Radiomics-based prediction of microsatellite instability in colorectal cancer at initial computed tomography evaluation

Jennifer S. Golia Pernicka, Johan Gagniere, Jayasree Chakraborty, Rikiya Yamashita, Lorenzo Nardo, John M. Creasy, Iva Petkovska, Richard R.K. Do, David D.B. Bates, Viktoriya Paroder, Mithat Gonen, Martin R. Weiser, Amber L. Simpson, Marc J. Gollub

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Purpose: To predict microsatellite instability (MSI) status of colon cancer on preoperative CT imaging using radiomic analysis. Methods: This retrospective study involved radiomic analysis of preoperative CT imaging of patients who underwent resection of stage II–III colon cancer from 2004 to 2012. A radiologist blinded to MSI status manually segmented the tumor region on CT images. 254 Intensity-based radiomic features were extracted from the tumor region. Three prediction models were developed with (1) only clinical features, (2) only radiomic features, and (3) “combined” clinical and radiomic features. Patients were randomly separated into training (n = 139) and test (n = 59) sets. The model was constructed from training data only; the test set was reserved for validation only. Model performance was evaluated using AUC, sensitivity, specificity, PPV, and NPV. Results: Of the total 198 patients, 134 (68%) patients had microsatellite stable tumors and 64 (32%) patients had MSI tumors. The combined model performed slightly better than the other models, predicting MSI with an AUC of 0.80 for the training set and 0.79 for the test set (specificity = 96.8% and 92.5%, respectively), whereas the model with only clinical features achieved an AUC of 0.74 and the model with only radiomic features achieved an AUC of 0.76. The model with clinical features alone had the lowest specificity (70%) compared with the model with radiomic features alone (95%) and the combined model (92.5%). Conclusions: Preoperative prediction of MSI status via radiomic analysis of preoperative CT adds specificity to clinical assessment and could contribute to personalized treatment selection.

Original languageEnglish (US)
JournalAbdominal Radiology
DOIs
StatePublished - Jan 1 2019

Fingerprint

Microsatellite Instability
Colorectal Neoplasms
Tomography
Area Under Curve
Colonic Neoplasms
Neoplasms
Microsatellite Repeats
Retrospective Studies
Sensitivity and Specificity

Keywords

  • Colon
  • Colonic neoplasms
  • Immunotherapy
  • Microsatellite instability
  • Microsatellite repeats

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Gastroenterology
  • Urology

Cite this

Radiomics-based prediction of microsatellite instability in colorectal cancer at initial computed tomography evaluation. / Golia Pernicka, Jennifer S.; Gagniere, Johan; Chakraborty, Jayasree; Yamashita, Rikiya; Nardo, Lorenzo; Creasy, John M.; Petkovska, Iva; Do, Richard R.K.; Bates, David D.B.; Paroder, Viktoriya; Gonen, Mithat; Weiser, Martin R.; Simpson, Amber L.; Gollub, Marc J.

In: Abdominal Radiology, 01.01.2019.

Research output: Contribution to journalArticle

Golia Pernicka, JS, Gagniere, J, Chakraborty, J, Yamashita, R, Nardo, L, Creasy, JM, Petkovska, I, Do, RRK, Bates, DDB, Paroder, V, Gonen, M, Weiser, MR, Simpson, AL & Gollub, MJ 2019, 'Radiomics-based prediction of microsatellite instability in colorectal cancer at initial computed tomography evaluation', Abdominal Radiology. https://doi.org/10.1007/s00261-019-02117-w
Golia Pernicka, Jennifer S. ; Gagniere, Johan ; Chakraborty, Jayasree ; Yamashita, Rikiya ; Nardo, Lorenzo ; Creasy, John M. ; Petkovska, Iva ; Do, Richard R.K. ; Bates, David D.B. ; Paroder, Viktoriya ; Gonen, Mithat ; Weiser, Martin R. ; Simpson, Amber L. ; Gollub, Marc J. / Radiomics-based prediction of microsatellite instability in colorectal cancer at initial computed tomography evaluation. In: Abdominal Radiology. 2019.
@article{a0dff623eced4ffcbfacf09ecb778c20,
title = "Radiomics-based prediction of microsatellite instability in colorectal cancer at initial computed tomography evaluation",
abstract = "Purpose: To predict microsatellite instability (MSI) status of colon cancer on preoperative CT imaging using radiomic analysis. Methods: This retrospective study involved radiomic analysis of preoperative CT imaging of patients who underwent resection of stage II–III colon cancer from 2004 to 2012. A radiologist blinded to MSI status manually segmented the tumor region on CT images. 254 Intensity-based radiomic features were extracted from the tumor region. Three prediction models were developed with (1) only clinical features, (2) only radiomic features, and (3) “combined” clinical and radiomic features. Patients were randomly separated into training (n = 139) and test (n = 59) sets. The model was constructed from training data only; the test set was reserved for validation only. Model performance was evaluated using AUC, sensitivity, specificity, PPV, and NPV. Results: Of the total 198 patients, 134 (68{\%}) patients had microsatellite stable tumors and 64 (32{\%}) patients had MSI tumors. The combined model performed slightly better than the other models, predicting MSI with an AUC of 0.80 for the training set and 0.79 for the test set (specificity = 96.8{\%} and 92.5{\%}, respectively), whereas the model with only clinical features achieved an AUC of 0.74 and the model with only radiomic features achieved an AUC of 0.76. The model with clinical features alone had the lowest specificity (70{\%}) compared with the model with radiomic features alone (95{\%}) and the combined model (92.5{\%}). Conclusions: Preoperative prediction of MSI status via radiomic analysis of preoperative CT adds specificity to clinical assessment and could contribute to personalized treatment selection.",
keywords = "Colon, Colonic neoplasms, Immunotherapy, Microsatellite instability, Microsatellite repeats",
author = "{Golia Pernicka}, {Jennifer S.} and Johan Gagniere and Jayasree Chakraborty and Rikiya Yamashita and Lorenzo Nardo and Creasy, {John M.} and Iva Petkovska and Do, {Richard R.K.} and Bates, {David D.B.} and Viktoriya Paroder and Mithat Gonen and Weiser, {Martin R.} and Simpson, {Amber L.} and Gollub, {Marc J.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s00261-019-02117-w",
language = "English (US)",
journal = "Abdominal Radiology",
issn = "2366-004X",
publisher = "Springer New York",

