- Email: [email protected]
Silent Pituitary Corticotroph Carcinoma in a Young Dog
J. Comp. Path. 2012, Vol. 146, 327e331
Available online at www.sciencedirect.com
www.elsevier.com/locate/jcpa
NEOPLASTIC DISEASE
Silent Pituitary Corticotroph Carcinoma in a Young Dog S. Gestier*, R. W. Cook*, W. Agnew† and M. Kiupel‡ †
* Veterinary Science Diagnostic Services, School of Veterinary Science, University of Queensland, Gatton 4343, Veterinary Specialist Services, Underwood 4119, Australia and ‡ Diagnostic Center for Population and Animal Health, Michigan State University, Lansing, MI 48910, USA
Summary An 11-month-old neutered female weimaraner was humanely destroyed 6 days after an acute onset of neurological signs. At necropsy examination the pituitary gland was replaced by a large neoplastic mass that compressed and infiltrated the overlying hypothalamus. Small nodules were detected in the spleen, kidneys and stomach. Adrenal, thyroid and parathyroid glands were normal in size. The primary pituitary mass, visceral nodules and microscopical metastases detected within the ventricles and leptomeninges of the brain comprised polygonal, chromophobic neoplastic cells, which labelled strongly for adrenocorticotrophic hormone (ACTH) on immunohistochemical examination. These findings, in the absence of clinical or pathological evidence of pituitary-dependent hyperadrenocorticism, support a diagnosis of endocrinologically-inactive (‘silent’) pituitary corticotroph (ACTH-containing) carcinoma. Ó 2011 Elsevier Ltd. All rights reserved. Keywords: ACTH; corticotroph; dog; pituitary carcinoma
Most tumours of the pituitary gland in man (Asa and Ezzat, 2002) and animals (Capen, 2007; La Perle and Capen, 2007; Kiupel, 2008) are adenomas (Asa and Ezzat, 2002). Endocrinologically-active pituitary adenomas in man most frequently secrete prolactin (PRL) or adrenocorticotrophic hormone (ACTH) (Asa and Ezzat, 2002; Saeger et al., 2007), while those in dogs generally secrete ACTH (Capen, 2007; La Perle and Capen, 2007). Endocrinologicallyinactive (‘silent’) human and animal pituitary adenomas are usually diagnosed as macroadenomas causing local mass effects (disrupted secretion of pituitary trophic hormones and/or neurological dysfunction) (Capen, 2007; Saeger et al., 2007). Pituitary carcinomas meeting the strict diagnostic criterion of cerebrospinal or systemic metastasis are rare in people, accounting for only 0.1e0.2% of pituitary tumours (Ragel and Couldwell, 2004; Saeger et al., 2007; Pinchot et al., 2009), and have been reported rarely in dogs, cats, sheep and cattle (Kiupel, 2008). Correspondence to: S. Gestier (e-mail: [email protected]). 0021-9975/$ - see front matter doi:10.1016/j.jcpa.2011.08.002
Here, we describe an endocrinologically-inactive (‘silent’) pituitary corticotroph carcinoma that caused acute neurological disease in a young dog. A healthy 11-month-old, neutered female weimaraner suddenly developed behavioural changes and within 4 days was anorexic, depressed and unable to stand. Magnetic resonance imaging (MRI) of the brain with gadolinium contrast showed mild enhancement of a circumscribed subdiencephalic midline mass. Neurological signs, including nystagmus, episodes of opisthotonus and seizures, progressed in severity and the dog was humanely destroyed 6 days after clinical onset of central nervous system (CNS) disease. At necropsy examination the pituitary gland was replaced by a large, nodular, off-white mass protruding caudally from a broad-based attachment to the hypothalamus, which was compressed and infiltrated by the tumour (Fig. 1A, B). On the dorsocaudal and medial surfaces of the occipital lobe and the ventral pyriform area of the temporal lobe of the left cerebral hemisphere there were prominent areas of leptomeningeal congestion and haemorrhage Ó 2011 Elsevier Ltd. All rights reserved.
328
S. Gestier et al.
Fig. 1. The pituitary mass protrudes caudally from its broad-based attachment to the hypothalamus. (A) There is haemorrhage within the leptomeninges and brain parenchyma of the pyriform area of the left temporal lobe of the cerebrum. (B) The left side of the brain overlying the mass is markedly swollen and haemorrhagic.
