Telomerase Activity as a Marker for Differential Diagnosis of Pancreatic Adenocarcinoma: A Systematic Review and Meta-Analysis

Abstract

Background

Studies evaluating the role of telomerase activity in pancreatic adenocarcinoma are inconsistent and a systemic review of the available literature may shed new light on this issue.

Objective

To systematically review the usefulness of telomerase activity in distinguishing pancreatic cancer from other pancreatic diseases.

Methods

A comprehensive search of the PubMed and Embase databases was conducted to identify eligible studies. Only studies evaluating telomerase activity in patients with suspected or previously diagnosed pancreatic adenocarcinomas versus nonpancreatic adenocarcinomas and published in English with a sufficient number of cases were included. The hierarchical summary receiver operating characteristic (HSROC) model was used to establish the potential value of telomerase activity in the diagnosis of pancreatic adenocarcinoma.

Results

A total of 19 studies qualified for this meta-analysis. In distinguishing pancreatic adenocarcinoma from benign diseases, the pooled sensitivity and specificity of telomerase activity were 0.81 (95% CI, 0.68-0.90) and 0.97 (95% CI, 0.93-0.98), respectively; the diagnostic odds ratio (DOR) was 126.62 (95% CI, 49.94-320.99); beta was −1.16 (95% CI, −3.62-1.29), Z was −0.93, p was 0.35>0.1, and lambda was 6.86 (95% CI, 1.01-12.70). In distinguishing pancreatic adenocarcinoma from chronic pancreatitis, the pooled sensitivity and specificity of telomerase activity were 0.77 (95% CI, 0.61-0.88) and 0.97 (95% CI, 0.91-0.99), respectively; DOR was 117.28 (95% CI, 32.25-426.53); beta was −0.38 (95% CI, −1.89-1.13), Z was −0.49, p was 0.62>0.1, and lambda was 5.30 (95% CI, 3.37-7.24).

Conclusions

The present meta-analysis demonstrates that telomerase activity could be a useful biomarker for the differential diagnosis of pancreatic adenocarcinoma and benign pancreatic diseases.

Keywords

Chronic pancreatitis Meta-analysis Pancreatic adenocarcinoma Telomerase activity

Introduction

Pancreatic cancer, with approximately 90% of pancreatic tumors being pancreatic adenocarcinoma, is a highly malignant disease and has an extremely poor prognosis with a 6.7% 5-year survival rate (1, 2). Its poor prognosis is mainly due to the failure of diagnosis at an early stage. Early diagnosis and radical resection can improve the survival rate significantly. The incidence of pancreatic adenocarcinoma ranks second after colorectal cancer among gastrointestinal malignancies in industrialized countries (3). It is estimated that there will be 46,420 new cases of pancreatic cancer and 39,590 deaths from this disease in the United States in 2014 (4). Unfortunately, only 10%-15% of patients with pancreatic cancer have resectable disease at diagnosis (5). Thus, early diagnosis is the key to improving survival, but preoperative diagnosis of pancreatic lesions remains challenging.

Telomerase, which is composed of telomerase RNA (hTR), telomerase protein 1 (TP1) and telomerase reverse transcriptase (hTERT), is an important reverse transcriptase that is responsible for the synthesis of telomeric DNA (6). Several studies suggest that hTERT is the most important unit and reflects the activity of nuclear telomerase (7). Telomerase activity is absent in most somatic cells, but not in germ cells and stem cells (8). Interestingly, it is highly expressed in the majority of human cancers including lung cancer, hepatocellular carcinoma and breast cancer (9), suggesting that it plays an important role in human carcinogenesis. Telomerase has also been reported to be aberrantly expressed in patients with pancreatic adenocarcinoma (10, 11). However, the findings from studies evaluating the role of telomerase activity in pancreatic adenocarcinoma are inconsistent. Some reports indicate that telomerase activity can predict the prognosis of pancreatic adenocarcinoma (12) and can be a therapeutic target (13). Although some reports have demonstrated that telomerase activity is an effective diagnostic biomarker for pancreatic adenocarcinoma (14, 15), bladder cancer (16), prostate cancer (17), and cervical cancer (18), its accuracy remains uncertain. To the best of our knowledge, there has been no comprehensive meta-analysis to evaluate the diagnostic accuracy of telomerase activity for pancreatic adenocarcinoma. Hence, it is necessary to conduct a systematic review and meta-analysis to compare telomerase activity with other diagnostic markers and to establish its value in the diagnosis of pancreatic adenocarcinoma.

Methods

Database Search

We searched the PubMed and Embase literature databases for publications prior to June 10, 2015, using the following keywords: “pancreatic cancer”, “pancreatic neoplasms”, “pancreatic carcinoma”, “pancreas cancer”, “telomerase”, “human telomerase reverse transcriptase”, “human telomerase RNA”, and their combinations. The search was limited to human studies that were published in English. If the same population was the subject of multiple publications, only the most complete or most recent publication was selected. We also hand-searched reference lists from publications to identify additional potentially relevant studies. This process was performed iteratively until no additional articles could be identified.

Inclusion and Exclusion Criteria

We included studies that met the following criteria: (1) the purpose of study was to evaluate telomerase activity in patients with suspected or previously diagnosed pancreatic adenocarcinomas, compared with chronic pancreatitis or benign lesions; (2) the total number of cases was more than 20; (3) the final diagnoses were confirmed by histology, pathology, and imaging data or follow-up longer than 6 months; and (4) the publication was in English. Studies were excluded on the basis of the following criteria: (1) the publication belonged to the categories of case reports, case-only studies, letters, reviews or meta-analyses; (2) the sample size was not sufficient (n<20); (3) the study concerned other kinds of pancreatic cancer or other controls; (4) the study was published in duplication or with data deficiency or duplication; and (5) the full publication was not in English.

