Abstract
Phaeochromocytomas are uncommon tumours of adrenal or extra-adrenal chromaffin tissue. About 2–26% of these have been reported to metastasize, but, on histological criteria, it is virtually impossible to predict malignant behaviour of the tumour. Using immunohistochemistry, we analysed the protein expression of SNAIL, a zinc-finger transcription factor, in a series of 50 phaeochromocytoma specimens from 42 patients. We found that SNAIL-expressing cells are frequent in metastatic primary tumours and their metastases, whereas in tumours without metastases, SNAIL expression is commonly absent. We conclude that the expression of SNAIL may be of use in predicting the metastatic potential of phaeochromocytoma.
Introduction
Phaeochromocytoma is a rare tumour of the adrenal medulla. We lack exact incidence figures in Finland, but in Sweden, the incidence is 2.1 per million (Stenstrom & Svardsudd 1986). It usually arises from adrenergic cells in the adrenal medulla. Extra-adrenal phaeochromocytomas arising from the paraganglion system are called paragangliomas. They represent ∼15–20% of all phaeochromocytomas (Lenders et al. 2005). Paragangliomas seem to be more aggressive and associated with malignant potential than phaeochromocytomas. Additionally, paragangliomas produce excessive amounts of noradrenaline, while phaeochromocytomas may produce just noradrenaline (i.e. von Hippel-Lindau tumours) or a combination of adrenaline and noradrenaline with distinct clinical manifestations.
Most phaeochromocytomas are non-metastatic. However, even local phaeochromocytomas may be symptomatic due to high catecholamine secretion, leading to high blood pressure, pallor, flushing and constipation. From 2.4 to 26% of all, phaeochromocytomas are metastatic ( Melicow 1977, Proye et al. 1992). The 5-year survival of non-metastatic tumours is >95%, but that of metastatic tumours only 44% (Salmenkivi et al. 2004). Owing to the risk of metastases, patients are followed up for a long time by monitoring catecholamine levels and flush symptoms. Positron emission tomography has been used to detect metastases (Timmers et al. 2007, Jager et al. 2008), but this technology is not in use in our centre. Another means of predicting the metastatic behaviour of the tumour would be of clinical value.
As many endocrine tumours, phaeochromocytomas lack reliable histological features that distinguish benign (i.e. non-metastatic) from malignant (i.e. metastatic) tumours (Salmenkivi et al. 2003), with perhaps the exception of extracapsular or intravascular growth and/or necrosis that are usually found in metastatic phaeochromocytomas (Salmenkivi et al. 2004). Classical histopathological features of malignancy like hyperchromasia and pleomorphism are useless, as they are frequently seen in phaeochromocytomas with either metastatic or non-metastatic behaviour (Salmenkivi et al. 2003, 2004). Malignant behaviour can currently only be defined by the appearance of the metastases.
Extra-adrenal phaeochromocytomas are reported to be malignant more often than adrenal phaeochromocytomas, whereas phaeochromocytomas found in the familiar bilateral form of the disease or as a manifestation of multiple endocrine neoplasia are equally malignant as sporadic phaeochromocytomas (John et al. 1999).
The zinc-finger transcription factor SNAIL is often overexpressed in metastatic carcinoma and melanoma (Poser et al. 2001, Blanco et al. 2002). By binding to E-box sequences in the promotor regions of several genes, it downregulates E-cadherin, claudin and occludin expression (Moreno-Bueno et al. 2008). These are key components of adherence/tight junctions of cells. SNAIL also regulates the expression of genes controlling cell motility, proliferation and differentiation (Prindull & Zipori 2004, Klymkowsky & Savagner 2009). SNAIL also promotes epithelial–mesenchymal transition and cell invasion (Oda et al. 1994, Oku et al. 2006). However, it is also a key regulator of neural crest migration, thus regulating the migration of chromaffin cells during the formation of the adrenal glands and paraganglionic system (Le Douarin et al. 1994), which links SNAIL to phaeochromocytomas and paragangliomas.
