August 30, 2008
Practical Pituitary Pathology: What Does the Pathologist Need to Know?
By Asa, Sylvia L
* Context.-The sellar region is the site of frequent pathology. The pituitary is affected by a large number of pathologic entities arising from the gland itself and from adjacent anatomical structures including brain, blood vessels, nerves, and meninges. The surgical pathology of this area requires the accurate characterization of primary adenohypophysial tumors, craniopharyngiomas, neurologic neoplasms, germ cell tumors, hematologic malignancies, and metastases as well as nonneoplastic lesions such as cysts, hyperplasias, and inflammatory disorders. Objective.-To provide a practical approach to the diagnosis of pituitary specimens.
Conclusions.-The initial examination requires routine hematoxylin- eosin to establish whether the lesion is a primary adenohypophysial proliferation or one of the many other types of pathology that occur in this area. The most common lesions resected surgically are pituitary adenomas. These are evaluated with a number of special stains and immunohistochemical markers that are now available to accurately classify these tumors. The complex subclassification of pituitary adenomas is now recognized to re- flect specific clinical features and genetic alterations that predict targeted therapies for patients with pituitary disorders.
(Arch Pathol Lab Med. 2008;132:1231-1240)
Anumber of pathologic processes occur in the region of the pituitary gland. They include primary pituitary lesions that are unique to this site, as well as disorders arising in adjacent anatomical structures such as brain, blood vessels, nerves, and meninges. The surgical pathology of this area requires the accurate characterization of neoplastic lesions, including pituitary adenoma and carcinoma, craniopharyngioma, neurologic neoplasms, germ cell tumors, and hematologic malignancies, and their distinction from nonneoplastic disorders such as cysts, hyperplasias, and inflammatory lesions.1,2 The spectrum of pituitary pathologies that represent the surgical pathology of the pituitary is outlined in Table 1.
The commonest disorder is the pituitary adenoma, a lesion that is increasingly recognized as a highly prevalent finding. A recent meta- analysis has shown that the postmortem prevalence of pituitary adenoma is 14.4% and that radiologic studies identify a lesion consistent with pituitary adenoma in 22.2% of the population, providing an overall estimated prevalence of 16.9%.3 Although many of these lesions are considered to be incidental findings, many have unrecognized impact on fertility, longevity, and quality of life, and their clinical significance is increasingly gaining attention. Moreover, the management of these lesions has seen major changes with the development of new pharmacotherapeutic agents, improved minimally invasive surgical approaches, and targeted radiotherapeutic techniques. The surgical pathologist must therefore recognize the important role of morphologic analysis in classifying sellar pathology for the diagnosis and management of the pituitary patient.
THE ROLE OF CLINICAL INFORMATION
The importance of clinical information cannot be overemphasized in this field. Patients with pituitary disease may present with symptoms and signs of hormone excess, or they may manifest features of a mass lesion, including headache, visual impairment, and hypopituitarism. The former usually indicates a primary adenohypophysial disorder, but it should be recognized that hyperprolactinemia may be a nonspecific finding because of a mass lesion that obstructs the pituitary stalk, interrupting blood flow that maintains prolactin (PRL) under tonic inhibition. The latter can be the result of any mass lesion in the region of the sella. The finding of diabetes insipidus or cranial nerve dysfunction make the diagnosis of a primary adenohypophysial cell proliferation unlikely and instead suggest other tumor types or inflammatory disorders.
Despite the importance of clinicopathologic correlation, the reality is that many pathologists are faced with diagnosing a lesion without clinical information. In most instances it is possible to determine a remarkable amount of information with careful morphologic evaluation using a targeted approach.
AT THE TIME OF SURGERY
The initial handling of tissue obtained at pituitary surgery should ensure adequate fixation in formalin for histology and immunohistochemistry. In rare cases, there may be a need for ultrastructural analysis; because this situation is not often predicted clinically, it is recommended that a small piece of tissue be routinely fixed for electron microscopy and retained in the event that it is needed. Currently there is no need for special handling of tissue for other diagnostic techniques.
Most pituitary specimens are very small and the tissue may be compromised by freezing artifact when surgeons request intraoperative consultation and frozen sections are performed. In some centers, pathologists use smear technology for intraoperative consultation to prevent this artifact; this method uses less tissue but requires experience for interpretation. Sometimes, the only diagnostic tissue is in the material used for the intraoperative procedure and it is fraught with artefact that precludes accurate evaluation. Because valid indication for intraoperative consultation is rare, this procedure should be restricted for use only when the clinical situation is unusual or when the surgeon encounters unexpected findings and the intraoperative diagnosis might change the surgical approach.
