August 30, 2008
Pheochromocytoma and Extra-Adrenal Paraganglioma: Updates
By Tischler, Arthur S
* Context.-Advances in genetics and gene expression profiling have led to new ways of thinking about the pathobiology of pheochromocytoma and extra-adrenal paraganglioma. These developments are concurrent with the publication and dissemination of the 2004 World Health Organization bluebook on pathology and genetics of endocrine tumors. Objective.-To summarize new information required by pathologists for effective participation in patient management and research.
Conclusions.-The World Health Organization reserves the term pheochromocytoma for tumors arising from chromaffin cells in the adrenal medulla. Closely related tumors in extra-adrenal sympathetic and parasympathetic paraganglia are classified as extra-adrenal paragangliomas. A pheochromocytoma is an intra-adrenal sympathetic paraganglioma. Although arbitrary, this nomenclature emphasizes important distinctive properties of intra-adrenal tumors, including an often adrenergic phenotype, relatively low rate of malignancy, and predilection to occur in particular hereditary syndromes. Malignancy is defined by presence of metastases not local invasion. Occult germline mutations characteristic of familial syndromes are now found in more than 20% of patients with apparently sporadic tumors, bringing the percentage of tumors with a known genetic basis close to 30%. In addition, tumor location and risk of malignancy vary with the underlying genetic defect. The "10 percent rule" for pheochromocytoma/paraganglioma-10% familial, 10% malignant, 10% extra-adrenal-is therefore no longer tenable. Current roles of pathology are limited to diagnosing primary or metastatic tumors and identifying features suggestive of malignant potential or hereditary disease. Future roles may involve more definitive assessment of malignancy, genotype-phenotype correlation, and identification of targets for therapy.
(Arch Pathol Lab Med. 2008;132:1272-1284)
Pheochromocytomas and extra-adrenal paragangliomas are tumors of neural crest-derived endocrine cells or organs, known as paraganglia, that exist throughout the distribution of the sympathetic nervous system and along supradiaphragmatic branches of the (parasympathetic) vagus and glossopharyngeal nerves during development or in adult life. Prototypical sympathetic paraganglia are the adrenal medulla and the organ of Zuckerkandl. The prototypical parasympathetic paraganglion is the carotid body. Other paraganglia are microscopic and highly variable in location.
The current World Health Organization classification reserves the term pheochromocytoma for intra-adrenal tumors. Similar tumors in other locations are defined as extra-adrenal paragangliomas, further coded according to their anatomic site. According to the World Health Organization Classification of Tumours, updated in 2004, "Phaeochromocytomas . . . arise in the adrenal medulla and are derived from chromaffin cells of neural crest origin,""Extra- adrenal paragangliomas arise from chromaffin cells in sympathoadrenal and parasympathetic paraganglia," and "A phaeochromocytoma is an intra-adrenal sympathetic paraganglioma."1 Restricted use of the term pheochromocytoma is an arbitrary convention originated in 1950 by H. T. Karsner, who wrote the first series of the Armed Forces Institute of Pathology Atlas of Tumor Pathology fascicle, Tumors of the Adrenal. Karsner noted the previous application of the term by other authors to various tumors in extraadrenal locations, including head and neck paragangliomas and even carcinoid tumors, and commented, "The confusion in terminology encumbers the exchange of experience which can be so constructive in fields where no one pathologist's experience is enough," adding "An agreement in nomenclature, even on an arbitrary basis, is helpful."2 Karsner's original definition, which encompassed tumors "in or near the adrenal," was refined in the second series fascicle of 1985 to include only intraadrenal neoplasms. By then, the Armed Forces Institute of Pathology fascicles had become the definitive vade mecum for pathologists in the United States, and for them the arbitrary nomenclature became the norm. Pathologists in other countries gradually adopted to the Armed Forces Institute of Pathology definition.
Although the current nomenclature has its origin in an arbitrary convention, "special" treatment of intra-adrenal tumors can in fact be justified on the basis of several distinctive characteristics. Those include a lower rate of malignancy (~5% overall vs ~20% for extra-adrenal paragangliomas associated with the sympathetic nervous system3), an often adrenergic phenotype (extra-adrenal paragangliomas are almost always noradrenergic4), and a proclivity to occur in association with particular genetic disorders, particularly multiple endocrine neoplasia (MEN) type 2 (MEN2).5 On the other hand, it can be reasonably argued that this classification may detract from efforts to understand the pathobiology of tumors that, overall, are more similar than different. Pheochromocytomas and extra-adrenal paragangliomas often have similar or identical morphology, share a mostly identical neuroendocrine phenotype, and sometimes have the same genetic predisposition.
