Prognostic value of microRNAs in pancreatic cancer: a meta-analysis.

BACKGROUND: The prognostic impact of microRNA (miRNA) expression levels in pancreatic cancer (PC) has been estimated for years, but the outcomes are controversial and heterogeneous. Therefore, we comprehensively reviewed the evidence collected on miRNA expression in PC to determine this effect.
RESULTS: PC patients with high miR-21 (HR=2.61, 95%CI=1.68-4.04), miR-451a (HR=2.23, 95%CI=1.23-4.04) or miR-1290 (HR=1.43, 95%CI=1.04-1.95) levels in blood had significantly poorer OS (P<0.05). Furthermore, PC patients with high miR-10b (HR=1.73, 95%CI=1.09-2.76), miR-17-5p (HR=1.91, 95%CI=1.30-2.80), miR-21 (HR=1.90, 95%CI=1.61-2.25), miR-23a (HR=2.18, 95%CI=1.52-3.13), miR-155 (HR=2.22, 95%CI=1.27-3.88), miR-203 (HR=1.65, 95%CI=1.14-2.40), miR-221 (HR=1.72, 95%CI=1.08-2.74), miR-222 levels (HR=1.72, 95%CI=1.02-2.91) or low miR-29c (HR=1.39, 95%CI=1.08-1.79), miR-126 (HR=1.55, 95%CI=1.23-1.95), miR-218 (HR=2.62, 95%CI=1.41-4.88) levels in tissues had significantly shorter OS (P<0.05).
CONCLUSIONS: In summary, blood miR-21, miR-451a, miR-1290 and tissue miR-10b, miR-17-5p, miR-21, miR-23a, miR-29c, miR-126, miR-155, miR-203, miR-218, miR-221, miR-222 had significant prognostic value.
METHODS: We searched PubMed, EMBASE, Web of Science and Cochrane Database of Systematic Reviews to recognize eligible studies, and 57 studies comprising 5445 PC patients and 15 miRNAs were included to evaluate the associations between miRNA expression levels and overall survival (OS) up to June 1, 2019. Summary hazard ratios (HR) with 95% confidence intervals (CI) were calculated to assess the effect.

INTRODUCTION

Much effort has been made over a long period of time to identify prognostic biomarkers in pancreatic cancer (PC) patients. Fortunately, a large body of literature has covered the survival of PC patients with abnormal microRNA (miRNA) expression [1-169]. Among all kinds of human cancers, PC has one of the worst prognoses, with a 5-year overall survival (OS) rate of lower than 5% [170]. Despite advances in clinical treatments and new surgical techniques, the survival rate of PC patients has been low for more than 30 years [171]. PC is highly aggressive; therefore, distant metastasis and tissue invasion may occur at early stages [172]. Since invasion and metastasis are the biggest obstacles to effective treatment of PC, it is imperative to explore the molecular biological mechanism leading to such invasive behavior to improve the survival time of patients.

miRNAs are small noncoding RNAs involved in gene regulation [173]. In cancers, a few upregulated miRNAs can serve as oncogenes (oncomiRs) [174], and downregulated miRNAs can serve as tumor suppressors [175]. Expression profiling data analyses have revealed signatures of diagnosis and prognosis that have been employed to stratify various tumor types [174, 176]. As a consequence, miRNAs have the potential to turn into clinical biomarkers for human tumors and into molecular therapeutic targets [177].

Despite comprehensive studies focused on illustrating the molecular biological mechanisms in PC, there are still challenges confronting the identification of minimally invasive and sensitive biomarkers of prognosis. Consequently, it is of vital significance to find prognostic signatures that can be conveniently and reliably applied in the clinical setting to improve the survival time of PC patients.

Increasing evidence indicates that miRNAs have the potential to act as PC prognostic biomarkers in clinical practice [1-169]. Regrettably, there has not been a meta-analysis to evaluate the relationship between dysregulated miRNA expression and survival in PC patients. In view of our previous work, meta-analyses of miRNA expression and cancer patients [178, 179], it is necessary to conduct the current work by searching the recently published literature about miRNAs as prognostic tools in PC tissue or blood.

RESULTS

Meta-analysis

An overview of the HR with 95%CI obtained from the overall comprehensive analysis for all included miRNAs is shown in Table 1. Based on the logical order of the miRNA names, the forest plot, Begg’s funnel plot, sensitivity analysis and funnel plot of the merged analysis adjusted with the trim and fill method are shown in Figures 1–7. The mean NOS score of the included studies was 7.0 (5.0-8.0), indicating that their quality was adequate (Table 2).