Data Extraction and Management

Data from each of the included studies were extracted independently by 2 reviewers (HG and XX) and any conflicting data were settled by discussion between the reviewers until consensus was reached. Data on first author, publication year, study region, ethnicity, period of study, age, detection method, specimen source, specimen collection, diagnostic criteria, source of controls, number of cases and controls, true positive, false positive, true negative, and false negative results were extracted and tabulated in a standard format. For publications including multiple diseases, cases of other kinds of pancreatic cancer were excluded.

Assessment of Quality of Evidence in Included Studies

The aforementioned 2 reviewers (HG and XX) independently assessed the quality of evidence in the included studies using the tool QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies-2) (19), which comprised 4 key domains including (i) patient selection, (ii) index test, (iii) reference standard flow and timing of the index tests, and (iv) reference standard for judging bias and applicability. The assessment process was conducted with the RevMan 5.2 software. We resolved disagreement about the selection of studies by discussion and consensus among the investigators.

Statistical Analysis

For each of the included studies, the data were abstracted as a 2 × 2 contingency table recording the presence of telomerase activity (positive vs. negative) and the histological diagnosis (benign or chronic pancreatitis). We calculated the true positive, true negative, false positive and false negative rates. When the table had one 0 cell, a correction of the calculation was made by adding a value of 0.5 to all cells of the table to avoid potential problems in odds calculation for sensitivities or specificities of 100%; if a study contained two 0 cells, it was excluded from the meta-analysis (20). The hierarchical summary receiver operating characteristic (HSROC) model was used to establish the potential value of telomerase activity in the diagnosis of pancreatic adenocarcinoma. The beta (β) and Z values assess the symmetry of SROC; the lambda (λ) value reflects the veracity of the diagnostic test discriminant ability. The pooled sensitivity, pooled specificity, diagnostic odds ratio (DOR), positive likelihood ratio (LR⁺), and negative likelihood ratio (LR^-) were used to demonstrate the potential value of telomerase activity in discriminating pancreatic cancer from benign diseases (21). Publication bias was evaluated using the Deeks test. A p value of less than 0.1 (p<0.1) indicated an existing publication bias.

The statistical analyses were performed using STATA 12.0 (StataCorp, College Station, TX, USA). The whole statistical process was conducted independently by the reviewers (HG and YP); discrepancy in results was resolved by discussion with XX. A p value of less than 0.05 (p<0.05) was considered statistically significant with the exception of the Deeks test, as mentioned above.

Results

Included Studies and Quality Assessment

The search of the aforementioned databases identified 207 articles. After the screening of titles and abstracts, 172 articles were excluded and 35 potentially eligible articles were retrieved for analysis of the full text. Among them, 19 eligible articles (8, 10, 11, 14, 15, 22-23-24-25-26-27-28-29-30-31-32-33-34-35) qualified for the meta-analysis. Of the 16 excluded, 9 were not appropriate in their design (12, 36-37-38-39-40-41-42-43), 3 were duplicated publications (9, 44, 45), 2 included fewer than 20 cases (46, 47), and 2 were review articles (48, 49) (Fig. 1).

Fig. 1

Flowchart of article screening and selection processes.

The characteristics of the 19 articles for pancreatic adenocarcinoma and index tests, detection method, specimen source, reference standard, and number of subjects are shown in Tables I and II. For the differential diagnosis of pancreatic cancer versus benign lesions, a total of 798 subjects were incorporated, while 567 subjects were included in the differential diagnosis of pancreatic cancer versus chronic pancreatitis.

Table I

Characteristics of included studies on telomerase activity and pancreatic adenocarcinoma

Author (ref.)	Year	Country	Period	Age range, mean age (years)	TA/hTERTmRNA detection method	Specimen source	Specimen collection	Final diagnosis	Total PACs	Benign lesions
Iwao et al (10)	1997	Japan	1996-1997	34-78 (62)	TRAP	Cytological specimens	ERPDB	1	15	17 (CP = 12, pancreatic cysts = 2, islet cell tumor = 1, pancreatic abscess = 1, solid and cystic tumor = 1)
Tsutsumi et al (35)	1997	Japan	Not reported	Not reported	TRAP	Tissue sections	Surgery	1	38	7 normal pancreas
Myung et al (8)	1999	Korea	Not reported	46-77 (63)	TRAP	Pancreatic juice	ERP	1, 2, 4, 5	12	19 (CP = 11, normal pancreas = 8)
Pearson et al (14)	2000	USA	Not reported	Not reported	TRAP	Tissue sections	Surgery/EUS-FNA	1	24	6 normal pancreas
Balcom et al (11)	2001	USA	Not reported	Not reported	TRAP/TRAP-PCR	Tissue sections	Surgery	1	21	16 (CP = 7, serous cystadenomas = 5, benign mucinous cystic neoplasms = 4)
Seki et al (25)	2001	Japan	Not reported	57-87	RT-PCR	Pancreatic juice	ERP	1, 2, 3, 4	17	19 (CP = 12, normal pancreas = 7)
Büchler et al (26)	2001	USA	Not reported	31-78 (52)	TRAP	Tissue sections	Surgery	1	29	57 (CP = 36, normal pancreas = 21)
Zhou et al (27)	2003	China	Not reported	55-74 (63)	TRAP	Tissue sections	Surgery	1	24	24 normal pancreas
Uemura et al (28)	2003	Japan	Not reported	31-76	TRAP	Tissue sections	Surgery	1	35	5 CP
Zhou et al (29)	2004	China	1999-2000	22-75 (58.5)	PCR-TRAP-ELISA	Cytological specimens	ERPDB	1, 2, 4, 5	18	3 CP
Ohuchida et al (30)	2004	Japan	2001-2002	43-78	TRAP	Pancreatic juice	ERCP	1, 5	11	13 CP
Mizumoto et al (15)	2005	Japan	Not reported	Not reported	TRAP	pancreatic juice	ERP	1	24	46 (CP = 23, adenoma = 23)
Mishra et al (31)	2005	USA	2002-2003	37-93 (67)	TRAP	Cytological specimens	EUS-FNA	1, 3	47	14 CP
Ohuchida et al (32)	2006	Japan	2001-2003	43-82	RT-PCR	Pancreatic juice	ERCP	1, 5	23	36 CP
Kawahara et al (33)	2007	Japan	2002-2003	70.5 ± 11.6	RT-PCR	Biliary lavage fluid	ERCP	1, 2	10	20 normal pancreas
Hashimoto et al (22)	2007	Japan	1990-2005	35-84 (66.5)	RT-PCR	Tissue sections	Surgery	1	15	10 CP
Hashimoto et al (24)	2007	Japan	1997-2005	31-83 (64)	IHC	Tissue sections	Surgery	1	42	12 CP
Nakashima et al (23)	2009	Japan	1999-2007	43-77	IHC	Pancreatic juice	ERCP	1	55	7 CP
Zhou et al (34)	2011	China	2007-2008	24-60 (48.9)	RT-PCR	Circulating cancer cells	Immunomagnetic separation	1	25	15 benign diseases

