Systematic Reviews
Copyright ©The Author(s) 2015. Published by Baishideng Publishing Group Inc. All rights reserved.
World J Gastroenterol. Apr 14, 2015; 21(14): 4323-4333
Published online Apr 14, 2015. doi: 10.3748/wjg.v21.i14.4323
Carbohydrate antigen 19-9 for differential diagnosis of pancreatic carcinoma and chronic pancreatitis
Si-Biao Su, Shan-Yu Qin, Wen Chen, Wei Luo, Hai-Xing Jiang
Si-Biao Su, Shan-Yu Qin, Wei Luo, Hai-Xing Jiang, Department of Gastroenterology, The First Affiliated Hospital of Guangxi Medical University, Nanning 530021, Guangxi Zhuang Autonomous Region, China
Wen Chen, Department of Educational Administration, The First Affiliated Hospital of Guangxi Medical University, Nanning 530021, Guangxi Zhuang Autonomous Region, China
Author contributions: Jiang HX designed the study, searched the databases, extracted the data, analyzed the results, and wrote the manuscript; Su SB helped design the study, search the databases, and write and revise the manuscript; Qin SY formulated the research question, and helped with database searches and analysis; Chen W and Luo W helped design the data abstraction form and served as second reviewers in extracting the data; all authors have read and approved the final manuscript.
Open-Access: This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Correspondence to: Dr. Hai-Xing Jiang, Department of Gastroenterology, The First Affiliated Hospital of Guangxi Medical University, No. 22 Shuangyong Road, Nanning 530021, Guangxi Zhuang Autonomous Region, China. Jianghaixing2001@163.com
Telephone: +86-771-5312726 Fax: +86-771-5356758
Received: September 20, 2014
Peer-review started: September 21, 2014
First decision: October 14, 2014
Revised: October 30, 2014
Accepted: January 8, 2015
Article in press: January 8, 2015
Published online: April 14, 2015

Abstract

AIM: To evaluate the utility of carbohydrate antigen 19-9 (CA19-9) for differential diagnosis of pancreatic carcinoma and chronic pancreatitis.

METHODS: We searched the literature for studies reporting the sensitivity, specificity, and other accuracy measures of serum CA19-9 levels for differentiating pancreatic carcinoma and chronic pancreatitis. Pooled analysis was performed using random-effects models, and receiver operating characteristic (ROC) curves were generated. Study quality was assessed using Standards for Reporting Diagnostic Accuracy and Quality Assessment for Studies of Diagnostic Accuracy tools.

RESULTS: A total of 34 studies involving 3125 patients with pancreatic carcinoma and 2061 patients with chronic pancreatitis were included. Pooled analysis of the ability of CA19-9 level to differentiate pancreatic carcinoma and chronic pancreatitis showed the following effect estimates: sensitivity, 0.81 (95%CI: 0.80-0.83); specificity, 0.81 (95%CI: 0.79-0.82); positive likelihood ratio, 4.08 (95%CI: 3.39-4.91); negative likelihood ratio, 0.24 (95%CI: 0.21-0.28); and diagnostic odds ratio, 19.31 (95%CI: 14.40-25.90). The area under the ROC curve was 0.88. No significant publication bias was detected.

CONCLUSION: Elevated CA19-9 by itself is insufficient for differentiating pancreatic carcinoma and chronic pancreatitis, however, it increases suspicion of pancreatic carcinoma and may complement other clinical findings to improve diagnostic accuracy.

Key Words: Pancreatic carcinoma, Chronic pancreatitis, Carbohydrate antigen, Diagnosis, Meta-analysis

Core tip: Pancreatic carcinoma and chronic pancreatitis show similar clinical manifestations. Differential diagnosis of pancreatic carcinoma and chronic pancreatitis remains a challenge, particularly in patients with pancreatic masses that may be benign (inflammatory) or malignant. Carbohydrate antigen 19-9 (CA19-9) shows promise for differentiating the diseases. We evaluated the usefulness of CA19-9 in this systematic review.



INTRODUCTION

Pancreatic carcinoma is the fourth leading cause of cancer deaths in the United States[1]. Currently, the most effective treatment is surgical resection[2,3]. However, approximately 80% of tumors are unresectable at diagnosis, and these patients show a 5-year survival rate below 5%[1]. The clinical manifestations of pancreatic carcinoma resemble those of chronic pancreatitis. In fact, chronic pancreatitis is strongly associated with pancreatic malignancy and may help to cause it. For example, individuals in parts of Southern India with idiopathic chronic pancreatitis unassociated with alcohol abuse show a high incidence of pancreatic carcinoma[4].

Differential diagnosis of pancreatic carcinoma and chronic pancreatitis remains a challenge, particularly in patients with pancreatic masses that may be benign (inflammatory) or malignant. This differentiation is important in order to avoid unnecessary resection in patients with inflammatory masses: 5%-10% of patients subjected to pancreatic resection are ultimately diagnosed with pancreatitis rather than pancreatic carcinoma[5]. Differentiation is also important in order to identify correctly pancreatic masses as cancerous and avoid leaving behind malignant masses. Pancreatic carcinoma is incurable in many patients who also have chronic pancreatitis, because the cancer is multicentric or advanced.

Carbohydrate antigen 19-9 (CA19-9) is the most popular serum-based marker for diagnosis of pancreatic cancer, and it is useful for detecting disease recurrence after surgery[6,7]. However, this biomarker has limited diagnostic power. CA19-9 level can be normal in patients with localized disease, therefore, it is less effective for screening for early pancreatic cancer. High CA19-9 levels can also occur in benign diseases, including chronic pancreatitis and nonmalignant jaundice[6-8].

