Pituitary DOI 10.1007/s11102-014-0552-2

Metastatic involvement of the pituitary gland: a systematic review with pooled individual patient data analysis Wenzhuan He • Fangxiang Chen • Brian Dalm Patricia A. Kirby • Jeremy D. W. Greenlee

•

Ó Springer Science+Business Media New York 2014

Abstract Purpose To report a rare case of pituitary metastasis (PM) from hepatocellular carcinoma (HCC) and help better understand the incidence of PM and its most common presenting symptoms through a pooled individual patient data analysis. Methods Literature regarding PM was systematically reviewed with a pooled individual patient data analysis conducted. Pooled individual data analysis result is also compared with the result in a most recent systematic review. Results Our results demonstrate that the incidence of PM among all intracranial metastases is 0.87 % (95 % CI 0.56, 1.18); it is 1.9 % (95 % CI 1.46, 2.34) among all autopsied cancer cases; it is 11.56 % (95 % CI 7.08, 16.04) among all breast cancer patients who had hypophysectomies and 12.83 % (95 % CI 10.5, 15.16) among all autopsied breast cancer patients. The fixed effect model showed that the incidence of PM in breast cancer patients group is significantly higher (p \ 0.001) with an odds ratio of 6.71 (95 % CI 4.24, 10.61). Breast and lung cancer are the most

W. He Department of Neurology, UMDNJ, Newark, NJ, USA F. Chen Department of Neurosurgery, St Louis University, St. Louis, MO, USA B. Dalm
J. D. W. Greenlee (&) Department of Neurosurgery, University of Iowa, 1827 JCP, 200 West Hawkins Drive, Iowa City, IA 52242, USA e-mail: [email protected] P. A. Kirby Department of Pathology, University of Iowa, Iowa City, IA, USA

common primary cancer of PM with a percentage of 37.2 and 24.2 respectively. The next most common primary sites are prostate and kidney respectively, although the percentages for each are only about 5. Diabetes insipidus (DI) remains the most common symptom among all reported PM cases with a pooled incidence of 42.34 % (95 % CI 36.15, 48.53). Although not significant (v2 = 2.846, df = 1, p = 0.061), it is less common in the most recent reported cases which has a pooled incidence of 32.76 % (95 % CI 20.31, 45.21). DI is extremely rare in the reported PM cases from HCC (none of the eight cases presented with DI). The symptoms of anterior hypopituitarism (23.68 vs 39.66 %, p = 0.015), visual deterioration (27.89 vs 41.38 %, p = 0.039), cranial nerve palsies (21.58 vs 41.38 %, p = 0.003) and headaches (15.79 vs 32.76 %, p = 0.005) were reported significantly higher than previously described in the literature. Conclusions Pituitary metastasis is rare in patients with cancer, and the pituitary gland is an uncommonly involved location in patients with intracranial metastases. With advanced diagnostic imaging techniques and increased awareness about the manifestation of sellar lesions, the incidence of cranial nerve palsies and anterior pituitarism are higher than reported. This information may allow earlier diagnosis of PM. Keywords Diabetes insipidus (DI)
Hepatocellular carcinoma (HCC)
Hypopituitarism
Intracranial metastases
Pituitary metastasis (PM)

Introduction Metastatic involvement of the pituitary gland is an uncommon disease. In 1857, Benjamin first reported a

123

Pituitary

pituitary metastasis (in German) from a melanoma discovered at autopsy while Harvey Cushing was the first to report a pituitary metastasis (PM) in the English literature in 1913 [1]. Although rare, PM has been increasingly reported during the past decades [2–12]. The reported incidence ranges from 0.14 to 28 percent with higher incidence observed in autopsy series [1, 4–7, 10, 13–22]. PMs are most often asymptomatic and therefore are generally found at autopsy specimens. It is reported that only about 7 % are symptomatic [23]. The most common presenting symptoms in patients with pituitary metastases are diabetes insipidus (DI) and/or oculomotor nerve palsies. The presence of visual loss and anterior pituitary insufficiency, while common presenting symptoms of pituitary adenoma, are less commonly seen with PM [14]. Although many kinds of malignancies can metastasize to the pituitary gland, breast and lung cancers are reported to be the most common [20, 21, 24, 25]. Other reported primary cancers producing PM include but are not limited to gastrointestinal, prostate, melanoma, pharynx, pancreas, thyroid, larynx, renal, liver, and ovarian. Pituitary metastasis from hepatocellular carcinoma (HCC) has been extremely rarely reported in literature, with only 7 cases to date [15, 26–30]. We report an additional case of a patient who had no previous history of cancer and presented with symptoms suggesting pituitary adenoma, yet was found to have a HCC pituitary metastasis. The patient was subsequently discovered to have spinal metastases requiring open decompression and stabilization. In addition to our case report, we have reviewed the published cases of pituitary metastasis and present a systematic review of the literature and additionally present an analysis of pooled individual patient data. Highlighting the rarity of the pituitary gland as a site of HCC metastasis, this case was recently the subject of a case of the month presentation in the journal Brian Pathology [70].

