Pancreatic cancer: Clinical aspects, assessment, and management

Clinical presentation

Pancreatic adenocarcinoma, or pancreas cancer, is one of the most lethal malignancies. While a relatively uncommon cancer, ranking 8th among cancer types in women and 10th in men, it is among the most common causes of cancer death in the United States. Pancreas cancer currently ranks third behind lung and colorectal cancer as a cause of cancer death, and is expected to become the second most common cause of cancer death in the coming decade. The increasing frequency of pancreas cancer in recent decades reflects both a true increase in its age-adjusted incidence, and the fact that pancreas cancer risk approximately doubles for each decade of life between age 50 and 90 ; thus the increasing average age of our population results in a larger number of patients with a new diagnosis of pancreas cancer (see Chapter 61 ).

The lethality of pancreas cancer reflects both its intrinsic biologic aggressiveness and that most patients are diagnosed when their tumor is relatively advanced. Surgical resection is the only treatment modality associated with long-term survival of pancreas cancer patients, but unfortunately only a small minority of pancreas cancer patients are candidates for surgery at the time of their diagnosis. Over half of pancreas cancer patients have systemic metastases identified at the time of diagnosis, and another third have locally advanced disease precluding immediate surgery, although some may become resectable with preoperative (neoadjuvant) chemotherapy, as will be discussed in the following sections. Only 15% to 20% of patients initially present with surgically operable pancreas cancer and, despite improvements in perioperative chemotherapy, cancer will recur in many of these patients. For unresectable patients, and those who recur after surgery, survival with current chemotherapy and radiation therapy approaches generally ranges from several months to several years. Recent therapeutic advances that have contributed to improved outcomes in other cancer types, in areas like targeted cancer therapy and immunotherapy, have thus far been of relatively limited benefit in pancreas cancer.

Pancreas cancer is rarely diagnosed “early” because it is often relatively asymptomatic in its initial stages. Tumors in the head of the pancreas can obstruct the bile duct, causing jaundice, and the phrase “painless jaundice” has been used to define a classic presentation of pancreas cancer. Most operable tumors occur in the pancreatic head and are identified in patients by the occurrence of jaundice. However, it is important to recognize that many tumors occur in areas of the pancreas that do not obstruct the bile duct, and that most pancreas cancers are not in fact painless by the time they are diagnosed. Many patients have a nonlocalized upper abdominal discomfort related to tumor growth and infiltration that is present for weeks or months before diagnosis. Pain radiating to the back is a particularly ominous symptom reflecting tumor involvement of retroperitoneal nerves. This pain is often worse at night when supine, and relieved with sitting. Other signs and symptoms of pancreas cancer include weight loss, nausea and vomiting reflecting gastric outlet obstruction, a palpable mass, gallbladder distension (Courvoisier’s sign), peripheral venous thrombosis (Trousseau’s sign), pancreatitis in the absence of other potential precipitants such as gallstones or regular alcohol use, and pancreatic insufficiency (either exocrine and/or endocrine). In fact, new-onset insulin-dependent diabetes after the age of 60 is associated with a 1% to 2% incidence of a previously undiagnosed pancreas cancer. Unfortunately, most of these signs and symptoms are nonspecific and, with the exception of jaundice, are generally subacute, contributing to the relative rarity of diagnosing pancreas cancer at an early stage.

Diagnosis

Laboratory tests

Patients in whom a suspicion of pancreas cancer is raised are often initially evaluated with serum blood tests. These laboratory studies may show a marked elevated bilirubin and alkaline phosphatase, confirming the clinical impression of obstructive jaundice, often with some elevation of liver alanine aminotransferase and aspartate aminotransferase. The serum tumor marker CA19-9 is often elevated in pancreas cancer, but it may also be markedly elevated in patients with biliary obstruction because of benign causes. Furthermore, CA19-9 is a carbohydrate antigen linked to the Lewis blood group antigens and approximately 15% to 20% of patients are incapable of synthesizing CA19-9. Thus the use of CA19-9 in making the diagnosis of pancreas cancer, particularly in patients with jaundice, is of limited value. Its primary utility is in following treatment response when it is elevated in nonjaundiced patients with a known pancreatic malignancy, whereas a reduction in CA19-9 levels is associated with improved survival. It is also a negative prognostic factor if markedly elevated in nonjaundiced patients and those who have undergone successful biliary decompression.

Other serum markers, such as carcinoembryonic antigen (CEA), are similarly too insensitive or nonspecific to be of use in the initial diagnosis of pancreas cancer. An evolving approach is the evaluation of circulating cancer cells or circulating tumor DNA in the diagnosis of pancreas cancer. The combination of such approaches with evaluation of CA19-9 and other blood markers may offer a relatively sensitive and specific approach to the early diagnosis of pancreas cancer, although this is still experimental.

Imaging studies

The identification of a pancreatic tumor is generally made by cross-sectional imaging (see Chapter 17 ). High-resolution triple-phase contrast-enhanced computed tomographic (CT) scanning of the abdomen is the primary tool used to identify a pancreatic mass. ^, Pancreatic carcinomas are generally hypoenhancing on the arterial phase and hypointense or isointense on the venous phase. CT scanning can also define regional adenopathy and potential extrapancreatic extension with involvement of adjacent vascular structures and is useful in detecting evidence of metastatic disease to the liver or peritoneum. Magnetic resonance imaging (MRI) scanning is of similar utility in assessing a pancreatic mass and may be more useful for defining small liver metastases. MRI may be preferentially used over CT in patients with limited renal function or allergy to the intravenous contrast used in CT. Patients rarely require both modalities except that MRI is more useful for characterizing small or indeterminate lesions (see Chapter 17 ). MRI is of particular utility in assessing cystic lesions of the pancreas and in their long-term follow-up, as MRI provides excellent resolution of high-risk features such as mural nodules, and the absence of radiation exposure is a benefit in long-term surveillance imaging (see Chapter 60 ).

