Meta analysis of the diagnostic efficacy of long non-coding RNA in ischemic stroke

Abstract

Objective To evaluate the diagnostic efficacy of long non-coding RNA (lncRNA) for acute ischemic stroke (IS) by Meta analysis.

Methods This research had obtained a registration number in PROSPERO (CRD 42021283867). The articles of lncRNA diagnosis of IS in PubMed, Web of Science, Cochrane, CBM, China National Knowledge Network, VIP and Wanfang were searched systematically. The papers was evaluated using the QUADAS-2 scale and a bias risk chart – RevMan5.3. Then, threshold effect, non-threshold effect, combined diagnostic efficacy and subgroup analysis of the articles were analyzed by Meta-DISC 1.4. Finally, the summary receiver operating characteristic (SROC) curve was drawn, sensitivity and publication bias were analyzed by stata16.

Results This paper covers eleven studies, including 1717 patients with IS and 1652 healthy subjects. The combined sensitivity of lncRNA diagnosis of IS was 0.70 (95%: 0.69～0.72), specificity was 0.71 (CI95%: 0.69～0.72), the positive likelihood ratio was 2.84 (CI95%: 2.25～3.59), the negative likelihood ratio was 0.39 (CI95%: 0.33～0.46), the diagnostic odds ratio was 7.69 (5.2～11.37) and the area under the curve (AUC) was 80%. Subgroup analysis found that among tissue sources, serum samples had the highest comprehensive diagnostic efficiency, up to 88.5%, and the comprehensive diagnostic efficiency within 24 hours was higher than that within 24 to 72 hours, up to 82.6%. In addition, low expression was slightly higher than high expression, reaching 79.8%.

Conclusion There is a high diagnosing efficiency shown in IncRNA, which could be expected to become a predictive tool for IS in the future

Background

The incidence of stroke bout 0.25% ~ 4%, mortality of about 30%, involving ischemic stroke (IS) accounted for about 80% of the stroke ^1,2. IS is usually caused by atherosclerotic thrombosis or cerebral artery embolism, which could lead to brain tissue death, focal neurological deficits, and disability, bringing an enormous burden to patients and society ³. Currently, IS has become the leading cause of death and disability as well as the second cause of dementia ⁴. Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) are the most common IS diagnostic tools, however, MRI is not suitable for patients with metal stents and is expensive ⁵, and CT can not explore early ischemic changes in the brain sensitively ⁶, therefore it is necessary to develop sensitive and affordable diagnostic tools. Long non-coding RNA(lncRNA) is non-coding RNA with a length of more than 200 nucleotides ⁷. In recent years, RNA-seq analysis has found that ischemia significantly changes the expression of a series of lncRNAs in brain tissue from human and adult rat brains. Accordingly, the abnormal expressions of IncRNA provide theoretical basics for IS diagnosis and treatment ^8–11. This study aims to systematically evaluate the comprehensive diagnostic efficacy of lncRNA for IS through quantitative meta-analysis and to provide evidence-based medical evidence for further exploring whether lncRNA can be used as a new marker of IS.

Materials And Methods

This study was conducted in strict accordance with relevant guidelines and regulations. The protocol of this study has been registered in the PROSPERO, an international database of prospectively registered systematic reviews in health and social care, with a registration No. CRD 42021283867 (https://www.crd.york.ac.uk/PROSPERO/).

Search strategy Keywords combined with a free word retrieval strategy were used to conduct lpapers retrieval on seven databases, namely Pubmed, Web of Science, Cochrane, CBM, CNKI, VIP and wanfang, from the establishment of the database to August 20, 2021. Database search terms involved "ischemic stroke/ acute ischemic strokes/acute ischemic stroke/acute cerebral infarction/cerebral Myocardial /ischemic cerebrovascular disease/cryptogenic embolism stroke /wake up stroke " and " lncRNA /long non-coding RNA".

Data extraction and screening Two professionals searched and screened independently on the seven databases, and conducted cross-checking. If there were any disagreements, a third person would get involved in negotiation and adjudication. The inclusion criteria of the literature were as follows: ①Subjects: all patients were diagnosed by imaging or met the diagnostic criteria of the guidelines for the diagnosis and treatment of IS ¹²; ②Diagnostic tools: MRI, CT, MRA, CTA or guidelines; ③Outcome indicators: Direct or indirect extraction of fourfold table data; ④Study type: diagnostic study. Exclusion criteria:①IS merges with hemorrhagic cerebrovascular disease; ②Animal experiments, reviews and conference papers; ③Studies on non-diagnostic lncRNA; ④Less than 20 cases were covered in the experiment or control group.

