2023 World Conference on Lung Cancer (Posters)-EP07.04. Real-World Biomarker Testing and Treatment Patterns Identified Using Machine Learning in Early Non-small Cell Lung Cancer

EP07.04. Real-World Biomarker Testing and Treatment Patterns Identified Using Machine Learning in Early Non-small Cell Lung Cancer - PDF(Slides)

Back to course

Pdf Summary

This document is a summary of a poster presentation at the International Association for the Study of Lung Cancer – WCLC Annual Meeting in 2023. The study aimed to evaluate biomarker testing and treatment patterns in patients with early non-small cell lung cancer (eNSCLC) using a machine learning (ML)-extracted electronic health record (EHR) dataset.<br /><br />The study population consisted of patients with stage I-IIIB eNSCLC who had surgery between January 2019 and August 2022. The ML-extracted dataset included 29,314 patients, while the human-abstracted dataset included 2,316 patients. The characteristics of the patients in both datasets were similar.<br /><br />The study found that biomarker testing rates improved over time, but there were still gaps in testing for PD-L1, ALK, and EGFR. Many patients resected in the latest quarter did not receive testing for these biomarkers. Among the resected patients who did receive biomarker testing, the majority had PD-L1 testing (69%), while EGFR and ALK testing rates were lower.<br /><br />In terms of treatment patterns, adjuvant treatment rates were generally higher than neoadjuvant or neoadjuvant-adjuvant treatment rates. The study also evaluated the use of atezolizumab, the only adjuvant immunotherapy approved during the study period, and found that 55% of eligible patients received it.<br /><br />The authors concluded that while biomarker testing rates have increased over time, there are still gaps in biomarker-informed treatment decisions. Further investigation is needed to understand the reasons for these gaps, such as cost or patient preference. The study also found that the ML-extracted dataset provided reliable results and allowed for larger sample sizes compared to the human-abstracted dataset.<br /><br />Overall, this study provides real-world evidence on biomarker testing and treatment patterns in patients with eNSCLC, highlighting the need for further research to improve the implementation of biomarker-informed treatments.

Asset Subtitle

Jay Lee

Meta Tag

Speaker Jay Lee

Topic Early-Stage NSCLC: Progress in Pathology

Keywords

biomarker testing

treatment patterns

non-small cell lung cancer

machine learning

electronic health record

PD-L1

ALK

EGFR

adjuvant treatment

atezolizumab