medwireNews: Researchers have identified micro (mi)RNA signatures that can potentially distinguish symptomatic patients with lung cancer from healthy people as well as those with other lung and non-lung diseases.
Andreas Keller (Saarland University, Saarbrücken, Germany) and colleagues say their findings “suggest that the identified patterns of circulating microRNAs may enable them to be used as biomarkers in a liquid biopsy to complement imaging tests, sputum cytology, and biopsies.”
Keller and team used machine-learning methods to create genome-wide miRNA profiles from the blood samples of 3046 individuals who were participants of the TREND cohort study and the COSYCONET case–control study.
Of these, 606 patients had symptomatic non-small-cell lung cancer (NSCLC) or small-cell lung cancer (SCLC), 593 had nontumor lung diseases such as chronic obstructive pulmonary disease, 883 had diseases not affecting the lung, and 964 were healthy controls.
The researchers split the study participants into two equal-sized training and validation sets that were matched for age, sex, and smoking status. The training set was used to identify the miRNAs with the most statistically significant differential expression between individuals with versus without lung cancer.
Keller and co-investigators found that a 15-miRNA signature gave the best discrimination between patients with lung cancer and all other individuals included in the study. When this signature was tested in the validation set, it identified blood samples from patients with lung cancer with an accuracy of 91.4%, a sensitivity of 82.8%, and a specificity of 93.5%.
Next, the team used a 14-miRNA signature from the training set to distinguish individuals in the validation set who had lung cancer from those with nontumor lung diseases. In this case, the accuracy was 92.5%, while the sensitivity and specificity were 96.4% and 88.6%, respectively.
Finally, Keller and team used another 14-miRNA signature from the training set to differentiate patients with stage I and II lung cancer in the validation set from individuals without lung cancer. The accuracy for the detection of these early-stage lung cancer patients was 95.9%, with a sensitivity of 76.3%, and a specificity of 97.5%.
The researchers also investigated whether they could distinguish between NSCLC and SCLC and identified a 9-miRNA signature that could do this with an accuracy of 84.6%, a sensitivity of 90.1%, and a specificity of 70.6% with the resampling approach.
Of note, none of the nine miRNAs from this comparison overlapped with those from the three other diagnostic scenarios, whereas there was “substantial overlap” among the other diagnostic scenarios.
Writing in JAMA Oncology, Keller et al conclude: “We believe this study presents a standardized approach that could be used to identify symptomatic patients with lung cancer based on blood-borne miRNA signatures.”
However, the researchers point out that several challenges still need to be overcome before miRNA profiling can be used clinically for the detection of lung cancer.
They say: “Prospective studies with large cohorts of patients with specific diseases, a format that is readily applicable in clinical in vitro diagnostic tests […], and an evaluation of the extent to which the miRNA signatures may complement imaging, sputum cytology, or biopsy are needed for clinical application of miRNA for diagnosis of lung cancer.”
medwireNews is an independent medical news service provided by Springer Healthcare. © 2020 Springer Healthcare part of the Springer Nature group