We have now analysed, by immunohistochemistry, 50 samples from 42 patients with phaeochromocytoma for their expression of SNAIL, a transcription factor essential for the migration of the neural crest. Our data show that high frequency of SNAIL-expressing cells is unambiguously associated with the metastatic phaeochromocytomas. Furthermore, SNAIL is lacking in a majority of non-metastatic phaeochromocytomas. Our data suggest that evaluation of SNAIL expression in phaeochromocytoma can be used to identify the group of patients that need to be followed intensely due to an enhanced risk for metastases.
Materials and methods
Tumour material
We collected a series of phaeochromocytoma surgical samples of both adrenal (n=48) and extra-adrenal (n=2) origin. The 50 surgical samples are from 42 individuals treated at the Department of Surgery, Helsinki University Central Hospital, which is a referral centre for endocrine surgery during 1985–2008. The material, including clinical records, was collected retrospectively during the year 2009. Data were obtained from the hospital records and from the Population Registry of Finland and Statistics Finland. The clinical diagnosis was non-metastatic phaeochromocytoma in 32 patients and metastatic phaeochromocytoma in 10 patients, including two patients with paraganglioma. From four individuals with a metastatic disease, tissue from both the primary tumour and metastases was available. Four patients developed a recurrence of the disease, and samples of the recurrent tumour were included in the material.
Four samples of normal adrenal medullary tissue were obtained from individuals who underwent adrenalectomy for unrelated reasons.
All samples underwent routine histopathological diagnostics during the time of treatment. This was performed by pathologists specialised in endocrine pathology (i.e. the authors K Salmenkivi, J Arola, P Heikkilä). Furthermore, all samples were re-evaluated for this study by K Salmenkivi and grouped according to previously published criteria as non-metastatic either (A) without or (B) with histological suspicious features or (C) malignant (Salmenkivi et al. 2003). Malignancy was defined as histologically or radiologically proven metastatic disease during or after the surgery of the primary tumour. All the malignant phaeochromocytomas presented at least one histologically suspicious feature.
Table 1 describes the demographic features of all patients included in the study: the size of the tumour; patient age at diagnosis; biochemical profile; side of tumour; location of distant metastases; time to metastasis (months) following operation of primary tumour; follow-up time of patients (months), which ones are alive and dead; the causes of death and the SNAIL expression score of primary tumour are tabulated. The mean and median follow-up times are both 99 months.
The demographic profile of the patients included