Methods Literature review Pertinent references were searched using Endnote 49 for windows remote search model on PubMed. The key words ‘‘pituitary metastasis’’ searched in all fields resulted in 105 articles, of which 22 were from either animal studies or studies of primary pituitary tumors and were excluded. Searching ‘‘metastatic cancer of the pituitary’’ retrieved 11 additional references. Searching ‘‘metastatic carcinoma of pituitary’’ retrieved 35 additional references. Using the combination of ‘‘metastasis’’ and ‘‘pituitary’’ in the title retrieved 106 references. This added 30 more new references to our database. About 20 new references were

123

identified from the bibliographies of the articles reviewed. Ultimately, we identified a total of 179 references from the search terms. All references were reviewed to retrieve relevant references for this study. Non-English articles were excluded. We found only two references [7, 31] that compared the incidence of PM among different groups of cancer patients. These two references were included for meta-analysis. One previous English literature review was reported in 2004 by Komninos et al. [28]. Articles published after that report were used to estimate the incidence of PM among all kinds of cancers and to calculate the incidence of different presenting symptoms of PM. These results were compared with the results in the article reported by Komninos et al. Statistical analysis Statistical Software SPSS 17.0 for windows was used for analysis. The data was pooled using the method described by Gissane et al. [32]. The incidence rates and the 95 % confidence intervals (CI) were then calculated with SPSS software. Chi square and independent t tests were used for categorical data and numerical data when appropriate. When meta-analysis was conducted, a Chi square test was conducted to test the heterogeneity of the studies. A fixed model was used if heterogeneity was not supported. Odds ratio and difference of percentage were then calculated with the pooled data. Case report A 49-year-old Caucasian male without previous history of malignancy presented with a history of sudden onset of worsening vision on the right side and diplopia for about 2-1/2 weeks. He also complained of headaches and problems with opening and moving his right eye. He denied nausea, vomiting, polyuria and polydipsia. He had no other complaints except chronic low back pain. Neurological examination identified right 3rd, 4th and 6th cranial nerve palsies with complete ptosis and ophthalmoplegia. There were no other neurological deficits noted. Formal visual field testing was unable to be completed prior to intervention. A pituitary hormonal serologic panel showed FSH, ACTH, TSH, free T4, prolactin, and growth hormone levels were all within normal limits. Brain MRI showed a 1.5 cm (AP) 9 3.0 cm (transverse) 9 2.9 cm (craniocaudal) homogenously enhancing mass centered in the sella turcica, with extension laterally into the right cavernous sinus and inferiorly into the superior aspect of the clivus (Fig. 1). The patient was taken to the operating room for endoscopic endonasal transsphenoidal tumor resection of the sellar mass. At surgery, soft tumor was encountered and

Pituitary Fig. 1 Brain MRI showing a metastatic pituitary tumor. Axial (a) and coronal (b) T1 weighted post-gadolinium magnetic resonance images and sagittal (c) T1 weighted pregadolinium MRI showing a homogenous pituitary mass which extends into the right cavernous sinus and inferiorly into the superior aspect of the clivus

tumor vascularity was increased compared to typical pituitary adenomas. Postoperatively, the patient recovered well. Vision and ptosis improved the second day after the surgery. He was discharged home on post-op day 3 without complications. On post-op day 5, the histopathology returned metastatic hepatocellular carcinoma (HCC; Figs. 2, 3). Subsequent systemic evaluation revealed alphafetoprotein (AFP) profoundly elevated (461,040 ng/mL, normal \0.9 ng/mL). AST was 43 l/L (normal \37 l/L) and ALT was 37 l/L (normal \35 l/L). Whole body CT scan was conducted and he was found to have hepatic cirrhosis, multifocal HCC, tumor thrombus extending into the distal left portal vein, bony metastases at multiple levels, most notably at T3 with a pathologic fracture as well as other fractures at T4, T11, T12, and L2. At this time, the patient was noting severe upper thoracic pain, worsened with activity and improved with rest, as well as

Fig. 2 The tumor has a vague glandular pattern with vacuolated cytoplasm. H&E stain, 9400 magnification

123

Pituitary

died 3 months after the transsphenoidal procedure because of his widely metastatic cancer.

Results

Fig. 3 Higher power of tumor demonstrating rare, tiny foci of bile. H&E stain, 91,000 magnification

progressive weakness in both lower extremities such that he lost the ability to ambulate due to 0/5 weakness in right dorsiflexion and debilitating pain. The pathologic fracture at T3 resulted in canal compromise with spinal cord compression. About 3 weeks after his transsphenoidal procedure, he underwent posterior thoracic decompression and instrumented fusion from T1 to T5 with pedicle screw and rod fixation. Following the spinal fusion, his back pain markedly improved, and he regained strength such that he could walk with a cane. He was discharged home with a plan to undergo combined adjuvant chemo- and radiotherapy for his widely metastatic cancer. However, he later declined both chemotherapy and radiation treatment and Table 1 425 reported cases of primary malignancies metastatic to pituitary gland

References from [6, 17, 26, 29, 41, 46, 47, 48–63]

123

Primary site

N

Table 1 shows the calculated incidence of PM among different primary cancers based on 425 reported PM cases. Breast and lung cancers are the most common primaries producing PM with a percentage of 37.2 and 24.2 respectively. The next most common primary sites are prostate and renal, yet the percentages of PM for these are only about 5 in each. These four cancers comprise more than 70 % of all PM cases. Table 2 further illustrates how the incidence of PM is quite different among different primary cancer groups. For example, the pooled incidence of PM is 1.9 % (95 % CI 1.5–2.3) among all autopsied cancer cases; it is 0.9 % (CI 0.6–1.2) among series reporting all intracranial metastases; it is 12.8 % (CI 10.5–15.2) among all autopsied breast cancer patients, and it is 11.6 % (CI 7.1–16.0) among all breast cancer patients who had hypophysectomies. Metaanalysis shows that the heterogeneity is not supported among included studies with a p value of 0.204. Thus the fixed effect model was applied. This analysis revealed that the incidence of PM is significantly higher (p \ 0.001) in autopsied breast cancer patients than that of autopsied all cancer patients. The difference of percentage is 9.3 % (CI 6.2–12.3, p \ 0.001; Fig. 4). The odds ratio of having a PM in autopsied breast cancer patients to odds of having a PM in autopsied unselected cancer patients is 6.7 (CI 4.2–10.6; p \ 0.001).

