Abstract
Background
The relationship between diabetes and pancreatic cancer has been established by more than several decades of research. However, serum levels of CEA and CA 19-9 in diabetic pancreatic cancer has not been shown.
Methods
Preoperative serum levels of CEA and CA 19-9 and clinicopathological characteristics were retrospectively analyzed in 79 with or 229 without diabetes in pancreatic ductal adenocarcinoma (PDA) patients.
Results
Of the 308 PDA patients enrolled, 79 (25.6%) patients had diabetes. The percentage of new-onset diabetes (i.e. <24 months in duration) was 57% (45/79) in PDA patients coupled with diabetes. Among diabetic PDA patients, mean total bilirubin and fasting blood glucose significantly increased in comparison with control groups (8.54 ± 14.88 vs. 4.16 ± 6.12; 170.22 ± 106.96 vs. 95.84 ± 15.76; P < 0.05). No significant differences were observed in mean levels of serum CA 19-9 and CEA levels between two groups. However, when the value of CEA and CA 19-9 was analyzed as a dichotomous variable, elevated CEA (≥5 ng/ml) and CA 19-9 (≥500 U/ml) levels were strongly correlated with the presence of diabetes in PDA patients.
Conclusion
Elevated CEA (≥5 ng/ml) and CA19-9 (≥500 U/ml) levels have an association with diabetic pancreatic cancer. New-onset diabetes combined with higher CA 19-9 and/or CEA might be regarded as a useful tool to screen early pancreatic cancer.
Keywords: Diabetes, Pancreatic cancer, CA19-9, CEA
Introduction
Pancreatic cancer is a highly aggressive malignant tumor, which is currently treated with disappointing results using conventional therapeutics. Pancreatic cancer ranks fourth and fifth as a cause of cancer mortality in the United States and Europe, with a median survival of 6 months (Jemal et al. 2007; Rivera et al. 2009). The dismal prognosis of pancreatic cancer is attributable to its tendency for late presentation, aggressive local invasion, early metastasis, and poor response to chemotherapy (Duffy et al. 2003).
The relationship between diabetes and pancreatic cancer has been established by more than several decades of research. Previous epidemiologic studies of the association between these two diseases have postulated long-standing diabetes is a risk factor for the development of pancreatic cancer (Everhart and Wright 1995). However, recently, a notion that diabetes might be an early accompanying symptom of pancreatic cancer was elucidated by a series of investigations (Chari et al. 2008; Ogawa et al. 2002; Wang et al. 2006). For example, in a retrospective study, the finding uncovered diabetes was more prevalent and predominant among pancreatic cancer patients as compared to controls, indicating diabetes was the manifestation of pancreatic cancer (Chari et al. 2008). Another study demonstrated the prevalence of pancreatic cancer in patients with new-onset diabetes mellitus (within 3 years) (13.9%) was significantly greater than patients with the history of diabetes mellitus for more than 3 years (2.0%) (Ogawa et al. 2002). Furthermore, a population-based case–control study of pancreatic cancer suggested recent-onset diabetes might be an early marker of pancreatic cancer, in which pancreatic cancer were more likely to report a history of diabetes (13%) than controls (9%) (Wang et al. 2006).
In an attempt to investigate the clinical relationships between diabetes and pancreatic cancer, retrospective clinical analysis of total 308 pancreatic ductal adenocarcinoma (PDA) has been done in present study. We found higher carcinoembryonic antigen (CEA) (≥5 ng/ml) and cancer antigen 19-9 (CA 19-9) (≥500 U/ml) levels had a remarkable association with the presence of diabetes in PDA patients.
