Comparison of Plasma Tu-M2-PK and CA19-9 in Pancreatic Cancer Maiken Thyregod Joergensen, MD, PhD,* Niels H.H. Heegaard, DMSci,Þ and Ove B. Schaffalitzky de Muckadell, DMSci* Objectives: The performance of the 2 tumor markers carbohydrate antigen 19-9 (CA19-9) and tumor M2 pyruvate kinase (Tu-M2-PK) sep- arately and in combination detecting pancreatic ductal adenocarcinoma (PDAC) was evaluated in a prospective study. Methods: The study comprised 103 patients referred because of sus- picion of pancreatic cancer. Of these, 51 patients had their conditions diagnosed as PDAC, whereas this diagnosis was ruled out in 52 after 12 months of follow-up. The performance of Tu-M2-PK was compared with that of CA19-9 using cutoff values 15 and 37 U/mL, respectively. Results: The sensitivity of Tu-M2-PK and CA19-9 in detecting PDAC was 55% and 86% at specificities of 52% and 73%, respectively. The area under the curve (AUC) of Tu-M2-PK was 0.55 and that of CA19-9 was 0.84. Combining the 2 markers did not significantly improve AUC (AUC = 0.85, P = 0.72) compared with CA19-9 when used alone. The presence of chronic pancreatitis or jaundice causes increased levels of CA19-9 but does not influence Tu-M2-PK. Conclusions: Tu-M2-PK was inferior to CA19-9 as marker of PDAC. Tu-M2-PK may have a role in diagnosing PDAC because it is not af- fected by cholestasis or Lewis phenotype. Neither tumor marker can stand alone in the diagnosis of PDAC. Key Words: pancreatic ductal adenocarcinoma, carbohydrate antigen 19-9, tumor M2 pyruvate kinase, receiver operating characteristic curves, bilirubin (Pancreas 2010;39: 243Y247) T he prevalence of pancreatic cancer in Northern Europe 1 is 8 to 12 per 100,000. The median survival time after diagnosis is only 4 to 6 months, and less than 5% are alive after 5 years. After radical surgery, the 5-year survival is 25%, but because of advanced tumor stage at diagnosis, less than 20% of the patients are offered surgery. 2,3 If pancreatic ductal adenocarcinoma (PDAC) is diagnosed at an early stage (tumor G1 cm and without lymph node metastases), the 5-year survival increases to 50%. 4 Typically, clinical symptoms are few and nonspecific at early stages, and thus, the development of biochemical tests for early detection of pancreatic cancers has high priority. Currently, carbohydrate antigen 19-9 (CA19-9) is the most important bio- marker for the diagnosis of PDAC in symptomatic patients. It is also used in monitoring the clinical course, as a prognostic factor, 5 and as a predictive factor for response to chemotherapy or chemoradiation. 6,7 CA19-9 is a tumor-associated antigen, defined by the monoclonal antibody 1116 NS 19-9, which reacts with the sialylated Lewis a blood group antigen present in the glycoprotein serum fraction. 8 However, approximately 5% to 10% of the general population has the Lewis ajbj phenotype, which means that they are unable to synthesize the CA19-9 antigen and will not have elevated levels secondary to pancreatic cancer or other malignancies. 9 Elevated CA19-9 levels are also found in patients with hepatocellular, colorectal, and ovarian neoplasms and in patients with cholestasis due to benign con- ditions such as pancreatitis and choledocholithiasis. 10 CA19-9 has been reported to detect PDAC with a sensitivity of 58% to 87% and a specificity of 93% in comparing patients with biopsy-proven adenocarcinoma in the pancreas with patients with benign pancreatic disease, nonpancreatic sources of ab- dominal pain, benign jaundice, nonpancreatic malabsorption, and patients with renal failure in dialysis. 11 A new tumor marker superior to CA19-9 should show a higher sensitivity and specificity when tested in the same pop- ulation and it should not be elevated in diseases that are differen- tial diagnoses to PDAC (eg, pancreatitis). The putative new PDAC marker tumor M2 pyruvate kinase (Tu-M2-PK) has been shown not to be affected by cholesta- sis. 12,13 Four tissue-specific isoforms of the pyruvate kinase exist: type L, in liver and kidney; type R, in erythrocytes; type M1, in muscle, brain, and heart; and type M2, in lung, kidney, and in undefined and proliferating tissue as in carcinogenesis. All these isoenzymes exist as enzymatically active tetramers. After phosphorylation of serine and tyrosine, the tetrameric M2- PK shifts to a dimeric form. The dimeric form has a low affinity for phosphoenole pyruvate, and consequently, energy-rich gly- colytic metabolites accumulate and enable cell proliferation even in hypoxic conditions, hence the denomination Tu-M2- PK. 14,15 The Tu-M2-PK values in the plasma of healthy indi- viduals are age- and sex-independent. 16 Three studies have found that the levels of plasma Tu-M2-PK correlated with the stage of pancreatic cancer. 12,17,18 Combining Tu-M2-PK and CA19-9 has been shown to increase sensitivity. 12,19 At the mo- ment, 5 studies on the role of Tu-M2-PK in diagnosing pan- creatic cancer 12,17,20Y22 have been published. A recent meta-analysis of these studies found a diagnostic odds ratio (DOR) of 35 (95% confidence interval [CI], 19.7Y62.3) for Tu-M2-PK and 44 (95% CI, 26.5Y73.1) for CA19-9 DOR. The pooled sensitivity and specificity for Tu-M2- PK were 60% and 95%, respectively. The few present studies on this subject did not allow calculation of DOR of Tu-M2-PK and CA19-9 in combination, but it was concluded that the combination of the 2 tumor markers will identify more patients with pancreatic cancer than when used separately. The upper reference limit for Tu-M2-PK varied from 15 to 28 U/mL, and asymmetry was found in the funnel plot. The authors recommend determination of either Tu-M2-PK alone or in combination with CA19-9 in patients with PDAC and patients suspected to have PDAC. 19 ORIGINAL ARTICLE Pancreas & Volume 39, Number 2, March 2010 www.pancreasjournal.com 243 From the *Department of Medical Gastroenterology Odense University Hos- pital, University of Southern Denmark, Odense; and †Department of Clinical Biochemistry and Immunology, Statens Serum Institut, Copenhagen, Denmark. Received for publication March 3, 2009; accepted August 5, 2009. Reprints: Maiken Thyregod Joergensen, MD, PhD, Department of Medical Gastroenterology, Odense University Hospital, Sdr. Boulevard 29, 5000 Odense C, Denmark (e-mail: maiken.t.joergensen@ ouh.regionsyddanmark.dk). This study was supported by grants from A.J. Andersen and Wife Foundation, the Illum Foundation, and the Foundation of 17.12.1981. Copyright * 2010 by Lippincott Williams & Wilkins Copyright @ 20 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 10