Final diagnosis: 1 = histological examination; 2 = diagnostic imaging; 3 = cytological examination; 4 = clinical history; 5 = followed up. PACs = pancreatic adenocarcinomas; CP = chronic pancreatitis; ERP = endoscopic retrograde pancreatography; ERPDB = endoscopic retrograde pancreatic duct brushing; EUS-FNA = endoscopic ultrasound-guided fine-needle aspiration; ERCP = endoscopic retrograde cholangiopancreatography; TRAP = telomerase repeat amplification protocol; RT-PCR = reverse transcriptase polymerase chain reaction; IHC = immunohistochemistry.

Table II

Contingency tables for pancreatic adenocarcinoma and chronic pancreatitis

Study (reference)	PC vs. benign lesions				PC vs. CP
Study (reference)	TP	FP	FN	TN	TP	FP	FN	TN
Iwao et al (10)	13	0	2	17	13	0	2	12
Tsutsumi et al (35)	32	0	6	7
Myung et al (8)	11	2	1	17	11	2	1	9
Pearson et al (14)	24	0	0	6
Balcom et al (11)	17	0	4	16	17	0	4	7
Seki et al (25)	15	2	2	17	15	2	2	10
Büchler et al (26)	10	1	19	56	10	1	19	35
Zhou et al (27)	21	3	3	21
Uemura et al (28)	33	0	4	5	33	0	4	5
Zhou et al (29)	14	0	4	3	14	0	4	3
Ohuchida et al (30)	1	0	10	13	1	0	10	13
Mizumoto et al (15)	20	1	4	45	20	0	4	23
Mishra et al (31)	37	1	10	13	37	1	10	13
Ohuchida et al (32)	5	0	18	36	5	0	18	36
Kawahara et al (33)	4	1	6	9
Hashimoto et al (22)	14	0	1	10	14	0	1	10
Hashimoto et al (24)	37	0	5	12	37	0	5	12
Nakashima et al (23)	46	0	9	7	46	0	9	7
Zhou et al (34)	25	0	0	15
Total	798				567

PC = pancreatic adenocarcinoma; CP = chronic pancreatitis; TP: true positive; FP: false positive; FN: false negative; TN: true negative.

Figure 2 shows the quality of the 19 eligible articles assessed by QUADAS-2 criteria. Thirteen studies (8, 14, 22-23-24-25-26-27-28-29, 33-34-35) were regarded as being at high risk of patient selection bias due to the case-control design; 9 studies (8, 11, 14-15, 25-26-27-28, 35) enrolled nonconsecutive patients; and 1 study (29) was classified as high risk in the index test domain due to the lack of the pre-specified threshold. All patients of 7 studies (8, 25, 29-30-31-32-33) received more than one reference standard and therefore these studies were classified as high risk in the flow and timing domain.

Fig. 2

Risk of bias and applicability concerns graph: the review authors’ judgments about each domain are presented as percentages across the included studies (A). Risk of bias and applicability concerns summary: the review authors’ judgments about each domain are presented for each included study (B).

Accuracy of Telomerase Activity in Differential Diagnosis of Pancreatic Adenocarcinoma versus Benign Lesions

The telomerase activity analysis was feasible in all 798 patients. The sensitivity for pancreatic adenocarcinoma detection ranged from 0.22 to 1.00 and the specificity ranged from 0.88 to 1.00; β = −1.16 (95% confidence interval [CI], −3.62-1.29), Z = −0.93, p = 0.35>0.1, λ = 6.86 (95% CI, 1.01-12.70). The pooled sensitivity, specificity, and DOR were 0.81 (95% CI, 0.68-0.90), 0.97 (95% CI, 0.93-0.98), and 126.62 (95% CI, 49.94-320.99), respectively. LR⁺ and LR^- were 24.51 (95% CI, 11.88-50.56) and 0.19 (95% CI, 0.11-0.34), respectively (Fig. 3A and 3C). The Deeks test revealed no significant publication bias (t = −0.07, p = 0.943; Fig. 4A).

