Keywords

Introduction

Pancreatic cancer is a fatal malignancy associated with rapid progression. One year relative survival rates are less than 30%, and nearly all patients die from the disease within 7 years of surgery [1, 2]. In 2012, it was estimated that 338,000 men and women were diagnosed with pancreatic cancer and 331,000 died of the disease [2]. Although there have been improvements in the diagnosis and prognosis of pancreatic cancer, these changes are minor [3]. Although smoking is the only established nonheritable risk factor for pancreatic cancer, only approximately 30% of the cases can be attributed to smoking [4]. Despite the inconclusive results, obesity, diabetes, alcohol consumption, chronic pancreatitis, diet, physical inactivity, and genetics have also been suggested as risk factors for pancreatic cancer [5, 6]. Given this poorly understood etiology, prevention of this deadly disease remains a challenge.

Etiological studies of pancreatic cancer have encountered methodological obstacles due to the highly aggressive nature of the disease. Disease and exposure misclassifications were major concerns as most studies had to rely upon death certificates or exposure information from next of kin. In addition, the majority of the cohort studies included very few pancreatic cancer cases (less than 50 exposed cases). Despite these challenges, many potential risk factors in occupational settings have been identified and are suspected to be associated with the pathogenesis of pancreatic cancer; approximately 12% of pancreatic cancer cases have been estimated to be attributable to occupational exposures [7, 8].

Occupational Risk Factors of Pancreatic Cancer

Current available studies which investigated occupational factors and the risk of pancreatic cancer have suggested a connection to working in industries such as chemical production, metal manufacturing, printing and paper manufacturing, transport and communication, and textiles. Other professions associated with an increased risk of pancreatic cancer also include solvent-related occupations such as mechanics, leather tanners, and dry cleaners as well as several silica dusts and asbestos-related occupations such as glass manufacturers, potters, and construction workers.

As shown in Table 6.1 (cohort studies) [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64] and Table 6.2 (case-control studies) [65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83], a number of studies investigated the association between specific occupations and industries and risk of pancreatic cancer. Although these studies have yielded inconsistent results, they do suggest that several occupations and industries may be associated with higher risk of pancreatic cancer.

Table 6.1 Cohort studies of occupational exposure and pancreatic cancer
Table 6.2 Case-control studies of occupational exposure and pancreatic cancer

Chemical, Petroleum, and Related Processing Industries

Previous studies have shown an increased risk of pancreatic cancer among men and women who worked in chemical industries. In a mortality study involving 3637 deaths from the American Chemical Society between 1948 and 1967, Li et al. [12] reported a significantly higher proportion of deaths from pancreatic cancer among male chemists aged 20–64 years compared to professional men in general. In standardized mortality ratio (SMR) studies, Hanis et al. [11] reported an increased risk of pancreatic cancer (SMR = 152) among refinery and chemical plant workers. Bond et al. [15] vreported an increased risk of pancreatic cancer (SMR = 233) among chemical workers. Wen et al. [41] reported an elevated risk among oil refinery workers (SMR = 167). Ott et al. [21] found an increased risk of pancreatic cancer associated with chemical manufacturing job. However, none of the results from the above studies were statistically significant. In a mortality study of chlorohydrin production workers, Benson and Teta [25] observed a statistically significantly elevated death due to pancreatic cancer (SMR = 492) in these workers who produced dichloromethane. An occupational mortality study in Washington State also indicated that chemists, chemical engineers, and chemical company workers experienced elevated proportional mortality rate (PMR) for pancreatic cancer [84].

A case-control study using the death certificates of 343 pancreatic cancer cases and 1315 other-cause-of-death cases as controls observed an odds ratio (OR) of 1.4 for people working in the chemical and allied industries [73]. A hospital-based case-control study of 198 pancreatic cancer cases and 209 controls reported a slightly elevated risk (OR = 1.2) among long-term workers in a chemical processing industry [68]. One case-control study of 625 pancreatic cancer cases and 1700 other cancer controls by Partanen et al. [80] reported a slightly reduced risk of pancreatic cancer associated with employment in the chemical and allied industries. In a high pancreatic cancer mortality region of Louisiana, 876 pancreatic cancer death records were matched to controls by age, race, sex, year of death, and parish of residence. The study found a twofold OR for workers in the oil refining industries [75]. A population-based case-control study in Iowa by Zhang et al. [83] observed a statistically significantly increased risk of pancreatic cancer associated with industries of chemical and allied products (OR = 3.5).