Diagnosis of pancreatic cancer at an early, resectable stage is especially difficult when the patient also presents with chronic pancreatitis[9,10], therefore, we wondered whether CA19-9 might be useful for differentiating the two diseases. We performed a systematic review and meta-analysis of the utility of CA19-9 as a serum tumor marker and its sensitivity and specificity for distinguishing pancreatic carcinoma and chronic pancreatitis.

MATERIALS AND METHODS
Search strategies

In June 2013 we searched MEDLINE (1980 to May 2013), EMBASE (1980 to May 2013), Web of Science (1990 to May 2013) and Cochrane databases. Although no language restrictions were imposed initially, for the full-text review and final analysis only English language articles were included. Additional articles were searched using the “Related articles” function in PubMed and by manually searching reference lists of identified articles and review articles. The following search terms were used: “pancreatic carcinoma” or “pancreatic cancer” and “chronic pancreatitis” and “carbohydrate antigen 19-9” and “diagnosis” or “sensitivity” or “specificity”. We contacted experts in the field to ask about studies that we may have missed in the databases. Conference abstracts and letters to the editor were excluded because of the limited data they contained.

Study inclusion criteria

A study was included when it provided both the sensitivity (true-positive rate) and specificity (true-negative rate) of using serum CA19-9 levels to diagnose pancreatic carcinoma or chronic pancreatitis in patients of any age. Studies were also included if they reported CA19-9 values in a scatter plot format that allowed patient-level data to be extracted. Studies had to involve at least 10 patients with pancreatic carcinoma or chronic pancreatitis in order to reduce selection bias due to a small number of participants. Patients had to be diagnosed with pancreatic carcinoma based on cytology and/or histology of pancreatic tissue, or diagnosed with chronic pancreatitis based on clinical information alone or in combination with histopathological resection, radiology (endoscopic retrograde cholangiopancreatography and computed tomography) and/or endoscopic ultrasonography. Two reviewers (Su SB and Jiang HX) independently determined study eligibility, and disagreements were resolved by consensus.

Data extraction and quality assessment

These same two reviewers independently confirmed the eligibility of the final set of studies and extracted the following data: first author, publication year, participant characteristics, assay methods, sensitivity and specificity data, cut-off values, and methodological quality. Serum CA19-9 values provided in scatter plots were extracted by placing scalar grids over the plots. A receiver operating characteristic (ROC) curve was calculated for each study.

To enable us to assess the methodological quality of the included studies, we extracted data on the following study design characteristics: (1) cross-sectional or case-control design; (2) consecutive or random sampling of patients; (3) blinded (single or double) or non-blinded interpretation of experimental and reference measurements; and (4) prospective or retrospective data collection. Su SB and Jiang HX independently assessed the methodological quality of studies using the Standards for Reporting Diagnostic Accuracy (STARD) guidelines[11] (maximum score of 25) and the Quality Assessment for Studies of Diagnostic Accuracy (QUADAS) guidelines[12] (maximum score of 14). Average inter-rater agreement on the methodological quality checklists was 0.96. If primary studies did not report information needed to assess methodological quality, we contacted the authors in an effort to obtain the data. If the authors did not respond, we changed the response for the relevant items from “not reported” to “no” on the assessment instruments.

Statistical analysis

Standard methods recommended for meta-analyses of diagnostic test evaluations were used[13]. Analyses were performed using Meta-DiSc for Windows (XI Cochrane Colloquium; Barcelona, Spain) and Stata 12.0 (Stata Corporation, College Station, TX, United States). The following measures of test accuracy were analyzed for each study: sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR) and diagnostic odds ratio (DOR). A summary ROC (SROC) curve[14] was generated for each study based on a single test threshold for sensitivity and specificity[13,15]. A random-effects model was adopted to calculate the average sensitivity, specificity, and other measures across studies[16,17].

To assess the effects of STARD and QUADAS scores on the diagnostic power of CA19-9, we included them as covariates in univariate, inverse variance-weighted meta-regression. We also analyzed the effects of other covariates on DOR, such as cross-sectional design, consecutive or random sampling of patients, single- or double-blinded interpretation of experimental and reference measurements, and prospective or retrospective data collection. The relative DOR (RDOR) was calculated to analyze the change in diagnostic precision in each study per unit increase in the covariate[18,19]. P < 0.05 was considered to show statistical significance.

The heterogeneity, or variability, across studies was assessed for statistical significance using the χ2 and Fisher exact tests. Publication bias can pose problems for meta-analyses of diagnostic studies, therefore, we tested for the potential presence of this bias using funnel plots and the Egger test[20].

RESULTS
Selection and summary of studies

We identified 345 citations via electronic searches, and 106 were retrieved for detailed analysis (Figure 1). Of these, 47 studies were excluded for failing to satisfy the inclusion criteria, and another 17 were excluded because they failed to provide sufficient information for meta-analysis. Five studies were duplicate publications. Two articles were meta-analyses, and one was excluded for involving fewer than 10 participants. In the end, 34 publications were included in the analysis[21-54], involving 3125 patients with pancreatic carcinoma and 2061 patients with chronic pancreatitis. The average sample size of the studies was 153 patients (range: 24-941). Table 1 summarizes the clinical characteristics of participants in each study; the numbers of true-positive, false-positive, false- negative and true-negative results; and STARD and QUADAS scores.