Materials and methods
Patients and methods
Among the 560 pancreatic tumor patients enlisted at the Second Affiliated Hospital, College of Medicine, Zhejiang University from January 2000 to 2007, 308 consecutive patients with PDA were included in this retrospective clinical study. Informed consent was obtained from all patients. Patients who underwent surgical procedure, including radical operation, palliative operation or exploratory operation were diagnosed by postoperative histology. Patients without surgical procedure were diagnosed by the elevated level of CA 19-9 combined with computed tomography, magnetic resonance imaging and ultrasound. Diagnostic criteria of diabetes were as follows: patients receiving prescription anti-diabetic medications for previously diagnosed diabetes or patients not reporting treatment for diabetes, but the fasting blood glucose (FBG) was ≥126 mg/dl. In this study, we referred to the pathology classification established by the International Union Against Cancer (UICC) TNM staging system [6th edition (2002)] for PDA (Sobin 2002). The tumor size was expressed by their greatest diameter. The preoperative serum indices were measured in early morning samples, which were taken before breakfast 2–3 days before surgery.
Statistical analysis
Comparisons of continuous variables in different subgroups were performed using Student’s t test. Relationships between categorical variables were compared using the Fisher exact test and χ2 test. Values of P < 0.05 were considered statistically significant for all statistical analyses. These tests were performed with the statistical package SPSS release 13.0 for Windows (SPSS Inc., Chicago, IL, USA).
Results
Clinicopathological characteristics of patients with or without diabetes in PDA
Of the 308 PDA patients enrolled, 79 (25.6%) patients had diabetes. However, the percentage of new-onset diabetes (i.e. <24 months in duration) was 57% (45/79) in PDA patients coupled with diabetes. The major complaints in two groups were epigastric pain and jaundice. The incidence of epigastric pain was 60.8 and 64.2% in PDA patients with or without diabetes, respectively. However, the percentage of patients with jaundice was significantly higher in PDA patients with diabetes than those without (38.0 vs. 23.1%; P < 0.05) (Table 1).
Table 1.
Clinicopathological features of PDA cases with or without diabetes
| Without diabetes (n = 229) | With diabetes (n = 79) | P value | |
|---|---|---|---|
| Age (years) | 62.65 ± 11.82 | 65.13 ± 10.12 | 0.1 |
| Gender | 0.675 | ||
| Male | 156 | 52 | |
| Female | 73 | 27 | |
| Complaints | |||
| Epigastric pain | 147 (64.2%) | 48 (60.8%) | 0.591 |
| Jaundice | 53 (23.1%) | 30 (38.0%) | 0.013 |
| Dorsalgia | 24 (10.5%) | 5 (6.3%) | 0.373 |
| Abdominal masses | 8 (3.5%) | 3 (3.8%) | 1 |
| Fever | 6 (2.6%) | 2 (2.5%) | 1 |
| Past medical history | |||
| Cigarette smoking | 94 (41.0%) | 28 (35.4%) | 0.425 |
| Alcohol intake | 76 (33.2%) | 21 (26.6%) | 0.326 |
| Operation | 140 (61.1%) | 52 (65.8%) | |
| Radical operation | 79 (34.5%) | 35 (44.3%) | |
| Non-radical operation | 61 (26.6%) | 17 (21.5%) | |
| Non-operation | 89 (38.9%) | 27 (34.2%) | |
| Metastasis | |||
| Hepatic metastasis | 59 | 19 | 0.881 |
| Pelvic metastasis | 8 | 2 | 1 |
| Omentum metastasis | 24 | 11 | 0.415 |
| Tumor location | 0.360 | ||
| Head | 131 | 42 | |
| Body and tail | 90 | 33 | |
| All pancreas | 8 | 4 | |
| Tumor size (cm) | 5.42 ± 2.80 | 5.62 ± 2.91 | 0.619 |
| Tumor differentiation | 0.917 | ||
| Well | 69 | 28 | |
| Moderately | 22 | 9 | |
| Poorly | 31 | 12 | |
| TNM stage | 0.599 | ||
| Stage I | 12 | 6 | |
| Stage II | 42 | 9 | |
| Stage III | 39 | 16 | |
| Stage IV | 88 | 33 | |
In clinicopathological characteristic aspect, 52 patients with and 140 without diabetes were diagnosed by postoperative histology. The number of patients underwent radical operation with or without diabetes were 35 and 79, respectively, in PDA patients. However, no significant differences were observed in tumor size, location, differentiation, TNM stage and hepatic metastasis between two groups (Table 1).
