Predicting pathological complete response in rectal cancer after chemoradiotherapy with a random forest using 18F-fluorodeoxyglucose positron emission tomography and computed tomography radiomics

Wei-Chih Shen; Shang-Wen Chen; Kuo-Chen Wu; Peng-Yi Lee; Chun-Lung Feng; Te-Chun Hsieh; Kuo-Yang Yen; Chia-Hung Kao

doi:10.21037/atm.2020.01.107

Predicting pathological complete response in rectal cancer after chemoradiotherapy with a random forest using ¹⁸F-fluorodeoxyglucose positron emission tomography and computed tomography radiomics

Ann Transl Med. 2020 Mar;8(5):207. doi: 10.21037/atm.2020.01.107.

Authors

Wei-Chih Shen^{1

2}, Shang-Wen Chen^{2

3

4

5}, Kuo-Chen Wu^{2

6}, Peng-Yi Lee^{3

7}, Chun-Lung Feng⁸, Te-Chun Hsieh^{9

10}, Kuo-Yang Yen^{9

10}, Chia-Hung Kao^{2

9

11

12}

Affiliations

¹ Department of Computer Science and Information Engineering, Asia University, Taichung.
² Center of Augmented Intelligence in Healthcare, China Medical University Hospital, Taichung.
³ Department of Radiation Oncology, China Medical University Hospital, Taichung.
⁴ School of Medicine, College of Medicine, China Medical University, Taichung.
⁵ Department of Radiology, School of Medicine, College of Medicine, Taipei Medical University, Taipei.
⁶ Department of Computer Science and Engineering, National Chung Hsing University, Taichung.
⁷ Department of Radiation Oncology, China Medical University Hospital, Yunlin.
⁸ Division of Hepato-Gastroenterology, Department of Internal Medicine, China Medical University Hospital, Taichung.
⁹ Department of Nuclear Medicine and PET Center, China Medical University Hospital, Taichung.
¹⁰ Department of Biomedical Imaging and Radiological Science, School of Medicine, College of Medicine, China Medical University, Taichung.
¹¹ Graduate Institute of Biomedical Sciences, School of Medicine, College of Medicine, China Medical University, Taichung.
¹² Department of Bioinformatics and Medical Engineering, Asia University, Taichung.

Abstract

Background: Neoadjuvant chemoradiotherapy (NCRT) followed by surgery is the standard treatment for patients with locally advanced rectal cancer. This study developed a random forest (RF) model to predict pathological complete response (pCR) based on radiomics derived from baseline ¹⁸F-fluorodeoxyglucose ([¹⁸F]FDG)-positron emission tomography (PET)/computed tomography (CT).

Methods: This study included 169 patients with newly diagnosed rectal cancer. All patients received ¹⁸F[FDG]-PET/CT, NCRT, and surgery. In total, 68 radiomic features were extracted from the metabolic tumor volume. The numbers of splits in a decision tree and trees in an RF were determined based on their effects on predictive performance. Receiver operating characteristic curve analysis was performed to evaluate predictive performance and ascertain the optimal threshold for maximizing prediction accuracy.

Results: After NCRT, 22 patients (13%) achieved pCR, and 42 features that could differentiate tumors with pCR were used to construct the RF model. Six decision trees and seven splits were suitable. Accordingly, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 81.8%, 97.3%, 81.8%, 97.3%, and 95.3%, respectively.

Conclusions: By using an RF, we determined that radiomics derived from baseline ¹⁸F[FDG]-PET/CT could accurately predict pCR in patients with rectal cancer. Highly accurate and predictive values can be achieved but should be externally validated.

Keywords: 18F-fluorodeoxyglucose ([18F]FDG); artificial intelligence; computed tomography (CT); machine learning; pathological complete response; positron emission tomography (PET); radiomics; random forest (RF); rectal cancer.