Clinical, histological, and transcriptomic clustering capture distinct clinically relevant aspects of carotid atherosclerosis

AIMS: Patients with carotid atherosclerosis exhibit substantial heterogeneity in clinical presentation, plaque characteristics, and future cardiovascular risk, which do not always align. Clinical variables, plaque morphology, and molecular profiles are all used for patient stratification, yet it remains unclear which aspects of disease these data types capture and how they relate to clinically relevant outcomes. We therefore aimed to determine how clustering based on clinical, histological, and transcriptomic data relates to current cerebrovascular symptoms, plaque biology, and future cardiovascular risk. METHODS AND RESULTS: We analysed 1075 patients undergoing carotid endarterectomy with matched clinical, histological, and transcriptomic data. Unsupervised clustering was applied separately to each data modality, and resulting clusters were evaluated for overlap and association with current symptom status and 3-year major adverse cardiovascular events (MACE). Clusters derived from different modalities showed limited overlap, but distinct associations with clinical outcomes. Clinical clusters were most strongly associated with 3-year MACE, whereas histological and transcriptomic clusters were more closely related to current cerebrovascular symptoms. Transcriptomic clusters further delineated molecular pathways and plaque biology. Integration of modalities provided modest improvement in outcome stratification, indicating that these data types capture complementary rather than redundant information. CONCLUSION: Clinical, histological, and transcriptomic data capture distinct but complementary aspects of carotid atherosclerosis, reflecting systemic cardiovascular risk, local plaque phenotype, and underlying biology, with differing relevance to current cerebrovascular symptoms and future cardiovascular risk. The choice of data modality should therefore be guided by the intended clinical or biological question, rather than assuming a single approach can capture the full disease spectrum.

Authors

Institutions

Publication Details

Journal
Cardiovascular Research
Published
2026-11-20
DOI
https://doi.org/10.1093/cvr/cvag226
Primary Topic
Cerebrovascular and Carotid Artery Diseases
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Clinical, histological, and transcriptomic clustering capture distinct clinically relevant aspects of carotid atherosclerosis

Joost Hoekstra, Minna Unelma Kaikkonen, Siting Wei, Gerard Pasterkamp et al.
2 links
Cardiovascular Research
Cerebrovascular and Carotid Artery Diseases
article

Clinical, histological, and transcriptomic clustering capture distinct clinically relevant aspects of carotid atherosclerosis

Joost Hoekstra, Minna Unelma Kaikkonen, Siting Wei, Gerard Pasterkamp, Hester M. den Ruijter, Michal Mokrý, Dominique P V de Kleijn, Sander W van der Laan
article en

Abstract

AIMS: Patients with carotid atherosclerosis exhibit substantial heterogeneity in clinical presentation, plaque characteristics, and future cardiovascular risk, which do not always align. Clinical variables, plaque morphology, and molecular profiles are all used for patient stratification, yet it remains unclear which aspects of disease these data types capture and how they relate to clinically relevant outcomes. We therefore aimed to determine how clustering based on clinical, histological, and transcriptomic data relates to current cerebrovascular symptoms, plaque biology, and future cardiovascular risk. METHODS AND RESULTS: We analysed 1075 patients undergoing carotid endarterectomy with matched clinical, histological, and transcriptomic data. Unsupervised clustering was applied separately to each data modality, and resulting clusters were evaluated for overlap and association with current symptom status and 3-year major adverse cardiovascular events (MACE). Clusters derived from different modalities showed limited overlap, but distinct associations with clinical outcomes. Clinical clusters were most strongly associated with 3-year MACE, whereas histological and transcriptomic clusters were more closely related to current cerebrovascular symptoms. Transcriptomic clusters further delineated molecular pathways and plaque biology. Integration of modalities provided modest improvement in outcome stratification, indicating that these data types capture complementary rather than redundant information. CONCLUSION: Clinical, histological, and transcriptomic data capture distinct but complementary aspects of carotid atherosclerosis, reflecting systemic cardiovascular risk, local plaque phenotype, and underlying biology, with differing relevance to current cerebrovascular symptoms and future cardiovascular risk. The choice of data modality should therefore be guided by the intended clinical or biological question, rather than assuming a single approach can capture the full disease spectrum.

Cardiovascular Research
University of Eastern Finland (FI), Utrecht University (NL), University Medical Center Utrecht (NL), University of Virginia (US)
Chan Zuckerberg Initiative, Health~Holland, Dutch Cardiovascular Alliance, Fondation Leducq, Universitair Medisch Centrum Utrecht, Horizon 2020 Framework Programme
Good health and well-being
Cerebrovascular and Carotid Artery Diseases
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon