Uncovering precision phenotype-biomarker associations in traumatic brain injury using topological data analysis

TRACK-TBI Investigators

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

Background: Traumatic brain injury (TBI) is a complex disorder that is traditionally stratified based on clinical signs and symptoms. Recent imaging and molecular biomarker innovations provide unprecedented opportunities for improved TBI precision medicine, incorporating patho-anatomical and molecular mechanisms. Complete integration of these diverse data for TBI diagnosis and patient stratification remains an unmet challenge. Methods and findings: The Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) Pilot multicenter study enrolled 586 acute TBI patients and collected diverse common data elements (TBI-CDEs) across the study population, including imaging, genetics, and clinical outcomes. We then applied topology-based data-driven discovery to identify natural subgroups of patients, based on the TBI-CDEs collected. Our hypothesis was two-fold: 1) A machine learning tool known as topological data analysis (TDA) would reveal data-driven patterns in patient outcomes to identify candidate biomarkers of recovery, and 2) TDA-identified biomarkers would significantly predict patient outcome recovery after TBI using more traditional methods of univariate statistical tests. TDA algorithms organized and mapped the data of TBI patients in multidimensional space, identifying a subset of mild TBI patients with a specific multivariate phenotype associated with unfavorable outcome at 3 and 6 months after injury. Further analyses revealed that this patient subset had high rates of post-traumatic stress disorder (PTSD), and enrichment in several distinct genetic polymorphisms associated with cellular responses to stress and DNA damage (PARP1), and in striatal dopamine processing (ANKK1, COMT, DRD2). Conclusions: TDA identified a unique diagnostic subgroup of patients with unfavorable outcome after mild TBI that were significantly predicted by the presence of specific genetic polymorphisms. Machine learning methods such as TDA may provide a robust method for patient stratification and treatment planning targeting identified biomarkers in future clinical trials in TBI patients.

Original languageEnglish (US)
Article numbere0169490
JournalPloS one
Volume12
Issue number3
DOIs
StatePublished - Mar 2017

Fingerprint

Biomarkers
biomarkers
Brain
data analysis
Phenotype
brain
phenotype
Brain Concussion
Genetic Polymorphisms
artificial intelligence
Polymorphism
Learning systems
Traumatic Brain Injury
genetic polymorphism
image analysis
Corpus Striatum
Precision Medicine
Molecular Imaging
Imaging techniques
Recovery

Cite this

Uncovering precision phenotype-biomarker associations in traumatic brain injury using topological data analysis. / TRACK-TBI Investigators.

In: PloS one, Vol. 12, No. 3, e0169490, 03.2017.

