Here we discuss the epidememiology of Pancreatic cancer (PCa), the eighth leading cause of death from cancer in men and the ninth leading cause of death from cancer in women throughout the world. Delay in diagnosis, surgically inaccessible location of the pancreas, absence of classic symptoms of the disease and poverty of molecular biomarkers result in a diagnostic challenge.
1Janeesh Sekkath Veedu, MD, 1Febin John, MD, 2CS Pitchumoni, MD, 1St. Peter’s University Hospital Rutgers RWJ Medical School, New Brunswick, NJ 2Chief of Gastroenterolgy, Hepatology and Clinical Nutrition, Saint Peter’s University Hospital, New Brunswick, NJ
INTRODUCTION
Pancreatic cancer (PCa) is the eighth leading cause of death from cancer in men and the ninth leading cause of death from cancer in women throughout the world.1 The American Cancer Society estimated that 45,220 Americans will be diagnosed with Pancreatic Cancer in 2013. Delay in diagnosis, surgically inaccessible location of the pancreas, absence of classic symptoms of the disease and poverty of molecular biomarkers result in a diagnostic challenge. PCa represents 2.4% of all cancers and 3.7% of cancer deaths. Early surgical resection has a survival benefit, however, since the disease is often diagnosed at late stages surgery is not curative and adjuvant therapy becomes palliative.2
PCa is usually seen in the elderly with a male predominance, the peak incidence being in those aged 65-75 years. Adenocarcinoma accounts for 95% of all cases, about 85% are sporadic with no family history or predisposing genetic syndromes. Although 5-year survival is low (<5%), high volume surgical centers have reported survival rates of up to 40%.3 Recent studies have shown that distinct molecular subtypes of PDAC exist and are associated with different prognosis and therapy response.4 Related to improved life-expectancy and probably adoption of cancer associated lifestyles the incidence is growing globally.3
ETIOLOGICAL ASSOCIATIONS
A. ENVIRONMENTAL
Few modifiable risk factors have been implicated in the etiology of PCa.
Cigarette Smoking
Cigarette smoking is a well-established risk factor for PCa and a co-factor in chronic pancreatitis secondary to alcoholism.5 It is attributable for about 20-30% of cases of PCa.6-13 Studies report a higher risk among current smokers compared to non-smokers, up to 6 fold depending on duration and intensity of cigarette smoking,14 (RR=1.74, 1.61—1.87) and also in former smokers with respect to never smokers (OR=1.20, 1.11-1.29).7,9 Smoking 1 pack/day increased the risk by 1% and the risk doubled for those with >40 pack years of smoking.7 Pipe/cigar smoking had lower risk when compared to cigarette smoking but passive smoking (workplace/household) did not increase the risk.12 Though the risk remained elevated for up to 15 years after quitting, a non-significant drop in risk was observed after 20 years.7,9
Cigarette smoke contains nearly 4000 chemicals of which more than 60 has been identified as carcinogens14 (polycyclic aromatic hydrocarbons, N-nitrosamines, aromatic amines, 1,3-butadiene, benzene, aldehydes and ethylene oxide). The toxins reach the pancreas indirectly via bloodstream or biliary regurgitation to exert carcinogenic effect.8,11,15 Recent studies have looked into genetic variations at carcinogen-metabolizing enzymes to further understand individual susceptibility to PCa.15 Of the carcinogens the most potent metabolite, NNK mediated pathways is well studied.14 There have been no major studies on the effect of e-cigarettes on PCa. E-cigarette users were more nicotine dependent than nonusers, had more prior quit attempts, and were more likely to be diagnosed with thoracic and head or neck cancers.16
Alcohol
An association between alcohol abuse and pancreatic injury was reported by Friedreich as early as 1878.17 Freidreich recognized an association of alcohol abuse with chronic pancreatic injury.17 Several studies have evaluated the association of alcohol and PCa but conclusive evidence is lacking.11,17-21 This could be due to interplay of significant confounders such as smoking, pancreatitis, nutritional and genetic factors.17,21 However, alcohol has been projected as an independent risk factor, attributable to 2-5% of all PCa cases (where population prevalence of heavy drinking is 10-15%).11,17,21 Heavy drinking (>40g or >3drinks/ day) is associated with moderate risk (RR=1.22, 1.12- 1.34) in women and up-to 3.5-fold risk in male binge drinkers (>70g or >5 drinks/day).17,19,21 The risk with type of beverage consumed (wine, beer, liquor/spirit) is variable but an increased risk with the duration of alcohol consumption is reported.18,19