(Fig. 1A). The ventral half of the left cerebral hemisphere extending from the thalamus through the basal ganglia was markedly swollen and haemorrhagic (Fig. 1B). The middle third of the spleen was contracted, with granular tan thickening of the capsule, which contained approximately 15 off-white nodules, 1e5 mm in diameter (Fig. 2). The left and right kidneys each contained a single, spherical, cortical nodule, 5 mm and 2 mm in diameter, respectively. On the gastric mucosal surface two circular plaques (2 mm and 4 mm in diameter) were present. The adrenal glands (20 10 4 mm), thyroid glands (approximately 20 5 4 mm) and parathyroid glands (approximately 1 mm in diameter) appeared normal. Samples of all lesions and major organs, including brain and spinal cord, were fixed in 10% neutral buffered formalin, processed routinely and embedded in paraffin wax. Sections were stained with haematoxylin and eosin (HE). Selected sections of the pituitary mass, brain and lesions in the spleen, kidneys and stomach were subject to immunohistochemistry (IHC) to determine expression of ACTH, follicle stimulating hormone (FSH) and growth hormone (GH). Dewaxing, antigen retrieval, labelling and counterstaining were performed with the BOND MAXÔ Automated Staining System
Fig. 2. Tumour nodules protrude from the capsular surface of the contracted middle segment of the spleen.
(Vision BioSystems/Leica, Bannockburn, Illinois, USA) using the Bond Compact Polymer Detection SystemÔ (Vision BioSystems/Leica). Antigen retrieval was achieved using the Bond Epitope Retrieval Solution 1Ô (Vision BioSystems/Leica) for 20 min. Sections were incubated with a rabbit polyclonal antibody against ACTH at a dilution of 1 in 500, a rabbit polyclonal antibody against GH (both from Dako North America, Carpinteria, California, USA) at a concentration of 1 in 1,000 and a mouse monoclonal antibody against FSH (Progen, Heidelberg, Germany) at a concentration of 1 in 100. Labelling was ‘visualized’ with 3, 30 -diaminobenzidine substrate (Vision BioSystems/Leica) and sections were counterstained with haematoxylin. For negative controls, the primary antibodies were replaced with homologous non-immune sera or buffer. Only cytoplasmic labelling within neoplastic cells was evaluated as positive. Effacing the normal microarchitecture of the pituitary gland and infundibular stalk and infiltrating the overlying hypothalamus and third ventricle were sheets of neoplastic cells with a fine fibrovascular stroma (Fig. 3A, B). Along one margin of the pituitary mass, a narrow zone of residual pars distalis was identified by its mixed population of adenohypophyseal cells. The neoplastic cells were polygonal with a small to moderate amount of faintlybasophilic (‘chromophobic’) cytoplasm and a large, oval, occasionally-notched nucleus with coarselystippled to vesicular chromatin and a prominent nuclear membrane (Fig. 3B). There was mild anisokaryosis, 0e2 mitotic figures per high power field (40 microscope objective) and localized areas of necrosis. Within the grossly-swollen left side of the brain there was marked neuropil oedema, congestion and localized haemorrhage. Segments of the cerebral and cerebellar leptomeninges were infiltrated by
Pituitary Carcinoma in a Dog
329
Fig. 3. (A) Hypothalamus, dog. Sheets of tumour cells infiltrate the hypothalamus and third ventricle. HE. 100. (B) Pituitary mass, dog. Tumour cells are polygonal, with chromophobic cytoplasm and a large, oval, occasionally-notched nucleus. There is one mitotic figure (arrowhead). HE. 350. (C) Left occipital cerebral cortex, dog. Leptomeningeal infiltration by metastatic tumour is associated with congestion and haemorrhage. HE. 200. (D) Spleen, dog. Metastatic tumour nodule within the capsule is surrounded by haemosiderosis and calcification (sidero-calcific plaque). HE. 100.
neoplastic cells, with associated congestion and haemorrhage (Fig. 3C). The central canal of the lumbar spinal cord contained a small nest of neoplastic cells. Within the capsule of the spleen, discrete nodules of neoplastic cells similar to those in the CNS were separated by areas of capsular haemosiderosis and mineralization (sidero-calcific plaque) (Fig. 3D). The renal nodules observed grossly comprised wellcircumscribed, unencapsulated sheets of similar neoplastic cells. In the stomach, the two plaques observed grossly each comprised a mucosal and submucosal infiltrate of neoplastic cells. Immunohistochemical labelling of the neoplastic cells of the pituitary mass, leptomeninges and tumour foci in the spleen, kidneys and stomach was strongly positive for ACTH (Fig. 4) and negative for FSH and GH. These findings supported a diagnosis of pituitary corticotroph (ACTH-containing) carcinoma, metastatic to cerebral and cerebellar leptomeninges, brain ventricles and spinal canal, spleen, kidneys and stomach, on the basis of a current histological classification system of endocrine tumours in domestic animals (Kiupel, 2008). Most reports of clinical disease attributable to pituitary tumours in dogs are in adult to aged animals (Meij
et al., 1998; Hanson et al., 2005; Capen, 2007; de Fornel et al., 2007). This large pituitary corticotroph carcinoma that caused CNS signs in an 11-month-old dog therefore appears to have developed earlier than previously reported canine pituitary tumours.
