—  SYMPOSIUM #17  —

Solid Tumors and Tumor-like Lesions of the Pancreas
Moderators: Ralph H. Hruban and Günter Klöppel

Section 1 - Ductal Adenocarcinoma of the Pancreas and Its Variants

Ralph H. Hruban
Professor of Pathology and Oncology
The Sol Goldman Pancreatic Cancer Research Center
The Johns Hopkins Medical Institutions
Baltimore, MD USA


Introduction
Infiltrating ductal adenocarcinoma of the pancreas (pancreatic cancer) is one of the most lethal of all of the solid malignancies. World-wide it is estimated that 227,000 will die from pancreatic cancer [1]. These numbers will only grow as the population ages.

Pancreatic cancer can be hard on the pathologist. The stakes are high and the diagnoses are difficult. Both the life expectancy and the most appropriate treatment for a patient diagnosed with chronic pancreatitis differ significantly from the life expectancy and appropriate treatment for a patient diagnosed with pancreatic cancer. Yet, in some cases, it can be extremely difficult, if not impossible, to distinguish pathologically between a benign reactive gland of chronic pancreatitis and an infiltrating gland of well-differentiated pancreatic cancer.

In this lecture we will discuss well-defined morphologic criteria that can be used to differentiate between neoplastic and reactive glands. In addition, the clinical characteristics, morphologic spectrum, and grading of infiltrating ductal adenocarcinoma will be presented. Emphasis will be placed on precursor lesions because these lesions can mimic infiltrating cancer, and because a better understanding of these lesions offers the best hope for treating pancreatic neoplasia before an incurable invasive cancer develops.

Ductal Adenocarcinoma
Infiltrating ductal adenocarcinoma of the pancreas is an invasive malignant epithelial neoplasm with glandular (ductal) differentiation and without a predominant component of any other type of carcinoma [2].

Grossly, most infiltrating ductal adenocarcinomas form poorly-defined firm fibrotic masses. They are white-yellow, and they obscure the normal lobular architecture of the pancreas [2]. This latter feature can be very helpful in distinguishing between chronic pancreatitis and infiltrating ductal adenocarcinoma. Some cancers undergo central necrosis and these cancers can form gross cysts. Cysts may also be found adjacent to the carcinoma due to localized dilatation of obstructed ducts (retention cysts) [2].

Microscopically, as defined above, infiltrating ductal adenocarcinoma of the pancreas are invasive malignant epithelial neoplasms with glandular (ductal) differentiation and without a predominant component of any of the other carcinoma types [2]. The degree of gland formation can vary from well-formed glands, to partially formed glands, to focal intracellular mucin production by poorly oriented cells infiltrating singly, to solid sheets of neoplastic cells [2]. Several features can be used to distinguish between benign reactive glands and infiltrating ductal adenocarcinoma, and most of these boil down to location, location, location (Table 1). In evaluating the location of a duct, recall that non-neoplastic glands, even when there is severe chronic pancreatitis, form predictable lobular units. The ducts are towards the center of the lobular unit and are surrounded by grape-like clusters of acini. The first feature useful in establishing a diagnosis of pancreatic cancer is that pancreatic carcinoma infiltrates in a haphazard pattern. The glands violate the lobular architecture, and are strewn in the parenchyma without rhyme or reason. Second, and this really is a manifestation of the first feature, is that neoplastic glands can be found adjacent to muscular arteries without intervening pancreatic parenchyma [3]. By contrast, in the non-neoplastic pancreas, muscular arteries run at the periphery of the lobules separated from the ducts by pancreatic parenchyma. The third and fourth features are that glands will be found where they simply don't belong. Perineural and vascular invasion are both features of an invasive adenocarcinoma. Benign glands abutting a nerve have been reported in very rare instances of severe chronic pancreatitis, but true perineural invasion is virtually diagnostic of invasive pancreatic cancer [2]. One feature of vascular invasion that should be noted is that when pancreatic cancer infiltrates into vessels it has a tendency to grow along the intimal surface of the vessel. In these instances the malignant glands lining the intima of the vessel can mimic a low-grade pancreatic intraepithelial neoplasia (PanIN) lesion. The fifth feature is called the "4 to 1 rule." The nuclei in non-neoplastic glands tend to be uniform, and in a single gland the area of the nuclei do not vary by more then 4 to 1. By contrast, significant nuclear pleomorphism can be seen in ductal adenocarcinomas, and the finding of nuclei in a single gland varying by more than 4 to 1 is indicative of a malignancy. Necrotic debris within the lumen of a gland is the sixth feature to support a diagnosis of pancreatic cancer. The seventh feature is incomplete lumina- that is lumina which are not completely lined by an epithelial layer such that the luminal contents appear to directly touch the stroma. I find the eighth feature, a gland directly touching fat without intervening stroma, difficult to apply. The ninth and final feature is the immunolabeling profile. The expression of the DPC4/MADH4 gene product, dpc4, is completely lost in slightly more than half of infiltrating ductal adenocarcinomas, and cancers express carcinoembryonic antigen (CEA). By contrast the vast majority of non-neoplastic glands have intact (normal) labeling for dpc4 and most do not express CEA. When rigorously applied, these nine features form a solid foundation for one of the most difficult differential diagnosis in surgical pathology, namely chronic pancreatitis versus pancreatic cancer.