Fig. 3

HSROC model for evaluating the potential value of telomerase activity in the diagnosis of pancreatic adenocarcinoma vs. benign lesions (A) HSROC curves summarizing the diagnostic performance of telomerase activity in distinguishing pancreatic adenocarcinoma from benign lesions (B) and pancreatic adenocarcinoma vs. chronic pancreatitis (C) and chronic pancreatitis (D). Each study is represented by an open circle whose size is proportional to the inverse standard error of sensitivity and specificity. The dark square represents the summary estimate of test accuracy, with the surrounding dashed zone outline denoting the 95% confidence region around this estimate.

Fig. 4

Deeks’ funnel plot of publication bias for pancreatic adenocarcinoma vs. benign lesions (A) and pancreatic adenocarcinoma vs. chronic pancreatitis (B).

Furthermore, the subgroup analysis was stratified by specimen source (tissue sections: β = −0.25 [95% CI, −2.60-2.10]; Z = −0.21, p = 0.84>0.1; λ = 5.98 [95% CI, 3.77-8.20]; pooled sensitivity = 0.86 [95% CI, 0.72-0.94]; pooled specificity = 0.98 [95% CI, 0.80-0.998]; DOR = 295.68 [95% CI, 22.43-3897.01]; LR⁺ = 41.11 [95% CI, 3.79-446.58]; LR^- = 0.14 [95% CI, 0.06-0.30]; pancreatic juice: β = −0.09 [95% CI, −1.49-1.31]; Z = −0.13, p = 0.90>0.1; λ = 4.92 [95% CI, 3.38-6.46]; pooled sensitivity = 0.68 [95% CI, 0.33-0.91]; pooled specificity = 0.98 [95% CI, 0.86-0.998]; DOR = 117.70 [95% CI, 18.34-755.53]; LR⁺ = 37.86 [95% CI, 5.43-263.97]; LR^- = 0.32 [95% CI, 0.12-0.88]) (Fig. 5). Other studied specimen sources involving cytological specimens, biliary lavage fluid, and circulating cancer cells could not be combined because the HSROC model requires a minimum of 4 studies.

Fig. 5

HSROC model for evaluating the potential value of telomerase activity in the diagnosis of pancreatic adenocarcinoma from tissue specimens (A) and pancreatic juice (C). HSROC curves summarizing the diagnostic performance of telomerase activity for the diagnosis of pancreatic adenocarcinoma from tissue specimens (B) and pancreatic juice (D). Each study is represented by an open circle whose size is proportional to the inverse standard error of sensitivity and specificity. The dark square represents the summary estimate of test accuracy, with the surrounding dashed zone outline denoting the 95% confidence region around this estimate.

Accuracy of Telomerase Activity in Differential Diagnosis of Pancreatic Adenocarcinoma versus Chronic Pancreatitis

Of the 19 studies, 14 (8, 10, 11, 15, 21-22-23-24-25, 27-28-29-30-31) involved 678 subjects in whom pancreatic cancer was differentiated from chronic pancreatitis; β = −0.38 (95% CI, −1.89-1.13), Z = −0.49, p = 0.62>0.1. Compared with the above results in studies differentiating pancreatic cancer from benign lesions, the λ value (5.30, 95% CI, 3.37-7.24) was slightly lower. The pooled sensitivity, specificity, and DOR were 0.77 (95% CI, 0.61-0.88), 0.97 (95% CI, 0.91-0.99), and 117.28 (95% CI, 32.25-426.53), respectively. LR⁺ and LR^- were 27.81 (95% CI, 8.15-94.87) and 0.24 (95% CI, 0.13-0.42), respectively (Fig. 3B and 3D). Deeks test again revealed no significant publication bias (t = 0.30, p = 0.767; Fig. 4B).

Discussion

The results of this meta-analysis indicated that telomerase activity had good diagnostic accuracy to distinguish benign lesions from malignant tumors. For the differential diagnosis between pancreatic adenocarcinoma and chronic pancreatitis, telomerase activity also demonstrated good diagnostic value, although the values were not as good as those observed for the differential diagnosis between pancreatic cancer and benign lesions. In the specimen subgroup analysis, telomerase activity had better diagnostic value with tissue specimens than pancreatic juice. However, all sensitivity values were not higher as specificity for the effective clinical application.

The relationship between telomerase activity and pancreatic adenocarcinoma has been intensively studied, but the conclusions are not consistent. Some studies showed no obvious link between telomere length and pancreatic cancer (21, 34). Others (11, 26, 34) found that increased telomerase activity was related to clinical manifestation of pancreatic cancer, a finding that was contradicted by yet other studies (25, 27, 45). With regard to the diagnostic value, a larger number of studies have indicated that positive telomerase activity is found in pancreatic adenocarcinoma and that detecting telomerase activity can help diagnose pancreatic adenocarcinoma (14, 30). The results of the present meta-analysis suggested that telomerase activity is a powerful diagnostic tool for pancreatic cancer. Several types of cells such as hematopoietic stem cells, lymphocytes, basal skin cells and intestinal stem cells can also be characterized by low positive telomerase activity. Therefore, when detection of telomerase activity is used in the diagnosis of pancreatic cancer, close attention should be paid to its false positivity, especially in chronic pancreatitis and in pancreatic parenchyma where extensive lymphocyte infiltration may occur. The specimen source may also be a contributing factor, and detection of telomerase activity is more reliable in tissue specimens than in pancreatic fluid, because the composition of the latter is complex and unstable.

