APSR2026-968

 
Late-Breaking
Submitted
Oral Presentation
AO31-006
 
 
https://storage.unitedwebnetwork.com/files/1285/1243179-46903-APSR2026-Fei Yu AO31-006_440291.pptx
Format: PPTX Only
 
Format: PDF Only
08. Lung Cancer
3

No more than 15 authors can be listed.
Please ensure the authors are listed in the right order.
If there are multiple affiliations, please use the "+" button next to "Select Country/Region" to add additional affiliation fields.

Fei Yu * Tianjin Medical University General Hospital Department of Oncology Tianjin China Shandong Medical and Pharmaceutical University Hospital Department of General Practice Binzhou China
Diansheng Zhong - Tianjin Medical University General Hospital Department of Oncology Tianjin China
Linlin Zhang - Tianjin Medical University General Hospital Department of Oncology Tianjin China
 
 
 
 
 
 
 
 
 
 
 
 
Presenting Author Details
Fei
Yu
feiyu1984@tmu.edu.cn
+8618605432689
41-50
China
Original Research
Development and internal validation of a clinical-immunological risk model for immune checkpoint inhibitor-related pneumonitis in patients with lung cancer
Maximum 25 words
Abstract Content
Immune checkpoint inhibitor-related pneumonitis (CIP) is a clinically important and potentially severe adverse event in patients with lung cancer receiving immunotherapy. Early identification of high-risk patients is essential for individualized surveillance and treatment safety. This study aimed to develop and internally validate a clinical-immunological prediction model for CIP in patients with lung cancer
We retrospectively included 619 patients with lung cancer treated with immune checkpoint inhibitors, among whom 75 developed CIP. After handling missing data by imputation, candidate predictors covering baseline immune-inflammatory indices, tumor characteristics, treatment-related factors, and pre-existing pulmonary conditions were screened using univariable analysis, collinearity assessment, and multivariable logistic regression. The final model incorporated baseline cluster of differentiation 3-positive (CD3+) T-cell count, lactate dehydrogenase, lymphocyte-to-monocyte ratio, histology, other immune-related adverse events, treatment regimen, programmed death-ligand 1 expression, and pre-existing lung disease. Model performance was assessed using receiver operating characteristic analysis, bootstrap internal validation, and risk stratification.
The incidence of CIP was 12.1% (75/619). Pre-existing lung disease and other immune-related adverse events were key predictors of increased CIP risk, while treatment regimen, programmed death-ligand 1 expression, and baseline immune-inflammatory markers provided additional predictive information. The model showed good discrimination in the imputed dataset, with an area under the receiver operating characteristic curve (AUC) of 0.8460; the bootstrap-corrected AUC was 0.8129. Using an optimal predicted-probability threshold of 0.145, the model separated patients into high- and low-risk groups, with CIP incidence of 36.9% and 3.5%, respectively.
A clinical-immunological prediction model integrating immune-inflammatory, tumor-related, treatment-related, and pulmonary factors identified patients with lung cancer at increased risk of CIP. This model may support early risk stratification, personalized monitoring, and safer immunotherapy management.
 
 
 
 
 
No
No
 
JPG or JPEG or PNG format ONLY
Lung cancer ,Immune checkpoint inhibitor-related pneumonitis ,Risk prediction
maximum 3 keywords and no abbreviations, separated by commas
The authors have no conflicts of interest to disclose
276