Pancreatic Cancer Treatments & Prevention

Pancreatic cancer is the fourth leading cause of cancer death in both men and women in the United States and will be responsible for an estimated 28,900 deaths in 2001. Relatively little is known of its etiology, and the only well-established risk factor is cigarette smoking. Studies over the past 3 decades have shown that 4%–16% of patients with pancreatic cancer have a family history of the disease. A small fraction of this aggregation can be accounted for in inherited cancer syndromes, including familial atypical multiple-mole melanoma, Peutz-Jeghers syndrome, hereditary breast-ovarian cancer, hereditary pancreatitis, and hereditary nonpolyposis colorectal cancer. These syndromes arise as a result of germline mutations in the BRCA2,  (familial atypical multiple-mole melanoma), mismatch repair (hereditary nonpolyposis colorectal cancer), and STK11 (Peutz-Jeghers syndrome) genes. In addition, hereditary plays a role in predisposing certain patients with apparently sporadic pancreatic cancer. Many patients with pancreatic cancers caused by a germline mutation in a cancer-causing gene do not have a pedigree that is suggestive of a familial cancer syndrome. A recent prospective analysis of the pedigrees in the National Familial Pancreatic Tumor Registry found that individuals with a family history of pancreatic cancer in multiple first-degree relatives have a high risk of pancreatic cancer themselves. The identification of such high-risk individuals will help clinicians target screening programs and develop preventive interventions with the hope of reducing the mortality of pancreatic cancer in these families.

Familial pancreatic cancer, BRCA2, p16, STK11

Pancreatic cancer is the fourth leading cause of cancer death in men and women in the United States. There will be an estimated 28,900 deaths due to pancreatic cancer in 2001. Worldwide, pancreatic cancer accounts for 2% of all cancer diagnoses. However, the etiology of pancreatic cancer remains poorly understood. A familial predisposition to pancreatic cancer has gradually been recognized over the past 25 years, first from case reports, then through epidemiological studies, and then through genetic analysis of families and the analysis of family-based registry data. More recently, several observational epidemiological studies have shown that individuals with a strong family history of pancreatic cancer are at increased risk of having pancreatic cancer themselves. Much has already been learned about the genetic changes involved in the development of pancreatic cancer through the study of sporadic (nonfamilial) pancreatic tumors. In particular, the genes targeted in these sporadic forms of cancers are also (when inherited in the germline) often the cause of familial cancers. For example, the p16 gene is somatically inactivated in virtually all sporadic pancreatic cancers, and germline p16 mutations are associated with a 20-fold increase in risk of pancreatic cancer. The identification and study of individuals with familial pancreatic cancer will help quantify cancer risk and form the basis for the development of screening and chemoprevention strategies for this disease.

EPIDEMIOLOGY AND BIOLOGY

Pancreas cancer accounts for 2% of all cancer diagnoses worldwide, approximately 185,000 new cases per year. In the United States, the overall age-adjusted incidence rate is 9/100,000. There is little (no more than twofold) variation in the incidence of pancreatic cancer worldwide. The incidence of pancreatic cancer is higher in Northern European than in Southern European nations and is higher in developed than in developing countries. However, a significant portion of this difference in incidence between developed and developing nations can be attributed to the difficulty in diagnosing this disease, which causes an underestimate of pancreatic cancer rates in many developing nations that lack appropriate diagnostic technology. The 5-year survival rate for pancreatic adenocarcinoma is 4%, the poorest of any major cancer, and has largely remained unchanged for decades. Pancreatic adenocarcinoma rarely occurs before the age of 40, and is more than 17 times more likely to develop in individuals older than 65 years than in those younger than 65. Males have a slightly higher risk (~1.5-fold) of developing pancreatic cancer than females; however, this difference between men and women decreases with increasing age. The age-adjusted incidence of pancreatic cancer in African-Americans is higher (13.6/100,000) than it is in Caucasians (9.8.100,000). United States Asian and Hispanic groups have lower rates at around 7.3 per 100,000, and the reported rate for American Indians is 4.5 per 100,000. Several countries, including the United States and Japan, have seen a dramatic increase in the incidence of pancreatic cancer since the 1930s. However, much of this increase can be attributed to improvements in diagnosis and the increase in cigarette smoking.

