Pancreatic Cancer: Epidemiology, Genetics, and Approaches to Screening
Pancreatic Cancer: Epidemiology, Genetics, and Approaches to Screening
ABSTRACT: Pancreatic cancer is a leading cause of cancer-related mortality. Treatment has limited efficacy, and 5-year survival rates remain less than 5%. Insights from epidemiology and discoveries in molecular genetics have laid the groundwork for a rational screening strategy for high-risk individuals. High-risk populations include those in their 6th to 8th decades of life, those with a family history of pancreatic cancer, and those with a personal history of tobacco smoking. Roughly 10% of cases are due to an inherited genetic susceptibility. Several familial syndromes with known genetic defects have been implicated, but the majority of familial cases result from as yet undefined genes. Acquired mutations have been identified in the oncogenes K-ras and HER2/neu, and in the tumor suppressor genes p16, p53, SMAD4, and BRCA2. No standard for screening or prevention exists, but strategies employing endoscopic, radiologic, and molecular methods to screen high-risk individuals are under investigation. [ONCOLOGY 16:1615-1638, 2002]
Pancreatic cancer is typically diagnosed at an advanced stage. Surgery is the only potentially curative option, but fewer than 20% of patients are candidates for curative resection at presentation. After resection, the 5-year survival rate is 35% to 40% for patients with small, node-negative tumors, but 12% to 15% overall. Among patients with unresectable disease, the 5-year survival rate is generally below 5%, even with optimal therapy.
Screening for early or preinvasive cancer in the pancreas has remained an elusive goal. Extensive epidemiologic studies have allowed us to better understand who is at risk for the disease. Advances in molecular genetics over the past 15 years have contributed substantially to characterizing the genetics of pancreatic cancer. These data provide opportunities to identify high-risk individuals with greater fidelity and design more sophisticated diagnostic and therapeutic approaches, with the hope of making earlier diagnoses and providing more effective treatment (Table 1).
In the United States, pancreatic cancer accounts for only 2% of all cancer diagnoses, with an incidence of just over 30,000 cases per year. It is responsible for up to 5% of cancer deaths, making it the fourth leading cause of cancer-related mortality among US men and women. The yearly incidence and mortality are roughly equivalent. From 1930 to 1970, incidence increased approximately twofold, to 10 cases per 100,000 men and 7.2 cases per 100,000 women, but has since stabilized and decreased modestly. This favorable trend has been observed primarily among white men, whereas rates among white women and black men and women have risen slightly.
As with cancer in general, the incidence and mortality rates are slightly higher among African-Americans, with poorer survival for each diagnostic stage. Black American men have the highest rate of pancreatic cancer in the world, whereas the rate in African men is somewhat low.
In 1990, approximately 171,500 cases of pancreatic cancers were diagnosed worldwide, accounting for approximately 168,000 deaths. Although it ranks 13th among malignancies in incidence, it is the 9th most common cause of cancer-related death, with a 98% mortality ratio. Other gastrointestinal cancers occur in about a 2:1 male-to-female ratio, but for pancreatic cancer, the ratio is nearly equally divided between the sexes.
Autopsy studies have consistently demonstrated that pancreatic cancers compose a significant proportion—as high as 25% to 40% in most studies—of carcinomas of unknown primary.[5-7] Unknown primaries account for about 2% of all cancer diagnoses, of which there are 8 million worldwide. The annual incidence of carcinomas of unknown primary may, therefore, be close to 200,000, and an additional 50,000 to 80,000 pancreatic cancers may be unaccounted for in compiled epidemiologic data.
Whereas several gastrointestinal cancers, including gastric, hepatocellular, esophageal, and colorectal, show striking geographic variation, that of pancreatic cancer is more modest but still noteworthy. The incidence is highest in developed nations—namely the United States, Canada, Scandinavia, Western Europe, and Australia—and less common in developing areas, such as Africa, South America, Mexico, the Caribbean, the Middle East, and Asia. A geographic variation that correlated positively with latitude worldwide has been noted. Within developed countries, the male-to-female disparity is more pronounced, with incidence rates ranging between 6 and 10 cases per 100,000 for men, and 4 to 6.5 cases for women.
