Protocol for the Examination of Specimens From Patients With Pheochromocytomas and Extra-Adrenal Paragangliomas

Beyond differential diagnosis, pathologists play important roles in identifying clues to hereditary disease and alerting clinicians to possible associated lesions and their significance. The proposed checklist aims to provide uniform and complete data to allow thorough evaluation of pheochromocytomas and extra-adrenal paragangliomas. This checklist will guide pathologists to issue standardized reports. It does not include the detailed information required to reach the diagnosis of pheochromocytoma or extra-adrenal paraganglioma; that is provided elsewhere.^1–5  A novel component of the checklist is a formatted clinicopathologic correlation.

+ Data elements marked with this symbol are not required. While they are important, some are not yet validated or regularly used in patient management and others may not be readily available to the pathologist examining the specimen.

Procedure (select all that apply) (note A)

 _Adrenalectomy

  _Right

  _Left

  _Bilateral

 _Extra-adrenal excision (specify):

 _Other (specify):

 _Not specified

+Biochemical Features (select all that apply) (note B)

 _Biochemically functioning

  _Metanephrine and/or adrenaline

  _Normetanephrine and/or noradrenaline

  _Methoxytyramine and/or dopamine

  _Other (specify):

 _Biochemically silent

 _Biochemical analysis not performed

 _Cannot be determined

+Tumor Scintigraphy or Positron Emission Tomography (PET) (select all that apply) (note C)

 _¹²³I-metaiodobenzylguanidine scintigraphy

 _¹⁸F-6-fluorodopamine PET

 _¹⁸F-6-fluorodihydroxphenylalanine PET

 _¹⁸F-fluorodeoxyglucose PET

 _Other (specify):

+Tumor Location and Size (from imaging) (note D)

Anatomic location (specify):

Greatest dimension: _ cm

+Additional dimensions: _ × _ cm

Second dominant tumor if multifocal

 _ × _ × _ cm

+Additional dimensions if more than 2 foci:

 _Cannot be determined

+Received

 _Fresh

 _In formalin

  +Fixation time:

 _Other (specify):

+Specimen Integrity

 _Intact

 _Fragmented

Specimen Size

 _ × _ × _ cm

+Additional dimensions:

+Specimen Weight

 _ grams

Tumor Focality

 _Unifocal or

 _Multifocal (specify number):

 _Cannot be determined

Tumor Size

Dominant tumor

 _ × _ × _ cm

Second dominant tumor if multifocal

 _ × _ × _ cm

+Additional dimensions if more than 2 foci:

Tumor Type (note E)

 _Pheochromocytoma(s), specify site(s):

 _Extra-adrenal paraganglioma(s), specify site(s):

 _Composite pheochromocytoma (specify): ___

 _Composite paraganglioma (specify site and components):

 _Gangliocytic paraganglioma

 _Metastatic pheochromocytoma, specify site:

 _Metastatic paraganglioma, specify site:

 _Other (specify):

Histologic Features (note F)

Growth pattern (select all that apply)

 _Nested (alveolar, zellballen) pattern

 _Trabecular pattern

 _Diffuse (solid) pattern

 _Expanded large confluent nests

 _Other (specify):

Composite tumor elements (select all that apply)

 _Absent

 _Present (select all that apply):

  _Neuroblastoma

   Specify extent (%):

   Degree of differentiation of the neuroblastic component (select all that apply)

    _Undifferentiated

    _Poorly differentiated

    _Differentiating

    _Cannot be assessed

  _Ganglioneuroblastoma

    Specify extent (%):

    Subtypes:

    _Nodular subtype

     Specify number of nodules:

     Specify the degree of differentiation for each neuroblastic nodule:

    _Intermixed subtype

  _Ganglioneuroma

   Specify extent (%):

  _Malignant peripheral nerve sheath tumor

   Specify extent (%):

  _Neuroblastic tumor, not otherwise specified

   Specify extent (%):

  _Other (specify):

+Cytologic variants of chromaffin and/or chief cells (select all that apply)

 _Epithelioid

 _Clear cell

 _Spindle cell

 _Lipid cell change

 _Oncocytic change

Necrosis

 _Not identified

 _Present, focal (small microscopic foci or single-cell necrosis)