Table 1

Summary about results of meta-analysis for miRNA expression in pancreatic cancer.

miRNA	Sample	Survival analysis	Number of articles	Included studies	HR	95%CI	Figure	P value	Heterogeneity (Higgins I2 statistic)	Total patients
High miR-21	Blood	OS	5	4-8	2.61	1.68-4.04	2	<0.01	I2=33.8%, P=0.20	326
High miR-196a	Blood	OS	2	16,17	1.61	0.50-5.23	2	0.43	I2=79.5%, P=0.03	66
High miR-451a	Blood	OS	3	7,8,23	2.23	1.23-4.04	2	<0.01	I2=2.1%, P=0.36	137
High miR-1290	Blood	OS	2	24,26	1.43	1.04-1.95	2	0.03	I2=0.0%, P=0.76	223
High miR-10b	Tissue	OS	4	35-38	1.73	1.09-2.76	3	0.02	I2=61.5%, P=0.03	375
High miR-17-5p	Tissue	OS	3	39-41	1.91	1.30-2.80	3	<0.01	I2=0.0%, P=0.96	164
High miR-21	Tissue	OS	19	5,43-60	1.90	1.61-2.25	3	<0.01	I2=43.9%, P=0.02	1947
High miR-21	Tissue	OSm	8	5,45-48,50-52	2.43	1.89-3.13	4	<0.01	I2=0.0%, P=0.73	592
High miR-21	Tissue	OSAdjusted			1.58	1.32-1.89		<0.01	I2=58.6%, P<0.01
High mIR-23a	Tissue	OS	4	50,53,61,62	2.18	1.52-3.13	8	<0.01	I2=0.0%, P=0.51	251
Low miR-29c	Tissue	OS	4	33,46,69,70	1.39	1.08-1.79	8	0.01	I2=51.8%, P=0.10	463
Low miR-126	Tissue	OS	3	27,68,82	1.55	1.23-1.95	8	<0.01	I2=0.0%, P=0.99	455
High miR-155	Tissue	OS	3	14,50,51	2.22	1.27-3.88	8	<0.01	I2=0.0%, P=0.47	211
Low mIR-200c	Tissue	OS	3	109-111	1.40	0.51-3.79	8	0.51	I2=87.2%, P<0.01	258
High miR-203	Tissue	OS	4	59,112-114	1.65	1.14-2.40	8	<0.01	I2=83.6%, P<0.01	619
Low miR-218	Tissue	OS	3	121-123	2.62	1.41-4.88	8	<0.01	I2=57.5%, P=0.10	248
High miR-221	Tissue	OS	4	46,50,125,126	1.72	1.08-2.74	8	0.02	I2=4.9%, P=0.37	187
High miR-222	Tissue	OS	3	28,126,127	1.72	1.02-2.91	8	0.04	I2=36.8%, P=0.21	322

HR: hazard ratios; CI: confidence intervals; OS: overall survival; mmultivariate analysis; AdjustedAdjusted with the trim and fill method.

Figure 1

Forest plot about OS of PC patients with high miR-21, miR-196a, miR-451a or miR-1290 level in blood

Figure 7

Forest plot about OS of PC patients with high miR-23a, miR-155, miR-203, miR-221, miR-222 or low miR-29c, miR-126, miR-200c, miR-218 level in tissue.

Table 2

Newcastle-Ottawa scale quality assessment results.

First author	Year	Reference	Selection	Comparability	Outcome	Total
Liu	2012	[4]	★★★	★★	★★	7
Wang	2013	[5]	★★★	★★	★★	7
Abue	2015	[6]	★★★	★★	★★	7
Goto	2018	[7]	★★★	★★	★★	7
Kawamura	2019	[8]	★★★	★★	★★★	8
Mikamori	2017	[14]	★★★	★★	★★★	8
Kong	2010	[16]	★★★	★★	★★	7
Yu	2017	[17]	★★★	★★	★★	7
Takahasi	2018	[23]	★★★	★★	★★	7
Li	2013	[24]	★★★	★★	★★★	8
Tavano	2013	[26]	★★★	★★	★★	7
Liao	2018	[27]	★★★	★★	★★	7
Schultz	2012	[28]	★★★	★★	★★	7
Wang	2019	[33]	★★	★	★★★	6
Nakata	2011	[35]	★★	★	★★★	6
Preis	2011	[36]	★★★	★★	★★	7
Nguyen	2016	[37]	★★★	★★	★★	7
Yang	2017	[38]	★★★	★★	★★★	8
Yu	2010	[39]	★★★	★★	★★★	8
Gu	2016	[40]	★★★	★★	★★	7
Zhu	2018	[41]	★★	★	★★	5
Dillhoff	2008	[43]	★★	★	★★★	6
Giovannetti	2010	[44]	★★★	★★	★★★	8
Hwang	2010	[45]	★★★	★★	★★★	8
Jamieson	2011	[46]	★★★	★★	★★	7
Nagao	2012	[47]	★★★	★★	★★	7
Caponi	2013	[48]	★★★	★★	★★★	8
Kadera	2013	[49]	★★★	★★	★★★	8
Ma	2013	[50]	★★★	★★	★★	7
Papaconstantinou	2013	[51]	★★★	★★	★★★	8
Donahue	2014	[52]	★★★	★★	★★★	8
Frampton	2014	[53]	★★★	★★	★★	7
Mitsuhashi	2015	[54]	★★★	★★	★★	7
Vychytilova-Faltejskova	2015	[55]	★★	★	★★	5
Morinaga	2016	[56]	★★★	★★	★★★	8
Benesova	2018	[57]	★★★	★★	★★	7
Xi	2018	[58]	★★	★	★★★	6
Zhang	2018	[59]	★★	★	★★★	6
Zhao	2018	[60]	★★★	★★	★★★	8
Diao	2018	[61]	★★	★	★★	5
Wu	2018	[62]	★★★	★★	★★	7
Liang	2018	[68]	★★	★	★★★	6
Jiang	2015	[69]	★★	★	★★	5
Zou	2015	[70]	★★★	★★	★★	7
Yu	2018	[82]	★★★	★★	★★★	8
Yu	2010	[109]	★★★	★★	★★★	8
Paik	2015	[110]	★★★	★★	★★★	8
Liu	2016	[111]	★★★	★★	★★★	8
Ikenaga	2010	[112]	★★★	★★	★★★	8
Shao	2017	[113]	★★	★	★★★	6
Shi	2018	[114]	★★	★	★★★	6
Li	2013	[121]	★★★	★★	★★	7
Zhu	2014	[122]	★★★	★★	★★	7
Li	2015	[123]	★★★	★★	★★★	8
Sarkar	2013	[125]	★★	★	★★★	6
Wang	2016	[126]	★★★	★★	★★	7
Lee	2013	[127]	★★★	★★	★★	7