Serum levels of blood indices in PDA patients with or without diabetes
Among diabetic PDA patients, mean total bilirubin (TBIL) and FBG significantly increased in comparison with control groups (8.54 ± 14.88 vs. 4.16 ± 6.12; 170.22 ± 106.96 vs. 95.84 ± 15.76; P < 0.05). There was no significant difference in alanine aminotransferase (ALT) and hemoglobin (Hb) levels between two groups (Table 2). Data of CA 19-9 and CEA were obtained from 215 and 202 PDA patients, respectively. Similar to ALT and Hb, no significant differences were observed in mean levels of serum CA 19-9 and CEA levels between two groups. However, when the value of CEA and CA 19-9 was analyzed as a dichotomous variable, elevated CEA (≥5 ng/ml) and CA 19-9 (≥500 U/ml) levels were strongly correlated with the presence of diabetes in PDA patients (Tables 3, 4).
Table 2.
Comparison of blood indices in PDA patients with or without diabetes
| Without diabetes (n = 229) | With diabetes (n = 79) | P | |
|---|---|---|---|
| TBIL (mg/dl) | 4.16 ± 6.12 | 8.54 ± 14.88 | 0.013 |
| ALT (U/l) | 70.67 ± 135.87 | 86.97 ± 154.90 | 0.353 |
| FBG (mg/dl) | 95.84 ± 15.76 | 170.22 ± 106.96 | 0.000 |
| Hb (g/dl) | 13.25 ± 11.30 | 12.38 ± 3.57 | 0.505 |
Table 3.
Comparison of serum CEA level in PDA cases with or without diabetes
| Without diabetes (n = 146) | With diabetes (n = 56) | P | |
|---|---|---|---|
| CEA (ng/ml) | 19.84 ± 72.10 | 58.34 ± 163.90 | 0.095 |
| <5 ng/ml | 99 | 24 | 0.002 |
| ≥5 ng/ml | 47 | 32 |
Table 4.
Comparison of serum CA 19-9 level in PDA cases with or without diabetes
| Without diabetes (n = 159) | With diabetes (n = 56) | P | |
|---|---|---|---|
| CA19-9 (U/ml) | 386.17 ± 717.94 | 867.11 ± 2,830.96 | 0.214 |
| <37 U/ml | 47 | 18 | 0.554 |
| ≥37 U/ml | 112 | 38 | |
| <100 U/ml | 61 | 24 | 0.331 |
| ≥100 U/ml | 98 | 32 | |
| <300 U/ml | 90 | 33 | 0.444 |
| ≥300 U/ml | 69 | 23 | |
| <500 U/ml | 123 | 36 | 0.033 |
| ≥500 U/ml | 36 | 20 |
Discussion
Well-known diabetes is a correlative factor with pancreatic cancer. A positive association between diabetes mellitus and pancreatic cancer risk has been reported by numerous epidemiologic studies (Everhart and Wright 1995). However, there has been some concerns that new-onset diabetes might be a concomitant manifestation of occult pancreatic cancer (Noy and Bilezikian 1994; Clark and Mitchell 1961). Indeed, impaired glucose tolerance or diabetes was discovered in a high proportion of pancreatic cancer patients. Removal of the tumor improved glucose metabolism among some patients with pancreatic cancer, providing evidence that altered glucose metabolism might be a stage in the development of the cancer (Permert et al. 1993).