 _Present, extensive (central, expansive, or “comedo” necrosis)

Mitotic rate (select all that apply)

 ++Based upon counting 50 high-power fields (HPFs: ×40 objective) and in the area of highest mitotic activity, and reported as number of mitoses per 10 HPFs)

 +++Alternatively the mitotic count can be given as per mm²

 _Specify mitoses (per 10 HPFs or mm²):

 _Atypical mitoses

 _Cannot be determined

+Additional features

 _Hyaline globules

 _Amyloid deposition

 _Neuromelanin deposition

 _Myxoid and/or hyaline stroma

 _Degeneration (specify):

Encapsulation and Invasion (note G)

 _Thick capsule

 _No capsule

 _Cannot be determined

Invasive Growth (select all that apply)

Tumor capsule invasion (transcapsular)

 _Present

  Specify the extent of invasion (number of foci):

 _Not identified

 _Indeterminate

 _Cannot be assessed

Adrenal capsule invasion (transcapsular)

 _Present

  Specify the extent of invasion (number of foci):

 _Not identified

 _Indeterminate

 _Cannot be assessed

Local invasion into surrounding tissues

 _Present

  Specify tissues:

  Specify extent (gross or microscopic):

 _Not identified

 _Indeterminate

 _Cannot be assessed

Vascular invasion (intravascular tumor cells associated with thrombus)

 Intracapsular

  _Present

   Specify the extent of invasion (number of vessels involved):

  _Not identified

  _Indeterminate

  _Cannot be assessed

 Beyond capsule

  _Present

   Specify the extent of invasion (number of vessels involved):

  _Not identified

  _Indeterminate

  _Cannot be assessed

 Lymphatic invasion

  _Present

  _Not identified

  _Indeterminate

  _Cannot be assessed

Surgical Margins

 _Uninvolved

  +Distance to closest margin:

 _Involved

  _Gross

  _Microscopic

 _Cannot be assessed

_Other (specify):

Metastases (note H)

 _Lymph node metastases

  _Present

  _Not identified

  _Indeterminate

  _Number of lymph nodes examined

  _Number of metastatic lymph nodes:

   +Number of lymph nodes with macrometastases (>2 mm):

   +Number of lymph nodes with micrometastases (≤0.2 mm):

  _Extranodal extension

   _Present

    _+Focal (microscopic)

    _+Extensive

   _Not identified

  _Indeterminate

 _Distant metastases

   _Present (specify site and data source):

   _Not identified

   _Cannot be determined

+Immunohistochemistry (check all positive or select all that apply) (note I)

 _Chromogranin A

 _Synaptophysin

 _Tyrosine hydroxylase

 _S100 protein (sustentacular cells)

 _Loss of SDHB expression

 _Loss of SDHA expression

MIB-1 (Ki-67) labeling index (percentage of positive tumor cells in area of highest nuclear labeling): %

Others (specify):

+Associated Lesions (note J)

 _Adrenal medullary hyperplasia

 _Current or past tumors in other organs (specify):

+Clinicopathologic Correlation (check all that apply)

 _Evidence of hereditary disease

  Clinical

   Family history (specify):

   Associated lesions (specify):

   Biochemical profile (specify):

   Cannot be assessed:

Pathologic

 _Multiple pheochromocytoma/paraganglioma

 _Adrenal medullary hyperplasia

 Immunohistochemistry (specify):

 Cannot be assessed:

+Comment(s):

Paraganglia are neural crest-derived neuroendocrine organs that produce predominantly catecholamines.^1–3  Paraganglia are typically divided into 2 groups by parasympathetic or sympathetic nervous system origin. Sympathetic paraganglia are also divided into 2 subgroups: the adrenal medulla, so-called sympathoadrenal paraganglia, and extra-adrenal sympathetic paraganglia.^3–5  The anatomic site impacts the nomenclature of tumors arising from paraganglia; while tumors arising from the adrenal medulla are termed pheochromocytomas, tumors arising from extra-adrenal locations are called paragangliomas regardless of their sympathetic or parasympathetic origins.^1–5  Furthermore, the anatomic site of a tumor predicts the risk of malignancy, since extra-adrenal paragangliomas exhibit a higher risk of malignancy.^1–8 

