Early non-invasive detection of stage Ia lung cancer using circulating tumor DNA methylation haplotypes (ctmDNA) in plasma.

e13159 Background: Lung cancer is one of the most common forms of cancer and is responsible for approximately 1.8 million deaths per year worldwide. The current 5-year survival rate for lung cancer is only 18%; however, this improves to 56% if the cancer is detected early. While low-dose CT scans have shown promise as an early detection method, only 16% of lung cancer is currently detected at an early stage. We therefore set out to develop a non-invasive blood-based screening assay to identify lung cancer at an early stage using ctmDNA (circulating tumor methylated DNA haplotypes). Methods: Blood samples were prospectively collected from two partner hospitals from 325 healthy individuals and 116 individuals diagnosed with lung nodules by low-dose CT scan in EDTA or Streck BCT tubes and immediately separated into plasma. Patients with lung nodules that appeared cancerous then underwent surgical resection, and cancer diagnosis was confirmed via pathology. Patients were matched between healthy and cancer groups by age, sex, and smoking status. Plasma samples were processed using the Singlera Genomics LUNA assay, a targeted bisulfite sequencing method which identifies methylation haplotype patterns related to early-stage lung cancer. 241 samples were used to train a classification model based on pathology results, and 200 samples were used as a test set to validate the model. Results: In the independent test set, the LUNA assay was able to show a sensitivity of 91.9% to detect early-stage lung cancer with a specificity of 93.3% in healthy patients. Even patients with stage Ia lung cancer were readily detected by the LUNA assay (sensitivity of 91.7%). Conclusions: We have shown that ctmDNA can be utilized to non-invasively screen for early-stage lung cancer with high sensitivity and specificity, paving the way for a blood-based lung cancer early screening assay.