1308P - Diameter threshold for lung nodules at low-dose CT screening: Chinese population screening experience

Background

High false positive rates in LDCT screening are an important concern. The study aims to explore the diagnostic performance of different diameter thresholds for various lung nodules to reduce unnecessary follow-up LDCT scans and the waste of medical resources.

Methods

Participants were recruited from the "Guangzhou Lung-Care Project" at the First Affiliated Hospital of Guangzhou Medical University in China between 2017 and 2021. The average transverse diameter of lung nodules was measured by semi-automatic segmentation. Positive outcomes were divided into two groups: outcome 1-lung cancer diagnosis with AAH/AIS, and outcome 2-lung cancer diagnosis without AAH/AIS. We calculated the area under the receiver-operating characteristic curves (AUCs), sensitivity, specificity, positive prediction rate(ppv) and negative prediction rate (NPV) to evaluate the diagnostic performance of lung cancer under each threshold between 5mm to 10mm.

Results

In the first year after LDCT screening, 126 cases of lung cancer (including 13 cases of AAH/AIS) were diagnosed among the participants of 11,705 (5,425 men and 6,280 women; median age 59). Increase the average transverse diameter threshold from 5 to 7mm in all lung nodules (solid, part-solid, non-solid), and gradually increase the AUC in outcome 1 (0.766 to 0.825) and outcome 2 (0.78 to 0.839). From 7 to 10mm, the AUC in outcome 1 (0.825 to 0.764) and outcome 2 (0.839 to 0.778) is gradually reduced. In the outcome 2, when the diameter threshold was increased from 5mm to 8mm, the specificity (60.06% to 90.82%) and ppv (6.57% to 19.02%) increased significantly, the sensitivity was reduced (94.02% to 76.07%). When the diameter threshold was increased from 8mm to 10mm, the sensitivity was significantly reduced (76.07% to 60.68%) and specificity (90.82% to 94.88%) and ppv (19.02% to 25.13%) increased slightly. There is no obvious difference between outcome 1 and outcome 2 in the above indicators.

Conclusions

Our study demonstrates that setting a diameter threshold of 8mm for LDCT screening of lung nodules achieves better overall diagnostic performance, reducing false positives and unnecessary follow-up scans while minimizing medical resource waste.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

Wenhua Liang.

Funding

The First Affiliated Hospital of Guangzhou Medical University; China State Key Laboratory of Respiratory Disease & National Clinical Research Center for Respiratory Disease.

Disclosure

All authors have declared no conflicts of interest.