INTRODUCTION — Surgical resection is the only potentially curative treatment for exocrine pancreatic cancer, but because of the late presentation of the disease, only 15 to 20 percent of patients are candidates for pancreatectomy. Furthermore, the prognosis of pancreatic cancer is poor even in those with potentially resectable disease. The five-year survival following pancreaticoduodenectomy is only approximately 25 to 30 percent for node-negative tumors and 10 percent for node-positive tumors. (See "Overview of surgery in the treatment of exocrine pancreatic cancer and prognosis".)
The nonfamilial risk factors for carcinoma of the exocrine pancreas will be reviewed here. Familial risk factors and screening for pancreatic cancer in high-risk individuals who have a family history of the disease, as well as the pathology, molecular pathogenesis, clinical manifestations, and diagnosis of this disorder are discussed separately. (See "Familial risk factors for pancreatic cancer and screening of high-risk patients" and "Pathology of exocrine pancreatic neoplasms" and "Molecular pathogenesis of exocrine pancreatic cancer" and "Clinical manifestations, diagnosis, and staging of exocrine pancreatic cancer".)
EPIDEMIOLOGY — In the United States, approximately 62,210 patients are diagnosed with cancer of the exocrine pancreas annually, and almost all are expected to die from the disease [1]. Pancreatic cancer is the fourth leading cause of cancer-related death in the United States among both men and women. The majority of these tumors (85 percent) are adenocarcinomas arising from the ductal epithelium. (See "Pathology of exocrine pancreatic neoplasms".)
Worldwide, pancreatic cancer is the seventh leading cause of cancer deaths in both men and women according to data from the World Health Organization (WHO) GLOBOCAN database and data from the 2017 Global Burden of Disease Study [2]. In general, pancreatic cancer affects more individuals inhabiting the Western/industrialized parts of the world; the highest incidence is reported in high-income North America, high-income Asia Pacific, and Western and Central Europe, while people living in South Asia and eastern and central Sub-Saharan Africa have the lowest reported incidence [2].
The disease is rare before the age of 45, but the incidence rises sharply thereafter. In data from the Global Burden of Disease Study, the number of incident cases peaked at age 65 to 69 for men and at 75 to 79 for women [2].
At least in the United States, incidence rates vary by sex and race [3]. The incidence is greater in males than in females (male-to-female ratio 1.3:1) and in Black Americans than in White Americans (16.7 per 100,000 for Black males compared with 14.8 per 100,000 for White males; the corresponding figures for women are 14.6 versus 11.5 cases per 100,000) [4].
The majority of cases are locoregionally advanced, and only 15 to 20 percent are potentially resectable at presentation, usually (but not always) because of vascular invasion. However, at least in the United States, the proportion of patients diagnosed with stage IA pancreatic cancer appears to be increasing over time. In an analysis of data from the National Cancer Institute Surveillance, Epidemiology, and End Results database, the incidence of stage IA pancreatic cancer increased significantly between 2004 and 2016 (annual percentage change 14.5), and their average age at diagnosis decreased by 3.5 years overall [5]. Perhaps more importantly, the five-year overall survival for stage IA pancreatic cancer nearly doubled from 45 percent in 2004 to 84 percent in 2012. These trends were attributed to improved early diagnosis and detection.
Notably, guidelines from the United States Preventive Services Task Force specifically recommend against screening for pancreatic cancer in asymptomatic adults not known to be at high risk because of family history or inherited genetic syndromes [6]. Screening for early detection in high-risk individuals who have familial syndromes predisposing them to pancreatic cancer is discussed elsewhere. (See "Familial risk factors for pancreatic cancer and screening of high-risk patients".)
RISK FACTORS — In the 2017 Global Burden of Disease Study, worldwide, deaths related to pancreatic cancer were primarily attributable to smoking (21 percent), high fasting plasma glucose (8.9 percent), and high body mass index (BMI; 6.2 percent) [2].