Special stains and immunohistochemical labeling can help in the evaluation of suspected infiltrating ductal adenocarcinoma. Most adenocarcinomas of the pancreas express mucin and therefore stain with periodic acid-Schiff stain and this staining is diastase resistant. They also stain with the mucicarmine stain. Most express the cytokeratins (CK) 7, 8, 13, 18 and 19 (2). CK 17 is expressed by 50% of infiltrating ductal adenocarcinoma, and CK 20 by <20%. The majority also express carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA 19-9), B72.3 (TAG-72), CA 125, and DUPAN 2. Most express MUC1 (a pan-epithelial mucin), MUC3, MUC4, and MUC5AC (a gastric foveolar mucin) [2]. A minority express MUC6 (a pyloric-gland mucin), and <10% MUC2.

Pancreatic Intraepithelial Neoplasia
Microscopic precursor lesions, known as pancreatic intraepithelial neoplasia, have been recognized for over a century. Recent molecular studies have helped solidify the importance of these lesions as precursors to invasive pancreatic cancer. PanINs are a hot topic because they suggest that it should be possible to detect and treat these non-invasive precursor lesions before an incurable invasive cancer develops. PanINs are defined as neoplastic epithelial proliferations in the smaller caliber pancreatic ducts, and PanINs were divided into three grades based on the degree of architectural and nuclear atypia present [4]. Just as there is a progression in the colorectum from adenoma, to adenoma with dysplasia, to invasive cancer, so too is there a histologic and genetic progression from PanIN-1, to PanIN-2, to PanIN-3, to invasive ductal adenocarcinoma in the pancreas [5].

PanINs are often encountered at surgical margins. Although there is scant evidence-based medicine to guide us, it is generally agreed that no therapy is needed for PanIN-1 or PanIN-2 at a margin [5]. There have been isolated case reports of PanIN-3 lesions which progressed to invasive ductal adenocarcinomas. Therefore the resection of additional pancreatic parenchyma to achieve a margin free of PanIN-3 may be warranted in some instances. When deciding whether or not to take an additional margin, it would be wise to first consider the patient's clinical situation. If the patient has a large invasive cancer with multiple lymph node metastases, then their invasive cancer posses a significantly greater threat to their life than a PanIN-3 [5]. By contrast, if it is a young patient with a potentially curable pancreatic cancer, the resection of additional parenchyma to achieve a margin free of PanIN-3 may be recommended [5].

Variants of Adenocarcinoma

Adenosquamous carcinoma is a malignant epithelial neoplasm with significant components of both glandular and squamous differentiation [2]. At least 30% of the neoplasm should have squamous differentiation [2]. The two components can be intimately admixed, or they can be topographically separate within the neoplasm (2). Adenosquamous carcinoma is an extremely aggressive neoplasm.

Colloid carcinoma is an infiltrating adenocarcinoma characterized by mucin producing neoplastic epithelial cells suspended ("floating") in large pools of extracellular mucin [2, 6]. The colloid component should comprise at least 80% of the neoplasm [2]. Most colloid carcinomas arise in association with an IPMN, and these IPMNs usually exhibit intestinal type papillae [2, 7].

