Objective: The role of Dicer in the prognosis of cancer patients remains controversial. This systematic review is attempted to assess the influence of Dicer as a prognostic predictor for survival in diverse types of cancers.
Methods: Studies were selected as candidates if they published an independent evaluation of Dicer expression level together with the correlation with prognosis in cancers. Random-effect model was applied in this meta-analysis. Heterogeneity between studies was assessed by Q-statistic with P < 0.10 to be statistically significant. Publication bias was investigated using funnel plot and test with Begg's and Egger's test. P < 0.05 was regarded as statistically significant.
Results: 24 of 44 articles revealed low Dicer status as a predictor of poor prognosis. The aggregate result of overall survival (OS) indicated that low Dicer expression level resulted in poor clinical outcomes, and subgroup of IHC and RT-PCR method both revealed the same result. Overall analysis of progression-free survival (PFS) showed the same result as OS, and both the two subgroups divided by laboratory method revealed positive results. Subgroup analysis by tumor types showed low dicer levels were associated with poor prognosis in ovarian cancer (HR = 1.93, 95% CI: 1.19-3.15), otorhinolaryngological tumors (HR = 2.39, 95% CI: 1.70-3.36), hematological malignancies (HR = 2.45, 95% CI: 1.69-3.56) and neuroblastoma (HR = 4.03, 95% CI: 1.91-8.50).
Conclusion: Low Dicer status was associated with poor prognosis in ovarian cancer, otorhinolaryngological tumors and ematological malignancies. More homogeneous studies with high quality are needed to further confirm our conclusion and make Dicer a useful parameter in clinical application.
Keywords: Dicer; cancer; hazard ratio; meta-analysis; prognosis.