The causal role of alcoholic pancreatitis which is responsible for <5% of PCa cases is not adequate enough to explain the link between alcohol and this PCa.21 Acetaldehyde (oxidative pathway) and fatty acid ethyl esters (non-oxidative pathway), the metabolic products of alcohol, activate pancreatic stellate cells leading to inflammation, immune response and cancer.17,21,22 Folate depletion leading to defective DNA synthesis/repair and carcinogen activation via induction of cytochrome P450 is also postulated means of alcohol injury.18
Diet
The role of diet in the pathogenesis of PCa is weak and contradictory. Mediterranean diet rich in plant-based foods, whole grains and fish with modest consumption of meat and dairy products was associated with a decreased risk (OR=0.51, 0.31-0.84).23-26 A 2.4-fold risk was reported in men on a Westernized diet (red/processed meat, potato, sugary beverage, refined grains, eggs and high-fat dairy).24 Red meat consumption was associated with an increased risk in men.27 A statistically significant 19% higher risk in those consuming processed meat (50g/day) was reported in a meta-analysis.27 Energy- dense diet consumption escalated the risk (up to 72%) while soft drinks did not.28,29
Although dietary fat is not associated with increased risk, a recent study attributed a diet rich in cholesterol with low fiber and folate to the increased incidence of PCa in Poland.30,31 Dietary magnesium especially in overweight men is found to decrease the risk (18% reduction with 100mg increased intake) and cruciferous vegetables (OR=0.90) were protective.32,33 There is no protective effect for antioxidant consumption while a 2-fold risk is seen in those with high serum levels of 25 (OH) vitamin D (=100 nmol/L).34,35 Based on data from the European Prospective Investigation into Nutrition and Cancer Cohort, coffee (total or decaffeinated) and tea consumption are not related to the risk of PCa.36
Occupation
Certain occupations are associated with an increased risk, especially with exposure to chlorinated hydrocarbons and polycyclic aromatic hydrocarbons (PAH).37 Dry-cleaning, metal-related work (Gold/silver smith) and electronic work have exposure to chlorinated compounds.37 Although an increased risk with herbicide and fungicide (not insecticide) was noted in one US study, others failed to demonstrate this risk.37,38 PAH associated risk was seen among metal workers and those in aluminum industry.37,39 Occupational exposure to inorganic dust, asbestos and ionizing radiation also amplified the risk.37,40 Assessment of PCa risk among night shift workers in Japan and food industry workers in Finland did not yield significant result except in Finnish males (SIR=1.5, 1.13-1.96).41,42
Exposure To Heavy Metals
There are reports in literature to support the risk for PCa from exposure to heavy metals. Higher incidence of PCa in the East Nile delta region of Egypt is now attributed to cadmium exposure from fertilizers and polluted river water.43,44 An epidemiological study from Louisiana reported this heavy metal exposure from food (pork, seafood, rice) as the cause of increased incidence of this cancer among the Cajuns.45 Cigarette smoke is another potent source of cadmium which could be implicated for the increased risk of PCa among smokers.45 Cadmium exerts its carcinogenic effect via impairment of DNA repair mechanisms to cause genomic instability.44
Arsenic with similar carcinogenic mechanism is also incriminated in PCa.44 A recent study from Florida reported a significant increase in the risk among those living within 1 mile radius of Arsenic-contaminated wells.46 There are reports about childhood arsenic exposure (from milk powder) and increased mortality associated with this cancer.44,47 Although asbestos exposure from drinking water was reported to increase PCa, subsequent follow-up and analysis failed to prove this.48 Similarly there are studies, which link exposure to lead and decreased levels of selenium (toenail concentrations) to PCa.44
Radiation
In a study of the Hiroshima and Nagasaki, the two sites of atomic bombings, no radiation effect was noted.49 However one study showed excess deaths from PCa in patients who received therapeutic irradiation for ankylosing spondylitis50 and another reported two cases of PCa in patients who got abdominal radiation for testicular cancer.51
Infection
Studies have revealed an infective etiology of PCa, the main agents being H pylori, HBV, HCV and HIV.