High miR-21, miR-451a and miR-1290 levels in the blood predict poor OS

Five studies [4-8] analyzed the connections between high blood miR-21 levels and OS, indicating that PC patients with high blood miR-21 levels had significantly poorer OS than those with low levels (HR=2.61, 95%CI=1.68-4.04, P<0.01, Figure 1).

Two studies [16, 17] reported the relationship between high blood miR-196a levels and OS, but no significant associations were found between high blood miR-196a and OS (HR=1.61, 95%CI=0.50-5.23, P=0.43, Figure 1).

Three studies [7, 8, 23] focused on the correlativity between high blood miR-451a levels and OS, indicating that PC patients with high miR-451a levels had significantly shorter OS than those with low levels (HR=2.23, 95%CI=1.23-4.04, P<0.01, Figure 1).

Two studies [24, 26] stressed the pertinence between high blood miR-1290 levels and OS, suggesting that PC patients with high miR-1290 levels had significantly worse OS than those with low levels (HR=1.43, 95%CI=1.04-1.95, P=0.03, Figure 1).

High miR-10b, miR-17-5P, miR-21, miR-23a, miR-155, miR-203, miR-221, and miR-222 levels or low miR-29c, miR-126, and miR-218 levels in tissues predict poor OS

The details are shown in Table 1 and Figures 2 and 7.

Figure 2

Forest plot about OS of PC patients with high miR-10b, miR-17-5P or miR-21 level in tissue.

High miR-21 levels in tissues predict poor OS (multivariate analysis)

The details are shown in Table 1 and Figure 3.

Figure 3

Forest plot about OS of PC patients with high miR-21 level in tissue (multivariate analysis).

Publication bias

Begg’s funnel plot was employed to estimate publication bias in the study of OS in PC patients with high tissue miR-21 levels (Figure 4). The results showed that the P value was less than 0.01, indicating the presence of publication bias.

Figure 4

Begg’s funnel plot about OS of PC patients with high miR-21 level in tissue.

Sensitivity analysis

Sensitivity analysis was used to estimate whether any single study had undue influence on the OS of PC patients with high tissue miR-21 levels (Figure 5). The outcome showed that no single investigation significantly affected the pooled HR and 95%CI.

Figure 5

Sensitivity analysis about OS of PC patients with high miR-21 level in tissue.

The trim and fill method

As such (Figure 4), the trim and fill method was conducted, and the pooled HR was recalculated with assumed lost studies to assess dissymmetry in the funnel plot (Figure 6), manifesting no publication bias (P=0.80). The recalculated HR did not change significantly for OS (HR=1.58, 95%CI=1.32-1.89, P<0.01).

Figure 6

Funnel plot about pooled analysis adjusted with trim and fill method of OS of PC patients with high miR-21 level in tissue. Circles: included studies; diamonds: presumed missing studies.