| Case | Diameter (mm) | Age at diagnosis | Metnef | Normet | Moma | Side | Location of metastases | Time to metastases (months) | Follow-up time (months) | Cause of deatha | SNAIL expression score |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 80 | 81 | 0 | 0 | 0 | R | 27 | Alive | 1 | ||
| 2 | 10 | 68 | 1.8 | 1.6 | n/a | R | 87 | Alive | 0 | ||
| 3 | 45 | 53 | 16.2 | 5.2 | 61.0 | R | 101 | Alive | 0 | ||
| 4b | 10 | 63 | 2.2 | 1.9 | n/a | L | 110 | Alive | 2 | ||
| 5 | 50 | 56 | 13.2 | 6.1 | n/a | R | 110 | Alive | 1 | ||
| 6 | 55 | 58 | 1.0 | 13.6 | 64.0 | R | 108 | Alive | 0 | ||
| 7 | 50 | 54 | 53.3 | 15.2 | n/a | R | 99 | Alive | 0 | ||
| 8 | 120 | 33 | 9 | 273.6 | n/a | R | 92 | Alive | 0 | ||
| 9 | 20 | 63 | 1.6 | 6.9 | n/a | R | 90 | Alive | 2 | ||
| 10 | 30 | 43 | 4.5 | 6 | n/a | L | 89 | Alive | 0 | ||
| 11 | 23 | 63 | 6 | n/a | n/a | R | 92 | Alive | 0 | ||
| 12 | 110 | 54 | 20.1 | 116 | n/a | R | 99 | Alive | 0 | ||
| 13 | n/a | 46 | 0.0 | 0.0 | n/a | n/a | Lymph node | 0 | 87 | Alive | 2 |
| 14 | 60 | 49 | 26.9 | 11.1 | n/a | L | 96 | Alive | 1 | ||
| 15 | 70 | 52 | 0.7 | 1.8 | n/a | L | Spine (bone) | 0 | 30 | t | 2 |
| 16 | 52 | 77 | 0.2 | 15.7 | n/a | R | 35 | u | 0 | ||
| 17 | 60 | 27 | 18.2 | 30.3 | n/a | R | Liver, omentum | 53 | 104 | Alive | 3 |
| 18 | 60 | 82 | 2.8 | 35.4 | n/a | R | 46 | Alive | 0 | ||
| 19 | 40 | 57 | 0 | 0 | 0 | L | 37 | Alive | 0 | ||
| 20 | 20 | 42 | 1.3 | 13.7 | n/a | R | 29 | Alive | 1 | ||
| 21 | n/a | 62 | 2.1 | 138.0 | n/a | L | Abdomen | 0 | 28 | Alive | 2 |
| 22 | 75 | 29 | 1.1 | 54 | n/a | R | 23 | Alive | 1 | ||
| 23 | 70 | 38 | 4 | 150 | n/a | R | 20 | Alive | 3 | ||
| 24 | 50 | 62 | 59.9 | 41.8 | n/a | R | 20 | Alive | 1 | ||
| 25 | n/a | 31 | 1.1 | 17.0 | 322.0 | n/a | Abdomen | 104 | 165 | Alive | 3 |
| 26 | n/a | 69 | 1.0 | 74.7 | 159.0 | R | Abdomen | 0 | 0 | t | 3 |
| 27 | n/a | 64 | 16.4 | 68.6 | 192.0 | L | 137 | u | 1 | ||
| 28 | n/a | 18 | 14.3 | n/a | 69.6 | R | Liver, abdomen | 183 | 284 | Alive | 3 |
| 29b | n/a | 69 | 4.7 | n/a | 34.3 | R | 166 | Alive | 0 | ||
| 30c | n/a | 47 | 0.8 | 32.0 | 59.0 | n/a | Spine (bone) | 0 | 104 | t | 1 |
| 31c | n/a | 39 | 1.0 | 10.6 | n/a | n/a | Lymph node | 0 | 171 | Alive | 2 |
| 32 | 60 | 17 | 0.4 | 34.5 | 55 | n/a | 151 | Alive | 1 | ||
| 33 | 100 | 64 | 1.6 | 80 | 96 | R | 158 | Alive | 0 | ||
| 34 | 25 | 50 | 3.9 | 3.7 | n/a | L | 147 | Alive | 1 | ||
| 35 | 50 | 51 | 11.6 | 24.8 | n/a | R | 147 | Alive | 0 | ||
| 36 | 50 | 51 | 13.5 | 6.3 | 42.0 | L | 135 | Alive | 0 | ||
| 37 | 35 | 45 | 14.0 | 20.0 | 133.0 | R | 145 | Alive | 0 | ||
| 38 | 60 | 49 | 6.6 | 50.0 | n/a | n/a | 135 | Alive | 2 | ||
| 39 | 55 | 56 | 32 | 70 | n/a | R | 52 | u | 0 | ||
| 40 | n/a | 45 | 1.7 | 108.0 | 387.0 | R | Liver, lung, lymph node | 90 | 178 | Alive | 1 |
| 41 | 50 | 82 | 33.2 | 8.9 | 51.0 | L | 73 | u | 1 | ||
| 42 | 43 | 46 | 10.3 | 27.7 | 97.0 | L | 132 | Alive | 0 |
Cause of death (t, tumour related; u, unrelated).
MEN2 syndrome.
Paraganglioma.
Metnef, 24 h urinary metanephrine (μmol); Normet, 24 h urinary normetanephrine (μmol); MOMA, 24 h urinary 3-methoxy-4-hydromandelic acid (μmol).
