2023 World Conference on Lung Cancer (Posters)-P2.05. An Artificial Neural Network System to Predict the Immune-related Adverse Events Based on the RNA Data of ORIENT-3 Study

P2.05. An Artificial Neural Network System to Predict the Immune-related Adverse Events Based on the RNA Data of ORIENT-3 Study - PDF(Abstract)

Back to course

Pdf Summary

This study presented the use of an artificial neural network system to predict immune-related adverse events (irAEs) in patients with squamous-cell non-small-cell lung cancer (sqNSCLC) treated with sintilimab. The researchers utilized the RNA data from the ORIENT-3 study, which was a phase 3 trial that included patients with advanced sqNSCLC. The RNA data was normalized and converted logarithmically for analysis.<br /><br />The study randomly divided the patients treated with sintilimab into a training cohort (70%) and a test cohort (30%). Univariate logistic regression and receiver operating characteristic curve (ROC) analysis were performed in the training cohort to select the most significant genes associated with irAEs. The top 15 genes were used to construct artificial neural network (ANN) models.<br /><br />The researchers conducted 300 rounds of three-fold cross-validation to test the accuracy of the ANN models. The final predicted value of the ANN system was obtained as the weighted average of output values from all ANN models. The accuracy of the ANN system was then assessed in the test cohort.<br /><br />Results showed that among the 61 patients treated with sintilimab, 35 developed irAEs. The ANN models constructed using gene selection revealed a high proportion (95.22%) with weights indicating good performance. The AUC in the training cohort was 1.00, while in the test cohort it was 0.84.<br /><br />In conclusion, this study demonstrated that the ANN system has the potential to predict irAEs in patients with sqNSCLC treated with sintilimab. This could have implications for identifying patients at risk for adverse events and guiding treatment decisions.

Asset Subtitle

Tongji Xie

Meta Tag

Speaker Tongji Xie

Topic Metastatic NSCLC: Immunotherapy - Biomarker

Keywords

artificial neural network system

immune-related adverse events

squamous-cell non-small-cell lung cancer

sintilimab

RNA data

ORIENT-3 study

logarithmic conversion

gene selection

accuracy assessment

adverse event risk