THE AUTOPSY PITUITARY
The high prevalence of pituitary adenomas means that pathologists will often identify these lesions as incidental findings in autopsies. The main concern is handling of the material for analysis. It is recommended that the sella turcica be examined after the brain is removed; the hypophysial stalk should be cut as high as possible to leave the gland intact. In the case of a large lesion, the sella may be eroded to the point where it must be resected en bloc. However, in most patients the sellar diaphragm can be opened and the dorsum sellae fractured to push it posteriorly, allowing the gland to be removed intact. The gland then can be evaluated grossly and sectioned for complete histologic evaluation.
There are 2 approaches to the sectioning and embedding of the pituitary (Figure 1). Many investigators use sagittal sections through the gland; others prefer transverse sections. The former permit examination of the stalk; the latter provide a more thorough examination of the gland and more accurate determination of the geographic distribution of the various cell types, at the expense of examining the stalk carefully.
The initial evaluation of a pituitary specimen involves review of material stained with hematoxylin-eosin. This routine stain allows the distinction of primary adenohypophysial pathologies from other entities. Rathke cleft cysts, arachnoid cysts, and dermoid cysts (Figure 2) are recognized based on preoperative clinical and radiologic findings and confirmed with the identification of the appropriate cyst lining.4 Hypophysitis of any type5 can be readily recognized with this conventional stain (Figure 3). The various tumors that arise in the sella-gliomas, meningiomas, schwannomas, and chordomas-are considered based on this analysis and their workup is different than that of a pituitary adenoma. Unusual hypothalamic neuronal gangliocytomas and gangliogliomas can give rise to clinical features of hormone excess that can mimic pituitary adenoma,6 but must be recognized, either distinct from an adenoma or associated with one.
Craniopharyngioma is a unique tumor of the sellar region that is derived from the oropharyngeal remnants of Rathke pouch. These lesions have a characteristic morphology that requires only routine hematoxylin-eosin staining for identification and classification (Figure 4). They are composed of cords or islands of squamoid epithelial cells in a loose fibrous stroma with varying degrees of desquamation and intervening cysts that often contain a thick oily fluid.1 They can be subclassified as adamantinomatous and papillary types; the former are known to harbor mutations of the beta-catenin gene as a specific molecular pathogenetic mechanism.7,8 These lesions have a bimodal distribution with peaks in childhood and in the sixth decade. Although adamantinomatous lesions predominate in childhood, most craniopharyngiomas in adults have a mixed pattern and the clinical significance of subclassification remains uncertain.
Germ cell tumors of the sella resemble germ cell tumors in other sites of the body.9 Hematologic malignancies of the sella are usually systemic disorders but occasionally arise as primary plasmacytomas or lymphomas.10-12
Metastatic malignancies are common in the pituitary,13,14 usually at a late stage of disseminated malignancy when the clinical diagnosis is known. However, occasionally these lesions are detected early, and particularly in patients with no known history of a primary malignancy, they can be clinically challenging. The lesions that most frequently give rise to pituitary metastasis are breast, lung, and prostate carcinomas. The metastatic tumor usually involves the neurohypophysis and extrasellar structures, creating a clinical presentation of diabetes insipidus and nerve palsies that is not consistent with pituitary adenoma. 15,16 The common lesions are readily diagnosed on routine histopathology, but a metastatic well- differentiated neuroendocrine carcinoma (Figure 5) can be a diagnostic dilemma.15,17 The expression of specific markers, such as thyroid transcription factor 1, CDX-2, or peptides of gut or lung origin, may be required to distinguish these lesions from a primary pituitary neoplasm. PRIMARY ADENOHYPOPHYSIAL CELL PROLIFERATIONS
Once a pituitary lesion is determined to be composed of epithelial cells with neuroendocrine differentiation and thought to be of primary pituitary origin, the classification of the lesion will require several steps. First, the lesion must be identified as hyperplasia or neoplasia. Second, the cell population responsible for the proliferation must be established. Finally, in the case of a neoplasm, the behavior, prognosis, and potential therapy of choice must be determined.
ADENOMA OR HYPERPLASIA?
Hyperplasia is controlled cell proliferation that is induced by a stimulus and stops when the stimulus is removed. Pituitary hyperplasia can be physiologic, as when lactotrophs proliferate during pregnancy, or pathologic when induced by excess hypophysiotropic hormones. Examples of the latter include cases of primary target organ failure, such as primary hypothyroidism,18,19 or hormone excess produced by neoplasms, such as hypothalamic gangliocytomas or ectopic sources of growth hormone-releasing hormone or corticotrophin-releasing hormone, such as bronchial, gastroenteropancreatic, adrenal, or prostatic endocrine tumors.20 Hyperplasia may be clinically indistinguishable from adenoma, usually in patients with acromegaly or Cushing disease. Radiologic imaging sometimes identifies differences; in hyperplasia, the proliferation is diffuse and there is no normal rim that enhances with gadolinium.21 However, this subtle difference is often overlooked and it falls to the pathologist to make the diagnosis. In this regard, the reticulin stain is a very useful tool. Normal adenohypophysis is composed of small acini of pituitary cells surrounded by an intact reticulin network (Figure 6, a). In hyperplasia, the acinar architecture is maintained and the reticulin network is preserved, but the acini are increased in size (Figure 6, b). In contrast, pituitary adenomas are characterized by complete disruption of the reticulin fiber network (Figure 6, c). Immunohistochemical stains are required to determine the hyperplastic cell population, and these stains will identify the admixed normal cells that contain all of the normal adenohypophysial hormones.1
PITUITARY ADENOMA CLASSIFICATION
Clinically pituitary adenomas are classified as hormonally active functioning adenomas and nonfunctioning adenomas that often present with visual impairment and hypopituitarism. Approximately two thirds of clinically diagnosed lesions are functioning adenomas.
