The lncRNAs Gas5, MALAT1 and SNHG8 as diagnostic biomarkers for epithelial malignant pleural mesothelioma in Egyptian patients

Abstract

Long noncoding RNAs have been shown to be involved in a myriad of physiological and pathological pathways. To date, malignant pleural mesothelioma (MPM) is considered an extremely aggressive cancer. One reason for this is the late diagnosis of the disease, which can occur within 30-40 years of asbestos exposure. There is an immense need for the development of new, sensitive, inexpensive and easy methods for the early detection of this disease other than invasive methods such as biopsy. The aim of this study was to determine the expression of circulating lncRNAs in mesothelioma patient plasma to identify potential biomarkers. Ten previously identified lncRNAs that were shown to be aberrantly expressed in mesothelioma tissues were selected as candidates for subsequent validation. The expression of the ten selected candidate lncRNAs was verified via quantitative PCR (qPCR) in human plasma samples from mesothelioma patients versus healthy controls. The expression levels of circulating GAS5, SNHG8 and MALAT1 were significantly greater in plasma samples from patients than in those from controls. The ROC analysis of both MALAT1 and SNHG8 revealed 88.89% sensitivity and 66.67% specificity. The sensitivity of these markers was greater than that of GAS5 (sensitivity 72.22% and specificity 66.67%). The regression model for GAS5 was statistically significant, while that for SNHG8 and MALAT1 was not significant due to the small sample size. The area under the curve (AUC) of the three ROC curves was acceptable and significant: 0.7519 for GAS5, 0.7352 for SNHG8 and 0.7185 for MALAT1. This finding confirmed their ability to be used as markers. The three lncRNAs were not affected by age, sex or smoking status. The three lncRNAs showed great potential as independent predictive diagnostic biomarkers. Although the prediction model for MALAT1 did not significantly differ, MALAT1 was significantly expressed in patients more than in controls (p = 0.0266), and the recorded sensitivity and specificity were greater than those of GAS5.

Keywords: Biomarker; Blood; Cancer; Circulating; Early detection; Marker; Mesothelioma; Plasma; Screening; lncRNA; qPCR.

Figure 1

Relative quantification (RQ) of MALAT1, GAS5 and SNHG8 expression compared to the rest of undetermined lncRNASs. Only MALAT1, GAS5 and SNHG8 were detected in the plasma of both groups. All three showed consistent upregulation in plasma of patients more than controls. The 2^−ΔΔCt method was used to calculate RQ as representation for expression levels. The RQ for MALAT1 was 7.16 in patients and 2.04 in controls. For GAS5, the RQ in patients was 3.44, while in controls 1.80. The SNHG8 RQs for patients and controls were 3.07 and 2.21, respectively. Statistical analysis of each expressed lncRNA was performed. All three were found to be statistically significant. *Indicates p < 0.05. Unpaired Student’s t test was used for parametric data, while Mann‒Whitney U test was used for nonparametric data. Results were presented as the mean ± SEM.

Figure 2

ROC curve of lncRNA expression levels and their subsequent diagnostic power. Results of Student’s test, Mann–Whitney U test used to analyse the difference between the patients and control groups (a) Unpaired Student’s t test was performed for GAS5 data. (b) SNHG8 and (c) MALAT1 data were analysed by Mann‒Whitney U test. All three were found to be statistically significant. *Indicates p < 0.05. The diagnostic power of the three lncRNAs was determined by ROC curve analysis. The three achieved an acceptable AUC. (d) The ROC curve of GAS5 had an AUC of 0.7519 (p value = 0.0140, 95% CI = 0.5864–0.9173). The sensitivity and specificity of GAS5 were 72.22% and 66.67%, respectively. (e) The discriminative power of SNHG8 was measured, and the AUC was 0.7352 (p value = 0.0217, 95% CI = 0.5401–0.9303). (f) The AUC of MALAT1 was 0.7185 (p value = 0.0329, 95% CI = 0.5279–0.9091). The sensitivity and specificity of SNHG8 and MALAT1 were 88.89% and 66.67%, respectively.