# Clinical-Radiomics Signature Predicts Pathologic Complete Response After Neoadjuvant Therapy in Oesophageal Squamous Cell Carcinoma

## TL;DR

This study developed a model combining clinical and radiomic data to predict which patients with esophageal cancer achieve a complete response after pre-surgery treatment.

## Contribution

A novel clinical-radiomics model was developed to predict pathologic complete response in ESCC after neoadjuvant therapy.

## Key findings

- The model achieved an AUC of 0.91 in the training cohort and 0.84 in the test cohort.
- Calibration and decision curve analyses confirmed the model's accuracy and clinical utility.

## Abstract

Neoadjuvant therapy (NAT) significantly improves the pathologic complete response (pCR) rates in patients with locally advanced esophageal squamous cell carcinoma (ESCC). Emerging evidence suggests that patients with pCR may experience favourable outcomes and could be considered for active surveillance strategies to delay surgery. This study aims to develop a clinical-radiomics model to predict pCR after NAT in ESCC.

We retrospectively enrolled 236 patients with locally advanced ESCC who received NAT at our centre and randomly assigned them to training and test cohorts (3:2 ratio). Radiomics features were extracted from tumour regions segmented on post-NAT contrast-enhanced computed tomography (CT) scans. After feature selection, a predictive model integrating radiomics and clinical variables was developed using logistic regression and visualized as a nomogram. Model performance was evaluated using area under the curve (AUC), accuracy, sensitivity, and specificity.

The clinical-radiomics model achieved an AUC of 0.91 (95% confidence interval [CI]: 0.86-0.95), accuracy of 0.84, sensitivity of 0.89, and specificity of 0.81 in the training cohort, and an AUC of 0.84 (95% CI: 0.76-0.92), accuracy of 0.78, sensitivity of 0.84, and specificity of 0.74 in the test cohort. Calibration curves demonstrated good agreement between predicted and observed outcomes, and decision curve analysis confirmed the model’s clinical utility.

The clinical-radiomics model accurately predicts pCR following NAT in ESCC and may guide personalized treatment strategies.

Oesophageal cancer is the seventh most common cancer and sixth leading cause of cancer-related death worldwide,1 with China accounting for over half of global ESCC cases.