A novel serum m⁷G-harboring microRNA signature for cancer detection

Background: Emerging evidence points to the exceptional importance and value of m⁷G alteration in the diagnosis and prognosis of cancers. Nonetheless, a biomarker for precise screening of various cancer types has not yet been developed based on serum m⁷G-harboring miRNAs. Methods: A total of 20,702 serum samples, covering 12 cancer types and consisting of 7,768 cancer samples and 12,934 cancer-free samples were used in this study. A m⁷G target miRNA diagnostic signature (m⁷G-miRDS) was established through the least absolute shrinkage and selection operator (LASSO) analyses in a training dataset (n = 10,351), and validated in a validation dataset (n = 10,351). Results: The m⁷G-miRDS model, a 12 m⁷G-target-miRNAs signature, demonstrated high accuracy and was qualified for cancer detection. In the training and validation cohort, the area under the curve (AUC) reached 0.974 (95% CI 0.971-0.977) and 0.972 (95% CI 0.969-0.975), respectively. The m⁷G-miRDS showed superior sensitivity in each cancer type and had a satisfactory AUC in identifying bladder cancer, lung cancer and esophageal cancer. Additionally, the diagnostic performance of m⁷G-miRDS was not interfered by the gender, age and benign disease. Conclusion: Our results greatly extended the value of serum circulating miRNAs and m⁷G in cancer detection, and provided a new direction and strategy for the development of novel biomarkers with high accuracy, low cost and less invasiveness for mass cancer screening, such as ncRNA modification.