Literature quality evaluation The QUADAS-2 scale was used to evaluate the quality of the included papers ¹³. The scale was divided into four parts: patient selection, index test, reference standard and flow and timing. The evaluation results were "high", "unclear", and "low" risk, corresponding to 0, 0 and 1 point respectively. RevMan5.3 was used to make the risk of bias and applicability concerns summary.

Statistical methods Meta - Disc 1.4 and Stata 16 analyzed the data¹⁴.Spearman correlation coefficient of meta-Disc1.4 was used to evaluate the heterogeneity caused by the threshold effect. Cochran's Q and I² tests were applied to assess the heterogeneity caused by the non-threshold effect. P < 0.05 or I² > 50% indicated significant heterogeneity between studies. A random-effect model or subgroup analysis was conducted, and a fixed-effect model was used when I² < 50%; The combined effect sizes were sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio and AUC. In addition, the reasons for heterogeneity and Deek's quantitative funnel plot test were analyzed by subgroup analysis and sensitivity analysis, and the difference was statistically significant at P < 0.05.

Results

Basic features of the included literature  There were 596 papers retrieved from seven databases managed by Endnote software. After removing duplicated papers, there were 462 articles. Then 100 papers, such as reviews and conference papers, were taken off after scanning the title. After that, the 317 unrelated papers were excluded by going through the abstract, including 272 non-conforming articles and 45 animal experiment papers. 34 papers with inconsistent outcome indicators and missing data were withdrawn after reading through the full text. Finally, 7 English and 4 Chinese articles were included (see Figure 1). Among these 11 studies involved 1717 was patients and 1652 normal controls, and outcomes included TP; FP. FN and TN. The basic characteristics were shown in Table 1

Table 1 Basic data of lncRNA in AIS diagnosis

Abbreviations: Peripheral blood mononuclear cell(PBMC);Magnetic resonance imaging(MRI)

Literature quality evaluation  Based on QUADAS-2 results: The papers’ scores were all above 3 points, and the quality was good. Due to the 11 articles being all case-control trials, there were high risks in selecting cases. Furthermore, four items ^15,19,23,25 did not give critical values leading to all being listed as high risks, and nine items ^{15,16,18-22,25} used more than two diagnostic methods resulting in being listed as high risks. Most of the articles did not describe whether reference standard judgment was blinded, therefore, it was unclear for inclusion. From the risk of bias map, it was found that the evaluation of the clinical applicability of the papers was high（Figure 2）.

Meta-analysis results

Threshold effect evaluation Meta DiSc software r=-0.282 obtained Spearman correlation coefficient between sensitivity logarithm and (1-specificity) logarithm (P=0.257>0.05), which was not significant, indicating that there was no threshold effect in this study. Furthermore, no "shoulder arm shape" appeared by drawing the symmetric sROC curve, demonstrating no heterogeneity caused by the threshold effect in this study (Figure 3).

Evaluation of the non-threshold effect   The Cochran-Q test of the diagnostic odds ratio (DOR) showed that Cochran-q =129.67, I²=86.9%, P<0.001, describing that heterogeneity caused by non-threshold effect exists in this study. Furthermore, this study's sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and I² of DOR were greater than 50%. Therefore, the random effect model was implemented to combine the above five effect sizes.

Effect size results were combined   According to stata16, the AUC of lncRNA for the diagnosis of IS was 80%. The Meta DiSc software showed that the combined sensitivity of combined IS was 0.70 (95%: 0.69-0.72), the specificity was 0.71 (CI95%: 0.69-0.72), and the positive likelihood was 0.71 (CI95%: 0.69-0.72). The ratio was 2.84 (CI95%: 2.25-3.59), the negative likelihood ratio was 0.39 (CI95%: 0.33-0.46), and the diagnostic odds ratio was 7.69 (5.2-11.37) (Figure 4).

Subgroup analysis results  Subgroup analysis found that: The heterogeneity of tissue source and blood collection time decreased after grouping, while the heterogeneity of lncRNA expression was almost unchanged after high-low grouping. Among tissue sources, the diagnostic efficiency of serum specimens was 88.5%. The diagnostic efficiency within 24h was 82.6% higher than that within 24 ~ 72h, and low expression was slightly higher than high expression (79.8%). The DOR of combined lncRNA was 20.33, much higher than the diagnostic performance of single lncRNA (Table 2).