Pituitary adenomas are also classified based on size and invasiveness. Microadenomas are defined as less than 1 cm; macroadenomas are larger than 1 cm. Large tumors growing upward are defined as showing suprasellar extension. Tumors are also classified radiologically and by the neurosurgeon as invasive or not, based on their infiltration into surrounding structures (dura, bone, sinuses, etc).
The pituitary is composed of at least 6 distinct cell types. Each cell is responsible for the production and secretion of at least 1 hormone. Recent advances in molecular biology have clarified 3 major pathways of cytodifferentiation of adenohypophyseal cells (Figure 7) that are determined by a complex pattern of transcription factor expression.22,23 These transcription factors can help in classifying adenomas.24-28
Corticotrophs differentiate first in the human fetal pituitary and the expression of the proopiomelanocortin (POMC) gene is regulated by the Tpit transcription factor29 that mediates its action in concert with Ptx1 and neuroD1,30,31 which were previously identified as corticotroph upstream transcription element-binding proteins. The second line of differentiation temporally in the human gland is determined by Pit-1, a protein that activates the growth hormone (GH), PRL, and beta-thyrotropin (beta-thyroid- stimulating hormone [TSH]) genes.32-37 Pit-1 initiates GH expression and somatotroph differentiation. Expression of estrogen receptor allows the expression of PRL and GH a bihormonal population of mammosomatotrophs.38 The development of mature lactotrophs is dependent on the presence of a putative GH repressor that has yet to be identified. Some of the Pit-1-expressing cells further express thyrotroph embryonic factor39 and develop into thyrotrophs in the presence of a GH repressor and GATA-2.40 In physiologic states, somatotrophs, mammosomatotrophs, and lactotrophs transdifferentiate in what is thought to be a reversible fashion.41 It has been shown in animal models that somatotrophs can also transdifferentiate into thyrotrophs in severe hypothyroidism and this too is thought to be reversible.42 These changes indicate fluidity of 4 cell types that are all dependent on Pit-1. The third line of cytodifferentiation is that of the gonadotrophs whose hormone production is dependent on steroidogenic factor 1 and GATA-2 in the presence of estrogen receptor. 28,40
Each cell type can give rise to tumors that are clinically functioning or silent. Some tumor types have morphologic variants based on patterns of immunoreactivity for hormones and subcellular structures and, in occasional cases, ultrastructural features1,2; the variants are thought to re- flect differing pathogenetic mechanisms and may predict differing responses to therapy. The current clinicopathologic classification of pituitary adenomas is shown in Table 2 and the detailed morphologic subclassification is outlined in Table 3.
The Pathology of Hormone Excess Syndromes
Most patients with Cushing disease have small lesions that are difficult to localize by magnetic resonance imaging. The differential diagnosis is pituitary adenoma versus corticotroph hyperplasia. The former is far more common, but the distinction is important and requires the use of reticulin staining and adrenocorticotropic hormone (ACTH) immunohistochemistry. The classical microadenoma is a densely granulated adenoma composed of strongly basophilic cells. These adenomas exhibit strong positivity with the periodic acid-Schiff stain. Immunohistochemically, they demonstrate expression of ACTH and generally have very strong reactivity with the CAM 5.2 antibody to keratins 7 and 8.
The pathologist should also examine the nontumorous gland to determine if there is Crooke hyaline change, a morphologic marker of feedback suppression that is usually found in nontumorous corticotrophs and confirms that the patient has elevated circulating glucocorticoids (Figure 8). This change is seen in patients with pituitary corticotroph adenoma, ectopic ACTH secretion, primary adrenal pathology, or iatrogenic administration of glucocorticoids. When this change is present but no tumor is seen, the pathologist must search for a microadenoma that may be only 1 to 2 mm and multiple sections through the specimen may be required to find the lesion. In the absence of an identified adenoma, the pathologist can only issue a report that indicates the presence of Crooke hyaline, consistent with Cushing syndrome, and the outcome of surgery alone will indicate the true nature of this disorder. Because very small microadenomas can be lost during surgery, perhaps suctioned during aspiration of blood in the operative field, an operative success can be assumed if biochemical normality is achieved along with regression of clinical signs and symptoms. In contrast, a surgical failure will require more careful clinical evaluation of the patient to exclude an ectopic source of ACTH or ACTH-like peptide, primary adrenal disease, or a missed pituitary lesion elsewhere in the gland.