Fig. 4. Pituitary mass, dog. Strong ACTH immunoreactivity in the cytoplasm of tumour cells. IHC. 350.
330
S. Gestier et al.
Despite the local invasion and compression of the hypothalamus by the large pituitary mass and the presence of intraventricular and leptomeningeal metastases, the dog appeared clinically normal until it suddenly developed CNS signs. This is consistent with the observation that endocrinologicallyinactive pituitary tumours in dogs often attain considerable size before they cause clinical signs from local mass effect (Capen, 2007). In this case, the acute haemorrhagic changes within the brain associated with the tumour are likely to have contributed to the sudden onset and progression of the neurological signs. The metastases detected in the spleen, kidneys and stomach were too small to be clinically important; the sidero-calcific plaques associated with the capsular metastases in the spleen may reflect past episodes of tumour-related capsular haemorrhage. Pituitary adenomas may be enclosed microadenomas (generally <5 mm diameter in dogs) and expansile or locally-invasive macroadenomas (>5 mm diameter in dogs) (Kiupel, 2008). The most frequently reported pituitary tumours in dogs are endocrinologically-active corticotroph (ACTH-secreting) chromophobe microadenomas (mCAs) or macroadenomas (MCAs) that cause pituitary-dependent hyperadrenocorticism (PDH) (Cushing’s-like disease), with or without subsequent local mass effect, and endocrinologically-inactive chromophobe macroadenomas that cause local mass effect only (Capen, 2007; La Perle and Capen, 2007). The last would include non-functioning (‘silent’) corticotroph (ACTH-containing) adenomas (SCAs), which are defined immunohistochemically (by hormone content), rather than histochemically (by cytoplasmic staining as acidophilic, basophilic or chromophobic) (Asa and Ezzat, 2002; Karavitaki et al., 2007). In a recent study of human mCAs and MCAs associated with Cushing’s disease and SCAs, the SCA was characterized as ‘a macroadenoma revealed by tumour mass-related symptoms with no clinical signs of hypercortisolism, normal plasma cortisol levels and elevated levels of ACTH, and ACTH or proopiomelanocortin (POMC) peptide immunostaining’ of the tumour (Raverot et al., 2010). In people, most clinically-relevant pituitary adenomas associated with Cushing’s disease present as mCAs, while SCAs have variable invasion or compression of adjacent structures (Karavitaki et al., 2007). It is unclear why SCAs fail to produce PDH, but hypotheses include the incomplete or aberrant processing of POMC (the large precursor of ACTH) and secretion of less active ACTH by the tumour (Raverot et al., 2010). Unfortunately, neither blood ACTH nor cortisol was measured in this dog to definitively discount
PDH. However, there were no clinical signs suggestive of hyperadrenocorticism and, importantly, no evidence at necropsy examination of the bilateral adrenal cortical hypertrophy expected with PDH. The morphology of the primary tumour and lack of clinical or pathological evidence of PDH in the face of strongly positive ACTH immunohistochemical labelling of the tumour cells are consistent with an endocrinologically-inactive (‘silent’) pituitary corticotroph (ACTH-containing) carcinoma. The stringent diagnostic criterion for pituitary carcinoma in human and veterinary medicine is the presence of cerebrospinal or systemic metastasis. Local invasion of the hypothalamus or surrounding soft or bony tissues is not necessarily diagnostic for malignancy and can be a feature of adenomas (Asa and Ezzat, 2002; Ragel and Couldwell, 2004; Kiupel, 2008). Some reported cases of pituitary carcinoma in dogs (Sarfaty et al., 1988; Puente, 2003) could be classified as invasive macroadenomas due to their lack of metastasis (Kiupel, 2008). In the present case, the cellular pleomorphism and mitotic activity within the primary tumour and metastases were minimal and similar to those seen in many corticotroph macroadenomas in dogs. Haemorrhage and necrosis can occur within both pituitary macroadenomas and carcinomas (Capen, 2007). There are therefore no reliable histological criteria to distinguish malignant from benign pituitary neoplasms (Kiupel, 2008). Most human pituitary carcinomas (88%) are endocrinologically active, usually secrete ACTH (42%) or PRL (33%), and along with endocrinologicallyinactive pituitary corticotroph carcinomas, present as invasive macrotumours causing mass effect (Ragel and Couldwell, 2004). Most pituitary carcinomas in dogs are endocrinologically-inactive, invasive, chromophobic macrotumours causing mass effect (Capen, 2007). However, there is a report in an 8-year-old cocker spaniel of a large functional pituitary corticotroph carcinoma with invasive growth into the third ventricle and microscopical metastases to brain and leptomeninges, causing PDH without clinical evidence of mass effect (Boujon et al., 1991). The morphological similarity of the primary pituitary tumour in the present case, a ‘silent’ corticotroph (ACTH-containing) tumour, to the endocrinologicallyinactive pituitary chromophobe macroadenoma reported in dogs suggests that cases of the latter tumour should be tested immunohistochemically to identify those that are SCAs.