Table 1 Summary of carbohydrate antigen 19-9 assay methods, results, and overall methodological quality of included studies.
Ref.Number of patientAssay methodCut-off for elevated CA19-9 (U/mL)Assay results
Quality score
TPFPFNTNSTARDQUADAS
Wang et al[21], 198658RIA372004341510
Safi et al[22], 1987191RIA3780287761610
Sakamoto et al[23], 198757RIA372614261812
Friess et al[24], 1993154ELISA3753146811711
Röthlin et al[25], 199397RIA3754814211712
Haglund et al[26], 1994199RIA37148331171611
Kuno et al[27], 1994117RIA3741106601913
Pasquali et al[28], 1994103RIA374721143129
Satake et al[29], 1994941RIA37454561182441913
Hámori et al[30], 199794RIA374841428117
Safi et al[31], 1997647RIA3729648512521812
Hayakawa et al[32], 199976RIA3721146351611
Kim et al[33], 1999160ELISA3769921611913
Manes et al[34], 199958RIA373034211711
Slesak et al[35], 2000122LIA37321414601812
Maire et al[36], 200278ELISA374344271711
Akashi et al[37], 200346RIA37157519129
Mu et al[38], 200324RIA37435121510
Cwik et al[39], 2004150RIA3782516471611
Jiang et al[40], 2004148ELISA3782714451712
Ventrucci et al[41], 200481EIA6045215191812
Teich et al[42], 200559ELISA22273313129
Chang et al[43], 2007111ELISA3763119281812
ELISA10057715321812
Kuhlmann et al[44], 200762EIA3717411301611
Liao et al[45], 2007150ELISA37841528231510
Bedi et al[46], 200984ELISA37231511351712
ELISA10014720431712
Firpo et al[47], 2009107ELISA3758217301812
Liao et al[48], 2009102RIA37472211221610
Morris-Stiff et al[49], 2009188ELISA3770313841913
Talar-Wojnarowska et al[50], 2010157ELISA37711814541712
Zapico-Muñiz et al[51], 2010102LIA10035712481611
Chung et al[52], 201178NR304021521129
Gold et al[53], 2013284EIA371801654341811
Kaur et al[54], 2013114RIA3776915141711
Figure 1
Figure 1 Flowchart of study selection.
Methodological quality of the included studies

Of the 34 studies in the meta-analysis, 30 had STARD scores ≥ 13, and 29 had QUADAS scores ≥ 10. All studies collected data from consecutive patients using a prospective design. No study reported interpretation of CA19-9 measurements in which analysts were blinded to the corresponding reference measurements (Table 2).

Table 2 Additional characteristics of patients and methodology in the included studies.
Ref.Country/areaPC/CP, nPC referenceCross-sectional designConsecutive or Random samplingBlinded designProspective design
Wang et al[21], 1986Taiwan24/34His or CytNoYesNoYes
Safi et al[22], 1987Germany87/104HisYesYesNoYes
Sakamoto et al[23], 1987Japan30/27HisNoYesNoYes
Friess et al[24], 1993Germany59/95HisYesYesNoYes
Röthlin et al[25], 1993Switzerland68/29HisNoYesNoYes
Haglund et al[26], 1994Finland179/20HisNoYesNoYes
Kuno et al[27], 1994Japan47/70HisYesYesNoYes
Pasquali et al[28], 1994Italy58/45HisNoNoNRYes
Satake et al[29], 1994Japan641/300HisYesYesNoYes
Hámori et al[30], 1997Hungary62/32HisNoYesNoYes
Safi et al[31], 1997Germany347/300His or BioYesYesNoYes
Hayakawa et al[32], 1999Japan27/49His (Bio, Aut)NoYesNoYes
Kim et al[33], 1999Korea90/70HisYesYesNoYes
Manes et al[34], 1999Italy34/24His or CytYesYesNoYes
Slesak et al[35], 2000Poland48/74HisNoYesNoYes
Maire et al[36], 2002France47/31His or CytNoYesNoYes
Akashi et al[37], 2003Japan20/26His or AutNoYesNoYes
Mu et al[38], 2003China9/15His or CytNoYesNoYes
Cwik et al[39], 2004Lublin98/52HisNRYesNRYes
Jiang et al[40], 2004China96/52HisYesYesNoYes
Ventrucci et al[41], 2004Italy60/21HisYesYesNoYes
Teich et al[42], 2005Germany30/16HisNoNoNoYes
Chang et al[43], 2007Taiwan72/39HisYesYesNoYes
New York, United States28/34HisNoYesNRYes
Kuhlmann et al[44], 2007China112/38HisNoYesNoYes
Liao et al[45], 2007India34/50His or BioYesYesNoYes
Bedi et al[46], 2009United States75/32His or CytYesYesNoYes
Taiwan58/44HisYesYesNoYes
Firpo et al[47], 2009United Kingdom73/115HisYesYesNoYes
Liao et al[48], 2009Poland85/72HisYesYesNoYes
Morris-Stiff et al[49], 2009Spain47/55HisYesYesNoYes
Talar-Wojnarowska et al[50], 2010Korea55/23HisYesYesNoYes
Zapico-Muñiz et al[51], 2010New York, United States234/50His or CytYesYesNoYes
Chung et al[52], 2011Germany91/23HisNoYesNoYes
Diagnostic accuracy

As shown in Figure 2, a Forest plot of serum CA19-9 levels in all 34 included studies showed that the sensitivity of this biomarker to differentiate between pancreatic carcinoma and chronic pancreatitis ranged from 0.44 to 0.96 [mean: 0.81, 95%CI: 0.80-0.83; χ2 = 77.23, P < 0.001), while the specificity ranged from 0.50 to 1.0 (mean: 0.81, 95%CI: 0.79-0.82; χ2 = 111.98, P < 0.001). The PLR was 4.08 (95%CI: 3.39-4.91; χ2 = 113.62, P < 0.001), NLR was 0.24 (95%CI: 0.21-0.28; χ2 = 86.13, P < 0.001) and DOR was 19.31 (95%CI: 14.4-25.9; χ2 = 94.02, P < 0.001). These χ2 values and associated P-values indicate significant heterogeneity among studies.