It is biologically plausible that an increased risk of pancreatic cancer can be associated with working in chemical industries, since many chemical agents have been suggested as carcinogens and some have been shown to increase the risk of pancreatic cancer. For example, a cohort study in Finland including 2050 male and 1924 female workers exposed to trichloroethylene, tetrachloroethylene, or 1,1,1-trichloroethane between 1967 and 1992 reported an increased risk of pancreatic cancer [44]. In a nested case-control study involving 28 pancreatic cancer deaths and 140 randomly selected controls, Selenskas et al. [82] observed an increased risk of pancreatic cancer associated with processing vinyl and polyethylene. Another nested case-control study by Garabrant et al. [69] involving 28 pancreatic cancer deaths and 112 matched controls reported that exposure to DDT was associated with an increased risk of pancreatic cancer. A population-based case-control study from Finland including 595 cases and 1622 controls reported an elevated risk associated with occupational exposure to solvents (including aliphatic and aromatic hydrocarbons) [77]. Two meta-analyses reported an elevated risk of pancreatic cancer associated with occupational exposure to chlorinated hydrocarbons [7, 85]. One examined 32 specific agents and found that chlorinated hydrocarbon solvents and related compounds had a meta-risk ratio (MRR) of 1.4 (95%CI: 1.0–1.8) [7]. Another one applied hierarchical Bayesian methods using both job title and exposure data; they observed a more than twofold increased risk of pancreatic cancer associated with occupational exposure to chlorinated hydrocarbon compounds (MRR = 2.21, 95%CI: 1.31–3.68) [85]. A recent hospital-based case-control study in Spain further supports a positive association between exposure to chlorinated hydrocarbon solvents and pancreatic cancer, but the association seemed stronger for ductal adenocarcinomas of the pancreas (OR = 4.11, 95%CI: 1.11–15.23), with a significant positive trend in risk with increasing duration of exposure (P for trend = 0.04) [81].

Metal Manufacturing Industries

Elevated risks of pancreatic cancer have been reported to be associated with metal manufacturing industries by a number of studies. Milham [13] reported an increased mortality of pancreatic cancer in aluminum mill workers and in sheet-metal workers. Maruchi et al. [86] reviewed all cases diagnosed in bona fide residents of Olmsted County, Minnesota, from 1935 to 1974 and found an overrepresentation of metal workers among patients with pancreatic cancer. A PMR study in workers from an automobile factory composed of forge, foundry, and engine (machine and assembly) plants reported a statistically significant PMR of pancreatic cancer in the engine plant (PMR = 1.9) [40]. Another PMR study in a bearing plant also reported an increased risk of pancreatic cancer [38]. A death certificate mortality study in Illinois reported an elevated risk of pancreatic cancer among metal workers [74]. Acquavella et al. [24] examined a metal work cohort (n = 3630) and found an excess in the mortality rate of pancreatic cancer. Ji et al. [70] reported an increased risk of pancreatic cancer among Chinese metal workers.

Studies have also investigated specific metals and metallic compounds in relation to pancreatic cancer. A study followed a group of Swedish battery workers exposed to nickel hydroxide and cadmium oxide and found an increased SIR and SMR for pancreatic cancer [60]. Rockette and Arena [22] followed a cohort of 21,829 workers with 5 or more years of employment in 14 aluminum reduction plants and found an elevated mortality for pancreatic cancer. A meta-analysis reported an excess in pancreatic cancer risk for nickel and nickel compounds and chromium and chromium compounds, but not for cadmium and cadmium compounds [7]. Individuals who work in metal manufacturing industries are exposed not only to different metals and metallic compounds but also to silica, lubricants, and chemical fumes [13]. For example, exposure to polycyclic aromatic hydrocarbons (PAHs), a class of chemicals including hundreds of compounds, was found in metal manufacturing industries such as aluminum production industry and iron and steel foundry [87]. While earlier meta-analyses showed a nonsignificant increased risk of pancreatic cancer associated with occupational exposure to PAHs [7, 85], subsequent studies supported a positive association between PAHs and pancreatic risk [88]. It is possible that the elevated risk of pancreatic cancer associated with metal manufacturing industries could be the joint effect of multiple exposures.