Research output: Contribution to journalArticle

@article{0dfea04e504c49d3a0c49162bc623294,
title = "Uncovering precision phenotype-biomarker associations in traumatic brain injury using topological data analysis",
abstract = "Background: Traumatic brain injury (TBI) is a complex disorder that is traditionally stratified based on clinical signs and symptoms. Recent imaging and molecular biomarker innovations provide unprecedented opportunities for improved TBI precision medicine, incorporating patho-anatomical and molecular mechanisms. Complete integration of these diverse data for TBI diagnosis and patient stratification remains an unmet challenge. Methods and findings: The Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) Pilot multicenter study enrolled 586 acute TBI patients and collected diverse common data elements (TBI-CDEs) across the study population, including imaging, genetics, and clinical outcomes. We then applied topology-based data-driven discovery to identify natural subgroups of patients, based on the TBI-CDEs collected. Our hypothesis was two-fold: 1) A machine learning tool known as topological data analysis (TDA) would reveal data-driven patterns in patient outcomes to identify candidate biomarkers of recovery, and 2) TDA-identified biomarkers would significantly predict patient outcome recovery after TBI using more traditional methods of univariate statistical tests. TDA algorithms organized and mapped the data of TBI patients in multidimensional space, identifying a subset of mild TBI patients with a specific multivariate phenotype associated with unfavorable outcome at 3 and 6 months after injury. Further analyses revealed that this patient subset had high rates of post-traumatic stress disorder (PTSD), and enrichment in several distinct genetic polymorphisms associated with cellular responses to stress and DNA damage (PARP1), and in striatal dopamine processing (ANKK1, COMT, DRD2). Conclusions: TDA identified a unique diagnostic subgroup of patients with unfavorable outcome after mild TBI that were significantly predicted by the presence of specific genetic polymorphisms. Machine learning methods such as TDA may provide a robust method for patient stratification and treatment planning targeting identified biomarkers in future clinical trials in TBI patients.",
author = "{TRACK-TBI Investigators} and Nielson, {Jessica L.} and Cooper, {Shelly R.} and Yue, {John K.} and Sorani, {Marco D.} and Tomoo Inoue and Yuh, {Esther L.} and Pratik Mukherjee and Petrossian, {Tanya C.} and Jesse Paquette and Lum, {Pek Y.} and Carlsson, {Gunnar E.} and Vassar, {Mary J.} and Lingsma, {Hester F.} and Gordon, {Wayne A.} and Valadka, {Alex B.} and Okonkwo, {David O.} and Manley, {Geoffrey T.} and Ferguson, {Adam R.} and Adeoye, {Opeolu M.} and Neeraj Badjatia and Boase, {Kimberly D.} and Yelena Bodien-Guller and Bullock, {Malcolm R.} and Chesnut, {Randall M.} and Corrigan, {John D.} and Crawford, {Karen L.} and Ramon Diaz-Arrastia and Dikmen, {Sureyya S.} and Duhaime, {Ann Christine} and Ellenbogen, {Richard G.} and Frank Ezekiel and Feeser, {Venkata R.} and Giacino, {Joseph T.} and Goldman, {Dana P.} and Luis Gonzales and Gopinath, {Shankar P.} and Gullapalli, {Rao P.} and Hemphill, {Jesse C.} and Hotz, {Gillian A.} and Kramer, {Joel H.} and Harvey Levin and Lindsell, {Christopher J.} and Joan Machamer and Christopher Madden and Markowitz, {Amy J.} and Alastair Martin and Mathern, {Bruce E.} and McAllister, {Thomas W.} and McCrea, {Michael A.} and Merchant, {Randall E.}",
year = "2017",
month = "3",
doi = "10.1371/journal.pone.0169490",
language = "English (US)",
volume = "12",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "3",

}

TY - JOUR

T1 - Uncovering precision phenotype-biomarker associations in traumatic brain injury using topological data analysis

AU - TRACK-TBI Investigators

AU - Nielson, Jessica L.

AU - Cooper, Shelly R.

AU - Yue, John K.

AU - Sorani, Marco D.

AU - Inoue, Tomoo

AU - Yuh, Esther L.

AU - Mukherjee, Pratik

AU - Petrossian, Tanya C.

AU - Paquette, Jesse

AU - Lum, Pek Y.

AU - Carlsson, Gunnar E.

AU - Vassar, Mary J.

AU - Lingsma, Hester F.

AU - Gordon, Wayne A.

AU - Valadka, Alex B.

AU - Okonkwo, David O.

AU - Manley, Geoffrey T.

AU - Ferguson, Adam R.

AU - Adeoye, Opeolu M.

AU - Badjatia, Neeraj

AU - Boase, Kimberly D.

AU - Bodien-Guller, Yelena

AU - Bullock, Malcolm R.

AU - Chesnut, Randall M.

AU - Corrigan, John D.

AU - Crawford, Karen L.

AU - Diaz-Arrastia, Ramon

AU - Dikmen, Sureyya S.

AU - Duhaime, Ann Christine

AU - Ellenbogen, Richard G.

AU - Ezekiel, Frank

AU - Feeser, Venkata R.

AU - Giacino, Joseph T.

AU - Goldman, Dana P.

AU - Gonzales, Luis

AU - Gopinath, Shankar P.

AU - Gullapalli, Rao P.

AU - Hemphill, Jesse C.

AU - Hotz, Gillian A.

AU - Kramer, Joel H.

AU - Levin, Harvey

AU - Lindsell, Christopher J.

