Clinical-Radiomics Signature Predicts Pathologic Complete Response After Neoadjuvant Therapy in Oesophageal Squamous...

Area of Science:

• Oncology
• Radiology
• Medical Imaging

Background:

• Neoadjuvant therapy (NAT) improves outcomes for locally advanced esophageal squamous cell carcinoma (ESCC).
• Pathologic complete response (pCR) after NAT is associated with favorable patient outcomes.
• Predicting pCR is crucial for tailoring treatment and considering surveillance strategies.

Purpose of the Study:

• To develop and validate a clinical-radiomics model for predicting pCR after NAT in ESCC patients.
• To integrate clinical variables with radiomics features for enhanced predictive accuracy.
• To provide a tool for personalized treatment decisions in ESCC management.

Main Methods:

• Retrospective analysis of 236 locally advanced ESCC patients treated with NAT.
• Extraction of radiomics features from post-NAT contrast-enhanced CT scans.
• Development of a logistic regression model combining clinical and radiomics data, visualized as a nomogram.

Main Results:

• The clinical-radiomics model demonstrated high predictive performance in both training (AUC 0.91) and test (AUC 0.84) cohorts.
• The model achieved good accuracy, sensitivity, and specificity in predicting pCR.
• Calibration curves and decision curve analysis confirmed the model's reliability and clinical utility.

Conclusions:

• The developed clinical-radiomics model accurately predicts pCR in ESCC patients receiving NAT.
• This predictive tool can aid in guiding personalized treatment strategies and potentially de-escalating therapy.
• Further validation may support its use in clinical decision-making for ESCC management.