Positron emission tomographic (PET) scanning using the 18-fluoro-2-deoxyglucose (FDG) tracer to assess tumor metabolic activity is of more limited utility because 10% to 30% of pancreas cancers fail to take up this tracer, and thus are PET negative. Furthermore, benign sites of inflammation may take up FDG and thus be erroneously PET positive. FDG-PET scanning is of greatest utility in helping to define small indeterminate lesions that cannot be well characterized by CT or MRI scanning, when the suspicion for a benign inflammatory process is low. Advances in the use of alternative tracers, such as 68-gallium dotatate for neuroendocrine malignancies, while useful in the diagnosis and staging of certain tumors, is outside the scope of this chapter (see Chapter 18 ).

Diagnostic biopsy

The definitive diagnosis of pancreatic adenocarcinoma requires evaluation of histopathology. However, not all operable pancreatic masses require a preoperative biopsy. A lesion that has radiographic characteristics of a pancreatic carcinoma and which is amenable to surgical resection after radiologic staging may, in some cases, be best managed by proceeding directly to surgery. For patients with a pancreas mass that has a high likelihood of being malignant based on radiologic findings alone and in whom surgery can be performed with low morbidity and mortality, performing preoperative biopsy may be of no benefit. In contrast, for patients who are at higher risk for surgery because of their age or underlying medical conditions and/or who wish to have a diagnosis made before consenting to surgery, obtaining a preoperative biopsy may be quite useful. Such biopsy-proven confirmation of the presence of pancreas cancer is also generally required for patients who will receive neoadjuvant chemotherapy.

For patients requiring a biopsy of a pancreas mass, endoscopic ultrasound (EUS)–guided fine-needle aspiration (FNA) has become the most common nonoperative or preoperative approach to identifying the presence of pancreas cancer ^, (see Chapter 22 ). Although occasionally limited by intervening vascular structures, in experienced hands the vast majority of pancreas lesions can be sampled with EUS-FNA. However, it is important to recognize that pancreas cancers are often highly fibrotic with relatively small areas of cancer cellularity. In some patients less than 10% of the area containing the cancer on histopathologic analysis will harbor nests of cancer cells, with the remainder composed of fibrous stroma. Sampling error in FNA of such lesions may fail to obtain material sufficient to make the diagnosis of cancer on cytology, accounting for the approximately 10% to 20% false-negative rate for EUS-FNA-guided biopsy of pancreas cancer. Although highly operator dependent, EUS may be more sensitive than axial imaging in identifying vascular involvement and suspicious peripancreatic lymph nodes, and thus may contribute to more accurate tumor staging as described next.

Alternative nonoperative approaches to obtaining a tissue diagnosis include endoscopic retrograde cholangiopancreatography (ERCP) with brushings for lesions causing biliary obstruction (see Chapter 20 ) and percutaneous biopsy under radiologic (ultrasound or CT) guidance (see Chapter 23 ). ERCP for diagnosis alone is rarely performed in the presence of a radiologically evident pancreatic mass. The sensitivity of ERCP with brushings is inferior to EUS-FNA and the complications of ERCP, particularly pancreatitis, which can be severe in a small fraction of patients, can be responsible for significant morbidity and rare mortality, and may complicate subsequent treatment decisions.

ERCP may be useful in placing a stent and decompressing the biliary tree to relieve the symptoms of jaundice, particularly pruritis, if patients are not able to have surgical resection with biliary reconstruction because of inoperable disease or planned neoadjuvant chemotherapy (see Chapters 20 and 30 ). Because plastic stents require replacement every 2 to 3 months, candidates for neoadjuvant treatment should generally have a metallic wall stent placed, as such stents provide a superior duration of biliary drainage that will generally not require another intervention during the period of neoadjuvant treatment. Data from randomized studies do not support the routine placement of biliary stents in jaundiced patients with resectable tumors who are anticipated to have surgery and biliary reconstruction within a relatively short time period. There is no evidence that endoscopic biliary decompression improves postsurgical outcomes. In addition to the small risk of severe pancreatitis mentioned earlier, there is growing literature demonstrating that preoperative biliary intervention is associated with a higher postoperative surgical infection rate, presumably caused by seeding of the biliary tree with enteric microorganisms.

In the relatively rare patient who requires a nonoperative biopsy and in whom EUS-FNA is unsuccessful because anatomic factors, percutaneous needle biopsy under radiologic guidance remains an option. Such biopsies may target the primary tumor mass, or suspected metastases in the liver or elsewhere. Seeding of tumor cells intraperitoneally as a result of percutaneous needle biopsy is a theoretical concern but is quite rare; in the largest series reported it occurred in 0.1% of cases. Given the relative difficulty in accessing the pancreas percutaneously, it is often easiest and most definitive to biopsy a suspected site of metastasis if such sites exist.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here