Given the changing demographics of Western populations, the numbers of pancreatic cancer cases are projected to increase during the next decade. Diabetes, recent cigarette smoking, and excess body weight are the cancer’s most consistent risk factors. The search for common and rare germline variants that influence risk of pancreatic cancer through genome-wide association studies and high-throughput-sequencing–based studies is underway and holds the promise of increasing the knowledge of variants and genes that play a role in inherited susceptibility of this disease. Research reported in this review has advanced the understanding of pancreatic cancer.
Key points
- •
Given the changing demographics of Western populations, the numbers of pancreatic cancer cases are projected to increase during the next decade.
- •
Diabetes, recent cigarette smoking, and excess body weight are the most consistent risk factors of pancreatic cancer.
- •
The search for common and rare germline variants that influence risk of pancreatic cancer through genome-wide association studies (GWAS) and high-throughput-sequencing–based studies is underway and holds the promise of increasing the knowledge of variants and genes that play a role in inherited susceptibility of this devastating disease.
- •
Although research gaps remain, research reported in this review has advanced the understanding of pancreatic cancer.
Epidemiology for pancreatic adenocarcinoma
Descriptive Epidemiology
Although pancreatic cancer accounts for less than 3% of cancer incidence, it ranks seventh for cancer mortality globally and fourth in the United States. Annually an estimated 338,000 and 46,420 people will be diagnosed with pancreatic cancer and 331,000 and 39,590 will die from the disease, worldwide and in the United States, respectively. There is no effective screening test for the malignancy; therefore, it is most often diagnosed at advanced stages with metastatic disease, which contributes to its high fatality with a mortality to incidence ratio of 0.98. The 5-year survival rate is 6.7%, which is poorer than that of other cancers. Only 9% of pancreatic cancer cases are diagnosed with localized disease when surgical resection may be an option for a cure. Those with localized disease have a 5-year survival of 25.8%. More than 90% of pancreatic cancers are ductal adenocarcinomas, with neuroendocrine tumors constituting about 5%.
Internationally, rates of pancreatic cancer vary by 7-fold, with higher rates in developed than in developing countries. In part the variation in incidence and mortality patterns worldwide may be explained by underascertainment of the disease and imperfect mortality data. Pancreatic cancer occurs slightly more often in men than in women and in African Americans than in Caucasians and other ethnicities in the United States. Its incidence and mortality rates increase with age. More than 88% of pancreatic cancer is diagnosed among people aged 55 years or older, with the median age at diagnosis of 71 years. In the United States, incidence and mortality rates for pancreatic cancer either decreased or remained stable during the latter part of the twentieth century but increased slightly since 2000 ( Figs. 1 and 2 ). Given the changing demographics in the United States, namely, the aging population and minority distribution, one recent study has projected that the number of pancreatic cancer deaths will surpass that of breast and colorectal cancer by the year 2030.
Risk Factors
Diabetes
Diabetes is associated with approximately a 2-fold risk for pancreatic cancer overall. Excess risks tend to be particularly high (2.5- to 10-fold) for diabetes diagnosed within 5 years of pancreatic cancer diagnosis and become attenuated with long duration of diabetes. A large pooled case-control study showed that the association with pancreatic cancer remained significant among those having diabetes for 20 years or more; however, the excess risk was diminished to 30%.
Biomarkers for diabetes and insulin resistance
Although diabetes as defined by fasting glucose, glucose intolerance, or hemoglobin A(1c), a marker for glucose control, has consistently been associated with pancreatic cancer in prospective epidemiologic studies, studies of other biomarkers hypothesized to mediate the diabetes-pancreatic cancer relationship have shown no association or inconsistent results ( Table 1 ). Insulin is a mitogen and has growth-promoting activity on pancreatic cancer cells, and patients with type 2 diabetes exhibit hyperinsulinemia during the early stages of the disease. Two studies that examined insulin and subsequent risk of pancreatic cancer showed significant elevated risks that were graded with significant trends across increasing categories of insulin. In both studies, the associations were stronger among cases occurring 10 years or more after their blood collection. These findings support that the positive associations observed among the later occurring cases are not the consequence of subclinical pancreatic cancer and that insulin resistance may play a role in pancreatic cancer etiology.