To further probe the relationship between pancreatic cancer and diabetes, a retrospective clinical study of 308 cases was performed. In this study, we found that 25.6% of PDA patients met the diagnostic standard for diabetes, which was in line with the findings (22.8%) reported by Gullo et al. (1994). Another data from a single institution in China demonstrated the incidence of diabetes was 34.63% (169/488) in pancreatic cancer (Kuang et al. 2009). Nevertheless, the reported percentage of diabetes in pancreatic cancer varies greatly according to the methods and criteria used for the diagnosis of diabetes. Using oral glucose tolerance tests and WHO criteria for diabetes, diabetes was diagnosed in nearly two-thirds of pancreatic cancer patients (Cersosimo et al. 1991; Permert et al. 1991, 1993). Moreover, adopting the 1997 ADA criteria, nearly half of the patients with pancreatic cancer at diagnosis had diabetes (Chari et al. 2001). Using the American Diabetes Association Criterion of FBG ≥126 mg/dl or FBG <126 mg/dl, the patients have been reported being on the treatment for diabetes at the time of recruitment to diagnose diabetes and the researcher found that 47% of pancreatic cancer subjects met criteria for diabetes (Pannala et al. 2008). However, Wang et al. (2006) showed (through extensive questions about the presence or absence of diabetes diagnosed by a physician) that only 13% of patients with pancreatic cancer had a history of diabetes.
Of these diabetic PDA patients, the percentage of new-onset diabetes was nearly 57% (45/79) in our study. The present results were consistent with the reference in which diabetes was diagnosed either concomitantly with the cancer (in 40.2%), or within 2 years before the diagnosis of cancer (in 15.9%) (Gullo et al. 1994). In addition, a research from China also showed more patients had diabetes (74.56%) within 2 years of pancreatic cancer diagnosis (Kuang et al. 2009). Thus, combining our findings with the previous studies, diabetes is an early clinical situation of pancreatic cancer and should be given more attentions.
In the current study, some novel results were found which has not been previously reported. The percentage of patients with jaundice and serum TBIL levels was significantly higher in PDA patients with diabetes than those without (40.0 vs. 23.1%; 8.54 ± 14.88 vs. 4.16 ± 6.12; P < 0.05). However, there was no significant difference in tumor size, location, differentiation, TNM stage, hepatic metastasis between these two groups, which was consistent with the former reports (Kuang et al. 2009; Pannala et al. 2008). Similar to the results of Hb and ALT, there were no statistically significant differences in mean serum CEA and CA 19-9 level between PDA patients with or without diabetes. However, when the value of CEA and CA 19-9 was analyzed as a binary variable according to the cut-off value 5 ng/ml and 500 U/ml, respectively, more interestingly, our data provide the first evidence that elevated CEA (≥5 ng/ml) and CA 19-9 (≥500 U/ml) levels correlated strongly with the presence of diabetes in PDA patients.
Slight elevation above the upper normal limit of serum CA 19-9 concentration in diabetic patients has been observed and correlated with blood glucose concentration (Nakamura et al. 1986; Petit et al. 1994). The mean level of serum CA19-9 in diabetic patients was 32.8 ± 25.8 or 46.0 ± 22.4 (U/ml) (Nakamura et al. 1986; Uygur-Bayramicli et al. 2007), far less than pancreatic cancer patients according to the literature (Chang et al. 2007; Guo et al. 2009; Kang et al. 2007). Paradoxically, no correlation between CA 19-9 and biochemical markers of metabolic compensation in diabetes was reported by Banfi et al. (1996). The mechanism for an elevation of serum CA 19-9 in patients with diabetes mellitus may be the increase in sialic acid levels and the accelerated sialylation of carbohydrate chains (Shimojo et al. 1990). However, in pancreatic cancer, the elevation of CA 19-9 is due to the increased production and secretion of this antigen from malignant cells (Fry et al. 2008). In present investigation, higher CA19-9 levels have a higher positive incidence in PDA patients with diabetes. Therefore, a theoretical conclusion could be drawn that higher CA19-9 level in diabetic pancreatic cancer patients might be caused by pancreatic cancer itself and its accompanying disease, new-onset diabetes.
