Genetic Testing in Pheochromocytoma or Functional Paraganglioma
http://www.100md.com
《临床肿瘤学》
the Université Paris-Descartes, Faculté de Médecine
Département de Génétique and Unité d’Hypertension Artérielle, H?pital Européen Georges Pompidou, and Service des Maladies Endocrininiennes et Mètaboliques, H?pital Cochin, Assistance Publique-H?pitaux de Paris
L’Institut National de la Santé et de la Recherche Médicale (INSERM) U36, Collège de France
INSERM U567
Centre National de la Recherche Scientifique Unitè Mixte de Recherche 8104, Institut Cochin
Service de Cancérologie Endocrinienne and Service de Génétique, Institut Gustave Roussy, Villejuif, France
Service de Mèdecine Interne et Endocrinologie H?pital Henri Mondor, Créteil, France
Unité d’Endocrinologie, Centre Hospitalier Universitaire (CHU) and INSERM Equipe Mixte 0105 Grenoble, France
Service de Medecine Interne et Hypertension Artérielle, CHU Rangueil
INSERM U558, 31059 Toulouse, France
Service d’Endocrinologie, CHU de Reims, Reims
Laboratoire de Génétique, H?pital Edouard Herriot, Lyon
Service d’Endocrinologie, H?pital de l’H?tel-Dieu, CHU de Nantes, Nantes
Service d’Endocrinologie, CHU de Marseille, Marseille
Génétique Oncologique Ecole Pratique des Hautes Etudes-Unitè Mixte de Recherche 8125, Faculté de Médecine Paris-Sud
Service d’Urologie, H?pital de Bicêtre, Le Kremlin-Bicêtre
Service d’Endocrinologie, CHU d’Angers, Angers
Service d’Endocrinologie, H?pital de l’Archet, CHU de Nice, Nice, France
ABSTRACT
PURPOSE: To assess the yield and the clinical value of systematic screening of susceptibility genes for patients with pheochromocytoma (pheo) or functional paraganglioma (pgl).
PATIENTS AND METHODS: We studied 314 patients with a pheo or a functional pgl, including 56 patients having a family history and/or a syndromic presentation and 258 patients having an apparently sporadic presentation. Clinical data and blood samples were collected, and all five major pheo-pgl susceptibility genes (RET, VHL, SDHB, SDHD, and SDHC) were screened. Neurofibromatosis type 1 was diagnosed from phenotypic criteria.
RESULTS: We have identified 86 patients (27.4%) with a hereditary tumor. Among the 56 patients with a family/syndromic presentation, 13 have had neurofibromatosis type 1, and germline mutations on the VHL, RET, SDHD, and SDHB genes were present in 16, 15, nine, and three patients, respectively. Among the 258 patients with an apparently sporadic presentation, 30 (11.6%) had a germline mutation (18 patients on SDHB, nine patients on VHL, two patients on SDHD, and one patient on RET). Mutation carriers were younger and more frequently had bilateral or extra-adrenal tumors. In patients with an SDHB mutation, the tumors were larger, more frequently extra-adrenal, and malignant.
CONCLUSION: Genetic testing oriented by family/sporadic presentation should be proposed to all patients with pheo or functional pgl. We suggest an algorithm that would allow the confirmation of suspected inherited disease as well as the diagnosis of unexpected inherited disease.