Environmental risk factors — Several environmental risk factors have been implicated in the risk of pancreatic cancer, including tobacco use, diet, alcohol consumption, and high caloric intake.
Cigarette smoking — Cigarette smoking increases the risk for pancreatic cancer [7,8], and the estimated population-attributable fraction of pancreatic cancer deaths to tobacco smoking is 11 to 32 percent [9]. In multiple cohort and case-control studies, the relative risk for developing pancreatic cancer among smokers was at least 1.5 [10-17]. The risk increases with the amount of cigarettes consumed [13,14,17], and it may be particularly high in heavy smokers who also have homozygous deletions of the gene for the carcinogen-metabolizing enzyme glutathione S-transferase theta 1 (GSTT1) [16].
Excess risk decreases with smoking cessation. In a large prospective study, the relative risk of pancreatic cancer among current smokers was 2.5 [18]. The risk fell by 48 percent by two years after discontinuing smoking and leveled off 10 to 15 years after stopping, eventually falling to the level of nonsmokers. It has been estimated that cessation of smoking could eliminate approximately 25 percent of pancreatic cancer deaths in the United States [15,18].
Obesity and physical inactivity — Several studies suggest a link between high body mass, lack of physical activity, and pancreatic cancer risk [19-28]. These relationships can be illustrated by the result of a representative cohort study that was based on follow-up data from the Health Professionals Follow-up Study and the Nurses' Health Study [20]. A BMI of at least 30 kg/m2 was associated with a significantly increased risk of pancreatic cancer compared with a BMI of less than 23 kg/m2 (relative risk 1.72, 95% CI 1.19-2.48). Height was also associated with an increased risk (relative risk 1.81, 95% CI 1.31-2.52). An inverse relationship was also observed for moderate physical activity when comparing the highest versus the lowest categories (relative risk 0.45, 95% CI 0.29-0.70), particularly among those with a BMI of at least 25 kg/m2.
Others have suggested that overweight and obese individuals develop pancreatic cancer at a younger age than do patients with a normal weight, and that they also have lower rates and shorter duration of survival once pancreatic cancer is diagnosed [23,29].
Diet — Studies evaluating the relationship between diet and pancreatic cancer are inconclusive, as evidenced by the following observations:
●A "Western" dietary pattern (high intake of saturated fat and/or meat, particularly smoked or processed meats) has been linked to the development of pancreatic cancer in many [14,30-32], but not all [33-35], studies.
●Several case-control studies [36-38] report a protective effect from the consumption of fresh fruits and vegetables, but prospective studies have not observed such an association [39].
●A case-control study from the National Institutes of Health (NIH)-American Association of Retired Persons (AARP) cohort found an inverse association between consumption of a "healthy diet" and risk of pancreatic cancer [40]. Compared with individuals meeting the fewest guidelines for a healthy diet according to the 2005 Dietary Guidelines for Americans (which included higher intake of fruits, vegetables, whole grains, milk, meat and beans, and oils [nonhydrogenated vegetable oils and those found in fish, nuts, and seeds] and lower intake of saturated fat, sodium, and calories from solid fat, alcohol, and added sugar), those who met the most dietary guidelines had a significantly reduced risk for pancreatic cancer (hazard ratio [HR] 0.85, 95% CI 0.74-0.97). Among men, the protective effect of a healthy diet was seen in men who were overweight/obese (BMI ≥25 kg/m2) but not those with a normal weight.
●Lower serum levels of lycopene (a carotenoid present in fruits) and selenium have been found in subjects who subsequently developed pancreatic cancer [41]. However, whether dietary or supplemental intake of these nutrients is associated with a reduced risk of pancreatic cancer is unclear [42,43].
Coffee and alcohol consumption — Data regarding the impact of coffee and alcohol ingestion on the risk of pancreatic cancer have been conflicting. Two pooled analyses suggest that, if there is an effect of alcohol consumption, it is small and limited to heavy drinkers [44,45]. The relationship between alcohol use and pancreatic cancer is confounded by cigarette smoking.