Figure 2
Figure 2 Forest plot showing sensitivity and specificity of carbohydrate antigen 19-9 in the diagnosis of pancreatic carcinoma. The point estimates of sensitivity and specificity from each study are shown as solid circles. Horizontal error bars indicate 95%CI. Numbers between the plots refer to references. Pooled estimates for the serum carbohydrate antigen 19-9 assay were 0.81 for sensitivity (95%CI: 0.80-0.83) and 0.81 for specificity (95%CI: 0.79-0.82).

These measures of differential diagnostic power varied with different CA19-9 assays and cut-off values used to define CA19-9 levels as elevated or normal (Table 3). Data from the 11 studies that relied on the enzyme-linked immunosorbent assay (ELISA) method, involving 1396 patients, gave a sensitivity of 0.83 and specificity of 0.79. Data from the 17 studies using the radioimmunoassay method, involving 3074 patients, gave a sensitivity of 0.82 and specificity of 0.81. Data from the three studies that relied on an enzyme immunosorbent assay (EIA) gave a sensitivity of 0.75 and specificity of 0.79. Data from the 30 studies (4879 patients) using a cut-off value of 37 U/mL gave a sensitivity of 0.82 and specificity of 0.80. Data from the three studies using a cut-off value of 100 U/mL gave corresponding values of 0.69 and 0.85. These variations in sensitivity and specificity with CA19-9 assay and cut-off values did not achieve statistical significance (P > 0.05, Table 4), suggesting that high cut-off values such as 100 U/mL may better increase the specificity for differential diagnosis of pancreatic carcinoma.

Table 3 Bivariate estimates of diagnostic precision based on different carbohydrate antigen 19-9 assay methods and cut-off values.
Assay method or cut-off valueNumber of studiesNumber of participantsSensitivity (95%CI)Specificity (95%CI)PLR (95%CI)NLR (95%CI)DOR (95%CI)
ELISA1113960.83 (0.80-0.86)0.79 (0.75-0.82)3.97 (2.96-5.33)0.20 (0.15-0.28)22.64 (12.44-41.22)
RIA1730740.82 (0.80-0.84)0.81 (0.79-0.83)4.16 (3.09-5.60)0.23 (0.19-0.27)20.14 (13.27-30.55)
EIA34270.75 (0.70-0.80)0.79 (0.70-0.86)3.84 (1.82-8.10)0.34 (0.27-0.43)10.29 (4.96-21.34)
LIA2224NDNDNDNDND
Cut-off of 37 U/mL3048790.82 (0.80-0.83)0.80 (0.78-0.82)3.94 (3.24-4.78)0.24 (0.21-0.28)18.79 (13.67-25.82)
Cut-off of 100 U/mL32970.69 (0.61-0.76)0.85 (0.79-0.91)4.35 (2.86-6.61)0.38 (0.18-0.77)11.53 (4.47-29.77)
All studies3451150.81 (0.80-0.83)0.81 (0.79-0.82)4.08 (3.39-4.91)0.24 (0.21-0.28)19.31 (14.40-25.90)
Table 4 Weighted meta-regression of the effects of study design, methodological quality and assay parameters on diagnostic accuracy of carbohydrate antigen 19-9.
CovariateNumber of studiesCoefficientRDOR (95%CI)P value
Study design and quality
STARD ≥ 13300.5641.76 (0.14-22.68)0.652
QUADAS ≥ 1029-0.6660.51 (0.06-4.11)0.514
Consecutive or random design320.9242.52 (0.26-24.68)0.411
Cross-sectional design18-0.5120.60 (0.28-1.28)0.178
Blinded design0NDNDND
Prospective design34NDNDND
Assay method or cut-off value
RIA17-0.6190.54 (0.12-2.51)0.413
ELISA11-0.7370.48 (0.10-2.26)0.336
EIA30.4251.53 (0.29-8.14)0.604
Cut-off of 37 U/mL300.5531.74 (0.36-8.36)0.474
Cut-off of 100 U/mL30.8902.43 (0.72-8.26)0.146

Instead of assessing diagnostic power using the traditional ROC plot, we calculated an SROC plot to reveal the effect of varying thresholds on sensitivity and specificity within each study. In this plot, different studies appear as different data points, allowing SROC curves to provide a global summary of test performance and illustrate the trade-off between sensitivity and specificity. Figure 3 shows an SROC curve for rates of true- and false-positive results obtained with the CA19-9 assay in individual studies. From this plot we determined the Q value, which was defined as the point of intersection of the SROC curve with a diagonal line extending from the left upper corner to the right lower corner of the plot. The Q value indicates the highest identical value of sensitivity and specificity, thereby serving as an overall measure of the discriminatory power of a test. Our SROC curve was desirably positioned near the upper left corner, and the maximum joint sensitivity and specificity was 0.81. The area under the curve (AUC) was 0.88 (Figure 3A), indicating high overall accuracy. SROC plots differed based on the CA19-9 assay method and cut-off values, but all plots were positioned near the upper left corner with AUCs near 0.88 (Figure 3B-F), again indicating high overall accuracy.

Figure 3
Figure 3 Summary receiver operating characteristic curves for carbohydrate antigen 19-9 assays for differential diagnosis of pancreatic carcinoma and chronic pancreatitis. Solid circles represent each study included in the meta-analysis, with circle size proportional to the number of participants in the study. SROC curves summarize the overall diagnostic accuracy for all included studies (A), studies using a cut-off of 37 U/mL carbohydrate antigen 19-9 (CA19-9) (B), studies using a cut-off of 100 U/mL CA19-9 (C), studies based on the radioimmunoassay method to assay CA19-9 (D), studies based on the ELISA method (E), and studies based on the enzyme immunoassay method (F). ELISA: Enzyme-linked immunosorbent assay; SROC: Summary receiver operating characteristic.
Multiple regression analysis and publication bias

Quality scores based on the STARD[11] and QUADAS[12] guidelines were generated for every study on the basis of the title and introduction, methods, results and discussion (Table 1). These scores were used in meta-regression to assess the effect of study quality on the RDOR of CA19-9 in the differential diagnosis of pancreatic carcinoma and chronic pancreatitis. Studies of higher quality (STARD score ≥ 13; QUADAS score ≥ 10) produced RDOR values similar to those of lower-quality studies. In addition, RDOR values did not differ significantly as a function of blinding, cross-sectional or case-control design, consecutive or random sampling, prospective or retrospective design, CA19-9 assay method, or cut-off values (P > 0.05). These results suggest that study design did not significantly affect diagnostic accuracy and that the risk of detection bias was low.

