Prognostic and diagnostic value of circRNA expression in cervical cancer: a meta analysis

Abstract

Introduction: Cervical cancer (CC) is a highly prevalent malignancy of the reproductive system. This study aimed to methodically assess the function of circular RNAs (circRNAs) as possible indicators of CC, with a specific emphasis on their usefulness in the identification, prediction, and correlation with clinicopathological elements.

Methods: A comprehensive literature search was conducted using databases such as PubMed, Cochrane Library, Web of Science, Embase, and the China National Knowledge Infrastructure (CNKI). The latest data were extracted on May 3^rd, 2024. The diagnostic potential of circRNA expression was evaluated using a range of metrics including sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). The importance of circRNAs was further evaluated in terms of clinical relevance, pathological features, and prognostic value using pooled odds ratios (ORs) and hazard ratios (HRs).

Results: The meta-analysis included 27 studies, which were categorised based on diagnostic applications (n=3), clinicopathological correlations (n=15), and prognostic evaluations (n=23). Elevated expression levels of oncogenic circRNAs were significantly associated with poor clinical indicators, including tumour size (odds ratio [OR] = 0.425, 95% confidence interval [CI]: 0.267-0.676), International Federation of Gynaecology and Obstetrics (FIGO) stage (OR = 0.315, 95% CI: 0.224-0.443), and lymph node metastasis (OR = 2.975, 95% CI: 1.816-4.872). This upregulation of oncogenic circRNA was also identified as a predictor of worse survival outcomes, with a hazard ratio (HR) of 2.13 (95% CI: 1.73-2.62, P < 0.001). The downregulation of circRNAs with tumour-suppressor properties was similarly associated with poor clinical parameters, such as tumour size (OR = 0.310, 95% CI: 0.102-0.941), FIGO stage (OR = 0.231, 95% CI: 0.101-0.527), and lymph node metastasis (OR = 2.430, 95% CI: 1.156-5.110), and was indicative of a worsened prognosis (HR = 2.20, 95% CI: 1.03-4.70, P = 0.042). In terms of diagnostic value, the pooled sensitivity, specificity, and area under the curve (AUC) were calculated to be 0.85, 0.83, and 0.91, respectively.

Conclusion: The results of our meta-analysis indicate that circRNAs have the potential to serve as promising biomarkers for CC diagnosis, prognosis, and clinicopathology.

Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/, identifier CRD42024544997.

Keywords: cervical cancer; circular RNAs; clinicopathological; diagnosis; meta-analysis; prognosis.

Figure 1

The PRISMA flowchart of the literature selection process.

Figure 2

Quality assessment of eligible studies for diagnostic meta-analysis.

Figure 3

Forest plots of the OS of CC patients for circRNAs. (A) Upregulated circRNAs. (B) Downregulated circRNAs. HR, hazard ratios.

Figure 4

Forest plots of the DFS of CC patients for circRNAs. (A) Upregulated circRNAs. (B) Downregulated circRNAs. HR, hazard ratios.

Figure 5

Subgroup analyses of OS for circRNAs, stratified by cut-off value, follow-up time, regulation and sample size.

Figure 6

Subgroup analyses of DFS for circRNAs, stratified by cut-off value, follow-up time and regulation.

Figure 7

Forest plots of summary sensitivity and specificity to illustrate the diagnostic value of circRNAs for CC.

Figure 8

The summary receiver operating characteristic (SROC) curve based on circRNAs for diagnosis analysis. ROC, receiver operator characteristic.

Figure 9

(A) Egger’s tests of OS (P = 0.003). (B) Egger’s tests of DFS(P = 0.942). (C) Sensitivity analysis of OS. (D) Sensitivity analysis of DFS.