In a patient in whom no tumor is found and there is no Crooke hyaline change of nontumorous corticotrophs, the diagnosis becomes complex. One possibility is that the patient has corticotroph hyperplasia; this diagnosis requires a careful evaluation of reticulin and corticotroph distribution that can be very difficult.43,44 Another possibility is pseudo-Cushing, a significant medical pitfall that can occasionally result in unnecessary pituitary surgery.
Generally, larger lesions tend to be obvious adenomas but not obviously basophilic adenomas, because they are usually sparsely granulated, chromophobic adenomas. The presence of periodic acid- Schiff positivity and weak ACTH reactivity makes the diagnosis evident, but these stains can also be equivocal in this setting. The addition of Tpit is very helpful.
Crooke cell adenoma is a rare variant of corticotroph adenoma. In this unusual lesion, the adenomatous cells exhibit the features of suppressed corticotrophs. These tumors are often associated with atypical clinical histories, and the diagnosis may be unclear, or there may be a history of cyclical Cushing syndrome. The morphology of these lesions is quite atypical, with prominent nuclear pleomorphism and large cells that can resemble gangliocytoma or metastatic carcinoma. Periodic acid-Schiff positivity and immunoreactivity for ACTH as well as the dense ring of keratin that fills the tumor cell cytoplasm and is identified with CAM 5.2 define this rare entity.
When the patient is known to have acromegaly or gigantism, the diagnosis of a GH-secreting adenoma is almost certain. However, rarely these patients can have somatotroph hyperplasia resulting from ectopic production of growth hormone-releasing hormone by endocrine tumors as described previously. Reticulin is a critical stain to exclude this possibility and ensure appropriate management for these patients.
Growth hormone-secreting adenomas may be monohormonal somatotroph adenomas, bihormonal mammosomatotroph adenomas, or plurihormonal adenomas of the Pit-1 family that also make TSH. Monohormonal somatotroph adenomas can be densely granulated or sparsely granulated. The distinctions can impact medical therapy in the event of surgical therapy45,46; therefore, accurate classification is important. Indeed, it appears that the most important distinction is between densely and sparsely granulated types, because the pathophysiology of these lesions will determine their response to the current therapies that are available: somatostatin analogues versus GH-antagonists and possibly dopamine agonists.47 All of these adenomas exhibit nuclear Pit-1 reactivity and variable cytoplasmic GH positivity. Mammosomatotrophs stain for PRL as well, and the unusual plurihormonal adenomas contain beta-TSH. All of the densely granulated variants are acidophilic and also contain alpha-subunit. The sparsely granulated somatotroph adenoma is the most difficult to diagnose, because it is often either negative or only weakly positive for GH. Indeed the most critical immunostain in this setting is the CAM 5.2 keratin stain. It identifies perinuclear keratin in densely granulated adenomas, including mammosomatotrophs and plurihormonal lesions (Figure 9, a). In contrast, it clearly decorates fi- brous bodies in the sparsely granulated adenomas (Figure 9, b). These fibrous bodies can often be recognized without the keratin stain. The tumor cells often have bilobed or concave, pleomorphic nuclei that are distorted by pale homogenous eosinophilic globules.
The patient presenting with hyperprolactinemia usually is treated with medical therapy. Patients who come to surgery either have failed medical therapy or suffer signifi- cant adverse effects induced by all of the dopaminergic agonists now available. Because most lactotroph adenomas respond well to these drugs with hormone normalization and tumor shrinkage, it is important for the pathologist to exclude the many other causes of hyperprolactinemia, including hypophysitis and all of the various nonadenohypophysial neoplasms identified previously.
Lactotroph adenomas are subclassified into sparsely granulated and densely granulated variants. Sparsely granulated adenomas are usually highly responsive to dopamine agonists and therefore usually exhibit major changes because of the previous therapy. Only untreated adenomas of this type exhibit the usually chromophobic morphology with abundant cytoplasm and characteristic juxtanuclear PRL immunoreactivity (Figure 10). More commonly, the lesions are composed of small cells in a fibrous stroma, resembling inflammation, plasmacytoma, or lymphoma. The diagnosis is confirmed by the identi- fication of strong nuclear positivity for Pit-1; usually they have at least focal PRL positivity. The rare densely granulated lactotroph adenomas are composed of acidophilic cells with strong and diffuse cytoplasmic positivity for PRL. Another unusual pituitary adenoma causing hyperprolactinemia is the so- called acidophil stem cell adenoma, an oncocytic lesion characterized by Pit-1 nuclear staining, variable PRL and GH reactivity, and fibrous bodies identified with the CAM 5.2 immunostain.