Most pathologists agree in principle with the need for consistency and accept the idea that the World Health Organization nomenclature, although imperfect, should be adhered to. Nonetheless, even in recent publications precise and consistent nomenclature has been difficult to attain. Contributing factors include both lack of familiarity with the World Health Organization nomenclature and resistance to its arbitrary distinctions. Extra-adrenal sympathetic paragangliomas are often still classified as pheochromocytomas, especially in papers written by clinicians, and tumors' site of origin is frequently not specified. Nomenclature preferences aside, the latter practice is particularly unfortunate because it obscures potential influences of anatomical and functional context on tumor pathogenesis and phenotype.6 Salient distinctions between the normal adrenal medulla and extra-adrenal sympathetic paraganglia include denser innervation, an adrenergic versus noradrenergic phenotype, an environment rich in adrenal cortical steroids, and maturation much later in fetal and postnatal development. Differences in maturation may be particularly important in providing different developmental windows7 in which tumorigenic events can occur. Fortunately, the nomenclature stand-off is resolving as a result of cooperation between pathologists and clinicians with a shared interest in pathobiology.
RECENT DEVELOPMENTS IN PHEOCHROMOCYTOMA AND PARAGANGLIOMA RESEARCH
Recent advances in genetics, gene expression profiling, and cell biology have led to new ways of thinking about the pathobiology of pheochromocytoma and extra-adrenal paraganglioma. Physicians have traditionally been taught to remember the clinical properties of sympathetic paragangliomas according to the "10 percent rule"-10% familial, 10% malignant, 10% extra-adrenal. That rule is no longer tenable. Occult germline mutations characteristic of familial pheochromocytoma/paraganglioma syndromes have recently been documented in more than 20% of patients presenting with apparently sporadic tumors,8,9 bringing the percentage of tumors with a known genetic basis close to 30%. In addition, tumor location and risk of malignancy vary according to the underlying genetic defect.
Hereditary disorders that are well known to be associated with development of pheochromocytomas/paragangliomas are MEN2A and MEN2B, von Hippel-Lindau (VHL) disease, and neurofibromatosis type 1 (NF1) resulting, respectively, from mutations of the RET (Rearranged in Transfection) proto-oncogene and the VHL and NF1 tumor suppressor genes (Table 1).1,9 von Hippel-Lindau disease is now divided into types 1 and 2, defined by the absence or presence of susceptibility to pheochromocytomas/paragangliomas (Table 2).1 In addition, the list of hereditary susceptibility disorders is now expanded to include familial paraganglioma (PGL) syndromes caused by mutations of succinate dehydrogenase (SDH) genes SDHD (PGLl), SDHC (PGL3), and SDHB (PGL4), which also appear to function as tumor suppressor genes10,11 (Table 1). A novel aspect of PGLl is a mode of transmission that involves genomic imprinting, that is, tumors occur only after paternal transmission of the mutated gene.10,11 Finally, several kindreds susceptible to tumors harbor mutations that have still not been identified.12 Somatic mutations of the genes responsible for hereditary pheochromocytomas/paragangliomas are uncommon in tumors that are truly sporadic.13-16
Striking genotype-phenotype correlations exist for tumors in each of the familial pheochromocytoma/paraganglioma syndromes with respect to malignancy, distribution, and function. Estimated rates of malignancy are quite low for most of the known mutations, ranging from 1% to 10%. However, at least 50% of tumors with SDHB mutations are malignant.9 In addition, tumors with any of the SDH mutations are often extra-adrenal, whereas those with mutated RET are confined to the adrenal medulla or immediate vicinity. SDH mutations are also suggested by the combined occurrence of sympathetic and parasympathetic paragangliomas.
Gene expression profiling studies complemented by immunohistochemical and biochemical analyses have revealed different clusters of markers in tumors with specific genetic backgrounds and in subsets of sporadic tumors.17,18 VHL, SDHB, or SDHD mutations are associated with a "transcription signature" characterized by genes related to hypoxia-driven transcription pathways. In contrast, the signature of tumors with RET mutations is consistent with increased activity of the Ras-mediated MAPK pathway.19 An additional distinctive characteristic of VHL tumors is that they usually do not express phenylethanolamine N-methyltransferase, the enzyme that synthesizes epinephrine from norepinephrine and are therefore noradrenergic even when intra-adrenal. Intra-adrenal tumors with SDH mutations are also reported to be noradrenergic, 20 whereas MEN2 and NF1 tumors in the adrenal typically produce both epinephrine and norepinephrine.17 The genotype-phenotype correlations in familial pheochromocytoma/paraganglioma syndromes and the high prevalence of unsuspected hereditary disease have led to gene-specific recommendations for genetic testing and subsequent patient management.16,21,22 Tumor location, presence of multiple tumors, presence of metastases, and type of catecholamine produced are useful as guides in deciding which genes to test. However, specific algorithms differ somewhat according to institutional preference and test availability. The most stringent recommendations favor genetic testing of all patients with apparently sporadic tumors for RET, VHL, and SDH mutations to avoid being misled by individual differences in presentation and penetrance, low penetrance of some mutations, de novo mutations, and imprinting of SDHD.22 Routine genetic testing is not currently available for NF1 mutations because the gene is extremely large and, in contrast to the other pheochromocytoma/ paraganglioma susceptibility genes, does not have discrete mutation "hot spots" leading to development of these tumors.23
ROLES OF PATHOLOGY IN MANAGEMENT OF PHEOCHROMOCYTOMAS AND PARAGANGLIOMAS
The current roles of pathology in management of pheochromocytomas and extra-adrenal paragangliomas are limited to diagnosis, documentation of malignant behavior, and recording of findings that may be clues to occult hereditary disease. It is anticipated that future roles will involve more definitive assessment of malignant potential, genotype-phenotype correlation, and identification of targets for therapy.
