Pancreatic Cancer: Epidemiology, Genetics, and Approaches to Screening

Drs. Konner and O’Reilly haveprovided a thorough review of current perspectives on pancreatic cancer. The disease is lethal, difficultto diagnose in its early stages, and resistant to standard chemotherapyregimens. Surgery can be curative if performed when the tumor is small (< 2cm), but only a minority of patients have small tumors.

Research Issues

Over the past 100 years, little headway has been made in understanding thenatural history of the disease or in strategies for its early diagnosis andtreatment. Among the reasons for this lack of progress: (1) the pancreas islocated in a remote region of the body, so that early diagnosis is problematic;(2) there have been few good animal models of the disease; (3) pancreatic tissueis difficult to obtain, as most patients do not undergo surgery; (4) the canceris rapidly lethal, and thus, it is difficult to quickly recruit patients forepidemiologic studies and chemotherapy trials; and (5) funding for research inpancreatic cancer has been limited, averaging $300/cancer diagnosis in theUnited States.

Nevertheless, this is an exciting time to be involved in pancreatic cancerresearch, as the developments of the past 5 years and the next 10 years willcombine to provide dramatic breakthroughs in our understanding and treatment ofthe disease. Our insight into its natural history is increasing significantly.Pancreatic dysplasia, or pancreatic intraductal neoplasia, is clearly theprecursor to the disease.[1] Patients who had dysplasia at partialpancreatectomy can develop cancer in the remaining organ 18 months to 10 yearslater.[2] Dysplasia shares the same genetic-mutation signatures as the cancer,and the stepwise histologic progression can be associated with the progressionof molecular defects.

Screening Considerations

Dysplasia appears to occur first in the small and medium-size ducts and, atleast in the familial syndromes, can be multifocal or widespread. This hasimplications for screening protocols: Computed tomography (CT) scans cannotassess changes in the small and medium-size ducts and thus are not helpful forvery early diagnosis or screening.

Endoscopic ultrasound, however, is suitable for detecting early changes inthe pancreatic parenchyma. These changes are similar to those seen in chronicpancreatitis, and in patients at high risk for cancer and no previous history ofpancreatitis, they can be a marker of neoplastic change.[3] Moreover, endoscopicultrasound is cost-effective in patients who have at least a 16% chance ofdeveloping the disease—thus, it would be useful for screening of high-riskpatients with a strong family history.[4] The modality, however, isoperator-dependent, and should be performed in centers with expertise in theearly diagnosis of pancreatic cancer.

What about early detection in the moderate- or low-risk population? A testthat is inexpensive, highly sensitive and specific, and can be performed onsamples that are easily available (such as blood or stool) would be ideal. Thecapability to develop such a test has been limited until the recent developmentof proteomics technology. It is now possible to look for signature proteins inserum samples, and early attempts reveal that the suitability of this type oftest for pancreatic cancer is promising.[5] Although ras mutational analysis isnot sensitive or specific enough for use as a screening marker, our increasingknowledge of the molecular events in pancreatic tumorigenesis may provide moresuitable markers.

Familial Syndromes

The causes of pancreatic cancer are becoming clearer. Genetic predispositionis important in cancer development: Not only do 10% of patients have a familyhistory consistent with an autosomal dominant inherited syndrome, but at least3% to 7% of apparently "sporadic" pancreatic cancer is caused by ahereditary genetic mutation. The recent discovery of a susceptibility gene onchromosome 4q should help identify a new gene that plays a role in the familial,and possibly, sporadic form of the disease.[6] However, linkage analysis willundoubtedly uncover more than one gene that causes the familial form. As in thefamilial adenomatous polyposis/colon cancer paradigm, familial cancer genes may,in turn, shed light on the sporadic form of the disease.

Treatment and Prevention

Although we are developing the capability for early detection and riskassessment, the most challenging areas of research remain treatment andprevention. Ultra-rapid ascertainment methods have been developed to identifyrecently diagnosed patients, ie, prime candidates for treatment trials. Becauseof the quick loss of stamina and health coupled with the 6-month averageprognosis, the need for rapid ascertainment is essential, as is the need forsurrogate markers of tumor response to treatment. Chemoprevention trials arebecoming a distinct possibility as we develop ways to detect the precancerousform of the disease and identify patients at highest risk.

Last, but not least, one of the most promising events in the field ofpancreatic cancer is the establishment of a Pancreatic Cancer Progress ReviewGroup by the National Cancer Institute.[7] The purpose of this group of expertsis to review the state of knowledge in the field and to recommend areas forfuture study. With this review and set of recommendations comes a commitment onthe part of the National Institutes of Health to increase support fordesperately needed studies and to recruit more scientists and physicians toengage in this battle.

References:

1. Hruban RH, Wilentz RE, Goggins M, et al: Pathology of incipient pancreaticcancer. Ann Oncol 10(suppl 4):S9-S11, 1999.

2. Brat DJ, Lillemoe KD, Yeo CJ, et al: Progression of pancreatic intraductalneoplasias to infiltrating adenocarcinoma of the pancreas. Am J Surg Pathol22:163-169, 1998.

3. Rulyak SJ, Brentnall TA: Inherited pancreatic cancer: Surveillance andtreatment strategies for affected families. Pancreatology 1:407-485, 2001.

4. Rulyak S, Kimmey MB, Veenstra DL, et al: Cost-effectiveness of pancreaticcancer screening in familial pancreatic cancer kindreds. Gastrointest Endosc. Inpress.

5. Rosty C, Christa L, Kuzdzal S, et al: Identification of hepatocarcinoma-intestine-pancreas/pancreatitis-associatedprotein I as a biomarker for pancreatic ductal adenocarcinoma by protein biochiptechnology. Cancer Res 62:1868-1875, 2002.

6. Eberle MA, Pfutzer R, Pogue-Geile KL, et al: A new susceptibility locusfor autosomal dominant pancreatic cancer maps to chromosome 4q32-34. Am J HumGenet 70:1044-1048, 2002.

7. Progress Review Groups: Pancreatic cancer. Available at http://prg.nci.nih.gov/pancreatic/default.html.Accessed November 5, 2002.