For a better comprehensive understanding of the effects of various diagnostic methods on the diagnostic accuracy, we systematically summarized the evidence from other meta-analyses and studies (Tab. III). Benign pancreatic diseases such as chronic pancreatitis and pancreatic cysts can be a confounding factor. Several imaging technologies including ultrasound (50-51-52), computed tomography (53, 54), magnetic resonance imaging (55), and endoscopic retrograde cholangiopancreatography (50) play an important role in identifying pancreatic cancer, but none of them is sufficiently sensitive and specific or has a high positive likelihood ratio. Meanwhile, a wide range of biochemical tumor markers have been found to be elevated in patients with pancreatic cancer (56). For example, CA199 is widely used for pancreatic cancer diagnosis, but it lacks specificity on account of its increased values in benign or malignant biliary diseases and the overall diagnostic accuracy is low (57, 58). CA242 (macrophage inhibitory cytokine-1) also has some disadvantages (58, 59). Additionally, of the genetic alterations, mutated TP53 is present in only 50%-70% of patients (60), while KRAS mutation is detectable in pancreatic juice of patients with pancreatic cancer but also in pancreatic juice of those with chronic pancreatitis, in smokers, and in patients with pancreatic intraepithelial neoplasia (61). A meta-analysis revealed that the sensitivity of KRAS mutation is limited, i.e., 0.68 (95% CI, 0.66-0.71) (62). Therefore, these methods are not recommended for the diagnosis of pancreatic carcinoma.

Table III

Accuracy of detection methods for pancreatic cancer (evidence from meta-analyses)

Detection method	Disease	Author (reference)	Year	Pooled sensitivity (95% CI)	Pooled specificity (95% CI)	Pooled positive likelihood ratio (95% CI)	Pooled negative likelihood ratio (95% CI)	Pooled diagnostic odds ratio (95% CI)	SROC	^*Q
CA199	Pancreatic cancer	Huang et al (57)	2014	0.80 (0.77-0.82)	0.80 (0.77-0.82)	3.82 (2.49-5.86)	0.28 (0.22-0.34)	14.79 (8.55-25.59)	0.87	0.80
EUS				0.767 (0.717-0.813)	0.917 (0.863-0.955)	-	-	38.94 (19.06-79.57)	0.927	0.862
ERCP	Pancreatic neoplasms	Li et al (50)	2014	0.579 (0.510-0.638)	0.906 (0.852-0.945)	-	-	23.18 (12.11-44.36)	0.897	0.828
EUS plus ERCP				0.799 (0.733-0.855)	0.942 (0.878-0.978)	-	-	74.88 (26.25-213.59)	0.954	0.896
MIC-1	Pancreatic cancer	Chen et al (59)	2014	0.79 (0.77-0.82)	0.86 (0.84-0.88)	6.20 (1.24-30.91)	-	35.73 (18.52-68.93)	0.92	-
EUS	PDAC vs. PIM	Li et al (51)	2013	0.96 (0.94-0.97)	0.73 (0.66-0.79)	-	-	-	0.8227	0.756
KRAS	Pancreatic carcinoma	Liu et al (62)	2013	0.68 (0.66-0.71)	0.87 (0.85-0.88)	4.54 (3.47-5.94)	0.37 (0.30-0.44)	14.90 (10.02-22.15)	0.86	0.79
PET	Pancreatic cancer	Wang et al (63)	2013	0.91 (0.88-0.93)	0.81 (0.75-0.85)	-	-	-	-	-
CA242	Pancreatic cancer	Gui et al (58)	2014	0.719 (0.690-0.746)	0.868 (0.849-0.885)	-	-	16.261 (0.939-26.605)	0.8340	0.7663
CA19-9	Pancreatic cancer	Gui et al (58)	2014	0.803 (0.777-0.826)	0.802 (0.780-0.823)	-	-	15.637 (0.279-26.352)	0.8667	0.7972
EUS	Pancreatic cancer	Hu et al (52)	2012	0.97 (0.95-0.98)	0.76 (0.69-0.82)	3.71 (2.72-5.07)	0.05 (0.02-0.13)	71.86 (29.03-177.91)	0.9529	0.8945
EUS-FNA	Solid pancreatic mass	Puli et al (65)	2013	0.868 (0.855-0.879)	0.958 (0.946-0.967)	15.2 (8.5-27.3)	0.17 (0.13-0.21)	98.5 (54.5-177.7)	0.94	0.8827
microRNA	Pancreatic cancer	Wan et al (64)	2012	0.89 (0.86-0.91)	0.93 (0.90-0.95)	11.62 (5.75-23.50)	0.14 (0.08-0.24)	115.13 (33.73-351.28)	0.97	0.9216
EUS-FNA + cytology classification 1	Solid pancreatic neoplasms	Hewitt et al (66)	2012	0.85 (0.84-0.86)	0.98 (0.97-0.99)	21 (12-36)	0.17 (014-0.21)	130 (74-229)	0.95	-
EUS-FNA + cytology classification 2	Solid pancreatic neoplasms	Hewitt et al (66)	2012	0.91 (0.90-0.92)	0.94 (0.93-0.96)	-	-	-	0.96	-
DWI	Pancreatic cancer	Wu et al (55)	2012	0.87 (0.82-0.81)	0.83 (0.71- 0.91)	5.84 (4.59-7.42)	0.24 (0.17-0.33)	-	-	-
PET/CT				0.85 (0.74-0.92)	0.91 (0.71-0.98)	9.53 (2.41-37.65)	0.17 (0.09-0.32)	-	-	-
CT	Pancreatic cancer	Orlando et al (53)	2004	0.81 (0.72-0.88)	0.66 (0.53-0.77)	-	-	-	0.8242
EUS	Vascular invasion in pancreatic cancer	Yang et al (54)	2014	0.72 (0.67-0.77)	0.89 (0.86-0.92)	5.14 (3.14-8.40)	0.36 (0.25-0.52)	-	0.9037	0.8352
CT				0.63 (0.58-0.67)	0.92 (0.90-0.94)	6.21 (3.96-9.71)	0.41 (0.31-0.55)	-	0.8948	0.8257

