College of American Pathologists

2013 — March Case of the Month

Posted March 18, 2013


CAP Foundation Online Case of the Month

Click Slide Image to View Case with DigitalScope

After reading the summary, try answering the three related multiple-choice questions below.

A 53-year-old man underwent a CT scan during the work up of abdominal pain, which revealed a tumor in the tail of the pancreas. A distal pancreatectomy revealed a 3.0 cm solid tumor.

Archive Case and Diagnosis:
This case first appeared as Performance Improvement Program in Surgical Pathology (PIP) 2010, case 07, and is a well-differentiated pancreatic endocrine tumor.

Criteria for Diagnosis and Comments:
The tumor is comprised of a uniform population of cells arranged in solid nests and trabeculae. Focal areas with pseudopapillary architecture are present. The tumor cells show uniform nuclei with a stippled chromatin pattern and inconspicuous nucleoli. There is mild nuclear pleomorphism, with <2 mitoses per 10 HPF. Necrosis and vascular invasion are not present. The stroma shows extensive hyalinization. The morphological features support the diagnosis of a well-differentiated pancreatic endocrine tumor (PET). The WHO grading scheme divides these tumors into three categories (Table 1). Based on morphology and mitotic activity, the tumor will be classified as grade 1 neuroendocrine tumor as per the WHO 2010 classification.

The outcome in well-differentiated endocrine neoplasms cannot be reliably predicted based on histological features; integration of clinical, pathological and imaging data is necessary to predict tumor behavior. Attempts have been made to utilize immunohistochemical markers such as CK19 and CD99 to predict aggressiveness, so far with disappointing results. Overall, the 5-year survival for nonmetastatic disease is 70-80%.

The WHO 2010 grading system is based on mitoses or Ki-67 index; other features like vascular invasion, necrosis, tumor size and functional status do not affect the grade of the tumor in this grading scheme. If the results are different based on mitoses and Ki-67, the higher grade should be assigned. Since the use of Ki-67 index tends to result in a higher grade than the mitotic activity, its use is recommended in all cases. In rare instances, the morphology is well-differentiated, but the Ki-67 index may exceed 20%. Initial evidence suggests that the outcome in these high grade tumors with well-differentiated morphology is intermediate between grade 2 NET and neuroendocrine carcinomas with large/small cell morphology. There are no definite recommendations for the treatment of these cases, but limited experience suggests that unlike large/small cell neuroendocrine carcinoma, platinum-based therapy may not work well in these cases.

The majority of PETs are sporadic, but they can occur in association with genetic syndromes, most commonly with autosomal dominant type 1 multiple endocrine neoplasia (MEN). Gastrinomas are the most common endocrine tumor associated with MEN 1. Other genetic diseases with PET include von Hippel-Lindau disease and type 1 neurofibromatosis.

PETs are separated into functioning and non-functioning categories based on the presence or absence of a clinical syndrome related to hormone secretion. The most frequent among the syndromic or functional tumors are insulinomas, glucagonomas, gastrinomas, and VIPomas (vasoactive intestinal peptide). Specific hormones can be detected in the serum in most cases. The non-functioning (non-syndromic) tumors are not associated with a clinical syndrome, but can show evidence of hormone production in the serum and/or by immunohistochemistry. Most tumors expressing pancreatic polypeptide, neurotensin and somatostatin fall in this category. These non-functioning lesions come to clinical attention due to large size, local invasion or distant metastases. Histological features are similar irrespective of type of hormone produced with a few exceptions. Stromal amyloid is commonly seen in insulinomas. Somatostatinomas are peculiar for containing glandular structures with psammoma bodies, but the latter are typically seen in duodenal rather than pancreatic tumors. Among functioning tumors, insulinomas are almost always benign. The other functioning tumors usually fall into the WHO categories of tumors of uncertain behavior (10-15%) or well-differentiated carcinomas (85-90%); the vast majority (90-95%) of non-functioning tumors are well-differentiated carcinomas.

PETs are typically positive for neuroendocrine markers such as synaptophysin (associated with synaptic vesicles) and/or chromogranin (associated with secretory granules). There are two families of chromogranin (A and B); antibodies directed against both should be used as some tumors like insulinomas react only with chromogranin B. Other neuroendocrine markers like CD56 (neural cell adhesion molecule, NCAM), CD57, neuron specific enolase (NSE) and protein gene product 9.5 (PGP 9.5) are variably positive, but are generally not necessary for the diagnosis. Most tumors also express epithelial markers like cytokeratin and MOC-31. The utility of immunohistochemistry for specific hormone products is not well established. Functioning tumors are defined by the clinical syndrome and not by immunohistochemical results. In the majority of cases, the hormone responsible for the syndrome can be detected immunohistochemically, but the intensity and degree of staining is unrelated to the severity of symptoms. Immunohistochemistry of non-functional tumors will often isolate a predominant cell type, but the clinical utility of this distinction is debatable. Many tumors possess more than one cell type (up to 18%). The transcription factor PAX-8 is often positive in pancreatic neuroendocrine tumors. Although not specific, its combined use with CDX-2 may be helpful in identifying site of origin in metastatic neuroendocrine tumors. Grossly, pancreatic endocrine tumors are typically solitary and solid. Cystic change can occur in rare instances. The differential diagnosis includes other pancreatic neoplasms with solid growth pattern such as solid-pseudopapillary tumor (SPT), acinar cell carcinoma (ACC) and pancreatoblastoma.