H pylori, extensively studied as a gastric carcinogen is being investigated for extra-gastric associations. Several studies including a meta-analysis have found significant association of PCa (AOR=1.38, 1.08- 1.75) with this highly prevalent infection (40% in developed countries and 70% in developing countries) and about 2-fold risk with CagA +/ VacA + strains.52,53 The antral colonization of H pylori and subsequent hyperchlorhydria leading to increased pancreatic secretions and hyperplasia is one of the plausible mechanisms.52,54 Inflammation (IL 8 and VEGF) and bacterial overgrowth from hypochlorhydria (increased N nitrosamine) are other suggested mechanisms.52-54
An increased risk was seen with active (RR=3.83), chronic (RR=1.39) and past (RR=1.41,1.06—1.87) HBV infection.55 There was a synergic increase in the risk in chronic/inactive HBsAg carriers with DM.56 HCV with mechanisms similar to HBV was found to double the risk for PCa (SIR=2.1, 1.4, 2.9).57-59
There is an increased incidence of PCa in HIV patients (SIR=2.2, 1.2-3.6).60 PCa was diagnosed at a younger age with advanced stages at presentation and had a higher likelihood of unfavorable performance status in HIV positive subjects.61 Periodontal disease, and Porphyromonas gingivalis, a pathogen for periodontal disease, are reported associations in PCa.62
B. CHRONIC PANCREATITIS
All types of chronic pancreatitis predispose to PCa (RR=5.1, 3.5-7.3), although <10% is attributed to it.63,64 A significant risk was associated with both acute (HR=9.1, 3.81-21.76) and chronic pancreatitis (RR=13.1, 6.1-28.9).63,65,66 Different forms of chronic pancreatitis such as hereditary, autoimmune and tropical pancreatitis are discussed in literature, all of which are significantly associated with PCa. Patients who underwent surgery for the treatment of chronic pancreatitis had significantly lower incidences of pancreatic cancer. Surgery for chronic pancreatitis may inhibit the development of pancreatic cancer in patients with chronic pancreatitis.67 Acute pancreatitis may be an initial manifestation of PCa.
1. Hereditary Pancreatitis (HP)
Hereditary Pancreatitis (HP) is an inherited form of chronic pancreatitis characterized by recurrent episodes of pancreatitis since childhood.68-71 Mutation in the cationic trypsinogen gene (PRSS1) was the first identified genetic defect.68,71 Subsequently several germline mutations such as protease serine 2 (PRSS2), pancreatic secretory trypsin inhibitor (SPINK1), CFTR, chymotrypsinogen C (CTRC) and calcium-sensing receptor (CASR) were discovered.68 Individuals with HP have a high risk for PCa (SIR=87, 42-114).69,70,72 The cumulative risk by age of 75 years is about 40%- 53.5%.69,70,72 Smoking and diabetes further increased the risk in these patients.71
2.Tropical Calcific Pancreatitis (TCP)
Also known as fibrocalculous pancreatic diabetes (FCPD), Tropical Calcific Pancreatitis (TCP) is a form of chronic pancreatitis in Afro-Asian countries.73,74
The exact etiology for this form of chronic pancreatitis has not been established. Studies have clearly shown that this is a high risk factor for PCa (RR=5, 1.03-14.6).73,75 Patients who develop PCa are younger compared to the denovo form. The entity although found in many states in India is well studied in large series of patients mostly from states of Kerala and Tamil Nadu.
