2022 World Conference on Lung Cancer (ePosters)-EP01.01-011. Utility of Arm-level cfDNA Fragment Size Distribution in the Early Detection of Lung Cancer and Pan Cancer

EP01.01-011. Utility of Arm-level cfDNA Fragment Size Distribution in the Early Detection of Lung Cancer and Pan Cancer

Pdf Summary

In this study, researchers explored the utility of arm-level circulating cell-free DNA (cfDNA) fragment size distribution (ARM-FSD) in the early detection of lung cancer and pan-cancers. They compared the performance of ARM-FSD to the commonly used window-level fragment size summary (WINDOW-FSS) approach.<br /><br />The researchers trained and validated the ARM-FSD model in a local lung cancer cohort consisting of 56 lung cancer patients and 106 healthy volunteers. They also tested the model in two independent external pan-cancer cohorts, a liver cancer cohort, and the local lung cancer cohort. They assessed the value of the two features in pan-cancer detection using an online pan-cancer cohort.<br /><br />The results showed that ARM-FSD outperformed WINDOW-FSS in predicting lung cancer. The ARM-FSD model achieved a high area under the ROC curve (AUC), indicating its strong performance. The performance of the model was validated in external lung cancer cohorts, further validating its robustness.<br /><br />In terms of pan-cancer detection, ARM-FSD also showed promising results. It achieved high AUCs in cross-validation using a pan-cancer training cohort and was validated in an external pan-cancer cohort.<br /><br />Overall, this study suggests that ARM-FSD is a robust and generalizable biomarker for the early detection of lung cancer and pan-cancers. It outperforms the WINDOW-FSS approach and shows potential for clinical application. However, further research and validation are needed to confirm these findings.

Asset Subtitle

Lifeng Wang, China

Meta Tag

Speaker Lifeng Wang, China

Topic Early Detection and Screening - Biomarkers

Keywords

arm-level circulating cell-free DNA

cfDNA fragment size distribution

early detection

lung cancer

pan-cancers

window-level fragment size summary

model training

model validation

area under the ROC curve

pan-cancer detection