References Asa SL, Ezzat S (2002) The pathogenesis of pituitary tumors. Nature Reviews Cancer, 2, 836e849.
331
Pituitary Carcinoma in a Dog
Boujon CE, Ritz U, Rossi GL, Bestetti GE (1991) A clinicopathological study of canine Cushing’s disease caused by a pituitary carcinoma. Journal of Comparative Pathology, 105, 353e365. Capen CC (2007) Endocrine glands. In: Pathology of Domestic Animals, 5th Edit., Vol. 3, MG Maxie, Ed., Saunders Elsevier, Edinburgh, pp. 334e351. de Fornel P, Delisle F, Devauchelle P, Rosenberg D (2007) Effects of radiotherapy on pituitary corticotroph macrotumors in dogs: a retrospective study of 12 cases. Canadian Veterinary Journal, 48, 481e486. Hanson JM, van’t Hoofd MM, Voorhout G, Teske E, Kooistra HS et al. (2005) Efficacy of transsphenoidal hypophysectomy in treatment of dogs with pituitarydependent hyperadrenocorticism. Journal of Veterinary Internal Medicine, 19, 687e694. Karavitaki N, Ansorge O, Wass JAH (2007) Silent corticotroph adenomas. Arquivos Brasileiros de Endocrinologia e Metabologia, 51, 1314e1318. Kiupel M (2008)Histological Classification of Tumors of the Endocrine System of Domestic Animals, Vol. 12. Armed Forces Institute of Pathology, Washington DC, pp. 17e25. La Perle KMD, Capen CC (2007) Endocrine system. In: Pathologic Basis of Veterinary Disease, 4th Edit., MD McGavin, JF Zachary, Eds., Elsevier Mosby, St Louis, pp. 693e716. Meij BP, Voorhout G, van den Ingh TSGAM, Hazewinkel HAW, Teske E et al. (1998) Results of transsphenoidal hypophysectomy in 52 dogs with pituitarydependent hyperadrenocorticism. Veterinary Surgery, 27, 246e261.
Pinchot SN, Sippel R, Chen H (2009) ACTH-producing carcinoma of the pituitary with refractory Cushing’s disease and hepatic metastases: a case report and review of the literature. World Journal of Surgical Oncology, 7, 39e45. Puente S (2003) Pituitary carcinoma in an Airedale terrier. Canadian Veterinary Journal, 44, 240e242. Ragel BT, Couldwell WT (2004) Pituitary carcinoma: a review of the literature. Neurosurgical Focus, 16, E7. Raverot G, Wierinckx A, Jouanneau E, Auger C, BorsonChazot F et al. (2010) Clinical, hormonal and molecular characterization of pituitary ACTH adenomas without (silent corticotroph adenomas) and with Cushing’s disease. European Journal of Endocrinology, 163, 35e43. Saeger W, Ludecke DK, Buchfelder M, Fahlbusch R, Quabbe HJ et al. (2007) Pathohistological classification of pituitary tumors: 10 years of experience with the German Pituitary Tumor Registry. European Journal of Endocrinology, 156, 203e216. Sarfaty D, Carrillo JM, Peterson ME (1988) Neurologic, endocrinologic, and pathologic findings associated with large pituitary tumors in dogs: eight cases (1976e1984). Journal of the American Veterinary Medical Association, 193, 854e856.