DISCUSSION

Foremost findings

The current meta-analysis included 57 English articles that incorporated 15 miRNAs and 5445 patients. As the most researched miRNA, PC patients with high blood or tissue miR-21 levels had significantly poorer OS than those with low levels. It also proved true among PC patients with high tissue miR-21 levels (multivariate analysis) and pooled analysis adjusted with the trim and fill method of OS, indicating that miR-21 is a stable and useful prognostic biomarker in PC. Moreover, a few other miRNAs had significant prognostic impact on PC, including blood miR-451a, and miR-1290 and tissue miR-10b, miR-17-5p, miR-29c, miR-126, miR-155, miR-203, miR-218, miR-221, and miR-222. Among these, blood miR-21, and miR-451a and tissue miR-23a, miR-155, and miR-218 were strong biomarkers of prognosis for PC.

Altered expression, potential targets and pathways for studied miRNAs

In addition, an overview of the 15 miRNAs with dysregulated levels, covering the validated targets and pathways, is shown in Table 3. Most of the included miRNAs showed stable expression levels, higher or lower than the control groups except miR-200c. In brief, Table 3 could support a better understanding of the molecular biological mechanisms of miRNAs in PC.

Table 3

Summary of miRNAs with altered expression, their validated targets and pathways entered this study.

miRNA	Reference	Expression	Potential target	Pathway
10b	[35–38]	Up	None	Cell invasion
17-5p	[39–41]	Up	PTEN,RBL2	Cell cycle, invasion and proliferation
21	[4–8,43–60]	Up	BTG2,FASL,PDCD4,SPRY2	Cell apopsotis, chemoresistance, cycle, proliferation, FASL/FAS, MAPK/ERK and PI3K/AKT signaling
23a	[50, 53, 61, 62]	Up	ESRP1,FOXP2,NEDD4L	Cell invasion, epithelial-mesenchymal transition, migration and proliferation
29c	[33, 46, 69, 70]	Down	MMP2	Cell invasion, migration and Wnt signaling
126	[27, 68, 82]	Down	None	None
155	[14, 50, 51]	Up	None	None
196a	[16,17]	Up	None	None
200c	[109–111]	Unstable	None	Cell invasion and proliferation
203	[59, 112–114]	Up	None	None
218	[121–123]	Down	UGT8,VOPP1	Cell proliferation
221	[46, 50, 125, 126]	Up	None	Cell migration and proliferation
222	[28, 126, 127]	Up	NOSTRIN	None
451a	[7, 8, 23]	Up	None	None
1290	[24, 26]	Up	None	None

PTEN: phosphatase and tensin homolog; RBL2: RB transcriptional corepressor like 2; BTG2: BTG anti-proliferation factor 2; FASL: Fas ligand; PDCD4: programmed cell death 4; SPRY2: sprouty RTK signaling antagonist 2; ESRP1: epithelial splicing regulatory protein 1; FOXP2: forkhead box P2; NEDD4L: NEDD4 like E3 ubiquitin protein ligase; UGT8: UDP glycosyltransferase 8; VOPP1: VOPP1 WW domain binding protein; NOSTRIN: nitric oxide synthase trafficking; FAS: Fas cell surface death receptor; MAPK: mitogen-activated protein kinase; ERK: extracellular regulated protein kinases; PI3K: phosphoinositide-3-kinase; AKT: AKT serine/threonine kinase 1.

Superiorities of the meta-analysis

The present work had two strengths : (1) we looked for and found out almost all studies with OS in PC patients with dysregulated miRNA levels. In addition, the recent miRNA expression pattern is shown in Tables 4 and 5 that differentiates miRNA names and the sample types. (2) The majority of included articles had large sample sizes (≥30, all but 4 studies [6, 41, 121, 125]), intensifying and widening the applicability of the prognostic outcomes for PC patients.

Table 4

Frequency of studies estimating prognostic value of blood miRNA expression in pancreatic cancer.

miR	N	R	miR	N	R	miR	N	R	miR	N	R
let-7b-5p	1	1	107	1	11	203	1	18	483-3p	1	6
16-2-3p	1	2	124	1	12	205	1	19	486-3p	1	24
19a-3p	1	1	125b-5p	1	13	210	1	17	602	1	2
19b-3p	1	1	150	1	10	222	1	20	629	1	25
21-5p	1	3	155	1	14	223-3p	1	1	877-5p	1	2
21	5	4-8	182	1	15	301a-3p	1	21	890	1	2
25-3p	1	1	191	1	7	373	1	22	1290	2	24,26
33a	1	9	192-5p	1	1	375	1	3	3201	1	2
34a	1	10	196a	2	16,17	451a	3	7,8,23	4525	1	8

Table 5

Frequency of studies estimating prognostic value of tissue miRNA expression in pancreatic cancer.