Immunohistochemistry
Sections of formalin-fixed, paraffin embedded tumour biopsies were stained with SNAIL antibody. Epitope retrieval was performed by pretreatment in a microwave oven for 7 min in a 10 mM sodium citrate buffer (pH 6.0). The rabbit polyclonal SNAIL antibody (1:600 dilution, catalogue # ab17732, AbCam plc. Cambridge, UK) was diluted in Powervision antibody blocking solution (Immunovision Technologies & Co., Burlingame, CA, USA) and incubated for 60 min at 21 °C. Binding of the primary antibody was detected with a Powervision+Poly-HRP histostaining kit (Immunovision Technologies & Co.). Tissue sections were counterstained with haematoxylin.
The immunostainings were scored blinded to the clinical diagnosis; first independently by two of the authors (H Sariola and V Häyry) and then jointly (H Sariola and V Häyry). The consensus score was used for the final analyses. SNAIL expression was scored according to the percentage of immunoreactive tumour cell nuclei: 0, no staining; 1, low (<20% of positive nuclei); 2, intermediate (20–70% of positive nuclei); and 3, high (>70% of positive nuclei). This method of scoring was selected after analysis of SNAIL in a test series of nine tumours, before the whole material was analysed and before subsequent statistical analyses.
Tissue sections of normal adult adrenal glands were used as negative controls and biopsies of breast carcinoma as a positive control for SNAIL expression in each staining set. Additional controls included were omittance of primary antibody and inclusion of normal rabbit IgG (Zymed Laboratories Inc. San Francisco, CA, USA) as mock.
Microscopy
For photography, a Nikon Eclipse 80i microscope with a Nikon Plan Fluor 60× objective, Nikon DS-Fi1 digital camera and NIS-Elements F imaging software version 2.3 were used (Nikon Corporation, Tokyo, Japan).
Statistical analysis
For categorical, non-ordered variables, cross tabulations were analysed using the χ2 test; when the χ2 test could not be used, the Fisher's exact test was chosen. All P values are two sided. SPSS version 12.0.1 software (SPSS, Chicago, IL, USA) was used for all statistical analyses. Sensitivity and specificity values were calculated as previously described (Altman & Bland 1994).
Ethical considerations
The project has been approved by the ethical committee of Helsinki University Central Hospital, and all work has been done in accordance with the Helsinki declaration.
Results
SNAIL expression is high in metastatic phaeochromocytomas
In the normal adrenal medulla, no SNAIL expression was detected (Fig. 1). Likewise, in 18 out of 32 specimens representing 32 cases of non-metastatic phaeochromocytomas, no SNAIL expression was seen. In 13 out of 32 specimens of non-metastatic tumours, SNAIL-positive tumour cells were observed at low or moderate frequency and only in one specimens the expression was high (Table 2). This patient was treated during 2007, and thus long-term clinical follow-up data are not yet available for this individual.
SNAIL expression in phaeochromocytoma and adrenal medulla. (A) Normal adrenal medulla with no immunoreactive cells. (B) Non-metastatic phaeochromocytoma without SNAIL expression. (C) Phaeochromocytoma that contains necrotic areas. No SNAIL is found. (D) Phaeochromocytoma with capsular invasion, but no clinically verified metastases. SNAIL-immunoreactive cells are frequent. (E) A metastatic tumour shows high frequency of SNAIL-positive cells. (F) A lymph node metastasis from the same patient as (E) displays an identical staining pattern as the primary tumour. All images are at 600× magnification.
Citation: Endocrine-Related Cancer 16, 4; 10.1677/ERC-09-0049
SNAIL protein expression in pheochromocytoma according to diagnosis
| SNAIL expression: number of samples | ||||||
|---|---|---|---|---|---|---|
| Clinical and histopathological diagnosis | No expression | Low | Moderate | High | Total | Number of patients |
| Non-metastatic | 18 | 10 | 3 | 1 | 32 | 32 |
| Metastatic, primary locationa | 0 | 2 | 5 | 7 | 14 | 10 |
| Metastatic, metastases | 0 | 0 | 1 | 3 | 4 | n/a |
| Total | 50 | 42 | ||||
Two of these were paragangliomas and the remaining 12 specimens represent eight patients with four recurrent diseases.
Out of 14 tumour specimens representing 10 cases with metastases, 12 showed high or intermediate SNAIL expression and only 2 out of 14 showed low expression. None of the metastatic tumours were SNAIL negative (Table 2).