Early studies identified TCP as a disease in young malnourished individuals with poor prognosis leading to diabetes and having a high risk for PCa. But recent research found strong genetic links to this disease (SPINK1/CFTR mutations) and dismissed the notion of regional predominance, links to nutrition and grave prognosis.76,77
The pathogenesis of malignancy in pancreatitis is postulated via inflammatory mediators, activation of signaling pathways (cyclooxygenase2 expression, Notch signaling, Hedgehog signaling) and oxidative damage.66,67 Ueda et al. reported a decreased risk in chronic pancreatitis patients managed surgically (HR=0.11, 0.014-0.80) which provides further evidence for the inflammatory etiology.67
3. Autoimmune Pancreatitis
Autoimmune Pancreatitis is a steroid responsive type of chronic pancreatitis, which mimics PCa.78-80 Although several case reports have been published, conclusive evidence regarding its association with cancer is lacking other than an increased occurrence of K-ras mutations.78-80 AIP features a significant inflammatory phase, and hence it is biologically plausible that AIP patients are similarly at increased risk for developing PCa. The potential for systemic inflammation in this multiorgan disease could also contribute to risk for extra pancreatic cancers. Finally, the late age at presentation of type 1 AIP and reports of cancer being discovered shortly before and after AIP diagnosis have fueled speculation that AIP is a paraneoplastic manifestation of an underlying cancer.81
C. DIABETES AND PANCREATIC CANCER
This is a clinically important but controversial topic. Type 2 Diabetes (DM) with its temporal association with PCa is described both as cause and result of the cancer.82-84 This is an independent risk factor with approximately two-fold increased risk compared to general population.82,83 Risk is inversely associated with the duration of DM, the highest risk being with <1year of DM (OR=5.38;3.49—8.30).83 No increased risk was seen in subjects with >9 years of DM (OR=1.02; 0.68- 1.52) which contradicted the findings of a previous meta-analyses.82,85,86 The association between DM and PCa was not modified by gender, smoking, age, or BMI.82 History of diabetes in a first degree relative increased the risk (OR=1.37, 1.10-1.71) per the PACIFIC study (pancreatic cancer: investigation into finding causes).87 A meta-analysis observed equal risk in diabetic men and women but some disparity exists in this regard.83,84,88 Higher risk was seen among those using insulin compared to those without (OR=3.34 vs. 1.50) in the Iowa Women’s Health Study (IWHS).84
Hyperglycemia associated with altered glucose metabolism, chronic inflammation, oxidative stress, and activation of insulin signaling cascades increases the risk of pancreatic cancer.89 The development of DM within a few years of a pancreatic cancer diagnosis is more likely to suggest an effect of the tumor, whereas diabetes of longer duration is more likely to contribute to the development of cancer.90
However in a study from Japan, PCa was diagnosed within 2 years of DM onset (new-onset) in 0% of the patients with early-onset DM, and in 33% of those with late-onset DM. Pre-existing type 2 diabetes, acute alcoholic hepatitis, acute pancreatitis, cholecystitis, and gastric ulcer independently or jointly predict subsequent pancreatic cancer risk.91
The notion that new-onset diabetes in pancreatic cancer is a paraneoplastic phenomenon caused by tumor secreted products was strengthened by a recent study that proposed adrenomedullin, a 52 amino- acid polypeptide, as a strong candidate for mediator of diabetes in pancreatic cancer. Adrenomedullin was also shown to be overexpressed in human pancreatic cancer and plasma levels of adrenomedullin were also increased in pancreatic cancer patients, especially those with diabetes.92 Earlier concept of beta cell destruction has given way to the role of hormonal secretions from the tumor causing insulin resistance, up-regulation of IGF- 1 leading to carcinogenesis via enhanced angiogenesis and cell growth without apoptosis.82,83,89 Supported by the fact is the observation that IGF receptor and insulin receptor substrate-2 (IRS-2) are over-expressed in cancer