Solid-pseudopapillary tumor (SPT) occurs predominantly in young women in the early to mid-20s. The tumor can show lobulated solid cell nests similar to PET. Formation of pseudopapillae with tumor cells attached to edematous or hyalinized fibrovascular stroma is a characteristic feature of SPT, but can be focally observed in PET. Nuclear grooves and PAS-D positive hyaline globules, both intracellular and extracellular, are often present. SPTs can resemble PET immunohistochemically as they are strongly positive for CD56 and can focally express synaptophysin, neuron specific enolase and pancytokeratin. Chromogranin is consistently negative. CD10 and vimentin are also diffusely positive. The immunohistochemically stains that are most helpful in the diagnosis of SPT are beta-catenin, progesterone receptor (PR) and E-cadherin. Majority of SPT harbor beta-catenin mutations that lead to nuclear localization of the protein (normal tissues show membrane location); this finding is uncommon in PET. Using antibodies to the extracellular domain, loss of E-cadherin is seen in nearly all SPTs (also in 30% of PETs), while antibodies to the cytoplasmic domain show nuclear expression. The combination of CD10+, PR+ (nuclear), beta-catenin + (nuclear) and complete membrane loss of E-cadherin is virtually diagnostic of SPT.

Acinar cell carcinoma (ACC) is an aggressive neoplasm that mostly occurs in late adulthood as a bulky, well-circumscribed, multinodular solid mass, most commonly in the pancreatic head. The tumor can show acinar, solid, or trabecular architecture. The tumor cells characteristically show a single prominent central nucleolus and eosinophilic cytoplasm with PAS-D positive zymogen granules. Neuroendocrine markers like chromogranin and synaptophysin are variably expressed in ACC. Immunohistochemical demonstration of pancreatic enzymes, especially trypsin, distinguishes it from PET.

Pancreatoblastoma is essentially a pediatric tumor that is histologically and immunohistochemically similar to ACC. The characteristic feature is the presence of squamoid corpuscles.

Table. The WHO Grading Scheme
TermMitosesKi-67 index
Neuroendocrine tumor, grade 1 (carcinoid)<2 per 10 high power fields<2%
Neuroendocrine tumor, grade 22-20 per 10 high power fields3-20%
Neuroendocrine carcinoma (large cell or small cell type)>20 per 10 high power fields>20%

Supplementary Questions:

Question Diagnostic Set
1. Which of the following is correct about well-differentiated pancreatic endocrine tumors? A. Most insulinomas are malignant
B. The distinction between low- and high-grade neuroendocrine carcinoma is based on vascular and perineural invasion
C. Tumors lacking nuclear atypia can show lymph node metastasis
D. Tumors larger than 2.0 cm are considered malignant
2. Which of the following immunohistochemical staining results is characteristic of solid pseudopapillary tumor? A. Cytoplasmic staining with chromogranin
B. Membrane staining with beta-catenin
C. Membrane staining with E-cadherin
D. Nuclear staining with progesterone receptor
3. Which of the following immunohistochemical staining results is correct about the diagnosis of pancreatic tumors? A. Chromogranin expression is characteristic of solid pseudopapillary tumors
B. Loss of membrane E-cadherin staining is characteristic of acinar cell carcinoma
C. Nuclear accumulation of beta-catenin is characteristic of neuroendocrine tumors
D. Trypsin expression is characteristic of pancreatoblastoma


  1. DeLellis R.A., Lloyd R.V., Heitz, P.U., Eng C. (Eds.): World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Endocrine Organs. IARC Press: Lyon 2004.
  2. Frankel WL. Update on pancreatic endocrine tumors. Arch Pathol Lab Med. 2006;130:963-966.
  3. Lam KY, Lo CY. Pancreatic endocrine tumour: a 22-year clinico-pathological experience with morphological, immunohistochemical observation and a review of the literature. Eur J Surg Oncol. 1997 Feb;23(1):36-42.
  4. Serra S, Chetty R. Revision 2: an immunohistochemical approach and evaluation of solid pseudopapillary tumour of the pancreas. J Clin Pathol. 2008;61:1153-1159.

Sanjay Kakar, MD
Surgical Pathology Committee
University of California
San Francisco, CA