Multimodal AI for Precision Diagnosis of Esophageal Cancer

This study intends to construct two multimodal deep learning models: one for the diagnosis of esophageal cancer and the prediction of invasive depth to assess suitability for endoscopic resection; the other model, based on this, classifies endoscopic non-resectable patients into different degrees of invasion to further explore the differences in the sensitivity and survival of AI-predicted benign and malignant tumors in patients' responses to NAT, thereby providing reliable decision support for precise individualized treatment. This aspect has rarely been addressed in previous studies.

Study Overview

• Status: Active, not recruiting
• Conditions: Esophageal Cancer
• Intervention / Treatment: Other: NAT; Other: surgery; Other: ESD

• Study Type: Observational
• Enrollment (Actual): 264

Contacts and Locations

Study Locations

• China
  • Shanghai, China
    • Zhongshan Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

The study population will consist of adult patients (≥18 years) with pathologically confirmed esophageal squamous cell carcinoma (ESCC) who underwent diagnostic endoscopy and contrast-enhanced CT examination at participating centers. Patients with early-stage disease who received endoscopic resection or esophagectomy and had pathological assessment of tumor invasion depth will be included for development and validation of a multimodal deep learning model for diagnosis and prediction of invasion depth. Patients with locally advanced disease who received neoadjuvant therapy followed by surgical resection, with available pathological response and follow-up data, will be included for development and validation of a multimodal model for predicting treatment response and prognosis. Clinical information, CT images, endoscopic images, pathological findings, and survival outcomes will be collected for analysis.

Description

Inclusion Criteria:

• Age ≥18 years.
• Histologically confirmed or clinically suspected esophageal squamous cell carcinoma (ESCC).
• Availability of pre-treatment endoscopic images and contrast-enhanced chest/upper abdominal CT scans.
• Availability of complete clinical and pathological data.
• Patients who underwent endoscopic resection (ESD/EMR) or esophagectomy with pathological assessment of tumor invasion depth.
• Adequate image quality for analysis.
• Written informed consent (for prospective cohorts, if applicable).

Exclusion Criteria:

• Histology other than squamous cell carcinoma.
• Prior treatment for esophageal cancer before baseline imaging, including chemotherapy, radiotherapy, immunotherapy, or endoscopic resection.
• Distant metastatic disease at diagnosis.
• Incomplete clinical, imaging, or pathological data.
• Poor-quality CT or endoscopic images unsuitable for analysis.
• History of another active malignancy within the past 5 years.
• Recurrent esophageal cancer.

Study Plan

How is the study designed?

Number of groups / cohorts

3

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
ESD	Other: ESD; Endoscopic submucosal dissection (ESD) is a minimally invasive endoscopic technique used for en bloc resection of superficial gastrointestinal neoplasms. The procedure is performed under conscious sedation or general anesthesia using a therapeutic endoscope. After lesion characterization and marking of the resection margins, a submucosal injection solution (e.g., saline mixed with epinephrine, dye, or viscous agents such as hyaluronic acid) is administered to lift the lesion from the muscularis propria. A circumferential mucosal incision is then made using an endoscopic knife, followed by meticulous submucosal dissection to separate the lesion from the underlying muscle layer. Hemostasis is achieved throughout the procedure using coagulation forceps or hemostatic devices as needed. The lesion is removed en bloc whenever possible, and the resected specimen is retrieved for histopathological evaluation. Post-resection inspection of the artificial ulcer is performed to assess for bleeding
ESD+surgery	Other: surgery; Additional surgery refers to esophagectomy with regional lymph node dissection performed after non-curative endoscopic resection. The procedure is recommended for patients with pathological risk factors for lymph node metastasis, including deep submucosal invasion, lymphovascular invasion, positive vertical resection margins, or other non-curative resection criteria. Surgical treatment is performed according to institutional standards and current clinical guidelines.; Other: ESD; Endoscopic submucosal dissection (ESD) is a minimally invasive endoscopic technique used for en bloc resection of superficial gastrointestinal neoplasms. The procedure is performed under conscious sedation or general anesthesia using a therapeutic endoscope. After lesion characterization and marking of the resection margins, a submucosal injection solution (e.g., saline mixed with epinephrine, dye, or viscous agents such as hyaluronic acid) is administered to lift the lesion from the muscularis propria. A circumferential mucosal incision is then made using an endoscopic knife, followed by meticulous submucosal dissection to separate the lesion from the underlying muscle layer. Hemostasis is achieved throughout the procedure using coagulation forceps or hemostatic devices as needed. The lesion is removed en bloc whenever possible, and the resected specimen is retrieved for histopathological evaluation. Post-resection inspection of the artificial ulcer is performed to assess for bleeding
NAT +surgery	Other: NAT; Neoadjuvant therapy has become a standard treatment strategy for patients with locally advanced esophageal squamous cell carcinoma (ESCC). Administered before surgical resection, neoadjuvant treatment aims to reduce tumor burden, downstage disease, increase the likelihood of complete (R0) resection, and improve long-term survival outcomes. Current neoadjuvant approaches include chemotherapy, chemoradiotherapy, and more recently, immune checkpoint inhibitor-based combination therapies. Among these, neoadjuvant immunochemotherapy has demonstrated promising pathological response rates and survival benefits, emerging as an important treatment option for ESCC. However, treatment response varies considerably among patients, highlighting the need for reliable predictive models to guide individualized therapeutic decision-making.; Other: surgery; Additional surgery refers to esophagectomy with regional lymph node dissection performed after non-curative endoscopic resection. The procedure is recommended for patients with pathological risk factors for lymph node metastasis, including deep submucosal invasion, lymphovascular invasion, positive vertical resection margins, or other non-curative resection criteria. Surgical treatment is performed according to institutional standards and current clinical guidelines.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Actual receipt of additional surgery	Actual receipt of additional surgery was defined as patients who underwent subsequent radical surgical treatment after ESD based on multidisciplinary team (MDT) recommendations, clinician judgment, and patient preference.	1 week
Presence of pathological indications for additional surgery	According to current guidelines, non-curative resection requiring consideration of additional radical treatment was defined by the presence of submucosal invasion (SM2 invasion), or lymphovascular invasion (LVI-positive), or a positive vertical resection margin (VM-positive).	1 week

Collaborators and Investigators

• Sponsor: Shanghai Zhongshan Hospital

Study record dates

Study Major Dates

• Study Start (Actual): March 1, 2023
• Primary Completion (Actual): January 1, 2026
• Study Completion (Estimated): June 7, 2026

Study Registration Dates

• First Submitted: June 1, 2026
• First Submitted That Met QC Criteria: June 1, 2026
• First Posted (Actual): June 5, 2026

Study Record Updates

• Last Update Posted (Actual): June 5, 2026
• Last Update Submitted That Met QC Criteria: June 1, 2026
• Last Verified: June 1, 2026

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Neoplasms by Site; Neoplasms; Gastrointestinal Neoplasms; Digestive System Neoplasms; Digestive System Diseases; Gastrointestinal Diseases; Head and Neck Neoplasms; Esophageal Diseases; Esophageal Neoplasms; Surgical Procedures, Operative; N-acetyltalosaminuronic acid
• Other Study ID Numbers: B2025-145(2)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No