Pancreatic cancer is the second-most-common gastrointestinal malignancy, and the fourth leading cause of cancer-related deaths in the United States. Less than 4% of patients are alive 5 years after diagnosis. Approximately 34,000 new cases were diagnosed in 2006, and 37,000 new cases were expected to be diagnosed in 2008.
The disease is more common in men than in women (1.3:1) and in certain ethnic and racial groups (eg, Blacks, Polynesians, and native New Zealanders). It is rare before the age of 45 years, but the incidence increases sharply after the seventh decade.
Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin.
Shaib YH, Davila JA, El-Serag HB. The epidemiology of pancreatic cancer in the United States: changes below the surface. Aliment Pharmacol Ther.
Pathogenesis & Risk Factors
Most pancreatic neoplasms arise from the three different types of the epithelial cells found in the pancreas. Acinar cells account for 80% of the volume of the gland but constitute 1% of exocrine tumors. Ductal cells constitute 10–15% of the volume but give rise to 90% of all tumors. Endocrine cells are 1–2% of volume and account for 1–2% of the tumors. Nonepithelial tumors are very rare.
Approximately 70% of ductal tumors are localized to the head of the pancreas, 5–10% to the body, and 10–15% to the tail. These tumors are hard in consistency due to the strong desmoplastic response they elicit.
Acinar tumors present as large pancreatic masses in the elderly, and distant metastasis is usually present at the time of diagnosis.
It has been proposed that pancreatic cancer develops from small intraductal precursor lesions (pancreatic intraepithelial neoplasia), following an evolution similar to the adenoma–carcinoma sequence seen in colorectal tumors. Although the precise mechanism and sequence of genetic mutations responsible for the development of pancreatic cancer remains unclear, genetic alterations found in these tumors can be classified into three categories: (1) activation of oncogenes; (2) inactivation of tumor suppressor genes; and (3) defect in DNA mismatch repair genes. The sonic hedgehog signaling pathway also appears to play a role. The sonic hedgehog gene, which is involved with embryonic development, appears to be unregulated in early and late stages of pancreatic carcinogenesis.
Mutations of the K-ras oncogene are seen in 90% of tumors and are the hallmark of pancreatic adenocarcinoma. Although this mutation may be seen in nonmalignant conditions such as chronic pancreatitis, it appears to be an early genetic alteration in pancreatic carcinogenesis.
Inactivation of Tumor Suppressor Genes
Inactivation and loss of function of tumor suppressor genes results in critical disruption of the cell cycle involving cellular differentiation, growth inhibition, regulation of transcription, DNA repair, and apoptosis. The genes most frequently involved are CDKN2A (95%), P53 (60%), DPC4 (50%), BRCA2, and STK11. About 10% of patients with hereditary pancreatic cancer harbor germline mutations of the BRCA2 gene.
Defect of DNA Mismatch Repair Gene
Mutations of mismatch repair genes, such as MLH1 and MSH2, have been found in 4% of pancreatic tumors.
Family History of Pancreatic Cancer
Genetic predisposition is the greatest risk factor for the development of pancreatic cancer. About 8–10% of patients with pancreatic cancer have a first-degree relative with the disease. These patients present at an earlier age, and smoking appears to contribute to the development of the cancer.
Hereditary Chronic Pancreatitis
This autosomal-dominant condition is strongly associated with pancreatic cancer although it accounts for only a small fraction of the total number of cases. It presents as recurrent attacks of acute pancreatitis early in life that may progress to chronic pancreatitis. The risk of an affected family member developing cancer is as high as 40% by age 70 and is highest among those who smoke.
The risk of pancreatic cancer is also increased in patients with certain familial cancer syndromes such as Peutz-Jeghers syndrome, ataxia–telangiectasia, familial adenomatous polyposis, and Lynch syndrome II.