The presentation of a patient with TSH excess requires the exclusion of thyrotroph hyperplasia (see "Adenoma or Hyperplasia?"). The rare thyrotroph adenomas are usually highly infiltrative macroadenomas with stromal fibrosis and marked nuclear atypia. Immunohistochemically, thyrotroph adenomas express alpha-subunit and beta-TSH.
Clinically Nonfunctioning Adenomas
The diagnosis of a clinically nonfunctioning adenoma requires appropriate classification for prognostication. The majority of these lesions are gonadotroph adenomas; these very rarely present with clinical or biochemical evidence of hormone excess. Nevertheless, they produce folliclestimulating hormone and/or luteinizing hormone and they express the transcription factors that prove gonadotroph differentiation. They have a highly characteristic histologic pattern, in which solid sheets, nests, and even sinusoidal patterns are interrupted by pseudopapillae and striking pseudorosettes around vascular channels (Figure 11). They usually are composed of admixtures of 2 cell types: tall columnar cells line pseudopapillae and rosettes, and polygonal cells comprise the bulk of the lesion. Oncocytic change can be observed in all patterns. Gonadotroph adenomas express alpha-subunit, beta-follicle- stimulating hormone, and beta-luteinizing hormone in scattered patterns and to variable degrees; they also express steroidogenic factor 1 with strong nuclear reactivity.
Occasional clinically silent adenomas are positive for Pit-1 and GH, PRL, or beta-TSH; these lesions should be classified as adenomas of the appropriate type as indicated previously, with the additional qualification of "silent" adenoma. Silent corticotroph adenomas are thought to arise from cells that fail to process the ACTH precursor, proopiomelanocortin, into the biologically active 1-39 ACTH. These lesions manifest Tpit and ACTH immunoreactivity and they are strongly positive for keratins 7 and 8. Indeed they resemble functioning corticotroph adenomas of the 2 types, sparsely and densely granulated variants; however, they are invariably macroadenomas and there is no associated Crooke hyaline in nontumorous corticotrophs. These lesions are generally much more aggressive than other silent adenomas, and recurrence is extremely common. The pathophysiology of the lack of clinical symptomatology of other silent adenomas is not known.
As immunohistochemical markers become more sophisticated, the number of truly unclassified adenomas is falling. The rare tumor that is completely negative for all hormones and transcription factors is classified as a "null cell adenoma." These usually behave like gonadotroph adenomas.
The Question of Plurihormonality
Reports of various combinations of hormones in unusual plurihormonal pituitary adenomas were extremely common in the past. However, the application of highly specific monoclonal antibodies and the understanding of cell differentiation have clarified many of the controversies. Reports of adenomas expressing GH or PRL with gonadotropins are now recognized to reflect nonspecific crossreactivity. 48 The fact that cells of the Pit-1 lineage express alpha-subunit and that many antisera raised against folliclestimulating hormone or luteinizing hormone recognized alpha- subunit explains many of these anomalies. The use of high-quality monoclonal antisera has made the occurrence of unusual plurihormonal profiles exceptionally rare. Indeed, some of these lesions represent double adenomas or "collision" tumors.49 Most lesions respect the lines of differentiation attributable to the 3 transcription factor lineages. Even the rare silent subtype 3 adenoma is usually positive for Pit-1, PRL, GH, and beta-TSH with other reactivities likely reflecting alpha-subunit cross-reactivity. This lesion is characterized by intense stromal fibrosis and high vascularity. Other distinct features are identified by electron microscopy.
The Role of Electron Microscopy
The classification of pituitary adenomas is based on careful studies that used immunohistochemistry, electron microscopy, and immunoelectron microscopy to identify structure-function correlations.50 Many of the ultrastructural features that were recognized as characterizing specific tumor types are now identified by immunohistochemistry. For example, fibrous bodies were considered the hallmark of the sparsely granulated somatotroph adenoma, and these are now readily identified with the CAM 5.2 immunostain.
There remain situations in which the histology and immunohistochemical profile are atypical and these cases require electron microscopy for accurate classification. The best approach to the use of this diagnostic tool is to fix and embed a small fragment of all pituitary tumors at the time of receipt in the event that electron microscopy may be required, recognizing that only a small number of specimens will ever require sectioning and ultrastructural examination. If this is impractical, certainly specimens from patients with atypical histories deserve this type of handling, because they will be the cases most likely to require this ancillary study.
