Precision Phenotyping of Dilated Cardiomyopathy Using Multidimensional Data

Upasana Tayal; Job A J Verdonschot; Mark R Hazebroek; James Howard; John Gregson; Simon Newsome; Ankur Gulati; Chee Jian Pua; Brian P Halliday; Amrit S Lota; Rachel J Buchan; Nicola Whiffin; Lina Kanapeckaite; Resham Baruah; Julian W E Jarman; Declan P O'Regan; Paul J R Barton; James S Ware; Dudley J Pennell; Bouke P Adriaans; Sebastiaan C A M Bekkers; Jackie Donovan; Michael Frenneaux; Leslie T Cooper; James L Januzzi; John G F Cleland; Stuart A Cook; Rahul C Deo; Stephane R B Heymans; Sanjay K Prasad

doi:10.1016/j.jacc.2022.03.375

Precision Phenotyping of Dilated Cardiomyopathy Using Multidimensional Data

Upasana Tayal^*, Job A J Verdonschot, Mark R Hazebroek, James Howard, John Gregson, Simon Newsome, Ankur Gulati, Chee Jian Pua, Brian P Halliday, Amrit S Lota, Rachel J Buchan, Nicola Whiffin, Lina Kanapeckaite, Resham Baruah, Julian W E Jarman, Declan P O'Regan, Paul J R Barton, James S Ware, Dudley J Pennell, Bouke P AdriaansSebastiaan C A M Bekkers, Jackie Donovan, Michael Frenneaux, Leslie T Cooper, James L Januzzi, John G F Cleland, Stuart A Cook, Rahul C Deo, Stephane R B Heymans, Sanjay K Prasad^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

BACKGROUND: Dilated cardiomyopathy (DCM) is a final common manifestation of heterogenous etiologies. Adverse outcomes highlight the need for disease stratification beyond ejection fraction.

OBJECTIVES: The purpose of this study was to identify novel, reproducible subphenotypes of DCM using multiparametric data for improved patient stratification.

METHODS: Longitudinal, observational UK-derivation (n = 426; median age 54 years; 67% men) and Dutch-validation (n = 239; median age 56 years; 64% men) cohorts of DCM patients (enrolled 2009-2016) with clinical, genetic, cardiovascular magnetic resonance, and proteomic assessments. Machine learning with profile regression identified novel disease subtypes. Penalized multinomial logistic regression was used for validation. Nested Cox models compared novel groupings to conventional risk measures. Primary composite outcome was cardiovascular death, heart failure, or arrhythmia events (median follow-up 4 years).

RESULTS: In total, 3 novel DCM subtypes were identified: profibrotic metabolic, mild nonfibrotic, and biventricular impairment. Prognosis differed between subtypes in both the derivation (P < 0.0001) and validation cohorts. The novel profibrotic metabolic subtype had more diabetes, universal myocardial fibrosis, preserved right ventricular function, and elevated creatinine. For clinical application, 5 variables were sufficient for classification (left and right ventricular end-systolic volumes, left atrial volume, myocardial fibrosis, and creatinine). Adding the novel DCM subtype improved the C-statistic from 0.60 to 0.76. Interleukin-4 receptor-alpha was identified as a novel prognostic biomarker in derivation (HR: 3.6; 95% CI: 1.9-6.5; P = 0.00002) and validation cohorts (HR: 1.94; 95% CI: 1.3-2.8; P = 0.00005).

CONCLUSIONS: Three reproducible, mechanistically distinct DCM subtypes were identified using widely available clinical and biological data, adding prognostic value to traditional risk models. They may improve patient selection for novel interventions, thereby enabling precision medicine.

Original language	English
Pages (from-to)	2219-2232
Number of pages	14
Journal	Journal of the American College of Cardiology
Volume	79
Issue number	22
DOIs	https://doi.org/10.1016/j.jacc.2022.03.375
Publication status	Published - 7 Jun 2022

Keywords

Cardiomyopathies
Cardiomyopathy, Dilated/diagnosis
Creatinine
Female
Fibrosis
Humans
Male
Middle Aged
Proteomics
Stroke Volume
proteomics
POSITION STATEMENT
machine learning
heart
STRATEGIES
ESC WORKING GROUP

Access to Document

10.1016/j.jacc.2022.03.375Licence: CC BY

Cite this

Tayal, U., Verdonschot, J. A. J., Hazebroek, M. R., Howard, J., Gregson, J., Newsome, S., Gulati, A., Pua, C. J., Halliday, B. P., Lota, A. S., Buchan, R. J., Whiffin, N., Kanapeckaite, L., Baruah, R., Jarman, J. W. E., O'Regan, D. P., Barton, P. J. R., Ware, J. S., Pennell, D. J., ... Prasad, S. K. (2022). Precision Phenotyping of Dilated Cardiomyopathy Using Multidimensional Data. Journal of the American College of Cardiology, 79(22), 2219-2232. https://doi.org/10.1016/j.jacc.2022.03.375