| Biomarker | Number of Studies | Association for Higher Concentration Compared with Lower Concentration a |
|---|---|---|
| HbA(1c) | 2 | ↑↑ |
| Insulin or proinsulin | 2 | ↑↑ Overall, stronger among cases occurring >10 y after blood collection |
| C-peptide | 2 | ↑↑ But not significant |
| Adiponectin (high) | 3 | ↓↓ Null |
| IGF-1, IGFBP-3 | 4 | Null |
| IGFBP-1 (low) | 1 | ↑ |
| IGF-2 | 1 | Null |
| sRAGE | 2 | ↓ Null |
| CML-AGE | 2 | Null |
| C-reactive protein | 4 | Null or ↓ in male smokers |
| IL6, sTNF-R1, sTNF-R2 | 2 | Null (↑sTNF-R1 but not significant in women only) |
| TGF-B | 1 | ↑ Among cases occurring >10 y after blood collection |
A nested case-control study of 4 cohorts examined the association between 83 metabolites and pancreatic cancer. The investigators observed significant positive associations between plasma branched-chain amino acids (ie, isoleucine, leucine, and valine) and subsequent risk of pancreatic cancer in both humans and in a mutant KRAS mouse model. The associations were independent of body mass index (BMI) and biomarkers related to insulin resistance and diabetes. Circulating branched-chain amino acid levels are elevated with obesity, glucose intolerance, and insulin resistance. The strongest association for these metabolites in this nested case-control study, however, was among subjects with samples collected between 2 and 5 years before pancreatic cancer diagnosis. The investigators hypothesized that this association was explained by breakdown of tissue protein that occurs in early-stage pancreatic cancer.
These studies support diabetes being an early manifestation of the malignancy as well as a risk factor involved in the cause of pancreatic cancer.
Pancreatitis and type 3c diabetes
Chronic pancreatitis is a known risk factor for pancreatic cancer, with risks ranging from 2- to 3-fold. The risk is particularly strong for familial pancreatitis. A large pooled analysis of 10 case-control studies estimated pancreatitis to account for approximately 1.34% of pancreatic cancer.
Type 3c diabetes (T3cDM) due to exocrine pancreatic disease has been recognized to play a role in pancreatic carcinogenesis. T3cDM is estimated to account for 5% to 10% of diabetes in Western populations. Patients with T3cDM have pancreatic inflammation, with 78% of patients having pancreatitis and 8% pancreatic cancer. Although pancreatic cancer contributes to T3cDM, it is also plausible that the pancreatitis associated with T3cDM participates in the development of the malignancy.
Maturity-onset diabetes of the young
Maturity-onset diabetes of the young (MODY) is estimated to account for 1% to 5% of all diabetes. Although MODY is typically diagnosed in young adults, it can be diagnosed at older ages also. MODY is caused by autosomal dominant gene mutations that alter islet beta-cell function.
Loci in genes linked to MODY have been associated with pancreatic cancer in studies using genome-wide genotyping data (see later section). In particular, common variants in 3 genes HNF1A, HNF1B, and PDX1 were associated with pancreatic cancer in individuals of European descent. These genes are also important components for transcriptional networks governing embryonic pancreatic development and differentiation, in addition to maintaining pancreatic homeostasis in adults.