Pheochromocytomas and other paragangliomas show enormous variability in cytology and histologic pattern and must be distinguished from a variety of endocrine and nonendocrine tumors. The classic pattern of "zellballen" formed by nests of uniform polygonal cells is often not evident, and one may instead observe diffuse architecture, spindle cells, admixtures of large and small cells, and extreme cytologic atypia (Figure 1, A through F; Figure 2, A and B). Areas of ganglioneuroma or ganglioneuroblastoma (Figure 3, A and B) are occasionally admixed with pheochromocytoma or paraganglioma in sympathoadrenal tumors (composite pheochromocytoma or paraganglioma). Parasympathetic paragangliomas often have more pronounced zellballen (Figure 2, B) and somewhat clearer cytoplasm than their sympathetic counterparts, but overlap exists between the 2 types of tumor.
Specific considerations in differential diagnosis vary according to anatomic site and diagnostic strategies vary accordingly. In the adrenal gland, the principal differential diagnosis is adrenal cortical adenoma or carcinoma. Elsewhere, possibilities include hepatic and hepatoid tumors, alveolar soft part sarcoma, melanoma, glomus tumors and other vascular neoplasms, and primary or metastatic carcinomas with endocrine or nonendocrine phenotype. It should be noted that the term glomus tumor has in the past been used as a synonym for paraganglioma in head and neck locations and occasionally still crops up as such in current publications. This use is unacceptable because it leads to confusion with true glomus tumors, which are derived from thermoregulatory myoarterial structures unrelated to paraganglia developmentally or functionally. The ambiguous application of the same term to different tumors is a vestige of 19th century nomenclature for head and neck paraganglia (eg, the "glomus caroticum,""glomus jugulare," and "glomus tympanicum").
Immunohistochemical staining procedures routinely available in most pathology laboratories can now reliably make the distinctions necessary for differential diagnosis, provided they are applied judiciously and with appreciation of potential artefacts. The latter include technical artefacts such as enhancement of endogenous biotin staining by heat-based antigen retrieval methods in staining protocols that use a biotin bridge,24 nonspecific staining of some neuroendocrine cell types by serum,25 and cross-reactivities of commercially touted antibodies that are often of dubious quality. Mitochondria often exhibit nonspecific interactions with antibodies, as does lipofuscin in the adrenal cortex. Consequently, oncocytic tumors (eg, of the adrenal cortex; Figure 4, A and B; Figure 5) can show very convincing but nonspecific staining for a variety of antigens, or normal cortical cells can show spurious staining for neuroendocrine markers. Although endogenous biotin is often cited as a cause of nonspecific staining, in my experience poorly validated primary antibodies or inadequate blocking of nonspecific staining is more often at fault. Nonspecific staining by any given antibody is not predictable. Moreover, no controls available in routine diagnostic immunohistochemistry are wholly adequate, including substitution of normal serum or immunoglobulin G in place of primary antibody.
The single most specific and reliable generic neuroendocrine marker currently used in pathology practice is chromogranin A (CgA), a major constituent of the matrix of catecholamine-containing secretory granules.26 Immunoreactivity for CgA will readily distinguish pheochromocytomas and other paragangliomas from tumors that are not neuroendocrine, such as those of the adrenal cortex. Staining of pheochromocytomas for CgA is usually extensive, and the diagnosis should be made with caution for tumors that show little or no staining. Distinction of pheochromocytomas and paragangliomas from other neuroendocrine tumors that also express CgA is a process of deduction that often involves panels of antibodies, for example, demonstration of immunoreactive tyrosine hydroxylase (TH; the rate- limiting enzyme in catecholamine biosynthesis) and exclusion of staining for keratin proteins that are often expressed in pulmonary and gastrointestinal neuroendocrine tumors. This distinction is particularly challenging when individual patients have both paragangliomas and carcinoid tumors (Figure 6, A through F; Figure 7, A through D), as can occur in NF1 or VHL disease.27-29 Excellent monoclonal antibodies are widely available for both CgA30 and TH. Although normal parasympathetic paraganglia express both TH and CgA, staining for either or both of those markers tends to be weaker and more variable in parasympathetic paragangliomas than in their sympathoadrenal counterparts and TH may actually be absent in some of those tumors4 (Figure 8, A through C). Low expression of TH correlates with the fact that the tumors were often classified as "non-chromaffin paragangliomas" in older literature.
