Predictive value of targeted proteomics for coronary plaque morphology in patients with suspected coronary artery disease

Michiel J Bom, Evgeni Levin, Roel S Driessen, Ibrahim Danad, Cornelis C Van Kuijk, Albert C van Rossum, Jagat Narula, James K Min, Jonathon A Leipsic, João P Belo Pereira, Charles A Taylor, Max Nieuwdorp, Pieter G Raijmakers, Wolfgang Koenig, Albert K Groen, Erik S G Stroes, Paul Knaapen

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

BACKGROUND: Risk stratification is crucial to improve tailored therapy in patients with suspected coronary artery disease (CAD). This study investigated the ability of targeted proteomics to predict presence of high-risk plaque or absence of coronary atherosclerosis in patients with suspected CAD, defined by coronary computed tomography angiography (CCTA).

METHODS: Patients with suspected CAD (n = 203) underwent CCTA. Plasma levels of 358 proteins were used to generate machine learning models for the presence of CCTA-defined high-risk plaques or complete absence of coronary atherosclerosis. Performance was tested against a clinical model containing generally available clinical characteristics and conventional biomarkers.

FINDINGS: A total of 196 patients with analyzable protein levels (n = 332) was included for analysis. A subset of 35 proteins was identified predicting the presence of high-risk plaques. The developed machine learning model had fair diagnostic performance with an area under the curve (AUC) of 0·79 ± 0·01, outperforming prediction with generally available clinical characteristics (AUC = 0·65 ± 0·04, p < 0·05). Conversely, a different subset of 34 proteins was predictive for the absence of CAD (AUC = 0·85 ± 0·05), again outperforming prediction with generally available characteristics (AUC = 0·70 ± 0·04, p < 0·05).

INTERPRETATION: Using machine learning models, trained on targeted proteomics, we defined two complementary protein signatures: one for identification of patients with high-risk plaques and one for identification of patients with absence of CAD. Both biomarker subsets were superior to generally available clinical characteristics and conventional biomarkers in predicting presence of high-risk plaque or absence of coronary atherosclerosis. These promising findings warrant external validation of the value of targeted proteomics to identify cardiovascular risk in outcome studies. FUND: This study was supported by an unrestricted research grant from HeartFlow Inc. and partly supported by a European Research Area Network on Cardiovascular Diseases (ERA-CVD) grant (ERA CVD JTC2017, OPERATION). Funders had no influence on trial design, data evaluation, and interpretation.

Original languageEnglish
Pages (from-to)109-117
JournalEBioMedicine
Volume39
DOIs
Publication statusPublished - Jan 2019