In all samples from which tissue from both the primary tumour and its metastases was available, intermediate or high SNAIL expression was seen in both the primary tumour and the metastasis (Fig. 1, Table 2).
The difference in SNAIL expression in specimens of non-metastatic versus metastatic tumours was statistically highly significant (Fisher's exact test P<0.001). Also when considering the patients rather than specimens, another statistical analysis was performed. Because in some patients more than one sample was available (i.e. recurrence, metastasis), so we used only the primary tumour specimen as basis of calculations. Also the two cases with paragangliomas were excluded, because of their dissimilar clinical behaviour compared with adrenal tumours. SNAIL expression correlated with metastatic behaviour (Fisher's exact test P<0.001).
SNAIL protein expression level was also compared with the clinical parameters presented in Table 1. No statistically significant correlations were observed.
Finally, the sensitivities and specificities of the diagnostic value of SNAIL expression were calculated to distinguish between malignant (metastatic) and benign (non-metastatic) tumour behaviour in all 40 patients in the follow-up. Here, we again excluded recurrent tumours and paragangliomas in the calculations. By setting the cut-off between non-immunoreactivity and any level of immunoreactivity, SNAIL expression had a 100% sensitivity of detecting the malignant tumours. However, by this method, the specificity was only 56%. If the cut-off level was set at 70% or more immunoreactive tumour cell nuclei, the sensitivity was 50%, but the specificity was 97% to detect metastatic disease.
For the group of tumours (B) with suspicious histological features (local invasive growth through the capsule or into blood vessels and/or necrosis), the frequency of SNAIL-positive samples was higher (50%) than in the group of non-metastatic tumours without aggressive histology (A) (40%), but less than in metastatic tumours (C) (100%). SNAIL protein expression correlated significantly with invasive growth seen in the histopathological material (Fisher's exact test P=0.006).
Discussion
Several histochemical and molecular markers (MIB-1, Cd34, VEGF, hTERT; p53, bcl-2 and c-erbB-2) have been investigated to identify the metastatic subgroup of phaeochromocytoma (de Krijger et al. 1999, van der Harst et al. 2000, Favier et al. 2002, Boltze et al. 2003), but the results have been inconclusive.
Ohta et al. (2005) has investigated, with real-time quantitative PCR, the RNA expression of 11 genes in phaeochromocytoma. Six of these were down-regulated in malignant versus benign tumours. One of these was E-cadherin, a downstream target of SNAIL. We could not confirm this on protein level. In our material, only 12 out of 50 samples expressed the E-cadherin protein, and even these only marginally (data not shown). Compared with other genes studied by Ohta et al. the relative E-cadherin RNA expression was two log-orders lower, which may relate to our own negative finding on protein level.
SNAIL, a zinc-finger transcription factor, was first discovered in Drosophila, where it regulates the dorsoventral patterning. In flies without Snail, the ventral organs fail to develop (Boulay et al. 1987, Kasai et al. 1992). The adrenal medulla emerges from neural crest. When the neural crest delaminates from neural tube, its cells start to migrate and proliferate (Le Douarin et al. 1994, Tucker 2004). The transcription factor SNAIL is one of the proteins that altogether make the specific molecular signature for migrating crest cells. The expression of SNAIL is first seen at the beginning of neural crest migration, it is maintained during migration, continues throughout the process of adrenal gland formation and is turned off in the mature adrenal medulla (Le Douarin et al. 1994, Tucker 2004).
The transcription factor SNAIL is also involved in epithelial–mesenchymal transition in tumour cells, suppressing, for instance, E-cadherin expression. (Batlle et al. 2000, Cano et al. 2000). High SNAIL expression incurs a mesenchymal, fibroblastic phenotype to tumour cells, in which epithelial cell adhesions are suppressed (De Craene et al. 2005). SNAIL expression is elevated in several malignancies of epithelial origin, e.g. oral squamous cell carcinoma and breast carcinoma (Blanco et al. 2002, Usami et al. 2008).
