Gastroenterology
Volume 134, Issue 1 , Pages 344-345, January 2008

Pancreatic Cancer: Cause and Result of Diabetes Mellitus

  • Niels Teich

      Affiliations

    • Corresponding Author InformationAddress requests for reprints to: Niels Teich, MD, University of Leipzig Medical Clinic II, Gastroenterology and Hepatology, Philipp-Rosenthal-Str. 27, D-04103 Leipzig, Germany. fax: (49) 341 9712209.

University of Leipzig, Medical Clinic II, Leipzig, Germany

Article Outline

 

See “Pancreatic cancer-associated diabetes mellitus: prevalence and temporal association with diagnosis of cancer” by Chari ST, Leibson CL, Rabe KG, et al on page 95.

Almost 175 years ago, a clinical case was published that described a patient who presented with symptoms of diabetes mellitus and died 6 months later from pancreatic cancer.1 Since this initial observation, numerous epidemiologic studies describe an association between diabetes mellitus and pancreatic cancer in humans and, occasionally, in other species.2 The first meta-analysis of 20 studies with predominantly type 2 diabetes patients estimated a 2-fold relative risk compared to patients without diabetes.3 An analogue association resulted from a very large meta-analysis, which included 9220 individuals with pancreatic cancer from the mid-1990s: the combined summary odds ratio was 1.82 (95% confidence interval, 1.66–1.89). This validated the premise that diabetes mellitus is a modest risk factor for pancreatic cancer. Furthermore, the latter study concluded that individuals in whom diabetes had been diagnosed within the last 4 years had a 50% greater risk as compared with individuals who had diabetes for at least 5 years.4 A similar magnitude of relative risk has been described for type 1 diabetes mellitus (2.0; 95% confidence interval, 1.37–3.01).5 However, many authors of these original articles and the subsequent meta-analyses excluded new-onset diabetes mellitus, as they considered new-onset diabetes mellitus to be an unlikely etiological factor of pancreatic cancer. One weakness of most association studies between pancreatic cancer and diabetes mellitus has been the diabetes mellitus diagnosis itself. Almost all investigations have involved retrospective designs, included random but not fasting glucose measurements, and assessed diabetes through self-reporting or by considering all patients who were taking an antidiabetic agent as having diabetes—without reporting objective findings for this diagnosis. These methodical approaches, however, do not fulfill the formal criteria for the diabetes mellitus definition as suggested by the American Diabetes Association (ADA)6 and other organizations.

In this issue of Gastroenterology, Chari et al7 present the results of an exhaustive chart review of 736 pancreatic cancer patients and 1875 matched controls with special consideration for the temporal dependence between diabetes mellitus onset and pancreatic cancer diagnosis. The ADA definition of diabetes was employed in the vast majority of patients, and fasting blood glucose levels for the 60 months preceding pancreatic cancer diagnosis were analyzed carefully. The authors report that prevalence of diabetes mellitus was similar in patients and controls 3 years before pancreatic cancer diagnosis. A continuous increase of diabetes mellitus prevalence was observed during the “countdown” to overt cancer symptoms and pancreatic cancer diagnosis. However, diabetes mellitus prevalence remained stable in the controls. The authors conclude that more than 40% of pancreatic cancer patients had diabetes mellitus at the time of cancer diagnosis, as compared with less than 20% of the controls. This means that half of the diabetes mellitus diagnoses in pancreatic cancer patients were new onset.

