Study characteristics and data quality
After searching PubMed, EMBASE and Cochrane library, a total of 3748 articles were retrieved. And 2632 articles were assessed after removing 839 duplicated records. Then, titles and abstracts were scanned, after which, 277 remaining articles entered the process of full-text reading. 165 articles were excluded for following reasons: insufficient data (n=32), foreign language (n=4), review article (n=18), meeting abstract (n=5), limited number of studies which focused on a certain lncRNA (n=108). After the above screening process, we identified 110 studies with 25 specific lncRNAs which might play an important part in the development of breast and gynecologic cancers. Afterwards, we extracted data of breast and gynecologic cancers form TCGA datasets according to the type of lncRNAs.
Finally, 111 eligible studies covering 25 lncRNAs associated with survival outcome or clinicopathological features of breast and gynecologic cancer patients were included in our meta-analysis (including TCGA datasets). The process of screening was shown in Fig. S1, and the characteristics of included studies were shown in Additional file 1. The results of quality assessment were shown in Additional file 2 in accordance with REMARK guideline.
Association between specific tissue lncRNA expression and prognostic outcome in breast and gynecologic cancers
Based on literature screening, we identified 25 specific lncRNAs which might be crucial in the development of breast and gynecologic cancers. And we conducted meta-analyses which covered nearly 20000 patients to systematically evaluate the association between specific lncRNA expression and prognostic outcome of breast and gynecologic cancers, and the results were shown in Fig. 1a and b and Table 1 and Supplementary Table S3 in Additional file 3. Our analyses indicated that 3 lncRNAs (HOTAIR, GAS5, PVT1) might act as promising prognosis biomarkers for breast and gynecologic cancers, because they significantly associated with both OS and DFS for breast and gynecologic cancer patients.
More concretely, higher tissue HOTAIR expression was significantly related to poorer OS (pooled HR=1.70, 95%CI: 1.32-2.19) and DFS (pooled HR=2.04, 95%CI: 1.04-4.03) in breast and gynecologic cancers, as shown in Fig. S2a and S2b. Also, there was a significant association between higher tissue PVT1 expression and poorer OS (pooled HR=1.47, 95%CI: 1.17-1.86) and DFS (pooled HR=1.74, 95%CI: 1.08-2.82) in breast and gynecologic cancers (Fig. S3a and S3b). Similarly, higher tissue SPRY4-IT1, HOXA11-AS, LINP1 and SNHG15 expression might result in poorer OS and DFS in breast and gynecologic cancers (Supplementary Table S3 in Additional file 3), but this conclusion still remained unclear due to the limited study numbers.
Interestingly, 2983 patients were included to evaluate the association between tissue GAS5 expression and prognostic outcome of breast and gynecologic cancer patients, and the pooled HR was 0.51 (95%CI: 0.34-0.77) for OS and 0.40 (95%CI: 0.25-0.63) for DFS, indicating that higher tissue GAS5 expression level predicted better prognostic outcome in breast and gynecologic cancers (Fig. S4a and S4b).
In addition, there was a significant association between tissue MALAT1 (pooled HR=1.51, 95%CI: 1.09- 2.08), NEAT1 (pooled HR=1.80, 95%CI: 1.25-2.58), CCAT2 (pooled HR=1.53, 95%CI: 1.09-2.14) expression and OS rather than DFS in breast and gynecologic cancers, as shown in Table 1. Nevertheless, there was a significant association between tissue UCA1 (HR=3.35, 95%CI: 1.31- 8.56), CRNDE (pooled HR=11.79, 95%CI: 4.29-32.46) expression and DFS rather than OS in breast and gynecologic cancers (Table 1).
Of note, no significant association was found between tissue ANRIL, CCAT1, FEZF1-AS1 and HOTTIP expression and prognostic outcome of breast and gynecologic cancers, indicating that they might not be effective biomarkers for breast and gynecologic cancers in clinical applications (Table1). And another included lncRNAs whose study number was less than 3 were shown in Additional file 1.