AU - Machamer, Joan

AU - Madden, Christopher

AU - Markowitz, Amy J.

AU - Martin, Alastair

AU - Mathern, Bruce E.

AU - McAllister, Thomas W.

AU - McCrea, Michael A.

AU - Merchant, Randall E.

PY - 2017/3

Y1 - 2017/3

N2 - Background: Traumatic brain injury (TBI) is a complex disorder that is traditionally stratified based on clinical signs and symptoms. Recent imaging and molecular biomarker innovations provide unprecedented opportunities for improved TBI precision medicine, incorporating patho-anatomical and molecular mechanisms. Complete integration of these diverse data for TBI diagnosis and patient stratification remains an unmet challenge. Methods and findings: The Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) Pilot multicenter study enrolled 586 acute TBI patients and collected diverse common data elements (TBI-CDEs) across the study population, including imaging, genetics, and clinical outcomes. We then applied topology-based data-driven discovery to identify natural subgroups of patients, based on the TBI-CDEs collected. Our hypothesis was two-fold: 1) A machine learning tool known as topological data analysis (TDA) would reveal data-driven patterns in patient outcomes to identify candidate biomarkers of recovery, and 2) TDA-identified biomarkers would significantly predict patient outcome recovery after TBI using more traditional methods of univariate statistical tests. TDA algorithms organized and mapped the data of TBI patients in multidimensional space, identifying a subset of mild TBI patients with a specific multivariate phenotype associated with unfavorable outcome at 3 and 6 months after injury. Further analyses revealed that this patient subset had high rates of post-traumatic stress disorder (PTSD), and enrichment in several distinct genetic polymorphisms associated with cellular responses to stress and DNA damage (PARP1), and in striatal dopamine processing (ANKK1, COMT, DRD2). Conclusions: TDA identified a unique diagnostic subgroup of patients with unfavorable outcome after mild TBI that were significantly predicted by the presence of specific genetic polymorphisms. Machine learning methods such as TDA may provide a robust method for patient stratification and treatment planning targeting identified biomarkers in future clinical trials in TBI patients.

AB - Background: Traumatic brain injury (TBI) is a complex disorder that is traditionally stratified based on clinical signs and symptoms. Recent imaging and molecular biomarker innovations provide unprecedented opportunities for improved TBI precision medicine, incorporating patho-anatomical and molecular mechanisms. Complete integration of these diverse data for TBI diagnosis and patient stratification remains an unmet challenge. Methods and findings: The Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) Pilot multicenter study enrolled 586 acute TBI patients and collected diverse common data elements (TBI-CDEs) across the study population, including imaging, genetics, and clinical outcomes. We then applied topology-based data-driven discovery to identify natural subgroups of patients, based on the TBI-CDEs collected. Our hypothesis was two-fold: 1) A machine learning tool known as topological data analysis (TDA) would reveal data-driven patterns in patient outcomes to identify candidate biomarkers of recovery, and 2) TDA-identified biomarkers would significantly predict patient outcome recovery after TBI using more traditional methods of univariate statistical tests. TDA algorithms organized and mapped the data of TBI patients in multidimensional space, identifying a subset of mild TBI patients with a specific multivariate phenotype associated with unfavorable outcome at 3 and 6 months after injury. Further analyses revealed that this patient subset had high rates of post-traumatic stress disorder (PTSD), and enrichment in several distinct genetic polymorphisms associated with cellular responses to stress and DNA damage (PARP1), and in striatal dopamine processing (ANKK1, COMT, DRD2). Conclusions: TDA identified a unique diagnostic subgroup of patients with unfavorable outcome after mild TBI that were significantly predicted by the presence of specific genetic polymorphisms. Machine learning methods such as TDA may provide a robust method for patient stratification and treatment planning targeting identified biomarkers in future clinical trials in TBI patients.

UR - http://www.scopus.com/inward/record.url?scp=85014551008&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85014551008&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0169490

DO - 10.1371/journal.pone.0169490

M3 - Article

C2 - 28257413

AN - SCOPUS:85014551008

VL - 12

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 3

M1 - e0169490

ER -