EJSO 33 (2007) 266e270

www.ejso.com

Review

Systematic review of carbohydrate antigen (CA 19-9) as a biochemical marker in the diagnosis of pancreatic cancer
K.S. Goonetilleke, A.K. Siriwardena*
Department of Surgery, Hepatobiliary Surgical Unit, Manchester Royal Inrmary, Oxford Road, Manchester M13 9WL, UK Accepted 3 October 2006 Available online 9 November 2006

Abstract Background: Although many biochemical markers have been examined in pancreatic cancer none are denitive for pre-operative diagnosis. This systematic review examines studies using biochemical markers for the diagnosis of pancreatic cancer in order to appraise their role in contemporary management algorithms. Methods: A search of the MEDLINE database was undertaken using the key words pancreatic neoplasm and serum tumour marker. Only studies providing original data on sensitivity and specicity are included and data are presented on diagnostic accuracy, effect of cholestasis and the relation of tumour stage to blood levels of markers. Results: CA 19-9 is the most extensively evaluated with pooled data from 2283 patients. The median sensitivity of CA 19-9 for diagnosis is 79 (70e90%) and median specicity 82 (68e91%). CA 19-9 elevation in non-malignant jaundice results in a fall in specicity. Combination with other markers improves accuracy. Conclusion: As the most extensively evaluated marker, CA 19-9 should be used in contemporary algorithms for the diagnosis of pancreatic cancer. Elevated values should be repeated after relief of jaundice. 2006 Published by Elsevier Ltd.
Keywords: Serum tumour markers; Pancreatic neoplasm; CA 19-9; CA 50; CEA

Review criteria Literature search and data extraction strategy A computerised search was made of the MEDLINE database for the period from January 1990 to December 2005 inclusive. The OVID search engine (Version 9; Ovid Technologies, New York, NY, USA) was employed. The MESH headings Pancreatic neoplasm, diagnosis, CA19-9, CA 50, CEA, Dupan-2, Span-1, Serum tumour markers were used. The search was restricted to articles in English and documents relating to humans. This resulted in 348 hits. The abstracts of these reports were then retrieved and studied. At this stage, studies were excluded if they were case reports, review articles and reports not providing data on
* Corresponding author. Tel.: 44 161 276 4250; fax: 44 161 276 4530. E-mail address: ajith.siriwardena@cmmc.nhs.uk (A.K. Siriwardena). 0748-7983/$ - see front matter 2006 Published by Elsevier Ltd. doi:10.1016/j.ejso.2006.10.004

serum tumour markers in pancreas cancer. These exclusions resulted in a nal study population of 26 manuscripts providing original data on biochemical markers for the diagnosis of pancreatic cancer. Data were extracted in the following categories: number of patients with pancreatic cancer, number with histological conrmation of malignancy, proportion of patients with jaundice, composition of control group (including proportion with jaundice in control), cut-off points for markers, type of assay used for marker analysis, sensitivity, specicity and nally positive and negative predictive value. All extracted data were crosschecked between authors to rule out discrepancy. Particular attention was paid to histological conrmation of malignancy: unless a report specically mentioned that there was a tissue diagnosis either by biopsy or resection, the diagnosis of pancreatic cancer was not regarded as histological conrmed for the purposes of this study. Data are presented in the following categories: CA 19-9 in the diagnosis of pancreatic cancer, other markers for pancreatic

