Evaluating the Efficacy and Safety of AI Localization Models in Multidisciplinary Team Care for NSCLC

Evaluating the Efficacy and Safety of AI Localization Models in Multidisciplinary Team Care for NSCLC: a Prospective, Controlled Clinical Trial Protocol

The goal of this clinical trial is to evaluate the effectiveness and safety of a locally deployed artificial intelligence (AI) decision-support model in the multidisciplinary team (MDT) process for patients with non-small cell lung cancer (NSCLC).

The main questions it aims to answer :

What is the level of agreement between treatment recommendations generated by the AI model and those made by a traditional MDT? How often do clinicians modify their final treatment decision after reviewing the AI model's recommendation? Researchers will compare treatment plans from the traditional MDT (Arm 1), the AI model (Arm 2), and the clinician's final decision after reviewing the AI output (Arm 3) to assess consistency, decision modification rates, and clinical efficiency.

Participants will:

Have their clinical, imaging, and molecular data submitted to both the traditional MDT and the AI model for independent treatment recommendations Receive a final treatment plan determined by clinicians after reviewing both recommendations, with follow-up for safety and survival outcomes

Study Overview

• Status: Recruiting
• Conditions: Nonsmall Cell Lung Cancer
• Intervention / Treatment: Diagnostic test: Treat Regimen

• Study Type: Interventional
• Enrollment (Estimated): 300
• Phase: Not Applicable

Contacts and Locations

• Study Contact: Name: qing liang, Dr.; Phone Number: +86 17863321987; Email: liangtsing99@163.com

Study Locations

• China
  • Guangdong
    • Guangzhou, Guangdong, China, 510000
      • Recruiting
      • Guangdong Provincial People's Hospital
      • Contact: Wenzhao Zhong, Dr.; Phone Number: +8613609777314; Email: 13609777314@163.com

Participation Criteria

Eligibility Criteria

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

Description

Inclusion Criteria:

1. Age ≥ 18 years;
2. MDT (Multidisciplinary Team) discussion deems a systemic treatment plan necessary;
3. Complete clinical, imaging, and molecular pathological data.

Exclusion Criteria:

1. Stage I patients;
2. Diagnosed with a thoracic tumor other than NSCLC;
3. Lack of detailed medical data, or missing data;

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: N/A
• Interventional Model: Single Group Assignment
• Masking: None (Open Label)

Number of Arms

1

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: AI-Assisted Multidisciplinary Team Decision-Making for Non-Small Cell Lung Cancer	Diagnostic test: Treat Regimen; The impact of artificial intelligence on clinicians' treatment plans

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Consistency rate	Consistency rate between Option 1 and Option 2 (calculated using Kappa value). Consistency rate between Option 1 and Option 3 (decision modification rate).	Baseline(MDT 1 Day)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
MDT Discussion Process Time	Time from start to end of multidisciplinary team (MDT) discussion, measured immediately after MDT end.	Baseline(MDT Day 1)
Quality of AI Recommendations	Physician-rated quality of AI recommendations using a Likert 5-point scale (1 = very poor, 5 = excellent).	Baseline(MDT Day 1)
Clinical Acceptability of AI	Physician-rated clinical acceptability of AI recommendations using a Likert 5-point scale (1 = unacceptable, 5 = fully acceptable).	Baseline(MDT Day 1)
MDT Discussion Efficiency	Physician-rated efficiency of MDT discussion process aided by AI using a Likert 5-point scale (1 = very inefficient, 5 = very efficient).	Baseline(MDT Day 1)
Process Convenience	Physician-rated convenience of the AI-integrated workflow using a Likert 5-point scale (1 = very inconvenient, 5 = very convenient).	Baseline(MDT Day 1)
Added Value to Clinical Decision	Physician-rated added value of AI to clinical decision-making using a Likert 5-point scale (1 = no added value, 5 = significant added value).	Baseline(MDT Day 1)
Learning and Training Value	Physician-rated learning and training value of AI system using a Likert 5-point scale (1 = no value, 5 = high value).	Baseline(MDT Day 1)
Overall Satisfaction	Physician-rated overall satisfaction with AI-assisted MDT using a Likert 5-point scale (1 = very dissatisfied, 5 = very satisfied).	Baseline(MDT Day 1)
Willingness to Use in Future	Physician-rated willingness to use AI system in future clinical practice using a Likert 5-point scale (1 = definitely not willing, 5 = definitely willing).	Baseline(MDT Day 1)
Disease-Free Survival (DFS)	Time from treatment initiation to disease recurrence or death from any cause, assessed every 3-6 months during 2-3 years follow-up.	3 years
Progression-Free Survival (PFS)	Time from treatment initiation to disease progression or death from any cause, assessed every 3-6 months during 2-3 years follow-up.	3 years
Overall Survival (OS)	Time from treatment initiation to death from any cause, assessed every 3-6 months during 2-3 years follow-up.	3 years

Collaborators and Investigators

• Sponsor: Wen-zhao ZHONG
• Collaborators: Guangdong Provincial People's Hospital

Publications and helpful links

General Publications

• Pillay B, Wootten AC, Crowe H, Corcoran N, Tran B, Bowden P, Crowe J, Costello AJ. The impact of multidisciplinary team meetings on patient assessment, management and outcomes in oncology settings: A systematic review of the literature. Cancer Treat Rev. 2016 Jan;42:56-72. doi: 10.1016/j.ctrv.2015.11.007. Epub 2015 Nov 24.
• Kim JK, Chua ME, Li TG, Rickard M, Lorenzo AJ. Novel AI applications in systematic review: GPT-4 assisted data extraction, analysis, review of bias. BMJ Evid Based Med. 2025 Sep 22;30(5):313-322. doi: 10.1136/bmjebm-2024-113066.
• Wiegand TLT, Jung LB, Gudera JA, Schuhmacher LS, Moehrle P, Rischewski JF, Mehrzad P, Jeong S, Nguyen LH, Poeschla M, Velezmoro LI, Kruk L, Dimitriadis K, Koerte IK. Demographic inaccuracies and biases in the depiction of patients by artificial intelligence text-to-image generators. NPJ Digit Med. 2025 Jul 19;8(1):459. doi: 10.1038/s41746-025-01817-6.

Study record dates

Study Major Dates

• Study Start (Actual): December 1, 2025
• Primary Completion (Estimated): October 31, 2027
• Study Completion (Estimated): December 31, 2028

Study Registration Dates

• First Submitted: March 25, 2026
• First Submitted That Met QC Criteria: May 31, 2026
• First Posted (Actual): June 4, 2026

Study Record Updates

• Last Update Posted (Actual): June 4, 2026
• Last Update Submitted That Met QC Criteria: May 31, 2026
• Last Verified: May 1, 2026

More Information

Terms related to this study

• Keywords: Non-Small Cell Lung Cancer; Treatment Decision-Making; Multidisciplinary Team; Large Language Model; Locally Deployed AI Model
• Additional Relevant MeSH Terms: Neoplasms by Site; Neoplasms; Respiratory Tract Diseases; Lung Diseases; Respiratory Tract Neoplasms; Thoracic Neoplasms; Lung Neoplasms; Carcinoma, Bronchogenic; Bronchial Neoplasms; Carcinoma, Non-Small-Cell Lung
• Other Study ID Numbers: KY2025-1003-02

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED
• IPD Plan Description: Patient information cannot be disclosed.

Drug and device information, study documents

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