Clinical Lab Manager - News, Editorial and Products for the Clinical Laboratory
Illustration of an aerial view of several tiny people standing in the shape of human lungs.
The Lung Cancer Artificial Intelligence Detector (LCAID) combines mass spectrometry-based lipid testing with AI.

Diagnosing Early-Stage Lung Cancer with Blood Samples and AI

The test shows promise for an early and widespread clinical detection method for lung cancer

A new blood test may now enable early-stage lung cancer detection, according to a recent study published in Science Translational Medicine. This test, the Lung Cancer Artificial Intelligence Detector (LCAID) v2.0, combines mass spectrometry-based lipid testing with artificial intelligence.

Lung cancer is the leading cause of cancer death, partially because it often goes undetected until later stages, decreasing chances of survival. This has spurred the search for reliable blood-based tests, such as LCAID v2.0, that could detect early stages of lung cancer in large populations.

To develop LCAID v2.0, the research team at the Peking University Peoples’ Hospital began by profiling metabolism-related transcriptional features from five patients awaiting treatment for non-small cell lung cancer (NSCLC) tumors. By using 10x Genomics’ droplet-based technology for single-cell RNA sequencing, they found cancer-associated dysregulation in lipid metabolism.

This led the researchers to perform a series of untargeted lipidomic analyses on blood plasma samples in an exploratory cohort of 311 participants with or without lung cancer using high-performance liquid chromatography-mass spectrometry (HPLC-MS) profiling. By applying support vector machine algorithm-based selection, they narrowed down nine lipids that were dysregulated in early-stage lung tumors. 

The researchers validated their approach on samples collected from a low-dose computed tomography (CT) lung cancer screening program and found that LCAID v2.0 was more than 90 percent sensitive and 92 percent specific for detecting early-stage lung cancer, demonstrating that it may be effective for large-scale screening of at-risk populations.