Abstract 204P
Background
Although neoadjuvant chemoradiotherapy followed by surgery is the standard treatment for esophageal cancer patients, most patients are unable to achieve pathological complete response with neoadjuvant therapy, resulting in poor outcomes. The aim of this study is to develop a method for selecting patients who can achieve pathological complete response through pre-neoadjuvant therapy chest-enhanced CT scans.
Methods
Two hundreds and one patients with esophageal cancer were enrolled and divided into a training set and a testing set in a 7:3 ratio. Radiomics features of intra-tumoral and peritumoral images were extracted from preoperative chest-enhanced CT scans of these patients. The features were dimensionally reduced in two steps. The selected intra-tumoral and peritumoral features, including marginal (with a distance of 0-3mm from the tumor) and adjacent (with a distance of 3-6mm from the tumor) ROI, were used to build models with four machine learning classifiers, including Support Vector Machine, XG-Boost, Random Forest and Naive Bayes. Models with satisfied accuracy and stability levels were considered to perform well. Finally, the performance of these well-performing models on the testing set was displayed using ROC curves.
Results
Among the 16 models, the best-performing models were the integrated (intra-tumoral and peritumoral features)-XGBoost and integrated-random forest models, which had average ROC AUCs of 0.906 and 0.918, respectively, with relative standard deviations (RSDs) of 6.26 and 6.89 in the training set. In the testing set, the AUCs were 0.845 and 0.871, respectively. There was no significant difference in the ROC curves between the two models. Table: 204P
The performance of the selected models on the testing set
Model | AUC (95% CI) | Specificity | Sensitivity |
Integrated-XGBoost | 0.845 (0.764, 0.928) | 0.864 | 0.777 |
Original-XGBoost | 0.759 (0.660, 0.857) | 0.900 | 0.592 |
Integrated-Random Forest | 0.871 (0.796, 0.946) | 0.682 | 0.933 |
Original-Random Forest | 0.795 (0.703, 0.887) | 0.825 | 0.673 |
Adjacent-Random Forest | 0.769 (0.671, 0.868) | 0.886 | 0.533 |
Integrated-Support Vector Machine | 0.719 (0.613, 0.825) | 0.795 | 0.622 |
Conclusions
The addition of peritumoral radiomics features to the radiomics analysis may improve the predictive performance of pathological response for esophageal cancer patients to neoadjuvant chemoradiotherapy.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
111P - Comparison of the efficacy and safety of fruquintinib and fruquintinib combined with immune checkpoint inhibitors in the treatment of metastatic microsatellite stable colorectal cancer: A real-world study
Presenter: Zhiqiang Wang
Session: Poster Display
Resources:
Abstract
112P - Optimal classification and treatment strategy based on technical and oncological futures in recurrence of colorectal liver metastases
Presenter: Kosuke Kobayashi
Session: Poster Display
Resources:
Abstract
113P - Phase I/II study of capecitabine(C)/oxaliplatin(O)/irinotecan(I) combined with bevacizumab(B) in the first-line treatment of metastatic colorectal cancer (mCRC)
Presenter: Kai Ou
Session: Poster Display
Resources:
Abstract
114P - The prognostic role of LAG-3 expression in metastatic colorectal cancer
Presenter: Yi-Hsuan Huang
Session: Poster Display
Resources:
Abstract
115P - Sidedness and survival of chemo-refractory metastatic colorectal cancer treated with lonsurf or regorafenib: A nationwide population-based study in Taiwan
Presenter: Meng-Che Hsieh
Session: Poster Display
Resources:
Abstract
116P - Burden and trends of colorectal cancer in high income Asia Pacific countries from 1990-2019 and its projections of deaths to 2040: A comparative analysis
Presenter: Monika Chhayani
Session: Poster Display
Resources:
Abstract
117P - Australasian real-world treatment selection and clinical outcomes for patients with left side (LS), RAS wildtype (RASwt) metastatic colorectal cancer (mCRC)
Presenter: Vanessa Wong
Session: Poster Display
Resources:
Abstract
119P - Neoadjuvant chemoradiotherapy in the mode of hypofractionation in locally advanced rectal cancer: Is it time to change standards of care?
Presenter: Abror Abdujapparov
Session: Poster Display
Resources:
Abstract
120P - Improved clinical outcomes with cetuximab maintenance therapy in left-sided RAS/BRAF wild-type metastatic colorectal cancer: A real-world study of Hunan cancer hospital
Presenter: Xiaolin Yang
Session: Poster Display
Resources:
Abstract
121P - Single-cell sequencing reveals the role of Treg cells with high expression of BIRC3 in regulating the progression of colorectal cancer
Presenter: Yuqiu Xu
Session: Poster Display
Resources:
Abstract