K.S. Goonetilleke, A.K. Siriwardena / EJSO 33 (2007) 266e270

267

cancer, effect of cholestasis on diagnostic accuracy and relation of tumour stage to blood levels of markers. Introduction A wide range of biochemical tumour markers has been reported to be elevated in patients with pancreatic cancer.1e6 These include tests based on monoclonal antibodies raised against tumour-associated glycoproteins such as carbohydrate antigen (CA 19-9)7 and carbohydrate antigen 242 (CA 242)8, to tests based on the detection of tumour isoenzymes such as tumour-M2-pyruvate kinase (Tu-M2PK)9,10 and oncofoetal antigens.11 The presence of a wide range of tests is an indicator that there is no single denitive diagnostic test for pancreatic cancer. Practical application of diagnostic tests has been limited by the confounding effect of co-existent jaundice on accuracy.12e16 Efforts to circumvent these problems have included the use of tests in combination17e19 or a bank of diagnostic tests.20,21 The relative paucity of development in serological diagnosis of pancreatic cancer has occurred in the context of major advances in diagnosis based on cross-sectional22e24 and endoscopic imaging.25e27 Current pre-operative diagnosis of pancreatic cancer relies on high-resolution, contrast-enhanced computed tomography22 with evidence that endoscopic ultrasonography can provide complementary information.25 Given the heterogeneous nature of the studies on biochemical markers for diagnosis of pancreatic cancer, uncertainty remains over the optimal test, timing of sampling and the role of serological markers in surveillance. In this setting, structured systematic reviews may provide information on consistent trends gleaned from a series of studies. Thus the aim of this study is to undertake a systematic review of biochemical markers used in the diagnosis of pancreatic cancer to examine the relative place of these tests in contemporary management algorithms. Carbohydrate antigen (CA 19-9) CA 19-9 is a predominantly carbohydrate antigen which was dened from the culture medium of a colorectal cancer cell line. It is a high molecular weight glycolipid derived from a monoclonal antibody isolated from mice, which is immunised with a human colon cell line. Twenty-two reports3,12,18e20,28e44 address the role of CA 19-9 in the diagnosis of pancreatic cancer. With the sole exception of Jangs study18, all reports use radioimmunoassay for measurement of CA 19-9 (Jangs report utilizes an enzymelinked immunosorbent assay ELISA). Data are reported on 2283 patients in whom there was unequivocal histological conrmation of malignancy in 945 (41%). The median sensitivity of CA 19-9 for the diagnosis of pancreatic cancer examining pooled data from all series is 79 (70e90%). The median specicity is 82 (68e91%). The median

positive predictive value (PPV) 72 (41e95) and the median negative predictive value 81 (65e98). Carbohydrate antigen CA 242, carbohydrate antigen CA 50, carcinoembryonic antigen (CEA), SPAN-1 and sialyl-lcat-N-tetraose (DUPAN-2) CA 242 is a cancer associated glycoconjugate expressed in mucin and found predominantly in the sera of pancreatic cancer patients. It is a sialylated carbohydrate antigen coexpressed with CA 50, and situated in the same macromolecule. Ten reports3,18,30,32,38,39,44e47examine the diagnostic accuracy of CA 242 in a total of 806 patients. From pooled data, the median sensitivity of CA 242 is 75 (65e82%), and the median specicity 90 (65e95%). The median PPV is 83 (37e96) and the median NPV 74 (52e96). CA 50 which may present as a high molecular weight glycolipid or glycoprotein was rst identied in 1983. It is detected in a number of epithelial tumours as well as in normal adult pancreatic and gallbladder tissue. Nine reports20,29,30,32,34,36,45e47 examine the diagnostic accuracy of CA 50 in a total of 1137 patients. From pooled data, CA 50 has a median sensitivity of 83 (65e96%) and a median specicity of 73 (34e90%) for the diagnosis of pancreatic cancer with a median PPV of 72 (26e95) and median NPV of 79 (61e99). Span-1 is a high molecular weight glycoprotein recognized by a murine monoclonal antibody produced against a human pancreatic cell line. Its epitope includes a sialic acid- like CA 19-9. A sandwich radioimmunoassay has recently been developed and clinically tested for this marker. Span-1 (4 reports on a pooled total of 952 patients)28,29,36,37 has a median sensitivity of 82 (81e89%) and a median specicity of 83 (67e85%). The median PPV is 50 (45e88) and median NPV is 93 (77e98). Dupan-2 is a monoclonal antibody raised against a human pancreatic adenocarcinoma cell line. The epitope of Dupan-2 antigen (sialyl-lcat-N-tetraose) is a glycosylated antigen expressed on a mucin like molecule, and is the precursor of CA 19-9. Dupan-2 (4 reports, 798 patients)12,20,28,36 has a median sensitivity of 65 (48e80%) and a median specicity of 80 (75e85%). The median PPV is 68 (45e83) with a median NPV of 78 (60e89). Carcinoembryonic antigen is a heterogeneous group of glycoproteins containing 50e80% carbohydrates and a single protein chain of about 800 amino acids. CEA (13 reports, 1323 patients)3,12,20,28,30,35-37,40,43e45,48 shows the lowest median sensitivity of 54 (40e92%) and a median specicity of 79 (59e90%). The median PPV is 65.5 (25e91) and median NPV is 72 (53e98). Effect of jaundice on sensitivity and specicity in pancreas cancer Three of four12,31,40,41 studies comparing CA 19-9 for diagnosis of pancreas cancer either in the presence or

