Quantifying Uncertainty and Robustness in a Biomathematical Model-Based Patient-Specific Response Metric for Glioblastoma

Andrea Hawkins-Daarud, Sandra K. Johnston, Kristin Swanson

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

PURPOSE: Glioblastomas, lethal primary brain tumors, are known for their heterogeneity and invasiveness. A growing body of literature has been developed demonstrating the clinical relevance of a biomathematical model, the proliferation-invasion model, of glioblastoma growth. Of interest here is the development of a treatment response metric, days gained (DG). This metric is based on individual tumor kinetics estimated through segmented volumes of hyperintense regions on T1-weighted gadolinium-enhanced and T2-weighted magnetic resonance images. This metric was shown to be prognostic of time to progression. Furthermore, it was shown to be more prognostic of outcome than standard response metrics. Although promising, the original article did not account for uncertainty in the calculation of the DG metric, leaving the robustness of this cutoff in question. METHODS: We harnessed the Bayesian framework to consider the impact of two sources of uncertainty: (1) image acquisition and (2) interobserver error in image segmentation. We first used synthetic data to characterize what nonerror variants are influencing the final uncertainty in the DG metric. We then considered the original patient cohort to investigate clinical patterns of uncertainty and to determine how robust this metric is for predicting time to progression and overall survival. RESULTS: Our results indicate that the key clinical variants are the time between pretreatment images and the underlying tumor growth kinetics, matching our observations in the clinical cohort. Finally, we demonstrated that for this cohort, there was a continuous range of cutoffs between 94 and 105 for which the prediction of the time to progression was over 80% reliable. CONCLUSION: Although additional validation must be performed, this work represents a key step in ascertaining the clinical utility of this metric.

Original languageEnglish (US)
Pages (from-to)1-8
Number of pages8
JournalJCO clinical cancer informatics
Issue number3
DOIs
StatePublished - Feb 1 2019

Fingerprint

Glioblastoma
Uncertainty
Gadolinium
Growth
Brain Neoplasms
Neoplasms
Magnetic Resonance Spectroscopy
Survival
Therapeutics

Cite this

Quantifying Uncertainty and Robustness in a Biomathematical Model-Based Patient-Specific Response Metric for Glioblastoma. / Hawkins-Daarud, Andrea; Johnston, Sandra K.; Swanson, Kristin.

In: JCO clinical cancer informatics, No. 3, 01.02.2019, p. 1-8.

Research output: Contribution to journalArticle

@article{5578c95ac6404f37a6a52f225939bc0b,
title = "Quantifying Uncertainty and Robustness in a Biomathematical Model-Based Patient-Specific Response Metric for Glioblastoma",
abstract = "PURPOSE: Glioblastomas, lethal primary brain tumors, are known for their heterogeneity and invasiveness. A growing body of literature has been developed demonstrating the clinical relevance of a biomathematical model, the proliferation-invasion model, of glioblastoma growth. Of interest here is the development of a treatment response metric, days gained (DG). This metric is based on individual tumor kinetics estimated through segmented volumes of hyperintense regions on T1-weighted gadolinium-enhanced and T2-weighted magnetic resonance images. This metric was shown to be prognostic of time to progression. Furthermore, it was shown to be more prognostic of outcome than standard response metrics. Although promising, the original article did not account for uncertainty in the calculation of the DG metric, leaving the robustness of this cutoff in question. METHODS: We harnessed the Bayesian framework to consider the impact of two sources of uncertainty: (1) image acquisition and (2) interobserver error in image segmentation. We first used synthetic data to characterize what nonerror variants are influencing the final uncertainty in the DG metric. We then considered the original patient cohort to investigate clinical patterns of uncertainty and to determine how robust this metric is for predicting time to progression and overall survival. RESULTS: Our results indicate that the key clinical variants are the time between pretreatment images and the underlying tumor growth kinetics, matching our observations in the clinical cohort. Finally, we demonstrated that for this cohort, there was a continuous range of cutoffs between 94 and 105 for which the prediction of the time to progression was over 80{\%} reliable. CONCLUSION: Although additional validation must be performed, this work represents a key step in ascertaining the clinical utility of this metric.",
author = "Andrea Hawkins-Daarud and Johnston, {Sandra K.} and Kristin Swanson",
year = "2019",
month = "2",
day = "1",
doi = "10.1200/CCI.18.00066",
language = "English (US)",
pages = "1--8",
journal = "JCO clinical cancer informatics",
issn = "2473-4276",
publisher = "American Society of Clinical Oncology",
number = "3",