Determination of Malignancy
According to the current World Health Organization classification, malignancy of pheochromocytomas and paragangliomas is defined by the presence of metastases1 not local invasion. Some pathologists challenge this convention, noting that other locally invasive tumors that have minimal metastatic potential are nonetheless classified as malignant (eg, basal cell carcinomas of the skin). The best argument to the contrary is based on tumor biology. Despite its potential lethality, local invasion alone is a poor predictor of metastases, and the absence of apparent invasion does not preclude development of metastases. The 2 types of aggressive behavior may therefore have different biologic underpinnings. Precise reporting and a consistent definition of malignancy are critical for assessing criteria that in the future will better define risk from either metastases or local invasion.
The most stringent definition of malignancy stipulates that metastases must be to a site where paraganglionic tissue is not normally present, for example, liver or bone (Figure 9), to avoid confusion with multiple primary tumors. This concern is certainly valid and is particularly applicable to clinical imaging studies. From the perspective of histopathology, if tumor is present within a clearly identifiable lymph node, it is reasonable to classify that tumor as malignant.
Determining whether a pheochromocytoma/paraganglioma is malignant before metastases have occurred is more problematic, and there is currently no consensus that it can reliably be done. An important obstacle to determining predictive criteria is the fact that pheochromocytomas/paragangliomas metastasize infrequently and often late. Consequently, it has been difficult to validate proposed criteria in large studies with adequate follow-up. Most existing analyses are therefore retrospective. In addition, the clustering of gene expression profiles in tumors with different underlying genetic predispositions31,32 suggests that any particular finding may carry different weight in different patient cohorts. For example, extensive local invasion might pose essentially no risk in an adequately excised tumor from a patient with MEN2A, whereas even minimal invasion might be ominous in a tumor with mutated SDHB. Because the prevalence of syndromic mutations varies geographically, studies reported from different countries or regions might offer very different conclusions. No single histologic feature is alone able to identify metastatic potential, including capsular or vascular invasion, extreme cytologic atypia, or areas resembling pediatric neuroblastoma. However, some evidence suggests that multifactorial scoring systems can help to histopathologically discriminate tumors that pose a significant risk of metastasis from those that do not. A seminal study by Linnoila et al3 in 1990 examined multiple histologic and nonhistologic parameters of 120 sympathoadrenal paragangliomas and developed a statistical model according to which more than 70% of the tumors could be classified correctly with more than 95% probability on the basis of 4 factors: extra-adrenal location, coarse nodularity, confluent necrosis, and absence of hyaline globules. Most malignant tumors had 2 or 3 of those features, whereas 89% of benign tumors had 1 or none. Unfortunately, a number of subsequent articles that deal with assessment of markers for malignancy blur the distinction between intra-adrenal and extra-adrenal tumors, so that the powerful independent predictive value of extra-adrenal location, clearly demonstrated as the most powerful predictor (P
In 2002 Thompson34 proposed the PASS system (pheochromocytoma of adrenal scaled score), specific to the adrenal gland, that scores multiple microscopic findings, including dependent and independent variables identified by Linnoila et al, to arrive at a total score correlated with metastatic potential. All tumors that metastasized had a score of greater than 4, but 17 of 50 with a score greater than 4 had not metastasized in a follow-up period of approximately 5 years, and 1 with a score of more than 15 had not metastasized in approximately 28 years. In his article, Thompson notes that the PASS score provides a threshold for risk but does not quantitate the risk above that threshold, and he further notes that a score of 3 or less does not guarantee that a patient will not at some point develop metastases. There is currently no agreement on the utility or reproducibility of a PASS score. Calculation of a score or reporting each of its components is optional, but a high score should not be considered equivalent to a diagnosis of malignancy.
Immunohistochemistry has been used as an ancillary technique for assessment of malignant potential, with mixed results. The marker most consistently reported to be correlated with malignancy is labeling index of tumor cells stained for the proliferation marker Ki-67, which is performed on paraffin sections using monoclonal antibody MIB-1.35-37 However, in some studies MIB-1 labeling does not correlate with malignancy. Studies of MIB-1 labeling show a striking lack of methodologic consistency, and many papers do not provide sufficient methodologic detail to permit replication. Although many pathologists now include an assessment of MIB-1 labeling in diagnostic reporting of pheochromocytomas/ paragangliomas, there is currently no prospect of standardization, and reporting of a Ki-67 index remains optional. Increasingly, numerous additional markers are reported to correlate with malignancy.9,35,38 However, their usefulness is still limited to research. As is the case for morphologic characteristics, it has not been ruled out that different immunohistochemical marker profiles carry different weights in tumors with different genetic backgrounds. For example, hypoxia-sensitive markers reported to be associated with malignancy, such as vascular endothelial growth factor and HIF1alpha, also characterize pheochromocytomas/ paragangliomas in patients with VHL disease, 31,32 and those tumors are usually benign.9
A 2005 scoring system proposed by Kimura et al36 for both pheochromocytomas and extra-adrenal sympathetic paragangliomas combines histologic, immunohistochemical, and biochemical characteristics to arrive at a score reported to predict both the metastatic potential of tumors and the prognosis for patients with tumors that metastasize. An interesting aspect of the system is scoring of whether a tumor is adrenergic or noradrenergic, thereby anticipating incorporation of this element into patient management algorithms later proposed by clinicians.21 Anatomic site in this system is indirectly given additional weight because extra-adrenal paragangliomas are almost invariably noradrenergic.
