Identification of Potential Prognostic Long Non-Coding RNA Biomarkers for Predicting Recurrence in Patients with...

Cervical cancer (CC) is one of the most common malignant tumors in women, and its treatment is often accompanied by high recurrence. We aimed to identify the long non-coding RNAs (lncRNAs) associated with CC recurrence.

We downloaded lncRNAs expression data of CC patients from The Cancer Genome Atlas (TCGA) dataset and used Cox regression models to analyze the lncRNAs relationship with CC recurrence. The significantly associated lncRNAs were utilized to construct a recurrence risk score (RRS) model. Bioinformatics analyses were used to assess the potential role of the critical lncRNAs in CC recurrence. The effect of critical lncRNAs on CC phenotype was determined by in vitro experiments.

Using Cox regression analysis, four lncRNAs, ie, HCG11, CASC15, LINC00189, and LINC00905, were markedly associated with worse recurrence-free survival (RFS) of CC, whereas three lncRNAs, including HULC, LINC00173, and MIR22HG, were the opposite. After constructing the RRS model, Kaplan-Meier analysis revealed that patients with high RRS had significantly increased risk of recurrence. Among the 20 types of tumors in the TCGA database which all had adjacent normal tissues, MIR22HG and HCG11were significantly downregulated in 18 and 10 types of tumors including CC, respectively. Increased MIR22HG was significantly relevant to decreased risks of recurrence among the subgroups of age at diagnosis < 45 (Hazard Ratio (HR) = 0.26), stage I/II (HR = 0.33), T stage I/II (HR = 0.30), chemotherapy (HR = 0.18), and molecular therapy (HR = 0.16). Functionally, elevated MIR22HG expression could suppress CC cell proliferation, migration and invasion.

MIR22HG has a fundamental role in CC recurrence and could be served as a potential prognostic biomarker.