Early detection of pancreatic cancer plays an important role in improving long-term survival. As a possible early manifestation of pancreatic cancer, recognition of new-onset diabetes used as a screening tool for diagnosis of asymptomatic, early stage pancreatic cancer (Pannala et al. 2009). Lean, new-onset diabetes cases, who lack family history of diabetes have been considered as a candidate for pancreatic cancer screening (Jemal et al. 2007). A study performed by Chari found that approximately 1% of patients had diabetes aged ≥50 years would be diagnosed as pancreatic cancer within 3 years of first meeting the criteria for diabetes, suggesting a set of new-onset diabetes as a marker for early pancreatic cancer detection (Chari et al. 2005). Furthermore, it is considered as a promising way to screen early pancreatic cancer patients in diabetes, especially when combined with the elevation of CA 19-9 in diabetes (Uygur-Bayramicli et al. 2007). Based on the data of present study, we speculate that when combined with the elevation of CA 19-9 and/or CEA in new-onset diabetes, it may be useful as a screening marker for undiagnosed pancreatic cancer. A full clinical investigation should be performed to exclude any possible pancreatic neoplasm when patients develop new-onset diabetes, but without positive family history of diabetes, and accompanied with escalated levels of CEA (≥5 ng/ml) and CA 19-9 (≥500 U/ml).
In summary, elevated CEA (≥ 5 ng/ml) and CA19-9 (≥500 U/ml) levels have an association with diabetic pancreatic cancer. We propose that new-onset diabetes combined with higher CA 19-9 and/or CEA might be regarded as a useful tool in screening early pancreatic cancer. However, more studies on a large scale investigation and research were needed before the method was used in clinical field of pancreatic cancer.
Acknowledgments
This work was supported by Grants from the National Natural Science Foundation of China (nos. 30672072 and 30872531), Natural Science Foundation of Zhejiang Province, China (no. Y206247), Foundation of Science and Technology Department of Zhejiang Province, China (no. 2006C23G2010216) and the Ministry of Science and Technology of People’s Republic of China (no. 2007AA02Z476).
Conflict of interest statement
None.
References
- Banfi G, Ardemagni A, Bravi S et al (1996) Are diabetic metabolic compensation and CA19.9 really correlated? Int J Biol Markers 11:207–210 [DOI] [PubMed] [Google Scholar]
- Cersosimo E, Pisters PW, Pesola G et al (1991) Insulin secretion and action in patients with pancreatic cancer. Cancer 67:486–493 [DOI] [PubMed] [Google Scholar]
- Chang MC, Chang YT, Su TC et al (2007) Adiponectin as a potential differential marker to distinguish pancreatic cancer and chronic pancreatitis. Pancreas 35:16–21 [DOI] [PubMed] [Google Scholar]
- Chari ST, Klee GG, Miller LJ et al (2001) Islet amyloid polypeptide is not a satisfactory marker for detecting pancreatic cancer. Gastroenterology 121:640–645 [DOI] [PubMed] [Google Scholar]
- Chari ST, Leibson CL, Rabe KG et al (2005) Probability of pancreatic cancer following diabetes: a population-based study. Gastroenterology 129:504–511 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chari ST, Leibson CL, Rabe KG et al (2008) Pancreatic cancer-associated diabetes mellitus: prevalence and temporal association with diagnosis of cancer. Gastroenterology 134:95–101 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Clark CG, Mitchell PE (1961) Diabetes mellitus and primary carcinoma of the pancreas. Br Med J 2:1259–1262 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Duffy JP, Eibl G, Reber HA et al (2003) Influence of hypoxia and neoangiogenesis on the growth of pancreatic cancer. Mol Cancer 2:12 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Everhart J, Wright D (1995) Diabetes mellitus as a risk factor for pancreatic cancer: a meta-analysis. J Am Med Assoc 273:1605–1609 [PubMed] [Google Scholar]