In addition to familial factors, numerous risk factors for pancreatic cancer have been investigated, and increased risk is associated with predisposing conditions, such as type II diabetes and chronic pancreatitis, and certain occupational chemical exposures. However, the most consistent and well-established environmental risk factor is cigarette smoking, which has been estimated to account for 25%–30% of this disease. The risk of pancreatic cancer in current smokers is about two to three times higher than that of nonsmokers. This increase in risk is present for current smokers only, and it rises with increasing cigarette consumption. Smoking cessation can result in as much as 48% reduction in risk within the first 2 years after quitting, and after about 10 years, the risk in former smokers is equivalent to that of those who have never smoked. Additionally, many studies have shown that diabetes mellitus can increase the risk of pancreatic cancers. However, diabetes mellitus develops in many pancreatic cancer patients as a complication of their pancreatic cancers. A meta-analysis indicated that diabetes mellitus of 5 or more years in duration is associated with a 2.0-fold (95% confidence interval [CI] 1.3–3.2) increased risk of developing pancreatic cancer.

CASE REPORTS OF FAMILIAL PANCREATIC CANCER

The first studies to explore what is now referred to as familial pancreatic cancer were case reports of families in which there were multiple cases of pancreatic cancer. In 1973, MacDermott and Kramer described a family in which pancreatic cancer developed in four of six siblings. In 1995, Evans and colleagues reported on a large family that had nine cases of pancreatic cancer among 50 individuals in four generations. Five of the nine individuals had a diagnosis of insulin-dependent diabetes mellitus at least 10 years before their diagnoses of pancreatic cancer, and six of the nine had a medical history of exocrine pancreatic insufficiency before diagnosis. Several studies have estimated the proportion of individuals with pancreatic cancer that report a positive family history of pancreatic cancer to be between 3.9% and 16%

PANCREATIC CANCER FAMILY REGISTRIES

One of first registries established to study families with multiple cases of pancreatic cancer was described by Lynch et al, who reported on 18 families with at least two first-degree relatives affected with pancreatic cancer. Overall, there were 47 cases of pancreatic cancer within these 18 families. When these 47 cases were compared with the United States population—based Surveillance, Epidemiology and End Results (SEER) registry data, they found that the age of onset of the familial cases was quite similar to that of pancreatic cancer in general. SEER study sites include Atlanta, Detroit, Los Angeles, Iowa, New Mexico, Seattle, Connecticut, San Francisco, and Utah. Additionally, Lynch et al reported that the proportion of ductal adenocarcinomas, survival time, and gender distribution in their familial cases were quite similar to those reported by SEER. In this regard, pancreatic cancer is unlike most other hereditary cancer syndromes, such as breast and colorectal cancers, in that individuals with hereditary forms of these cancers typically present at an earlier age.

The National Familial Pancreatic Tumor Registry (NFPTR) was established at Johns Hopkins in 1994 and is now probably the largest registry of familial pancreatic cancer, with more than 680 families enrolled from all over the world, including the United States, Europe, and Australia. More than 270 of these families have familial pancreatic cancer, defined as pancreatic cancer in an individual with at least a pair of first-degree relatives affected with pancreatic cancer. The NFPTR has investigated various research questions to help understand the causes of familial pancreatic cancer.

PROSPECTIVE RISK OF PANCREATIC CANCER AMONG PANCREATIC CANCER KINDREDS

A recent prospective study of the first 341 families enrolled in the NFPTR demonstrated that first-degree relatives of the probands in familial pancreatic cancer kindreds (defined as at least a pair of first-degree relatives affected with pancreatic cancer in the kindreds) have an increased risk of developing pancreatic cancer when compared with what is expected on the basis of SEER incidence rates. In this study, 150 families with familial pancreatic cancer and 191 families with sporadic pancreatic cancer (kindreds with a pancreatic cancer but without a pair of affected first-degree relatives) were followed up to study the prospective risk of pancreatic cancer. Some sporadic pancreatic cancer kindreds did have pancreatic cancer reported in more distantly related members. These NFPTR families were followed up from their date of enrollment in the registry (January 1, 1994 through September 1, 1998) until the study closed (December 31, 1998). Overall, nine incident (prospective) cases of pancreatic cancer developed in 1240 at-risk first-degree relatives who contributed 2683 person-years of follow-up over the observation period. Seven of the nine cases occurred among 598 at-risk first-degree relatives in familial pancreatic cancer kindreds. The other two new cases occurred among the 642 at-risk first-degree relatives in sporadic pancreatic cancer kindreds. The incidence of new pancreatic cancer in these two groups of families was then compared with the expected incidence of pancreatic cancer based on SEER data. Age- and sex-specific rates were used. Sporadic pancreatic cancer kindreds had a higher number of incident pancreatic cancers than expected, with an observed-to-expected ratio (O : E) of 6.5 (95% CI 0.78–23.3), although this was not statistically significant. However, familial pancreatic cancer kindreds had a statistically higher number of incident pancreatic cancer cases than expected (O : E = 18.3, 95% CI 4.74–44.5). Estimated incidence rates were 24.5 per 100,000 in sporadic pancreatic cancer families and 76 per 100,000 in familial pancreatic cancer families. When the analysis was limited to familial pancreatic cancer families with three or more affected first-degree relatives, the O : E increased to 56.6 (95% CI 12.4–175.0), and the incidence rate was estimated to be 301.4 per 100,000.[ 36] This study provided the first quantified estimate of prospective risk of pancreatic cancer in the familial setting and established that apparently healthy relatives of patients with familial pancreatic cancer have a significantly increased risk of developing pancreatic cancer themselves.