Diabetes mellitus is associated with pancreatic cancer, and in most cases there is no family history. In most patients the diagnosis of pancreatic cancer is made within 2 years of the onset of diabetes mellitus. Although some studies have shown that high prediagnosis serum concentrations of glucose and insulin, along with insulin resistance, correlate with an increased risk of pancreatic cancer, other studies suggest that the tumor may be responsible for the diabetogenic state.
Environmental Risk Factors
The best-established environmental risk factor associated with pancreatic cancer is cigarette smoking. The relative risk of pancreatic cancer among current smokers is 2.5. The risk increases with the number of cigarettes consumed and returns to baseline 15 years after the patient stops smoking.
Diet appears to be another important environmental factor. Diets high in fat and meat appear to be linked to the development of pancreatic cancer, while consumption of fruits and vegetables seem to have a protective effect. Low levels of selenium and lycopene have been associated with the development of pancreatic cancer.
Data on coffee and alcohol consumption, use of aspirin and other nonsteroidal anti-inflammatory drugs, and the development of pancreatic cancer have been conflicting, and recent studies show no definite relationship. In a recent study, obesity significantly increased the risk of pancreatic cancer. Some studies have shown an association between Helicobacter pylori infection, particularly the Cag A strain, and pancreatic cancer.
Nonhereditary Risk Factors
The risk of pancreatic adenocarcinoma is about 4% in patients with nonhereditary chronic pancreatitis 20 years after disease onset.
Owing to the lack of characteristic signs and symptoms, most patients with pancreatic tumors present late in the course of the disease. As a result, less than 15% of tumors are resectable at the time of diagnosis. Even after surgical resection with negative margins, the 5-year survival rate is between 10% and 25%, with a median survival between 10 and 20 months. Patients may present with vague, low-intensity, dull abdominal discomfort or pain that radiates to the back and may be associated with weight loss, anorexia, weakness, diarrhea, and vomiting (Table 28–1). The pain is primarily due to the invasion of the celiac and superior mesenteric plexus.
Table 28–1. Clinical Manifestations that May Be Associated with Pancreatic Cancer. ||Download (.pdf)
Table 28–1. Clinical Manifestations that May Be Associated with Pancreatic Cancer.
|Symptoms and Signs||Laboratory Findings|
Abdominal pain or discomfort
Irritable bowel syndrome (diarrhea)
New-onset diabetes after age 50 y
New onset mood changes
↑ Serum alkaline phosphatase
↑ Erythrocyte sedimentation rate
↑ Serum CA 19-9
Location of the tumor also defines the symptoms and the prognosis. Tumors of the head of the pancreas produce symptoms early, and painless jaundice is seen in more than 50% of cases due to obstruction of the extrahepatic bile duct. In less than one third of patients, obstruction of the bile duct by pancreatic neoplasm is accompanied by a palpable, nontender gallbladder referred to as Courvoisier sign. This finding also may be seen in bile duct obstruction by cholangiocarcinoma, duodenal carcinoma, and carcinoma of the ampulla of Vater.
Tumors of the body and tail are either "asymptomatic" or manifest with nonspecific symptoms, such as abdominal discomfort, and the diagnosis is mostly made after metastatic disease has developed.
Obstruction of the pancreatic duct may lead to pancreatic exocrine insufficiency in the form of steatorrhea and malabsorption. New-onset diabetes mellitus after the age of 50 years has been associated with the development of pancreatic cancer, especially in the absence of a family history of diabetes. Other uncommon manifestations of pancreatic neoplasm include acute pancreatitis in the elderly without any obvious cause, thrombophlebitis, psychiatric disturbances, pruritus due to cholestasis, signs and symptoms of gastrointestinal bleeding, and obstruction due to erosion and growth of the pancreatic neoplasm into the duodenal lumen. In patients older than age 50, pancreatic cancer can present with features of irritable bowel syndrome or early-onset diabetes.
Laboratory Findings (See Table 28–1)
Malignant obstruction of the distal bile duct by a neoplasm of the pancreatic head characteristically produces an alkaline phosphatase level that is four to five times the upper limits of normal, and the increase is disproportionate to the bilirubin level until late in the course of the disease. The rise in transaminases is usually mild. Despite biliary stasis, cholangitis is uncommon. Lipase and amylase are elevated in tumors that cause pancreatic duct obstruction and present as acute pancreatitis.
