2023 World Conference on Lung Cancer (Posters)-P2.07. Checkpoint Inhibitor Pneumonitis Model for NSCLC based onPreexistingLung Ground Glass Opacities: Development and Validation

P2.07. Checkpoint Inhibitor Pneumonitis Model for NSCLC based onPreexistingLung Ground Glass Opacities: Development and Validation - PDF(Slides)

Back to course

Pdf Summary

This study aimed to identify high-risk factors for checkpoint inhibitor pneumonitis (CIP) among non-small cell lung cancer (NSCLC) patients using quantitative imaging and computational analysis. The study included a retrospective cohort of NSCLC patients who received immune checkpoint inhibitor (ICI) therapy. A deep learning algorithm labeled and measured the volume of interstitial lesions on pretreatment CT scans. Two prediction models were developed to estimate the probability of CIP grade 2 or grade 3. The models were validated and their performance was quantified. <br /><br />The results showed that age, histology (non-squamous/squamous), and ground glass opacity (GGO) percentage in the whole lung were significant predictors of CIP grade 2. Histology and GGO percentage in the right lower lung were significant predictors of CIP grade 3. Assessing the volume and distribution of GGOs on pretreatment CT scans could help monitor and manage the risk of CIP in NSCLC patients receiving ICI treatment. Cox proportional hazard models were also used to examine predictors of progression-free survival (PFS). <br /><br />The study concluded that the use of quantitative imaging and computational analysis could help identify high-risk factors for CIP in NSCLC patients receiving ICI therapy. This approach could assist in monitoring and managing the risk of CIP and potentially improve patient outcomes. Further research is needed to validate these findings and incorporate them into clinical practice.

Asset Subtitle

Xinyue Wang

Meta Tag

Speaker Xinyue Wang

Topic Metastatic NSCLC: Immunotherapy - Retrospective

Keywords

checkpoint inhibitor pneumonitis

non-small cell lung cancer

NSCLC

quantitative imaging

computational analysis

retrospective cohort

immune checkpoint inhibitor therapy

deep learning algorithm

interstitial lesions

prediction models