Percentage

Primary site

N

Percentage

Breast

158

37.2

Multiple myeloma

3

0.7

Lung

103

24.2

Paranasal sinus

3

0.7

Prostate

22

5.2

Oral cavity

3

0.7

Renal

21

4.9

Lymphoma

2

0.5

Melanoma/Skin

13

3.1

Larynx

2

0.5

Thyroid

13

3.1

Germ cell tumor

2

0.5

Colon

12

2.8

Ovary

2

0.5

Unknown

12

2.8

Ileum

1

0.2

Stomach

9

2.1

Retroperitoneum

1

0.2

Pancreas

7

1.6

Bile duct

1

0.2

Liver

6

1.4

Lymphosarcoma

1

0.2

Pharynx

5

1.2

Penis

1

0.2

Endometrium

5

1.2

Thymus

1

0.2

Leukemia

5

1.2

Nasal cavities

1

0.2

Urinary/bladder Uterine/cervix

4 4

0.9 0.9

Salivary glands Merkel cell carcinoma

1 1

0.2 0.2

Pituitary Table 2 Pooled analysis of pituitary gland metastases Author

Journal

Year

# Patients with PM

# Total patients

Percentage

95 % CI

All cancers: autopsy series Kovacs [20]

Oncology

1973

18

1,857

1.0

0.5 to 1.4

Abrams et al. [31]

Cancer

1950

18

1,000

1.8

1.0 to 2.6

Hagerstrand and Schonebeck [7]

Acta Pathol Microbiol Scand

1969

29

763

3.8

2.4 to 5.2

Roessmann et al. [64]

Cancer

1970

5

60

8.3

1.1 to 15.5

70

3,680

1.9

1.5 to 2.3

Pooled result All cancers: intracranial metastases Marsh et al. [57]

Pituitary

2010

1

155

0.6

-0.6 to 1.9

Kano et al. [54]

Surgical neurology

2009

18

2,522

0.7

0.4 to 1.0

Mori et al. [65]

J Neurosurg

2006

8

623

1.3

0.4 to 2.2

Neurochirurgia

1990

3 30

150 3,450

2.0 0.9

-0.3 to 4.3 0.6 to 1.2

Houck et al. [66]

Cancer

1970

2

44

4.5

-1.9 to 11.0 4.6 to 13.4

Chiang et al. [1] Pooled result Breast cancer: autopsy series Abrams et al. [31]

Cancer

1950

15

167

9.0

Grisoli et al. [67]

Neurochirurgie

1974

19

205

9.3

5.3 to 13.3

Hagerstrand and Schonebeck [7]

Acta Pathol Microbiol Scand

1969

46

308

14.9

10.9 to 18.9

Smulders and Smets [22]

Bull. Ass. Franc. Cancer

1960

Pooled result

20

71

28.2

17.5 to 38.9

102

795

12.8

10.5 to 15.2

Breast cancer: cases with hypophysectomy Duchen [68]

J Pathol Bacteriol

1966

12

155

7.7

3.5 to 12.0

Gurling et al. [69]

Br J Cancer

1957

11

44

25.0

11.7 to 38.3

23

199

11.6

7.1 to 16.0

Pooled result

Fig. 4 Meta-analysis to compare the incidence of PM in breast cancer patients and the incidence of PM in all cancer patients. Pooled data demonstrates that the incidence of PM in all autopsied breast

cancer patients is significantly higher (p \ 0.001) than that among all type of autopsied cancer patients with an odds ratio of 6.7 (CI 4.2–10.6)

To date, only seven cases of PM from HCC have been reported (Table 3). Examining the present case that we report in addition to the previous reports, DI is not a common presenting symptom in the HCC PM patient subpopulation (one out of eight patients). Cranial nerve palsies (7/8 pts) and headaches (5/8 pts) are more common in this patient population. There is a strong male preponderance in HCC PM patients (7/8 pts), which is consistent with a previous study [33] of systemic HCC patients, in which 87 % of patients were male with a male to female ratio of 6.7:1. With respect to the presentation of PM from HCC patients compared to the other types of PM, the HCC

PM patients demonstrate significantly lower incidence of anterior pituitary insufficiency (40 vs 0 %, p = 0.027). There are no significant differences in the incidence of DI (33 vs 0 %, p = 0.055), visual disturbance (41 vs 38 %, p = 0.834), cranial nerve palsies (41 vs 63 %, p = 0.259) and headaches (33 vs 63 %, p = 0.101). In examining the presentation of PM patients in recent reports compared to the earlier review [28], the more recent PM patients demonstrate less DI (p = 0.061) than other PM subgroups, but the incidence of visual disturbance, anterior pituitary insufficiency, cranial nerve palsies and headaches are all significantly more common in the most

123

Died 3 Surgery

123

recent reported cases (all p \ 0.05, Fig. 5). Furthermore, there is no significant difference (v2df = 4, = 1.97; p = 0.741) between the five parameters mentioned above in the newly reported PM cases. Table 4 illustrates the incidence of different symptoms in previously reported PM cases, reported by Komninos et al. [28], and subsequently reported 58 PM cases.