The Egger test showed no significant evidence of publication bias in reports about CA19-9 assays for differential diagnosis of pancreatic carcinoma (P = 0.944).

DISCUSSION

Timely and accurate diagnosis of pancreatic carcinoma is critical for patient prognosis, but it remains a challenge because the signs and symptoms of pancreatic cancer overlap considerably with those of chronic pancreatitis. Compounding this challenge is the fact that acute or chronic pancreatitis increases the risk of pancreatic carcinoma, as well as the fact that this cancer can induce secondary inflammatory processes. In this systematic review, we find evidence that although CA19-9 levels on their own are inadequate for differentiating pancreatic carcinoma and chronic pancreatitis, elevated CA19-9 may complement other clinical tests to help confirm a diagnosis of pancreatic carcinoma.

CA19-9 is a sialylated Lewis (Lea) blood-group antigen, which was first identified as a ligand bound by monoclonal antibody 1116 NS 19-9[55]. CA19-9 levels are elevated in > 80% of patients with advanced pancreatic cancer[56]. However, up to 40% of patients with chronic pancreatitis also have elevated CA19-9 levels, suggesting that these levels do not reliably differentiate between patients with pancreatic carcinoma and those with chronic pancreatitis[57]. In contrast to these earlier findings, our meta-analysis shows that the mean sensitivity of a CA19-9 assay was 0.81; mean specificity, 0.81; maximum joint sensitivity and specificity, 0.81; and AUC, 0.88. These values suggest high overall accuracy. These sensitivity and specificity values are similar to the corresponding values of 0.79-0.81 and 0.82-0.90 reported in two previous meta-analyses[6,58]. Interestingly, both previous meta-analyses examined the ability of serum CA19-9 to differentiate pancreatic carcinoma from benign pancreatic diseases in general, not specifically chronic pancreatitis.

DOR is an indicator of test accuracy that combines sensitivity and specificity data into a single number[59]. The DOR is the ratio of the odds of positive test results in the patient with disease relative to the odds of positive test results in the patient without disease. Thus, higher DOR values indicate better discriminatory test performance. The mean DOR in our study was 19.31, implying that CA19-9 levels may be useful in diagnosing pancreatic carcinoma.

Although SROC and DOR meta-analyses provide evidence that CA19-9 can help differentiate between pancreatic cancer and chronic pancreatitis, these diagnostic indicators are difficult to interpret and relate to clinical practice. Therefore, we examined the differential diagnostic power of CA19-9 using the more clinically meaningful likelihood ratio[60]. PLRs and NLRs of > 10 or < 0.1 indicate high accuracy. The overall PLR value in our meta-analysis was 4.08, indicating that patients with pancreatic carcinoma are ~ 4-fold more likely to have elevated CA19-9 than patients with chronic pancreatitis. On the other hand, NLR in our meta-analysis was 0.24, meaning that a patient without elevated CA19-9 would still have a 24% chance of having pancreatic carcinoma, or that 24% of patients with pancreatic carcinoma would not have elevated CA19-9. This proportion is too high to rule out pancreatic cancer in patients who do not have elevated CA19-9. These findings suggest that serum CA19-9 levels are insufficient on their own to differentiate between pancreatic carcinoma and chronic pancreatitis. A better approach may be a combined diagnostic strategy drawing on clinical information as well as findings from cytology and histology of pancreatic tissue, radiology and/or endoscopic ultrasonography.

The present meta-analysis had several limitations. First, the exclusion of conference abstracts, letters to the editor, and non-English-language studies may have led to publication bias, although our bias analysis suggests that this was not a significant problem. Second, nonrandom misclassification bias may have occurred given that different studies used different approaches to diagnose chronic pancreatitis, including histology of pancreatic tissue, radiology, endoscopic ultrasonography and/or clinical information alone. Third, CA19-9 is not routinely measured when patients present with chronic pancreatitis, so the individuals in our meta-analysis may not be completely representative of this patient population. Fourth, 5%-10% of patients lacked the Lewis enzyme, fucosyltransferase, and so cannot present elevated CA19-9 even when tumor burden is high. Finally, we did not identify any large, blinded randomized controlled trials that satisfied our inclusion criteria.

In conclusion, our meta-analysis suggests that although CA19-9 showed considerable sensitivity and specificity for differentiating pancreatic carcinoma and chronic pancreatitis, the relatively high NLR means that CA19-9 levels by themselves have insufficient diagnostic accuracy. At the same time, elevated CA19-9 should increase suspicion of pancreatic carcinoma and may complement other clinical and histological findings to help confirm a diagnosis of cancer.

COMMENTS
Background

Pancreatic carcinoma and chronic pancreatitis show similar clinical manifestations. Carbohydrate antigen 19-9 (CA19-9) shows promise for differentiating the diseases.

Research frontiers

Differential diagnosis of pancreatic carcinoma and chronic pancreatitis remains a challenge, particularly in patients with pancreatic masses that may be benign (inflammatory) or malignant.

Innovations and breakthroughs

This is believed to be the first systematic review and meta-analysis of the utility of CA19-9 as a serum tumor marker and its sensitivity and specificity for distinguishing pancreatic carcinoma and chronic pancreatitis.

