Meta-analysis of microRNA expression profiling studies in human cervical cancer.

Abstract

Cervical cancer is one of the most common malignant tumors in women, and numerous studies have associated the disease with changes in microRNA (miRNA) expression. This meta-analysis aimed to consolidate and assess the results of these studies in order to identify potential miRNA biomarkers of cervical cancer. We systematically searched the literature for studies comparing miRNA expression between cervical cancer tissues and normal cervical tissues, and we meta-analyzed the result of 27 studies comprising 1,132 cancer samples and 943 normal samples. We used a vote-counting strategy that took into account total sample and mean fold-change, in order to comprehensively assess associations between certain miRNAs and cervical cancer occurrence and progression. The studies described 195 miRNAs that were significantly up-regulated and 96 microRNAs that were down-regulated in cervical cancer tissues (stage I-IV) relative to normal cervical tissues. Vote-counting analysis showed that up-regulation was most consistently reported for miR-20a and miR-21 (four studies), followed by miR-10a, miR-15b, miR-20b, miR-141, miR-200a, and miR-224 (three studies). Down-regulation was reported most consistently for miR-143 (seven studies), followed by miR-203 and miR-145 (six studies). Fourteen miRNA, respectively, showed a significantly correlated lymphatic node metastasis in eight studies. This meta-analysis has identified several miRNAs whose expression correlates reliably with cervical cancer. These should be probed in further studies to explore their potential as diagnostic biomarkers.