Identification of Occult Hereditary Disease
Pathology plays an important role in identifying clues that may lead to identification of patients with previously undetected familial pheochromocytoma/paraganglioma syndromes. In general, a finding of multicentric tumors is suggestive of familial disease. In the adrenal gland, familial disease may be manifest as multiple pheochromocytomas and/or diffuse and nodular hyperplasia. These changes should be carefully searched for and their presence or absence should be documented. They are most likely to occur in MEN2 but are often subtle or absent in MEN2 and are usually absent in other syndromes.
Adrenal medullary hyperplasia is usually detected by gross examination of the specimen. Adrenal medullary tissue is gray, in contrast to the brown zona reticularis of the cortex, and is normally concentrated in the head and body of the gland. Hyperplasia is suggested when medulla extends into the alae or tail and is often confirmed by subtle nodularity seen grossly or microscopically (Figure 10; Figure 11, A and B). In the absence of diffuse hyperplasia, separate small nodules are sometimes found widely separated by apparently normal medulla (Figure 11, B). A practical approach to search for these changes is to "breadloaf" the entire fresh adrenal into thin slices, allow the slices to adhere to a dry paper towel, and immerse the towel flat in fixative. This approach provides complete sections without retraction artefact.
Some evidence suggests that microscopic findings may also provide clues to underlying genotypes. Pheochromocytomas that occur in patients with VHL disease have been reported to exhibit distinctive features consisting of a thick vascular capsule, myxoid and hyalinized stroma, small cells with clear or amphophilic cytoplasm, intermixed small vessels, and absence of hyaline globules.39 These changes are not always present and it not known whether they correlate with specific VHL mutations. They may also overlap with findings in other familial or sporadic pheochromocytomas. In addition, the residual vascular scaffold of collapsed normal adrenal medulla adjacent to a pheochromocytoma may resemble a thick vascular capsule. Immunohistochemical staining for phenylethanolamine N- methyltransferase, the enzyme that synthesizes epinephrine from norepinephrine, may be a useful ancillary technique because pheochromocytomas in VHL disease are usually noradrenergic32 and, therefore, phenylethanolamine N-methyltransferase negative (Figure 12, A through C). A caveat, however, is that high-quality monoclonal antibodies such as those long-available for CgA or TH are not yet available for phenylethanolamine N-methyltransferase.
A PRACTICAL SUMMARY
Recent advances in genetics and gene expression profiling have dramatically increased our understanding of the previously mysterious combinations of tumors in hereditary pheochromocytoma/ paraganglioma syndromes. Studies of familial syndromes, especially VHL and familial PGL syndromes resulting from SDH mutations, have provided a platform from which to explore both familial and sporadic tumors. Understanding of genotype-phenotype correlations is important for both pathology and clinical management.
Despite these breakthroughs, the current practice of pathology with respect to these tumors is still based on careful gross examination of specimens and histologic study based principally on hematoxylin-eosin sections. Because pheochromocytomas and extra- adrenal paragangliomas often metastasize late and their behavior is unpredictable, patients require long-term follow-up. These tumors should therefore never be signed out as "histologically benign." They should also not be signed out as malignant unless there is unequivocal documentation of metastasis to lymph nodes or distant sites. Care must be taken to avoid interpreting separate primary tumors as lymph nodes. Local invasion should be documented and does have some correlation with metastatic potential, at least if it is extensive.3 However, by itself it is a poor predictor. In the classic 1990 study by Linnoila et al,3 30 of 120 sympathetic paragangliomas had metastasized but only 2 of those also showed extensive local invasion, whereas 2 additional tumors with extensive local invasion showed no evidence of metastasis.
In addition to local invasion, other features correlated with malignancy in large series3,34,36 should be noted. The listing of potentially unfavorable findings will presumably flag a tumor for some type of follow-up, but the nature of the required follow-up remains unclear. Most pathologists do not currently use any formal scoring system in routine practice, for reasons including the lack of prospective validation, poor concordance in recognizing some of the scored parameters, and a high false-positive rate evident from the fact that many tumors with high scores apparently never metastasize.3 If a numerical score is assigned it must be emphasized that a high score does not constitute a diagnosis of malignancy either for clinical or research purposes.