This study has important strengths: the pancreatic cancer diagnosis was histologically proven in 84% of the patient cohort and was convincingly probable in the remaining patients who were significantly older and refused attempts to obtain tissue diagnosis. Almost all cases and controls had documented full-fasting blood glucose levels of at least 126 mg/dL, and therefore, fulfilled the ADA definition of diabetes mellitus. Despite its retrospective design, this study is of the highest methodical quality that has been published thus far in this field. Consequently, the study has important implications for further research, as it convincingly suggests new-onset diabetes mellitus as an early manifestation of pancreatic cancer. The observation that diabetes mellitus is not a cause but a result of pancreatic cancer has been substantiated mainly by preclinical studies. For example, mice treated with pancreatic cancer cell–conditioned media develop hyperglycemia (Basso 1995).8 In pancreatic cancer patients, glucose tolerance often ameliorates or returns to normal after tumor resection.9

From the data presented in this issue of Gastroenterology, possible directions for future research can be derived: Is pancreatic cancer-associated new-onset diabetes mellitus a different entity from new-onset type 2 diabetes mellitus? Could novel diagnostic screening markers be able to differentiate new-onset diabetes mellitus as an “alarm symptom” of asymptomatic pancreatic cancer in an older subject? Investigations to answer to these questions are hampered by the relatively low incidence of pancreatic cancer in patients with new-onset diabetes mellitus. A recent population-based study reports that only 1% of diabetes subjects aged 50 years and older were diagnosed with pancreatic cancer within 3 years of diabetes mellitus diagnosis.10 This finding is comparable to a less-stringently designed retrospective cohort study from San Francisco that found only 1 additional pancreatic cancer case for every 332 new diabetics.11

Only 1 study has prospectively investigated pancreatic cancer in patients with severe, early decompensated, new-onset diabetes mellitus. In that study, 115 patients aged 50 and older, who were emergently hospitalized for very new-onset insulin-dependent diabetes mellitus shorter than 30 days’ duration, were investigated by abdominal imaging. In this highly selected cohort, 6 patients had pancreatic adenocarcinoma, almost all in an advanced state.12 This observation underscores that pancreatic cancer is usually diagnosed late, which contributes to disastrous survival rates. Even in the setting of completely resected, node-negative pancreatic cancer, the majority of patients die from their pancreatic cancer. A landmark Finnish study came to the pessimistic conclusion that more than two thirds of the reported long-term survivors of assumed pancreatic cancer did not have the disease. In cases with a validated histology, the overall 5-year survival rate of pancreatic cancer patients was only 0.2%; all of these patients had a very early cancer stage.13

To date, there is no practical clinical, radiological, or laboratory criterion that can eliminate the possibility of early pancreatic cancer in new-onset diabetic patients. The early observation that islet amyloid polypeptide (a hormone that reduces insulin sensitivity) is secreted by pancreatic cancer cells did not lead to a valid marker of pancreatic cancer in patients with diabetes.14 Recent attempts to identify early diabetogenic serum changes in pancreatic cancer patients have used proteomic applications. In 1 analysis by matrix-assisted laser desorption ionization time of flight, a group identified a pancreatic cancer-derived 14 amino acid peptide, which significantly reduced glucose uptake in myoblasts.15 Whether this pancreatic cancer-associated diabetogenic factor has a sufficient diagnostic accuracy for pancreatic cancer in new-onset diabetes or prediabetes has yet to be determined.

The worldwide rapid increase of diabetes mellitus incidence demands diabetes prevention programs, which have not been achieved so far.16 Furthermore, an appropriate agenda for early detection of dangerous conditions complicating diabetes, such as pancreatic cancer, is needed. A recent study pointed to a previously unrecognized association between pancreatic cancer and diabetes mellitus. Increasingly, women of childbearing age are overweight or obese and, subsequently, the prevalence of gestational diabetes mellitus is rapidly increasing.17 The first population-based cohort study from Israel showed a greater than 7-fold increased risk of pancreatic cancer in women with gestational diabetes mellitus in at least 1 pregnancy.18

The data available today clearly suggest that diabetes mellitus can be both a long-standing cause of pancreatic cancer and, as shown now, an early manifestation of the disease.