Cite this

Bom, Michiel J ; Levin, Evgeni ; Driessen, Roel S ; Danad, Ibrahim ; Van Kuijk, Cornelis C ; van Rossum, Albert C ; Narula, Jagat ; Min, James K ; Leipsic, Jonathon A ; Belo Pereira, João P ; Taylor, Charles A ; Nieuwdorp, Max ; Raijmakers, Pieter G ; Koenig, Wolfgang ; Groen, Albert K ; Stroes, Erik S G ; Knaapen, Paul. / Predictive value of targeted proteomics for coronary plaque morphology in patients with suspected coronary artery disease. In: EBioMedicine. 2019 ; Vol. 39. pp. 109-117.
@article{ae211d96e0c64bc5ac5e3f7a941364ab,
title = "Predictive value of targeted proteomics for coronary plaque morphology in patients with suspected coronary artery disease",
abstract = "BACKGROUND: Risk stratification is crucial to improve tailored therapy in patients with suspected coronary artery disease (CAD). This study investigated the ability of targeted proteomics to predict presence of high-risk plaque or absence of coronary atherosclerosis in patients with suspected CAD, defined by coronary computed tomography angiography (CCTA).METHODS: Patients with suspected CAD (n = 203) underwent CCTA. Plasma levels of 358 proteins were used to generate machine learning models for the presence of CCTA-defined high-risk plaques or complete absence of coronary atherosclerosis. Performance was tested against a clinical model containing generally available clinical characteristics and conventional biomarkers.FINDINGS: A total of 196 patients with analyzable protein levels (n = 332) was included for analysis. A subset of 35 proteins was identified predicting the presence of high-risk plaques. The developed machine learning model had fair diagnostic performance with an area under the curve (AUC) of 0·79 ± 0·01, outperforming prediction with generally available clinical characteristics (AUC = 0·65 ± 0·04, p < 0·05). Conversely, a different subset of 34 proteins was predictive for the absence of CAD (AUC = 0·85 ± 0·05), again outperforming prediction with generally available characteristics (AUC = 0·70 ± 0·04, p < 0·05).INTERPRETATION: Using machine learning models, trained on targeted proteomics, we defined two complementary protein signatures: one for identification of patients with high-risk plaques and one for identification of patients with absence of CAD. Both biomarker subsets were superior to generally available clinical characteristics and conventional biomarkers in predicting presence of high-risk plaque or absence of coronary atherosclerosis. These promising findings warrant external validation of the value of targeted proteomics to identify cardiovascular risk in outcome studies. FUND: This study was supported by an unrestricted research grant from HeartFlow Inc. and partly supported by a European Research Area Network on Cardiovascular Diseases (ERA-CVD) grant (ERA CVD JTC2017, OPERATION). Funders had no influence on trial design, data evaluation, and interpretation.",
author = "Bom, {Michiel J} and Evgeni Levin and Driessen, {Roel S} and Ibrahim Danad and {Van Kuijk}, {Cornelis C} and {van Rossum}, {Albert C} and Jagat Narula and Min, {James K} and Leipsic, {Jonathon A} and {Belo Pereira}, {Jo{\~a}o P} and Taylor, {Charles A} and Max Nieuwdorp and Raijmakers, {Pieter G} and Wolfgang Koenig and Groen, {Albert K} and Stroes, {Erik S G} and Paul Knaapen",
note = "Copyright {\circledC} 2018 The Authors. Published by Elsevier B.V. All rights reserved.",
year = "2019",
month = "1",
doi = "10.1016/j.ebiom.2018.12.033",
language = "English",
volume = "39",
pages = "109--117",
journal = "EBioMedicine",
issn = "2352-3964",
publisher = "Elsevier BV",

}

Predictive value of targeted proteomics for coronary plaque morphology in patients with suspected coronary artery disease. / Bom, Michiel J; Levin, Evgeni; Driessen, Roel S; Danad, Ibrahim; Van Kuijk, Cornelis C; van Rossum, Albert C; Narula, Jagat; Min, James K; Leipsic, Jonathon A; Belo Pereira, João P; Taylor, Charles A; Nieuwdorp, Max; Raijmakers, Pieter G; Koenig, Wolfgang; Groen, Albert K; Stroes, Erik S G; Knaapen, Paul.

In: EBioMedicine, Vol. 39, 01.2019, p. 109-117.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Predictive value of targeted proteomics for coronary plaque morphology in patients with suspected coronary artery disease

AU - Bom, Michiel J

AU - Levin, Evgeni

AU - Driessen, Roel S

AU - Danad, Ibrahim

AU - Van Kuijk, Cornelis C

AU - van Rossum, Albert C

AU - Narula, Jagat

AU - Min, James K

AU - Leipsic, Jonathon A

AU - Belo Pereira, João P

AU - Taylor, Charles A

AU - Nieuwdorp, Max

AU - Raijmakers, Pieter G

AU - Koenig, Wolfgang

AU - Groen, Albert K

AU - Stroes, Erik S G

AU - Knaapen, Paul

N1 - Copyright © 2018 The Authors. Published by Elsevier B.V. All rights reserved.