Table 2 Subgroup analysis of different categorical variables

Note: Abbreviations:Positive likelihood ratio(PLR);Negative likelihoodratio(NLR);Diagnostic odds ratio(DOR);ROC curve(AUC);Peripheral blood mononuclear cell(PBMC)

Sensitivity analysis  A sensitivity analysis was performed to understand the combined effect size changes after removing an individual study (Figure.5). The results demonstrated that the bivariate model was moderately robust. Two outliers were identified by impact analysis, and two outliers were found through outlier detection.

Publication bias test  Through Deek's quantitative funnel plot test, P=0.83, there was no publication bias in the literature (Figure 6).

Discussion

In recent years, many studies have found that lncRNAs have potential as biomarkers for IS, providing a theoretical basis for diagnosing and treating IS ^15-25. This study combines the effect sizes of 11 papers on lncRNA diagnosis of IS through meta-analysis. The area under the merger curve was 80%, the sensitivity of the merger was 0.70, and the specificity was 0.71, indicating that lncRNA could distinguish IS from normal patients with high diagnostic value. Additionally, the positive likelihood ratio was 2.84, reporting that lncRNA was 2.84 times more likely to judge positive than incorrectly judged positive in diagnosing IS correct. The higher the ratio, the higher the probability of true positive. The negative likelihood ratio was 0.39, and it is noted that the smaller the value, the lower the false-negative rate of diagnosis. When DOR < 1, the diagnostic test efficiency is low, and the larger the value, the higher the diagnostic efficiency²⁶. The DOR of this study was 7.69, demonstrating that the overall diagnostic efficiency was high. Furthermore, through the sensitivity analysis to judge sources of heterogeneity, finding one of the research beyond the limit, a diagnosis of abnormal value, found by reading articles. The research object was 1100 patients, the sensitive and specific degree was not high, especially the specific degree of only 0.538, which might cause heterogeneity in the calculation of the overall research¹⁵. 

Among these lncRNAs, some have also been reported in vitro and in vivo studies of IS. For example, Guo et al. ²⁷ reported that the expression of lncRNA MIAT increased in PC12 cell injury, induced by oxygen-glucose deprivation/reoxygenation (OGD/R) in IS rats, which was consistent with the expression of MIAT in this Meta-analysis. MIAT silencing promoted the Regulated development and DNA damage responses 1 (REDD1) and reduced neuronal autophagy and apoptosis, thereby reducing IS-induced neuronal damage²⁷. Zhang et al. ²⁸ found that the expression level of serum lncRNA ANRIL in IS patients was higher than people who were in the control group, moreover, the expression level of serum ANRIL in patients with severe neurological dysfunction was significantly higher than that in patients with moderate or mild neurological dysfunction, which could be used as a molecular marker for the diagnosis of IS and play an important role in the pathogenesis and development of IS. Other studies indicated that the sensitivity and specificity of serum LncRNA ANRIL in the differentiation of ischemic stroke atrial fibrillation were 76.6% and 81.4%, respectively ²⁹, showing that ANRIL has a great predictive value for ischemic stroke atrial fibrillation. 

In a clinical study of acute IS, it was found that LncRNA NEAT1 expression was also up-regulated and had a good predictive value for the risk of acute IS (AUC=0.804), which is consistent with the Meta result. In addition, NEAT1 expression was positively correlated with NIHSS stroke scale score and inflammatory factors, however, it was negatively correlated with anti-inflammatory factors ²⁴, which can be used as a new biomarker to participate in the prognosis and development of acute IS. Han et al. ³⁰ found that LncRNA NEAT1 increased the protein level of the core factors of the Wnt/β-catenin signaling pathway, indicating that NEAT1 was involved in this activation. The Upregulation of NEAT1 could regulate OGD/R through Wnt/β-catenin signaling to induce microglia and neuroinflammatory damage, which may be a new therapeutic target for IS.

Limitations of this study: ① Although most of the items included in this study are English articles, these articles were published by Chinese, which might be limited to regional population; ②There were differences in tissue source, collection time and expression level of the included papers, indicating high heterogeneity. Then through subgroup analysis, heterogeneity did not decrease significantly, in addition, there were limited studies involving low expression and combined lncRNA; ③The diagnostic efficacy of all lincRNAs was combined in this study, and it is not clear which lncRNA had the highest diagnostic efficacy for IS.