Discussion

N/R = Not report; N = No; Y = Yes; RT = Radiation therapy; HA = Headache

Current

2011

50

M

N

0.5

Bone

diplopia, 3rd, 6th. Nerve palsies, HA

Y

N

N

Died 5 Surgery Y J Endocrinol Invest. Moreno-Perez et al. [29]

2007

65

M

N

3

None

ptosis, visual loss, diplopia, 3rd CN palsy, HA, decreased libido

Y

N

Died 12 Surgery N J Endocrinol Invest. Hirsch et al. [26]

2005

46

M

N

0.5

Adrenal

sudden HA, blurred vision, cav sinus syndrome including 3–6 CNs palsies

Y

N

Died 3 Surgery Y J Clin Endocrinol Metab Komninos et al. [28]

2004

68

M

N

1

None

ptosis, diplopia, vision loss

Y

Y

Alive 24 Surgery J Neuroncol Karamouzis et al. [27]

2003

59

M

Y

N/R

Bone

HA, vision disturbance,

N/R

N

N

Died 12 Surgery ? RT Hong Kong Med J Aung et al. [15]

2002

71

M

N

0.5

Bone

diplopia, ptosis, HA, 3rd and 6th CN palsies

Y

Y

N

Died 3 Surgery Hong Kong Med J Aung et al. [15]

2002

40

M

Y

2

Y

diplopia, retro-orbital HA, right 3rd, 6th nerve palsy, HA

Y

Y

N

N/R N/R N/R N/R J Neurosurg Morita et al. [30]

1998

N/R

N/R

N/R

N/R

N/R

N/R

N/R

N/R

Survival (mo) Treatment CA History Sex Age Year Journal Author

Table 3 Reported cases of PM from hepatocellular carcinoma

Symptom duration (mo)

Other mets

Symptoms

Cav sinus invasion

Ele PRL

DI

Outcome

Pituitary

Although rare, PM has been increasingly reported during the past few decades. These reports consist mainly of single case reports, or small case series [15, 23, 30, 34]. There are limited large series published data and most were autopsybased studies on varied patient populations, such as breast cancer patients, unselected cancer patients, or all patients with intracranial metastases. As a result of analyzing these disparate patient groups, the reported incidence of PM ranges between 0.14 and 28.17 % [1, 4, 5, 7, 10, 13–22, 28]. After careful review of these original studies, we subsequently calculated the incidence of PM based on stratified patient populations. To the best of our knowledge, this is the first time the incidence was calculated through a pooled data analysis. Such an analysis is helpful in that it likely produces a more accurate estimation of true incidence of PM, by using larger numbers of patients. The literature and our meta-analysis clearly demonstrate that breast cancer is the most common cancer to involve the pituitary gland. We found that the incidence of PM in breast cancer patients is about 9.3 % higher than the incidence of PM in the unselected cancer population. Therefore, the odds ratio of a breast cancer patient having a PM is about 7 times greater than a patient with a different primary cancer having a PM. This further extends the previous observation that breast and lung cancers are reported to be the most common primary sites of malignancies that metastasize to pituitary gland [20, 21, 24, 25]. Metastases from HCC to the pituitary gland are extraordinarily rare, and the case we present is only the eighth reported. Our patient was very similar to the other reported cases of PM from HCC in that he was male, had cranial nerve palsies due to cavernous sinus invasion, and had headache at presentation. Additionally, once the pathological diagnosis was made from the surgical specimen, he was found to have advanced stage cancer with widely metastatic disease including a destructive spinal lesion that eventually required surgical decompression and stabilization [47]. Although the patient died 3 months after he was diagnosed, he symptomatically responded well to both surgical treatments. Of note, the patient declined systemic adjuvant treatments. This short survival time is consistent with the survival time of systemic HCC patients with a median

Pituitary

Fig. 5 Comparison of the incidence of different symptoms in recently reported PM patients (pooled data) and an early reviewed pooled data

survival time of 5 months. This demonstrates that despite treatment, the survival time is not changed after treatment of the PM, which is related with the primary tumor, and goals of treating these lesions are clearly palliative in nature.

Pituitary metastasis occurred mostly in patients with systemic dissemination. Among all patients with pituitary metastasis, only 2.5–18.2 % of cases will demonstrate symptoms referable to the gland or parasellar structures [28]. Latent pituitary metastases are reported in about 5 % of patients with a known history of malignancy, but as noted this may depend on the primary cancer [29]. Symptomatic pituitary metastasis can be the first indication of malignancy and can closely mimic a benign pituitary adenoma symptomatically as demonstrated in our case. It is reported that pituitary metastasis is the initial presentation of an unknown primary tumor in up to 20–30 % patients [35, 36]. Most investigators support that the clinical presentation may help in differentiating the pituitary metastasis from the pituitary adenoma if the patient develops symptoms of DI. DI due to pituitary adenoma are rare with an incidence of about 1 % and is typically a late finding [37, 38]. When a patient first presents with a pituitary mass and DI, pituitary metastasis should be included in the initial differential diagnosis [30]. Sioutos [37] suggested that the possibility of pituitary metastasis is high when a patient with known metastatic cancer develops DI and has radiographic evidence of a

Table 4 Pooled data showing the incidence of different symptoms of PM Author

Journal

Year

# DI

# Anterior insufficiency

# Visual loss

# CN palsies

# Headaches

Total

Komninos et al. [28]

J Clin Endocrinol Metab

2004

86

45

53

Iwai et al. [34]

Neurol Med Chir

2004

2

2

1

41

30

190

3

0

Hirsch et al. [26]

J Endocrinol Invest

2005

0

0

1

1

7

1

Gopan et al. [6]