CASE-CONTROL AND COHORT STUDIES

Recently, several studies have shown that a family history of any cancer and a family history of pancreatic cancer are risk factors for pancreatic cancer. The first of these studies was published in 1988 by Falk et al, who performed a case-control study in Louisiana, a region with a particularly high incidence of pancreatic cancer. They found an increased risk of pancreatic cancer in individuals who reported a close relative with pancreatic cancer (odds ratio [OR] = 5.25 [95% CI: 2.08–13.21]). A subsequent study by Ghadirian et al found that 7.8% of pancreatic cancer cases and only 0.6% of controls had a family history of pancreatic cancer, a 13-fold difference. There were no differences in environmental risk factors between the two groups. Fernandez et al conducted a case-control study in Italy and found an increased risk of pancreatic cancer in individuals who reported a relative with pancreatic cancer, after the investigators adjusted for tobacco, dietary factors, and history of diabetes or pancreatitis (OR = 2.8 [95% CI 1.3–6.3]). Additionally, individuals with a close relative with any cancer had a 1.86 odds (95% CI 1.42–2.44) of developing pancreatic cancer, as compared with those without a family history. A recent study of 484 cases and 2099 controls, ascertained through population-based registries in three regions of the United States (Atlanta, Detroit, and New Jersey), found that individuals with a first-degree relative of pancreatic cancer had an OR of 3.2 (95% CI 1.8–5.6), and the risk was higher (OR = 3.6; 95% CI 1.5–8.7) in those with an affected sibling than in those with an affected parent (OR = 2.6; 95% CI 1.2–5.4). A higher risk due to family history was apparent in individuals who smoked for more than 20 years (OR = 5.3; 95% CI 2.1–13.4) compared with individuals who did not smoke or smoked for less than 20 years (OR = 2.2; 95% CI 1.0–7.9). However, there did not appear to be a significant interaction between smoking and family history. Additionally, risk due to family history did not vary by the age at onset of the cases. Schenk et al studied 247 cases and 420 controls and found a higher risk of pancreatic cancer in smokers with a history of pancreatic cancer in first-degree relatives (relative risk [RR] = 8.25; 95% CI 2.18–31.07) than in nonsmokers with a positive family history (RR = 2.15; 95% CI 0.63–7.27). Overall, they found that having a family history of pancreatic cancer increased the risk of pancreatic cancer 2.49-fold (95% CI 1.32–4.69).[ 37] Recently, as part of the American Cancer Society's Cancer Prevention Study II, Coughlin et al reported an analysis of 1,102,308 individuals, of whom 3751 developed pancreatic cancer during 14 years of follow-up. After investigators adjusted for age, the relative risk of developing pancreatic cancer in individuals who reported a positive family history of pancreatic cancer at baseline was 1.5 (95% CI 1.1–2.1). This risk estimate was not changed when it was further adjusted for history of gallstones, body mass index, smoking history, alcohol consumption, history of diabetes, and several dietary factors. These epidemiological studies establish the importance of a family history of pancreatic cancer even when other risk factors, including smoking, are accounted for, and emphasize that inherited factors do play a role in the development of some pancreatic cancers.

GENETIC MUTATIONS IN SPORADIC TUMORS

More than 90 percent of infiltrating pancreatic adenocarcinomas have activating mutations in the K-ras oncogene.[ 38] The p16 tumor suppressor gene is inactivated by somatic deletion, methylation, or mutation in more than 95% of sporadic pancreatic cancers. The DPC4 and p53 tumor suppressor genes are also frequently somatically inactivated in sporadic pancreatic adenocarcinoma, p53 encodes for a transcription factor that is activated in response to DNA damage and is inactivated in 50%–70% of pancreatic cancers. DPC4 is a tumor suppressor gene that codes for a protein that functions in the transforming growth factor-ß (TGF-ß) pathway and is inactivated in 55% of pancreatic cancers. Several other genetic alterations occur in pancreatic cancer, including inactivation of the BRCA2 (7%), LKB1/STK11 (4%), MKK4 (4%), TGFß-R1 and TGFß-R2 (<5% combined),ALK4 (<3%),[ and RB1 genes (<5%).

