AI-assisted staging and treatment decisions for hepatocellular carcinoma

A Prospective, Randomized, Controlled, Crossover Study of Artificial Intelligence-Assisted Multi-Dimensional Staging and Treatment Decision-Making for Hepatocellular Carcinoma

This project will test whether an AI tool helps doctors more accurately stage HCC and choose treatments for adult patients with new or suspected primary liver cancer.

Quick facts

• Study type: Observational
• Enrollment: 108 (estimated)
• Ages: 18 Years and up
• Sex: All
• Sponsor: Beijing Tsinghua Chang Gung Hospital (other)
• Locations: 1 site (Beijing, Changping)
• Trial ID: NCT07538882 on ClinicalTrials.gov

What this trial studies

This prospective, multi-center study enrolls adults with suspected or newly diagnosed primary hepatocellular carcinoma and compares physician decisions made with and without an AI assistance tool using a balanced multi-rater, multi-case crossover design. Physicians from different hospital tiers review the same imaging and baseline clinical data under unassisted and AI-assisted conditions, and their staging (CNLC, TNM, BCLC) and treatment recommendations are recorded. An independent three-expert panel establishes the reference standard using full imaging, clinical data, and multidisciplinary discussion. The design specifically examines whether AI reduces diagnostic and therapeutic variability between primary/secondary hospitals and higher-tier centers.

Who should consider this trial

Good fit: Adults (≥18 years) with suspected or newly diagnosed primary HCC who can provide consent and have complete baseline clinical data and contrast-enhanced abdominal CT are ideal candidates.

Not a fit: Patients with secondary (metastatic) liver cancer, concurrent other active malignancies, missing required imaging or lab data, or prior anti-tumor treatment are unlikely to benefit or be eligible.

Why it matters

Potential benefit: If successful, the AI could improve staging accuracy and lead to more consistent, evidence-aligned treatment choices across hospitals, especially benefiting patients at lower-tier centers.

How similar studies have performed: Prior AI work in liver imaging has shown promise for lesion detection and staging, but prospective multi-center MRMC comparisons of AI-assisted clinical decision-making remain limited.

Eligibility criteria

Inclusion Criteria:

* Age \>= 18 years.
* Patients prospectively presenting with suspected or newly diagnosed primary hepatocellular carcinoma (HCC) later confirmed by pathology or meeting the China Liver Cancer (CNLC) guidelines.
* Complete baseline clinical data acquired during the prospective enrollment period, including complete history of present/past illness, ECOG PS score, comprehensive laboratory tests (liver function, coagulation, tumor markers such as AFP, etc.), and baseline abdominal contrast-enhanced CT.
* Patients (or their legal representatives) must provide written informed consent for their clinical data to be used in this trial.

Exclusion Criteria:

* Patients with secondary (metastatic) liver cancer or concurrent severe malignancies of other systems.
* Patients who fail to complete the required baseline imaging or laboratory tests, preventing accurate staging calculation (e.g., missing data for Child-Pugh score).
* Patients who have previously received anti-tumor therapies for liver cancer prior to enrollment.

Where this trial is running

Beijing, Changping

• Beijing Tsinghua Changgung Hospital — Beijing, Changping, China (RECRUITING)

Study contacts

• Study coordinator: Jitao Wang

Conditions: Hepatocellular Carcinoma, Artificial Intelligence, Staging, Clinical Decision-Making, Diagnostic Accuracy