Back to Article Outline

References 

  1. Bright R. Cases and observations connected with disease of the pancreas and duodenum. Med Chir Tr. 1833;18:1–56
  2. Poel WE, Yergganian G. Adenocarcinoma of the pancreas in diabetes-prone Chinese hamsters. Am J Med. 1961;31:861–863
  3. Everhart J, Wright D. Diabetes mellitus as a risk factor for pancreatic cancer (A meta-analysis). JAMA. 1995;273:1605–1609
  4. Huxley R, Ansary-Moghaddam A, Berrington de Gonzalez A, et al. Type-II diabetes and pancreatic cancer: a meta-analysis of 36 studies. Br J Cancer. 2005;92:2076–2083
  5. Stevens RJ, Roddam AW, Beral V. Pancreatic cancer in type 1 and young-onset diabetes: systematic review and meta-analysis. Br J Cancer. 2007;96:507–509
  6. American Diabetes Association. Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 2007;30:S42–S47
  7. Chari ST, Leibson CL, Rabe KG, et al. Pancreatic cancer-associated diabetes mellitus: prevalence and temporal association with diagnosis of cancer. Gastroenterology. 2008;134:95–101
  8. Basso D, Brigato L, Veronesi A, et al. The pancreatic cancer cell line MIA PaCa2 produces one or more factors able to induce hyperglycemia in SCID mice. Anticancer Res. 1995;15:2585–2588
  9. Fogar P, Pasquali C, Basso D, et al. Diabetes mellitus in pancreatic cancer follow-up. Anticancer Res. 1994;14:2827–2830
  10. Chari ST, Leibson CL, Rabe KG, et al. Probability of pancreatic cancer following diabetes: a population-based study. Gastroenterology. 2005;129:504–511
  11. Gupta S, Vittinghoff E, Bertenthal D, et al. New-onset diabetes and pancreatic cancer. Clin Gastroenterol Hepatol. 2006;4:1366–1372
  12. Damiano J, Bordier L, Le Berre JP, et al. Should pancreas imaging be recommended in patients over 50 years when diabetes is discovered because of acute symptoms?. Diabetes Metab. 2004;30:203–207
  13. Carpelan-Holmstrom M, Nordling S, Pukkala E, et al. Does anyone survive pancreatic ductal adenocarcinoma? (A nationwide study re-evaluating the data of the Finnish Cancer Registry). Gut. 2005;54:385–387
  14. Permert J, Larsson J, Westermark GT, et al. Islet amyloid polypeptide in patients with pancreatic cancer and diabetes. N Engl J Med. 1994;330:313
  15. Basso D, Greco E, Fogar P, et al. Pancreatic cancer-derived S-100A8 N-terminal peptide: a diabetes cause?. Clin Chim Acta. 2006;372:120–128
  16. Absetz P, Valve R, Oldenburg B, et al. Type 2 diabetes prevention in the “real world”: one-year results of the GOAL Implementation Trial. Diabetes Care. 2007;30:2465–2470
  17. Dabelea D, Snell-Bergeon JK, Hartsfield CL, et al. Increasing prevalence of gestational diabetes mellitus (Gdiabetes mellitus) over time and by birth cohort: Kaiser Permanente of Colorado Diabetes Mellitus Screening Program. Diabetes Care. 2005;28:579–584
  18. Perrin MC, Terry MB, Kleinhaus K, et al. Gestational diabetes as a risk factor for pancreatic cancer: A prospective cohort study. BMC Med. 2007;5:25

PII: S0016-5085(07)02033-1

doi:10.1053/j.gastro.2007.11.020

Refers to article:

  • Pancreatic Cancer–Associated Diabetes Mellitus: Prevalence and Temporal Association With Diagnosis of Cancer , 29 October 2007

    Suresh T. Chari, Cynthia L. Leibson, Kari G. Rabe, Lawrence J. Timmons, Jeanine Ransom, Mariza de Andrade, Gloria M. Petersen
    Gastroenterology January 2008 (Vol. 134, Issue 1, Pages 95-101)

Gastroenterology
Volume 134, Issue 1 , Pages 344-345, January 2008

Article Tools