Similarly, the relationship between coffee intake and pancreatic cancer risk has been controversial, with some studies indicating an elevated risk with higher levels of consumption, and others indicating no relationship. A meta-analysis of 37 case-control and 17 cohort studies concluded that the pooled relative risk of pancreatic cancer for highest versus lowest intake of coffee was 1.13 (95% CI 0.99-1.29), and it remained not statistically significant when the analysis was limited to those studies that adjusted for smoking [46].
Aspirin and NSAID use — Laboratory data suggest a possible inhibitory effect of aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) on pancreatic tumorigenesis. However, the available epidemiologic data in humans are conflicting [47-55]. In fact, preliminary data from the Nurses' Health study, in which 88,378 women without cancer were followed for up to 18 years, raise the possibility that extended regular use of aspirin might increase the risk of pancreatic cancer [52]. Compared with nonusers, participants who ingested 14 or more aspirin tablets weekly (but not fewer) for four or more years had a significantly increased relative risk of pancreatic cancer (relative risk 1.86). The findings did not change appreciably when obesity and smoking were taken into account.
Despite these data, studies of other large cohorts have not found any link between aspirin use and higher pancreatic cancer risk, even among individuals using aspirin ≥30 times monthly for over 20 years [53].
Helicobacter pylori — A weak but statistically significant association between Helicobacter pylori infection and pancreatic cancer has been reported. (See "Association between Helicobacter pylori infection and gastrointestinal malignancy", section on 'Pancreatic cancer'.)
HBV and HCV infection — Several reports indicate an association between infection with hepatitis B virus (HBV) and hepatitis C virus (HCV), and pancreatic cancer [56-58]. The magnitude of risk is less than that for hepatocellular cancer. (See "Epidemiology and risk factors for hepatocellular carcinoma", section on 'Hepatitis B virus' and "Epidemiology and risk factors for hepatocellular carcinoma", section on 'Hepatitis C virus'.)
Diabetes mellitus, glucose metabolism, and insulin resistance — Numerous epidemiologic studies describe an association between diabetes mellitus and pancreatic cancer [59-67]. In a meta-analysis of 88 studies (50 cohort and 30 case-control studies, predominantly of patients with type 2 diabetes), the pooled relative risk for pancreatic cancer in diabetics compared with patients without diabetes was 2.08 (95% CI 1.87-2.32) [67]. Patients with diabetes may also have unfavorable pathologic features and poorer long-term outcomes compared with those without diabetes [68].
It has been suggested that diabetes may be a consequence rather than a cause of pancreatic cancer [69-71]. As an example, one study compared 512 patients with newly diagnosed pancreatic cancer with 933 controls of a similar age [71]. Compared with the controls, diabetes was more prevalent in pancreatic cancer cases (47 versus 7 percent) and was more likely to have been diagnosed in the preceding two years (74 versus 53 percent). After pancreaticoduodenectomy, diabetes resolved in 17 of 30 patients (57 percent) with new-onset diabetes, while its prevalence was unchanged in the 11 patients who had longstanding diabetes. One intriguing possibility derived from in vitro work is that pancreatic cancers might induce paraneoplastic beta cell dysfunction and inhibition of insulin secretion by shedding exosomes containing adrenomedullin (a vasodilator peptide hormone that regulates insulin balance and may participate in the development of diabetes [72]) into the systemic circulation [73]. An observational study is underway in Italy examining the utility of adrenomedullin measurement to see if it may identify patients with adult-onset diabetes who are at risk for pancreatic cancer.
