Work

Dosiomics and Radiomics Guided Precision Neoadjuvant Chemoradiotherapy for Esophageal Squamous Cell Carcinoma Patients

Year: 2024

Type: article

Abstract: Purpose/Objective(s) This investigation aims to development and validation a dual-omics model to predict the pathology complete response (pCR) of esophageal squamous cell carcinoma (ESCC) patients after neoadjuvant chemoradiotherapy. Materials/Methods This retrospective study recruited patients who was diagnosis with ESCC by biopsy between 01.01.2021 to 12.31.2023 at one hospital. All of the involved patients were divided into training cohort and validation cohort according to the ratio as 6:4. The radiomics features and dosiomics features were extracted from CT image and dose image by Pyradiomcs. Least absolute shrinkage and selection operator (LASSO) was used to select the radiomics and dosiomics features for building dual-omics model. Logistic regression model was used to select clinical factors. Combined radiomics, dosiomics features and clinical factors to develop the dual-omics model. Receiver operate characteristic (ROC) curve and calibration curve were constructed to display the model's performance. Nomogram demonstrated the predictive ability of the model. Results 276 ESCC patients were recruited after including and excluding criterions (61.5±6.29 years and 232 men). There were one hundred and twenty-seven patients (46%) achieved pCR. 944 radiomic features and 930 dosiomics features were extracted, respectively. Finally, three radiomics features and one dosiomics feature were selected for model's development. One clinical factor (Red Blood Cell, RBC) had also been used to construct the model. The combined model showed superior performance in training cohort and validation cohort, the area under the curve (AUC) values were 0.820,0.796 respectively. The calibration curve also demonstrated the superior performance of the model. Conclusion The dual-omics model combined clinical factor displays superior pCR prediction ability, which would help the clinician optimize treatment decision and prolong patient's overall survival.

Source: International Journal of Radiation Oncology*Biology*Physics

DOI: 10.1016/j.ijrobp.2024.07.2031

Author: Y. Cui

Language: English

Topic: Radiomics and Machine Learning in Medical Imaging

Subfield: Radiology, Nuclear Medicine and Imaging

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: hybrid

Estimated APC paid: $3,940

Funder: National Cancer Institute