PY - 2019/1

Y1 - 2019/1

N2 - BACKGROUND: Risk stratification is crucial to improve tailored therapy in patients with suspected coronary artery disease (CAD). This study investigated the ability of targeted proteomics to predict presence of high-risk plaque or absence of coronary atherosclerosis in patients with suspected CAD, defined by coronary computed tomography angiography (CCTA).METHODS: Patients with suspected CAD (n = 203) underwent CCTA. Plasma levels of 358 proteins were used to generate machine learning models for the presence of CCTA-defined high-risk plaques or complete absence of coronary atherosclerosis. Performance was tested against a clinical model containing generally available clinical characteristics and conventional biomarkers.FINDINGS: A total of 196 patients with analyzable protein levels (n = 332) was included for analysis. A subset of 35 proteins was identified predicting the presence of high-risk plaques. The developed machine learning model had fair diagnostic performance with an area under the curve (AUC) of 0·79 ± 0·01, outperforming prediction with generally available clinical characteristics (AUC = 0·65 ± 0·04, p < 0·05). Conversely, a different subset of 34 proteins was predictive for the absence of CAD (AUC = 0·85 ± 0·05), again outperforming prediction with generally available characteristics (AUC = 0·70 ± 0·04, p < 0·05).INTERPRETATION: Using machine learning models, trained on targeted proteomics, we defined two complementary protein signatures: one for identification of patients with high-risk plaques and one for identification of patients with absence of CAD. Both biomarker subsets were superior to generally available clinical characteristics and conventional biomarkers in predicting presence of high-risk plaque or absence of coronary atherosclerosis. These promising findings warrant external validation of the value of targeted proteomics to identify cardiovascular risk in outcome studies. FUND: This study was supported by an unrestricted research grant from HeartFlow Inc. and partly supported by a European Research Area Network on Cardiovascular Diseases (ERA-CVD) grant (ERA CVD JTC2017, OPERATION). Funders had no influence on trial design, data evaluation, and interpretation.

AB - BACKGROUND: Risk stratification is crucial to improve tailored therapy in patients with suspected coronary artery disease (CAD). This study investigated the ability of targeted proteomics to predict presence of high-risk plaque or absence of coronary atherosclerosis in patients with suspected CAD, defined by coronary computed tomography angiography (CCTA).METHODS: Patients with suspected CAD (n = 203) underwent CCTA. Plasma levels of 358 proteins were used to generate machine learning models for the presence of CCTA-defined high-risk plaques or complete absence of coronary atherosclerosis. Performance was tested against a clinical model containing generally available clinical characteristics and conventional biomarkers.FINDINGS: A total of 196 patients with analyzable protein levels (n = 332) was included for analysis. A subset of 35 proteins was identified predicting the presence of high-risk plaques. The developed machine learning model had fair diagnostic performance with an area under the curve (AUC) of 0·79 ± 0·01, outperforming prediction with generally available clinical characteristics (AUC = 0·65 ± 0·04, p < 0·05). Conversely, a different subset of 34 proteins was predictive for the absence of CAD (AUC = 0·85 ± 0·05), again outperforming prediction with generally available characteristics (AUC = 0·70 ± 0·04, p < 0·05).INTERPRETATION: Using machine learning models, trained on targeted proteomics, we defined two complementary protein signatures: one for identification of patients with high-risk plaques and one for identification of patients with absence of CAD. Both biomarker subsets were superior to generally available clinical characteristics and conventional biomarkers in predicting presence of high-risk plaque or absence of coronary atherosclerosis. These promising findings warrant external validation of the value of targeted proteomics to identify cardiovascular risk in outcome studies. FUND: This study was supported by an unrestricted research grant from HeartFlow Inc. and partly supported by a European Research Area Network on Cardiovascular Diseases (ERA-CVD) grant (ERA CVD JTC2017, OPERATION). Funders had no influence on trial design, data evaluation, and interpretation.

U2 - 10.1016/j.ebiom.2018.12.033

DO - 10.1016/j.ebiom.2018.12.033

M3 - Article

VL - 39

SP - 109

EP - 117

JO - EBioMedicine

JF - EBioMedicine

SN - 2352-3964

ER -