Applications

Available evidence suggests that elevated CA19-9 by itself is insufficient for differentiating pancreatic carcinoma and chronic pancreatitis. Nevertheless, elevated CA19-9 should increase suspicion of pancreatic carcinoma and therefore may complement other clinical findings to improve the accuracy of differential diagnosis.

Peer-review

This study evaluated the role of CA19-9 for differentiating pancreatic carcinoma and chronic pancreatitis and concluded that CA19-9 itself is insufficient, but elevated CA19-9 should increase suspicion of pancreatic carcinoma and therefore may complement other clinical findings to improve the accuracy of differential diagnosis. This result is reasonable and valuable for clinical practice.

Footnotes

P- Reviewer: Chen JQ, Kim TH S- Editor: Yu J L- Editor: Wang TQ E- Editor: Liu XM

References
1.  Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, Thun MJ. Cancer statistics, 2008. CA Cancer J Clin. 2008;58:71-96.  [PubMed]  [DOI]
2.  Sener SF, Fremgen A, Menck HR, Winchester DP. Pancreatic cancer: a report of treatment and survival trends for 100,313 patients diagnosed from 1985-1995, using the National Cancer Database. J Am Coll Surg. 1999;189:1-7.  [PubMed]  [DOI]
3.  Sohn TA, Yeo CJ, Cameron JL, Koniaris L, Kaushal S, Abrams RA, Sauter PK, Coleman J, Hruban RH, Lillemoe KD. Resected adenocarcinoma of the pancreas-616 patients: results, outcomes, and prognostic indicators. J Gastrointest Surg. 2000;4:567-579.  [PubMed]  [DOI]
4.  Ramesh H, Augustine P. Surgery in tropical pancreatitis: analysis of risk factors. Br J Surg. 1992;79:544-549.  [PubMed]  [DOI]
5.  Taylor B. Carcinoma of the head of the pancreas versus chronic pancreatitis: diagnostic dilemma with significant consequences. World J Surg. 2003;27:1249-1257.  [PubMed]  [DOI]
6.  Goonetilleke KS, Siriwardena AK. Systematic review of carbohydrate antigen (CA 19-9) as a biochemical marker in the diagnosis of pancreatic cancer. Eur J Surg Oncol. 2007;33:266-270.  [PubMed]  [DOI]
7.  Zhang S, Wang YM, Sun CD, Lu Y, Wu LQ. Clinical value of serum CA19-9 levels in evaluating resectability of pancreatic carcinoma. World J Gastroenterol. 2008;14:3750-3753.  [PubMed]  [DOI]
8.  Sawabu N, Watanabe H, Yamaguchi Y, Ohtsubo K, Motoo Y. Serum tumor markers and molecular biological diagnosis in pancreatic cancer. Pancreas. 2004;28:263-267.  [PubMed]  [DOI]
9.  Li D, Xie K, Wolff R, Abbruzzese JL. Pancreatic cancer. Lancet. 2004;363:1049-1057.  [PubMed]  [DOI]
10.  Agarwal B, Correa AM, Ho L. Survival in pancreatic carcinoma based on tumor size. Pancreas. 2008;36:e15-e20.  [PubMed]  [DOI]
11.  Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, Lijmer JG, Moher D, Rennie D, de Vet HC. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative. The Standards for Reporting of Diagnostic Accuracy Group. Croat Med J. 2003;44:635-638.  [PubMed]  [DOI]
12.  Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol. 2003;3:25.  [PubMed]  [DOI]
13.  Devillé WL, Buntinx F, Bouter LM, Montori VM, de Vet HC, van der Windt DA, Bezemer PD. Conducting systematic reviews of diagnostic studies: didactic guidelines. BMC Med Res Methodol. 2002;2:9.  [PubMed]  [DOI]
14.  Lau J, Ioannidis JP, Balk EM, Milch C, Terrin N, Chew PW, Salem D. Diagnosing acute cardiac ischemia in the emergency department: a systematic review of the accuracy and clinical effect of current technologies. Ann Emerg Med. 2001;37:453-460.  [PubMed]  [DOI]
15.  Moses LE, Shapiro D, Littenberg B. Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations. Stat Med. 1993;12:1293-1316.  [PubMed]  [DOI]
16.  Irwig L, Tosteson AN, Gatsonis C, Lau J, Colditz G, Chalmers TC, Mosteller F. Guidelines for meta-analyses evaluating diagnostic tests. Ann Intern Med. 1994;120:667-676.  [PubMed]  [DOI]
17.  Vamvakas EC. Meta-analyses of studies of the diagnostic accuracy of laboratory tests: a review of the concepts and methods. Arch Pathol Lab Med. 1998;122:675-686.  [PubMed]  [DOI]
18.  Suzuki S, Moro-oka T, Choudhry NK. The conditional relative odds ratio provided less biased results for comparing diagnostic test accuracy in meta-analyses. J Clin Epidemiol. 2004;57:461-469.  [PubMed]  [DOI]
19.  Westwood ME, Whiting PF, Kleijnen J. How does study quality affect the results of a diagnostic meta-analysis? BMC Med Res Methodol. 2005;5:20.  [PubMed]  [DOI]
20.  Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315:629-634.  [PubMed]  [DOI]