Antidiabetic drugs
Several studies have examined the association between antidiabetic drugs and pancreatic cancer. Metformin works by suppressing glucose production in the liver, so there has been a particular interest in whether metformin has antitumor effects by decreasing circulating insulin concentrations. In contrast to metformin, sulfonylureas lower blood glucose levels by increasing the release of insulin from the pancreas. A large well-conducted case-control study that used drug prescription and other data derived from the UK General Practice Research Database showed that compared with those with no prior respective drug treatment, no associations were observed for metformin use, whereas significant positive associations were observed overall for long duration of sulfonylureas and insulin use as quantified by the number of recorded prescriptions. The same pattern of association was observed when this analysis was restricted to participants with a history of diabetes. A meta-analysis of clinical trials and observational studies similarly showed no association between metformin use and pancreatic cancer. In contrast, a large pooled case-control study showed a significant inverse association with long duration (15 years or more) of oral antidiabetic drug use and positive association for less than 10 years duration of insulin treatment compared with their respective nontreatment groups. A limitation of these studies is that the referent group for the observation studies includes participants receiving treatment of the other antidiabetic drugs, which could potentially influence the individual drug risk estimates.
Modifiable risk factors
Tobacco use
Current cigarette smoking is the most consistent potentially modifiable factor associated with pancreatic cancer with a close to a 2-fold elevated risk observed in most studies and has been estimated to account for up to 24% of pancreatic cancer. Compared with never smokers, the risk starts to decrease 10 years after smoking cessation, with the risk approaching that of never smokers 15 or more years after smoking cessation. Cigar and pipe smoking and smokeless tobacco are also associated with increased risk for pancreatic cancer in some but not all studies.
Overweight and obesity
Obesity is known to contribute to insulin resistance and type 2 diabetes. Most prospective studies show positive associations between middle-aged and older adult BMI and pancreatic cancer, with risk stronger among nonsmokers. Cigarette smoke contains a multitude of carcinogens, so it is possible that obesity does not play a substantial role in the cause of pancreatic cancer in smokers. A meta-analysis of 23 prospective cohort studies, totaling 9500 cases, showed that per 5 unit increase in BMI was associated with a significant 10% increased risk for pancreatic cancer. There was some evidence for a nonlinear relationship between BMI and pancreatic cancer, with the most pronounced increased risk being among participants with a BMI greater than 35 kg/m 2 . Waist circumference and waist to hip ratio are measures of abdominal adiposity that have consistently been associated with type 2 diabetes and are also associated with pancreatic cancer.
Adiposity at younger ages and across a lifetime in the same individuals and risk of pancreatic cancer has not been examined extensively. Ten studies, 5 case-control and 5 prospective, examined BMI during adolescence or early adulthood and risk of pancreatic cancer, with 6 showing statistically significant positive associations with excess risks ranging from 20% to 2-fold. Three studies showed significant positive associations between adiposity in early, middle, or older age in the same individuals and subsequent pancreatic cancer. One large prospective study of AARP members indicated longer duration of overweight was significantly associated with pancreatic cancer such that for every 10 years of having a BMI 25 kg/m 2 or more, pancreatic cancer risk increased 6% overall and 18% among those diagnosed with diabetes. The association was not as strong among participants without diabetes. This pattern of association with long duration of adiposity and diabetes supports the notion that diabetes may be on the causal pathway between obesity and pancreatic cancer. The proportion of pancreatic cancer explained by ever being overweight or obese at any age was substantial in the AARP population, 14% overall and 21% in never smokers.
Although weight change has inconsistently been associated with pancreatic cancer, one pooled study of 14 cohorts showed that compared with subjects who maintained their BMI from early adulthood to older age, those who gained more than 10 kg/m 2 had a significantly elevated risk of pancreatic cancer. The National Institutes of Health AARP study showed that a BMI gain of more than 10 kg/m 2 after age 50 years was associated with pancreatic cancer but not gain earlier in life. A BMI gain of 10 kg/m 2 corresponds to an increase from normal weight to obese.
Maintaining a normal BMI and avoiding adiposity at any age may prevent pancreatic cancer.
