High-dimensional detection of imaging response to treatment in multiple sclerosis

Baris Kanber, Parashkev Nachev, Frederik Barkhof, Alberto Calvi, Jorge Cardoso, Rosa Cortese, Ferran Prados, Carole H Sudre, Carmen Tur, Sebastien Ourselin, Olga Ciccarelli

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Changes on brain imaging may precede clinical manifestations or disclose disease progression opaque to conventional clinical measures. Where, as in multiple sclerosis, the pathological process has a complex anatomical distribution, such changes are not easily detected by low-dimensional models in common use. This hinders our ability to detect treatment effects, both in the management of individual patients and in interventional trials. Here we compared the ability of conventional models to detect an imaging response to treatment against high-dimensional models incorporating a wide multiplicity of imaging factors. We used fully-automated image analysis to extract 144 regional, longitudinal trajectories of pre- and post- treatment changes in brain volume and disconnection in a cohort of 124 natalizumab-treated patients. Low- and high-dimensional models of the relationship between treatment and the trajectories of change were built and evaluated with machine learning, quantifying performance with receiver operating characteristic curves. Simulations of randomised controlled trials enrolling varying numbers of patients were used to quantify the impact of dimensionality on statistical efficiency. Compared to existing methods, high-dimensional models were superior in treatment response detection (area under the receiver operating characteristic curve = 0.890 [95% CI = 0.885-0.895] vs. 0.686 [95% CI = 0.679-0.693], P < 0.01]) and in statistical efficiency (achieved statistical power = 0.806 [95% CI = 0.698-0.872] vs. 0.508 [95% CI = 0.403-0.593] with number of patients enrolled = 50, at α = 0.01). High-dimensional models based on routine, clinical imaging can substantially enhance the detection of the imaging response to treatment in multiple sclerosis, potentially enabling more accurate individual prediction and greater statistical efficiency of randomised controlled trials.

Original languageEnglish
Pages (from-to)49
JournalNPJ digital medicine
Volume2
DOIs
Publication statusPublished - 2019

Cite this

Kanber, Baris ; Nachev, Parashkev ; Barkhof, Frederik ; Calvi, Alberto ; Cardoso, Jorge ; Cortese, Rosa ; Prados, Ferran ; Sudre, Carole H ; Tur, Carmen ; Ourselin, Sebastien ; Ciccarelli, Olga. / High-dimensional detection of imaging response to treatment in multiple sclerosis. In: NPJ digital medicine. 2019 ; Vol. 2. pp. 49.
@article{315328818dc64cb4a0883af56a939399,
title = "High-dimensional detection of imaging response to treatment in multiple sclerosis",
abstract = "Changes on brain imaging may precede clinical manifestations or disclose disease progression opaque to conventional clinical measures. Where, as in multiple sclerosis, the pathological process has a complex anatomical distribution, such changes are not easily detected by low-dimensional models in common use. This hinders our ability to detect treatment effects, both in the management of individual patients and in interventional trials. Here we compared the ability of conventional models to detect an imaging response to treatment against high-dimensional models incorporating a wide multiplicity of imaging factors. We used fully-automated image analysis to extract 144 regional, longitudinal trajectories of pre- and post- treatment changes in brain volume and disconnection in a cohort of 124 natalizumab-treated patients. Low- and high-dimensional models of the relationship between treatment and the trajectories of change were built and evaluated with machine learning, quantifying performance with receiver operating characteristic curves. Simulations of randomised controlled trials enrolling varying numbers of patients were used to quantify the impact of dimensionality on statistical efficiency. Compared to existing methods, high-dimensional models were superior in treatment response detection (area under the receiver operating characteristic curve = 0.890 [95{\%} CI = 0.885-0.895] vs. 0.686 [95{\%} CI = 0.679-0.693], P < 0.01]) and in statistical efficiency (achieved statistical power = 0.806 [95{\%} CI = 0.698-0.872] vs. 0.508 [95{\%} CI = 0.403-0.593] with number of patients enrolled = 50, at α = 0.01). High-dimensional models based on routine, clinical imaging can substantially enhance the detection of the imaging response to treatment in multiple sclerosis, potentially enabling more accurate individual prediction and greater statistical efficiency of randomised controlled trials.",
author = "Baris Kanber and Parashkev Nachev and Frederik Barkhof and Alberto Calvi and Jorge Cardoso and Rosa Cortese and Ferran Prados and Sudre, {Carole H} and Carmen Tur and Sebastien Ourselin and Olga Ciccarelli",
year = "2019",
doi = "10.1038/s41746-019-0127-8",
language = "English",
volume = "2",
pages = "49",
journal = "NPJ digital medicine",
issn = "2398-6352",

