Prediction of epidermal growth factor receptor mutation status by textural features in stage IV lung adenocarcinoma
- 28.02.2024
- original report
Erschienen in:
- Verfasst von
-
Changbin Wang
Changbin Wang
- Department of Radiation Oncology, Shengli Oilfield Central Hospital, Dongying, China
-
Ran Zhang
Ran Zhang
- Department of Radiation Oncology, Shanghai First Maternity and Infant Hospital, School of Medicine, Tongji University, Shanghai, China
- Tongji University, Shanghai, China
-
Xiaorong Sun
Xiaorong Sun
- Department of Nuclear Medicine, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, China
-
Ligang Xing
Korrespondierender Autor Ligang Xing
- Department of Radiation Oncology, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, China
- Erschienen in
- memo - Magazine of European Medical Oncology | Ausgabe 2/2024
Summary
Background
The purpose of this study was to assess the correlation between textural features from 18fluorodeoxyglucose positron emission tomography (18F-FDG-PET) or computed tomography (CT) and EGFR mutation status in patients with stage IV adenocarcinoma lung cancer.
Methods
In all, 71 patients who were diagnosed with stage IV adenocarcinoma lung cancer between April 2014 and August 2018 were included in this study. 18F-FDG-PET/CT scanning and EGFR mutation tests were performed before targeted molecular therapy. Textural features were extracted from manually segmented volumes of tumors, and highly dependent features were excluded. Multivariate logistic regression analysis was used to establish predictive models for detection of EGFR mutations. Receiver operating characteristic (ROC) curves were applied to evaluate areas under the curves (AUCs) of each model.
Results
Of the 71 patients, 39 (54.9%) were EGFR mutation and 32 (45.1%) showed wild-type. EGFR mutation status was significantly associated with female sex (P = 0.026). In multivariate analysis, three PET (co-occurrence contrast, intensity-size-zone low-intensity large-zone emphasis, and texture spectrum max spectrum) and two CT quantitative features (intensity-size-zone high-intensity zone emphasis and normalized co-occurrence second angular moment) were independent predictors of EGFR mutation status. The predictive model generated from combined clinical and textural features showed a better predictive value than the model from textural features alone (AUC 0.897 vs 0.864).
Conclusions
Textural features combined with clinical features could establish a model for improving the predicting power of EGFR mutation status in patients with stage IV adenocarcinoma lung cancer.
Anzeige
- Titel
- Prediction of epidermal growth factor receptor mutation status by textural features in stage IV lung adenocarcinoma
- Verfasst von
-
Changbin Wang
Ran Zhang
Xiaorong Sun
Ligang Xing
- Publikationsdatum
- 28.02.2024
- Verlag
- Springer Vienna
- Erschienen in
-
memo - Magazine of European Medical Oncology / Ausgabe 2/2024
Print ISSN: 1865-5041
Elektronische ISSN: 1865-5076 - DOI
- https://doi.org/10.1007/s12254-024-00961-1
Dieser Inhalt ist nur sichtbar, wenn du eingeloggt bist und die entsprechende Berechtigung hast.