While pheochromocytomas and most sympathetic paragangliomas are often associated with clinical symptoms, only a small percentage of parasympathetic paragangliomas are symptomatic.¹  Many clinically silent paragangliomas, particularly of the sympathoadrenal type, will produce metanephrines and/or methoxytyramine and therefore be amenable to biochemical testing.^7,9  However, parasympathetic paragangliomas often lack tyrosine hydroxylase, the enzyme required for catecholamine synthesis, and are therefore usually nonfunctional.² 

Similar to other neuroendocrine neoplasms, pheochromocytomas and extra-adrenal paragangliomas are also capable of producing and secreting other peptides that can cause clinical syndromes. Production of adrenocorticotropic hormone, β-endorphin, corticotropin-releasing hormone, calcitonin gene–related peptide, vasoactive intestinal peptide, growth hormone–releasing hormone, neuropeptide Y, peptide YY, insulin-like growth factor–1, galanin, adrenomedullin, serotonin, somatostatin, and gastrinlike neuropeptide have been reported.^{1–5,10–17} 

Recent molecular data demonstrate genotype-phenotype correlations in paragangliomas with respect to tumor distribution, catecholamine production, and risk of metastasis.^2,7,18–23  It is now recognized that at least 30% of paragangliomas and pheochromocytomas are associated with familial syndromes.^2,3,18  Specific genotype-biochemical correlations highlight the importance of laboratory testing to characterize patterns of catecholamine excess. The biochemical profiles of tumors associated with mutations of genes encoding succinate dehydrogenase subunits, (collectively referred to as SDHx; x refers to all subunits, SDHA refers to subunit A, etc) are characterized by dopamine and/or norepinephrine production. VHL (von Hippel-Lindau)–related tumors are associated with norepinephrine production, and RET (rearranged during transfection)- and NF1 (neurofibromin 1)–related tumors are associated with epinephrine production.^2,7  Moreover, the risk of metastasis is significantly higher in SDHB (succinate dehydrogenase subunit B)–related paragangliomas, which are usually observed in extraadrenal locations and reach larger sizes with much lower tissue concentrations of catecholamines than other paragangliomas.^7,19–23  Catecholamines (dopamine, norepinephrine, and epinephrine) are not continuously secreted in normal conditions, and undergo intracellular methylation by the tumor cells that produce them. Biochemical testing for the O-methylated metabolites of dopamine, norepinephrine, and epinephrine (methoxytyramine, normetanephrine, and metanephrine, respectively) in plasma and/or urine is therefore superior to measurement of the parent catecholamines.^7,22,23  These data are of clinical significance in that integration of the biochemical profile with other information, such as tumor location and dimensions, becomes an important part of comprehensive synoptic reporting.

Similar to the genotype-biochemical profile correlations of paragangliomas, the functional status of a paraganglioma has an impact on imaging modalities that are used to localize these lesions.^{18 123}I-metaiodobenzylguanidine scintigraphy and ¹⁸F-6-fluorodopamine or ¹⁸F-6-fluorodihydroxphenylalanine (18F-FDOPA) positron emission tomography (PET), are superior to other functional imaging modalities for detecting pheochromocytomas.^18,24  In contrast, ¹⁸F-fluorodeoxyglucose PET is more useful than other modalities for diagnostic localization of SDHB-driven metastatic paragangliomas,^18,24  whereas 18F-FDOPA PET has been reported to be the most effective functional imaging modality for localization of SDHx-related head and neck paragangliomas.²⁵ Recently, it has been shown that 18F-FDOPA PET is most useful for the detection of head and neck paragangliomas and neuroendocrine neoplasms arising in patients with VHL syndrome.^26,27  When available, the integration of functional imaging data is of clinical interest and will ascertain the completeness of the synoptic report.

The significance of tumor location with respect to the parasympathetic/sympathetic origin of the tumor, and correlation with the biochemical profile and the appropriate terminologies, are discussed in detail in notes A and B. Therefore, the anatomic location of the tumor must be clearly specified in the synoptic report with the appropriate classification based on location.