SROC = summary receiver operating characteristic; CI = confidence interval; PDAC = pancreatic ductal adenocarcinoma; PIM = pancreatic inflammatory masses; MIC-1 = macrophage inhibitory cytokine-1; DWI = diffusion-weighted magnetic resonance imaging; ERCP = endoscopic retrograde cholangiopancreatography; EUS = endoscopic ultrasound; EUS-FNA = endoscopic ultrasound-guided fine-needle aspiration; PET = positron emission tomography; CT = computed tomography; cytology classification 1: suspicious/atypical results on EUS-FNA, negative; cytology classification 2: suspicious/atypical results on EUS-FNA, positive.

Q: It is an intersection point of SROC curve and a international method of test efficacy, which corresponds the highest value of sensitivity and specificity.

Hence, it is necessary to investigate a new diagnostic marker with high sensitivity, specificity and stability for early diagnosis of pancreatic cancer. Many studies have demonstrated that PET (63), PET/CT (55), and microRNA (64) may be useful in the early diagnosis of pancreatic cancer. The diagnostic efficacy of 2 or more methods combined may be better than that of 1 method alone. For example, endoscopic ultrasound-guided fine-needle aspiration cytology yields better results (65, 66) than some of the biomarkers including telomerase activity. Nevertheless, telomerase activity can be a candidate marker, and its diagnostic effect can be improved when combined with other methods.

In addition to the small sample size, there were several other limitations in the present meta-analysis. First, English language-only publications were included, which potentially caused publication bias. Second, some low-quality studies included may have influenced the reliability of the results. Third, the cutoff values of telomerase among the included studies differed or were not reported. In addition, different methods were used in the analysis of different specimens, which may lead to significant variations in the results obtained. Publication bias is an inherent limitation of all meta-analyses, irrespective of the outcomes of the Deeks test.

In conclusion, telomerase activity is likely to play an important role in the diagnosis of pancreatic adenocarcinoma, and the diagnostic effect would be improved when it is combined with other diagnostic tools. Future prospective, large-scale multicenter studies are needed to generate higher quality evidence necessary to verify these conclusions. In addition, PET/CT and microRNA may be useful for the early diagnosis of pancreatic cancer.

Footnotes

Financial support: The work was supported by International S&T Cooperation Program of China (Grant No. 2014D FA31420).

Conflict of interest: The authors declare no competing interest.

References

Gudjonsson

Cancer of the pancreas. 50 years of surgery. Cancer 1987; 60 (9): 2284–2303

Carpelan-Holmström

Nordling

Pukkala

. Does anyone survive pancreatic ductal adenocarcinoma? A nationwide study re-evaluating the data of the Finnish Cancer Registry. Gut 2005; 54 (3): 385–387

Jemal

Siegel

Ward

Hao

Thun

Cancer statistics, 2009. CA Cancer J Clin 2009; 59 (4): 225–249

Siegel

Zou

Jemal

Cancer statistics, 2014. CA Cancer J Clin 2014; 64 (1): 9–29

Katz

Savides

Moossa

Bouvet

An evidence-based approach to the diagnosis and staging of pancreatic cancer. Pancreatology 2005; 5 (6): 576–590

Blackburn

Telomeres and telomerase: their mechanisms of action and the effects of altering their functions. FEBS Lett 2005; 579 (4): 859–862

Abe

Tanaka

Notsu

. Localization of an hTERT repressor region on human chromosome 3p21.3 using chromosome engineering. Genome Integr 2010; 1 (1): 6

Myung

Kim

. Telomerase activity in pure pancreatic juice for the diagnosis of pancreatic cancer may be complementary to K-ras mutation. Gastrointest Endosc 2000; 51 (6): 708–713

Hiyama

Kodama

Shinbara

. Telomerase activity is detected in pancreatic cancer but not in benign tumors. Cancer Res 1997; 57 (2): 326–331

10.

Iwao

Hiyama

Yokoyama

. Telomerase activity for the preoperative diagnosis of pancreatic cancer. J Natl Cancer Inst 1997; 89 (21): 1621–1623

11.

Balcom

IV Keck

Warshaw

Antoniu

Graeme-Cook

Fernández-del Castillo

Telomerase activity in periampullary tumors correlates with aggressive malignancy. Ann Surg 2001; 234 (3): 344–351

12.

Grochola

Greither

Taubert

. Prognostic relevance of hTERT mRNA expression in ductal adenocarcinoma of the pancreas. Neoplasia 2008; 10 (9): 973–976

13.

Guittat

Alberti

Gomez

. Targeting human telomerase for cancer therapeutics. Cytotechnology 2004; 45 1-2 75–90

14.

Pearson

Chiao

Zhang

. The detection of telomerase activity in patients with adenocarcinoma of the pancreas by fine needle aspiration. Int J Oncol 2000; 17 (2): 381–385

15.

Mizumoto

Tanaka

Detection of telomerase activity in patients with pancreatic cancer. Methods Mol Med 2005; 103 199–205

16.

Glukhov

Grigorieva

Gordeev

Vinarov

Potoldykova NV: Development of noninvasive bladder cancer diagnosis on basis of telomerase and it's subunits hTERT and hTR detection. Biomed Khim 2015; 61 (1): 150–160

17.

