Discussion
Recent evidence has shown that 98% of the human genome could be transcribed into non-coding RNAs (Tao et al., 2015). A large body of studies emphasizes the potential functional roles of lncRNAs in various biological mechanisms and tumorigenesis (Yan et al., 2017, Kong et al., 2016). Altered and aberrant expression of lncRNAs in numerous tumor samples has been well documented, which drives several putative carcinogenesis mechanisms (Li et al., 2018a, Li et al., 2018c, Yan et al., 2017). The lncRNA GAS5plays a vital role as a tumor suppressor lncRNA in almost all cancers (Zheng et al., 2016, Li et al., 2017, Li et al., 2018c, Guo et al., 2017). LncRNA GAS5also encodes small nucleolar RNAs (snoRNAs) and acts as a glucocorticoid riborepressor. Previous investigations have shown that GAS5overexpression inhibits osteosarcoma’s tumorigenesis and epithelial-mesenchymal transition (EMT) by regulating the miR-221/ARHI pathway (Smith and Steitz, 1998, Xu et al., 2018). There is substantial literature reporting clinical cancer therapy approaches for lncRNAGAS5 . Several reports have shown a down-regulation of GAS5in various cancer cells, associated with a poorer prognosis (Cao et al., 2014, Kong et al., 2016, Mourtada-Maarabouni et al., 2009, Li et al., 2018a). Whereas it has been demonstrated the upregulation of GAS5 is associated with a more promising outcome in prostate cancer therapy (Yan et al., 2017). It has been demonstrated that GAS5 can inhibit cell growth by spongingmiR-21, which results in the upregulation of phosphatase and tensin homologs (PTEN ) and inhibition of the mTOR signaling pathway in trastuzumab-resistant HER2-positive breast cancer (Oldoni et al., 2016). Furthermore, the downregulation of lncRNA GAS5 confers tamoxifen resistance by activating miR-222, which targets PTEN in breast cancer (Gu et al., 2018).
The underlying mechanisms of lncRNA GAS5 are not fully understood. Numerous investigations have reported that polymorphisms in lncRNA might have a crucial impact on the lncRNA functions (Kong et al., 2016). Current evidence has shown that lncRNA GAS5 rs145204276 polymorphism could predispose patients to cancer. However, the results concerning the association of the lncRNA GAS5 rs145204276 polymorphism and cancer risk have been inconsistent (Li et al., 2018c, Li et al., 2017, Tao et al., 2015, Xu et al., 2018, Yuan et al., 2018, Zheng et al., 2016, Zhu et al., 2017, Zhu et al., 2016). For instance, Zhuet al. showed that the del allele elevated colorectal cancer risk in the patients (Zhu et al., 2016). At the same time, Zheng et al. found that the del allele was significantly associated with decreased colorectal cancer risk (Zheng et al., 2016). In addition, previous studies have indicated that the lncRNA GAS5 rs145204276 polymorphism could reduce the risk of gastric cancer (Aminian et al., 2019, Li et al., 2018b). Different factors, including variation in populations and ethnicity, different cancer types or lack of statistical power, could cause this discrepancy. Therefore, we conducted this meta-analysis by ten eligible studies, including 6917 cases and 8977 controls, to increase the strength of assessing the association of the lncRNA GAS5 rs145204276 polymorphism and cancer risk. To the best of our knowledge, this meta-analysis is the first meta-analysis which systematically considers all eligible studies in the association of the lncRNA GAS5 rs145204276 polymorphism with cancer risk. In this study, we did not find any significant association between the lncRNA GAS5 rs145204276 polymorphism with cancer risk in the different genetic models. Interestingly, in subgroup analysis based on cancer types, we found that the polymorphism was associated with decreased gastric cancer risk in the patients. It has been demonstrated that the lncRNA GAS5 rs145204276 del allele positively correlated to P27Kip1 expression in gastric cancer patients (Aminian et al., 2019). Besides, P27Kip1 has been shown to act as a tumor suppressor gene, and its down regulation has been reported in various cancers (Ogino et al., 2009, Zhuang et al., 2011).
However, in subgroup analysis based on the source of control, the data did not show a significant association between the polymorphism and cancer risk. Sensitivity analyses revealed that the results were stable and robust among the different genetic models. No publication bias was found in the meta-analysis. In fact, it indicates that our meta-analysis is reliable. However, this meta-analysis has some limitations. These limitations may be a guide for future studies to overcome them. First, only studies written in English were included. Second, the subgroup analysis was limited to cancer types and sources of control. We did not consider an analysis of other variables such as lifestyle factors (e.g., smoking, exercise, and alcohol consumption), environmental factors, gender, stage of cancer, metastasis, and chemotherapy response due to inadequate and inaccessible data of the eligible articles. Third, the sample sizes of the studies were relatively small, which could have a dramatic impact on the power of the study. Fourth, there was heterogeneity in our pooled results which may originate from different types of cancers. All of the included population samples were from China. Finally, genotyping errors may be possible due to applying other genotyping methods. Hence, it is necessary to conduct further studies in different ethnicities with large sample sizes and consider more comprehensive criteria for assessing lncRNA GAS5 and cancer risks.