CA 19-9 is a sialylated Lewis antigen that has been found to be clinically useful in both diagnosis and monitoring of the treatment response. The sensitivity and specificity, using a cutoff level of 37 units/mL, is 86% and 87%, respectively. Biliary tract obstruction with cholangitis due to a nonmalignant cause can result in high levels of CA 19-9. In one study, levels as high as 32,000 were reported.
Imaging and Other Diagnostic Studies (See Chapter 9)
Pancreatic adenocarcinoma is staged using a TNM classification (Table 28–2). It consists of evaluating the characteristics of the primary tumor, namely tumor size and infiltration into major vessels (T stage), regional lymph node involvement (N stage), and the presence and absence of distant metastasis (M stage). Various modalities are available for the diagnosis and staging of pancreatic tumors, including those described below.
Table 28–2. Staging of Pancreatic Exocrine Cancer. ||Download (.pdf)
Table 28–2. Staging of Pancreatic Exocrine Cancer.
|Definition of TNM|
|Primary Tumor (T)|
|TX||Primary tumor cannot be assessed|
|T0||No evidence of primary tumor|
|Tis||In situ carcinoma|
|T1||Tumor limited to the pancreas, ≤2 cm in greatest dimension|
|T2||Tumor limited to the pancreas, >2 cm in greatest dimension|
|T3||Tumor extends beyond pancreas but without involvement of celiac axis or superior mesenteric artery|
|T4||Tumor involves celiac axis or superior mesenteric artery (unresectable primary tumor)|
|Regional Lymph Nodes (N)|
|NX||Regional lymph nodes cannot be assessed|
|N0||No regional lymph node metastasis|
|N1||Regional lymph node metastasis|
|Distant Metastasis (M)|
|MX||Distant metastasis cannot be assessed|
|M0||No distant metastasis|
|Stage 0||Tis, N0, M0|
|Stage IA||T1, N0, M0|
|Stage IB||T2, N0, M0|
|Stage IIA||T3, N0, M0|
|Stage IIB||T1-3, N1, M0|
|Stage III||T4, any N, M0|
|Stage IV||Any T, any N, M1|
Computed Tomography (CT) Scan
Transabdominal ultrasound and the CT scan remain the most common imaging modalities used in patients with suspected pancreatic cancer. A pancreatic protocol helical CT scan is probably the best initial modality for diagnosis and staging of pancreatic adenocarcinoma. Based on the imaging features, a correct diagnosis of pancreatic cancer is made in more than 90% of cases. In addition to detecting metastasis, based on the involvement of adjacent organs, and vascular invasion, the helical CT scan can predict unresectability with more than 90% accuracy. The major drawback is that a significant proportion of patients found to have resectable disease on CT scan have unresectable disease at laparotomy due to small tumor implants in the liver and peritoneum.
Endoscopic Ultrasound (EUS)
EUS has a diagnostic sensitivity similar to helical CT scan but may be superior in diagnosing small pancreatic tumors, and portal and splenic vein invasion. EUS-guided fine needle aspiration (EUS–FNA) has a diagnostic sensitivity of about 85–90% with a false-negative rate of 10–15%. It is safe procedure with minimal risk of tumor seeding. EUS is less accurate in predicting superior mesenteric vein and superior mesenteric artery involvement by the tumor.
Endoscopic Retrograde Cholangiopancreatography (ERCP)
Pancreatic tumors appear as strictures of the pancreatic duct or the bile duct on ERCP. This stricturing of both the bile duct and the pancreatic duct is referred to as the "double duct sign." Advances in pancreatic imaging such as helical CT have made ERCP unnecessary as an initial test. Major limitations of ERCP are the limited ability to obtain a tissue diagnosis in malignant bile duct obstruction (positive in <50% of cases); limited utility for pancreatic tumor staging, as it provides no information about tumor extent, vascular invasion, or involvement of the lymph nodes; and risk of complications such as pancreatitis, perforation, and rarely, death.