• Age and Gender—Age is an extremely important determinant of risk. Pancreatic cancer rarely occurs before the age of 50 years, and is seen most frequently in the 7th and 8th decades, regardless of etiology (sporadic, familial, or smoking-related).
In the United States, males have a relative risk of 1.3 compared with females. A family history of pancreatic cancer significantly increases an individual’s risk for the disease, particularly in the setting of a familial syndrome.
• Environment—Environmental factors also appear to play a significant role. Tobacco smoking has been the most consistently demonstrated risk factor and is implicated as a cause in roughly 30% of all cases.[11-13] The relative risk depends on the number of cigarettes smoked and ranges from 1.5 in light smokers to 10 for two-pack-per-day smokers. Rivenson et al induced pancreatic carcinomas in rats by injecting them with various nitrosamines (tobacco-specific chemical carcinogens), which are probably culpable in humans, as well. Risk decreases with time following smoking cessation.
• Race—Incidence among African-Americans is about 30% higher than among white Americans; therefore, race has been implicated as a risk factor. This trend, however, may reflect disparities in wealth, education, and access to health care rather than an inherent tendency toward pancreatic cancer, as incidence among black Africans is low.
• Diabetes—Various comorbidities have been purported to pose an increased risk of pancreatic cancer, with pancreatic diseases being implicated in particular. A clear association with type 2 diabetes has been recognized for decades; however, the cause-and-effect relationship has not been fully elucidated. Among studies that have examined whether rates of pancreatic cancer are higher in patients with long-standing vs recent-onset diabetes, some have demonstrated a long-term risk[18,19] and others, a short-term risk, with recent data supporting both theories.[21,22] Long-standing diabetics who develop pancreatic cancer frequently experience an exacerbation of their disease in the period preceding their diagnosis.
Diabetes does not just result from a destructive mechanical effect of pancreatic cancer, as illustrated by the fact that although most cancers occur in the head (70%) or body (20%) of the pancreas, most of the insulin-producing beta cells are found in the tail. A physiologic basis for the insulin-resistant state was suggested by Permert et al, who demonstrated abnormalities in islet hormone levels in response to fasting and hyperglycemic states in diabetics with pancreatic and nonpancreatic cancers, as well as in healthy controls. Interestingly, in this study, responses of C-peptide and islet amyloid polypeptide to glucagon normalized after subtotal pancreatectomy for pancreatic cancer.
• Pancreatitis—Chronic pancreatitis is also associated with pancreatic cancer, but again, whether one is a consequence of the other or if both result from a common cause has not been determined. Additionally, overlapping signs and symptoms as well as common risk factors may confound the diagnosis. Some studies have observed an 18- to 28-fold increase in incidence compared to controls among patients with pancreatitis,[24,25] while others have revealed more moderate associations.[26,27] A clear risk has been established in the case of hereditary pancreatitis, resulting from mutations in the trypsinogen gene, an autosomal dominant trait.
• Surgery—Surgical procedures including partial gastrectomy and cholecystectomy have been evaluated as risk factors. Several studies have shown a two- to sevenfold increased risk, which manifests 1 or 2 decades following surgery, usually as peptic ulcer disease.[28-30] Theories to explain this connection include hypergastrinemia, elevated levels of cholecystokinin, and decreased gastric digestion of carcinogens. Other studies have not demonstrated such a clear risk and suggest that smoking may be a confounder.
• Diet—Several large human studies have demonstrated positive associations with meat consumption and carbohydrate intake and a protective effect for dietary fiber and consumption of fruits and vegetables.[32-34] Coffee drinking is no longer considered a risk factor; in fact, recent studies suggest that it may be protective.
• Occupation—Occupational risks, usually involving exposure to industrial chemicals, have also been identified as minor contributors to the development of pancreatic cancer. An elevated risk has been identified among coke plant workers, chemists, and construction workers. Exposure to pesticides and ionizing radiation has also been implicated.
• Height and Weight—Two US cohort studies identified obesity and height as independent risk factors. Individuals with a body mass index of at least 30 kg/m² had a relative risk of 1.7 compared to those with a body mass index less than 23 kg/m². Height was associated with a relative risk of 1.81 for the highest vs lowest categories.