Similar to other guidelines, tumor size is a required field in surgical pathology reports. Numerous reports^{3,6,23,28–31}  have indicated that malignant tumors are heavier and larger than tumors with benign behavior. Although the tumor size and weight are not universally considered independent parameters, a cutoff of 5- to 6-cm diameter and 80- to150-g weight have been suggested to predict malignant behavior.^23,28–31 

The issue of multifocality is important and should be included in the synoptic report.^1–5  Patients with multiple paragangliomas should be evaluated for the possibility of underlying genetic susceptibility and thus genetic testing for RET, NF1, VHL, SHDx, TMEM127 (transmembrane protein 127), MAX (MYC-associated factor-X), and KIF1Bβ (kinesin family member 1B) mutations should be considered.^{1,3,19,20,32–36}  While the value of systematic genetic screening for “sporadic” cases remains controversial, clinical features including family history, along with the biochemical and morphologic features (multifocality, adrenal medullary hyperplasia, thick capsule, clear cell morphology) and immunoprofile (loss of SDHB and SDHA expression)^36,37  (see notes B, F, I and J), can provide important insight to determine which gene(s) should be screened preferentially in patients with pheochromocytomas and/or extra-adrenal paragangliomas. Multifocality includes multiple pheochromocytomas in the same adrenal gland.

Anatomic location impacts the terminology used for these tumors. In the presence of metastatic disease, the term metastatic should be used. The term composite is used when a tumor combines features of paraganglioma or pheochromocytoma with those of malignant peripheral nerve sheath tumor, ganglioneuroma, ganglioneuroblastoma, and neuroblastoma. Comprehensive data related to neuroblastic and related components should be reported by using the designated synoptic checklist.³⁸  The histologic classification generated from the recommendations of the 2004 World Health Organization Classification of Tumours of endocrine organs¹  is listed below; however, for simplicity the format proposed is shortened to allow a practical approach for synoptic reporting.

 _Pheochromocytoma

 _Metastatic pheochromocytoma

 _Composite pheochromocytoma (specify components):

 _Carotid body paraganglioma

 _Jugulotympanic paraganglioma

 _Vagal paraganglioma

 _Laryngeal paraganglioma

 _Aorticopulmonary paraganglioma

 _Gangliocytic paraganglioma

 _Cauda equina paraganglioma

 _Orbital paraganglioma

 _Nasopharyngeal paraganglioma

 _Extra-adrenal sympathetic paraganglioma

 _Superior and inferior para-aortic paraganglioma

 _Urinary bladder paraganglioma

 _Intrathoracic and cervical paravertebral paraganglioma

 _Metastatic paraganglioma

 _Composite paraganglioma (specify site and components):

 _Others (specify):

Regardless of sympathetic or parasympathetic origin, paragangliomas usually exhibit overlapping morphologic features. They display a variety of growth patterns and cytologic features.^1–5  While sympathetic paragangliomas and pheochromocytomas consist of polygonal cells, so-called chromaffin cells that exhibit amphophilic to basophilic cytoplasm, parasympathetic tumors consist of polygonal cells, so-called chief cells that often have relatively clearer cytoplasm than their sympathetic counterparts. However, overlapping of these cells is often seen in these tumors. Similar to other endocrine lesions, oncocytic change, spindle cell morphology, and lipid cell degeneration leading a clear cell morphology that mimics cortical lesions can also be seen in these neoplasms.^1,3,39 

Genotype-phenotype correlations have highlighted the fact that VHL-related tumors often exhibit a thick vascular capsule, hyalinized and myxoid stroma, round tumor cells intermingled with small vessels, cells with predominantly amphophilic and clear cell cytoplasm, absence of intracytoplasmic hyaline globules, lipid degeneration,³⁹  and lack of nuclear atypia or mitoses.^1,3 