@article{7942837d7fe34118bb56a68348bd77fa,

title = "Precision Phenotyping of Dilated Cardiomyopathy Using Multidimensional Data",

abstract = "BACKGROUND: Dilated cardiomyopathy (DCM) is a final common manifestation of heterogenous etiologies. Adverse outcomes highlight the need for disease stratification beyond ejection fraction.OBJECTIVES: The purpose of this study was to identify novel, reproducible subphenotypes of DCM using multiparametric data for improved patient stratification.METHODS: Longitudinal, observational UK-derivation (n = 426; median age 54 years; 67% men) and Dutch-validation (n = 239; median age 56 years; 64% men) cohorts of DCM patients (enrolled 2009-2016) with clinical, genetic, cardiovascular magnetic resonance, and proteomic assessments. Machine learning with profile regression identified novel disease subtypes. Penalized multinomial logistic regression was used for validation. Nested Cox models compared novel groupings to conventional risk measures. Primary composite outcome was cardiovascular death, heart failure, or arrhythmia events (median follow-up 4 years).RESULTS: In total, 3 novel DCM subtypes were identified: profibrotic metabolic, mild nonfibrotic, and biventricular impairment. Prognosis differed between subtypes in both the derivation (P < 0.0001) and validation cohorts. The novel profibrotic metabolic subtype had more diabetes, universal myocardial fibrosis, preserved right ventricular function, and elevated creatinine. For clinical application, 5 variables were sufficient for classification (left and right ventricular end-systolic volumes, left atrial volume, myocardial fibrosis, and creatinine). Adding the novel DCM subtype improved the C-statistic from 0.60 to 0.76. Interleukin-4 receptor-alpha was identified as a novel prognostic biomarker in derivation (HR: 3.6; 95% CI: 1.9-6.5; P = 0.00002) and validation cohorts (HR: 1.94; 95% CI: 1.3-2.8; P = 0.00005).CONCLUSIONS: Three reproducible, mechanistically distinct DCM subtypes were identified using widely available clinical and biological data, adding prognostic value to traditional risk models. They may improve patient selection for novel interventions, thereby enabling precision medicine.",

keywords = "Cardiomyopathies, Cardiomyopathy, Dilated/diagnosis, Creatinine, Female, Fibrosis, Humans, Male, Middle Aged, Proteomics, Stroke Volume, proteomics, POSITION STATEMENT, machine learning, heart, STRATEGIES, ESC WORKING GROUP",

author = "Upasana Tayal and Verdonschot, {Job A J} and Hazebroek, {Mark R} and James Howard and John Gregson and Simon Newsome and Ankur Gulati and Pua, {Chee Jian} and Halliday, {Brian P} and Lota, {Amrit S} and Buchan, {Rachel J} and Nicola Whiffin and Lina Kanapeckaite and Resham Baruah and Jarman, {Julian W E} and O'Regan, {Declan P} and Barton, {Paul J R} and Ware, {James S} and Pennell, {Dudley J} and Adriaans, {Bouke P} and Bekkers, {Sebastiaan C A M} and Jackie Donovan and Michael Frenneaux and Cooper, {Leslie T} and Januzzi, {James L} and Cleland, {John G F} and Cook, {Stuart A} and Deo, {Rahul C} and Heymans, {Stephane R B} and Prasad, {Sanjay K}",

year = "2022",

month = jun,

day = "7",

doi = "10.1016/j.jacc.2022.03.375",

language = "English",

volume = "79",

pages = "2219--2232",

journal = "Journal of the American College of Cardiology",

issn = "0735-1097",

publisher = "Elsevier Science",

number = "22",

}

Tayal, U, Verdonschot, JAJ, Hazebroek, MR, Howard, J, Gregson, J, Newsome, S, Gulati, A, Pua, CJ, Halliday, BP, Lota, AS, Buchan, RJ, Whiffin, N, Kanapeckaite, L, Baruah, R, Jarman, JWE, O'Regan, DP, Barton, PJR, Ware, JS, Pennell, DJ, Adriaans, BP, Bekkers, SCAM, Donovan, J, Frenneaux, M, Cooper, LT, Januzzi, JL, Cleland, JGF, Cook, SA, Deo, RC, Heymans, SRB & Prasad, SK 2022, 'Precision Phenotyping of Dilated Cardiomyopathy Using Multidimensional Data', Journal of the American College of Cardiology, vol. 79, no. 22, pp. 2219-2232. https://doi.org/10.1016/j.jacc.2022.03.375