Dietary exposures
The low incidence and rapid fatality of pancreatic cancer have made it difficult for epidemiologists to examine dietary risk factors. Pancreatic cancer cases often have gastrointestinal problems, weight loss, or diabetes before being diagnosed because of subclinical cancer, which can influence dietary intake and related biomarkers. Retrospective ascertainment of diet for this cancer site in particular may have biases including control selection, recall, and proxy reporting that can result in inaccurate risk estimates. Prospective studies are less likely subject to these methodological problems. Dietary exposures are also subject to measurement error. Large prospective studies, including pooled efforts that harmonize data, have provided adequately powered analyses to detect small magnitudes of risk that are typical of nutritional exposures.
Among dietary exposures, coffee consumption is consistently not associated with pancreatic cancer. There is suggestive evidence for heavy alcohol use (>3 drinks per day) and high red meat, total and saturated fat, and fructose intake playing a role in pancreatic cancer. Some but not all studies have shown protective associations between healthier lifestyle and dietary patterns and pancreatic cancer.
Heavy alcohol use is the most common cause of both acute and chronic pancreatitis, and high levels of alcohol use could plausibly contribute to pancreatic cancer development via pancreatitis. Large studies have observed 35% to 80% increased risk between heavy alcohol use and pancreatic cancer. Individuals who are heavy drinkers often tend to also smoke cigarettes, and there is a concern about residual confounding by smoking. One large prospective study with more than 6500 cases that was conducted for 24 years showed a significant increased pancreatic cancer risk for participants who consumed more than 3 drinks per day compared with nondrinkers, with the association remaining significant in never smokers.
Several nested case-control studies that examined the association between vitamin D status and subsequent pancreatic cancer have shown varying results. 25-Hydroxyvitamin D [25(OH)D] is the best measure of vitamin D exposure, as it represents that from both the diet and that synthesized endogenously from solar ultraviolet B light exposure. A large pooled analysis of 8 cohorts showed a statistically significant 2-fold increased risk among participants with circulating 25(OH)D concentrations greater than 100 nmol/L compared with those with concentrations of 50 to 75 nmol/L and no association for concentrations less than 50 nmol/L. A subsequent pooled study using comparable clinically relevant 25(OH)D cut points and referent category observed a statistically significant 40% increased risk for participants with concentrations 37.5 to less than 50 nmol/L but did not observe associations for participants with concentrations less than 37.5 nmol/L or greater than or equal to 75 nmol/L.
Other nutrition-related exposures, such as glycemic load and index; total carbohydrate, sucrose, fiber, folate, vitamin C, vitamin E, sweetened beverage, soft drink, and cholesterol intake; and folate and multivitamin supplement use, and physical activity are not consistently associated with pancreatic cancer.
Epidemiology for pancreatic adenocarcinoma
Descriptive Epidemiology
Although pancreatic cancer accounts for less than 3% of cancer incidence, it ranks seventh for cancer mortality globally and fourth in the United States. Annually an estimated 338,000 and 46,420 people will be diagnosed with pancreatic cancer and 331,000 and 39,590 will die from the disease, worldwide and in the United States, respectively. There is no effective screening test for the malignancy; therefore, it is most often diagnosed at advanced stages with metastatic disease, which contributes to its high fatality with a mortality to incidence ratio of 0.98. The 5-year survival rate is 6.7%, which is poorer than that of other cancers. Only 9% of pancreatic cancer cases are diagnosed with localized disease when surgical resection may be an option for a cure. Those with localized disease have a 5-year survival of 25.8%. More than 90% of pancreatic cancers are ductal adenocarcinomas, with neuroendocrine tumors constituting about 5%.
Internationally, rates of pancreatic cancer vary by 7-fold, with higher rates in developed than in developing countries. In part the variation in incidence and mortality patterns worldwide may be explained by underascertainment of the disease and imperfect mortality data. Pancreatic cancer occurs slightly more often in men than in women and in African Americans than in Caucasians and other ethnicities in the United States. Its incidence and mortality rates increase with age. More than 88% of pancreatic cancer is diagnosed among people aged 55 years or older, with the median age at diagnosis of 71 years. In the United States, incidence and mortality rates for pancreatic cancer either decreased or remained stable during the latter part of the twentieth century but increased slightly since 2000 ( Figs. 1 and 2 ). Given the changing demographics in the United States, namely, the aging population and minority distribution, one recent study has projected that the number of pancreatic cancer deaths will surpass that of breast and colorectal cancer by the year 2030.
