Correspondence

Pancreatitis and the Risk of Pancreatic Cancer

N Engl J Med 1993; 329:1502-1503November 11, 1993

Article

To the Editor:

The 14- to 16-fold increase in the risk of pancreatic cancer among patients with chronic pancreatitis, noted by Lowenfels et al. (May 20 issue),1 is striking. However, if the true risk were increased this much, one must wonder why earlier studies did not uncover it.

From the Methods section, it is not clear how the cohort was assembled. The authors mention that recruitment of patients with pancreatitis began in 1946. Does this mean that the protocol and study design were in place in 1946? Or does it mean that the first diagnosis of chronic pancreatitis was retrospectively ascertained by a review of medical records back to 1946? A separate presentation of the Swedish component of the study2 indicates that the latter approach was used. The cohort thus probably consists of two types of patients: those with incident cases of pancreatitis, who had their first hospitalization during the study's active recruitment period (which is undefined in the paper), and those with prevalent cases of pancreatitis diagnosed before the active recruitment period.

Selection bias is a concern for the prevalent cases. Patients could have been included in the study because of recent abdominal pain and a diagnosis of chronic pancreatitis in the distant past. If the symptoms at the time of recruitment were actually the first signs of pancreatic cancer, even the two- or five-year exclusion period would not be enough to prevent inclusion of these patients. Consequently, patients with prevalent pancreatitis and an assumed flare-up who actually had pancreatic cancer would have been more likely to be enrolled in the cohort than patients with “silent” prevalent pancreatitis and no cancer, whose pancreatitis could have been diagnosed as far back as 1946. The latter patients would not have been enrolled in the study because they would not have presented with symptoms during the active recruitment phase. Indeed, if 27 patients with chronic pancreatitis and pancreatic cancer were identified in this retrospective manner over the 43-year period, the reported excess risk would be explained.

The cumulative mortality rate of 27 percent among patients with a history of chronic pancreatitis exceeding two years and a mean duration of disease of 7.4 years (range, 2 to 43 years) suggest that a substantial portion of the patients studied represent prevalent cases. Separate analyses therefore appear to be warranted for prevalent and incident cases of pancreatitis in relation to the occurrence of pancreatic cancer.

Anders Ekbom, M.D.
Uppsala University, S-751 85 Uppsala, Sweden

Joseph K. McLaughlin, Ph.D.
National Cancer Institute, Bethesda, MD 20892

Olof Nyren, M.D.
Uppsala University, S-751 85 Uppsala, Sweden

2 References

1. 1

   Lowenfels AB, Maisonneuve P, Cavallini G, et al. Pancreatitis and the risk of pancreatic cancer. N Engl J Med 1993;328:1433-1437
   Full Text | Web of Science | Medline

2. 2

   Domellof L, Matthiessen P, Andren-Sandberg A. Kronisk pankretit -- epidemiologiska data fran en internationell kohortstudie. Svensk Kirurgi 1993;51:41-42
   

Author/Editor Response

Dr. Lowenfels replies:

To the Editor: Ekbom et al. ask why earlier studies failed to show an increased risk of pancreatic cancer in patients with pancreatitis. No previous study had sufficient power to detail the relation between these two diseases. One previous report suggested an increased risk of pancreatic cancer after alcoholic or nonalcoholic pancreatitis,1 and another noted that the overall risk of pancreatic cancer in subjects with hereditary pancreatitis was about 5 percent,2 which is similar to our findings.

With respect to study design, in 1991 we initiated our historical cohort study with data recorded in several international centers, including data on some prevalent cases of pancreatitis. However, we have reanalyzed the data from the Danish center, which consisted only of patients with pancreatitis from a well-defined area of Copenhagen who had had hospital contact between April 1, 1978, and March 31, 1982. We restricted the analysis to patients with incident cases of pancreatitis first diagnosed during this period who had been followed for two or more years. Four cancers were diagnosed as compared with an expected number of 0.16, yielding a risk ratio (standardized incidence ratio) of 25 (95 percent confidence interval, 6.8 to 64), which is higher than originally reported.