268

K.S. Goonetilleke, A.K. Siriwardena / EJSO 33 (2007) 266e270

absence of jaundice show that jaundice was associated with increased sensitivity. However, any benet is offset by all studies showing a fall in specicity in the presence of jaundice. Ohshios study shows that similar ndings were seen with CEA and Dupan-2.12 Results for the interpretation of CA19-9, CEA and Dupan-2 in the presence of cholestasis are shown for the studies available. Effect of tumour stage on accuracy of tumour markers (UICC TNM classication) Ten studies18,29,31,33,36,38e40,44,48 examined the effect of tumour stage on the accuracy of tumour markers (nine evaluated CA 19-9 either alone or in combination). In all studies, the trend was for greater sensitivity with advancing stage of cancer at the time of initial presentation. Discussion A tumour marker that allows for the reliable diagnosis of pancreatic cancer has proved elusive. Many studies have examined markers in a large number of patients but despite this it remains difcult to draw denitive conclusions. This systematic review has examined the available reports on the use of biochemical markers for the diagnosis of pancreatic cancer. The search parameters and history are described so that the yield can be independently duplicated. As with all systematic reviews it remains possible that studies containing potentially valuable data may not have eluded detection. Other potential limitations include publication bias in favour of studies reporting positive data and the risk of compounding inaccuracy when pooling predictive data (sensitivity and specicity). Acknowledging these limitations, an important strength of this systematic review is that it highlights the considerable variation in study design and composition of patient cohorts that exists between studies. Key variations are in the nature of the control group (non-pancreatic cancer controls pooled with non-cancer controls), the rigour used to obtain a tissue diagnosis of malignancy in patients with suspected pancreatic cancer and in the proportion of patients with jaundice. There are also considerable variations in the cut-off values used to distinguish the normal range. Considering these limitations, what can be learnt from this study, thought to be the largest pooled cohort data report on the utility of biochemical markers for the diagnosis of pancreatic cancer. First, it is readily apparent that CA 19-9 is the most thoroughly evaluated test. With an overall pooled sensitivity from 2283 patients of 79% and a specicity of 82% across a range of reports over a relatively prolonged period of time it is the standard for comparison of other tests. The principal limitation of CA 19-9 is that it can be elevated in patients with non-malignant obstructive jaundice resulting in impaired specicity. Efforts to circumvent this limitation include the use of higher cut-off values41