Risk Factors
Diabetes
Diabetes is associated with approximately a 2-fold risk for pancreatic cancer overall. Excess risks tend to be particularly high (2.5- to 10-fold) for diabetes diagnosed within 5 years of pancreatic cancer diagnosis and become attenuated with long duration of diabetes. A large pooled case-control study showed that the association with pancreatic cancer remained significant among those having diabetes for 20 years or more; however, the excess risk was diminished to 30%.
Biomarkers for diabetes and insulin resistance
Although diabetes as defined by fasting glucose, glucose intolerance, or hemoglobin A(1c), a marker for glucose control, has consistently been associated with pancreatic cancer in prospective epidemiologic studies, studies of other biomarkers hypothesized to mediate the diabetes-pancreatic cancer relationship have shown no association or inconsistent results ( Table 1 ). Insulin is a mitogen and has growth-promoting activity on pancreatic cancer cells, and patients with type 2 diabetes exhibit hyperinsulinemia during the early stages of the disease. Two studies that examined insulin and subsequent risk of pancreatic cancer showed significant elevated risks that were graded with significant trends across increasing categories of insulin. In both studies, the associations were stronger among cases occurring 10 years or more after their blood collection. These findings support that the positive associations observed among the later occurring cases are not the consequence of subclinical pancreatic cancer and that insulin resistance may play a role in pancreatic cancer etiology.
| Biomarker | Number of Studies | Association for Higher Concentration Compared with Lower Concentration a |
|---|---|---|
| HbA(1c) | 2 | ↑↑ |
| Insulin or proinsulin | 2 | ↑↑ Overall, stronger among cases occurring >10 y after blood collection |
| C-peptide | 2 | ↑↑ But not significant |
| Adiponectin (high) | 3 | ↓↓ Null |
| IGF-1, IGFBP-3 | 4 | Null |
| IGFBP-1 (low) | 1 | ↑ |
| IGF-2 | 1 | Null |
| sRAGE | 2 | ↓ Null |
| CML-AGE | 2 | Null |
| C-reactive protein | 4 | Null or ↓ in male smokers |
| IL6, sTNF-R1, sTNF-R2 | 2 | Null (↑sTNF-R1 but not significant in women only) |
| TGF-B | 1 | ↑ Among cases occurring >10 y after blood collection |
A nested case-control study of 4 cohorts examined the association between 83 metabolites and pancreatic cancer. The investigators observed significant positive associations between plasma branched-chain amino acids (ie, isoleucine, leucine, and valine) and subsequent risk of pancreatic cancer in both humans and in a mutant KRAS mouse model. The associations were independent of body mass index (BMI) and biomarkers related to insulin resistance and diabetes. Circulating branched-chain amino acid levels are elevated with obesity, glucose intolerance, and insulin resistance. The strongest association for these metabolites in this nested case-control study, however, was among subjects with samples collected between 2 and 5 years before pancreatic cancer diagnosis. The investigators hypothesized that this association was explained by breakdown of tissue protein that occurs in early-stage pancreatic cancer.
These studies support diabetes being an early manifestation of the malignancy as well as a risk factor involved in the cause of pancreatic cancer.
Pancreatitis and type 3c diabetes
Chronic pancreatitis is a known risk factor for pancreatic cancer, with risks ranging from 2- to 3-fold. The risk is particularly strong for familial pancreatitis. A large pooled analysis of 10 case-control studies estimated pancreatitis to account for approximately 1.34% of pancreatic cancer.