However, increasing data support the view that abnormal glucose metabolism, insulin resistance, and hyperinsulinemia are etiologic factors for pancreatic cancer rather than the result of a subclinical cancer [19,74-78]. As examples:
●In a prospective cohort study, employees of 84 Chicago-area businesses were screened for postload plasma glucose levels and followed for an average of 25 years [74]. Compared with controls with a postload plasma glucose of 119 mg/dL (6.6 mmol/L) or less, the relative risk of mortality from pancreatic cancer was 1.65, 1.6, and 2.15 for those with levels of 120 to 159 mg/dL (6.7 to 8.8 mmol/L), 160 to 199 mg/dL (8.9 to 11.1 mmol/L), and ≥200 mg/dL (11.1 mmol/L), respectively, even after adjusting for age, race, cigarette smoking, and BMI.
●An association between prediagnosis serum levels of glucose and insulin, insulin resistance, and pancreatic cancer risk was suggested in a case-cohort study within the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study, a primary prevention trial enrolling 29,133 male Finnish smokers aged 50 to 69 [75]. Fasting serum concentrations of glucose and insulin, and levels of insulin resistance were assessed. The interval between serum collection and follow-up was up to 16.7 years. The study included 169 incident cases of pancreatic cancer that were diagnosed after the fifth year of follow-up, and 400 randomly selected controls.
After adjustment for age, years of smoking, and BMI, higher prediagnosis serum concentrations of glucose and insulin, as well as insulin resistance were significantly correlated with the risk of pancreatic cancer. The positive associations were stronger among the cases that occurred more than 10 years after baseline (highest versus lowest quartile for glucose, HR 2.16, 95% CI 1.05-4.42; for insulin, HR 2.90, 95% CI 1.22-6.92; for insulin resistance, HR 2.71, 95% CI 1.19-6.18). The prospective study design, with a minimum five-year follow-up prior to the detection of incident pancreatic cancer, minimizes the possibility that the identified insulin and glucose abnormalities resulted from "subclinical" pancreatic cancer.
●On the other hand, an analysis derived from the Nurses' Health Study and the Health Professionals Follow-up Study examining the temporal pattern of cancer risk after a first diagnosis of type 2 diabetes mellitus determined that risk for pancreatic cancer was elevated within the first eight years after diagnosis, but decreased thereafter [78]. A central role for hyperinsulinemia rather than hyperglycemia was suggested by analysis of a subset of subjects in whom blood was available, which showed that levels of C-peptide (a marker of endogenous insulin secretion) fell beyond eight years after diagnosis, while hemoglobin A1C (a marker of glucose levels over the preceding three months) was relatively stable out to 15 years postdiagnosis.
The mechanism underlying this association is unclear. However, at least some data suggest that the increased risk of pancreatic cancer in patients with metabolic diseases, such as type 2 diabetes mellitus and other states of insulin resistance, as well as obesity may be mediated by reduced levels of plasma adiponectin, a fat-derived hormone that has insulin-sensitizing and anti-inflammatory properties [79].
The data linking type 2 diabetes with an elevated risk of pancreatic cancer have led some to suggest that the new onset of diabetes in a thin older adult should prompt consideration of screening for early diagnosis of a potentially resectable pancreatic cancer. At least three studies have addressed the utility of computed tomography (CT) screening for early detection of pancreatic cancer in adults with new-onset diabetes. Two uncovered mainly unresectable tumors, but they selected patients for screening based on the presence of cancer-related symptoms. A third series from the Mayo Clinic suggested that CT scans done at the time of newly diagnosed diabetes in otherwise asymptomatic patients were more likely to show potentially resectable tumors than scans performed six months later [80]. Whether higher resectability rates translated into higher cure rates was not addressed.
CT screening of all older subjects with new-onset diabetes in order to discover a small number of pancreatic cancers is not feasible. Identification of those features that differentiate pancreatic cancer-associated diabetes from other cases of new-onset diabetes would help direct efforts to the subset of individuals who would most benefit from screening CT, but these factors have not yet been established. At least some data suggest that the risk of pancreatic cancer is especially elevated in older adults with new-onset diabetes and a previously healthy weight with otherwise unexplained unintentional weight loss [81], but whether these patients have a high enough risk of pancreatic cancer to justify screening is not established.