cells of the pancreas.93 Other+ studies have shown the presence of diabetogenic factors (2030 MW peptide, Amylin/IAPP) in the serum.83,94 Thus patients with new-onset DM with a family history of DM should be screened for underlying malignancy.87 Similarly new-onset DM in older patients (>65 years) with a negative family history and low BMI (<25) or recent weight loss (>2kg) also have a likelihood for associated PCa.95 Reducing diabetes by controlling obesity could benefit pancreatic cancer rates, in addition to the many other known health benefits.82 One study showed that dyslipidemia, but not diabetes, is a significant risk factor for PCa. Patients with new-onset diabetes and a history of dyslipidemia are at an especially high risk of PCa.96 DM is also an independent risk factor for liver, colorectal and breast cancers but decreases the risk of prostate cancer.97
The use of metformin, the most commonly prescribed drug for type 2 diabetes, was repeatedly associated with the decreased risk of the occurrence of various types of cancers, especially of pancreas and colon and hepatocellular carcinoma.98
D. OBESITY
Obesity, a rising epidemic, has association with multiple cancers99-101 and has been discussed in detail in an earlier paper in this series. Most of the studies have found an association of increased BMI (marker of obesity) with PCa (RR=1.2-3).84,102,103 A meta-analysis observed 19% increased risk in obese people (BMI>30 kg/m2).101 Obesity related PCa had a population attributable fraction of 26.9% and 19.3% in US and EU respectively.102,104 An earlier age of onset was seen in those who were obese/ overweight during their adolescence (HR=2.09, 1.25- 3.50).102,105 Similarly in the elderly, obesity was found to reduce survival in PCa patients.102 Metabolic syndrome (MetS) is associated with many more consequences than generalized obesity. MetS was found to be associated with PCa in both men (SIR=178,144-266) and women (RR=1.58, p<0.0001).106,107 MetS components were also found to increase the risk [fasting blood glucose (OR=4.24), total cholesterol (OR=1.79), apolipoprotien A (OR=36.06)].108 One European study reported significant risk with several metabolic factors in women (mid-blood pressure, glucose, triglycerides, BMI).109
Physical activity was found to decrease the risk [e.g.: history of sports (HR=0.80, 0.64-0.99), occupational physical activity (RR=0.75, 0.58-0.96)].103,110 Release of cytokines (IL-6, TNF a, CRP) leading to insulin resistance and higher insulin levels result in increased IGF-I: IGFBP-3 (insulin growth factor and binding protein ratio) is probably related to carcinogenesis in obesity.101,106,111
INHERITED PANCREATIC CANCER
Genetic predisposition accounts for 5-10% of all pancreatic cancers.72,87,112,113
Familial pancreatic cancer (FPCA) -is defined as a family with more than one first degree relative (FDR) with history of PCa without any inherited syndromes.112 A 2.3 to 4.5-fold increased risk with 1 FDR, 6.4-fold with 2 FDRs and up-to 32-fold with =3 FDRs with pancreatic cancer has been noted.72,114,115 FPCA which is influenced by race (Ashkenazi Jews), smoking and diabetes, and genetic anticipation (younger age or worse prognosis with successive generations).72 Pancreatic intraepithelial neoplasia with mutations in the K-ras (codon 12) was more frequently (2.75 fold) observed in familial pancreatic cancer when compared to sporadic.72
Hereditary pancreatic cancer- is a genetic syndrome with mutations that increase the risk for PCa.112 Peutz- Jeghers syndrome with STK11/LKB1 gene mutation is associated with up-to 132-fold increased risk for PCa.72,112,116,117 Hereditary Non Polyposis Colon Cancer (HNPCC) is associated with a lifetime risk of 1.3-4% for PCa.112 Majority of the Hereditary Breast Ovarian Cancer (HBOC) is due to mutations in BRCA1 and BRCA2 genes.112 The risk for PCa in BRCA1 carriers is minimally elevated compared to general population (RR=2.8 vs. 1.3%).112 BRCA2 mutation has a 5-7% lifetime risk in carriers and is the most common inherited gene for development of PCa.112 Families with Familial Atypical Multiple Mole Melanoma syndrome (FAMMM) are at increased risk (13-22%) for this cancer.72,112 Individuals with p16/ CDKN2A (FAMMM gene) mutation have a 38-fold higher risk in comparison to general population.72 Studies have shown a mild elevation of risk in FAP with APCA gene mutation (RR=4.5), Cystic fibrosis (CFTR gene) and Ataxia telangiectasia (ATM gene).72,112,116 Associations of PCa with PALB2 (partner and localizer of BRCA2) and palladin (cytoskeleton associated protein) mutations are being observed.72,116,118. (See table X)