Type 3c diabetes (T3cDM) due to exocrine pancreatic disease has been recognized to play a role in pancreatic carcinogenesis. T3cDM is estimated to account for 5% to 10% of diabetes in Western populations. Patients with T3cDM have pancreatic inflammation, with 78% of patients having pancreatitis and 8% pancreatic cancer. Although pancreatic cancer contributes to T3cDM, it is also plausible that the pancreatitis associated with T3cDM participates in the development of the malignancy.
Maturity-onset diabetes of the young
Maturity-onset diabetes of the young (MODY) is estimated to account for 1% to 5% of all diabetes. Although MODY is typically diagnosed in young adults, it can be diagnosed at older ages also. MODY is caused by autosomal dominant gene mutations that alter islet beta-cell function.
Loci in genes linked to MODY have been associated with pancreatic cancer in studies using genome-wide genotyping data (see later section). In particular, common variants in 3 genes HNF1A, HNF1B, and PDX1 were associated with pancreatic cancer in individuals of European descent. These genes are also important components for transcriptional networks governing embryonic pancreatic development and differentiation, in addition to maintaining pancreatic homeostasis in adults.
Antidiabetic drugs
Several studies have examined the association between antidiabetic drugs and pancreatic cancer. Metformin works by suppressing glucose production in the liver, so there has been a particular interest in whether metformin has antitumor effects by decreasing circulating insulin concentrations. In contrast to metformin, sulfonylureas lower blood glucose levels by increasing the release of insulin from the pancreas. A large well-conducted case-control study that used drug prescription and other data derived from the UK General Practice Research Database showed that compared with those with no prior respective drug treatment, no associations were observed for metformin use, whereas significant positive associations were observed overall for long duration of sulfonylureas and insulin use as quantified by the number of recorded prescriptions. The same pattern of association was observed when this analysis was restricted to participants with a history of diabetes. A meta-analysis of clinical trials and observational studies similarly showed no association between metformin use and pancreatic cancer. In contrast, a large pooled case-control study showed a significant inverse association with long duration (15 years or more) of oral antidiabetic drug use and positive association for less than 10 years duration of insulin treatment compared with their respective nontreatment groups. A limitation of these studies is that the referent group for the observation studies includes participants receiving treatment of the other antidiabetic drugs, which could potentially influence the individual drug risk estimates.
Modifiable risk factors
Tobacco use
Current cigarette smoking is the most consistent potentially modifiable factor associated with pancreatic cancer with a close to a 2-fold elevated risk observed in most studies and has been estimated to account for up to 24% of pancreatic cancer. Compared with never smokers, the risk starts to decrease 10 years after smoking cessation, with the risk approaching that of never smokers 15 or more years after smoking cessation. Cigar and pipe smoking and smokeless tobacco are also associated with increased risk for pancreatic cancer in some but not all studies.
Overweight and obesity
Obesity is known to contribute to insulin resistance and type 2 diabetes. Most prospective studies show positive associations between middle-aged and older adult BMI and pancreatic cancer, with risk stronger among nonsmokers. Cigarette smoke contains a multitude of carcinogens, so it is possible that obesity does not play a substantial role in the cause of pancreatic cancer in smokers. A meta-analysis of 23 prospective cohort studies, totaling 9500 cases, showed that per 5 unit increase in BMI was associated with a significant 10% increased risk for pancreatic cancer. There was some evidence for a nonlinear relationship between BMI and pancreatic cancer, with the most pronounced increased risk being among participants with a BMI greater than 35 kg/m 2 . Waist circumference and waist to hip ratio are measures of abdominal adiposity that have consistently been associated with type 2 diabetes and are also associated with pancreatic cancer.