April 1st, 2011 ½ Received, Accepted, August 7th, 2011
Available online at www.sciencedirect.com
www.elsevier.com/locate/jcpa
NEOPLASTIC DISEASE
Silent Pituitary Corticotroph Carcinoma in a Young Dog S. Gestier*, R. W. Cook*, W. Agnew† and M. Kiupel‡ †
* Veterinary Science Diagnostic Services, School of Veterinary Science, University of Queensland, Gatton 4343, Veterinary Specialist Services, Underwood 4119, Australia and ‡ Diagnostic Center for Population and Animal Health, Michigan State University, Lansing, MI 48910, USA
Summary An 11-month-old neutered female weimaraner was humanely destroyed 6 days after an acute onset of neurological signs. At necropsy examination the pituitary gland was replaced by a large neoplastic mass that compressed and infiltrated the overlying hypothalamus. Small nodules were detected in the spleen, kidneys and stomach. Adrenal, thyroid and parathyroid glands were normal in size. The primary pituitary mass, visceral nodules and microscopical metastases detected within the ventricles and leptomeninges of the brain comprised polygonal, chromophobic neoplastic cells, which labelled strongly for adrenocorticotrophic hormone (ACTH) on immunohistochemical examination. These findings, in the absence of clinical or pathological evidence of pituitary-dependent hyperadrenocorticism, support a diagnosis of endocrinologically-inactive (‘silent’) pituitary corticotroph (ACTH-containing) carcinoma. Ó 2011 Elsevier Ltd. All rights reserved. Keywords: ACTH; corticotroph; dog; pituitary carcinoma
Most tumours of the pituitary gland in man (Asa and Ezzat, 2002) and animals (Capen, 2007; La Perle and Capen, 2007; Kiupel, 2008) are adenomas (Asa and Ezzat, 2002). Endocrinologically-active pituitary adenomas in man most frequently secrete prolactin (PRL) or adrenocorticotrophic hormone (ACTH) (Asa and Ezzat, 2002; Saeger et al., 2007), while those in dogs generally secrete ACTH (Capen, 2007; La Perle and Capen, 2007). Endocrinologicallyinactive (‘silent’) human and animal pituitary adenomas are usually diagnosed as macroadenomas causing local mass effects (disrupted secretion of pituitary trophic hormones and/or neurological dysfunction) (Capen, 2007; Saeger et al., 2007). Pituitary carcinomas meeting the strict diagnostic criterion of cerebrospinal or systemic metastasis are rare in people, accounting for only 0.1e0.2% of pituitary tumours (Ragel and Couldwell, 2004; Saeger et al., 2007; Pinchot et al., 2009), and have been reported rarely in dogs, cats, sheep and cattle (Kiupel, 2008). Correspondence to: S. Gestier (e-mail: [email protected]). 0021-9975/$ - see front matter doi:10.1016/j.jcpa.2011.08.002
Here, we describe an endocrinologically-inactive (‘silent’) pituitary corticotroph carcinoma that caused acute neurological disease in a young dog. A healthy 11-month-old, neutered female weimaraner suddenly developed behavioural changes and within 4 days was anorexic, depressed and unable to stand. Magnetic resonance imaging (MRI) of the brain with gadolinium contrast showed mild enhancement of a circumscribed subdiencephalic midline mass. Neurological signs, including nystagmus, episodes of opisthotonus and seizures, progressed in severity and the dog was humanely destroyed 6 days after clinical onset of central nervous system (CNS) disease. At necropsy examination the pituitary gland was replaced by a large, nodular, off-white mass protruding caudally from a broad-based attachment to the hypothalamus, which was compressed and infiltrated by the tumour (Fig. 1A, B). On the dorsocaudal and medial surfaces of the occipital lobe and the ventral pyriform area of the temporal lobe of the left cerebral hemisphere there were prominent areas of leptomeningeal congestion and haemorrhage Ó 2011 Elsevier Ltd. All rights reserved.
328
S. Gestier et al.
Fig. 1. The pituitary mass protrudes caudally from its broad-based attachment to the hypothalamus. (A) There is haemorrhage within the leptomeninges and brain parenchyma of the pyriform area of the left temporal lobe of the cerebrum. (B) The left side of the brain overlying the mass is markedly swollen and haemorrhagic.