The term composite should be used when a tumor combines features of paraganglioma or pheochromocytoma with those of malignant peripheral nerve sheath tumor, ganglioneuroma, ganglioneuroblastoma, and neuroblastoma. Comprehensive data related to neuroblastic and related component should be reported by using the designated synoptic checklist.³⁸  In this setting, corticomedullary tumors, cauda equina paragangliomas showing ependymal differentiation, as well as gangliocytic paragangliomas that include Schwann-like cells and ganglion cells, do not qualify as composite tumors.^1,4  Moreover, scattered mature ganglion cells seen in pheochromocytomas/paragangliomas should not be misinterpreted as a component of a composite tumor.^1–4 

No single histologic parameter is able to predict malignant behavior in paragangliomas and pheochromocytomas.^1–5  Tumor necrosis is uncommon in these tumors and degenerative changes should not be mistaken as necrosis.^1–5  However, expanded large confluent nests with central comedo necrosis, which are at least 3 times greater than conventional small nests, have been described in some malignant pheochromocytomas/paragangliomas.^1–4  Therefore, a distinction should be made between focal (small microscopic foci or single-cell necrosis) and extensive (central, expansive, or “comedo” necrosis). Increased mitoses (>3 per 10 high-power fields [HPFs]) and atypical mitotic figures have been reported in some malignant cases,^1,3,4  but mitoses are usually very rare even in malignant cases. There is currently no standard approach to mitotic count in pheochromocytoma/paraganglioma. On the basis of established methodology for other neuroendocrine tumors, it has been recommended that mitotic count should be based upon counting 50 HPFs (×40 objective) and in the area of highest mitotic activity, and reported as number of mitoses per 10 HPFs. However, taking into consideration the variations in field size, providing the number of mitoses per mm² seems to be more appropriate. The College of American Pathologists' Breast Cancer Protocol⁴⁰  recommends that the size of HPFs be measured by using a micrometer. Alternatively, it has been suggested that the HPF diameter per area can also be calculated by using the following formulas⁴⁰: (1) Measure the diameter of a low-power field by using a ruler; (2) Calculate a constant by using the following formula: Eyepiece Magnification × Objective Magnification × Microscopic Field Diameter = A Constant; (3) Calculate the diameter of an HPF for other objectives by using the following formula: Unknown Field Diameter = Constant/(Eyepiece Magnification × Objective Magnification); and (4) Calculate the area of the HPF as follows: Half of the field diameter is the radius of the field (r), and 3.1415 × r ²  = Area of Microscopic Field. By doing this, one can also provide the mitotic activity per mm².

According to the 2004 World Health Organization (WHO) classification of endocrine neoplasms, malignancy of pheochromocytomas and extra-adrenal paragangliomas is defined by the presence of metastases to sites where paraganglial tissue is not normally found.¹  Although local gross invasion into the adjacent organs is considered in the definition of malignancy proposed by the 2007 Armed Forces Institute of Pathology fascicle,⁴  this is not regarded as a strong predictor of metastases and therefore it is not integrated in the 2004 WHO classification.^1–3  Moreover, unlike other neoplasms, vascular invasion is also not universally accepted as an unequivocal predictor of malignant potential in paragangliomas and pheochromocytomas.^1–3  However, it is important to document the invasiveness of these tumors. Strict criteria to diagnose vascular invasion (intravascular tumor cells associated with thrombus) and capsular invasion (transcapsular) should be applied as they are in other endocrine organs.⁴¹  As discussed in note F, the presence of a thick vascular capsule may raise the suspicion of a VHL-related paraganglioma.^1–3 

An extra-adrenal location, large size, and the presence of SDHB mutations are all important risk factors for metastatic spread.^6,7,21,23,42  High rates of malignancy in tumors associated with SDHB mutations can be fully accounted by both their typically extra-adrenal location and large size.²³  While tumors arising from head and neck paragangliomas are much less often metastatic, mediastinal and intra-abdominal paragangliomas appear to often be associated with metastatic disease.^1,3,6,8 

The diagnosis of metastasis is appropriate when pheochromocytoma or paraganglioma is present in a site where normal paraganglia do not exist. It is crucial to remember the normal anatomic distribution of paraganglia as discussed in note A, in order to consider the possibility of multifocal primary tumors. The pathology report should state the total number of lymph nodes examined and the number of nodes with metastases; nodal involvement should be reported as macrometastasis (>2 mm) or micrometastasis (≤2 mm and including isolated tumor cells) from the size of the metastatic deposit. While the determination of the nodal disease is easy, the assessment of distant metastasis can be challenging in the setting of multifocal disease, since primary paragangliomas do also occur in rare anatomic sites such as thyroid, pituitary, gallbladder, and lung.^2,4,43–45  Therefore, these rare locations should not be considered metastatic ab initio.