miR	N	R	miR	N	R	miR	N	R	miR	N	R	miR	N	R
let-7a-3	1	27	92b-3p	1	75	155	3	14,50,51	301a-3p	1	129	509-5p	1	151
let-7g*	1	28	93	1	38	181c	1	100	301b	1	38	539	1	152
let-7g	1	29	96-5p	1	76	182-5p	1	76	323-3p	1	130	545	1	153
1	1	30	100	2	50,77	183	1	101	326	1	71	548an	1	154
7-5p	1	31	101	1	78	191	1	102	328	1	68	590-5p	1	38
9-5p	1	32	103	1	79	192	2	33,103	329	1	131	613	1	155
9	1	33	107	1	80	195	1	104	337	1	132	615-5p	1	156
10a-5p	1	34	124	1	81	196a-2	1	105	342-3p	2	53,133	661	1	157
10b	4	35-38	125a-3p	1	29	196b	2	59,106	361-3p	1	134	663	1	158
15b	1	38	125a	1	68	198	2	55,107	367	1	135	664a	1	68
17-5p	3	39-41	125b	1	77	199a-3p	1	53	371-5p	1	136	664	1	159
19a	1	42	126	3	27,68,82	200c-3p	1	108	374b-5p	1	137	675-5p	1	160
21	19	5,43-60	130b	1	83	200c	3	109-111	375	1	50	675	1	28
23a	4	50,53,61,62	132	2	33,84	203	4	59,112-114	376b	1	68	708-5p	1	161
24-1	1	27	133a-1	1	27	204-5p	1	115	376c	1	68	744	1	162
25-3p	1	63	133a	2	33,85	204	1	95	377	1	138	891b	1	163
26a	1	64	135b-5p	2	86,87	205-5p	1	29	410-3p	1	139	940	1	164
27a	1	53	135b	1	88	205	2	19,116	421	1	27	1181	1	165
29a-5p	1	29	137	1	89	211	1	117	424	2	82,114	1246	1	166
29a	1	65	139-5p	1	90	212-3p	1	29	429	1	140	1247	1	167
29b-2-5p	1	66	139	1	91	212	2	28,118	448	1	141	1266	1	168
29b-3p	1	67	140	1	33	214	1	30	450b-5p	1	28	1293	1	114
29b	2	33,68	141	2	92,93	216b-5p	1	119	451	1	142	1301	1	68
29c	4	33,46,69,70	142-3p	2	53,94	216b	1	120	454	1	68	3157	1	27
30a	1	71	142-5p	1	95	217	1	50	483-3p	1	143	3613	1	68
30b	1	72	143	1	50	218	3	121-123	491	1	33	3656	1	169
30d	1	46	146a	1	28	219	1	71	494	3	144-146	4521	1	27
30e	1	27	148a*	1	28	221-3p	1	124	495	1	147	4709	1	27
31	2	50,54	148a	1	50	221	4	46,50,125,126	497	1	148	5091	1	27
34a-5p	2	29,73	148b	1	96	222	3	28,126,127	501-3p	1	149
34a	1	46	150	1	97	223	1	128	501	1	27
34b	1	74	153	2	98,99	224	2	46,71	506	1	150

Highlighted studies were included in the present meta-analysis; N: Number of studies estimating prognostic value; R: References.

Drawbacks

The following drawbacks of the current meta-analysis should considered: (1) there were numerous variables, consisting of dissimilar sample types from PC patients at different stages, cutoffs, and miRNA detection methods, among which the differences in sample type and cutoffs were the main drawbacks; (2) we only selected English articles, perhaps excluding potential papers published in other languages about PC patients with miRNA expression levels and prognostic outcomes; (3) we only chose studies estimating OS, perhaps excluding potential investigations reporting prognosis with other survival results, such as disease-free and recurrence-free survival; (4) the prognostic impact of miRNA expression levels in pancreatic cancer should be adjusted for risk factors that have an important influence on pancreatic cancer prognosis, such as age, educational level, sex, smoking, obesity, heavy alcohol intake, underlying illnesses and family history of cancer, which indicates possible mutations. However, the searched papers may not all contain the very concerned information. Therefore, the impact of bias in predicting miRNAs involved in pancreatic cancer prognosis may occur due to the lack of adjustment for risk factors in a rigorous conclusion.

Insight for future clinical and experimental studies

Notably, this study was the first meta-analysis of the associations between abnormal miRNA levels and prognosis in PC patients. This study provides direction for further clinical and experimental study: (1) joint detection of various miRNA levels could be utilized by clinical workers and other health care providers, which might extremely expand the ability to assess the prognosis of PC patients such that immediate treatment might be supplied; (2) advances and trends regarding miRNA expression levels and the survival time of PC patients could be obviously acquired by the experimental researchers mentioned in Tables 4 and 5. In addition, miRNA molecular mechanisms could be obtained by assessing the data in Table 3; and (3) several contradictory outcomes concerning the prognostic value of miRNAs might be resolved on account of the present work.

CONCLUSIONS

In summary, blood miR-21, miR-451a, miR-1290 and tissue miR-10b, miR-17-5p, miR-21, miR-23a, miR-29c, miR-126, miR-155, miR-203, miR-218, miR-221, miR-222 had significant prognostic value.

