Background Sensitive and specific biomarkers for identifying early stage of non-small cell lung cancer (NSCLC) are urgently needed to improve the therapeutic outcome and reduce the mortality. Small non-coding microRNAs (miRNAs) are key components of cancer development and are considered as potential biomarkers for cancer diagnosis and for monitoring treatment. The aim of this study was to determine whether aberrant miRNA expression can be used as a marker in peripheral blood mononuclear cells (PBMC) for the diagnosis of NSCLC. Methods The levels of two mature miRNAs (miR-143 and miR-150) were detected by probe-based stem-loop quantitative reverse-transcriptase PCR (RT-qPCR) in PBMC of 64 patients with NSCLC and 26 healthy individuals, and the relationship between miR-143 and miR-150 levels and clinical and pathological factors was explored. Results All endogenous miRNAs were present in peripheral blood in a remarkably stable form and detected by RT-qPCR. MiR-143 expression in the PBMC specimens was significantly lower in NSCLC patients than in healthy individuals (P<0.0001). MiR-150 expression in the PBMC specimens was not significantly different between NSCLC patients and healthy individuals (P=0.260). MiR-150 expression was significantly higher in lung adenocarcinoma patients than in healthy individuals (P=0.001). There was a very strong difference in the expression level of miR-150 between lung adenocarcinoma patients and lung squamous cell carcinoma patients (P<0.0001). In receiver operating characteristic curve (ROC) analysis, low expression of miR-143 showed the area under the ROC (AUC) of 0.885 for distinguishing cancer patients from healthy subjects. High expression of miR-150 had an AUC of 0.834 for distinguishing lung adenocarcinoma patients from healthy subjects. High expression of miR-150 had an AUC of 0.951 for distinguishing lung adenocarcinoma from lung squamous cell carcinoma. The miR-150 level was significantly associated with distant metastasis (P=0.014). Conclusions It is indicated that there is a potential for using miR-143 as a novel diagnostic biomarker for NSCLC. Moreover, miR-150 can be a highly accurate marker for differentiating adenocarcinoma from squamous cell carcinoma.