In conclusion, lncRNA has a high comprehensive diagnostic efficiency for IS, with low q-PCR detection cost, convenient and fast blood sample collection. It is believed that with the development of lncRNA technology, its diagnostic value in IS will be confirmed and widely promoted in clinical practice.

Declarations

CRediT authorship contribution statement

Jiqing Zheng:Literature screening, prepared figures1-5, Writing -original draft

Yaobin Long:Conception, Manuscript revision, Funding acquisition

Yun Liu:Literature screening, Writing -original draft, Funding acquisition

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

Guangxi Science and Technology Major Project (NO: Gui Ke AA17204017-5), Guangxi Natural Science Foundation of China (NO: 2022GXNSFAA035511), National Natural Science Foundation of China (NO: 81960768), Development and Promotion of Appropriate Medical and Health Technologies in Guangxi(NO: S2019010）

Availability of Data and Materials

The datasets used and/or analysed during the current study available from the corresponding author (Jiqing Zheng) on reasonable request. 

References

1. Mozaffarian, D. et al. Heart Disease and Stroke Statistics-2016 Update: A Report From the American Heart Association. Circulation 133, e38-360, doi:10.1161/cir.000000000000035 0 (2016).
2. Donkor, E. S. Stroke in the 21(st) Century: A Snapshot of the Burden, Epidemiology, and Quality of Life. Stroke research and treatment 2018, 3238165, doi:10.1155/2018/3238165 (2018).
3. Prabhakaran, S., Ruff, I. & Bernstein, R. A. Acute stroke intervention: a systematic review. Jama 313, 1451–1462, doi:10.1001/jama.2015.3058 (2015).
4. Dong, X. et al. γδ T cells aggravate blood-brain-barrier injury via IL-17A in experimental ischemic stroke. Neuroscience letters 776, 136563, doi:10.1016/j.neulet.2022.136563 (2022).
5. Zhang, W., Cheng, J., Zhang, Y., Wang, K. & Jin, H. Analysis of CT and MRI Combined Examination for the Diagnosis of Acute Cerebral Infarction. Journal of the College of Physicians and Surgeons–Pakistan: JCPSP 29, 898–899, doi:10.29271/jcpsp.2019.09.898 (2019).
6. Gomolka, R. S. et al. Quantification of image contrast of infarcts on computed tomography scans. The neuroradiology journal 30, 15–22, doi:10.1177/1971400916678226 (2017).
7. Liang, H. et al. LncRNA PFL contributes to cardiac fibrosis by acting as a competing endogenous RNA of let-7d. Theranostics 8, 1180–1194, doi:10.7150/thno.20846 (2018).
8. He, W. et al. Altered Long Non-Coding RNA Transcriptomic Profiles in Ischemic Stroke. Human gene therapy 29, 719–732, doi:10.1089/hum.2017.064 (2018).
9. Zhu, W. et al. LncRNA Expression Profiling of Ischemic Stroke During the Transition From the Acute to Subacute Stage. Frontiers in neurology 10, 36, doi:10.3389/fneur.2019.00036 (2019).
10. Duan, X. et al. Bioinformatics analysis of a long non–coding RNA and mRNA regulation network in rats with middle cerebral artery occlusion based on RNA sequencing. Molecular medicine reports 20, 417–432, doi:10.3892/mmr.2019.10300 (2019).
11. Yang, Y. et al. Comprehensive analysis of lncRNA expression profiles in rats with cerebral ischemia-reperfusion injury after treatment with 20(R)-ginsenoside Rg3. Journal of integrative neuroscience 21, 16, doi:10.31083/j.jin2101016 (2022).
12. Powers, W. J. et al. Guidelines for the Early Management of Patients With Acute Ischemic Stroke: 2019 Update to the 2018 Guidelines for the Early Management of Acute Ischemic Stroke: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke 50, e344-e418, doi:10.1161/str.0000000000000211 (2019).
13. Whiting, P. F. et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Annals of internal medicine 155, 529–536, doi:10.7326/0003-4819-155-8-201110180-00009 (2011).
14. Wu, Y. et al. Non-Coding RNAs as Circulating Biomarkers for the Diagnosis of Intracranial Aneurysm: A Systematic Review and Meta-Analysis. Journal of stroke and cerebrovascular diseases: the official journal of National Stroke Association 30, 105762, doi:10.1016/j.jstrokecerebrovasdis.2021.105762 (2021).