Magnetic Resonance Imaging (MRI)
Accuracy of MRI to determine resectability appears comparable to the dual-phase helical CT scan. Magnetic resonance cholangiopancreatography (MRCP) is as sensitive as ERCP in the diagnosis of pancreatic tumors.
Positron Emission Tomography (PET)
PET scanning is not routinely used in diagnosis of pancreatic cancers. It can be useful in diagnosing tumor reoccurrence after pancreatic resection.
Studies have shown that standard preoperative imaging modalities, including CT scan in 15–40% cases, are unable to detect small peritoneal and liver metastasis. In select patients, staging laparoscopy along with peritoneal cytologic examination can detect unsuspected metastasis and prevent unwarranted surgery.
Figure 28–1 presents an algorithm for the diagnosis and treatment of suspected pancreatic cancer. The need for tissue to establish a diagnosis in a suspected pancreatic cancer remains controversial. Although consensus on an optimal approach is lacking, any diagnostic strategy should aim to reliably and safely establish a diagnosis, if necessary; determine resectability to avoid costs; and avoid morbidity and mortality associated with unnecessary surgical intervention.
Algorithm for the diagnosis and treatment of suspected pancreatic carcinoma. CT, computed tomography; ERCP, endoscopic retrograde cholangiopancreatography; EUS, endoscopic ultrasound; FNA, fine needle aspiration; MRCP, magnetic resonance cholangiopancreatography.
Comparing imaging modalities such as (EUS, MRI, and CT) has methodological limitations, such as patient selection, study design, and quality. EUS has been generally found to be superior to CT and MRI in detection and characterization of smaller lesions. It has been found to be superior to helical CT scan and MRI for tumor and nodal staging, with a sensitivity of greater than 90%. EUS does not offer any advantage over helical CT in determining resectability of preoperatively suspected pancreatic adenocarcinoma, and the two modalities are considered complementary.
Helical pancreatic protocol CT (PPCT) scan is usually the preferred initial test for loco-regional staging and detection of distant metastasis. Based on the CT scan findings, pancreatic masses fall into one of the following four broad categories: (1) no mass, (2) unresectable mass, (3) resectable mass, and (4) equivocal findings.
Based on clinical suspicion and pretest probability, further imaging by MRCP, EUS, ERCP, or a combination of these methods may be necessary, even if the initial CT findings are negative for any mass lesion. EUS can identify lesions not seen on CT or MRI.
If a clearly unresectable mass is identified on CT scan, a tissue diagnosis is usually indicated prior to starting any palliative chemotherapy or radiotherapy. EUS–FNA, or a percutaneous approach using a CT or transabdominal ultrasound, can be used. In patients with pancreatic cancer who experience pain, EUS can be used to perform celiac plexus neurolysis during the initial diagnostic and staging examination.
Patients identified as having a resectable mass on CT scan should be considered for EUS as a complementary test because it is more accurate for local tumor staging and predicting vascular invasion. The need to obtain a tissue diagnosis in this group is controversial. Beyond the patient's desire for a tissue diagnosis, pretest probability of a tumor, and available local expertise, the major consideration determining whether to obtain a tissue diagnosis should be its impact on clinical decision-making and management.
The major drawback of pursuing a tissue diagnosis is that the sensitivity of EUS–FNA is 85–90%, giving it an unacceptably high false-negative rate of 10–15%. If the pretest probability based on the clinical presentation laboratory data and radiological findings for cancer is high, then a negative tissue diagnosis does not influence the decision to proceed with surgery. Although minimal, the potential for complications such as bleeding, pancreatitis, and tumor seeding during fine needle aspiration should also be considered.
Besides confirming the diagnosis of pancreatic cancer in the majority of patients, the advantages of obtaining tissue are that it helps identify other malignancies, such as neuroendocrine tumors, lymphoma, and small-cell carcinoma, as well as nonmalignant conditions such as autoimmune pancreatitis and chronic pancreatitis, resulting in changes in treatment and prognosis.
