Study design

A total of 102 LSCC patients and 101 healthy controls were enrolled in our study. The experimental process was designed into four stages (Fig. 1). In the initial screening stage, thirty peripheral plasma samples from LSCC patients and 10 from NCs were randomly selected and pooled as three LSCC samples and one NC sample (10 samples were pooled as one pool sample) for miRNA microarrays via Exiqon‐miRCURY‐Ready‐to‐Use‐PCR‐Human‐panel‐I+II‐V1.M (Exiqon miRNA qPCR panel, Vedbaek, Denmark). Then, in order to verify reproducibility and reliability of the results which acquired from arrays, plasma samples from 32 LSCC samples and 31 NC samples were used to confirm the dysregulated miRNAs in training stage. In following testing stage, the selected miRNAs were refined from 55 LSCC samples and 55 NC samples. In external validation stage, 15 LSCC samples and 15 NC samples were enrolled to further assess the diagnostic performance of the selected miRNAs. These identified miRNAs were validated by 23 pairs of formalin‐fixed paraffin‐embedded (FFPE) LSCC tissue and matched adjacent normal tissue specimens from the same surgery patients. The potential form of miRNAs in the peripheral circulation was investigated via testing the expression of plasma exosomal miRNAs in 16 LSCC samples and 16 NC samples.

The flowchart of the experiment design. LSCC, lung squamous cell carcinoma; NC, normal control.

Venous blood samples of NC and LSCC samples were collected and placed in an ethylenediaminetetraacetic acid (EDTA)‐containing tube (Becton, Dickinson and Company). Peripheral plasma was isolated from whole blood within 12 hours after collection and then stored at −80°C for further analysis. Tissue specimens were stored in liquid nitrogen.