To optimize reporting of information required for care of patients with cancer, guidelines and templates for reporting of malignant tumors are provided by the College of American Pathologists (www.cap.org) and the Association of Directors of Anatomic and Surgical Pathology (ADASP) (www.ADASP.org) in the United States and the Royal College of Pathologists (www.rcpath.org) in the United Kingdom. Those recommendations are available in pathology journals39,40 and textbooks, as well as on the organizations' Web sites. The College of American Pathologists and the ADASP checklists are based on staging criteria in the American Joint Committee on Cancer 2002 staging manual and incorporate elements required by the American College of Surgery Commission on Cancer, which accredits cancer centers in the United States. Strictly speaking, most pheochromocytomas and extra-adrenal paragangliomas do not require template reporting because they are not malignant. However, the templates available for these tumors incorporate major elements of the various proposed scoring systems and therefore serve as useful checklists of features that should be noted, while leaving flexibility to add additional elements. Most elements listed are generic descriptors required in all tumor reporting (size, weight, necrosis, mitotic activity, and vascular or capsular invasion), but some are not necessarily highly correlated with the behavior of pheochromocytomas or extra-adrenal paragangliomas. For example, tumor size and weight may show some correlation with metastatic potential but not as independent variables.35 The adrenal checklist currently proposed for use in the United Kingdom (Table 3) is somewhat more detailed than that provided by ADASP, whereas ADASP provides a separate paraganglioma checklist not available on the Royal College Web site (Table 4). As noted on the ADASP Web site, ". . . ADASP realizes that specimens and practices vary and it will not be possible to report these elements in every case." Quality pathology reporting starts with the approach to the gross specimen. Sampling should be liberal and should represent all morphologically variant areas of a tumor, the tumor capsule, and the interface with adjacent soft tissue. A general guideline of at least one section per centimeter of tumor is suggested, as for many types of tumor. In principle, surfaces should be inked prior to sectioning to assess margins. However, laparoscopy is now the method of choice for resecting pheochromocytomas, and adrenals are often lacerated during laparoscopic removal. Routine inking is not a substitute for thoughtful gross assessment and can be confusing if ink is daubed onto lacerated surfaces. An additional potential source of confusion is that paragangliomas, particularly of the head and neck, are frequently embolized by clinicians prior to surgery to reduce operative bleeding, thereby creating gross or microscopic areas of necrosis that should not be interpreted as tumor necrosis. Foreign material used for this purpose can usually be identified in tumor blood vessels (Figure 13).
An exciting prospect is that pathologists in the near future might play a role in management of malignant pheochromocytomas/ paragangliomas by immunohistochemically identifying proteins for targeted therapy, as is already routine for a number of other malignancies. One immediate possibility is targeting of heat shock proteins41 using drugs that are already in clinical trials.42 There is presently little discussion of how sections should be used for immunohistochemistry or other ancillary studies. If the appearance of a tumor varies from slide to slide, as is often the case, should all of the slides be stained for each marker studied? If a tumor is large, should staining be performed on sections representing both the center and the periphery? Although the logical answer to both questions is yes, that approach may be precluded by constraints of cost and time. These issues remain to be resolved both in the design of protocols for clinical studies and in clinical practice.
I thank James F. Powers, PhD, for assistance with preparation of this article and for performing the immunoblot study shown in Figure 5.
Table 4. Summary of Association of Directors of Anatomic and Surgical Pathology Recommendations for Reporting of Extra-adrenal Paragangliomas*
Anatomic site of tumor pathologic diagnosis (terminology based on anatomic site, eg, "urinary bladder paraganglioma")
Type of resection
Tumor size (_____cm x _____cm x _____cm) and weight (_____g)
Gross description (external and cut surfaces)
Microscopic description (including presence and quantitation of mitotic figures)
Margins of resection (free of tumor, or tumor present at margins)
Presence and approximate extent of necrosis
Presence or absence of invasion (vessels, adjacent tissues or organs)
Unusual pathologic features
* Excerpted from Lack et al,43 provided as a checklist at http:// www.adasp.org/ (accessed February 5, 2008). Reprinted from Human Pathology with permission from Elsevier.
1. DeLellis RA, Lloyd RV, Heitz PU, Eng C, eds. Pathology and Genetics of Tumours of Endocrine Organs. Lyon, France: IARC Press; 2004. World Health Organization Classification of Tumours.
2. Karsner HT. Tumors of the Adrenal.Washington, DC: Armed Forces Institute of Pathology; 1950. Atlas of Tumor Pathology; 1st series, fascicle 29.
3. Linnoila RI, Keiser HR, Steinberg SM, Lack EE. Histopathology of benign versus malignant sympathoadrenal paragangliomas: clinicopathologic study of 120 cases including unusual histologic features. Hum Pathol. 1990;21:1168-1180.
4. Lloyd RV, Sisson JC, Shapiro B, Verhofstad AA. Immunohistochemical localization of epinephrine, norepinephrine, catecholamine-synthesizing enzymes, and chromogranin in neuroendocrine cells and tumors. Am J Pathol. 1986;125: 45-54.
5. Bryant J, Farmer J, Kessler LJ, Townsend RR, Nathanson KL. Pheochromocytoma: the expanding genetic differential diagnosis. J Natl Cancer Inst. 2003; 95:1196-1204.