Table 1Table 1Risk of Pancreatic Cancer for Patients with Early and Delayed Diagnosis of Pancreatitis. shows the risk of pancreatic cancer according to the duration of symptoms of pancreatitis. The risk of pancreatic cancer was nearly identical in subjects who may be considered to have had incident cases of pancreatitis because their disease had been diagnosed within one year after the onset of symptoms and subjects in whom pancreatitis had been diagnosed after a longer interval. It seems unlikely that the elevated risk ratios in our study are related to selection bias among prevalent cases.

Finally, our data for the pancreas resemble findings for other gastrointestinal organs, such as the esophagus,3 colon,4 and liver,5 in which an elevated risk of cancer has also been associated with antecedent benign disease.

Albert B. Lowenfels, M.D.
New York Medical College, Valhalla, NY 10595

5 References

1. 1

   Ammann RW, Schueler G. Chronic pancreatitis, pancreatic cancer, alcohol, and smoking. Gastroenterology 1984;87:744-745
   Web of Science | Medline

2. 2

   Gross JB. Hereditary pancreatitis. In: Go VLW, ed. The exocrine pancreas: biology, pathobiology, and diseases. New York: Raven Press, 1986:829-39.
   

3. 3

   Williamson WA, Ellis FH Jr, Gibb SP, et al. Barrett's esophagus: prevalence and incidence of adenocarcinoma. Arch Intern Med 1991;151:2212-2216
   CrossRef | Web of Science | Medline

4. 4

   Ekbom A, Helmick C, Zack M, Adami H-O. Ulcerative colitis and colorectal cancer -- a population-based study. N Engl J Med 1990;323:1228-1233
   Full Text | Web of Science | Medline

5. 5

   Adami HO, Hsing AW, McLaughlin JK, et al. Alcoholism and liver cirrhosis in the etiology of primary liver cancer. Int J Cancer 1992;51:898-902
   CrossRef | Web of Science | Medline

Citing Articles (8)

Citing Articles

1. 1

   Sara H. Olson. (2012) Selected medical conditions and risk of pancreatic cancer. Molecular Carcinogenesis 51:1, 75-97
   CrossRef

2. 2

   M. T. Rosenfeldt, K. M. Ryan. (2011) The multiple roles of autophagy in cancer. Carcinogenesis 32:7, 955-963
   CrossRef

3. 3

   Simone Reuter, Subash C. Gupta, Madan M. Chaturvedi, Bharat B. Aggarwal. (2010) Oxidative stress, inflammation, and cancer: How are they linked?. Free Radical Biology and Medicine 49:11, 1603-1616
   CrossRef

4. 4

   A. J. Schetter, N. H. H. Heegaard, C. C. Harris. (2010) Inflammation and cancer: interweaving microRNA, free radical, cytokine and p53 pathways. Carcinogenesis 31:1, 37-49
   CrossRef

5. 5

   Wiktor Bednarz, Robert Olewinski. (2002) European Journal of Gastroenterology & Hepatology 14:6, 671-677
   CrossRef

6. 6

   George H. Sakorafas, Adelais G. Tsiotou. (1999) Pancreatic cancer in patients with chronic pancreatitis: a challenge from a surgical perspective. Cancer Treatment Reviews 25:4, 207-217
   CrossRef

7. 7

   Britt–Marie Karlson, Anders Ekbom. (1998) Reply. Gastroenterology 114:4, 861
   CrossRef

8. 8

   David Hsiang, Helmut Friess, Markus W. Büchler, Matthias Ebert, John Butler, Murray Korc. (1997) Absence of K-ras mutations in the pancreatic parenchyma of patients with chronic pancreatitis. The American Journal of Surgery 174:3, 242-246
   CrossRef