Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Computed tomography-based radiomic analysis for predicting pathological response and prognosis after neoadjuvant chemotherapy in patients with locally advanced esophageal cancer

Computed tomography-based radiomic analysis for predicting pathological response and prognosis... PurposeAccurate prediction of prognosis and pathological response to neoadjuvant chemotherapy (NAC) is crucial for optimizing treatment strategies for patients with locally advanced esophageal cancer (LA-EC). This study aimed to investigate the use of radiomics for pretreatment CT in predicting the pathological response of patients with LA-EC to NAC.MethodsOverall, 144 patients (145 lesions) with LA-EC who underwent pretreatment contrast-enhanced CT and then received NAC followed by surgery with pathological tumor regression grade (TRG) analysis were enrolled. The obtained dataset was randomly divided into training and validation cohorts using fivefold cross-validation. CT-based radiomic features were extracted followed by the feature selection process using the variance threshold, SelectKBest, and least absolute shrinkage and selection operator methods. The radiomic model was constructed using six machine learning classifiers, and predictive performance was evaluated using ROC curve analysis in the training and validation cohorts.ResultsAll patients were divided into responders (n = 40, 28%) and non-responders (n = 104, 72%) based on the TRG results and a statistically significant split by overall survival analysis (0.899 [0.754–0.961] vs. 0.630 [0.510–0.729], respectively). There were no significant differences between responders and non-responders in terms of age, sex, tumor size, tumor location, or histopathology. The mean AUC of fivefold in the validation cohort was 0.720 (confidence interval [CI]: 0.594–0.982), and the best AUC of the radiomic model using logistic regression to predict the non-responders was 0.815 (CI: 0.626–1.000, sensitivity 0.620, specificity 0.860).ConclusionA radiomic model derived from contrast-enhanced CT may help stratify chemotherapy effect prediction and improve clinical decision-making.Graphical Abstract[graphic not available: see fulltext] http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Abdominal Radiology Springer Journals

Computed tomography-based radiomic analysis for predicting pathological response and prognosis after neoadjuvant chemotherapy in patients with locally advanced esophageal cancer

Loading next page...
 
/lp/springer-journals/computed-tomography-based-radiomic-analysis-for-predicting-0D80X4GnG9

References (60)

Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
ISSN
2366-004X
eISSN
2366-0058
DOI
10.1007/s00261-023-03938-6
Publisher site
See Article on Publisher Site

Abstract

PurposeAccurate prediction of prognosis and pathological response to neoadjuvant chemotherapy (NAC) is crucial for optimizing treatment strategies for patients with locally advanced esophageal cancer (LA-EC). This study aimed to investigate the use of radiomics for pretreatment CT in predicting the pathological response of patients with LA-EC to NAC.MethodsOverall, 144 patients (145 lesions) with LA-EC who underwent pretreatment contrast-enhanced CT and then received NAC followed by surgery with pathological tumor regression grade (TRG) analysis were enrolled. The obtained dataset was randomly divided into training and validation cohorts using fivefold cross-validation. CT-based radiomic features were extracted followed by the feature selection process using the variance threshold, SelectKBest, and least absolute shrinkage and selection operator methods. The radiomic model was constructed using six machine learning classifiers, and predictive performance was evaluated using ROC curve analysis in the training and validation cohorts.ResultsAll patients were divided into responders (n = 40, 28%) and non-responders (n = 104, 72%) based on the TRG results and a statistically significant split by overall survival analysis (0.899 [0.754–0.961] vs. 0.630 [0.510–0.729], respectively). There were no significant differences between responders and non-responders in terms of age, sex, tumor size, tumor location, or histopathology. The mean AUC of fivefold in the validation cohort was 0.720 (confidence interval [CI]: 0.594–0.982), and the best AUC of the radiomic model using logistic regression to predict the non-responders was 0.815 (CI: 0.626–1.000, sensitivity 0.620, specificity 0.860).ConclusionA radiomic model derived from contrast-enhanced CT may help stratify chemotherapy effect prediction and improve clinical decision-making.Graphical Abstract[graphic not available: see fulltext]

Journal

Abdominal RadiologySpringer Journals

Published: Aug 1, 2023

Keywords: Esophageal neoplasm; Clinical decision-making; Survival analysis; Neoadjuvant chemotherapy; Computed tomography

There are no references for this article.