6. Tischler AS. Molecular and cellular biology of pheochromocytomas and extra-adrenal paragangliomas. Endocr Pathol. 2006;17:321-328.
7. Lee S, Nakamura E, Yang H, et al. Neuronal apoptosis linked to EglN3 prolyl hydroxylase and familial pheochromocytoma genes: developmental culling and cancer. Cancer Cell. 2005;8:155-167.
8. Neumann HP, Bausch B, McWhinney SR, et al. Germ-line mutations in nonsyndromic pheochromocytoma. N Engl J Med. 2002;346:1459-1466.
9. Eisenhofer G, Bornstein SR, Brouwers FM, et al. Malignant pheochromocytoma: current status and initiatives for future progress. Endocr Relat Cancer. 2004;11:423-436.
10. Baysal BE. Genomic imprinting and environment in hereditary paraganglioma. Am J Med Genet C Semin Med Genet. 2004;129:85-90.
11. Baysal BE. Role of mitochondrial mutations in cancer. Endocr Pathol. 2006; 17:203-212.
12. Dahia PL, Hao K, Rogus J, et al. Novel pheochromocytoma susceptibility loci identified by integrative genomics. Cancer Res. 2005;65:9651-9658.
13. Dannenberg H, Komminoth P, Dinjens WN, Speel EJ, de Krijger RR. Molecular genetic alterations in adrenal and extra-adrenal pheochromocytomas and paragangliomas. Endocr Pathol. 2003;14:329- 350.
14. Astuti D, Morris M, Krona C, et al. Investigation of the role of SDHB inactivation in sporadic phaeochromocytoma and neuroblastoma. Br J Cancer. 2004;91:1835-1841.
15. Braun S, Riemann K, Kupka S, et al. Active succinate dehydrogenase (SDH) and lack of SDHD mutations in sporadic paragangliomas. Anticancer Res. 2005; 25:2809-2814.
16. Korpershoek E, Petri BJ, van Nederveen FH, et al. Candidate gene mutation analysis in bilateral adrenal pheochromocytoma and sympathetic paraganglioma. Endocr Relat Cancer. 2007;14:453-462.
17. Eisenhofer G, Goldstein DS, Kopin IJ, Crout JR. Pheochromocytoma: rediscovery as a catecholamine-metabolizing tumor. Endocr Pathol. 2003;14:193-212.
18. Brouwers FM, Glasker S, Nave AF, et al. Proteomic profiling of von Hippel-Lindau syndrome and multiple endocrine neoplasia type 2 pheochromocytomas reveals different expression of chromogranin B. Endocr Relat Cancer. 2007;14: 463-471.
19. Dahia PL, Ross KN, Wright ME, et al. A HIF1alpha regulatory loop links hypoxia and mitochondrial signals in pheochromocytomas. PLoS Genet. 2005;1:e8.
20. Timmers HJ, Kozupa A, Eisenhofer G, et al. Clinical presentations, biochemical phenotypes, and genotype-phenotype correlations in patients with succinate dehydrogenase subunit B- associated pheochromocytomas and paragangliomas. J Clin Endocrinol Metab. 2007;92:779-786.
21. Pacak K, Eisenhofer G, Ahlman H, et al. Pheochromocytoma: recommendations for clinical practice from the First International Symposium. October 2005. Nat Clin Pract Endocrinol Metab. 2007;3:92- 102.
22. Benn DE, Robinson BG. Pheochromocytoma-quo vadis? Nat Clin Pract Endocrinol Metab. 2007;3:377.
23. Bausch B, Borozdin W, Mautner VF, et al. Germline NF1 mutational spectra and loss-of-heterozygosity analyses in patients with pheochromocytoma and neurofibromatosis type 1. J Clin Endocrinol Metab. 2007;92:2784-2792.
24. Srivastava A, Tischler AS, Delellis RA. Endogenous biotin staining as an artifact of antigen retrieval with automated immunostaining. Endocr Pathol. 2004; 15:175-178.
25. Tischler AS, Tsokas P, Shahsavari M, Powers JF. Immunoreactivity of normal rabbit serum with epinephrine (E) cells of the rat adrenal medulla after microwave antigen retrieval. Cell Tissue Res. 1998;293:563-566.
26. Feldman SA, Eiden LE. The chromogranins: their roles in secretion from neuroendocrine cells and as markers for neuroendocrine neoplasia. Endocr Pathol. 2003;14:3-23.
27. Fuller CE, Williams GT. Gastrointestinal manifestations of type 1 neurofibromatosis (von Recklinghausen's disease). Histopathology. 1991;19:1-11.
28. Sinkre PA, Murakata L, Rabin L, Hoang MP, Albores-Saavedra J. Clear cell carcinoid tumor of the gallbladder: another distinctive manifestation of von Hippel-Lindau disease. Am J Surg Pathol. 2001;25:1334-1339.
