between proteomics-based cluster membership and disease progression in people with HFrEF.
Follow-up was truncated at three years. The model was derived in 379 participants of the BioSHiFT cohort and externally validated in 511 participants of the HFPGR. Bars represent the 95% confidence interval. Numerical data underlying the figure are presented in Supplementary Data 9. CI confidence interval, HF heart failure, HFPGR Henry Ford HF PharmacoGenomic Registry, HR hazard ratio.
Fig. 5
Fig. 5. Association of differentially expressed proteins and canonical cardiac proteins with clinical outcomes.
Follow-up was truncated at three years. The model was derived in 379 participants of the BioSHiFT cohort. Bars represent the 95% confidence interval. CI confidence interval, HF heart failure, HR hazard ratio.
See this image and copyright information in PMC

References

    1. McDonagh, T. A. et al. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: Developed by the Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC) With the special contribution of the Heart Failure Association (HFA) of the ESC. Eur. Heart J.42, 3599–3726 (2021). - DOI - PubMed
    1. Butler, J., Fonarow, G. C. & Gheorghiade, M. Strategies and opportunities for drug development in heart failure. JAMA309, 1593–1594 (2013). - DOI - PubMed
    1. Vaduganathan, M., Greene, S. J., Ambrosy, A. P., Gheorghiade, M. & Butler, J. The disconnect between phase II and phase III trials of drugs for heart failure. Nat. Rev. Cardiol.10, 85–97 (2013). - DOI - PubMed
    1. Ahmad, T. et al. Clinical implications of chronic heart failure phenotypes defined by cluster analysis. J. Am. Coll. Cardiol.64, 1765–1774 (2014). - DOI - PMC - PubMed
    1. Daniels, L. B. & Maisel, A. S. Natriuretic peptides. J. Am. Coll. Cardiol.50, 2357–2368 (2007). - DOI - PubMed

LinkOut - more resources