Pituitary

2007

2

5

3

1

3

Moreno-Perez et al. [29]

J Endocrinol Invest

2007

0

1

1

5

1

1

Pozzessere et al. [17]

Tumori

2007

0

1

0

Takei et al. [46]

Acta cytologica

2007

0

0

1

0

0

1

1

1

Bhatoe et al. [49]

J Neurooncol

2008

0

1

1

0

1

0

1

Feiz-Erfan et al. [47]

Skull Base

2008

1

1

4

Goglia et al. [52]

Pituitary

2008

1

1

1

4

0

7

1

1

Gur et al. [53]

Pituitary

2008

0

0

0

1

1

1

Lin et al. [56]

Minim Invasive Neurosurg

2008

1

1

1

0

0

0

Williams et al. [63]

Ann Diagn Pathol

2008

1

0

1

1

0

1

1

1

Kano et al. [54]

Surg Neurol

2009

7

7

2

8

Peppa et al. [58]

Clin Breast Cancer

2009

1

1

0

0

1

1

Riemenschneider et al. [60]

Arch Neurol

2009

0

0

0

1

1

1

Santarpia et al. [61]

Head and neck

2009

1

1

1

0

1

1

Feletti et al. [50]

J Neurooncol

2010

0

0

1

0

0

1

Luu et al. [41]

J Clin Neurosci

2010

1

0

4

0

3

4

Marsh et al. [57]

Pituitary

2010

1

1

0

1

1

1

Prodam et al. [59]

Thyroid

2010

0

0

1

1

1

1

Fridley et al. [51]

J Clin Neurosci

2011

0

0

0

0

1

1

Current

0

0

1

1

1

Pooled

105

68

77

65

49

42.34

27.42

31.05

26.21

19.76

(21.83, 33.01)

(25.25, 36.85)

(20.70, 31.72)

(14.77, 24.75)

Percentage 95 % CI

(36.15, 48.53)

18

1 248

123

Pituitary

pituitary mass. Cranial neuropathies are another reported symptom with an incidence as high as 15 % in symptomatic PM but\2 % in pituitary adenoma patients [14]. Due to the high incidence in PM, some authors suggest that the manifestations of DI and/or cranial neuropathies point toward PM rather than pituitary adenoma especially when these symptoms develop in a rapid course and in patients older than 50 year old [39]. Anterior hypopituitarism, visual disturbance, and headaches are less common in PM, and occur in pituitary adenoma patients such that they provide no help in differentiating PM from an adenoma. Most investigators are in agreement with the explanation that the reported PM were more frequent in the posterior lobe and infundibulum leading to the most common symptom of DI [1]. However, some other investigators reported that the anterior lobe was actually involved in a majority (78 %) of the metastatic cases while the posterior lobe was involved in only 44 % of cases [40]. Most authors believe the fact that the anterior lobe is less involved is mainly attributed to the lack of direct arterial blood supply to the anterior lobe as direct hematogenous spread is one of the major routes for pituitary metastases [1, 21, 23, 24, 28, 30]. Another contributing factor is likely that extensive destruction of the anterior lobe is usually required to produce symptomatic anterior pituitary insufficiency. However, with more sensitive radiology techniques and immunochemistry tests, more and more anterior pituitary deficiencies are detected earlier reflecting the increasing incidences of reported symptoms of anterior hypopituitarism [30]. Our current analysis is in support of this aspect with an incidence of about 40 % for anterior pituitary deficiencies, which is significantly different from the previously reported incidence of 23.68 % (p = 0.015) [28]. Furthermore, our study results from HCC suggests that almost all incidences of major symptoms of PM are significantly increased when compared to the result of Komninos et al. [28] except for the incidence of DI (Fig. 3). This is consistent with a previous study result in which 5 pituitary metastases from renal cell carcinoma were reported and found that RCC caused more severe hypopituitarism and visual dysfunction but less DI when compared to those from other primary tumors [6]. This perhaps suggests that HCC is different from other primary tumors. Indeed, in the reported 8 patients with PM from HCC, almost all patients are male except one whose gender was not reported. This may suggest that there is a predilection of male gender over female in HCC metastatic to pituitary gland. Alternatively, this may just represent the overall male predominance observed in HCC. Although it is difficult to differentiate PM from adenoma both symptomatically and radiologically, some neuroimaging characteristics may provide insight. Some investigators have suggested that the presence of bony

123

erosion without sellar enlargement indicates PM more than a pituitary adenoma [41]. Others suggest a metastatic pituitary mass may appear dumbbell-shaped due to indentation by the diaphragma sella and may help differentiate from an adenoma [42]. Studies have shown that MR spectroscopy may provide information on proliferation and hemorrhage of macroadenoma with dimensions[20 mm in at least two directions [43] as well as in PM. This may suggest that the MR spectroscopy positive finding in a pituitary mass less than 20 mm in diameter indicates PM more than a pituitary adenoma. Ryu et al. [44] reported that micro- or macro- pituitary adenoma can manifest as hypermetabolic foci on FDG-PET imaging which makes it difficult when evaluating patterns of brain metastasis. Because it is difficult to differentiate symptomatically and radiographically between PM and adenoma, most PMs were initially suspected during surgery in which the tumor appeared firm, invasive, vascular and hemorrhagic, like the one in our case. Therefore, the diagnosis is ultimately confirmed by histopathology. Occasionally, a PM may occur in a pituitary adenoma such that adenoma and PM tissue may coexist [45]. Takei et. al [46] suggested that cytological imprints can be used reliably to make a diagnosis not only of pituitary adenoma but also of metastatic lesions. With regard to the management of PM, available management modalities include: surgery, radiosurgery, whole brain radiation therapy, chemotherapy, medical therapy such as endocrinological replacement, or some combinations of the above modalities. The prognosis of pituitary metastasis is generally related to the histological subtype and the stage of the primary malignancy rather than to the presence of such metastasis [9]. Literature shows no significant differences in survival time between surgical and nonsurgical treatment groups [30]. Therefore, the management of patients with pituitary metastasis is palliative, and should be directed at minimizing morbidity from symptomatic lesions and maximizing the remaining quality of life.