Printing and Paper Manufacturing Industries

A PMR study of 1401 commercial pressmen showed a significant PMR of pancreatic cancer among those employed 20 years or longer [43]. Similar results were found in another study of printing pressmen [89]. The Third National Cancer Survey of 7518 incident cancer cases found an elevated risk of pancreatic cancer associated with printing workers [23]. Wingren et al. [90] investigated mortality patterns among Swedish pulp and paper mill workers and reported excess risk of pancreatic cancer. The Louisiana study found twofold odds ratios for workers in the paper manufacturing industries [75]. Kernan et al. [78] reported a statistically significant increase in risk of pancreatic cancer associated with printing and paper manufacturing. In the Swedish population, Alguacil et al. [50] reported an elevated risk of pancreatic cancer among printing workers in women. While most studies reported an elevated risk, some studies did not observe an association with pancreatic cancer among those workers [62, 64]. It was suggested that exposures to solvents might be the most likely explanation for the association even though specific solvents were not identified [78].

Transport and Communication Industries

A prospective mortality study of cancer by the American Cancer Society involving 461,981 males aged 40–79 years with known smoking habits reported an elevated risk of pancreatic cancer among truck drivers [26]. The Finland study, using other cancer patients as controls, reported an elevated risk of pancreatic cancer for male transport inspectors and supervisors [80]. A hospital-based case-control study of 198 cases and 209 controls indicated an increased risk of pancreatic cancer for truck drivers [68]. A population-based study in Iowa reported that men who worked as heavy truck drivers, or as railroad brake, signal, and switch operators, had an increased risk of pancreatic cancer [83]. A recent hospital-based case-control study in Spain found an approximately twofold increased risk associated with diesel engine exhaust and two to threefold increased risk among truck drivers [81]. Workers in these occupations may be heavily exposed to motor exhaust, which contains PAHs that have been classified as human carcinogens [91] and have been linked to an increased risk of pancreatic cancer [7, 85, 88]. In addition to PAHs, individuals who worked in such industries may also be exposed to a variety of hazardous materials such as cutting oils, solvents, and metal dust, which have been suggested as risk factors [38, 85, 92].

Textile Industries

An occupational mortality study in Washington State reported a threefold increase in pancreatic cancer mortality in both men and women fabric workers under 65 years old [93]. A case-control study involving 625 pancreatic cancer cases and 1700 other cancer controls in Finland found an increased risk among female textile workers [80]. A hospital-based case-control study in Spain observed an elevated risk among female textile and garment workers [65]. A hospital-based case-control study in France reported an increased risk of pancreatic cancer associated with textile industry [76]. A population-based case-control study in Iowa observed an increased risk of pancreatic cancer for female textile sewing machine operators and tenders, and the risk was greater with longer duration of employment in this occupation [83]. A population-based case-control study in Shanghai China also found an elevated risk among female textile workers [70]. It has been speculated that the excessive risk associated with textiles workers may be related to exposure to spinning oils or textile dusts [68]. In contrast, a cohort study in Shanghai China reported that occupational exposure to cotton dust and endotoxin in the textile industry was associated with a reduced risk of pancreatic cancer [94].

Other Occupations and Industries

In addition to the abovementioned industries and occupations that have been relatively well studied, an increased risk of pancreatic cancer has been linked to several other occupational settings. Results from these epidemiological studies, however, have been inconsistent. For example, an elevated risk in glass manufacturers, potters, and construction workers was suggested by some studies [70, 76]. It was unclear whether the association was due to exposures to silica dusts, asbestos, or other industrial dusts [68, 93]. Several solvent-related occupations or industries such as mechanics [33, 65, 68, 80], leather tanners or other leather industries [29, 43, 73, 76], and dry cleaners [71] have been associated with an increased risk of pancreatic cancer. Although farmers are typically exposed to pesticides which have been linked to an increased risk of pancreatic cancer [69, 95, 96], studies have not observed an increased risk of pancreatic cancer among farmers [78, 82]. Employment in furniture and home furnishing stores, medical and other health-related services, educational services, purchasing agents and buyers, supervisors of sales occupations, and insurance sales people have also been suggested to be associated with pancreatic cancer risk [78, 83]. In the absence of exposure to environmental hazards, lifestyle risk factors, such as lack of physical activity [97, 98], may play a role in the development of pancreatic cancer among these workers. While occupational physical activity was associated with a reduced risk of pancreatic cancer based on a meta-analysis of four prospective cohort studies [99], another study found that a reduced risk of pancreatic cancer associated with occupational exposure to physical activity became null after adjusting for body mass index (BMI), suggesting that the observed reduced risk associated with occupational physical activity may be due to confounding factors [94]. It is also possible that exposure to infectious agents may play a role in the development of pancreatic cancer in these professions, since they require extensive personal contacts [83].