}

TY - JOUR

T1 - Quantifying Uncertainty and Robustness in a Biomathematical Model-Based Patient-Specific Response Metric for Glioblastoma

AU - Hawkins-Daarud, Andrea

AU - Johnston, Sandra K.

AU - Swanson, Kristin

PY - 2019/2/1

Y1 - 2019/2/1

N2 - PURPOSE: Glioblastomas, lethal primary brain tumors, are known for their heterogeneity and invasiveness. A growing body of literature has been developed demonstrating the clinical relevance of a biomathematical model, the proliferation-invasion model, of glioblastoma growth. Of interest here is the development of a treatment response metric, days gained (DG). This metric is based on individual tumor kinetics estimated through segmented volumes of hyperintense regions on T1-weighted gadolinium-enhanced and T2-weighted magnetic resonance images. This metric was shown to be prognostic of time to progression. Furthermore, it was shown to be more prognostic of outcome than standard response metrics. Although promising, the original article did not account for uncertainty in the calculation of the DG metric, leaving the robustness of this cutoff in question. METHODS: We harnessed the Bayesian framework to consider the impact of two sources of uncertainty: (1) image acquisition and (2) interobserver error in image segmentation. We first used synthetic data to characterize what nonerror variants are influencing the final uncertainty in the DG metric. We then considered the original patient cohort to investigate clinical patterns of uncertainty and to determine how robust this metric is for predicting time to progression and overall survival. RESULTS: Our results indicate that the key clinical variants are the time between pretreatment images and the underlying tumor growth kinetics, matching our observations in the clinical cohort. Finally, we demonstrated that for this cohort, there was a continuous range of cutoffs between 94 and 105 for which the prediction of the time to progression was over 80% reliable. CONCLUSION: Although additional validation must be performed, this work represents a key step in ascertaining the clinical utility of this metric.

AB - PURPOSE: Glioblastomas, lethal primary brain tumors, are known for their heterogeneity and invasiveness. A growing body of literature has been developed demonstrating the clinical relevance of a biomathematical model, the proliferation-invasion model, of glioblastoma growth. Of interest here is the development of a treatment response metric, days gained (DG). This metric is based on individual tumor kinetics estimated through segmented volumes of hyperintense regions on T1-weighted gadolinium-enhanced and T2-weighted magnetic resonance images. This metric was shown to be prognostic of time to progression. Furthermore, it was shown to be more prognostic of outcome than standard response metrics. Although promising, the original article did not account for uncertainty in the calculation of the DG metric, leaving the robustness of this cutoff in question. METHODS: We harnessed the Bayesian framework to consider the impact of two sources of uncertainty: (1) image acquisition and (2) interobserver error in image segmentation. We first used synthetic data to characterize what nonerror variants are influencing the final uncertainty in the DG metric. We then considered the original patient cohort to investigate clinical patterns of uncertainty and to determine how robust this metric is for predicting time to progression and overall survival. RESULTS: Our results indicate that the key clinical variants are the time between pretreatment images and the underlying tumor growth kinetics, matching our observations in the clinical cohort. Finally, we demonstrated that for this cohort, there was a continuous range of cutoffs between 94 and 105 for which the prediction of the time to progression was over 80% reliable. CONCLUSION: Although additional validation must be performed, this work represents a key step in ascertaining the clinical utility of this metric.

UR - http://www.scopus.com/inward/record.url?scp=85061571318&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061571318&partnerID=8YFLogxK

U2 - 10.1200/CCI.18.00066

DO - 10.1200/CCI.18.00066

M3 - Article

SP - 1

EP - 8

JO - JCO clinical cancer informatics

JF - JCO clinical cancer informatics

SN - 2473-4276

IS - 3

ER -