21.  Wang TH, Lin JT, Chen DS, Sheu JC, Sung JL. Noninvasive diagnosis of advanced pancreatic cancer by real-time ultrasonography, carcinoembryonic antigen, and carbohydrate antigen 19-9. Pancreas. 1986;1:219-223.  [PubMed]  [DOI]
22.  Safi F, Roscher R, Bittner R, Schenkluhn B, Dopfer HP, Beger HG. High sensitivity and specificity of CA 19-9 for pancreatic carcinoma in comparison to chronic pancreatitis. Serological and immunohistochemical findings. Pancreas. 1987;2:398-403.  [PubMed]  [DOI]
23.  Sakamoto K, Haga Y, Yoshimura R, Egami H, Yokoyama Y, Akagi M. Comparative effectiveness of the tumour diagnostics, CA 19-9, CA 125 and carcinoembryonic antigen in patients with diseases of the digestive system. Gut. 1987;28:323-329.  [PubMed]  [DOI]
24.  Friess H, Büchler M, Auerbach B, Weber A, Malfertheiner P, Hammer K, Madry N, Greiner S, Bosslet K, Beger HG. CA 494--a new tumor marker for the diagnosis of pancreatic cancer. Int J Cancer. 1993;53:759-763.  [PubMed]  [DOI]
25.  Röthlin MA, Joller H, Largiadèr F. CA 242 is a new tumor marker for pancreatic cancer. Cancer. 1993;71:701-707.  [PubMed]  [DOI]
26.  Haglund C, Lundin J, Kuusela P, Roberts PJ. CA 242, a new tumour marker for pancreatic cancer: a comparison with CA 19-9, CA 50 and CEA. Br J Cancer. 1994;70:487-492.  [PubMed]  [DOI]
27.  Kuno N, Kurimoto K, Fukushima M, Hayakawa T, Shibata T, Suzuki T, Sakakibara A, Katada N, Nakano S, Takayama T. Effectiveness of multivariate analysis of tumor markers in diagnosis of pancreatic carcinoma: a prospective study in multiinstitutions. Pancreas. 1994;9:725-730.  [PubMed]  [DOI]
28.  Pasquali C, Sperti C, D’Andrea AA, Costantino V, Filipponi C, Pedrazzoli S. Clinical value of serum TAG-72 as a tumor marker for pancreatic carcinoma. Comparison with CA 19-9. Int J Pancreatol. 1994;15:171-177.  [PubMed]  [DOI]
29.  Satake K, Takeuchi T. Comparison of CA19-9 with other tumor markers in the diagnosis of cancer of the pancreas. Pancreas. 1994;9:720-724.  [PubMed]  [DOI]
30.  Hámori J, Arkosy P, Lenkey A, Sápy P. The role of different tumor markers in the early diagnosis and prognosis of pancreatic carcinoma and chronic pancreatitis. Acta Chir Hung. 1997;36:125-127.  [PubMed]  [DOI]
31.  Safi F, Schlosser W, Kolb G, Beger HG. Diagnostic value of CA 19-9 in patients with pancreatic cancer and nonspecific gastrointestinal symptoms. J Gastrointest Surg. 1997;1:106-112.  [PubMed]  [DOI]
32.  Hayakawa T, Naruse S, Kitagawa M, Ishiguro H, Kondo T, Kurimoto K, Fukushima M, Takayama T, Horiguchi Y, Kuno N. A prospective multicenter trial evaluating diagnostic validity of multivariate analysis and individual serum marker in differential diagnosis of pancreatic cancer from benign pancreatic diseases. Int J Pancreatol. 1999;25:23-29.  [PubMed]  [DOI]
33.  Kim HJ, Kim MH, Myung SJ, Lim BC, Park ET, Yoo KS, Seo DW, Lee SK, Min YI. A new strategy for the application of CA19-9 in the differentiation of pancreaticobiliary cancer: analysis using a receiver operating characteristic curve. Am J Gastroenterol. 1999;94:1941-1946.  [PubMed]  [DOI]
34.  Manes G, Spada OA, Rabitti PG, Feola B, Misso S, Minerva A, Uomo G. Neopterin serum levels in pancreatic adenocarcinoma. Int J Pancreatol. 1999;25:31-37.  [PubMed]  [DOI]
35.  Slesak B, Harlozinska-Szmyrka A, Knast W, Sedlaczek P, van Dalen A, Einarsson R. Tissue polypeptide specific antigen (TPS), a marker for differentiation between pancreatic carcinoma and chronic pancreatitis. A comparative study with CA 19-9. Cancer. 2000;89:83-88.  [PubMed]  [DOI]
36.  Maire F, Micard S, Hammel P, Voitot H, Lévy P, Cugnenc PH, Ruszniewski P, Puig PL. Differential diagnosis between chronic pancreatitis and pancreatic cancer: value of the detection of KRAS2 mutations in circulating DNA. Br J Cancer. 2002;87:551-554.  [PubMed]  [DOI]
37.  Akashi T, Oimomi H, Nishiyama K, Nakashima M, Arita Y, Sumii T, Kimura T, Ito T, Nawata H, Watanabe T. Expression and diagnostic evaluation of the human tumor-associated antigen RCAS1 in pancreatic cancer. Pancreas. 2003;26:49-55.  [PubMed]  [DOI]
38.  Mu DQ, Wang GF, Peng SY. p53 protein expression and CA19.9 values in differential cytological diagnosis of pancreatic cancer complicated with chronic pancreatitis and chronic pancreatitis. World J Gastroenterol. 2003;9:1815-1818.  [PubMed]  [DOI]
39.  Cwik G, Wallner G, Skoczylas T, Krzyzanowski M, Ciechainski A, Madro P. Elevated tumor marker CA 19-9 in the differential diagnosis of pancreatic mass lesions. Ann Univ Mariae Curie Sklodowska Med. 2004;59:213-218.  [PubMed]  [DOI]