General Considerations

When interpreting results from occupational studies, it is important to take the “healthy worker effect” into consideration. Individuals able to sustain employment require a minimum level of health. Employed individuals tend to be healthier than the general population that includes both healthy and sick people. In studies comparing the incidence or mortality of occupational settings to those of the general population, true associations are likely to be underestimated.

Several other issues needed to be considered as well, when interpreting the occupational risk factors.

First, studies using occupation/industry titles to evaluate occupational exposures are likely to introduce exposure misclassification. Occupation/industry titles lack information on specific environmental hazardous agents. Workers classified under a specific occupational title or employed in a specific industry can be exposed to more than one agent. On the other hand, exposure to one agent can occur at multiple occupations or industries. The same occupational title may vary between different industries and may have different exposure levels with regard to agents. A job-exposure matrix, linking information from both occupation and industry titles with specific exposure, would therefore minimize exposure misclassification.

Second, many occupational studies were based on deceased cases due to the clinically aggressive nature of the disease. This limits the quality and quantity of information available. As a result, many previous studies have failed to control for potentially confounding factors such as smoking.

Third, given the rarity of pancreatic cancer, most available studies had limited power to detect small to moderate associations between certain occupational exposures and risk of pancreatic cancer. Thus, many studies were likely unpublished because they were unable to detect meaningful associations. For this reason, pooling of data from projects and replication of studies is very important.

Fourth, nonoccupational risk factors may play a synergistic role with occupational factors in the risk of pancreatic cancer. Integration of occupational and nonoccupational risk factors would provide a more precise profile for predicting individuals’ risks. Finally, genetic susceptibility should also be considered when investigating occupational risk factors.

Non-occupational Risk Factors of Pancreatic Cancer

Smoking

A positive association between cigarette smoking and pancreatic cancer has been demonstrated by nearly all studies published since the 1960s. In a large meta-analysis, current smokers experienced a 70% increased risk of pancreatic cancer compared to nonsmokers, and the risk showed clear dose–responses [100]. After cessation of cigarette smoking, the risk remains elevated for a minimum of 10 years [100]. A recent pooled analysis from the International Pancreatic Cancer Cohort Consortium further demonstrated that current smokers had significantly elevated risk of pancreatic cancer (OR = 1.77) compared to nonsmokers and the risk increased significantly with greater intensity, duration, and cumulative smoking dose [101]. This pooled analysis also indicated that risks after more than 15 years after smoking cessation were similar to that for never smokers [101], which highlights the importance of smoking cessation in disease prevention. Environmental tobacco smoke or passive smoke contains many of the same carcinogenetic chemicals as active smoke [102]. However, very few studies have investigated the association between passive smoke and pancreatic cancer risk. Results from the limited studies have provided mixed results [103,104,105,106].

Alcohol Consumption

Based on the results from most case-control and cohort studies, an International Agency for Research on Cancer (IARC) Monograph working group in 2007 concluded that there was an inadequate evidence of the role of alcohol in pancreatic cancer in humans [107]. However, a positive association between heavy alcohol consumption and pancreatic cancer has been suggested by studies that collected detailed information on alcohol consumption [108,109,110,111,112,113,114,115,116,117,118,119]. A recent pooled analysis using data from the International Pancreatic Cancer Case-Control Consortium further demonstrated that heavy drinkers experienced an increased risk of pancreatic cancer, whereas light to moderate alcohol consumption was not associated with an increased risk of pancreatic cancer [120].

