Deep Learning Signature for Predicting Complete Pathological Response to Neoadjuvant Chemoimmunotherapy in Non-small Cell Lung Cancer

Non-small Cell Lung Cancer Clinical Trial

Official title:

An Integration of a Computed Tomography/Positron Emission Tomography/Whole Slide Image (CT/PET/WSI) Based Deep Learning Signature for Predicting Complete Pathological Response to Neoadjuvant Chemoimmunotherapy in Non-small Cell Lung Cancer: A Multicenter Study

Clinical Trial Details — Status: Recruiting

Administrative data

• NCT number: NCT05925751
• Other study ID #: DLCPR
• Status: Recruiting
• Start date: May 1, 2023
• Est. completion date: October 31, 2023

Study information

• Verified date: June 2023
• Source: Shanghai Pulmonary Hospital, Shanghai, China
• Contact: n/a
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

The purpose of this study is to evaluate the performance of a CT/PET/ WSI-based deep learning signature for predicting complete pathological response to neoadjuvant chemoimmunotherapy in non-small cell lung cancer

Recruitment information / eligibility

• Status: Recruiting
• Enrollment: 100
• Est. completion date: October 31, 2023
• Est. primary completion date: October 31, 2023
• Gender: All
• Age group: 20 Years to 75 Years

Eligibility

Inclusion Criteria:

1. Age ranging from 20-75 years;

2. Patients who underwent curative surgery after neoadjuvant chemoimmunotherapy for NSCLC;

3. Obtained written informed consent.

Exclusion Criteria:

1. Missing image data;

2. Pathological N3 disease.

Related Conditions & MeSH terms

• Carcinoma, Non-Small-Cell Lung
• Lung Neoplasms
• Non-small Cell Lung Cancer

Intervention

Diagnostic Test:
• CT/PET/WSI-based Deep Learning Signature
CT/PET/WSI-based Deep Learning Signature for Predicting Complete Pathological Response to Neoadjuvant Chemoimmunotherapy in Non-small Cell Lung Cancer

Locations

Country	Name	City	State
China	The First Affiliated Hospital of Nanchang University	Nanchang	Jiangxi
China	Ningbo HwaMei Hospital	Ningbo	Zhejiang
China	Affiliated Hospital of Zunyi Medical University	Zunyi	Guizhou

Sponsors (4)

Lead Sponsor	Collaborator
Shanghai Pulmonary Hospital, Shanghai, China	Ningbo HwaMei Hospital, Zhejiang, China, The First Affiliated Hospital of Nanchang University, Jiangxi, China, Zunyi Medical College

Country where clinical trial is conducted

China, 

Outcome

Type	Measure	Description	Time frame	Safety issue
Other	Specificity	The specificity of the deep learning model in predicting complete pathological response. CPR was defined as no residual tumor in both resected primary tumor and lymph nodes. Patients with non-small cell lung cancer receiving neoadjuvant chemoimmunotherapy will achieve either CPR or non-CPR, which can be confirmed by pathological examination after surgical resection. And the model will output the predictive value (CPR/non-CPR) for each patient receiving neoadjuvant chemoimmunotherapy.	2023.5.1-2023.10.31
Other	Positive predictive value	The positive predictive value of the deep learning model in predicting complete pathological response. CPR was defined as no residual tumor in both resected primary tumor and lymph nodes. Patients with non-small cell lung cancer receiving neoadjuvant chemoimmunotherapy will achieve either CPR or non-CPR, which can be confirmed by pathological examination after surgical resection. And the model will output the predictive value (CPR/non-CPR) for each patient receiving neoadjuvant chemoimmunotherapy.	2023.5.1-2023.10.31
Other	Negative predictive value	The negative predictive value of the deep learning model in predicting complete pathological response. CPR was defined as no residual tumor in both resected primary tumor and lymph nodes. Patients with non-small cell lung cancer receiving neoadjuvant chemoimmunotherapy will achieve either CPR or non-CPR, which can be confirmed by pathological examination after surgical resection. And the model will output the predictive value (CPR/non-CPR) for each patient receiving neoadjuvant chemoimmunotherapy.	2023.5.1-2023.10.31
Other	Accuracy	The accuracy of the deep learning model in predicting complete pathological response. CPR was defined as no residual tumor in both resected primary tumor and lymph nodes. Patients with non-small cell lung cancer receiving neoadjuvant chemoimmunotherapy will achieve either CPR or non-CPR, which can be confirmed by pathological examination after surgical resection. And the model will output the predictive value (CPR/non-CPR) for each patient receiving neoadjuvant chemoimmunotherapy.	2023.5.1-2023.10.31
Primary	Area under the receiver operating characteristic curve	The area under the receiver operating characteristic curve (ROC) of the deep learning model in predicting complete pathological response (CPR). CPR was defined as no residual tumor in both resected primary tumor and lymph nodes. Patients with non-small cell lung cancer receiving neoadjuvant chemoimmunotherapy will achieve either CPR or non-CPR, which can be confirmed by pathological examination after surgical resection. And the model will output the predictive value (CPR/non-CPR) for each patient receiving neoadjuvant chemoimmunotherapy.	2023.5.1-2023.10.31
Secondary	Sensitivity	The sensitivity of the deep learning model in predicting complete pathological response. CPR was defined as no residual tumor in both resected primary tumor and lymph nodes. Patients with non-small cell lung cancer receiving neoadjuvant chemoimmunotherapy will achieve either CPR or non-CPR, which can be confirmed by pathological examination after surgical resection. And the model will output the predictive value (CPR/non-CPR) for each patient receiving neoadjuvant chemoimmunotherapy.	2023.5.1-2023.10.31