(Fig. 1A). The ventral half of the left cerebral hemisphere extending from the thalamus through the basal ganglia was markedly swollen and haemorrhagic (Fig. 1B). The middle third of the spleen was contracted, with granular tan thickening of the capsule, which contained approximately 15 off-white nodules, 1e5 mm in diameter (Fig. 2). The left and right kidneys each contained a single, spherical, cortical nodule, 5 mm and 2 mm in diameter, respectively. On the gastric mucosal surface two circular plaques (2 mm and 4 mm in diameter) were present. The adrenal glands (20 10 4 mm), thyroid glands (approximately 20 5 4 mm) and parathyroid glands (approximately 1 mm in diameter) appeared normal. Samples of all lesions and major organs, including brain and spinal cord, were fixed in 10% neutral buffered formalin, processed routinely and embedded in paraffin wax. Sections were stained with haematoxylin and eosin (HE). Selected sections of the pituitary mass, brain and lesions in the spleen, kidneys and stomach were subject to immunohistochemistry (IHC) to determine expression of ACTH, follicle stimulating hormone (FSH) and growth hormone (GH). Dewaxing, antigen retrieval, labelling and counterstaining were performed with the BOND MAXÔ Automated Staining System
Fig. 2. Tumour nodules protrude from the capsular surface of the contracted middle segment of the spleen.
(Vision BioSystems/Leica, Bannockburn, Illinois, USA) using the Bond Compact Polymer Detection SystemÔ (Vision BioSystems/Leica). Antigen retrieval was achieved using the Bond Epitope Retrieval Solution 1Ô (Vision BioSystems/Leica) for 20 min. Sections were incubated with a rabbit polyclonal antibody against ACTH at a dilution of 1 in 500, a rabbit polyclonal antibody against GH (both from Dako North America, Carpinteria, California, USA) at a concentration of 1 in 1,000 and a mouse monoclonal antibody against FSH (Progen, Heidelberg, Germany) at a concentration of 1 in 100. Labelling was ‘visualized’ with 3, 30 -diaminobenzidine substrate (Vision BioSystems/Leica) and sections were counterstained with haematoxylin. For negative controls, the primary antibodies were replaced with homologous non-immune sera or buffer. Only cytoplasmic labelling within neoplastic cells was evaluated as positive. Effacing the normal microarchitecture of the pituitary gland and infundibular stalk and infiltrating the overlying hypothalamus and third ventricle were sheets of neoplastic cells with a fine fibrovascular stroma (Fig. 3A, B). Along one margin of the pituitary mass, a narrow zone of residual pars distalis was identified by its mixed population of adenohypophyseal cells. The neoplastic cells were polygonal with a small to moderate amount of faintlybasophilic (‘chromophobic’) cytoplasm and a large, oval, occasionally-notched nucleus with coarselystippled to vesicular chromatin and a prominent nuclear membrane (Fig. 3B). There was mild anisokaryosis, 0e2 mitotic figures per high power field (40 microscope objective) and localized areas of necrosis. Within the grossly-swollen left side of the brain there was marked neuropil oedema, congestion and localized haemorrhage. Segments of the cerebral and cerebellar leptomeninges were infiltrated by
Pituitary Carcinoma in a Dog
329
Fig. 3. (A) Hypothalamus, dog. Sheets of tumour cells infiltrate the hypothalamus and third ventricle. HE. 100. (B) Pituitary mass, dog. Tumour cells are polygonal, with chromophobic cytoplasm and a large, oval, occasionally-notched nucleus. There is one mitotic figure (arrowhead). HE. 350. (C) Left occipital cerebral cortex, dog. Leptomeningeal infiltration by metastatic tumour is associated with congestion and haemorrhage. HE. 200. (D) Spleen, dog. Metastatic tumour nodule within the capsule is surrounded by haemosiderosis and calcification (sidero-calcific plaque). HE. 100.
neoplastic cells, with associated congestion and haemorrhage (Fig. 3C). The central canal of the lumbar spinal cord contained a small nest of neoplastic cells. Within the capsule of the spleen, discrete nodules of neoplastic cells similar to those in the CNS were separated by areas of capsular haemosiderosis and mineralization (sidero-calcific plaque) (Fig. 3D). The renal nodules observed grossly comprised wellcircumscribed, unencapsulated sheets of similar neoplastic cells. In the stomach, the two plaques observed grossly each comprised a mucosal and submucosal infiltrate of neoplastic cells. Immunohistochemical labelling of the neoplastic cells of the pituitary mass, leptomeninges and tumour foci in the spleen, kidneys and stomach was strongly positive for ACTH (Fig. 4) and negative for FSH and GH. These findings supported a diagnosis of pituitary corticotroph (ACTH-containing) carcinoma, metastatic to cerebral and cerebellar leptomeninges, brain ventricles and spinal canal, spleen, kidneys and stomach, on the basis of a current histological classification system of endocrine tumours in domestic animals (Kiupel, 2008). Most reports of clinical disease attributable to pituitary tumours in dogs are in adult to aged animals (Meij
et al., 1998; Hanson et al., 2005; Capen, 2007; de Fornel et al., 2007). This large pituitary corticotroph carcinoma that caused CNS signs in an 11-month-old dog therefore appears to have developed earlier than previously reported canine pituitary tumours.