(300 U/ml increases the specicity to 87% in patients with cholestasis) and the use of an array of tumour markers. There is no additional advantage in sampling bile CA 19-9 as the bile to serum ratio is 1:1.12 Knowledge of the Lewis blood group status may aid interpretation of CA 19-9 in cancer as Rosen reported that the 10% of the Caucasian population who are Lewis genotype negative are unable to express CA 19-9.34 Serial CA 19-9 analysis after relief of jaundice may be useful although in Ohshios study, CA 19-9 remained elevated. In an interesting report, Malesci and colleagues reported that a normal CA 19-9 combined with normal imaging effectively ruled out pancreatic cancer. A further study done by Ritts and colleagues combining diagnostic imaging and preoperative CA19-9 levels suggest CA19-9 to be a good adjunct when combined with radiological imaging in non-jaundiced patients.49 In 84 patients with histologically conrmed pancreatic cancer, CA19-9 values >40 u/ml in the presence of positive or equivocal radiological ndings was highly suggestive of a diagnosis of pancreatic cancer. (positive predictive value 0.94). These ndings are worthy of further study. A common approach is to utilize CA 19-9 in combination with other markers. It is difcult to draw conclusions from these reports but individual studies have suggested that the addition of CAM 17.1/WGA31 or CA 24232 increased diagnostic yield. More dramatically, composite marker systems such as CAMPAS-P, which are based on CA 19-9 plus 9 other markers, differentiated all 35 cancers from the 32 patients with benign disease in Saitos study. The sensitivity of biochemical markers is increased in patients with advanced disease but in contemporary hepato-pancreato-biliary practice it would be expected that these patients would be readily identied by imaging modalities. Increasing evidence suggests that Tu-M2-PK may become a valuable plasma marker for the diagnosis of cancer. Numerous investigators have reported on the utility of TuM2-PK in the early diagnosis of pancreatic cancer. The sensitivity (71%e79%) and specicity (64%e95%) reported in various series were encouraging.50e54 Tu-M2-PK has an advantage over CA19-9 in that it is not affected by cholestasis. The potential promise of tumour metabolism-based markers such as Tu-M2-PK as markers of disseminated malignancy in patients with no overt evidence of metastasis require further study.55 Finally, although several studies have examined the role of CA 19-9 as a population screening tool56e58 it is not sufciently reliable for this role. There have been large studies done in Japan and Korea to assess the usefulness of CA19-9 as screening tool to diagnose pancreatic cancer in both asymptomatic and symptomatic patients. Mass screening was conducted in Japan over two different periods, 1984e1985 and 1987e1988 on asymptomatic and symptomatic outpatients, all over 40, in 17 hospitals. All patients underwent either US or serum CA19-9 and

K.S. Goonetilleke, A.K. Siriwardena / EJSO 33 (2007) 266e270

269

elastase1 levels. When results were abnormal, further studies were carried out, including repeat US, ERCP, CT etc. A nal diagnosis was made either clinically or histologically. Mass screening of 10 162 asymptomatic persons >40 resulted in the detection of only 4 cases of pancreatic cancer. A screening of 2678 asymptomatic persons >40 in another later survey produced no cases of pancreatic cancer. By contrast, in screening 4506 symptomatic outpatients >40, they found 85 pancreatic cancers (2%) of which 28 (32%) were resected. 56 of more than 4200 symptomatic outpatients(1%) in the later survey were diagnosed with cancer of pancreas, 19 (34%) of whom underwent resection. Similar study undertaken in Korea involving 70 940 asymptomatic patients detected only 4 patients with pancreatic cancer. Therefore, using serum CA19-9 with US to detect pancreatic cancer through mass screening is ineffective in asymptomatic patients but effective to screen symptomatic outpatients in nding resectable tumours. In summary, the recommendations made as a result of this systematic review are that CA 19-9 as the most thoroughly evaluated biochemical marker should be used in current management algorithms. Intelligent, contemporary use of CA 19-9 would suggest that elevated values be repeated after relief of jaundice (bearing in mind that false positives may persist), higher cut-offs considered in jaundiced patients and a normal CA 19-9 taken to add weight to the exclusion of pancreatic cancer in a patient with normal cross-sectional imaging. Although other markers have been evaluated, none at present can be regarded as consistently superior to CA 19-9. A nal important lesson to learn from this systematic review is that future studies must look to obtain a diagnosis of cancer more rigorously and pay greater detail to protocol design in terms of inclusion of jaundiced patients and the nature of the control group.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18. 19.

20.

21.

References
1. Niederau C, Grendell JH. Diagnosis of pancreatic carcinoma. Imaging techniques and tumor markers. Pancreas 1992;7:6686. 2. Louhimo J, Alfthan H, Stenman UH, Haglund C. Serum HCG beta and CA 72-4 are stronger prognostic factors than CEA, CA 19-9 and CA 242 in pancreatic cancer. Oncology 2004;66:12631. 3. Ni XG, Bai XF, Mao YL, et al. The clinical value of serum CEA, CA19-9, and CA242 in the diagnosis and prognosis of pancreatic cancer. Eur J Surg Oncol 2005;31:1649. 4. Jiang XT, Tao HQ, Zou SC. Detection of serum tumor markers in the diagnosis and treatment of patients with pancreatic cancer. Hepatobiliary Pancreat Dis Int 2004;3:4648. 5. Molina LM, Diez M, Cava MT, et al. Tumor markers in pancreatic cancer: A comparative clinical study between CEA, CA19-9 and CA 50. Int J Biol Markers 1990;5:12732. 6. Ban G, Zerbi A, Pastori S, Parolini D, Di-Carlo V, Bonini P. Behaviour of tumor markers CA19-9, CA195, CAM 43, CA242, and TPS in the diagnosis and follow up of pancreatic cancer. Clin Chem 1993;39: 4203. 7. Magnani JL, Steplewski Z, Koprowski H, Ginsburg V. Identication of the gastrointestinal and pancreatic cancer-associated antigen detected 22.

