Deep Learning Signature for Predicting Aggressive Histological Pattern in Resected Non-small Cell Lung Cancer

Positron Emission Tomography/ Computed Tomography (PET/CT) Based Deep Learning Signature for Predicting Aggressive Histological Pattern in Resected Non-small Cell Lung Cancer

The purpose of this study is to evaluate the performance of a PET/ CT-based deep learning signature for predicting aggressive histological pattern in resected non-small cell lung cancer based on a multicenter prospective cohort.

Study Overview

• Status: Recruiting
• Conditions: Non-small Cell Lung Cancer; Spread Through Air Space; Visceral Pleural Invasion; Lymphovascular Invasion
• Intervention / Treatment: Diagnostic test: PET/CT-based Deep Learning Signature

• Study Type: Observational
• Enrollment (Estimated): 1500

Contacts and Locations

Study Locations

• China
  • Guizhou
    • Zunyi, Guizhou, China
      • Recruiting
      • Affiliated Hospital of Zunyi Medical University
      • Contact: Yongxiang Song, Dr; Phone Number: 15505177258; Email: zhong961008@163.com
  • Jiangxi
    • Nanchang, Jiangxi, China
      • Recruiting
      • The First Affiliated Hospital Of Nanchang University
      • Contact: Bentong Yu, Dr; Phone Number: 021-65115006; Email: 1151697503@qq.com
  • Zhejiang
    • Ningbo, Zhejiang, China
      • Recruiting
      • Ningbo HwaMei Hospital
      • Contact: Minglei Yang, Dr; Phone Number: 021-65115006; Email: almondjj@163.com

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

Resected Stage I-III Non-small Cell Lung Cancer

Description

Inclusion Criteria:

(1) Participants scheduled for surgery for radiological finding of pulmonary lesions from the preoperative thin-section CT scans; (2) Pathological confirmation of primary NSCLC; (3) Age ranging from 20-75 years; (4) Obtained written informed consent.

Exclusion Criteria:

(1) Multiple lung lesions; (2) Poor quality of PET-CT images; (3) Participants with incomplete clinical information; (4) Participants who have received neoadjuvant therapy.

Study Plan

How is the study designed?

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Area under the receiver operating characteristic curve	The area under the receiver operating characteristic curve (ROC) of the deep learning model in predicting the presence or absence of the aggressive histological pattern. The aggressive histological pattern includes spread through air space (STAS), visceral pleural invasion (VPI), and lymphovascular invasion (LVI). And the model will output all predictive values (presence or absence) of the three kinds of aggressive histological patterns.	2023.5.1-2023.10.31

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Sensitivity	The sensitivity of the deep learning model in predicting the presence or absence of the aggressive histological pattern. The aggressive histological pattern includes spread through air space (STAS), visceral pleural invasion (VPI), and lymphovascular invasion (LVI). And the model will output all predictive values (presence or absence) of the three kinds of aggressive histological patterns.	2023.5.1-2023.10.31

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Specificity	The specificity of the deep learning model in predicting the presence or absence of the aggressive histological pattern. The aggressive histological pattern includes spread through air space (STAS), visceral pleural invasion (VPI), and lymphovascular invasion (LVI). And the model will output all predictive values (presence or absence) of the three kinds of aggressive histological patterns.	2023.5.1-2023.10.31
Positive predictive value	The positive predictive value of the deep learning model in predicting the presence or absence of the aggressive histological pattern. The aggressive histological pattern includes spread through air space (STAS), visceral pleural invasion (VPI), and lymphovascular invasion (LVI). And the model will output all predictive values (presence or absence) of the three kinds of aggressive histological patterns.	2023.5.1-2023.10.31
Negative predictive value	The negative predictive value of the deep learning model in predicting the presence or absence of the aggressive histological pattern. The aggressive histological pattern includes spread through air space (STAS), visceral pleural invasion (VPI), and lymphovascular invasion (LVI). And the model will output all predictive values (presence or absence) of the three kinds of aggressive histological patterns.	2023.5.1-2023.10.31
Accuracy	The accuracy of the deep learning model in predicting the presence or absence of the aggressive histological pattern. The aggressive histological pattern includes spread through air space (STAS), visceral pleural invasion (VPI), and lymphovascular invasion (LVI). And the model will output all predictive values (presence or absence) of the three kinds of aggressive histological patterns.	2023.5.1-2023.10.31

Collaborators and Investigators

• Sponsor: Shanghai Pulmonary Hospital, Shanghai, China
• Collaborators: Zunyi Medical College; Ningbo HwaMei Hospital, Zhejiang, China; The First Affiliated Hospital of Nanchang University, Jiangxi, China

Study record dates

Study Major Dates

• Study Start (Actual): May 1, 2023
• Primary Completion (Estimated): October 31, 2023
• Study Completion (Estimated): October 31, 2023

Study Registration Dates

• First Submitted: May 12, 2023
• First Submitted That Met QC Criteria: June 27, 2023
• First Posted (Actual): June 29, 2023

Study Record Updates

• Last Update Posted (Actual): June 29, 2023
• Last Update Submitted That Met QC Criteria: June 27, 2023
• Last Verified: June 1, 2023

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Respiratory Tract Diseases; Neoplasms; Lung Diseases; Neoplasms by Site; Respiratory Tract Neoplasms; Thoracic Neoplasms; Carcinoma, Bronchogenic; Bronchial Neoplasms; Lung Neoplasms; Carcinoma, Non-Small-Cell Lung
• Other Study ID Numbers: DLAHP

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