The blood-borne miRNA signature of lung cancer patients is independent of histology but influenced by metastases

Abstract

Objectives: In our previous studies we reported a panel of 24 miRNAs that allowed discrimination between blood of lung tumor patients independent of the histological subtype and blood of healthy controls with an accuracy of 95.4% [94.9%-95.9%]. Here, we now separately analyzed the miRNA expression in blood of non-small cell lung cancer (NSCLC), including squamous cell lung cancer and adenocarcinoma, and small cell lung cancer (SCLC) patients. Patients and methods: In total, we examined the expression levels of 1,205 miRNAs in blood samples from 20 patients from each of the three histological groups and determined differentially expressed miRNAs between histological subtypes and metastatic and non-metastatic lung cancer. We further determined the overlap of miRNAs expressed in each subgroup with the 24-miRNA signature of lung tumor patients. Results: Based on a raw p-value < 0.05, only 18 blood-borne miRNAs were differentially expressed between patients with adenocarcinoma and with squamous cell lung carcinoma, 11 miRNAs between adenocarcinoma and SCLC, and 2 between squamous cell lung carcinoma and SCLC. Likewise, the comparison based on a fold change of 1.5 did not reveal major differences of the blood-borne miRNA expression pattern between NSCLC and SCLC. In addition, we found a large overlap between the blood-borne miRNAs detected in the three histological subgroups and the previously described 24-miRNA signature that separates lung cancer patients form controls. We identified several miRNAs that allowed differentiating between metastatic and non-metastatic tumors both in blood of patients with adenocarcinoma and in blood of patients with SCLC. Conclusion: There is a common miRNA expression pattern in blood of lung cancer patients that does not allow a reliable further subtyping into NSCLC or SCLC, or into adenocarcinoma and squamous cell lung cancer. The previously described 24-miRNA signature for lung cancer appears not primarily dependent on histological subtypes. However, metastatic adenocarcinoma and SCLC can be predicted with 75% accuracy.