MATERIALS AND METHODS

Search strategy

Two independent authors (Fei Zhao and Chao Wei) performed the literature search from 4 online databases, PubMed, EMBASE, Web of Science and Cochrane Database of Systematic Reviews. Afterwards, Yue Zhang reassessed undetermined information. An extensive and comprehensive search was performed utilizing the keywords: ‘microRNA’, ‘miRNA’, ‘miR’, and ‘pancreatic cancer’, ‘pancreatic carcinoma’ and ‘pancreatic adenocarcinoma’. After duplicates were eliminated, 875 reports remained. Accordingly, 671 articles were excluded by titles and abstracts. For the residual 204 studies, 35 full-text studies were removed. The details of the literature selection are shown in Figure 8. The search deadline was June 1, 2019.

Figure 8

Flow diagram of literature search and selection.

Inclusion criteria

The inclusion criteria were as follows: (1) articles on the correlation between miRNA expression level and survival time of PC patients; (2) inclusion of estimated OS outcomes; and (3) full-text in English.

Exclusion criteria

The exclusion criteria were as follows: (1) articles without original data (reviews, letters or laboratory studies); (2) nondichotomous miRNA level; and (3) frequency of studies evaluating OS of miRNA expression level equal or less than 2 in tissue. In addition, on the condition that more than one article was published on the same subjects, the most well-rounded paper was chosen for the present work. Likewise, if both univariate and multivariate analysis of OS were covered, the latter was chosen, as this type of analysis considers interferential factors.

Quality assessment

Fei Zhao and Chao Wei confirmed all qualified studies that analyzed the prognostic value of miRNAs in PC, and Yue Zhang reevaluated undetermined information. Quality assessment for each paper was performed employing the modified Newcastle–Ottawa Scale (NOS) [180]. NOS scores were calculated according to selection, comparability, and outcome. Articles with NOS scores ≥6 were considered high-quality articles [181].

Study selection

The flow chart with details of the study selection process is given in Figure 8.

Study frequency

The frequency of studies estimating the OS of PC patients with and miRNA expressions of PC patients is presented in Tables 4 (blood) and 5 (tissue), and includes the miRNA names, the frequency of included miRNAs, and the reference number.

Study characteristics

The fundamental particulars of the included literature are fully listed in Table 6. On the condition that the data were not offered in the article but just as Kaplan–Meier survival curves, the data were abstracted from the curves, and the generation of HR with 95% CI was next carried out employing the software Engauge Digitizer version 4.1.

Table 6

Characteristics of included studies about pancreatic cancer.

miRNA

Study

Country

Sample

Number

Stage

Cut-off

Method

Follow-up (month)

Result

HR (L/H)

HR (H/L)

95%CI

21

Liu, 2012 [4]

China

Serum

38

I-IV

Median

qRT-PCR

24

OSu

3.26

1.47-7.23

21

Wang, 2013 [5]

China

Serum

177

III-IV

Median

qRT-PCR

30

OSm

1.71

1.15-2.54

21

Abue, 2015 [6]

Japan

Plasma

24

I-IV

850

qRT-PCR

>20

OSu

5.99

1.95-18.40

21

Goto, 2018 [7]

Japan

Serum

32

I-IV

Median

qRT-PCR

>40

OSu

2.57

0.90-7.35

21

Kawamura, 2019 [8]

Japan

Plasma

55

I-II

Mean

qRT-PCR

60

OSm

3.10

1.19-9.10

196a

Kong, 2010 [16]

China

Serum

35

I-IV

-5.22

qRT-PCR

>16

OSu

3.37

1.14-9.97

196a

Yu, 2017 [17]

China

Plasma

31

None

Median

qRT-PCR

15

OSm

0.99

0.92-1.06

451a

Goto, 2018 [7]

Japan

Serum

32

I-IV

Median

qRT-PCR

>40

OSu

1.45

0.63-3.31

451a

Takahasi, 2018 [23]

Japan

Plasma

50

I-II

Median

qRT-PCR

54

OSm

3.20

1.07-11.94

451a

Kawamura, 2019 [8]

Japan

Plasma

55

I-II

Mean

qRT-PCR

60

OSm

3.60

1.13-11.31

1290

Li, 2013 [24]

USA

Serum

56

I-III

Median

qRT-PCR

>80

OSu

1.63

0.66-3.98

1290

Tavano, 2013 [26]

Italy

Plasma

167

I-IV

ROC

ddPCR

>40

OSu

1.40

1.00-1.96

10b

Nakata, 2011 [35]

Japan

FFPE

115

None

None

qRT-PCR

101

OSu

2.19

1.37-3.50

10b

Preis, 2011 [36]

Lebanon

FFPE

95

I-IV

5000

ISH

36

OSu

3.59

1.73-7.43

10b

Nguyen, 2016 [37]

USA

Frozen

55

I-II

1.5 fold

qRT-PCR

34.25

OSu

1.12

0.54-2.32

10b

Yang, 2017 [38]