Conclusions In conclusion, the incidence of pituitary metastasis is about 1 % among all patients with intracranial metastasis. It is about 2 % in all autopsied cancer patients. Breast and lung cancer are the most common primary carcinomas producing pituitary metastasis. Breast cancer patients are more likely to have a pituitary metastasis than unselected cancer patients, with an odds ratio about 6.7. DI remains the most common presenting symptom of PM. However, with modern advances of diagnostic imaging techniques and symptoms of anterior pituitary insufficiency readily

Pituitary

recognized, the incidence of cranial nerve palsies and anterior hypopituitarism are higher than the previously reported values. This information may allow clinicians to more efficiently diagnose pituitary metastasis, which is very important given the poor prognosis that these patients face. Such recognition can direct treatments aimed at maximizing quality of life. Conflict of interest The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

References 1. Chiang MF, Brock M, Patt S (1990) Pituitary metastases. Neurochirurgia (Stuttg) 33:127–131. doi:10.1055/s-2008-1053571 2. Beltrami CA, Grandi E (1970) Pituitary metastases. Riv Patol Clin Sper 11:197–211 3. De Merlier Y, Duprez T, Maiter D, Cosnard G (1996) MR features of pituitary metastases in two patients with central diabetes insipidus. Acta Neurol Belg 96:141–142 4. Delarue J, Chomette G, Pinaudeau Y, Brocheriou C, Auriol M (1964) Pituitary metastases. Frequency. Histopathologic study. Arch Anat Pathol (Paris) 12:179–182 5. Fassett DR, Couldwell WT (2004) Metastases to the pituitary gland. Neurosurg Focus 16:E8 6. Gopan T, Toms SA, Prayson RA, Suh JH, Hamrahian AH, Weil RJ (2007) Symptomatic pituitary metastases from renal cell carcinoma. Pituitary 10:251–259. doi:10.1007/s11102-007-0047-5 7. Hagerstrand I, Schonebeck J (1969) Metastases to the pituitary gland. Acta Pathol Microbiol Scand 75:64–70 8. Legait E, Legait H (1955) Pituitary, pineal and choroid metastases of an ovarian lymphoma in a Rhode Island hen. Bull Assoc Fr Etud Cancer 42:564–573 9. Metivier AC, Mennecier B, Grunenberger F, Schumacher C, Schott R, Jeung MY, Schlienger JL, Pauli G, Quoix E (2006) Pituitary metastases from lung cancer. Five cases. Rev Pneumol Clin 62:231–236 10. Ntyonga-Pono MP, Thomopoulos P, Luton JP (1999) Pituitary metastases. 3 cases. Presse Med 28:1567–1571 11. Rao SR, Rao RS (2001) Pituitary metastases in carcinoma breast. J Postgrad Med 47:135–136 12. Struk DW, Knapp TR, Munk PL, Poon PY (1995) Pituitary and intradural spinal metastases: an unusual initial presentation of lung cancer. Can Assoc Radiol J 46:118–121 13. Takatsugi K, Komuta K, Hosen N, Kitada S, Iida S, Nishihara K, Kimura R, Maeda K, Igarashi T (1998) Metastasis of small cell lung cancer to the parotid gland as the initial clinical manifestation, followed by metastases to the pituitary gland and lumber spinal cord. Nihon Kokyuki Gakkai Zasshi 36:246–250 14. Max MB, Deck MD, Rottenberg DA (1981) Pituitary metastasis: incidence in cancer patients and clinical differentiation from pituitary adenoma. Neurology 31:998–1002 15. Aung TH, Po YC, Wong WK (2002) Hepatocellular carcinoma with metastasis to the skull base, pituitary gland, sphenoid sinus, and cavernous sinus. Hong Kong Med J 8:48–51 16. Kistler M, Pribram HW (1975) Metastatic disease of the sella turcica. Am J Roentgenol Radium Ther Nucl Med 123: 13–21 17. Pozzessere D, Zafarana E, Buccoliero AM, Pratesi C, Fargnoli R, Di Leo A (2007) Gastric cancer metastatic to the pituitary gland: a case report. Tumori 93:217–219