Thus, screening for pancreatic cancer is not warranted in older, otherwise asymptomatic adults with new-onset atypical diabetes. This recommendation is consistent with guidelines from the United States Preventive Services Task Force, which specifically recommend against screening for pancreatic cancer in asymptomatic adults not known to be at high risk because of family history or inherited genetic syndromes, including those with preexisting or new-onset diabetes [6]. Screening for high-risk individuals who have familial syndromes predisposing them to pancreatic cancer is discussed elsewhere. (See "Familial risk factors for pancreatic cancer and screening of high-risk patients", section on 'Pancreatic cancer screening'.)
Hereditary risk factors
Family history and genetic predisposition syndromes — Pancreatic cancer aggregates in some families; approximately 5 to 10 percent of individuals with pancreatic cancer have a family history of the disease [82,83]. There are two broad categories of hereditary risk for pancreatic cancer: defined syndromes in which patients are at risk for a pancreatic cancer (and frequently for other malignancies), and familial pancreatic cancer, for which a specific molecular basis has not been identified. These syndromes and the approximate risk for pancreatic cancer are outlined in the table (table 1). A separate topic is available that discusses these syndromes and screening for pancreatic cancer in high-risk individuals. (See "Familial risk factors for pancreatic cancer and screening of high-risk patients".)
At least some data suggest that germline mutations in pancreatic cancer susceptibility genes are not rare, even in the absence of a significant family history of pancreatic cancer [84-86]. In the largest analysis, of 854 patients undergoing pancreatic cancer resection at a single institution over a 15-year period, 33 were identified as having a potentially deleterious germline mutation, 31 of which affected a known pancreatic cancer susceptibility gene (including BRCA, ATM, PALB2, CDKN2A, and MLH1) [84].
Routine genetic testing of patients with newly diagnosed, apparently sporadic pancreatic cancer may yield some clinical benefits (eg, identifying mutation carriers for whom specific forms of enhanced cancer screening are recommended to reduce the risk of associated cancers, identifying family members of the index case who might benefit from screening for the cancer-predisposing mutation, identifying cancer susceptibility mutations that might be therapeutically targetable). As an example, the POLO trial demonstrated benefit from maintenance therapy with the poly(ADP-ribose) polymerase (PARP) inhibitor olaparib in patients with advanced BRCA-mutated pancreatic cancer. (See "Initial systemic chemotherapy for metastatic exocrine pancreatic cancer", section on 'Patients with homologous recombination repair deficiency'.)
A year 2019 Provisional Clinical Opinion from the American Society of Clinical Oncology (ASCO) states that germline genetic testing may be offered to patients with pancreatic cancer and an unremarkable family history if an informative result could directly benefit the patient or his or her family members [87]. This practice is also supported by updated year 2019 consensus-based guidelines from the National Comprehensive Cancer Network (NCCN) [88]. (See "Familial risk factors for pancreatic cancer and screening of high-risk patients", section on 'Genetic predisposition syndromes'.)
Other inherited risk factors
ABO blood group — ABO blood type is an inherited characteristic that has been associated with the risk of several gastrointestinal malignancies, including pancreatic cancer [89-92]. Blood type is determined by the presence or absence of human blood group antigens, glycoproteins that are expressed on the surface of red blood cells and several other types of cells, including pancreatic cancer cells.
The relationship between blood type and pancreatic cancer risk was examined in two large independent prospective cohorts (the Nurses' Health Study and the Health Professionals Follow-up Study) [89]. Compared with blood group O, individuals with non-O blood group (type A, AB, or B) were significantly more likely to develop pancreatic cancer (adjusted HR for incident pancreatic cancer 1.32, 1.51, and 1.72, respectively), and the association was nearly identical in the two cohorts. Overall, 17 percent of the 316 pancreatic cancer cases were attributable to inheriting a non-O blood type.