Glybochko

Zezerov

Glukhov

. Telomerase as a tumor marker in diagnosis of prostatic intraepithelial neoplasia and prostate cancer. Prostate 2014; 74 (10): 1043–1051

18.

Wang

Park

Kim

. Use of hTERT and HPV E6/E7 mRNA RT-qPCR TaqMan assays in combination for diagnosing high-grade cervical lesions and malignant tumors. Am J Clin Pathol 2015; 143 (3): 344–351

19.

Whiting

Rutjes

Westwood

et al; QUADAS-2 Group. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med 2011; 155 (8): 529–536

20.

Schlesselman

Stolley

Case control studies. Design, conduct, analysis. New York Oxford University Press 1982: 174-177.

21.

Bossuyt

Davenport

Deeks

Hyde

Leeflang

Scholten

Chapter 11: Interpreting results and drawing conclusions. In: Deeks

, Bossuyt

, Gatsonis

, eds. Cochrane handbook for systematic reviews of diagnostic test accuracy version 0.9. The Cochrane Collaboration, 2013 Available from: http://srdta.cochrane.org/.

22.

Hashimoto

Murakami

Uemura

. Telomere shortening and telomerase expression during multistage carcinogenesis of intraductal papillary mucinous neoplasms of the pancreas. J Gastrointest Surg 2008; 12 (1): 17–29

23.

Nakashima

Murakami

Uemura

. Usefulness of human telomerase reverse transcriptase in pancreatic juice as a biomarker of pancreatic malignancy. Pancreas 2009; 38 (5): 527–533

24.

Hashimoto

Murakami

Uemura

. Detection of human telomerase reverse transcriptase (hTERT) expression in tissue and pancreatic juice from pancreatic cancer. Surgery 2008; 143 (1): 113–125

25.

Seki

Suda

Aoyagi

. Diagnosis of pancreatic adenocarcinoma by detection of human telomerase reverse transcriptase messenger RNA in pancreatic juice with sample qualification. Clin Cancer Res 2001; 7 (7): 1976–1981

26.

Büchler

Conejo-Garcia

Lehmann

. Real-time quantitative PCR of telomerase mRNA is useful for the differentiation of benign and malignant pancreatic disorders. Pancreas 2001; 22 (4): 331–340

27.

Zhou

Zhang

Chen

. Relationship between telomerase activity and its subunit expression and inhibitory effect of antisense hTR on pancreatic carcinoma. World J Gastroenterol 2003; 9 (8): 1808–1814

28.

Uemura

Hiyama

Murakami

. Comparative analysis of K-ras point mutation, telomerase activity, and p53 overexpression in pancreatic tumours. Oncol Rep 2003; 10 (2): 277–283

29.

Zhou

Huang

Liu

Zhang

Detection of K-ras point mutation and telomerase activity during endoscopic retrograde cholangiopancreatography in diagnosis of pancreatic cancer. World J Gastroenterol 2004; 10 (9): 1337–1340

30.

Ohuchida

Mizumoto

Ishikawa

. A highly sensitive and quantitative telomerase activity assay with pancreatic juice is useful for diagnosis of pancreatic carcinoma without problems due to polymerase chain reaction inhibitors: analysis of 100 samples of pancreatic juice from consecutive patients. Cancer 2004; 101 (10): 2309–2317

31.

Mishra

Zhao

Sweeney

. Determination of qualitative telomerase activity as an adjunct to the diagnosis of pancreatic adenocarcinoma by EUS-guided fine-needle aspiration. Gastrointest Endosc 2006; 63 (4): 648–654

32.

Ohuchida

Mizumoto

Yamada

. Quantitative analysis of human telomerase reverse transcriptase in pancreatic cancer. Clin Cancer Res 2006; 12 (7 Pt 1): 2066–2069

33.

Kawahara

Odo

Kinoshita

Shirouzu

Aoyagi

Analysis of hTERT mRNA expression in biliary tract and pancreatic cancer. J Hepatobiliary Pancreat Surg 2007; 14 (2): 189–193

34.

Zhou

. Marker expression in circulating cancer cells of pancreatic cancer patients. J Surg Res 2011; 171 (2): 631–636

35.

Tsutsumi

Tsujiuchi

Ishikawa

. Increased telomerase activities in human pancreatic duct adenocarcinomas. Jpn J Cancer Res 1997; 88 (10): 971–976

36.

Lam

Fan

Luk

Telomerase activity in pancreatic endocrine tumours: a potential marker for malignancy. Mol Pathol 2000; 53 (3): 133–136

37.

Mizumoto

Suehara

Muta

. Semi-quantitative analysis of telomerase in pancreatic ductal adenocarcinoma. J Gastroenterol 1996; 31 (6): 894–897

38.

Suehara

Mizumoto

Muta

. Telomerase elevation in pancreatic ductal carcinoma compared to nonmalignant pathological states. Clin Cancer Res 1997; 3 (6): 993–998

39.

Yeh

Cheng

Chen

. Telomerase activity is a useful marker to distinguish malignant pancreatic cystic tumors from benign neoplasms and pseudocysts. J Surg Res 1999; 87 (2): 171–177

40.

Niiyama

Mizumoto

Kusumoto

. Activation of telomerase and its diagnostic application in biopsy specimens from biliary tract neoplasms. Cancer 1999; 85 (10): 2138–2143

41.

Morales

Burdick

Saboorian

Wright

Shay

In situ hybridization for telomerase RNA in routine cytologic brushings for the diagnosis of pancreaticobiliary malignancies. Gastrointest Endosc 1998; 48 (4): 402–405

42.