INTRODUCTION
Pheochromocytomas (pheo) are catecholamine-secreting adrenal medulla tumors, and functional paragangliomas (pgl) are catecholamine-secreting extra-adrenal tumors most often located in the sympathetic ganglias. In recent years, significant progress has been made in understanding the genetic determinism of these tumors. The following four inherited disorders have so far been identified: multiple endocrine neoplasia type 2, von Hippel-Lindau disease, neurofibromatosis type 1, and hereditary pheo-pgl; these syndromes are caused by mutations in the RET, VHL, NF1, and SDH (SDHD, SDHB, and SDHC) genes, respectively.1 It has been recently shown that a germline mutation may be present in 12% to 24% of apparently sporadic pheo. The Freiburg-Warsaw-Columbus Pheochromocytoma Group genotyped 271 patients with a nonsyndromic pheo and identified mutations in 66 patients (24%); 30 patients had mutations on VHL, 13 had mutations on RET, 11 had mutations on SDHD, and 12 had mutations on SDHB.2 Our group genotyped an initial series of 84 patients with an apparently sporadic pheo. These patients had been observed for an average duration of 8.8 ± 5.7 years, and a germline mutation was identified in 12% of the patients principally on the SDHB and VHL genes. A germline mutation on SDHB was associated with a high risk of recurrence or malignancy, showing the importance of SDHB genetic testing.3 Here, we have extended our recruitment and analyzed the SDHB, VHL, SDHD, RET, and SDHC genes in a series of 314 patients recruited nationally.
PATIENTS AND METHODS
In this study, 314 patients (140 men and 174 women) with a pheo or a functional pgl were recruited from several different French clinical centers participating in several clinical research networks (Cortico et Medullosurrènale: les Tumeurs Endocrines, ParaganGlioma Network, Groupe des Tumeurs Endocrines). The whole series contains the 84 patients from the COMETE network who were previously reported.3 The diagnosis of pheo or functional pgl was based on the presence of an adrenal or an extra-adrenal thoracoabdominal tumor associated with a high level of total urinary metanephrines. For every patient, the diagnosis has been confirmed by histologic examination after surgery except for three patients having an extensive malignant metastatic dissemination who could not be operated on. Malignancy was defined by the presence of extra-paraganglionic metastases, including lung, liver, bone, and lymph nodes. For each patient, a family history and clinical features included in hereditary pheo syndromes were investigated, as previously described.3 The diagnosis of neurofibromatosis type 1 was based on phenotypic criteria.4 As described elsewhere3 and in accordance with the national ethical rules, a written informed consent was obtained from every patient before drawing blood peripheral sample and DNA extraction. For each patient, we directly sequenced all the coding exons of SDHB, SDHD, SDHC, and VHL and exons 10, 11, 13, 14, 15, and 16 of RET, as previously described.5-7
The statistical analysis was performed with Staview version 5.0 (SAS Institute Inc, Cary, NC) statistical software. We used the Fisher’s exact test for small samples, the standard 2 test for larger groups, and the analysis of variance test to compare more than two variables. P < .05 was considered statistically significant.
RESULTS
Clinical Characteristics
In this series of 314 patients with pheo or functional pgl, 256 patients (81.5%) had an adrenal tumor (pheo) exclusively (41 of these patients had a bilateral pheo), 50 patients (15.9%) had a functional extra-adrenal pgl exclusively, and eight patients (2.5%) had both pheo and functional pgl. Among the 314 patients, 52 (16.6%) had a malignant form. We found a positive family history and/or a syndromic presentation in 56 patients (17.8%); 13 of these patients had a family history of pheo, nine patients had a phenotype of von Hippel-Lindau disease, 14 patients had multiple endocrine neoplasia type 2, and seven patients had hereditary pgl. Thirteen patients were diagnosed as having neurofibromatosis type 1 from clinical criteria. The other 258 patients (82.2%) had an apparently sporadic presentation. Patients with a family history and/or a syndromic presentation were younger at diagnosis than sporadic presentation patients (30.2 ± 13.1 v 43.6 ± 15.6 years of age, respectively; P < .0001) and more often had bilateral tumors (55.4% v 3.9%, respectively; P < .0001; Table 1).