Coffee Consumption

Since McMahon et al. [121] in 1981 reported a strong positive association between coffee consumption and risk of pancreatic cancer, numerous studies have subsequently investigated the relationship and have provided inconsistent results. A meta-analysis of 14 cohort studies conducted in 2011 showed a significant inverse association between coffee consumption and risk of pancreatic cancer [122]. A subsequent meta-analysis including 37 case-control studies and 17 cohort studies suggested a nonsignificant increase of such risk associated with coffee consumption [123]. A recent updated meta-analysis including 20 cohort studies reported a protective effect of high coffee consumption for pancreatic cancer risk (OR = 0.75; 95%CI: 0.63–0.86) [124].

Obesity

World Cancer Research Fund (WCRF) and American Institute of Cancer Research (AICR) panel concluded that there was a dose–response relationship between BMI and pancreatic cancer risk based on 23 cohort studies (RR = 1.14; 95% CI, 1.07, 1.22 per 5 kg/m2 increase in BMI) and 15 case-control studies (OR = 1.00; 95% CI, 0.87, 1.15 per 5 kg/m2 increase in BMI) [125]. A pooled analysis including 14 cohort studies reported that the risk of pancreatic cancer was 47% greater among obese (BMI ≥30 kg/m2) individuals compared to individuals with BMIs between 21 and 22.9 kg/m2 [126]. It was estimated that approximately 12.8% of pancreatic cancers in men and 11.5% in women could be attributed to overweight/obesity [4]. A meta-analysis confirmed that both general and abdominal obesity were associated with increased pancreatic cancer risk [127].

Nutrition

Although studies linking dietary intake and risk of pancreatic cancer have provided inconclusive results, a majority of studies have suggested a reduced risk of pancreatic cancer associated with high fruit and vegetable intake [98, 128,129,130,131,132]. Studies also suggested that certain nutrients found in fruits and vegetables (i.e., vitamin C, vitamin E, carotenoids, and other antioxidants) were associated with a reduced risk of pancreatic cancer [133,134,135,136,137,138]. High fat and red meat intake was associated with an increased risk of pancreatic cancer in some studies [98, 139,140,141] but not in others [132, 136, 142, 143]. A meta-analysis of 11 prospective studies found a positive association between pancreatic cancer incidence and processed meat consumption [144]. However, subsequent cohort studies did not support such findings [145,146,147]. A large cohort study detected no association between intakes of red and processed meat and risk of pancreatic cancer, but the study found that poultry consumption was associated with an increased risk of pancreatic cancer [145]. Another cohort study suggested that processed meat sources of dietary nitrate and nitrite might be associated with pancreatic cancer among men only [147]. A recent large cohort study reported that low meat eaters and vegetarians and vegans had lower mortality for pancreatic cancer compared with regular meat eaters [148]. Frequent nut consumption had been inversely associated with risk of pancreatic cancer in women [149, 150]. Findings from the latest meta-analysis supported that fruit and vegetable intake was inversely associated with the risk of pancreatic cancer [151]. Furthermore, another study suggested that 0–12% of pancreatic cancer cases could be prevented by increasing fruit or folate intake [152].

Diabetes

Diabetes has been considered to be associated with the risk of pancreatic cancer, but the causal relationship between diabetes and pancreatic cancer remains controversial. A recent meta-analysis including 35 cohort studies reported that diabetes was associated with 90% increased risk of pancreatic cancer. The risk was inversely correlated with the duration of diabetes with the highest risk found among patients diagnosed within less than a year [153]. Several studies reported that type I and type II diabetes doubled the risk of pancreatic cancer [154,155,156]. The United States National Cancer Institute estimates that diabetes is associated with a 1.8-fold increased risk of pancreatic cancer in Hispanic men and Asians compared to whites and blacks [67]. Pancreatic cancer risk decreased with the duration of diabetes, but a 30% excess risk persists for those with more than two decades of diabetes diagnosis [70]. Oral antidiabetics or insulin use were associated with a reduced risk of pancreatic cancer [67, 70].