Complications from surgery are another major concern. Pancreatoduodenectomy is associated with significant morbidity, and the mortality rate can range from less than 1% in a large-volume center to about 15% in centers performing few such surgeries per year. This must be taken into account by the physicians and the patient prior to surgical resection of the pancreas.
Patients with tumors localized to the body and tail of the pancreas and resectable by helical CT criteria should be considered for laparoscopy prior to surgery as these tumors are usually advanced at presentation and frequency of unsuspected spread to the peritoneum can be as high as 50%.
In patients whose CT findings are equivocal regarding the presence of a mass or potential for resectability, EUS is indicated, as it is the most sensitive method to detect small tumors of the pancreas and evaluate for vascular invasion.
Adamek HE, Albert J, Breer H, et al. Pancreatic cancer detection with magnetic resonance cholangiopancreatography and endoscopic retrograde cholangiopancreatography: a prospective controlled study. Lancet
ASGE Standards of Practice Committee, Gan SI, Rajan E, et al. Role of EUS: communication from the ASGE standards of practice committee. Gastrointest Endosc
Dewitt J, Devereaux BM, Lehman GA, et al. Comparison of endoscopic ultrasound and computed tomography for the pre-operative evaluation of pancreatic cancer: a systemic review. Clin Gastroenterol Hepatol
Mertz HR, Sechopoulos P, Delbeke D, et al. EUS, PET, and CT scanning for evaluation of pancreatic adenocarcinoma. Gastrointest Endosc
Soriano A, Castells A, Ayuso C, et al. Preoperative staging and tumor resectability assessment of pancreatic cancer: prospective study comparing endoscopic ultrasonography, helical computed tomography, magnetic resonance imaging, and angiography. Am J Gastroenterol
Wiersema MJ, Norton ID, Clain JE. Role of EUS in the evaluation of pancreatic adenocarcinoma. Gastrointest Endosc
. 2000;52: 578–582.
Pancreatic adenocarcinoma must be differentiated from the following conditions of the pancreas that can mimic its symptoms: autoimmune pancreatitis, chronic pancreatitis, pancreatic lymphoma, neuroendocrine tumors of the pancreas, and cystic lesions of the pancreas.
Autoimmune pancreatitis is a rare disorder of presumed autoimmune etiology that can manifest with obstructive jaundice, weight loss, abdominal pain, and new-onset diabetes. Imaging studies show diffuse enlargement of the pancreas, irregular narrowing of the pancreatic duct, strictures of the common bile duct, and stricture of the intrahepatic radicals similar to that seen in primary sclerosing cholangitis (refer to Table 27–3 in the preceding chapter). Increased levels of serum γ-globulins, especially immunoglobulin G4, are present. Lymphoplasmacytic infiltration with fibrosis and obliterative phlebitis is seen on histologic examination. Corticosteroids are effective in alleviating symptoms and reversing histopathologic changes (see Chapter 27).
The incidence of pancreatic adenocarcinoma in patients with chronic pancreatitis increases about 2% per decade after onset of the disease. Abdominal pain, weight loss, and jaundice may occur in patients with chronic pancreatitis, making it difficult to differentiate chronic pancreatitis complicated by adenocarcinoma.
Imaging studies are not always useful as patients with chronic pancreatitis can present with strictures of the pancreatic and bile duct on ERCP, a mass lesion on CT (variant of AIP), or with changes in the echo texture of pancreatic tissue, making EUS images difficult to interpret. EUS appears to be superior to CT scan for detection of coexistent malignancy.
Although CA 19-9 is elevated in about 80% of patients with adenocarcinoma, it can also be elevated in patients with chronic pancreatitis, and very high values may occur in patients with obstructive jaundice and cholangitis.
K-ras mutations do not seem to be clinically useful in differentiating chronic pancreatitis and pancreatic cancer.