GENETIC SYNDROMES ASSOCIATED WITH PANCREATIC CANCER

Several genetic disorders are associated with an increased risk of pancreatic cancer . As Knudson demonstrated for retinoblastoma, individuals with a germline mutation in a tumor suppressor gene require only one mutational event, or “hit,” to induce tumor formation, whereas those without an inherited mutation require two hits, or two separate somatic events that inactivate both copies of the gene to induce tumor formation. Thus, an understanding of the aforementioned molecular alterations that occur in sporadic pancreatic cancers can provide clues to the genetic alterations that predispose some families to pancreatic cancer.

BRCA2

Germline mutations in the BRCA2 gene are associated with an increased risk of pancreatic cancer. BRCA2 is a tumor suppressor gene, and its protein product binds to the Rad51 protein, which is involved in the repair of DNA strand breaks. Murphy et al found that approximately 10% of patients with familial pancreatic cancer harbor germline mutations in BRCA2, and germline mutations in the BRCA2 gene have been identified in 7% of apparently sporadic pancreatic cancer. Remarkably, many individuals with germline BRCA2 mutations do not come from typical hereditary breast cancer kindreds, and some have no family cancer history at all. Individuals who inherit a germline BRCA2 mutation have a lifetime risk of developing pancreatic cancer that is about 10 times higher than that of the general population. These results highlight the fact that an inherited predisposition to developing pancreatic cancer in many patients with this disease cannot be identified from family history alone.

p16 and Familial Atypical Multiple-Mole Melanoma Syndrome

Familial atypical multiple-mole melanoma (FAMMM) is a rare inherited syndrome characterized by multiple nevi and melanomas and is caused by germline mutations in p16. The prevalence of germline p16 mutations is not known. However, one study has indicated that less than 10% of melanoma families had a germline p16 mutation. p16 is a tumor suppressor gene, and the protein product of this gene acts as cyclin-dependent kinase inhibitor. p16 is one of the most common targets for mutations in human cancers and is somatically inactivated in almost 100% of all pancreatic cancers. The wild-type p16 acts to arrest normal diploid cells in the G1 phase of the cell cycle. FAMMM kindreds also have an increased risk of pancreatic cancer. Goldstein et al[ 4] found that the lifetime risk of pancreatic cancer in individuals carrying a germline mutation in p16 (also called CDKN2A) is 13 to 22 times higher than the general population.

Hereditary Nonpolyposis Colorectal Cancer

Lynch et al reported that families with hereditary nonpolyposis colorectal cancer (HNPCC) have an increased risk of developing pancreatic cancer. HNPCC is typically caused by germline mutations in DNA mismatch repair genes, including hMSH2, hMLH1, hPMS1, hPMS2, and hMSH6. HNPCC is an autosomal dominant condition in which families have an increased risk of developing multiple cancers, especially of the colorectum, endometrium, and ureters. Although approximately 4%-11% of all resected pancreatic carcinomas have a defect in a DNA mismatch repair gene, most of these are caused by somatic, not inherited, mutations. Of interest, mismatch repair—deficient pancreatic carcinomas can often be recognized by their unique histology, the so-called medullary pathology.

Hereditary Pancreatitis

Hereditary pancreatitis is an autosomal dominant disorder that is associated with germline mutations in the cationic trypsinogen gene (PRSS1), although about 30% of families with hereditary pancreatitis do not have a mutation in this gene. The defining characteristic of this disease is recurrent episodes of pancreatitis, typically beginning at a young age, and often in the teenage years. Several studies have estimated that the lifetime risk of pancreatic carcinoma in individuals affected by hereditary pancreatitis is 40% by age 70, or 50 to 60 times higher than expected. Pancreatic carcinoma risk in this group has also been shown to be higher in patients who are smokers, with the onset of the cancer occurring about 20 years earlier in smokers than in nonsmokers. It is believed that pancreatic cancer develops in these individuals through the recurrent episodes of injury and repair to the pancreatic epithelial tissue. Injury to the pancreas occurs because the associated PRSS1 mutation makes the associated enzyme resistant to cleavage by trypsin, resulting in auto-activation of the enzyme and digestion of the pancreatic tissue.

Peutz-Jeghers Syndrome [SYK11]

Peutz-Jeghers syndrome, an autosomal dominant disorder characterized by hamartomatous polyps in the gastrointestinal tract and pigmented macules of the lips, buccal mucosa, and digits, has also been associated with an increased risk of several gastrointestinal cancers, including esophageal, stomach, small intestine, colon, lung, breast, uterine, ovarian, and pancreatic cancer. Peutz-Jeghers syndrome is caused by mutations in the LKB1/STK11 tumor suppressor gene, on chromosome 19p13, which encodes for a serine-threonine kinase. Little is known about the function of STK11.