Subgroup analyses for the association between specific lncRNA expression and OS in breast and gynecologic cancers
In order to evaluate potential modifying effect of variables and explore the source of heterogeneity, subgroup analyses were conducted according to population and cancer types, which were shown in Fig. 1a, Table 2 and Supplementary Table S4 in Additional file 3. And to make the results of subgroup analyses more convincing, the number of concluded study in each subgroup must be greater than or equal to 2.
Firstly, subgroup analyses were conducted according to population. The analyses indicated that higher tissue HOTAIR expression was significantly associated with poorer OS of breast and gynecologic cancer patients whether in an Asian population (pooled HR=2.05, 95%CI: 1.35-3.11) or in European and American populations (pooled HR=1.49, 95%CI: 1.01-2.20). However, significant association between lncRNAs and survival outcome was uniquely observed in an Asian population for PVT1 (pooled HR=2.07, 95%CI: 1.44-3.00), MALAT1 (pooled HR=2.90, 95%CI: 2.03-4.14) and CCAT2 (pooled HR=2.48, 95%CI: 1.82-3.37).
As for the cancer types, the analyses indicated that higher tissue GAS5 expression was significantly associated with better OS of both breast (pooled HR=0.42, 95%CI: 0.20-0.86) and gynecologic cancer (pooled HR=0.56, 95%CI: 0.34-0.94) patients. Besides, we proposed that higher tissue expression of HOTAIR (pooled HR=1.98, 95%CI: 1.43-2.74), PVT1 (pooled HR = 1.45, 95%CI: 1.12-1.88) and NEAT1 (pooled HR=1.78, 95%CI: 1.15-2.78) might predicted OS of gynecologic cancer patients. Besides, MALAT1 (pooled HR=1.83, 95%CI: 1.13-2.96) expression was significantly associated with OS in breast cancer, indicating it might act as potential prognostic biomarkers for breast cancer with protective effects. And the subgroup analyses of another included lncRNAs were shown in Supplementary Table S4 in Additional file 3.
In addition, subgroup analyses suggested that heterogeneity was mainly influenced by different populations.
Association between specific lncRNA expression and clinicopathological features in breast and gynecologic cancers
As shown in Table 3, we also explored the association between specific lncRNA expression and clinicopathological features in breast and gynecologic cancers, such as age, FIGO stage, histology grade, lymph node metastasis (LNM), lymphovascular space invasion (LVSI) and tumor size, which might provide clues why some lncRNAs associated with survival outcome of breast and gynecologic cancers. Higher tissue expression of 10 lncRNAs was significantly associated with FIGO Stage (I-II vs III-IV) of breast and gynecologic cancers, including HOTAIR, UCA1, GAS5, ANRIL, PVT1, CCAT1, CCAT2, CRNDE, FEZF1-AS1 and HOTTIP. In addition, there was a significant association between 3 lncRNAs (ANRIL, CCAT1, FEZF1-AS1) and histology grade (G1+G2 vs G3) in breast and gynecologic cancers. And higher tissue expression of HOTAIR, UCA1, GAS5, ANRIL, CCAT1, NEAT1, CRNDE, HOTTIP was all significant associated with LNM of breast and gynecologic cancers. Besides, there was a trend that UCA1 and MALAT1 were correlated with LVSI in breast and gynecologic cancer patients. Furthermore, only CRNDE was significantly related to tumor size of patients with breast and gynecologic cancers. And the subgroup analyses of another included lncRNAs were shown in Supplementary Table S4 in Additional file 3. In general, these results indicated that different lncRNAs might function differently in clinicopathological features in breast and gynecologic cancers, which might influenced their prognostic values.
Publication bias
Funnel plots were shown in Fig. S5. No publication bias was found for the included studies except for the assessment of association between tissue MALAT1 expression and DFS in breast and gynecologic cancers. Nevertheless, it was still difficult to confirm whether the publication bias really existed due to the limited number of studies.