Germany I

Frozen

69

I-IV

None

qRT-PCR

>60

OSu

1.99

1.07-3.73

Germany II

Frozen

41

I-IV

None

qRT-PCR

>60

OSu

0.81

0.39-1.67

17-5p

Yu, 2010 [39]

Japan

FFPE

80

I-IV

5.69

qRT-PCR

100

OSu

1.85

1.08-3.15

17-5p

Gu, 2016 [40]

China

Tissue

58

I-IV

None

qRT-PCR

>50

OSu

1.89

0.98-3.64

17-5p

Zhu, 2018 [41]

China

Tissue

26

None

None

qRT-PCR

>50

OSu

2.18

0.77-6.17

21

Dillhoff, 2008 [43]

USA

FFPE

80

None

Median

ISH

>60

OSu

4.23

2.17-8.25

21

Giovannetti, 2010 [44]

Italy

Frozen

59

I-IV

Median

qRT-PCR

60.5

OSu

2.31

1.30-4.10

21

Hwang, 2010 [45]

Korea and Italy

Tissue

97

II-IV

Median

qRT-PCR

>60

OSm

3.16

1.67-6.02

21

Jamieson, 2011 [46]

UK

Frozen

48

None

Median

qRT-PCR

>50

OSm

3.22

1.21-8.58

21

Nagao, 2012 [47]

Japan

FFPE

65

None

Mean

qRT-PCR

>40

OSm

2.12

1.07-4.20

21

Caponi, 2013 [48]

Italy and UK

FFPE

57

None

Median

qRT-PCR

117.3

OSm

3.28

1.52-7.05

21

Kadera, 2013 [49]

USA

Tissue

145

I-II,IV

Median

ISH

100

OSu

1.06

0.70-1.60

21

Ma, 2013 [50]

China

Frozen

78

I-IV

2 fold

qRT-PCR

>25

OSm

2.60

1.15-5.87

21

Papaconstantinou, 2013 [51]

Greece

FFPE

88

None

Mean

qRT-PCR

>60

OSm

3.93

1.25-12.35

21

Wang, 2013 [5]

China

Tissue

65

III-IV

Median

qRT-PCR

60

OSm

2.24

1.14-4.37

21

Donahue, 2014 [52]

USA I

FFPE

94

I-IV

Median

ISH

72

OSm

1.70

1.03-2.82

USA II

FFPE

87

I-IV

Median

ISH

72

OSu

0.94

0.59-1.49

21

Frampton, 2014 [53]

UK

Frozen

91

IIA,IIB

Median

qRT-PCR

>48

OSu

1.85

1.08-3.18

21

Mitsuhashi, 2015 [54]

Japan

FFPE

283

I-IV

75%

qRT-PCR

48

OSu

1.60

1.07-2.39

21

Vychytilova-Faltejskova, 2015 [55]

Czech

FFPE

74

None

27.15

qRT-PCR

>40

OSu

1.76

1.08-2.86

21

Morinaga, 2016 [56]

Japan

FFPE

39

None

Median

ISH

114.1

OSu

1.80

0.90-3.60

21

Benesova, 2018 [57]

Czech

FFPE

91

II-IV

Median

qRT-PCR

18

OSu

1.60

1.02-2.50

21

Xi, 2018 [58]

TCGA

Tissue

169

I-IV

Median

Downloaded

60

OSu

1.47

1.00-2.16

21

Zhang, 2018 [59]

GEO

Tissue

174

I-IV

Median

Downloaded

>80

OSu

1.89

1.37-2.62

21

Zhao, 2018 [60]

Japan

Tissue

63

0-IV

None

qRT-PCR

>60

OSu

2.99

1.25-7.14

23a

Ma, 2013 [50]

China

Frozen

78

I-IV

2 fold

qRT-PCR

>25

OSu

1.64

0.71-3.79

23a

Frampton, 2014 [53]

UK

Frozen

91

IIA,IIB

Median

qRT-PCR

>48

OSu

1.87

1.07-3.16

23a

Diao, 2018 [61]

China

Frozen

30

None

Median

qRT-PCR

25

OSu

2.55

1.10-5.92

23a

Wu, 2018 [62]

China

Tissue

52

None

3.5

qRT-PCR

>50

OSu

3.64

1.56-8.47

29c

Jamieson, 2011 [46]

UK

Frozen

48

None

Median

qRT-PCR

>50

OSm

1.89

0.68-5.26

29c

Jiang, 2015 [69]

TCGA

Frozen

132

I-IV

None

Downloaded

>50

OSu

1.59

1.15-2.18

29c

Zou, 2015 [70]

China

FFPE

105

I-IV

Median

qRT-PCR

30

OSm

1.14

1.00-1.29

29c

Wang, 2019 [33]

GEO

Tissue

178

I-IV

None

Downloaded

>80

OSu

1.67

1.05-2.63

126

Liang, 2018 [68]