Fig. 4. Pituitary mass, dog. Strong ACTH immunoreactivity in the cytoplasm of tumour cells. IHC. 350.
330
S. Gestier et al.
Despite the local invasion and compression of the hypothalamus by the large pituitary mass and the presence of intraventricular and leptomeningeal metastases, the dog appeared clinically normal until it suddenly developed CNS signs. This is consistent with the observation that endocrinologicallyinactive pituitary tumours in dogs often attain considerable size before they cause clinical signs from local mass effect (Capen, 2007). In this case, the acute haemorrhagic changes within the brain associated with the tumour are likely to have contributed to the sudden onset and progression of the neurological signs. The metastases detected in the spleen, kidneys and stomach were too small to be clinically important; the sidero-calcific plaques associated with the capsular metastases in the spleen may reflect past episodes of tumour-related capsular haemorrhage. Pituitary adenomas may be enclosed microadenomas (generally <5 mm diameter in dogs) and expansile or locally-invasive macroadenomas (>5 mm diameter in dogs) (Kiupel, 2008). The most frequently reported pituitary tumours in dogs are endocrinologically-active corticotroph (ACTH-secreting) chromophobe microadenomas (mCAs) or macroadenomas (MCAs) that cause pituitary-dependent hyperadrenocorticism (PDH) (Cushing’s-like disease), with or without subsequent local mass effect, and endocrinologically-inactive chromophobe macroadenomas that cause local mass effect only (Capen, 2007; La Perle and Capen, 2007). The last would include non-functioning (‘silent’) corticotroph (ACTH-containing) adenomas (SCAs), which are defined immunohistochemically (by hormone content), rather than histochemically (by cytoplasmic staining as acidophilic, basophilic or chromophobic) (Asa and Ezzat, 2002; Karavitaki et al., 2007). In a recent study of human mCAs and MCAs associated with Cushing’s disease and SCAs, the SCA was characterized as ‘a macroadenoma revealed by tumour mass-related symptoms with no clinical signs of hypercortisolism, normal plasma cortisol levels and elevated levels of ACTH, and ACTH or proopiomelanocortin (POMC) peptide immunostaining’ of the tumour (Raverot et al., 2010). In people, most clinically-relevant pituitary adenomas associated with Cushing’s disease present as mCAs, while SCAs have variable invasion or compression of adjacent structures (Karavitaki et al., 2007). It is unclear why SCAs fail to produce PDH, but hypotheses include the incomplete or aberrant processing of POMC (the large precursor of ACTH) and secretion of less active ACTH by the tumour (Raverot et al., 2010). Unfortunately, neither blood ACTH nor cortisol was measured in this dog to definitively discount
PDH. However, there were no clinical signs suggestive of hyperadrenocorticism and, importantly, no evidence at necropsy examination of the bilateral adrenal cortical hypertrophy expected with PDH. The morphology of the primary tumour and lack of clinical or pathological evidence of PDH in the face of strongly positive ACTH immunohistochemical labelling of the tumour cells are consistent with an endocrinologically-inactive (‘silent’) pituitary corticotroph (ACTH-containing) carcinoma. The stringent diagnostic criterion for pituitary carcinoma in human and veterinary medicine is the presence of cerebrospinal or systemic metastasis. Local invasion of the hypothalamus or surrounding soft or bony tissues is not necessarily diagnostic for malignancy and can be a feature of adenomas (Asa and Ezzat, 2002; Ragel and Couldwell, 2004; Kiupel, 2008). Some reported cases of pituitary carcinoma in dogs (Sarfaty et al., 1988; Puente, 2003) could be classified as invasive macroadenomas due to their lack of metastasis (Kiupel, 2008). In the present case, the cellular pleomorphism and mitotic activity within the primary tumour and metastases were minimal and similar to those seen in many corticotroph macroadenomas in dogs. Haemorrhage and necrosis can occur within both pituitary macroadenomas and carcinomas (Capen, 2007). There are therefore no reliable histological criteria to distinguish malignant from benign pituitary neoplasms (Kiupel, 2008). Most human pituitary carcinomas (88%) are endocrinologically active, usually secrete ACTH (42%) or PRL (33%), and along with endocrinologicallyinactive pituitary corticotroph carcinomas, present as invasive macrotumours causing mass effect (Ragel and Couldwell, 2004). Most pituitary carcinomas in dogs are endocrinologically-inactive, invasive, chromophobic macrotumours causing mass effect (Capen, 2007). However, there is a report in an 8-year-old cocker spaniel of a large functional pituitary corticotroph carcinoma with invasive growth into the third ventricle and microscopical metastases to brain and leptomeninges, causing PDH without clinical evidence of mass effect (Boujon et al., 1991). The morphological similarity of the primary pituitary tumour in the present case, a ‘silent’ corticotroph (ACTH-containing) tumour, to the endocrinologicallyinactive pituitary chromophobe macroadenoma reported in dogs suggests that cases of the latter tumour should be tested immunohistochemically to identify those that are SCAs.