Prognostication remains a major challenge in pituitary pathology. Proliferative activity51-53 using markers such as proliferating cell nuclear antigen, Ki-67/MIB-1, and antiapoptotic Bcl-2 have unfortunately demonstrated no consistent correlation with tumor invasiveness or recurrence.52 Although invasive pituitary adenomas and carcinomas exhibit a high DNA topoisomerase IIalpha index, this indicator has no significant advantage over MIB-1 as a prognostic marker.54 Cyclooxygenase 2 expression correlates with patient age, but not with tumor size or invasiveness.55 Detection of telomerase expression may predict recurrence in pituitary adenomas.56 Galectin- 3, a beta-galactoside-binding protein implicated in cellular differentiation and proliferation as well as angiogenesis, tumor progression, and metastasis, may play a role in pituitary tumor progression.57
Unfortunately, none of these is a true marker of biologic behavior. The best predictive marker remains the tumor classification based on hormone content and cell structure. For example, among acromegalics who fail surgical resection, response to long-acting somatostatin analogues is best predicted by the subtype of somatotroph adenoma as densely or sparsely granulated.45,46 This finding renders the value of a CAM 5.2 keratin stain more important than almost any other immunostain in this setting. A silent corticotroph adenoma will recur more often and more aggressively than a silent gonadotroph adenoma. A silent subtype 3 adenoma will almost certainly behave invasively, infiltrating the base of the skull, whereas a silent adenoma of the gonadotroph lineage will usually grow by expansion upward. Pituitary carcinoma, by definition a lesion that exhibits distant cerebrospinal and/or systemic metastasis, is an exceptionally rare lesion that cannot be defined by morphologic parameters of the primary tumor.
The approach to pituitary pathology is complex and requires recognition of many pathologic entities. Familiarity with inflammatory and neurologic diseases must be coupled with a detailed understanding of pituitary hyperplasia and adenoma classification. In the past, a diagnosis of "adenoma" was considered sufficient for many patients, but the advances in pituitary medicine demand a more thorough clinicopathologic diagnosis that will guide patient management.
Table 1. Classification of Pituitary Pathology*
Granular cell tumor
Vascular and mesenchymal tumors
Germ cell tumor
Lymphoma/leukemia/Langerhans cell histiocytosis
Vascular and mesenchymal tumors
Miscellaneous (salivary gland lesions, melanoma, etc)
Rathke cleft cysts
Brown tumor of bone
* Surgical pathology only, not including developmental and metabolic lesions that are not biopsied.
[dagger] Although classified as benign, many of these lesions are locally invasive and cause significant morbidity and mortality.
1. Asa SL. Tumors of the Pituitary Gland. Washington, DC: Armed Forces Institute of Pathology; 1998. Atlas of Tumor Pathology ; 3rd series, fascicle 22.
2. DeLellis RA, Lloyd RV, Heitz PU, Eng C. Tumours of Endocrine Organs. Lyon, France: IARC Press; 2004. World Health Organization Classification of Tumours.
3. Ezzat S, Asa SL, Couldwell WT, et al. The prevalence of pituitary adenomas: a systematic review. Cancer. 2004;101:613-619.
4. Shin JL, Asa SL, Woodhouse LJ, Smyth HS, Ezzat S. Cystic lesions of the pituitary: clinicopathological features distinguishing craniopharyngioma, Rathke's cleft cyst, and arachnoid cyst. J Clin Endocrinol Metab. 1999;84:3972-3982.
5. Cheung CC, Ezzat S, Smyth HS, Asa SL. The spectrum and significance of primary hypophysitis. J Clin Endocrinol Metab. 2001;86:1048-1053.
6. Puchner MJA, Ludecke DK, Saeger W, Riedel M, Asa SL. Gangliocytomas of the sellar region-a review. Exper Clin Endocrinol. 1995;103:129-149.
7. Sarubi JC, Bei H, Adams EF, et al. Clonal composition of human adamantinomatous craniopharyngiomas and somatic mutation analyses of the patched (PTCH), Gsalpha and Gi2alpha genes. Neurosci Lett. 2001;310(1):5-8.
8. Sekine S, Shibata T, Kokubu A, et al. Craniopharyngiomas of adamantinomatous type harbor beta-Catenin gene mutations. Am J Pathol. 2002;161:1997- 2001.
9. Jennings MT, Gelman R, Hochberg F. Intracranial germ-cell tumors: natural history and pathogenesis. J Neurosurg. 1985;63:155- 167.
10. Samaratunga H, Perry-Keene D, Apel RL. Primary lymphoma of the pituitary gland: a neoplasm of acquired MALT? Endocr Pathol. 1997;8:335-341.
11. Kuhn D, Buchfelder M, Brabletz T, Paulus W. Intrasellar malignant lymphoma developing within pituitary adenoma. Acta Neuropathol (Berl). 1999;97: 311-316.
12. Landman RE, Wardlaw SL, McConnell RJ, Khandji AG, Bruce JN, Freda PU. Pituitary lymphoma presenting as fever of unknown origin. J Clin Endocrinol Metab. 2001;86:1470-1476.
13. Roessmann U, Kaufman B, Friede RL. Metastatic lesions in the sella turcica and pituitary gland. Cancer. 1970;25:478-480.
