Search

Disease-independent sources of biomarker variability include pre-analytical, analytical and biological variance. The aim of the present study was to evaluate whether the pre-analytical phase has any impact on the emerging heart disease TWEAK and HMGB1 protein markers and miRNA biomarkers, and whether peptidome profiling allows the identification of pre-analytical quality markers.An assessment was made of sample type (serum, EDTA-Plasma, Citrate-Plasma, ACD-plasma, Heparin-plasma), temperature of sample storage (room temperature or refrigerated), time of sample storage (0.5, 3, 6 and 9h) and centrifugation (one or two-step). Aliquots of all processed samples were immediately frozen (-80°C) before analysis. Proteins were assayed by ELISAs, miRNA expression profile by microarray and peptidome profiling by MALDI-TOF/MS.Temperature, time and centrifugation had no impact on TWEAK and HMGB1 results, which were significantly influenced by matrix type, TWEAK levels being significantly higher (F=194.7, p0.0001), and HMGB1 levels significantly lower (F=36.32, p0.0001) in serum than in any other plasma type. Unsuitable miRNA results were obtained using Heparin-plasma. Serum miRNA expression profiles depended mainly on temperature, while EDTA-plasma miRNA expression profiles were strongly affected by the centrifugation method used. MALDI-TOF/MS allowed the identification of seven features as indices of pre-analytical serum (m/z at 1206, 1350, 1865 and 2021) or EDTA-plasma (m/z 1897, 2740 and 2917) degradation.Serum and EDTA-plasma allow the analysis of both proteins and miRNA emerging biomarkers of heart diseases. Refrigerated storage prevents an altered miRNA expression profile also in cases of a prolonged time-interval between blood drawing and processing.

Validation of biomarkers is essential to advance the translational process to clinical application. Although there exists an increasing number of reports on small non-coding RNAs (microRNAs) as minimally-invasive markers from blood, serum or plasma, just a limited number is verified in follow-up studies. We used qRT-PCR to evaluate a known miRNA signature measured from blood that allowed for separation between patients with non-small cell lung cancer (NSCLC), COPD and healthy controls. From the data of our previous microarray studies we selected a panel of 235 miRNAs related to lung cancer and COPD. We observed a high concordance between the AUC values of our initial microarray screening and the qRT-PCR data (correlation of 0.704, p < 10−16). Overall, 90.3% of markers were successfully validated. Among the top markers that were concordant between both studies we found hsa-miR-20b-5p, hsa-miR-20a-5p, hsa-miR-17-5p, and hsa-miR-106a-5p. The qRT-PCR analysis also confirmed that non-small cell lung cancer patients could be accurately differentiated from unaffected controls: a subset of five markers was sufficient to separate NSCLC patients from unaffected controls with accuracy of 94.5% (specificity and sensitivity of 98% and 91%). Beyond differentiation from controls, we also succeeded in separating NSCLC patients from patients with COPD. MiRNAs that were identified as relevant for the separation between lung cancer and COPD by both qRT-PCR and the array-based studies included hsa-miR-26a-5p, hsa-miR-328-3p and hsa-miR-1224-3p. Although for differentiation between NSCLC patients from COPD patients more markers were required, still high accuracy rates were obtained (5 markers: 78.8%; 10 markers: 83.9%; 50 markers: 87.6%).