OTHER FACTORS
Several studies among different populations across the world have reported an increased risk associated with non-O blood groups for PCa (OR=1.37, 1.02- 1.83).59,119-121 Mechanisms though not clear; relevance of physiological differences in inflammatory mediators (TNFa, cellular adhesion molecules) is being postulated.119 Few studies have found a significant co- relation of this disease with a history of cholelithiasis (HR=3.12, 2.05-4.78) and cholecystectomy (higher prevalence 6.2% vs. 2.9%).108,122 Anti-diabetic medications and NSAID’s are found to have an effect on PCa risk. Metformin (HR=0.73, 0.66—0.80) and thiazolidinediones were associated with reduced risk but insulin (HR=4.63, 2.64—8.10) and sulphonylureas (HR=4.95, 2.74—8.96) aggravate the risk.123-125 DPPV IV inhibitors (sitagliptin) have a theoretical risk for carcinogenesis126 but a recent meta-analysis on this issue reported conflicting results.127
Although not conclusive enough, there is evidence to suggest that high-dose aspirin reduces the risk for PCa (OR=0.78, 0.64-0.95).128-130 Similarly there is lack of satisfactory evidence for other NSAIDs.130 A combination of aspirin, curcumin and sulphoraphane has been found to be beneficial against PCa in animal studies.131 One study from Netherlands observed an inverse association of PCa with hypertension.132 Metformin offers a potential novel approach for pancreatic ductal adenocarcinoma prevention and therapy.133
ALLERGIES
A recent meta-analysis reported 30% drop in pancreatic cancer risk among those with history of allergies.134 Statistically significant risk reduction was observed with hay fever (OR=0.74, 0.56-0.96) and allergy to animals (OR=0.62, 0.41-0.94).134,135 Other allergies, such as those to foods and medications, have been less well studied and associations with risk are unclear.136 Heightened immune surveillance is suggested as the plausible explanation.
GLOBAL EPIDEMIOLOGY
The annual incidence and mortality of PCa is the same (ASR incidence =7.2 vs. 2.8 and ASR mortality=6.8 vs. 2.7).137 Analysis of global data based on human development suggest a higher incidence in areas with high human development (ASR incidence= 4.6) as opposed to areas with less development (ASR incidence= 1.2) (Fig-1).137 Although reasons are not fully elucidated, it is linked to human lifestyle and diet. Immigrant studies, which found increased risk among Indians who migrated to Australia and UK, support this observation.138,139 The highest incidence for women is reported in North American and northern Europe. A high incidence of PCa in Ashkenazi Jews140 and a lower incidence among the Utah Mormons141 has been noted. A brief summary of the incidence of PCa in different parts of the world is given below.
A. AMERICAS
North America, with isolated exceptions, has the highest incidence and mortality for PCa in the world (Incidence ASR= 7.4 and mortality= 6.9) (Fig-1).137 Even though rates in South America are lower, French Guyana and Uruguay are ahead of US and Canada.137 Lowest estimates are seen in Guatemala, Haiti, Panama and the Bahamas (Central America).137
Although PCa ranks only 13th among cancers, it is the 4th major cause of cancer-related death137 with a 5-year survival rate of 6%.142 Blacks have a higher incidence (33% more) and mortality (32% more) than Whites.143 Asia Pacific Islanders have the lowest rates and better survival.142 The rates for the indigenous groups fall between the Blacks and Asia- Pacific Islanders.142 This racial disparity could not be attributed to any of the known risk factors (smoking, BMI, family history, diabetes and cholecystectomy).143 Latitudinal variation in the incidence and mortality of this cancer was attributed to solar UV-B exposure.144
B. EUROPE