These findings are supported by the results of a genome-wide association study that identified variants in the ABO locus that were associated with susceptibility to pancreatic cancer [93].
While these data suggest that ABO blood group status may represent a common inherited susceptibility for pancreatic cancer, additional studies are needed to confirm these findings and establish the potential mechanisms by which ABO blood group influences the risk of pancreatic cancer. An interaction between non-O blood type and colonization with H. pylori has been suggested [91]. (See 'Helicobacter pylori' above.)
Cystic fibrosis — Patients with cystic fibrosis (most of whom have inherited the predisposing gene in an autosomal recessive pattern) have an increased risk of pancreatic cancer. In a meta-analysis, the standardized incidence ratio was 6.18, 95% CI 1.31-29.27, and it was two- to fivefold higher in patients who had undergone lung transplantation [94]. (See "Cystic fibrosis: Overview of gastrointestinal disease", section on 'Gastrointestinal cancer'.)
Nonhereditary chronic pancreatitis — Chronic inflammation of the pancreas is a risk factor for pancreatic cancer [95-98]. In a report from the International Pancreatitis Study Group, for example, 2015 patients with chronic pancreatitis were followed for a mean of 7.4 years, during which a total of 56 pancreatic cancers were identified [95]. The expected number of cases of cancer calculated from country-specific incidence data and adjusted for age and sex was 2.13, yielding a standardized incidence ratio (the ratio of observed to expected cases) of 26.3. The cumulative risk reached 1.8 percent at 10 years and 4 percent at 20 years, independent of the type of pancreatitis. A less pronounced relationship has been noted by others [96]. In any event, the population-attributable risk has been estimated to be 1.3 percent, suggesting that a relatively small proportion of pancreatic cancers might be avoided if pancreatitis could be prevented [97]. (See "Chronic pancreatitis: Clinical manifestations and diagnosis in adults".)
The signaling mechanisms that underlie the transition from chronic pancreatitis to invasive cancer are largely undefined, but they are beginning to be elucidated [99].
Pancreatic cysts — Patients with intraductal papillary mucinous neoplasm of the pancreas (IPMN), which is the most common type of neoplastic pancreatic cyst, are at risk for malignant degeneration and are commonly managed with surveillance. When an IPMN develops invasive malignancy, it is usually referred to as IPMN-associated adenocarcinoma. These patients, however, are also at risk of developing conventional pancreatic cancer (concurrent or distinct ductal adenocarcinoma), which arises away from the cyst, suggesting the existence of a pancreatic field defect. This phenomenon has been described in 2 to 9 percent of patients who are being followed for IPMN [100,101]. An in-depth discussion of IPMNs, including a recommended surveillance strategy, is discussed in detail elsewhere. (See "Intraductal papillary mucinous neoplasm of the pancreas (IPMN): Pathophysiology and clinical manifestations" and "Intraductal papillary mucinous neoplasm of the pancreas (IPMN): Evaluation and management".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Pancreatic cancer".)
SUMMARY — The major risk factors for pancreatic cancer are the following:
●Cigarette smoking. (See 'Cigarette smoking' above.)
●High body mass and lack of physical activity. (See 'Obesity and physical inactivity' above.)
●Numerous epidemiologic studies describe an association between diabetes mellitus and pancreatic cancer, although it is not clear whether abnormal glucose metabolism is a consequence or cause of pancreatic cancer. (See 'Diabetes mellitus, glucose metabolism, and insulin resistance' above.)
●Hereditary risk factors, such as hereditary pancreatitis, other highly penetrant conditions caused by germline mutations in known cancer-causing genes, and familial pancreatic cancer, for which a specific genetic abnormality has not yet been identified (table 1). (See "Familial risk factors for pancreatic cancer and screening of high-risk patients".)
●Nonhereditary chronic pancreatitis. (See 'Nonhereditary chronic pancreatitis' above.)
●Pancreatic cysts. (See 'Pancreatic cysts' above.)