Long Non-Coding RNA HOTAIR as Ideal Biomarker for the Diagnosis of Various Carcinomas.

Abstract

HOTAIR is a variety of long non-coding RNA that has been recognized as a predictive factor for most cancers. This meta-analysis examined the complete investigative effectiveness of the level of HOTAIR expression for various cancers. Research on the diagnostic value of HOTAIR in different carcinomas was acquired by searching the online databases. Twelve studies consisting of 927 cancer cases were chosen in our research. The sensitivity as well as specificity of the involved articles was helpful to establish the summary receiver operator characteristic (SROC) curve in addition to compute the area under the SROC curve (AUC). In addition, a meta-regression test was done to determine the heterogeneity sources among available studies. The combined effect sizes calculated from involved studies were as follows: sensitivity, 0.73; specificity, 0.83; PLR, 4.4; NLR, 0.32; DOR, 14; and an AUC of 0.85. Deeks' funnel plot asymmetry test showed no probable publication bias. The meta-regression analyses signified that the type of ethnicity is the major cause of heterogeneity. The present meta-analysis suggested that elevated HOTAIR can be considered as a relatively accurate marker for cancer diagnosis and can be applied to support the diagnosis of various cancers.