Western Europe, Central and Eastern Europe have higher incidence (ASR=6.6-7.3) and mortality (ASR=6.6-6.8) for PCa compared to Northern and Southern Europe (Fig-1).137 Highest incidence is seen in Czech Republic, Slovakia, Hungary, Slovenia and Finland with Czech Republic having the highest incidence rate in the world (ASR=9.7).137 Sweden, Albania, Cyprus and Bosnia have the lowest rates. Hungary has the highest mortality rate for PCa in Europe (ASR=8.8).137
The overall cancer mortality with the exception of PCa has decreased in this region since 1980.145,146 A higher incidence was reported among people living in most deprived areas, partly linked to high prevalence of smoking.146 In England, total number of cases was higher in Whites compared to Non-Whites (South-Asian, Blacks, Chinese 17% of population). However, age standardized incidence was higher in Blacks (ASR=5.7 in Blacks vs. 4.9 in Whites).139 The risk was lower in South-Asians compared to Whites while no significant risk was demonstrated in Blacks and Chinese.139 Immigrant population in England (non- White) had higher incidence of this cancer compared to their counterparts living in homeland.139 A 138% increase in incidence of PCa was reported in the Inuit population (SIR=2.38, 1.97-2.86; p<0.0001) plausibly due to the high prevalence of diabetes and smoking.147 Even though a Nordic country, Finland has a high incidence of PCa (ASR=8.7).137 This is the 5th most fatal cancer in the country (ASR=7.8).137 The 5-year survival has not improved much over the past 50 years (from 3 to 5%).148 Although coffee consumption is very high, it is not associated with an increased risk for PCa (HR=0.82, 0.38-1.76).149
C. ASIA
Eastern Asia (People’s Republic of China, Japan, North Korea, South Korea, Mongolia and Taiwan) followed by Western Asia (Armenia, Azerbaijan, Middle East, Cyprus, Sinai Peninsula of Egypt, Georgia, Turkey) has the highest incidence of PCa in Asia (ASR=4.5 and 3.9 respectively) (Fig-1).137 Similar is the trend in mortality (ASR=4.3 and 3.8 respectively). The Central and South- East Asia have lower rates (Fig-1).137 Highest mortality rate for PCa is in Armenia worldwide (ASR=8.9). Also Armenia has the highest incidence and mortality in this continent followed by Japan, Israel, Kazakhstan and Korea.137 Lowest rates are seen in India, Nepal, Bangladesh, Pakistan and Sri Lanka.137
1. Japan
PCa has a high incidence in Japan (ASR=8.5) and is the 5th common cause of cancer-related mortality in both men and women (ASR=9.5 and 6.1 respectively).137 Although the 5-year survival is around 5%, resected cases have better prognosis (5-year survival=18.8%).150 Positive family history and presence of diabetes were reported as major risk factors apart from smoking.151 Northern Japan has a higher mortality from this cancer as opposed to south which was linked to variations in exposure to solar irradiation and temperature.152 A high non-linear relationship of PCa death with low- dose external irradiation (<20mSv) was reported in Japanese-A bomb survivors.153
2. Korea
With a high incidence (ASR=6.7), PCa in Korea is the 5th most fatal cancer (ASR=6.2).137 Analysis of trends in incidence from 1999-2010 showed that PCa is increasing in Korea (APCA in both sexes=1.4) with a greater increase in women (APCA=2.2 vs. 0.6 in men).154 Although hepatitis B is endemic in Korea, it was not found to increase the risk for this malignancy (OR 1.03, 0.69-1.53; p=0.91) as opposed to hepatitis C (OR=2.30, 1.30-4.08; p< 0.01) and non-O blood group (OR=1.29, 1.05-1.58; p=0.01).59
3. China
The incidence of PCa is not very high in China (ASR= 3.6) but it is the 8th most common cause of cancer related death.137 Cigarette smoking (44%), pancreatitis (16%) and family history of PCa (8%) were the major etiologies in young patients with PCa.155 Incidence of diabetes among PCa patients was much higher (34.6% vs. 8.8%) and 74.56% (in the cancer group) had onset of DM within 2 years of diagnosis of cancer.156 Energy dense foods increased the risk (OR: 1.72; 95% CI: 1.25, 2.35; P = 0.001) in Chinese.28 Regular green tea drinking was found to be protective in Chinese women (OR 0.68, 95% CI 0.48—0.96).157 For reasons not clear, in one study acute pancreatitis patients had a high risk of developing PCa within 5 years of index pain (HR=9.10; 3.81-21.76).66
