Abstract 294P
Background
Low-dose CT screening for high-risk individuals has demonstrated improvements in lung cancer-specific mortality. However, significant challenges remain, including low specificity and poor adherence rates, which hinder the widespread implementation of such tests. In contrast, liquid biopsy has emerged as a promising non-invasive tool for diagnosis and monitoring.
Methods
We have developed an ultra-sensitive and cost-effective blood test for lung cancer diagnosis and monitoring. The process involves extracting cell-free DNA from the blood, fluorescently labeling epigenetic marks, and hybridizing them onto a custom, targeted microarray chip. The microarray spots illuminate based on the epigenetic content of the captured DNA, generating unique signatures for different cancer types and stages.
Results
A total of 103 individuals were tested, comprising 51 diagnosed with lung cancer and 52 healthy controls. Hundreds of genomic loci that distinguish non-small cell lung cancer (NSCLC) patients from healthy individuals were identified using a subset of 43 samples. The remaining 60 samples were blindly tested, yielding a sensitivity of 93.1% and a specificity of 93.5%, with an Area Under the Curve (AUC) of 0.943. Furthermore, the test successfully differentiated between adenocarcinoma (N=40) and squamous cell carcinoma (N=8), achieving a sensitivity of 82.5%, specificity of 87.5%, and an AUC of 0.922.
Conclusions
These findings highlight the effectiveness of the blood test, based on epigenetic signatures, in detecting and classifying lung cancer samples. Prospective trials based on these results will be conducted to validate these results.
Funding
Jaxbio Technologies LTD.
Disclosure
A. Agbarya: Financial Interests, Personal, Advisory Board: Jaxbio Technologies LTD. S. Zikrin, Y. Michaeli, N. Gilat, L. Itelson: Financial Interests, Personal, Officer: Jaxbio Technologies LTD. Y. Ebenstein: Financial Interests, Personal, Leadership Role: Jaxbio Technologies LTD. All other authors have declared no conflicts of interest.