- Fry LC, Monkemuller K, Malfertheiner P (2008) Molecular markers of pancreatic cancer: development and clinical relevance. Langenbecks Arch Surg 393:883–890 [DOI] [PubMed] [Google Scholar]
- Gullo L, Pezzilli R, Morselli-Labate AM (1994) Diabetes and the risk of pancreatic cancer. N Engl J Med 331:81–84 [DOI] [PubMed] [Google Scholar]
- Guo Q, Zhang B, Dong X et al (2009) Elevated levels of plasma fibrinogen in patients with pancreatic cancer: possible role of a distant metastasis predictor. Pancreas 38:e75–e79 [DOI] [PubMed] [Google Scholar]
- Jemal A, Siegel R, Ward E et al (2007) Cancer statistics, 2007. CA Cancer J Clin 57:43–66 [DOI] [PubMed] [Google Scholar]
- Kang CM, Kim JY, Choi GH et al (2007) The use of adjusted preoperative CA 19-9 to predict the recurrence of resectable pancreatic cancer. J Surg Res 140:31–35 [DOI] [PubMed] [Google Scholar]
- Kuang TT, Jin D, Wang D et al (2009) Clinical epidemiological analysis of the relationship between pancreatic cancer and diabetes mellitus: data from a single institution in China. J Digest Dis 10:26–29 [DOI] [PubMed] [Google Scholar]
- Nakamura N, Aoji O, Yoshikawa T et al (1986) Elevated serum CA19-9 levels in poorly controlled diabetic patients. Jpn J Med 25:278–280 [DOI] [PubMed] [Google Scholar]
- Noy A, Bilezikian JP (1994) Clinical review 63: diabetes and pancreatic cancer: clues to the early diagnosis of pancreatic malignancy. J Clin Endocrinol Metab 79:1223–1231 [DOI] [PubMed] [Google Scholar]
- Ogawa Y, Tanaka M, Inoue K et al (2002) A prospective pancreatographic study of the prevalence of pancreatic carcinoma in patients with diabetes mellitus. Cancer 94:2344–2349 [DOI] [PubMed] [Google Scholar]
- Pannala R, Leirness JB, Bamlet WR et al (2008) Prevalence and clinical profile of pancreatic cancer-associated diabetes mellitus. Gastroenterology 134:981–987 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pannala R, Basu A, Petersen GM et al (2009) New-onset diabetes: a potential clue to the early diagnosis of pancreatic cancer. Lancet Oncol 10:88–95 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Permert J, Larsson J, Ihse I et al (1991) Diagnosis of pancreatic cancer: alteration of glucose metabolism. Int J Pancreatol 9:113–117 [DOI] [PubMed] [Google Scholar]
- Permert J, Ihse I, Jorfeldt L et al (1993a) Improved glucose metabolism after subtotal pancreatectomy for pancreatic cancer. Br J Surg 80:1047–1050 [DOI] [PubMed] [Google Scholar]
- Permert J, Ihse I, Jorfeldt L et al (1993b) Pancreatic cancer is associated with impaired glucose metabolism. Eur J Surg 159:101–107 [PubMed] [Google Scholar]
- Petit JM, Vaillant G, Olsson NO et al (1994) Elevated serum CA19-9 levels in poorly controlled diabetic patients. Relationship with Lewis blood group. Gastroenterol Clin Biol 18:17–20 [PubMed] [Google Scholar]
- Rivera F, Lopez-Tarruella S, Vega-Villegas ME et al (2009) Treatment of advanced pancreatic cancer: from gemcitabine single agent to combinations and targeted therapy. Cancer Treat Rev 35:335–339 [DOI] [PubMed] [Google Scholar]
- Shimojo N, Naka K, Nakajima C et al (1990) The effect of non-insulin-dependent diabetes on serum concentrations of tumor-associated carbohydrate antigens of CA19-9, CA-50, and sialyl SSEA-1 in association with the Lewis blood phenotype. Clin Chim Acta 190:283–289 [DOI] [PubMed] [Google Scholar]
- Sobin LH, Wittekind C (eds) (2002) UICC: TNM classification of malignant tumours. Wiley, London
- Uygur-Bayramicli O, Dabak R, Orbay E et al (2007) Type 2 diabetes mellitus and CA 19-9 levels. World J Gastroenterol 13:5357–5359 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wang F, Gupta S, Holly EA (2006) Diabetes mellitus and pancreatic cancer in a population-based case-control study in the San Francisco Bay Area, California. Cancer Epidemiol Biomarkers Prev 15:1458–1463 [DOI] [PubMed] [Google Scholar]