29. Griffiths DF, Williams GT, Williams ED. Duodenal carcinoid tumours, phaeochromocytoma and neurofibromatosis: islet cell tumour, phaeochromocytoma and the von Hippel-Lindau complex: two distinctive neuroendocrine syndromes. Q J Med. 1987;64:769-782. 30. Wilson BS, Lloyd RV. Detection of chromogranin in neuroendocrine cells with a monoclonal antibody. Am J Pathol. 1984;115:458-468.
31. Dahia PL, Ross KN, Wright ME, et al. A HIF1alpha regulatory loop links hypoxia and mitochondrial signals in pheochromocytomas. PLoS Genet. 2005;1:72-80.
32. Eisenhofer G, Huynh TT, Pacak K, et al. Distinct gene expression profiles in norepinephrine- and epinephrine-producing hereditary and sporadic pheochromocytomas: activation of hypoxia- driven angiogenic pathways in von Hippel-Lindau syndrome. Endocr Relat Cancer. 2004;11:897-911.
33. Amar L, Baudin E, Burnichon N, et al. Succinate dehydrogenase B gene mutations predict survival in patients with malignant pheochromocytomas or paragangliomas. J Clin Endocrinol Metab. 2007;92:3822-3828.
34. Thompson LD. Pheochromocytoma of the Adrenal gland Scaled Score (PASS) to separate benign from malignant neoplasms: a clinicopathologic and immunophenotypic study of 100 cases. Am J Surg Pathol. 2002;26:551-566.
35. August C, August K, Schroeder S, et al. CGH and CD 44/MIB-1 immunohistochemistry are helpful to distinguish metastasized from nonmetastasized sporadic pheochromocytomas. Mod Pathol. 2004;17:1119- 1128.
36. Kimura N,Watanabe T, Noshiro T, Shizawa S, MiuraY. Histological grading of adrenal and extra-adrenal pheochromocytomas and relationship to prognosis: a clinicopathological analysis of 116 adrenal pheochromocytomas and 30 extraadrenal sympathetic paragangliomas including 38 malignant tumors. Endocr Pathol. 2005;16:23-32.
37. Liu TH, Chen YJ, Wu SF, et al. [Distinction between benign and malignant pheochromocytomas]. Zhonghua Bing Li Xue Za Zhi. 2004;33:198-202.
38. Portela-Gomes GM, Stridsberg M, Grimelius L, Falkmer UG, Falkmer S. Expression of chromogranins A, B, and C (secretogranin II) in human adrenal medulla and in benign and malignant pheochromocytomas: an immunohistochemical study with region- specific antibodies. APMIS. 2004;112:663-673.
39. Koch CA, Mauro D,Walther MM, et al. Pheochromocytoma in von Hippel- Lindau disease: distinct histopathologic phenotype compared to pheochromocytoma in multiple endocrine neoplasia type 2. Endocr Pathol. 2002;13:17-27.
40. Lack EE. Recommendations for the reporting of tumors of the adrenal cortex and medulla. Association of Directors of Anatomic and Surgical Pathology. Virchows Arch. 1999;435:87-91.
41. Scholz T, Eisenhofer G, Pacak K, Dralle H, Lehnert H. Clinical review: current treatment of malignant pheochromocytoma. J Clin Endocrinol Metab. 2007;92:1217-1225.
42. Dymock BW, Barril X, Brough PA, et al. Novel, potent small- molecule inhibitors of the molecular chaperone Hsp90 discovered through structure-based design. J Med Chem. 2005;48:4212-4215.
43. Lack EE, Lloyd RV, Carney JA, Woodruff JW. Recommendations for the reporting of extra-adrenal paragangliomas. The Association of Directors of Anatomic and Surgical Pathology. Hum Pathol. 2003;34:112-113.
44. Peczkowska M, Cascon A, Prejbisz A, et al. Extra-adrenal and adrenal pheochromocytomas associated with a germline SDHC mutation. Nat Clin Pract Endocrinol Metab. 2008;4:111-115.
45. Ercolino T, Becherini L, Valeri A, et al. Uncommon clinical presentations of pheochromocytoma and paraganglioma in two different patients affected by two distinct novel VHL germline mutations. Clin Endocrinol (Oxf). 2008;68:762-768.
46. DeAngelis LM, Kelleher MB, Post KD, Fetell MR. Multiple paragangliomas in neurofibromatosis: a new neuroendocrine neoplasia. Neurology. 1987;37:129-133.
Arthur S. Tischler, MD
Accepted for publication March 11, 2008.
From the Department of Pathology, Tufts New England Medical Center, Boston, Mass.
The author has no relevant financial interest in the products or companies described in this article.
Reprints: Arthur S. Tischler, MD, Department of Pathology, Tufts Medical Center, 750 Washington St, Boston, MA 02111 (e-mail: [email protected]).
Copyright College of American Pathologists Aug 2008
(c) 2008 Archives of Pathology & Laboratory Medicine. Provided by ProQuest LLC. All rights Reserved.