Purpose: To investigate a novel composite methodology of using targeted serum microRNAs (micro ribonucleic acid; miRNA) and urine metabolites for the accurate detection of early stage non-small cell lung cancer
(NSCLC).
Methods: Consecutively consenting NSCLC patients and matched control subjects were recruited to provide samples of serum for miRNA and/or urine for metabolite analyses. Serum miRNA levels were measured using quantitative real-time reverse-transcription with exogenous control, and the comparative delta cycle threshold (△CT) method was used to calculate relative miRNA expression of two targeted miRNAs (miR-21 and miR-223). The concentrations of six targeted urinary metabolites in patients and healthy controls were measured using proton nuclear magnetic resonance (1H NMR) spectroscopy. A composite methodology of using the 35 accruals with both serum and urine biomarkers was then established with binary logistic regression, receiver operating characteristic (ROC) models with or without artificial intelligence (AI).
Results: The ROC analysis of miRNA expression yielded a sensitivity of 96.4% and a specificity of 88.2% for the detection of early stage NSCLC, with area under the curve (AUC) = 0.91 (CI 95%: 0.80-1.0). Relative urinary concentrations of 4-methoxyphenylacetic acid (4MPLA) were significantly different between NSCLC and healthy control (p=0.008). The ROC analysis of 4MPLA yielded a sensitivity of 82.1% and a specificity of 88.2%, with AUC = 0.85. The composite process combining miRNA and metabolite expression demonstrated a sensitivity and specificity of nearly 100% and AUC=1.
Conclusions: A highly specific, sensitive and non-invasive detection method for NSCLC was developed. Pending validation, this can potentially improve the early detection and, hence, the treatment and survival outcomes of patients.
Genomic analysis of drug resistant small cell lung cancer cell lines by combining mRNA and miRNA expression profiling