Positivity for tyrosine hydroxylase, which is the rate limiting enzyme in the synthesis of catecholamines,⁴⁶  is very helpful to distinguish paragangliomas from other neuroendocrine carcinomas, which can also be negative for cytokeratins.²  However, positivity for chromogranin A and tyrosine hydroxylase is usually weaker and more variable in parasympathetic paragangliomas than in sympathetic paragangliomas and is sometimes absent.² Some of these tumors selectively express chromogranin B.

S100 protein is typically used to highlight the sustentacular network in paragangliomas; however, the reactivity pattern is usually variable. It is of note that epithelioid endocrine cells and spindled Schwann-like cells of gangliocytic paragangliomas can be positive for cytokeratin and S100 protein, respectively.^1,4  Moreover, cauda equina paragangliomas, which are usually intradural lesions limited to the filum terminale, may show ependymal and neuronal differentiation and can be positive for cytokeratin.^1,4,47 

There is currently no standard approach to scoring Ki-67 in pheochromocytoma and paraganglioma. On the basis of established methodology for other neuroendocrine tumors, it is recommended that Ki-67 index should be reported as percentage of positive tumor cells in area of highest nuclear labeling.⁴⁸ 

Loss of SDHB expression is regarded as a surrogate marker for some of the familial paraganglioma syndromes caused by SDHx mutations³⁶ ; therefore, immunohistochemical testing for SDHB has become a part of the routine assessment of these lesions in many centers. Moreover, the use of SDHB antibody not only allows the identification of SDHx-related tumors, but also provides prognostic data, owing to the high rate of malignancy associated with SDHB-driven paragangliomas.^7,21,24,42  Recently, it was demonstrated that SDHA immunohistochemistry is also very useful to reveal the presence of SDHA germline mutations³⁷ ; paragangliomas associated with germline SDHA mutation show negative staining for SDHA as well as SDHB.

It is widely accepted that adrenal medullary hyperplasia is a precursor lesion of pheochromocytomas arising in MEN 2 (multiple endocrine neoplasia type 2) syndromes and is characterized by a nodular and/or diffuse enlargement of the adrenal medulla.^1–5  Although other predisposing genetic syndromes are not usually associated with adrenal medullary hyperplasia, it is noteworthy that a 61-year-old man with an SDHB mutation was found to have bilateral adrenal medullary hyperplasia characterized by an increased cortex to medulla ratio of 1:1 in both glands.⁴⁹  Other exceptions might also exist.

The determination of underlying adrenal medullary hyperplasia is one of the clinical responsibilities of pathologists examining adrenal glands. When examining diffuse hyperplasia, it is important to remember that medulla is normally present only in the head and body, but not in the tail of the gland with only minimal extension into the alae.^3–5  Although it is sometimes hard to define the tail of the adrenal gland owing to distortion of the gland by tumor, the presence of adrenal medullary tissue in the tail qualifies as adrenal medullary hyperplasia.^3–5  In general, medulla should not represent more than one-third of the gland thickness, with cortex on each side comprising the other two thirds. The distinction of pheochromocytoma from nodular adrenal medullary hyperplasia is arbitrary since even microscopic nodules observed in the setting of MEN 2 syndrome represent clonal proliferations; therefore, they are indeed neoplastic lesions.^3,5  However, nodules less than 1 cm can be practically considered to represent hyperplasia,⁴  provided that they grossly and microscopically resemble the rest of the medulla. It should be remembered that adrenal medullary nodules and pheochromocytomas can occur in adrenals in MEN 2 syndrome without an obvious background of diffuse hyperplasia. The adrenal gland adjacent to an apparently sporadic pheochromocytoma should therefore be “breadloafed” and carefully examined for small nodules.