23. 24.

25.

26.

27.

by monoclonal antibody 19-9 in the sera of patients as a mucin. Cancer Res 1983;43:548992. Lindholm L, Johansson C, Jansson EL, Hallberg C, Nilsson O. An immunometric assay (IRMA) for the CA50 antigen. In: Holmgren J, editor. Tumour marker antigens. Lund, Sweden: Studentlitteratur; 1985, p. 122. Hardt PD, Ngoumou BK, Rupp J, Schnell-Kretschmer H, Kloer HU. Tumor M2-pyruvate kinase: a promising tumor marker in the diagnosis of gastro-intestinal cancer. Anticancer Res 2000;20:49658. Schneider J, Schulze G. Comparison of Tumor M2-Pyruvate Kinase (Tumor M2-PK), Carcinoembryonic Antigen (CEA), Carbohydrate antigens CA19-9 and CA 72-4 in the diagnosis of gastrointestinal cancer. Anticancer Res 2003;23:508993. Nishida K, Sugiura M, Yoshikawa T, Kondo M. Enzyme immunoassay of pancreatic oncofetal antigen (POA) as marker of pancreatic cancer. Gut 1985;26:4505. Ohshio G, Manabe T, Watanabe Y, et al. Comparative studies of DUPAN-2, carcinoembryonic antigen, and CA19-9 in the serum and bile of patients with pancreatic and biliary tract diseases: Evaluation of the inuence of obstructive jaundice. Am J Gastroenterol 1990;85: 13706. Basso D, Meggiato T, Fabris C, et al. Extra-hepatic cholestasis determines a reversible increase of glycoproteic tumour markers in benign and malignant diseases. Eur J Clin Invest 1992;22:8004. Pearce S, Thornes H, Carr D, Tanner A. Diagnostic pitfall; interpretation of CA19-9 concentrations in the presence of hepatic dysfunction. Gut 1994;35:7078. Basso D, Fabris C, Meggiato T, et al. Serum levels of CA19-9 and tissue polypeptide antigen (TPS) are related to the presence of a benign extra-hepatic cholestasis. Med Sci Res 1989;17:134. Mann DV, Edwards R, Ho S, Lau WY, Glazer G. Elevated tumour marker CA19-9: clinical interpretation and inuence of obstructive jaundice. Eur J Surg Oncol 2000;26:4749. Carpelan-Holmstrom M, Louhimo J, Stenman UH, Alfthan H, Haglund CCEA. CA19-9 and CA 72-4 improve the diagnostic accuracy in gastrointestinal cancers. Anticancer Res 2002;22:23116. Jiang J, Wu C, Deng H, et al. Serum level of TGSF, CA242 and CA199 in pancreatic cancer. World J Gastroenterol 2004;10:16757. Ventrucci M, Cipolla A, Racchini C, Casadei R, Simoni P, Gullo L. Tumor M2-Pyruvate Kinase, a new metabolic marker for pancreatic cancer. Digest Dis Sci 2004;49:114955. Saito S, Taguchi K, Nishimura N, et al. Clinical usefulness of computer-assisted diagnosis using combination assay of tumor markers for pancreatic carcinoma. Cancer 1993;72:3818. Hayakawa T, Naruse S, Kitagawa M, et al. 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;25:239. McCarthy MJ, Evans J, Sagar G, Neoptolemos JP. Prediction of resectability of pancreatic malignancy by computed tomography. Br J Surg 1998;85:3205. Chong M, Freeny PC, Schmiedl UP. Pancreatic arterial anatomy: depiction with dual-phase helical CT. Radiology 1998;208:53742. Vedantham S, Lu DS, Reber HA, Kadell B. Small peripancreatic veins: improved assessment in pancreatic cancer patients using thinsection pancreatic phase helical CT. AJR Am J Roentgenol 1998; 170:37783. Rosch T, Lightdale C, Botet JF, et al. Localization of pancreatic endocrine tumors by endoscopic ultrasonography. N Engl J Med 1992;326: 17216. Howard TJ, Chin AC, Streib EW, Kopecky KK, Wiebke EA. Value of helical computed tomography, angiography and endoscopic ultrasound in determining resectability of periampullary carcinoma. Am J Surg 1997;174:23741. Mertz HR, Sechopoulos P, Delbeke D, Leach SD. EUS, PET, and CT scanning for evaluation of pancreatic adenocarcinoma. Gastrointest Endosc 2000;52:36771.