4. India
India has a low incidence (ASR=1.2) and mortality (ASR=1.1) for PCa.137 Recent studies note an increasing incidence of PCa in both men and women.158 A greater risk was observed in educated males with about 3-fold risk in those with >12 years of education.159 Tropical calcific pancreatitis, which has a very high risk for this cancer (RR=100, 37-218), is highly prevalent in some parts of India.73,75
5. Israel
PCa in Israel ranks 3rd in men (ASR=8.6) and 4th in women (ASR=6.2) in malignancy related deaths.137 Jews had a higher incidence than Arabs (ASR males 7.45 vs. 5.61) with the highest incidence (ASR males 8.11 vs. 7.45) in immigrant Jews (European-born).160 Nevertheless, a decreasing trend is seen in the Jewish population160 Mutation in BRCA1/2 was the major cause of this cancer in Ashkenazi Jews.161
D. AFRICA
The southern part of the continent, which includes South Africa, Mauritius and Zimbabwe, has the highest estimates for PCa in Africa (ASR incidence=4.3, ASR mortality=4.2) (Fig-1).137 Libya and Egypt, though in the northern part has incidence and mortality rates similar to Southern Africa.137 There are few studies, which attribute pollution of the Nile for the increased incidence in Egypt.43,162 Serum cadmium is suggested as the etiologic agent for the occurrence of early- onset PCa in the East Nile delta region.43 A Moroccan study reported higher incidence (17%) of pancreatic adenocarcinoma in adults <45 years (3% in the US) which did not correlate with smoking, alcoholic pancreatitis or family history.163 Lowest incidence (ASR<2) and mortality (ASR<1.9) for this malignancy is seen in Malawi, Guinea and Tanzania.137
E. OCEANA
Australia has the highest incidence (ASR=6.6) of PCa in this region followed by New Caledonia (ASR=6.5) and New Zealand (ASR=5.9) while island countries such as Samoa and Vanuatu have the least (Fig-1).137
1. Australia
PCa is the 5th common cause of cancer-related death in Australia.137 A rise in incidence (ASR 7.67 to 8.24) and mortality (ASR 7.02 to 7.58) of PCa was observed in women (from 1977-2006) while these estimates dropped in men164 attributed to variation in smoking habits.164 An interesting observation is that mortality from PCa was 9% less in Brisbane (Queensland) when compared to Melbourne (Victoria), which was linked to variations in UV exposure between the two capital cities.165
2. New Zealand
PCa is the 5th most fatal cancer in this country.137 A higher incidence and worse prognosis was reported among the Maori tribe which was attributed to smoking.166 Blakely et al. analyzed cancer incidence in New Zealand by dividing the population into 4 ethnic groups (Maori, Pacific, Asian and European/others) and found a 1.5-times higher rates for PCa in the Maori and Pacific groups compared to the European group.167
CONCLUSION
PCa continues to be a major clinical challenge because of a trend in increase in incidence with no major improvements in survival. Recent studies have reported miRNA-21, c-Myc, L-type amino acid 1 transporter (LAT1), K-ras codon 12 mutation, p38? Mitogen-activated Protein Kinase and SMAD4 as biomarkers in predicting prognosis and survival in these patients.168-174 Circulating Tumor Cell (CTC) detection in peripheral blood with a diagnostic accuracy of 70% (EUS-FNA=85%) is a promising noninvasive early diagnostic procedure.175 A study from MD Anderson, Texas reported an association between NAFLD and pancreatic cancer.175 If this observation is confirmed in further studies, it is concerning since MetS and obesity are rapidly increasing. Furthermore they observed simultaneous pancreatitis and liver cirrhosis in obese pancreatic cancer patients providing additional evidence for the role of obesity.175
Endocrine (islet cell tumors) and rare non-endocrine tumors (acinar cell carcinoma, adenosquamous carcinoma, colloid carcinoma, giant cell tumor, hepatoid carcinoma, intraductal papillary-mucinous neoplasm, mucinous cystic neoplasm, pancreatoblastoma) of the pancreas are not discussed in this article.
Regardless of the advances in medical science, PCa remains a challenge. More desirable survival outcomes rely on novel research that focuses on finer diagnostic and therapeutic approach, yet to materialize.