14. Kovacs K. Metastatic cancer of the pituitary gland. Oncology. 1973;27: 533-542.
15. McCormick PC, Post KD, Kandji AD, Hays AP. Metastatic carcinoma to the pituitary gland. Br J Neurosurg. 1989;3:71-79.
16. Fassett DR, Couldwell WT. Metastases to the pituitary gland. Neurosurg Focus. 2004;16(4):E8.
17. Branch CL Jr, Laws ER Jr. Metastatic tumors of the sella turcica masquerading as primary pituitary tumors. J Clin Endocrinol Metab. 1987;65:469-474.
18. Khalil A, Kovacs K, Sima AAF, Burrow GN, Horvath E. Pituitary thyrotroph hyperplasia mimicking prolactin-secreting adenoma. J Endocrinol Invest. 1984;7: 399-404.
19. Kubota T, Hayashi M, Kabuto M, et al. Corticotroph cell hyperplasia in a patient with Addison disease: case report. Surg Neurol. 1992;37:441-447.
20. Sano T, Asa SL, Kovacs K. Growth hormone-releasing hormone- producing tumors: clinical, biochemical, and morphological manifestations. Endocr Rev. 1988;9:357-373.
21. Ezzat S, Asa SL, Stefaneanu L, et al. Somatotroph hyperplasia without pituitary adenoma associated with a long standing growth hormone-releasing hormone- producing bronchial carcinoid. J Clin Endocrinol Metab. 1994;78:555- 560.
22. Asa SL, Ezzat S. The cytogenesis and pathogenesis of pituitary adenomas. Endocr Rev. 1998;19:798-827.
23. Asa SL, Ezzat S. Molecular basis of pituitary development and cytogenesis. Front Horm Res. 2004;32:1-19.
24. Asa SL, Puy LA, Lew AM, Sundmark VC, Elsholtz HP. Cell type- specific expression of the pituitary transcription activator Pit-1 in the human pituitary and pituitary adenomas. J Clin Endocrinol Metab. 1993;77:1275-1280.
25. Friend KE, Chiou Y-K, Laws ER Jr, Lopes MBS, Shupnik MA. Pit- 1 messenger ribonucleic acid is differentially expressed in human pituitary adenomas. J Clin Endocrinol Metab. 1993;77:1281-1286.
26. Friend KE, Chiou YK, Lopes MBS, Laws ER Jr, Hughes KM, Shupnik MA. Estrogen receptor expression in human pituitary: correlation with immunohistochemistry in normal tissue, and immunohistochemistry and morphology in macroadenomas. J Clin Endocrinol Metab. 1994;78:1497-1504.
27. Zafar M, Ezzat S, Ramyar L, Pan N, Smyth HS, Asa SL. Cell- specific expression of estrogen receptor in the human pituitary and its adenomas. J Clin Endocrinol Metab. 1995;80:3621-3627.
28. Asa SL, Bamberger A-M, Cao B, Wong M, Parker KL, Ezzat S. The transcription activator steroidogenic factor-1 is preferentially expressed in the human pituitary gonadotroph. J Clin Endocrinol Metab. 1996;81:2165-2170.
29. Lamolet B, Pulichino AM, Lamonerie T, et al. A pituitary cell- restricted T box factor, Tpit, activates POMC transcription in cooperation with Pitx homeoproteins. Cell. 2001;104:849-859.
30. Lamonerie T, Tremblay JJ, Lanctot C, Therrien M, GauthierY, Drouin J. Ptx1, a bicoid-related homeo box transcription factor involved in transcription of the pro-opiomelanocortin gene. Genes Dev. 1996;10:1284-1295.
31. Poulin G, Turgeon B, Drouin J. NeuroD1/beta2 contributes to cell-specific transcription of the proopiomelanocortin gene. Mol Cell Biol. 1997;17:6673- 6682.
32. Ingraham HA, Chen R, Mangalam HJ, et al. A tissue-specific transcription factor containing a homeodomain specifies a pituitary phenotype. Cell. 1988;55: 519-529.
33. Mangalam HJ, Albert VR, Ingraham HA, et al. A pituitary POU domain protein, Pit-1, activates both growth hormone and prolactin promoters transcriptionally. Genes Dev. 1989;3:946-958.
34. Li S, Crenshaw EB III, Rawson EJ, Simmons DM, Swanson LW, Rosenfeld MG. Dwarf locus mutants lacking three pituitary cell types result from mutations in the POU-domain gene pit-1. Nature. 1990;347:528-533.
35. Ingraham HA, Albert VR, Chen R, et al. A family of POU- domain and Pit-1 tissue-specific transcription factors in pituitary and neuroendocrine development. Annu Rev Physiol. 1990;52:773-791.
36. Rosenfeld MG. POU-domain transcription factors: pou-er-ful developmental regulators. Genes Dev. 1991;5:897-907.
37. Yan G, Pan WT, Bancroft C. Thyrotropin-releasing hormone action on the prolactin promotor is mediated by the POU protein Pit- 1. Mol Endocrinol. 1991; 5:535-541.