Inoue

Tsuchida

Kawasaki

Fujimoto

Yamasaki

Kajiyama

Preoperative diagnosis of intraductal papillary-mucinous tumors of the pancreas with attention to telomerase activity. Cancer 2001; 91 (1): 35–41

43.

Serikawa

Sasaki

Fujimoto

Kuwahara

Chayama

Management of intraductal papillary-mucinous neoplasm of the pancreas: treatment strategy based on morphologic classification. J Clin Gastroenterol 2006; 40 (9): 856–862

44.

Ohuchida

Mizumoto

Ogura

. Telomerase reverse transcriptase messenger RNA levels in quantitative assessment of telomerase activity and human pancreatic juice samples for the diagnosis of pancreatic cancer. Clin Cancer Res 2005; 11 (6): 2285–2292

45.

Zhou

Huang

Zhang

Chen

Diagnosis of pancreatic cancer by cytology and telomerase activity in exfoliated cells obtained by pancreatic duct brushing during endoscopy. Hepatobiliary Pancreat Dis Int 2007; 6 (3): 308–311

46.

Uehara

Nakaizumi

Tatsuta

. Diagnosis of pancreatic cancer by detecting telomerase activity in pancreatic juice: comparison with K-ras mutations. Am J Gastroenterol 1999; 94 (9): 2513–2518

47.

Uehara

Nakaizumi

Iishi

. In situ telomerase activity in pancreatic juice may discriminate pancreatic cancer from other pancreatic diseases. Pancreas 2008; 36 (3): 236–240

48.

Zisuh

Han

Zhan

Expression of telomerase & its significance in the diagnosis of pancreatic cancer. Indian J Med Res 2012; 135 (1): 26–30

49.

Wang

Chen

Tang

Quantitative assessment of the diagnostic role of human telomerase activity from pancreatic juice in pancreatic cancer. Tumour Biol 2014; 35 (8): 7897–7904

50.

Chen

Guo

Comparison of ERCP, EUS, and ERCP combined with EUS in diagnosing pancreatic neoplasms: a systematic review and meta-analysis. Tumor Biol 2014; 35 (9): 8867–8874

51.

Shi

Lin

Zeng

Endoscopic ultrasound elastography for differentiating between pancreatic adenocarcinoma and inflammatory masses: a meta-analysis. World J Gastroenterol 2013; 19 (37): 6284–6291

52.

Gong

Zhu

Endoscopic ultrasound elastography for differential diagnosis of pancreatic masses: a meta-analysis. Dig Dis Sci 2013; 58 (4): 1125–1131

53.

Orlando

Kulasingam

Matchar

Meta-analysis: the detection of pancreatic malignancy with positron emission tomography. Aliment Pharmacol Ther 2004; 20 (10): 1063–1070

54.

Yang

Qian

. Diagnostic accuracy of EUS and CT of vascular invasion in pancreatic cancer: a systematic review. J Cancer Res Clin Oncol 2014; 140 (12): 2077–2086

55.

Hua

Liu

Chen

Diagnostic value of diffusion-weighted magnetic resonance imaging compared with fluorodeoxyglucose positron emission tomography/computed tomography for pancreatic malignancy: a meta-analysis using a hierarchical regression model. J Gastroenterol Hepatol 2012; 27 (6): 1027–1035

56.

Dabizzi

Assef

Raimondo

Diagnostic management of pancreatic cancer. Cancers (Basel) 2011; 3 (1): 494–509

57.

Huang

Liu

Diagnostic value of serum carbohydrate antigen 19-9 in pancreatic cancer: a meta-analysis. Tumour Biol 2014; 35 (8): 7459–7465

58.

Gui

Yan

Liu

CA19-9 and CA242 as tumor markers for the diagnosis of pancreatic cancer: a meta-analysis. Clin Exp Med 2014; 14 (2): 225–233

59.

Chen

Liu

Zhao

Wang

Gao

Chen

Diagnostic performance of serum macrophage inhibitory cytokine-1 in pancreatic cancer: a meta-analysis and meta-regression analysis. DNA Cell Biol 2014; 33 (6): 370–377

60.

Ishimaru

Itoh

Hanada

Tsuchida

Iwao

Kajiyama

Immunocytochemical detection of p53 protein from pancreatic duct brushings in patients with pancreatic carcinoma. Cancer 1996; 77 (11): 2233–2239

61.

Löhr

Müller

Mora

. p53 and K-ras mutations in pancreatic juice samples from patients with chronic pancreatitis. Gastrointest Endosc 2001; 53 (7): 734–743

62.

Liu

Chen

Zhao

Zhang

Diagnostic accuracy of K-ras mutation for pancreatic carcinoma: a meta-analysis. Hepatobiliary Pancreat Dis Int 2013; 12 (5): 458–464

63.

Wang

Chen

Liu

Qin

Huang

FDG-PET in diagnosis, staging and prognosis of pancreatic carcinoma: a meta-analysis. World J Gastroenterol 2013; 19 (29): 4808–4817

64.

Wan

Shen

Yang

Wang

Chen

Wen

Diagnostic value of microRNA for pancreatic cancer: a meta-analysis. Arch Med Sci 2012; 8 (5): 749–755

65.

Puli

Bechtold

Buxbaum

Eloubeidi

How good is endoscopic ultrasound-guided fine-needle aspiration in diagnosing the correct etiology for a solid pancreatic mass?

Pancreas 2013; 42 (1): 20–26

66.

Hewitt

McPhail

Possamai

Dhar

Vlavianos

Monahan

EUS-guided FNA for diagnosis of solid pancreatic neoplasms: a meta-analysis. Gastrointest Endosc 2012; 75 (2): 319–331