Pancreatitis

Chronic pancreatitis is another established risk factor for pancreatic cancer. A six-country historical cohort study consisting of 2015 subjects with chronic pancreatitis reported 1.8% 10-year and 4.0% 20-year cumulative risks of pancreatic cancer [157]. About 4% of chronic pancreatitis patients developed pancreatic cancer [158]. The risk of pancreatic cancer associated with pancreatitis was two times higher among people who were younger than 65 years old compared to those who were 65 years or older [159]. Patients with hereditary pancreatitis a rare, autosomal-dominant disease that usually occurs at a young age had a risk that was 50–60 times greater than expected [160].

Helicobacter pylori

Studies have shown that Helicobacter pylori infection, a major risk factor associated with pancreatic cancer, has an estimated population attributable fraction of 4–25% [152]. According to a recent follow-up study, these results were not supported [161].

Clinical and Pathological Features of Pancreatic Cancer

Clinical Features

Pancreatic cancer is rare before the age of 40, and the median age at diagnosis is approximately age 70. Pancreatic cancer is difficult to detect and diagnose because of the insidious nature of early stage signs and symptoms as well as the relatively inaccessible anatomic location of the pancreas. The presenting symptoms of pancreatic cancer depend on the location of the tumor within the gland. For tumors located in the head and body of the pancreas, symptoms are generally precipitated by compression of surrounding structures such as the bile duct, the mesenteric and celiac nerves, the pancreatic duct, and the duodenum [162]. As a result, classic symptoms include unexplained weight loss, jaundice, and pain in the upper or middle abdomen and back. Other symptoms may include dyspepsia, nausea, vomiting, and fatigue. Pain is the most common presenting symptom in patients with pancreatic cancer. As a result of tumor invasion of the celiac and mesenteric plexus, the pain may take on a gnawing nature. Besides abdominal pain, patients with pancreatic head cancer usually suffer from jaundice caused by biliary tract obstruction that can increase levels of conjugated bilirubin and alkaline phosphatase. As a result, the patient’s urine darkens. In addition, the stool may be pale from decreased stercobilinogen in the bowel. On rare occasions, a pancreatic tumor may cause duodenal obstruction or gastrointestinal bleeding. Obstruction of the pancreatic duct may lead to pancreatitis. Patients with pancreatic cancer often have dysglycemia. As such, pancreatic cancer should be considered in the differential diagnoses of acute pancreatitis and newly diagnosed diabetes.

Pathological Features

Pancreatic cancer tumors can arise anywhere in the pancreas with the most frequent focus being in the head, followed by the body and tail. Pancreatic cancer grossly produces a firm, poorly demarcated, multinodular mass with an intense desmoplastic reaction [163]. In addition to ductal adenocarcinomas, a number of histological types of pancreatic cancer have been recognized, including adenosquamous carcinoma, colloid carcinoma, hepatoid carcinoma, medullary carcinoma, signet-ring cell carcinoma, undifferentiated carcinoma, and undifferentiated carcinoma with osteoclast-like giant cells. Pancreatic cancers are extremely infiltrative neoplasms. Vascular and perineural invasion are present in the majority of surgically resected cancers. Pancreatic cancer metastasizes most commonly to regional lymph nodes and the liver. Other frequent metastatic sites include the peritoneum, lungs, adrenals, and bones [163].

Molecular Markers

The most widely utilized tumor marker for pancreatic cancer in the clinic is cancer antigen (CA) 19–9. The serum marker CA 19–9 is useful in confirming the diagnosis in symptomatic patients and in predicting prognosis and recurrence after resection [164, 165]. Due to its lack of sensitivity and specificity, this antigen is not useful in screening asymptomatic patients [162].

Global gene expression studies of pancreatic cancers have suggested several potential new serum markers for pancreatic cancer. One such marker is the macrophage inhibitory cytokine 1 (MIC1) [166]. Elevated serum MIC1 antigen levels significantly outperformed CA 19–9 and other tumor markers in distinguishing patients with resectable pancreatic cancers from healthy controls [167]. In addition to MIC1, gene products of osteopontin [168], tissue inhibitor of metalloproteinase-1 [169], and mesothelin genes [170] have also been suggested as potential novel tumor markers of pancreatic cancer.