We now show that a high frequency of SNAIL-expressing tumour cells in primary phaeochromocytomas suggests metastatic potential. SNAIL expression detects with high sensitivity metastatic disease. Likewise, lack of SNAIL expression is a characteristic of non-metastatic tumours. High and diffuse SNAIL expression in phaeochromocytomas with metastases (frequency >70% positive nuclei) suggests that SNAIL expression can be used for identification of patients at high risk of metastases requiring intense clinical follow-up. Thus, the data indicate that the majority of patients with SNAIL-negative phaeochromocytoma may be followed up less intensely than those tumours with SNAIL expression.
SNAIL expression thus detects or confirms malignancy in metastatic lesions, but, in primary tumours, moderate or high SNAIL expression may be considered predictive for metastasis. It is also noteworthy that during follow-up (median 99 months), none of the SNAIL-negative cases developed metastases or recurrence.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
V Häyry has received grant support from Finska läkaresällskapet and Nona ja Kullervo Väreen säätiö, both non-profit organisations supporting medical research in Finland.
Acknowledgements
We thank Ms Lea Armassalo for her expert technical assistance, Dr Olli Tynninen for his help with photography and registered nurse Elina Aspiala for her help with collecting patient data.
References
Altman DG & Bland JM 1994 Diagnostic tests. 1: sensitivity and specificity. BMJ 308 1552.
Batlle E, Sancho E, Franci C, Dominguez D, Monfar M, Baulida J & Garcia De Herreros A 2000 The transcription factor snail is a repressor of E-cadherin gene expression in epithelial tumour cells. Nature Cell Biology 2 84–89.
Blanco MJ, Moreno-Bueno G, Sarrio D, Locascio A, Cano A, Palacios J & Nieto MA 2002 Correlation of snail expression with histological grade and lymph node status in breast carcinomas. Oncogene 21 3241–3246.
Boltze C, Mundschenk J, Unger N, Schneider-Stock R, Peters B, Mawrin C, Hoang-Vu C, Roessner A & Lehnert H 2003 Expression profile of the telomeric complex discriminates between benign and malignant pheochromocytoma. Journal of Clinical Endocrinology and Metabolism 88 4280–4286.
Boulay JL, Dennefeld C & Alberga A 1987 The drosophila developmental gene snail encodes a protein with nucleic acid binding fingers. Nature 330 395–398.
Cano A, Perez-Moreno MA, Rodrigo I, Locascio A, Blanco MJ, del Barrio MG, Portillo F & Nieto MA 2000 The transcription factor snail controls epithelial–mesenchymal transitions by repressing E-cadherin expression. Nature Cell Biology 2 76–83.
De Craene B, Gilbert B, Stove C, Bruyneel E, van Roy F & Berx G 2005 The transcription factor snail induces tumor cell invasion through modulation of the epithelial cell differentiation program. Cancer Research 65 6237–6244.
Favier J, Plouin PF, Corvol P & Gasc JM 2002 Angiogenesis and vascular architecture in pheochromocytomas: distinctive traits in malignant tumors. American Journal of Pathology 161 1235–1246.
van der Harst E, Bruining HA, Jaap Bonjer H, van der Ham F, Dinjens WN, Lamberts SW, de Herder WW, Koper JW, Stijnen T & Proye C et al. 2000 Proliferative index in phaeochromocytomas: does it predict the occurrence of metastases? Journal of Pathology 191 175–180.
Jager PL, Chirakal R, Marriott CJ, Brouwers AH, Koopmans KP & Gulenchyn KY 2008 6-L-18F-Fluorodihydroxyphenylalanine PET in neuroendocrine tumours: basic aspects and emerging clinical applications. Journal of Nuclear Medicine 49 573–586.
John H, Ziegler WH, Hauri D & Jaeger P 1999 Pheochromocytomas: can malignant potential be predicted? Urology 53 679–683.
Kasai Y, Nambu JR, Lieberman PM & Crews ST 1992 Dorsal–ventral patterning in drosophila: DNA binding of snail protein to the single-minded gene. PNAS 89 3414–3418.