18. Nomizu T, Kanno M, Watanabe T, Watanabe F, Suzuki S, Tsuchiya A, Yamaki Y, Takeda Y, Ohuchi N, Saito T, Abe R (1996) A case of breast cancer metastatic to the pituitary gland. Breast Cancer 3:71–74 19. Cox EV 3rd (1979) Chiasmal compression from metastatic cancer to the pituitary gland. Surg Neurol 11:49–50 20. Kovacs K (1973) Metastatic cancer of the pituitary gland. Oncology 27:533–542 21. McCormick PC, Post KD, Kandji AD, Hays AP (1989) Metastatic carcinoma to the pituitary gland. Br J Neurosurg 3:71–79 22. Smulders J, Smets W (1960) Metastases from carcinoma of the breast. Incidence of pituitary metastases. Bull Assoc Fr Cancer 000:434–456 23. Teears RJ, Silverman EM (1975) Clinicopathologic review of 88 cases of carcinoma metastatic to the pituitary gland. Cancer 36:216–220 24. Branch CL Jr, Laws ER Jr (1987) Metastatic tumors of the sella turcica masquerading as primary pituitary tumors. J Clin Endocrinol Metab 65:469–474 25. Nelson PB, Robinson AG, Martinez AJ (1987) Metastatic tumor of the pituitary gland. Neurosurgery 21:941–944 26. Hirsch D, Benbassat CA, Drozd T, Okon E, Blum I (2005) Pituitary and bilateral adrenal enlargement: an unusual presentation of hepatocellular carcinoma. J Endocrinol Invest 28:454–458 27. Karamouzis MV, Melachrinou M, Fratzoglou M, LabropoulouKaratza C, Kalofonos HP (2003) Hepatocellular carcinoma metastasis in the pituitary gland: case report and review of the literature. J Neurooncol 63:173–177 28. Komninos J, Vlassopoulou V, Protopapa D, Korfias S, Kontogeorgos G, Sakas DE, Thalassinos NC (2004) Tumors metastatic to the pituitary gland: case report and literature review. J Clin Endocrinol Metab 89:574–580 29. Moreno-Perez O, Peiro FM, Lopez P, Boix E, Meoro A, SernaCandel C, Aranda FI, Pico AM (2007) An isolated pituitary metastasis as presentation of a differentiated hepatocellular carcinoma mimicking a nonfunctioning macroadenoma. J Endocrinol Invest 30:428–433 30. Morita A, Meyer FB, Laws ER Jr (1998) Symptomatic pituitary metastases. J Neurosurg 89:69–73. doi:10.3171/jns.1998.89.1. 0069 31. Abrams HL, Spiro R, Goldstein N (1950) Metastases in carcinoma; analysis of 1000 autopsied cases. Cancer 3:74–85 32. Gissane C, Jennings D, Kerr K, White JA (2002) A pooled data analysis of injury incidence in rugby league football. Sports Med 32:211–216 33. Tangkijvanich P, Mahachai V, Suwangool P, Poovorawan Y (2004) Gender difference in clinicopathologic features and survival of patients with hepatocellular carcinoma. World J Gastroenterol 10:1547–1550 34. Iwai Y, Yamanaka K, Honda Y, Matsusaka Y (2004) Radiosurgery for pituitary metastases. Neurol Med Chir (Tokyo) 44:112–116 discussion 117 35. Freda PU, Wardlaw SL, Post KD (1996) Unusual causes of sellar/ parasellar masses in a large transsphenoidal surgical series. J Clin Endocrinol Metab 81:3455–3459 36. Heshmati HM, Khosla S, Robins SP, O’Fallon WM, Melton LJ 3rd, Riggs BL (2002) Role of low levels of endogenous estrogen in regulation of bone resorption in late postmenopausal women. J Bone Miner Res 17:172–178. doi:10.1359/jbmr.2002.17.1.172 37. Sioutos P, Yen V, Arbit E (1996) Pituitary gland metastases. Ann Surg Oncol 3:94–99 38. Hollenhorst R, Younge B (1973) Ocular manifestations produced by adenomas of the pituitary gland: analysis of 1000 cases. In: Kohler PO, Ross GT (eds) Diagnosis and treatment of pituitary tumors. Excerpta Medica, Amsterdam, pp 53–68

123

Pituitary 39. Zager EL, Hedley-Whyte ET (1987) Metastasis within a pituitary adenoma presenting with bilateral abducens palsies: case report and review of the literature. Neurosurgery 21:383–386 40. Oi S, Ciric I, Mayer TK (1978) Metastatic breast carcinoma in the pituitary gland (author’s transl). No To Shinkei 30:69–73 41. Luu ST, Billing K, Crompton JL, Blumbergs P, Lee AW, Chen CS (2010) Clinicopathological correlation in pituitary gland metastasis presenting as anterior visual pathway compression. J Clin Neurosci 17:790–793. doi:10.1016/j.jocn.2009.09.044 42. Freda PU, Wardlaw SL (1999) Clinical review 110: diagnosis and treatment of pituitary tumors. J Clin Endocrinol Metab 84:3859–3866 43. Stadlbauer A, Buchfelder M, Nimsky C, Saeger W, Salomonowitz E, Pinker K, Richter G, Akutsu H, Ganslandt O (2008) Proton magnetic resonance spectroscopy in pituitary macroadenomas: preliminary results. J Neurosurg 109:306–312. doi:10. 3171/JNS/2008/109/8/0306 44. Ryu SI, Tafti BA, Skirboll SL (2010) Pituitary adenomas can appear as hypermetabolic lesions in (18) F-FDG PET imaging. Journal of Neuroimaging 20:393–396. doi:10.1111/j.1552-6569. 2008.00347.x 45. Bret P, Jouvet A, Madarassy G, Guyotat J, Trouillas J (2001) Visceral cancer metastasis to pituitary adenoma: report of two cases. Surg Neurol 55:284–290 46. Takei H, Buckleair L, Goodman JC, Powell SZ (2007) Intraoperative cytologic diagnosis of symptomatic carcinoma (pulmonary small cell carcinoma) metastatic to the pituitary gland: a case report. Acta Cytol 51:637–641 47. Feiz-Erfan I, Rao G, White WL, McCutcheon IE (2008) Efficacy of trans-septal trans-sphenoidal surgery in correcting visual symptoms caused by hematogenous metastases to the sella and pituitary gland. Skull Base 18:77–84. doi:10.1055/s-2007-991105 48. Alacacioglu A, Oztop I, Fidan F, Akkoclu A, Kargi A, Osma E, Ada E, Yilmaz U (2008) Diabetes insipidus caused by pituitary gland metastasis accompanied by iris metastasis of small cell lung cancer: case presentation and review of the literature. Tumori 94:765–768 49. Bhatoe HS, Badwal S, Dutta V, Kannan N (2008) Pituitary metastasis from medullary carcinoma of thyroid: case report and review of literature. J Neurooncol 89:63–67. doi:10.1007/s11060008-9586-5 50. Feletti A, Marton E, Rossi S, Canal F, Longatti P, Billeci D (2010) Pituitary metastasis of Merkel cell carcinoma. J Neurooncol 97:295–299. doi:10.1007/s11060-009-0025-z 51. Fridley J, Adams G, Rao V, Patel A, Humphries W, Goodman C, Gopinath S (2011) Small cell lung cancer metastasis in the pituitary gland presenting with seizures and headache. J Clin Neurosci 18:420–422. doi:10.1016/j.jocn.2010.05.006 52. Goglia U, Ferone D, Sidoti M, Spaziante R, Dadati P, Ravetti JL, Villa G, Bodei L, Paganelli G, Minuto F, Giusti M (2008) Treatment of a pituitary metastasis from a neuroendocrine tumour: case report and literature review. Pituitary 11:93–102. doi:10.1007/s11102-007-0038-6 53. Gur C, Lalazar G, Salmon A, Dubiner V, Gross DJ (2008) Metastatic pancreatic neuroendocrine tumor presenting as a pituitary space occupying lesion: a case report. Pituitary 11:293–297. doi:10.1007/s11102-007-0053-7

