Use of radiomics for the prediction of local control of brain metastases after stereotactic radiosurgery.
brain metastases
local control
neuro-oncology
radiomics
response prediction
Journal
Neuro-oncology
ISSN: 1523-5866
Titre abrégé: Neuro Oncol
Pays: England
ID NLM: 100887420
Informations de publication
Date de publication:
09 06 2020
09 06 2020
Historique:
pubmed:
21
1
2020
medline:
28
4
2021
entrez:
21
1
2020
Statut:
ppublish
Résumé
Local response prediction for brain metastases (BM) after stereotactic radiosurgery (SRS) is challenging, particularly for smaller BM, as existing criteria are based solely on unidimensional measurements. This investigation sought to determine whether radiomic features provide additional value to routinely available clinical and dosimetric variables to predict local recurrence following SRS. Analyzed were 408 BM in 87 patients treated with SRS. A total of 440 radiomic features were extracted from the tumor core and the peritumoral regions, using the baseline pretreatment volumetric post-contrast T1 (T1c) and volumetric T2 fluid-attenuated inversion recovery (FLAIR) MRI sequences. Local tumor progression was determined based on Response Assessment in Neuro-Oncology‒BM criteria, with a maximum axial diameter growth of >20% on the follow-up T1c indicating local failure. The top radiomic features were determined based on resampled random forest (RF) feature importance. An RF classifier was trained using each set of features and evaluated using the area under the receiver operating characteristic curve (AUC). The addition of any one of the top 10 radiomic features to the set of clinical features resulted in a statistically significant (P < 0.001) increase in the AUC. An optimized combination of radiomic and clinical features resulted in a 19% higher resampled AUC (mean = 0.793; 95% CI = 0.792-0.795) than clinical features alone (0.669, 0.668-0.671). The increase in AUC of the RF classifier, after incorporating radiomic features, suggests that quantitative characterization of tumor appearance on pretreatment T1c and FLAIR adds value to known clinical and dosimetric variables for predicting local failure.
Sections du résumé
BACKGROUND
Local response prediction for brain metastases (BM) after stereotactic radiosurgery (SRS) is challenging, particularly for smaller BM, as existing criteria are based solely on unidimensional measurements. This investigation sought to determine whether radiomic features provide additional value to routinely available clinical and dosimetric variables to predict local recurrence following SRS.
METHODS
Analyzed were 408 BM in 87 patients treated with SRS. A total of 440 radiomic features were extracted from the tumor core and the peritumoral regions, using the baseline pretreatment volumetric post-contrast T1 (T1c) and volumetric T2 fluid-attenuated inversion recovery (FLAIR) MRI sequences. Local tumor progression was determined based on Response Assessment in Neuro-Oncology‒BM criteria, with a maximum axial diameter growth of >20% on the follow-up T1c indicating local failure. The top radiomic features were determined based on resampled random forest (RF) feature importance. An RF classifier was trained using each set of features and evaluated using the area under the receiver operating characteristic curve (AUC).
RESULTS
The addition of any one of the top 10 radiomic features to the set of clinical features resulted in a statistically significant (P < 0.001) increase in the AUC. An optimized combination of radiomic and clinical features resulted in a 19% higher resampled AUC (mean = 0.793; 95% CI = 0.792-0.795) than clinical features alone (0.669, 0.668-0.671).
CONCLUSIONS
The increase in AUC of the RF classifier, after incorporating radiomic features, suggests that quantitative characterization of tumor appearance on pretreatment T1c and FLAIR adds value to known clinical and dosimetric variables for predicting local failure.
Identifiants
pubmed: 31956919
pii: 5709657
doi: 10.1093/neuonc/noaa007
pmc: PMC7283017
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
797-805Commentaires et corrections
Type : CommentIn
Informations de copyright
© The Author(s) 2020. Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Références
Cancer Discov. 2015 Nov;5(11):1164-1177
pubmed: 26410082
Radiology. 1983 Sep;148(3):839-43
pubmed: 6878708
J Neurosurg. 2002 Dec;97(6):1276-81
pubmed: 12507123
Invest Radiol. 2010 Jan;45(1):49-56
pubmed: 19996757
J Craniomaxillofac Surg. 2019 Mar;47(3):510-515
pubmed: 30642733
Magn Reson Med. 2017 Sep;78(3):1110-1120
pubmed: 27690156
IEEE Trans Med Imaging. 2010 Jun;29(6):1310-20
pubmed: 20378467
Cancer. 2006 Jun 15;106(12):2672-81
pubmed: 16700040
J Neurooncol. 2017 Sep;134(2):433-441
pubmed: 28674974
Front Neuroinform. 2013 Dec 30;7:45
pubmed: 24416015
Neurosurgery. 2005 May;56(5):936-45; discussion 936-45
pubmed: 15854241
IEEE Trans Med Imaging. 2010 Jan;29(1):196-205
pubmed: 19923044
Acta Oncol. 2012 May;51(5):629-35
pubmed: 22537310
Radiother Oncol. 2003 Jul;68(1):15-21
pubmed: 12885447
J Neurosurg. 2015 Aug;123(2):373-86
pubmed: 25978710
Neurosurgery. 2018 Sep 1;83(3):345-353
pubmed: 29126142
Sci Rep. 2018 Feb 6;8(1):2475
pubmed: 29410469
Radiology. 2016 Feb;278(2):563-77
pubmed: 26579733
BMC Bioinformatics. 2011 Mar 17;12:77
pubmed: 21414208
Cancer Res. 2017 Nov 1;77(21):e104-e107
pubmed: 29092951
Lancet Oncol. 2015 Jun;16(6):e270-8
pubmed: 26065612
Neuro Oncol. 2017 Apr 1;19(suppl_2):ii2-ii15
pubmed: 28380635
AJNR Am J Neuroradiol. 2011 Nov-Dec;32(10):1885-92
pubmed: 21920854
J Thorac Oncol. 2017 Mar;12(3):467-476
pubmed: 27903462