40.  Jiang JT, Wu CP, Deng HF, Lu MY, Wu J, Zhang HY, Sun WH, Ji M. Serum level of TSGF, CA242 and CA19-9 in pancreatic cancer. World J Gastroenterol. 2004;10:1675-1677.  [PubMed]  [DOI]
41.  Ventrucci M, Cipolla A, Racchini C, Casadei R, Simoni P, Gullo L. Tumor M2-pyruvate kinase, a new metabolic marker for pancreatic cancer. Dig Dis Sci. 2004;49:1149-1155.  [PubMed]  [DOI]
42.  Teich N, Kleeff J, Lochs H, Mössner J, Keim V, Friess H, Ockenga J. The presence of the proteolysis-inducing factor in urine does not predict the malignancy of a pancreatic tumour. BMC Gastroenterol. 2005;5:20.  [PubMed]  [DOI]
43.  Chang MC, Chang YT, Su TC, Yang WS, Chen CL, Tien YW, Liang PC, Wei SC, Wong JM. Adiponectin as a potential differential marker to distinguish pancreatic cancer and chronic pancreatitis. Pancreas. 2007;35:16-21.  [PubMed]  [DOI]
44.  Kuhlmann KF, van Till JW, Boermeester MA, de Reuver PR, Tzvetanova ID, Offerhaus GJ, Ten Kate FJ, Busch OR, van Gulik TM, Gouma DJ. Evaluation of matrix metalloproteinase 7 in plasma and pancreatic juice as a biomarker for pancreatic cancer. Cancer Epidemiol Biomarkers Prev. 2007;16:886-891.  [PubMed]  [DOI]
45.  Liao Q, Zhao YP, Yang YC, Li LJ, Long X, Han SM. Combined detection of serum tumor markers for differential diagnosis of solid lesions located at the pancreatic head. Hepatobiliary Pancreat Dis Int. 2007;6:641-645.  [PubMed]  [DOI]
46.  Bedi MM, Gandhi MD, Jacob G, Lekha V, Venugopal A, Ramesh H. CA 19-9 to differentiate benign and malignant masses in chronic pancreatitis: is there any benefit? Indian J Gastroenterol. 2009;28:24-27.  [PubMed]  [DOI]
47.  Firpo MA, Gay DZ, Granger SR, Scaife CL, DiSario JA, Boucher KM, Mulvihill SJ. Improved diagnosis of pancreatic adenocarcinoma using haptoglobin and serum amyloid A in a panel screen. World J Surg. 2009;33:716-722.  [PubMed]  [DOI]
48.  Liao WC, Wu MS, Wang HP, Tien YW, Lin JT. Serum heat shock protein 27 is increased in chronic pancreatitis and pancreatic carcinoma. Pancreas. 2009;38:422-426.  [PubMed]  [DOI]
49.  Morris-Stiff G, Teli M, Jardine N, Puntis MC. CA19-9 antigen levels can distinguish between benign and malignant pancreaticobiliary disease. Hepatobiliary Pancreat Dis Int. 2009;8:620-626.  [PubMed]  [DOI]
50.  Talar-Wojnarowska R, Gasiorowska A, Olakowski M, Lekstan A, Lampe P, Malecka-Panas E. Clinical value of serum neopterin, tissue polypeptide-specific antigen and CA19-9 levels in differential diagnosis between pancreatic cancer and chronic pancreatitis. Pancreatology. 2010;10:689-694.  [PubMed]  [DOI]
51.  Zapico-Muñiz E, Farré-Viladrich A, Rico-Santana N, González-Sastre F, Mora-Brugués J. Standardized peptidome profiling of human serum for the detection of pancreatic cancer. Pancreas. 2010;39:1293-1298.  [PubMed]  [DOI]
52.  Chung HW, Lim JB. Clinical significance of serum levels of immune-associated molecules, uric acid and soluble MHC class I chain-related molecules A and B, as diagnostic tumor markers for pancreatic ductal adenocarcinoma. Cancer Sci. 2011;102:1673-1679.  [PubMed]  [DOI]
53.  Gold DV, Gaedcke J, Ghadimi BM, Goggins M, Hruban RH, Liu M, Newsome G, Goldenberg DM. PAM4 enzyme immunoassay alone and in combination with CA 19-9 for the detection of pancreatic adenocarcinoma. Cancer. 2013;119:522-528.  [PubMed]  [DOI]
54.  Kaur S, Chakraborty S, Baine MJ, Mallya K, Smith LM, Sasson A, Brand R, Guha S, Jain M, Wittel U. Potentials of plasma NGAL and MIC-1 as biomarker(s) in the diagnosis of lethal pancreatic cancer. PLoS One. 2013;8:e55171.  [PubMed]  [DOI]
55.  Koprowski H, Herlyn M, Steplewski Z, Sears HF. Specific antigen in serum of patients with colon carcinoma. Science. 1981;212:53-55.  [PubMed]  [DOI]
56.  Boeck S, Stieber P, Holdenrieder S, Wilkowski R, Heinemann V. Prognostic and therapeutic significance of carbohydrate antigen 19-9 as tumor marker in patients with pancreatic cancer. Oncology. 2006;70:255-264.  [PubMed]  [DOI]
57.  Rosty C, Goggins M. Early detection of pancreatic carcinoma. Hematol Oncol Clin North Am. 2002;16:37-52.  [PubMed]  [DOI]
58.  Ballehaninna UK, Chamberlain RS. The clinical utility of serum CA 19-9 in the diagnosis, prognosis and management of pancreatic adenocarcinoma: An evidence based appraisal. J Gastrointest Oncol. 2012;3:105-119.  [PubMed]  [DOI]
59.  Glas AS, Lijmer JG, Prins MH, Bonsel GJ, Bossuyt PM. The diagnostic odds ratio: a single indicator of test performance. J Clin Epidemiol. 2003;56:1129-1135.  [PubMed]  [DOI]
60.  Deeks JJ. Systematic reviews in health care: Systematic reviews of evaluations of diagnostic and screening tests. BMJ. 2001;323:157-162.  [PubMed]  [DOI]