}

Kanber, B, Nachev, P, Barkhof, F, Calvi, A, Cardoso, J, Cortese, R, Prados, F, Sudre, CH, Tur, C, Ourselin, S & Ciccarelli, O 2019, 'High-dimensional detection of imaging response to treatment in multiple sclerosis' NPJ digital medicine, vol. 2, pp. 49. https://doi.org/10.1038/s41746-019-0127-8

High-dimensional detection of imaging response to treatment in multiple sclerosis. / Kanber, Baris; Nachev, Parashkev; Barkhof, Frederik; Calvi, Alberto; Cardoso, Jorge; Cortese, Rosa; Prados, Ferran; Sudre, Carole H; Tur, Carmen; Ourselin, Sebastien; Ciccarelli, Olga.

In: NPJ digital medicine, Vol. 2, 2019, p. 49.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - High-dimensional detection of imaging response to treatment in multiple sclerosis

AU - Kanber, Baris

AU - Nachev, Parashkev

AU - Barkhof, Frederik

AU - Calvi, Alberto

AU - Cardoso, Jorge

AU - Cortese, Rosa

AU - Prados, Ferran

AU - Sudre, Carole H

AU - Tur, Carmen

AU - Ourselin, Sebastien

AU - Ciccarelli, Olga

PY - 2019

Y1 - 2019

N2 - Changes on brain imaging may precede clinical manifestations or disclose disease progression opaque to conventional clinical measures. Where, as in multiple sclerosis, the pathological process has a complex anatomical distribution, such changes are not easily detected by low-dimensional models in common use. This hinders our ability to detect treatment effects, both in the management of individual patients and in interventional trials. Here we compared the ability of conventional models to detect an imaging response to treatment against high-dimensional models incorporating a wide multiplicity of imaging factors. We used fully-automated image analysis to extract 144 regional, longitudinal trajectories of pre- and post- treatment changes in brain volume and disconnection in a cohort of 124 natalizumab-treated patients. Low- and high-dimensional models of the relationship between treatment and the trajectories of change were built and evaluated with machine learning, quantifying performance with receiver operating characteristic curves. Simulations of randomised controlled trials enrolling varying numbers of patients were used to quantify the impact of dimensionality on statistical efficiency. Compared to existing methods, high-dimensional models were superior in treatment response detection (area under the receiver operating characteristic curve = 0.890 [95% CI = 0.885-0.895] vs. 0.686 [95% CI = 0.679-0.693], P < 0.01]) and in statistical efficiency (achieved statistical power = 0.806 [95% CI = 0.698-0.872] vs. 0.508 [95% CI = 0.403-0.593] with number of patients enrolled = 50, at α = 0.01). High-dimensional models based on routine, clinical imaging can substantially enhance the detection of the imaging response to treatment in multiple sclerosis, potentially enabling more accurate individual prediction and greater statistical efficiency of randomised controlled trials.

AB - Changes on brain imaging may precede clinical manifestations or disclose disease progression opaque to conventional clinical measures. Where, as in multiple sclerosis, the pathological process has a complex anatomical distribution, such changes are not easily detected by low-dimensional models in common use. This hinders our ability to detect treatment effects, both in the management of individual patients and in interventional trials. Here we compared the ability of conventional models to detect an imaging response to treatment against high-dimensional models incorporating a wide multiplicity of imaging factors. We used fully-automated image analysis to extract 144 regional, longitudinal trajectories of pre- and post- treatment changes in brain volume and disconnection in a cohort of 124 natalizumab-treated patients. Low- and high-dimensional models of the relationship between treatment and the trajectories of change were built and evaluated with machine learning, quantifying performance with receiver operating characteristic curves. Simulations of randomised controlled trials enrolling varying numbers of patients were used to quantify the impact of dimensionality on statistical efficiency. Compared to existing methods, high-dimensional models were superior in treatment response detection (area under the receiver operating characteristic curve = 0.890 [95% CI = 0.885-0.895] vs. 0.686 [95% CI = 0.679-0.693], P < 0.01]) and in statistical efficiency (achieved statistical power = 0.806 [95% CI = 0.698-0.872] vs. 0.508 [95% CI = 0.403-0.593] with number of patients enrolled = 50, at α = 0.01). High-dimensional models based on routine, clinical imaging can substantially enhance the detection of the imaging response to treatment in multiple sclerosis, potentially enabling more accurate individual prediction and greater statistical efficiency of randomised controlled trials.

U2 - 10.1038/s41746-019-0127-8

DO - 10.1038/s41746-019-0127-8

M3 - Article

VL - 2

SP - 49

JO - NPJ digital medicine

JF - NPJ digital medicine

SN - 2398-6352

ER -