Klymkowsky MW & Savagner P 2009 Epithelial–mesenchymal transition a cancer researcher's conceptual friend and foe. American Journal of Pathology 174 1588–1593.
de Krijger RR, van der Harst E, van der Ham F, Stijnen T, Dinjens WN, Koper JW, Bruining HA, Lamberts SW & Bosman FT 1999 Prognostic value of p53, bcl-2, and c-erbB-2 protein expression in phaeochromocytomas. Journal of Pathology 188 51–55.
Le Douarin NM, Dupin E & Ziller C 1994 Genetic and epigenetic control in neural crest development. Current Opinion in Genetics and Development 4 685–695.
Lenders JMW, Eisenhofer G, Mannelli M & Pacak K 2005 Phaeochromocytoma. Lancet 366 665–675.
Melicow MM 1977 One hundred cases of pheochromocytoma (107 tumors) at the columbia-presbyterian medical center, 1926–1976: a clinicopathological analysis. Cancer 40 1987–2004.
Moreno-Bueno G, Portillo F & Cano A 2008 Transcriptional regulation of cell polarity in EMT and cancer. Oncogene 27 6958–6969.
Oda T, Kanai Y, Oyama T, Yoshiura K, Shimoyama Y, Birchmeier W, Sugimura T & Hirohashi S 1994 E-cadherin gene mutations in human gastric carcinoma cell lines. PNAS 91 1858–1862.
Ohta S, Lai EW, Pang ALY, Brouwers FM, Chan WY, Eisenhofer G, de Krijger R, Ksinatova L, Breza J & Blazicek P et al. 2005 Downregulation of metastasis suppressor genes in malignant pheochromocytoma. International Journal of Cancer 114 139–143.
Oku N, Sasabe E, Ueta E, Yamamoto T & Osaki T 2006 Tight junction protein claudin-1 enhances the invasive activity of oral squamous cell carcinoma cells by promoting cleavage of laminin-5 gamma2 chain via matrix metalloproteinase (MMP)-2 and membrane-type MMP-1. Cancer Research 66 5251–5257.
Poser I, Dominguez D, de Herreros AG, Varnai A, Buettner R & Bosserhoff AK 2001 Loss of E-cadherin expression in melanoma cells involves up-regulation of the transcriptional repressor snail. Journal of Biological Chemistry 276 24661–24666.
Prindull G & Zipori D 2004 Environmental guidance of normal and tumor cell plasticity: epithelial–mesenchymal transitions as a paradigm. Blood 103 2892–2899.
Proye C, Vix M, Goropoulos A, Kerlo P & Lecomte-Houcke M 1992 High incidence of malignant pheochromocytoma in a surgical unit, 26 cases out of 100 patients operated from 1971 to 1991. Journal of Endocrinological Investigation 15 651–663.
Salmenkivi K, Heikkila P, Haglund C, Louhimo J & Arola J 2003 Lack of histologically suspicious features, proliferative activity, and p53 expression suggests benign diagnosis in phaeochromocytomas. Histopathology 43 62–71.
Salmenkivi K, Heikkila P, Haglund C & Arola J 2004 Malignancy in pheochromocytomas. APMIS 112 551–559.
Stenstrom G & Svardsudd K 1986 Pheochromocytoma in Sweden 1958–1981. An analysis of the national cancer registry data. Acta Medica Scandinavica 220 225–232.
Timmers HJLM, Kozupa A, Chen CC, Carrasquillo JA, Ling A, Eisenhofer G, Adams KT, Solis D, Lenders JWM & Pacak K 2007 Superiority of fluorodeoxyglucose positron emission tomography to other functional imaging techniques in the evaluation of metastatic SDHB-associated pheochromocytoma and paraganglioma. Journal of Clinical Oncology 25 2262–2269.
Tucker RP 2004 Neural crest cells: a model for invasive behavior. International Journal of Biochemistry and Cell Biology 36 173–177.
Usami Y, Satake S, Nakayama F, Matsumoto M, Ohnuma K, Komori T, Semba S, Ito A & Yokozaki H 2008 Snail-associated epithelial–mesenchymal transition promotes oesophageal squamous cell carcinoma motility and progression. Journal of Pathology 215 330–339.