270

K.S. Goonetilleke, A.K. Siriwardena / EJSO 33 (2007) 266e270 45. Pasanen P, Eskelinen M, Partanen K, Pikkarainen P, Pentilla I, Alhava E. Clinical evaluation of a new serum tumour marker CA 242 in pancreatic carcinoma. Br J Cancer 1992;65:7314. 46. Nilsson O, Johansson C, Glimelius B, et al. Sensitivity and specicity of CA242 in gastro-intestinal cancer. A comparison with CEA, CA 50 and CA19-9. Br J Cancer 1992;65:21521. 47. Palsson B, Masson P, Andren-Sandberg A. The inuence of cholestasis on CA50 and CA242 in pancreatic cancer and benign biliopancreatic diseases. Scan J Gastroenterol 1993;28:9817. 48. Andicoechea A, Vizoso F, Alexandre E, et al. Comparative study of Carbohydrate antigen 195 and Carcinoembryonic antigen for the diagnosis of pancreatic carcinoma. World J Surg 1999;23: 22732. 49. Ritts Jr E, Nagorney D, Jacobsen D, Talbot R, Zurawski Jr V. Comparison of preoperative serum CA19-9 levels with results of diagnostic imaging modalities in patients undergoing laparotomy for suspected pancreatic or gallbladder disease. Pancreas 1994;9:70716. 50. Cerwenka H, Aigner R, Bacher H, et al. Tu M2-PK (Pyruvate kinase type tumour M2), CA 19-9 and CEA in patients with benign, malignant and metastasizing pancreatic lesions. Anticancer Res. 1999;19: 84952. 51. Hardt PD, Ngoumou BK, Rupp J, Schnell-Kretschemer H, Kloer HU. Tumour M2-Pyruate kinase: a promising tumour marker in the diagnosis of gastrointestinal cancer. Anticancer Res 2000;20:49658. 52. Schneider J, Schulze G. Comparison of tumour M2-pyruvate kinase (Tumour M2-PK), carcinoembryonic antigen (CEA), carbohydrate antigen CA 19-9 and CA 72-4 in the diagnosis of gastrointestinal cancer. Anticancer Res 2003;23:508994. 53. Schulze G. The tumour marker Tumour M2-PK: an application in the diagnosis of gastrointestinal cancer. Anticancer Res 2000;20: 49614. 54. Oremek GM, Eigenbrodt E, Radle J, Zeuzem ST, Seiffert UB. Value of the serum levels of the Tumour marker Tu M2-PK in pancreatic cancer. Anticancer Res. 1997;17:30314. 55. Siriwardana HPP, King NKK, France MW, Siriwardena AK. Plasma tumour-M2-pyruvate kinase and serum carbohydrate antigen for the preoperative diagnosis of pancreatic cancer. Br J Surg 2005;92:99 101. (abstract). 56. Satake K, Takeuchi T, Homma T, Ozaki H. CA19-9 as a screening and diagnostic tool in symptomatic patients: the Japanese experience. Pancreas 1994;9:7036. 57. Kim J, Lee KT, Lee JK, Paik SW, Rhee JC, Choi KW. Clinical usefulness of carbohydrate antigen 19-9 as a screening test for pancreatic cancer in an asymptomatic population. J Gastroenterol Hepatol 2004;19:1826. 58. Homma T, Tsuchiya R. The study of the mass screening of persons without symptoms and of the screening of outpatients with gastrointestinal complaints or icterus for pancreatic cancer in Japan, using CA19-9 and elastase-1 or ultrasonography. Int J Pancreatol 1991;9: 11924.