Patients with pancreatic cancer can be subdivided into three categories based on the extent of tumor spread: (1) tumor confined to the pancreas (resectable disease at diagnosis), representing approximately 15–20% of patients; (2) locally advanced disease (unresectable), 40%; and (3) metastatic disease, 40%.
Surgical resection is the only curative treatment for pancreatic cancer. Owing to the characteristically late presentation, only 15% of patients are candidates for pancreatectomy. The most common surgery is the Whipple pancreaticoduodenectomy, used in patients with cancers located in the head of the pancreas. The procedure involves en bloc removal of the gastric antrum, pancreas, and duodenum. Removal of regional lymph nodes has not been shown to be associated with improved survival. Mortality rates for this procedure are about 1–3% at centers that perform a large number of these procedures.
As noted earlier, even after surgical resection with negative margins, patients have a 5-year survival rate of only 10–25%, with median survival of 10–20 months. Involvement of the lymph nodes in resected patients is the most important prognostic factor; 5-year survival after the Whipple procedure is 10% for patients with node-positive disease and 25–30% for those with node-negative disease. Tumor size less than 3 cm, well-differentiated tumors, negative surgical margins, and absence of lymph node metastasis are associated with improved survival.
Surgical resection of tumors located on the body and the tail consists of distal subtotal pancreatectomy along with splenectomy.
Endoscopic stent placement in patients with resectable tumors to relieve jaundice prior to surgery does not alter postoperative morbidity and mortality and is not recommended. If the surgery is scheduled for several weeks later, a plastic stent can be placed to relieve jaundice and reduce the risk of cholangitis.
Birkmeyer JD, Stukel TA, Siewers AE, et al. Surgeon volume and operative mortality in the United States. N Engl J Med.
Blackstock AW, Mornex F, Partensky C, et al. Adjuvant gemcitabine
and concurrent radiation for patients with resected pancreatic cancer: a phase II study. Br J Cancer.
Neoptolemos JP, Stocken DD, Friess H, et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med.
Stojadinovic A, Brooks A, Hoos A, et al. An evidence-based approach to the surgical management of resectable pancreatic approach to the surgical management of resectable pancreatic adenocarcinoma. J Am Coll Surg.
Optimal treatment in these patients remains controversial. The combination of chemotherapy and radiation is associated with modest improvements in median survival but may produce significant side effects. 5–Fluorouracil, gemcitabine, and paclitaxel are some of the common chemotherapeutic agents used with radiotherapy.
Symptom palliation is an essential component of care for patients with locally advanced disease. Patients with tumors localized to the pancreatic head can develop obstructive jaundice and gastric outlet obstruction; severe abdominal pain is more common with tumors involving the body and tail of the pancreas.
Sohn TA, Yeo CJ, Cameron JL, et al. Do preoperative biliary stents increase post pancreaticoduodenectomy complications? J GastrointestSurg.
Compression of the bile duct by the tumor can cause obstructive jaundice. Endoscopic stent placement is the most effective approach to relieve malignant bile duct obstruction. Metal stents are preferred over plastic ones for palliation once the patient is no longer considered a surgical candidate. Unlike plastic stents, which require frequent periodic changes due to occlusion, metal stents have a significantly higher patency rate; however, once placed they cannot be removed endoscopically. The stents may be placed percutaneously if an endoscopic approach is not possible as a result of tumor in-growth or previous surgery. Rarely, a surgical biliary enteric bypass is required to relieve the obstruction.
Gastric Outlet Obstruction
Obstruction of the gastric outlet is caused by extension of the pancreatic cancer and can be treated by gastrojejunostomy or by endoscopically placed expanding metal stents.
Narcotic medications are usually required to control pain associated with pancreatic cancers. Celiac plexus block (chemical neurolysis) can be achieved though a percutaneous approach, or the block can be performed using EUS. Radiation can be used to control intractable pain.
Chemotherapy is used for palliation and results in slight survival benefits. Gemcitabine-based combination therapy is the treatment of choice and associated with better pain control, decreased weight loss, and improvement in performance status.