123

54. Kano H, Niranjan A, Kondziolka D, Flickinger JC, Lunsford LD (2009) Stereotactic radiosurgery for pituitary metastases. Surg Neurol 72:248–255. doi:10.1016/j.surneu.2008.06.003 discussion 255–246 55. Koshimoto Y, Maeda M, Naiki H, Nakakuki K, Ishii Y (1995) MR of pituitary metastasis in a patient with diabetes insipidus. AJNR Am J Neuroradiol 16:971–974 56. Lin EY, Hsieh CT, Lin CS, Tsai TH, Chiang YH (2008) Keyhole surgery for isolated pituitary stalk metastatic tumors: a case report and review of the literature. Minim Invasive Neurosurg 51:222–224. doi:10.1055/s-2008-1062742 57. Marsh JC, Garg S, Wendt JA, Gielda BT, Turian JV, Herskovic AM (2010) Intracranial metastatic disease rarely involves the pituitary: retrospective analysis of 935 metastases in 155 patients and review of the literature. Pituitary 13:260–265. doi:10.1007/ s11102-010-0229-4 58. Peppa M, Papaxoinis G, Xiros N, Raptis SA, Economopoulos T, Hadjidakis D (2009) Panhypopituitarism due to metastases to the hypothalamus and the pituitary resulting from primary breast cancer: a case report and review of the literature. Clin Breast Cancer 9:E4–E7. doi:10.3816/CBC.2009.n.047 59. Prodam F, Pagano L, Belcastro S, Golisano G, Busti A, Sama M, Caputo M, Bellone S, Voci A, Valente G, Aimaretti G (2010) Pituitary metastases from follicular thyroid carcinoma. Thyroid 20:823–830. doi:10.1089/thy 2009.0256 60. Riemenschneider MJ, Beseoglu K, Hanggi D, Reifenberger G (2009) Prostate adenocarcinoma metastasis in the pituitary gland. Arch Neurol 66:1036–1037. doi:10.1001/archneurol.2009.146 61. Santarpia L, Gagel RF, Sherman SI, Sarlis NJ, Evans DB, Hoff AO (2009) Diabetes insipidus and panhypopituitarism due to intrasellar metastasis from medullary thyroid cancer. Head Neck 31:419–423. doi:10.1002/hed.20911 62. Thomason T, Oxford LE, Ducic Y (2007) Metastatic acinic cell carcinoma presenting as a recurrent pituitary adenoma. J Otolaryngol 36:E91–E92 63. Williams MD, Asa SL, Fuller GN (2008) Medullary thyroid carcinoma metastatic to the pituitary gland: an unusual site of metastasis. Ann Diagn Pathol 12:199–203. doi:10.1016/j. anndiagpath.2006.08.007 64. Roessmann U, Kaufman B, Friede RL (1970) Metastatic lesions in the sella turcica and pituitary gland. Cancer 25:478–480 65. Mori Y, Kobayashi T, Shibamoto Y (2006) Stereotactic radiosurgery for metastatic tumors in the pituitary gland and the cavernous sinus. J Neurosurg 105(Suppl):37–42. doi:10.3171/ sup.2006.105.7.37 66. Houck WA, Olson KB, Horton J (1970) Clinical features of tumor metastasis to the pituitary. Cancer 26:656–659 67. Grisoli F, Leclercq TA, Robert F, Hardy J (1974) Pituitary metastases of breast cancer. Neurochirurgie 20:583–592 68. Duchen LW (1966) Metastatic carcinoma in the pituitary gland and hypothalamus. J Pathol Bacteriol 91:347–355. doi:10.1002/ path.1700910208 69. Gurling KJ, Scott GB, Baron DN (1957) Metastases in pituitary tissue removed at hypophysectomy in women with mammary carcinoma. Br J Cancer 11:519–522 70. Wilson T, Kirby P (2013) Hepatocellular carcinoma metastatic to the sella. Brain Pathol 23(3):365–366