Using pancreatic juice as a potential source of biomarkers of early stage pancreatic cancer has attracted significant interest [171, 172]. Because of its direct relationship to the ductal system of the pancreas, it would undoubtedly contain enriched fractions of tumor markers unadulterated by serum components [173]. However, pancreatic juice can only be obtained during an invasive endoscopic procedure. Thus, pancreatic juice-based biomarkers are not feasible for screening.

Carcinogenic Mechanisms

During the past two decades, the rapid accumulation of knowledge of the molecular biology of this disease has significantly advanced our understanding of pancreatic carcinogenesis. Like many other malignancies, pancreatic carcinogenesis involves multiple subsets of genes undergoing genetic changes [174]. Pancreatic cancer develops from normal ductular epithelium through a sequential worsening of precursor lesions that can be identified through histology and genetic testing [175, 176]. Overexpression of HER2/neu and point mutations in the K-ras gene present in more than 90% of pancreatic cancer cases at early stages of the disease [175, 177, 178]. The p16 tumor suppressor gene is inactivated in more than 80–90% of pancreatic cancer cases at an intermediate stage [179]. The P53 and DPC4 genes are inactivated in about 50% of pancreatic cancer cases and BRCA2 in about 7–10% at a relatively later stage [174, 180, 181].

Several genetic syndromes (i.e., hereditary pancreatitis, hereditary nonpolyposis colorectal cancer, ataxia-telangiectasia, Peutz–Jehers syndrome, familial breast cancer, and familial atypical multiple-mole melanoma) have been associated with pancreatic cancer risk [182]. However, the carriers of these genetic disorders in the general population are rare. It has been recognized that single-nucleotide polymorphisms (SNPs) in common and low-penetrance genes influence both the response and susceptibility to carcinogens and may play important roles in pancreatic carcinogenesis. Exogenous and endogenous carcinogens can alter gene expression, proliferation, or differentiation through mechanisms such as aberrant DNA methylation, oxidative effects, impaired DNA repair pathways, and abnormal activation of receptors, transcription factors, and cell cycle proteins [183]. While major advances have been made to better understand the interaction between environmental factors and genetic susceptibility to human cancers, the gene–environment interaction for pancreatic cancer has not yet been fully evaluated. There are currently several studies investigating the association between genetic polymorphisms and risk of pancreatic cancer.

Genetic Susceptibility

Studies using candidate gene approaches have mainly focused on genes in the following pathways: carcinogen metabolism [184,185,186,187,188,189,190,191,192,193], DNA repair [186, 194,195,196,197,198,199], inflammatory response [200, 201], alcohol-metabolizing enzymes [202, 203], methylation [117, 202,203,204,205,206], and protease inhibitors [191, 207,208,209]. Associations between polymorphisms in metabolic genes (i.e., GSTM1, GSTT1, CYP1A1, CYP1A2, NAT1 NAT2, and UGT1A7) and risk of pancreatic cancer were generally null from a meta-analysis [175]. However, studies suggested that the combination of GSTT1-null and GSTP1-codon 105 Val variants significantly increased the risk for pancreatic cancer [193]. Individuals who were heavy smokers and carried GSTT1-null genotype significantly increased their risk of pancreatic cancer compared to nonsmokers with GSTT1-present genotype [185]. Heavy smokers with the CYP1A2∗1F(A-163C) C allele or NAT1 rapid alleles experienced a significantly elevated risk of pancreatic cancer as compared to never smokers carrying non-at-risk alleles [188].

A case-control study conducted at the MD Anderson Cancer Center investigated genetic variants in glucose metabolism genes and risk of pancreatic cancer in 1654 cases and 1182 controls [210]. The study genotyped 26 SNPs of five glucose metabolism genes, GCK, GFPT1, GPI, HK2, and OGT, and found a significant association of HK2 R844K GA/AA genotype with reduced pancreatic cancer risk (OR = 0.78). A significant interaction with diabetes was observed. The HK2 R844K GA/AA genotype was associated with a reduced risk of pancreatic cancer among nondiabetic individuals (OR = 0.68) but with increased risk among diabetic patients (OR = 3.69). These risk associations remained statistically significant when the analysis was restricted to whites or after exclusion of recent-onset diabetes. No significant effect of other genes or significant interaction of genotype with other risk factors was observed.