}

TY - JOUR

T1 - Radiomics-based prediction of microsatellite instability in colorectal cancer at initial computed tomography evaluation

AU - Golia Pernicka, Jennifer S.

AU - Gagniere, Johan

AU - Chakraborty, Jayasree

AU - Yamashita, Rikiya

AU - Nardo, Lorenzo

AU - Creasy, John M.

AU - Petkovska, Iva

AU - Do, Richard R.K.

AU - Bates, David D.B.

AU - Paroder, Viktoriya

AU - Gonen, Mithat

AU - Weiser, Martin R.

AU - Simpson, Amber L.

AU - Gollub, Marc J.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Purpose: To predict microsatellite instability (MSI) status of colon cancer on preoperative CT imaging using radiomic analysis. Methods: This retrospective study involved radiomic analysis of preoperative CT imaging of patients who underwent resection of stage II–III colon cancer from 2004 to 2012. A radiologist blinded to MSI status manually segmented the tumor region on CT images. 254 Intensity-based radiomic features were extracted from the tumor region. Three prediction models were developed with (1) only clinical features, (2) only radiomic features, and (3) “combined” clinical and radiomic features. Patients were randomly separated into training (n = 139) and test (n = 59) sets. The model was constructed from training data only; the test set was reserved for validation only. Model performance was evaluated using AUC, sensitivity, specificity, PPV, and NPV. Results: Of the total 198 patients, 134 (68%) patients had microsatellite stable tumors and 64 (32%) patients had MSI tumors. The combined model performed slightly better than the other models, predicting MSI with an AUC of 0.80 for the training set and 0.79 for the test set (specificity = 96.8% and 92.5%, respectively), whereas the model with only clinical features achieved an AUC of 0.74 and the model with only radiomic features achieved an AUC of 0.76. The model with clinical features alone had the lowest specificity (70%) compared with the model with radiomic features alone (95%) and the combined model (92.5%). Conclusions: Preoperative prediction of MSI status via radiomic analysis of preoperative CT adds specificity to clinical assessment and could contribute to personalized treatment selection.

AB - Purpose: To predict microsatellite instability (MSI) status of colon cancer on preoperative CT imaging using radiomic analysis. Methods: This retrospective study involved radiomic analysis of preoperative CT imaging of patients who underwent resection of stage II–III colon cancer from 2004 to 2012. A radiologist blinded to MSI status manually segmented the tumor region on CT images. 254 Intensity-based radiomic features were extracted from the tumor region. Three prediction models were developed with (1) only clinical features, (2) only radiomic features, and (3) “combined” clinical and radiomic features. Patients were randomly separated into training (n = 139) and test (n = 59) sets. The model was constructed from training data only; the test set was reserved for validation only. Model performance was evaluated using AUC, sensitivity, specificity, PPV, and NPV. Results: Of the total 198 patients, 134 (68%) patients had microsatellite stable tumors and 64 (32%) patients had MSI tumors. The combined model performed slightly better than the other models, predicting MSI with an AUC of 0.80 for the training set and 0.79 for the test set (specificity = 96.8% and 92.5%, respectively), whereas the model with only clinical features achieved an AUC of 0.74 and the model with only radiomic features achieved an AUC of 0.76. The model with clinical features alone had the lowest specificity (70%) compared with the model with radiomic features alone (95%) and the combined model (92.5%). Conclusions: Preoperative prediction of MSI status via radiomic analysis of preoperative CT adds specificity to clinical assessment and could contribute to personalized treatment selection.

KW - Colon

KW - Colonic neoplasms

KW - Immunotherapy

KW - Microsatellite instability

KW - Microsatellite repeats

UR - http://www.scopus.com/inward/record.url?scp=85068220750&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068220750&partnerID=8YFLogxK

U2 - 10.1007/s00261-019-02117-w

DO - 10.1007/s00261-019-02117-w

M3 - Article

AN - SCOPUS:85068220750

JO - Abdominal Radiology

JF - Abdominal Radiology

SN - 2366-004X

ER -