28. Kiriyama S, Hayakawa T, Kondo T, et al. Usefulness of a new tumor marker, Span-1, for the diagnosis of pancreas cancer. Cancer 1990;65: 155761. 29. Kobayashi T, Kawa S, Tokoo M, et al. Comparative study of CA 50 (time resolved uoroimmunoassay), Span-1, and CA 19-9 in the diagnosis of pancreatic cancer. Scan J Gastroenterol 1991;26:78797. 30. Kuusela P, Haglund C, Roberts P. Comparison of a new tumour marker CA242 with CA19-9, CA 50 and carcinoembryonic antigen (CEA) in digestive tract diseases. Br J Cancer 1991;63:63640. 31. Parker N, Makin C, Ching C, et al. A new enzyme linked lectin/mucin antibody sandwich assay (CAM 17.1/WGA) assessed in combination with CA19-9 and peanut lectin binding assay for the diagnosis of pancreatic cancer. Cancer 1992;70:10628. 32. Rothlin M, Joller H, Largiader F. CA 242 is a new tumor marker for pancreatic cancer. Cancer 1993;71:7017. 33. Malesci A, Montorsi M, Mariani A, et al. Clinical utility of the serum CA19-9 test for diagnosing pancreatic cancer in symptomatic patients: a prospective study. Pancreas 1992;7:497502. 34. Rosen A, Linder S, Harmenberg U, Pegert S. Serum levels of CA 19-9 and CA50 in relation to Lewis blood cell status in patients with malignant and benign pancreatic disease. Pancreas 1993;8:1605. 35. Friess H, Buchler M, Auerbach B, et al. CA 494- a new tumor marker for the diagnosis of pancreatic cancer. In J Cancer 1993;53:75963. 36. Satake K, Takeuchi T. Comparison of CA19-9 with other tumor markers in the diagnosis of cancer of the pancreas. Pancreas 1994;9:7204. 37. Kuno N, Kurimoto K, Fukushima M, et al. Effectiveness of multivariate analysis of tumor markers in diagnosis of pancreatic carcinoma: a prospective study in multiinstitutions. Pancreas 1994;9:72530. 38. Haglund C, Lundin J, Kuusela P, Roberts P. CA 242, a new tumour marker for pancreatic cancer: a comparison with CA19-9, CA 50 and CEA. Br J Cancer 1994;70:48792. 39. Kawa S, Tokoo M, Hasabe O, et al. Comparative study of CA242 and CA19-9 for the diagnosis of pancreatic cancer. Br J Cancer 1994;70: 4816. 40. Gansauge F, Gansauge S, Parker N, et al. CAM 17.1- a new diagnostic marker in pancreatic cancer. Br J Cancer 1996;74:19972002. 41. Kim H, Kim M, Myung S, et al. A new strategy for the application of CA19-9 in the differentiation of pancreaticobiliary cancer: analysis using a receiver operating characteristic curve. Am J Gatroenterol 1999; 94:19416. 42. Slesak B, Harlozinska-Szmyrka A, Knast W, Sedlaczek P, Dalen A, Einarsson R. Tissue polypeptide specic antigen (TPS), a marker for differentiation between pancreatic carcinoma and chronic pancreatitis. Cancer 2000;89:838. 43. Nazli O, Bozdag A, Tansug T, Kir R, Kaymak E. The diagnostic importance of CEA and CA19-9 for the early diagnosis of pancreatic carcinoma. Hepatogatroenterology 2000;47:17502. 44. Ozkan H, Kaya M, Cengiz A. Comparison of tumor marker CA242 with CA19-9 and carcinoembryonic antigen (CEA) in pancreatic cancer. Hepatogastroenterology 2003;50:166974.