A review of 210 individuals with Peutz-Jeghers syndrome indicated that the risk of pancreatic cancer was greatly increased (132-fold; 95% CI 44–261) compared with that of the general population. A recent study that combined 53 Peutz-Jeghers syndrome cases from four studies found that these patients had an 11% lifetime risk of developing pancreatic cancer. Genetic analysis of 100 resected pancreatic adenocarcinomas showed bi-allelic inactivation of the LKB1/STK11 gene in 4% of resected sporadic pancreatic adenocarcinomas. These studies demonstrate the importance of mutations in this gene to both the inherited and sporadic forms of pancreatic cancer.

SEGREGATION ANALYSIS

The familial clustering of pancreatic cancer observed overwhelmingly across different populations suggests that in many families, there is an inherited basis for their disease. Segregation is a statistical methodology aimed at assessing whether this observed familial aggregation is due to either the involvement of a major gene or shared environmental factors. A recent hospital-based segregation analysis of 287 pancreatic cancer cases and their families recruited from Johns Hopkins Hospital and Johns Hopkins Bayview Medical Center provided evidence for a major gene that influences the age at onset of pancreatic cancer in some families. Approximately 6/10,000 individuals were estimated to carry this high-risk genotype. This study excluded shared environmental factors as the cause of this familial aggregation. Cigarette consumption, defined as ever/never smoking, did not significantly alter this risk. Because of its established importance as a risk factor of pancreatic cancer, it is important that future studies continue to consider the influence of smoking in genetic causes of pancreatic cancer. Along with this study, future segregation analyses will provide a foundation on which linkage studies aimed at finding the genes responsible for this aggregation can be planned.

DISCUSSION AND FUTURE DIRECTIONS

In order to determine the basis of inherited pancreatic cancer, two main approaches are available to researchers. The candidate gene approach relies on sequencing germline DNA from patients with familial pancreatic cancer for mutations in selected genes. Gene are selected either because they are known to be somatically inactivated in pancreatic cancer or because they have been associated with other inherited cancer syndromes. Through this approach, we have learned about the importance of germline mutations in BRCA2 and p16 in predisposing some families to pancreatic cancer. However, some genes are only rarely inactivated somatically (i.e., BRCA1 and BRCA2), and some inherited tumor suppressor genes give rise to cancer only in a tissue-specific manner. We therefore must also look to other means of identifying novel tumor suppressor genes that may be involved in pancreatic cancer susceptibility. Many difficulties hinder efforts to identify these germline mutations by use of traditional means, including linkage studies. The low penetrance of cancer-predisposing genes can make many cases that are caused by an inherited predisposition appear to be sporadic. Further, the rapidly fatal course of pancreatic cancer and the frequent lack of information on the accuracy of diagnoses make identification of cases difficult. Therefore, the best chance for these linkage studies to succeed lies in the development of multicenter collaborations aimed at finding the genes responsible for familial aggregation of pancreatic cancer in high-risk families.

Future studies must also examine how shared environmental factors influence the risk of pancreatic cancer in these families, including potential gene and environment interactions. Finally, several centers have begun clinical investigations to determine the prevalence of pancreatic abnormalities in individuals at high risk for pancreatic cancer because of a strong family history of the disease (at least three affected first-degree relatives). Current studies rely on imaging methods, particularly endoscopic ultrasound. In preliminary studies, we have found silent neoplasia and mass lesions in asymptomatic high-risk individuals through the use of endoscopic ultrasonography (M. Canto, Johns Hopkins Hospital, personal communication, 2001). Several such high-risk individuals have undergone successful pancreatic resection. The morbidity and mortality from partial pancreatic resection are now relatively low (~2% mortality) in experienced centers. These results encourage hope that screening strategies for high-risk individuals followed by resection of those with asymptomatic cancers could reduce the mortality from this disease in high-risk individuals. Currently, early detection and surgery offer the only hope of survival for this disease.

Although we have learned much over the past several decades about the aggregation of pancreatic cancer in some families, the genetic basis for most of this familial aggregation is still poorly understood. Increasing our understanding of the genetic basis of this familial aggregation will improve our ability to provide accurate genetic counseling to these families. A better understanding will provide greater insight into the biologic basis of both familial and sporadic disease, which could lead to improved chemotherapy for pancreatic adenocarcinoma and ultimately to the development of preventive measures for this devastating disease.

Last updated Jan 2/07