Adiposity at younger ages and across a lifetime in the same individuals and risk of pancreatic cancer has not been examined extensively. Ten studies, 5 case-control and 5 prospective, examined BMI during adolescence or early adulthood and risk of pancreatic cancer, with 6 showing statistically significant positive associations with excess risks ranging from 20% to 2-fold. Three studies showed significant positive associations between adiposity in early, middle, or older age in the same individuals and subsequent pancreatic cancer. One large prospective study of AARP members indicated longer duration of overweight was significantly associated with pancreatic cancer such that for every 10 years of having a BMI 25 kg/m 2 or more, pancreatic cancer risk increased 6% overall and 18% among those diagnosed with diabetes. The association was not as strong among participants without diabetes. This pattern of association with long duration of adiposity and diabetes supports the notion that diabetes may be on the causal pathway between obesity and pancreatic cancer. The proportion of pancreatic cancer explained by ever being overweight or obese at any age was substantial in the AARP population, 14% overall and 21% in never smokers.
Although weight change has inconsistently been associated with pancreatic cancer, one pooled study of 14 cohorts showed that compared with subjects who maintained their BMI from early adulthood to older age, those who gained more than 10 kg/m 2 had a significantly elevated risk of pancreatic cancer. The National Institutes of Health AARP study showed that a BMI gain of more than 10 kg/m 2 after age 50 years was associated with pancreatic cancer but not gain earlier in life. A BMI gain of 10 kg/m 2 corresponds to an increase from normal weight to obese.
Maintaining a normal BMI and avoiding adiposity at any age may prevent pancreatic cancer.
Dietary exposures
The low incidence and rapid fatality of pancreatic cancer have made it difficult for epidemiologists to examine dietary risk factors. Pancreatic cancer cases often have gastrointestinal problems, weight loss, or diabetes before being diagnosed because of subclinical cancer, which can influence dietary intake and related biomarkers. Retrospective ascertainment of diet for this cancer site in particular may have biases including control selection, recall, and proxy reporting that can result in inaccurate risk estimates. Prospective studies are less likely subject to these methodological problems. Dietary exposures are also subject to measurement error. Large prospective studies, including pooled efforts that harmonize data, have provided adequately powered analyses to detect small magnitudes of risk that are typical of nutritional exposures.
Among dietary exposures, coffee consumption is consistently not associated with pancreatic cancer. There is suggestive evidence for heavy alcohol use (>3 drinks per day) and high red meat, total and saturated fat, and fructose intake playing a role in pancreatic cancer. Some but not all studies have shown protective associations between healthier lifestyle and dietary patterns and pancreatic cancer.
Heavy alcohol use is the most common cause of both acute and chronic pancreatitis, and high levels of alcohol use could plausibly contribute to pancreatic cancer development via pancreatitis. Large studies have observed 35% to 80% increased risk between heavy alcohol use and pancreatic cancer. Individuals who are heavy drinkers often tend to also smoke cigarettes, and there is a concern about residual confounding by smoking. One large prospective study with more than 6500 cases that was conducted for 24 years showed a significant increased pancreatic cancer risk for participants who consumed more than 3 drinks per day compared with nondrinkers, with the association remaining significant in never smokers.
Several nested case-control studies that examined the association between vitamin D status and subsequent pancreatic cancer have shown varying results. 25-Hydroxyvitamin D [25(OH)D] is the best measure of vitamin D exposure, as it represents that from both the diet and that synthesized endogenously from solar ultraviolet B light exposure. A large pooled analysis of 8 cohorts showed a statistically significant 2-fold increased risk among participants with circulating 25(OH)D concentrations greater than 100 nmol/L compared with those with concentrations of 50 to 75 nmol/L and no association for concentrations less than 50 nmol/L. A subsequent pooled study using comparable clinically relevant 25(OH)D cut points and referent category observed a statistically significant 40% increased risk for participants with concentrations 37.5 to less than 50 nmol/L but did not observe associations for participants with concentrations less than 37.5 nmol/L or greater than or equal to 75 nmol/L.
Other nutrition-related exposures, such as glycemic load and index; total carbohydrate, sucrose, fiber, folate, vitamin C, vitamin E, sweetened beverage, soft drink, and cholesterol intake; and folate and multivitamin supplement use, and physical activity are not consistently associated with pancreatic cancer.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