Hepatoid carcinoma is a malignant epithelial neoplasm with significant hepatocellular differentiation [2]. Only a few cases have been reported and most also have areas with a more common direction of differentiation, particularly ductal adenocarcinoma. [2].

Medullary carcinoma is a malignant epithelial neoplasm characterized by poor differentiation, pushing borders, a syncytial growth pattern and necrosis [8]. Most are microsatellite instable (MSI+), and immunolabeling will often reveal the loss of the expression of one of the DNA mismatch repair proteins, Mlh1 and Msh2 [8]. Patients with medullary cancers are more likely to have a family history of cancer.

Signet ring carcinoma is a malignant epithelial neoplasm in which the predominant component is composed of infiltrating round non-cohesive (isolated) cells containing intracytoplasmic mucin [2]. Metastases from a breast or gastric primary should be ruled out before making the diagnosis of a signet ring carcinoma primary to the pancreas.

Undifferentiated carcinoma is anepithelial neoplasm with a significant component showing no glandular structures or other features to indicate a definite direction of differentiation [2]. The neoplastic cells range from pleomorphic epithelioid mononuclear cells containing abundant eosinophilic cytoplasm admixed with bizarre frequently multinucleated giant cells, to relatively monomorphic spindle cells [2]. As one would expect, these are extremely aggressive almost uniformly fatal malignancies.

The undifferentiated carcinoma with osteoclast-like giant cells is a malignant epithelial neoplasm composed of reactive multinucleated giant cells admixed with atypical neoplastic mononuclear cells [2]. These neoplasms often arise in association with an adenocarcinoma or mucinous cystic neoplasm.

TABLE 1: H and E features useful in distinguishing benign glands and invasive adenocarcinoma+

Benign gland Invasive Carcinoma
Haphazard growth pattern No Yes
Glands adjacent to vessels No Yes
Perineural invasion No Yes
Intravascular invasion No Yes
Nuclear variation > 4 to 1 No Yes
Intraluminal necrosis Usually no Often
Incomplete glands No Often
Gland touching fat No Yes

+Adapted from [2]

Reference List
  1. Parkin DM, Bray F, Ferlay J, Pisani P. Estimating the world cancer burden: Globocan 2000. Int J Cancer 2001; 94(2):153-156.

  2. Hruban RH, Klimstra DS, Pitman MB. Atlas of tumor pathology. Tumors of the pancreas. Fourth Series ed. Washington, DC: Armed Forces Institute of Pathology, 2006.

  3. Sharma S, Green KB. The pancreatic duct and its arteriovenous relationship: an underutilized aid in the diagnosis and distinction of pancreatic adenocarcinoma from pancreatic intraepithelial neoplasia. A study of 126 pancreatectomy specimens. Am J Surg Pathol 2004; 28(5):613-620.

  4. Hruban RH, Adsay NV, Albores-Saavedra J, Compton C, Garrett E, Goodman SN et al. Pancreatic intraepithelial neoplasia: a new nomenclature and classification system for pancreatic duct lesions. Am J Surg Pathol 2001; 25(5):579-586.

  5. Maitra A, Fukushima N, Takaori K, Hruban RH. Precursors to invasive pancreatic cancer. Adv Anat Pathol 2005; 12(2):81-91.

  6. Adsay NV, Pierson C, Sarkar F, Abrams J, Weaver D, Conlon K et al. Colloid (mucinous noncystic) carcinoma of the pancreas. Am J Surg Pathol 2001; 25(1):26-42.

  7. Seidel G, Zahurak M, Iacobuzio-Donahue CA, Sohn TA, Adsay NV, Yeo CJ et al. Almost all infiltrating colloid carcinomas of the pancreas and periampullary region arise from in situ papillary neoplasms: a study of 39 cases. Am J Surg Pathol 2002; 26(1):56-63.

  8. Wilentz RE, Goggins M, Redston M, Marcus VA, Adsay NV, Sohn TA et al. Genetic, immunohistochemical, and clinical features of medullary carcinoma of the pancreas: a newly described and characterized entity. Am J Pathol 2000; 156(5):1641-1651.