15. Yang, J. et al. LncRNA ANRIL Expression and ANRIL Gene Polymorphisms Contribute to the Risk of Ischemic Stroke in the Chinese Han Population. Cellular and molecular neurobiology 38, 1253–1269, doi:10.1007/s10571-018-0593-6 (2018).
16. Feng, L., Guo, J. & Ai, F. Circulating long noncoding RNA ANRIL downregulation correlates with increased risk, higher disease severity and elevated pro-inflammatory cytokines in patients with acute ischemic stroke. Journal of clinical laboratory analysis 33, e22629, doi:10.1002/jcla.22629 (2019).
17. Guo, X. et al. Identification of Novel LncRNA Biomarkers and Construction of LncRNA-Related Networks in Han Chinese Patients with Ischemic Stroke. Cellular physiology and biochemistry: international journal of experimental cellular physiology, biochemistry, and pharmacology 50, 2157–2175, doi:10.1159/000495058 (2018).
18. Zhu, M. et al. Peripheral Blood Leukocyte Expression of lncRNA MIAT and Its Diagnostic and Prognostic Value in Ischemic Stroke. Journal of stroke and cerebrovascular diseases: the official journal of National Stroke Association 27, 326–337, doi:10.1016/j.jstrokecerebrovasdis.2017.09.009 (2018).
19. Deng, Q. W. et al. Differential long noncoding RNA expressions in peripheral blood mononuclear cells for detection of acute ischemic stroke. Clinical science (London, England: 1979) 132, 1597–1614, doi:10.1042/cs20180411 (2018).
20. Hong, C.Y. et al. Expression level of serum Lnc RNA CAI2 and its clinical significance in patients with acute ischemic stroke, Guangxi Medical Journal 41, 2998–3001 (2019).
21. 21..He, P. et al. The diagnostic and prognostic value of Long non -coding RNA ANRIL in ischemic stroke, Prevention and Treatment of Cardio-Cerebral-Vascular Disease 19, 397–400 (2019).
22. Guo, P. et al. The differential diagnosis value of serum miR-29 combined with lncRNATUG1 and ADMA in ischemic stroke, Stroke and Nervous Diseases 26, 277–281 (2019).
23. Li, S. et al. Expression Profile and Potential Functions of Circulating Long Noncoding RNAs in Acute Ischemic Stroke in the Southern Chinese Han Population. Frontiers in molecular neuroscience 12, 290, doi:10.3389/fnmol.2019.00290 (2019).
24. Li, P., Duan, S. & Fu, A. Long noncoding RNA NEAT1 correlates with higher disease risk, worse disease condition, decreased miR-124 and miR-125a and predicts poor recurrence-free survival of acute ischemic stroke. Journal of clinical laboratory analysis 34, e23056, doi:10.1002/jcla.23056 (2020).
25. Liu, M. et al. Zhou, Serum levels of Lnc RNA SNHG14 and MiR-145-5p in patients with acute cerebral infarction and their relationships with risk of disease, Stroke and Nervous Diseases 27, 47–51 + 55 (2020).
26. Glas, A. S., Lijmer, J. G., Prins, M. H., Bonsel, G. J. & Bossuyt, P. M. The diagnostic odds ratio: a single indicator of test performance. Journal of clinical epidemiology 56, 1129–1135, doi:10.1016/s0895-4356(03)00177-x (2003).
27. Guo, X., Wang, Y., Zheng, D., Cheng, X. & Sun, Y. LncRNA-MIAT promotes neural cell autophagy and apoptosis in ischemic stroke by up-regulating REDD1. Brain research 1763, 147436, doi:10.1016/j.brainres.2021.147436 (2021).
28. Zhang, K. et al. Circulating lncRNA ANRIL in the Serum of Patients with Ischemic Stroke. Clinical laboratory 65, doi:10.7754/Clin.Lab.2019.190143 (2019).
29. Zeng, W. & Jin, J. The correlation of serum long non-coding RNA ANRIL with risk factors, functional outcome, and prognosis in atrial fibrillation patients with ischemic stroke. Journal of clinical laboratory analysis 34, e23352, doi:10.1002/jcla.23352 (2020).
30. Han, D. & Zhou, Y. YY1-induced upregulation of lncRNA NEAT1 contributes to OGD/R injury-induced inflammatory response in cerebral microglial cells via Wnt/β-catenin signaling pathway. In vitro cellular & developmental biology. Animal 55, 501–511, doi:10.1007/s11626-019-00375-y (2019).