38. Day RN, Koike S, Sakai M, Muramatsu M, Maurer RA. Both Pit-1 and the estrogen receptor are required for estrogen responsiveness of the rat prolactin gene. Mol Endocrinol. 1990;4:1964-1971.
39. Drolet DW, Scully KM, Simmons DM, et al. TEF, a transcription factor expressed specifically in the anterior pituitary during embryogenesis, defines a new class of leucine zipper proteins. Genes Dev. 1991;5:1739-1753.
40. Scully KM, Rosenfeld MG. Pituitary development: regulatory codes in mammalian organogenesis. Science. 2002;295:2231-2235.
41. Frawley LS, Boockfor FR. Mammosomatotropes: presence and functions in normal and neoplastic pituitary tissue. Endocr Rev. 1991;12:337-355.
42. Horvath E, Lloyd RV, Kovacs K. Propylthiouracil-induced hypothyroidism results in reversible transdifferentiation of somatotrophs into thyroidectomy cells: a morphologic study of the rat pituitary including immunoelectron microscopy. Lab Invest. 1990;63:511-520.
43. Trouillas J, Guigard MP, Fonlupt P, Souchier C, Girod C. Mapping of corticotropic cells in the normal human pituitary. J Histochem Cytochem. 1996;44: 473-479.
44. McNicol AM. Patterns of corticotropic cells in the adult human pituitary in Cushing's disease. Diag Histopathol. 1981;4:335- 341.
45. Ezzat S, Kontogeorgos G, Redelmeier DA, Horvath E, Harris AG, Kovacs K. In vivo responsiveness of morphological variants of growth hormone-producing pituitary adenomas to octreotide. Eur J Endocrinol. 1995;133:686-690.
46. Bhayana S, Booth GL, Asa SL, Kovacs K, Ezzat S. The implication of somatotroph adenoma phenotype to somatostatin analog responsiveness in acromegaly. J Clin Endocrinol. Metab. 2005;90:6290- 6295.
47. Asa SL, DiGiovanni R, Jiang J, et al. A growth hormone receptor mutation impairs growth hormone autofeedback signaling in pituitary tumors. Cancer Res. 2007;67:7505-7511.
48. Labat-Moleur F, Trouillas J, Seret-Begue D, Kujas M, Delisle M-B, Ronin C. Evaluation of 29 monoclonal and polyclonal antibodies used in the diagnosis of pituitary adenomas: a collaborative study from pathologists of the Club Franc?ais de l'Hypophyse. Pathol Res Pract. 1991;187:534-538.
49. Jastania RA, Alsaad KO, Al Shraim M, Kovacs K, Asa SL. Double adenomas of the pituitary: transcription factors Pit-1, T-pit, and SF-1 identify cytogenesis and differentiation. Endocr Pathol. 2005;16:187-194. 50. Kovacs K, Horvath E. Tumors of the Pituitary Gland. Washington, DC: Armed Forces Institute of Pathology; 1986. Atlas of Tumor Pathology; 2nd series, fascicle 21.
51. Knosp E, Kitz K, Perneczky A. Proliferation activity in pituitary adenomas: measurement by monoclonal antibody Ki-67. Neurosurgery. 1989;25:927-930.
52. Amar AP, Hinton DR, Krieger MD,Weiss MH. Invasive pituitary adenomas: significance of proliferation parameters. Pituitary. 1999;2(2):117-122.
53. Thapar K, Kovacs K, Scheithauer BW, et al. Proliferative activity and invasiveness among pituitary adenomas and carcinomas: an analysis using the MIB-1 antibody. Neurosurgery. 1996;38:99-107.
54. Vidal S, Kovacs K, Horvath E, et al. Topoisomerase IIalpha expression in pituitary adenomas and carcinomas: relationship to tumor behavior. Mod Pathol. 2002;15:1205-1212.
55. Vidal S, Kovacs K, Bell D, Horvath E, Scheithauer BW, Lloyd RV. Cyclooxygenase- 2 expression in human pituitary tumors. Cancer. 2003;97:2814-2821.
56. Yoshino A, Katayama Y, Fukushima T, et al. Telomerase activity in pituitary adenomas: significance of telomerase expression in predicting pituitary adenoma recurrence. J Neurooncol. 2003;63:155-162.
57. Riss D, Jin L, Qian X, et al. Differential expression of galectin-3 in pituitary tumors. Cancer Res. 2003;63:2251-2255.
Sylvia L. Asa, MD, PhD
Accepted for publication January 30, 2008.
From the Department of Pathology, University Health Network, Toronto, Ontario.
The author has no relevant financial interest in the products or companies described in this article.
Reprints: Sylvia L. Asa, MD, PhD, Department of Pathology, University Health Network, 200 Elizabeth St, 11th Floor, Toronto, Ontario, Canada M5G 2C4 (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.