TY - JOUR

T1 - Precision Phenotyping of Dilated Cardiomyopathy Using Multidimensional Data

AU - Tayal, Upasana

AU - Verdonschot, Job A J

AU - Hazebroek, Mark R

AU - Howard, James

AU - Gregson, John

AU - Newsome, Simon

AU - Gulati, Ankur

AU - Pua, Chee Jian

AU - Halliday, Brian P

AU - Lota, Amrit S

AU - Buchan, Rachel J

AU - Whiffin, Nicola

AU - Kanapeckaite, Lina

AU - Baruah, Resham

AU - Jarman, Julian W E

AU - O'Regan, Declan P

AU - Barton, Paul J R

AU - Ware, James S

AU - Pennell, Dudley J

AU - Adriaans, Bouke P

AU - Bekkers, Sebastiaan C A M

AU - Donovan, Jackie

AU - Frenneaux, Michael

AU - Cooper, Leslie T

AU - Januzzi, James L

AU - Cleland, John G F

AU - Cook, Stuart A

AU - Deo, Rahul C

AU - Heymans, Stephane R B

AU - Prasad, Sanjay K

PY - 2022/6/7

Y1 - 2022/6/7

N2 - BACKGROUND: Dilated cardiomyopathy (DCM) is a final common manifestation of heterogenous etiologies. Adverse outcomes highlight the need for disease stratification beyond ejection fraction.OBJECTIVES: The purpose of this study was to identify novel, reproducible subphenotypes of DCM using multiparametric data for improved patient stratification.METHODS: Longitudinal, observational UK-derivation (n = 426; median age 54 years; 67% men) and Dutch-validation (n = 239; median age 56 years; 64% men) cohorts of DCM patients (enrolled 2009-2016) with clinical, genetic, cardiovascular magnetic resonance, and proteomic assessments. Machine learning with profile regression identified novel disease subtypes. Penalized multinomial logistic regression was used for validation. Nested Cox models compared novel groupings to conventional risk measures. Primary composite outcome was cardiovascular death, heart failure, or arrhythmia events (median follow-up 4 years).RESULTS: In total, 3 novel DCM subtypes were identified: profibrotic metabolic, mild nonfibrotic, and biventricular impairment. Prognosis differed between subtypes in both the derivation (P < 0.0001) and validation cohorts. The novel profibrotic metabolic subtype had more diabetes, universal myocardial fibrosis, preserved right ventricular function, and elevated creatinine. For clinical application, 5 variables were sufficient for classification (left and right ventricular end-systolic volumes, left atrial volume, myocardial fibrosis, and creatinine). Adding the novel DCM subtype improved the C-statistic from 0.60 to 0.76. Interleukin-4 receptor-alpha was identified as a novel prognostic biomarker in derivation (HR: 3.6; 95% CI: 1.9-6.5; P = 0.00002) and validation cohorts (HR: 1.94; 95% CI: 1.3-2.8; P = 0.00005).CONCLUSIONS: Three reproducible, mechanistically distinct DCM subtypes were identified using widely available clinical and biological data, adding prognostic value to traditional risk models. They may improve patient selection for novel interventions, thereby enabling precision medicine.

AB - BACKGROUND: Dilated cardiomyopathy (DCM) is a final common manifestation of heterogenous etiologies. Adverse outcomes highlight the need for disease stratification beyond ejection fraction.OBJECTIVES: The purpose of this study was to identify novel, reproducible subphenotypes of DCM using multiparametric data for improved patient stratification.METHODS: Longitudinal, observational UK-derivation (n = 426; median age 54 years; 67% men) and Dutch-validation (n = 239; median age 56 years; 64% men) cohorts of DCM patients (enrolled 2009-2016) with clinical, genetic, cardiovascular magnetic resonance, and proteomic assessments. Machine learning with profile regression identified novel disease subtypes. Penalized multinomial logistic regression was used for validation. Nested Cox models compared novel groupings to conventional risk measures. Primary composite outcome was cardiovascular death, heart failure, or arrhythmia events (median follow-up 4 years).RESULTS: In total, 3 novel DCM subtypes were identified: profibrotic metabolic, mild nonfibrotic, and biventricular impairment. Prognosis differed between subtypes in both the derivation (P < 0.0001) and validation cohorts. The novel profibrotic metabolic subtype had more diabetes, universal myocardial fibrosis, preserved right ventricular function, and elevated creatinine. For clinical application, 5 variables were sufficient for classification (left and right ventricular end-systolic volumes, left atrial volume, myocardial fibrosis, and creatinine). Adding the novel DCM subtype improved the C-statistic from 0.60 to 0.76. Interleukin-4 receptor-alpha was identified as a novel prognostic biomarker in derivation (HR: 3.6; 95% CI: 1.9-6.5; P = 0.00002) and validation cohorts (HR: 1.94; 95% CI: 1.3-2.8; P = 0.00005).CONCLUSIONS: Three reproducible, mechanistically distinct DCM subtypes were identified using widely available clinical and biological data, adding prognostic value to traditional risk models. They may improve patient selection for novel interventions, thereby enabling precision medicine.

KW - Cardiomyopathies

KW - Cardiomyopathy, Dilated/diagnosis

KW - Creatinine

KW - Female

KW - Fibrosis

KW - Humans

KW - Male

KW - Middle Aged

KW - Proteomics

KW - Stroke Volume

KW - proteomics

KW - POSITION STATEMENT

KW - machine learning

KW - heart

KW - STRATEGIES

KW - ESC WORKING GROUP

U2 - 10.1016/j.jacc.2022.03.375

DO - 10.1016/j.jacc.2022.03.375

M3 - Article

C2 - 35654493

SN - 0735-1097

VL - 79

SP - 2219

EP - 2232

JO - Journal of the American College of Cardiology

JF - Journal of the American College of Cardiology

IS - 22

ER -