TCGA

FFPE

175

I-IV

Median

Downloaded

>83.3

OSm

1.58

1.04-2.39

126

Liao, 2018 [27]

TCGA

Tissue

112

I-II

None

Downloaded

>40

OSu

1.51

0.98-2.32

126

Yu, 2018 [82]

TCGA

Tissue

168

I-II

Median

Downloaded

72.4

OSm

1.55

1.07-2.24

155

Ma, 2013 [50]

China

Frozen

78

I-IV

2 fold

qRT-PCR

>25

OSm

1.37

0.52-3.58

155

Papaconstantinou, 2013 [51]

Greece

FFPE

88

None

Mean

qRT-PCR

>60

OSm

3.14

1.09-9.09

155

Mikamori, 2017 [14]

Japan

Tissue

45

I-II

Mean

qRT-PCR

>72

OSm

2.63

1.07-6.46

200c

Yu, 2010 [109]

Japan

FFPE

99

I-IV

0.64

qRT-PCR

101

OSm

2.25

1.10-4.60

200c

Paik, 2015 [110]

Korea

FFPE

84

IB-III

0.65

qRT-PCR

140

OSm

0.56

0.34-0.93-

200c

Liu, 2016 [111]

China

Tissue

75

I-IV

Mean

qRT-PCR

60

OSm

2.31

1.73-6.38

203

Ikenaga, 2010 [112]

Japan

FFPE

107

I-IV

0.054

qRT-PCR

98

OSm

1.21

0.72-2.07

203

Shao, 2017 [113]

TCGA

Tissue

161

I-IV

None

Downloaded

>80

OSu

2.18

1.31-2.49

203

Shi, 2018 [114]

TCGA

Tissue

177

None

Median

Downloaded

>72

OSu

1.24

1.10-1.39

203

Zhang, 2018 [59]

GEO

Tissue

174

I-IV

Median

Downloaded

>80

OSu

2.27

1.57-3.27

218

Li, 2013 [121]

China

FFPE

28

None

1.5 fold

qRT-PCR

>20

OSu

1.86

0.80-4.35

218

Zhu, 2014 [122]

China

Frozen

113

I-IV

Mean

qRT-PCR

>50

OSm

2.12

1.51-2.50

218

Li, 2015 [123]

China

Frozen

107

I-IV

Median

qRT-PCR

60

OSm

7.24

2.01-18.28

221

Jamieson, 2011 [46]

UK

Frozen

48

None

Median

qRT-PCR

>50

OSm

0.92

0.34-2.54

221

Ma, 2013 [50]

China

Frozen

78

I-IV

2 fold

qRT-PCR

>25

OSm

2.00

0.87-4.62

221

Sarkar, 2013 [125]

USA

FFPE

24

None

None

qRT-PCR

>83.3

OSu

1.36

0.52-3.51

221

Wang, 2016 [126]

Germany

Frozen

37

I-II

66.7%

qRT-PCR

>40

OSu

2.85

1.20-6.77

222

Schultz, 2012 [28]

Denmark

FFPE

225

I-II

Median

qRT-PCR

24

OSm

1.39

1.06-1.84

222

Lee, 2013 [127]

China

Frozen

60

I-IV

Median

qRT-PCR

15

OSm

5.16

1.16-22.91

222

Wang, 2016 [126]

Germany

Frozen

37

I-II

None

qRT-PCR

>40

OSu

1.86

0.79-4.37

HR (L/H): hazard ratios of low expression versus high expression of miRNAs; HR (H/L): hazard ratios of high expression versus low expression of miRNAs; CI: confidence intervals; TCGA: The Cancer Genome Atlas; GEO: Gene Expression Omnibus; FFPE: formalin-fixed paraffin-embedded; qRT-PCR: quantitative real-time polymerase chain reaction; ddPCR: droplet digital polymerase chain reaction; ISH: in-situ hybridization; OS: overall survival; uUnivariate analysis; mMultivariate analysis.

Statistical analysis

All analyses were carried out employing Stata version 13.0 (StataCorp, College Station, TX, USA). OS was the primary and unique guideline for the prognosis of PC patients with miRNAs. The HR was regarded as significant at the P <0.05 level in case of the 95% CI not including the value 1. Furthermore, a single miRNA was considered a strong candidate if its HR was over 2. Most analyses used random-effects models other than fixed-effects models because of the dissimilarity of sample types from PC patients at dissimilar stages, cutoffs, and miRNA methods in single studies. Begg’s funnel plot was used to estimate publication bias. A two-tailed P value less than 0.05 was regarded as significant. If publication bias occurred, the trim and fill method was conducted. The sensitivity analysis was employed to assess how sensitive the entire effect size was to remove the impact of single investigations. If the point estimation was outside of the 95% CI of the entire effect value after it was excluded from the entire analysis, a single study was deemed to have undue influence.