The Prognostic Role of miR-31 in Head and Neck Squamous Cell Carcinoma: Systematic Review and Meta-Analysis with Trial Sequential Analysis

Abstract

Background: Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer worldwide with high recurrence, metastasis, and poor treatment outcome. Prognostic survival biomarkers can be a valid tool for assessing a patient's life expectancy and directing therapy toward specific targets. Recent studies have reported microRNA (miR) might play a critical role in regulating different types of cancer. The main miR used as a diagnostic and prognostic biomarker and reported in the scientific literature for HNSCC is miR-21. Other miRs have been investigated to a lesser extent (miR-99a, miR-99b, miR-100, miR-143, miR-155, miR-7, miR-424, miR-183), but among these, the one that has attracted major interest is the miR-31.

Methods: The systematic review was conducted following the PRISMA guidelines using electronic databases, such as PubMed, Scopus, and the Cochrane Central Register of Controlled Trials, with the use of combinations of keywords, such as miR-31 AND HNSCC, microRNA AND HNSCC, and miR-31. The meta-analysis was performed using the RevMan 5.41 software (Cochrane Collaboration, Copenhagen, Denmark).

Results: This search produced 721 records, which, after the elimination of duplicates and the application of the inclusion and exclusion criteria, led to 4 articles. The meta-analysis was conducted by applying fixed-effects models, given the low rate of heterogeneity (I² = 40%). The results of the meta-analysis report an aggregate hazard ratio (HR) for the overall survival (OS), between the highest and lowest miR-31 expression, of 1.59, with the relative intervals of confidence (1.22 2.07). Heterogeneity was evaluated through Chi² = 5.04 df = 3 (p = 0.17) and the Higgins index I² = 40; testing for the overall effect was Z = 3.44 (p = 0.00006). The forest plot shows us a worsening HR value of OS, in relation to the elevated expression of miR-31.

Conclusions: In conclusion, the data resulting from the current meta-analysis suggest that miR-31 is associated with the prognosis of patients with HNSCC and that elevated miR-31 expression could predict a poor prognosis in patients with this type of neoplasm.

Keywords: HNSCC; OSCC; miR-31; microRNA; noncoding RNA; oral cancer.

Figure 1

Entire selection and screening procedures are described in the PRISMA flowchart; tables with the red lines are the searches performed subsequently (on 3 March 2022), with the addition of Web of Science and with an update of the keywords and results on PubMed and Scopus.

Figure 2

Forest plot of the fixed-effects model of the meta-analysis; HR = 1.58, 95% CI: [1.21, 2.06]; df = degrees of freedom; I² = Higgins heterogeneity index, I² < 50%, heterogeneity irrelevant; I² > 75%, significant heterogeneity; C.I. = confidence intervals; P = p value; SE = standard error. The graph for each study shows the first author and the date of publication, hazard ratio with confidence intervals, log HR standard error, and weight of each study expressed as a percentage. The final value is expressed in bold with the relative confidence intervals. The black line shows the position of the average value, and the rhombus in light black shows the measure of the average effect.

Figure 3

Funnel plot. The absence of heterogeneity is evident graphically. I² = 38%, SE: standard error.

Figure 4

Forest plot of the random-effects model of the meta-analysis.

Figure 5

AIS, APIS, light green line (Z = 1.98), dashed red line (monitoring boundary), blue line (cumulative z curve), red line (sample size).