Two studies from Japan examined polymorphisms in alcohol-metabolizing enzyme genes and risk of pancreatic cancer [202, 203]. Miyasaka et al. [203] reported that the risk of pancreatic cancer associated with smoking was enhanced in subjects with an inactive form of ALDH2 in a male population. Kanda et al. [202] found that drinkers carrying both ADH1B His/His and ALDH2 Lys+ had significantly increased risk of pancreatic cancer as compared to nondrinkers with both ADH1B His/His and ALDH2 Glu/Glu.

Li et al. [197] investigated nine SNPs of seven DNA repair genes (LIG3, LIG4, OGG1, ATM, POLB, RAD54L, and RECQL) and found SNPs in ATM and LIG3 genes significantly associated with the risk of pancreatic cancer and suggested significant interactions between SNPs in ATM or LIG4 genes and diabetes to pancreatic cancer. Several studies suggested that polymorphisms of XRCC2 and XPD genes modified smoking-related pancreatic cancer [186, 196, 198]. Some studies also suggested potential gene–gene interactions within the same pathway (i.e., XRCC1 with APE1, XRCC1 with MGMT, OGG1 with XPC, XPA with ERCC2) [195] or cross different pathways (i.e., XRCC1 with GSTT1/GSTM1) [194] in relation to pancreatic cancer risk.

A case-control study from Mayo Clinic of 1354 Caucasian pancreatic cancer patients and 1189 healthy Caucasian controls investigated 1538 SNPs in 102 inflammatory pathway genes [201]. After adjusting for known risk factors for pancreatic cancer, single SNP analysis revealed an association between four SNPs in NOS1 and one in the CD101 gene with pancreatic cancer risk. These results, however, were not replicated in other pancreatic cancer case-control and cohort populations. A population-based case-control study with 308 cases and 964 controls from the San Francisco Bay Area suggested that proinflammatory gene polymorphisms in combination with proinflammatory conditions might influence pancreatic cancer development [200].

Suzuki et al. [117] investigated polymorphisms in MTHFR, MTR, MTRR, and TS genes and found that heavy drinkers carrying MTHFR 667 CC, MTR 2756 AA, or MTRR 66G allele had significantly increased risk of pancreatic cancer compared to nondrinkers, suggesting that folate-related enzyme polymorphisms modify the association between alcohol consumption and pancreatic cancer risk. Wang et al. [206] reported an increased risk of pancreatic cancer associated with MTHFR 677CT or TT genotypes compared to MTHFR CC genotype and with TS 3Rc/3RC genotype compared to TS 3Rg/3Rg genotype. This study also suggested an interaction between MTHFR C677T polymorphism and smoking and drinking. Similar interactions were also reported in another study [204].

Recently, genome-wide association studies (GWAS) among the population of European ancestry identified common SNPs in several genomic regions (i.e., 1q32.1, 2p14, 3q28, 5p15.33, 7p14.1, 7q32.3, 8q24.21, 9q34.2, 12q24.31, 13q22.1, 16q23.1, 17q24.3, 22112.1) that are associated with pancreatic cancer risk [211,212,213,214]. A GWAS from China identified five significant genomic regions (5p13.1, 10q26.11, 21q21.3, 21q22.3, and 22q13.32) that are associated with risk of pancreatic cancer [215]. A Japanese GWAS reported three significant loci (6p25.3, 7q36.2, and 12p11.21) associated with pancreatic cancer risk [216]. Future studies are needed to investigate gene–environmental interactions with a broad spectrum of occupational and environmental factors in addition to smoking and alcohol consumption.

Conclusion

Although the overall incidence of pancreatic cancer is low in comparison to other cancers, this devastating disease is associated with a low survival rate, often claiming the life of its victims within the first year. From previous studies, a wide array of contributing occupational and nonoccupational risk factors has been suggested. Some of these include smoking, excessive alcohol consumption, obesity, physical inactivity, diabetes, chronic pancreatitis, nutritional considerations, and complex genetic predispositions and interactions. Further studies and data pooling may help gain a better understanding of such risk factors, ultimately leading to effective awareness and prevention programs.

Since delays in early diagnosis may contribute to poor prognosis, misclassification of initial symptoms may be prevented and earlier diagnosis accomplished through the use of specific molecular markers. Thus, the identification and implementation of pancreatic tumors markers has potential to be an important diagnostic tool.