References Asa SL, Ezzat S (2002) The pathogenesis of pituitary tumors. Nature Reviews Cancer, 2, 836e849.
331
Pituitary Carcinoma in a Dog
Boujon CE, Ritz U, Rossi GL, Bestetti GE (1991) A clinicopathological study of canine Cushing’s disease caused by a pituitary carcinoma. Journal of Comparative Pathology, 105, 353e365. Capen CC (2007) Endocrine glands. In: Pathology of Domestic Animals, 5th Edit., Vol. 3, MG Maxie, Ed., Saunders Elsevier, Edinburgh, pp. 334e351. de Fornel P, Delisle F, Devauchelle P, Rosenberg D (2007) Effects of radiotherapy on pituitary corticotroph macrotumors in dogs: a retrospective study of 12 cases. Canadian Veterinary Journal, 48, 481e486. Hanson JM, van’t Hoofd MM, Voorhout G, Teske E, Kooistra HS et al. (2005) Efficacy of transsphenoidal hypophysectomy in treatment of dogs with pituitarydependent hyperadrenocorticism. Journal of Veterinary Internal Medicine, 19, 687e694. Karavitaki N, Ansorge O, Wass JAH (2007) Silent corticotroph adenomas. Arquivos Brasileiros de Endocrinologia e Metabologia, 51, 1314e1318. Kiupel M (2008)Histological Classification of Tumors of the Endocrine System of Domestic Animals, Vol. 12. Armed Forces Institute of Pathology, Washington DC, pp. 17e25. La Perle KMD, Capen CC (2007) Endocrine system. In: Pathologic Basis of Veterinary Disease, 4th Edit., MD McGavin, JF Zachary, Eds., Elsevier Mosby, St Louis, pp. 693e716. Meij BP, Voorhout G, van den Ingh TSGAM, Hazewinkel HAW, Teske E et al. (1998) Results of transsphenoidal hypophysectomy in 52 dogs with pituitarydependent hyperadrenocorticism. Veterinary Surgery, 27, 246e261.
Pinchot SN, Sippel R, Chen H (2009) ACTH-producing carcinoma of the pituitary with refractory Cushing’s disease and hepatic metastases: a case report and review of the literature. World Journal of Surgical Oncology, 7, 39e45. Puente S (2003) Pituitary carcinoma in an Airedale terrier. Canadian Veterinary Journal, 44, 240e242. Ragel BT, Couldwell WT (2004) Pituitary carcinoma: a review of the literature. Neurosurgical Focus, 16, E7. Raverot G, Wierinckx A, Jouanneau E, Auger C, BorsonChazot F et al. (2010) Clinical, hormonal and molecular characterization of pituitary ACTH adenomas without (silent corticotroph adenomas) and with Cushing’s disease. European Journal of Endocrinology, 163, 35e43. Saeger W, Ludecke DK, Buchfelder M, Fahlbusch R, Quabbe HJ et al. (2007) Pathohistological classification of pituitary tumors: 10 years of experience with the German Pituitary Tumor Registry. European Journal of Endocrinology, 156, 203e216. Sarfaty D, Carrillo JM, Peterson ME (1988) Neurologic, endocrinologic, and pathologic findings associated with large pituitary tumors in dogs: eight cases (1976e1984). Journal of the American Veterinary Medical Association, 193, 854e856.
April 1st, 2011 ½ Received, Accepted, August 7th, 2011