Genetic Data According to the Presentation
We identified a heterozygous germline mutation in 73 patients and diagnosed neurofibromatosis type 1 in 13 patients. Thus, 86 (27.4%) of the 314 patients had a hereditary pheo or functional pgl. Among the 86 patients with a hereditary pheo or functional pgl, 56 (65.1%) had a syndromic presentation and/or a positive family history, whereas 30 (34.9%) had apparently sporadic presentation. Among the 258 sporadic presentation patients, 30 (11.6%) had a germline mutation. Most mutations (27 of 30 mutations, 90%) were identified on the SDHB (n = 18, 60%) or VHL (n = 9, 30%) genes. We did not find any mutation on the SDHC gene. Despite the fact that all patients were systematically screened for the entire set of genes, we found only one causative mutation for each patient. The type and the number of identified germline mutations according to the presentation are listed in Tables 2 and 3, respectively.
Genetic Data According to the Clinical Characteristics
Patients with a mutation were younger at time of diagnosis than patients without a mutation (P < .0001; Table 4). The younger patients were affected by the von Hippel-Lindau disease. Extra-adrenal tumors were present in 58 patients. Among them, 28 patients (48.3%) had a mutation located on the SDHB (17 of 28 patients, 60.7%), SDHD (seven of 28 patients, 25.0%), or VHL (four of 28 patients, 14.3%) genes. The tumors were more often bilateral in patients with a mutation than in patients without a mutation (36 of 86 patients, 41.9% v five of 228 patients, 2.2%; P < .0001). Most patients with bilateral tumors harbored a mutation (36 of 41 patients, 87.8%). These mutations were mainly on the VHL (17 of 36 patients, 47.2%) and RET (11 of 36 patients, 30.5%) genes. In the subgroup of 52 patients with a malignant tumor, 18 patients had a mutation, with 15 of the mutations (83.3%) on the SDHB gene. These data suggest that earlier onset, bilateral, extra-adrenal, and malignant tumors indicate the presence of a genetic disease.
NF1 Patients
All patients with neurofibromatosis type 1 had classic cutaneous lesions (café au lait pigmented spots) and neurofibromas. The average age of the 13 patients when diagnosed with pheo (40.1 years) was higher than the average age observed in the other patients with a hereditary disease. Every patient had an adrenal medullary tumor, with five patients having bilateral disease and one having a malignant pheo (Table 4).
RET Mutation Carriers
Most mutations were located in codon 634 (Table 2). The only patient with an apparently sporadic presentation harbored a mutation in exon 13 (Lys790Phe). Among the 16 patients with a RET mutation, 15 (93.8%) had a personal or family history of medullary thyroid cancer (Table 3). Among these patients, the tumors were always adrenal and often bilateral (11 of 15 patients, 73%; Table 4).
SDHD Mutation Carriers
We found nine different mutations that were nonsense or frameshift mutations (Table 2). An SDHD mutation was identified in 11 patients, among whom nine (81.8%) had a family history on the father’s side of the family (six of nine patients, 66.7%) and/or an associated head and neck pgl (seven of nine patients, 77.7%; Table 3). The majority of these patients (seven of 11 patients, 63.6%) had extra-adrenal tumors (Table 4).
VHL Mutation Carriers
Twelve different mutations were identified in exons 1 and 3 of the VHL gene (Table 2). They were missense, frameshift, and splicing mutations. We observed two different mutations in codon 98 of exon 1 (Tyr98His and Tyr98Cys) and frequent mutated codons and recurrent mutations in exon 3 (Tyr156Cys, Arg161Gln, Arg167Trp, and Arg167Gln). The 25 patients with a VHL mutation included 16 patients (64%) with family history and/or a syndromic presentation and nine patients (36%) with sporadic presentation, representing 30% (nine of 30 patients) of the patients with an apparently sporadic presentation who had mutations (Table 3). At diagnosis, a positive family history was found in 10 patients, and other syndromic lesions (renal tumors and hemangioblastomas) were found in nine patients. The tumors were bilateral in 17 patients (70.8%; Table 4). All nine sporadic presentation patients had a mutation located in exon 3 of the VHL gene. In the Tyr156Cys mutation carriers, we did not find any other syndromic von Hippel-Lindau disease lesions, except pheo or pgl.
SDHB Mutation Carriers
We found 15 different mutations with recurrent mutated codons at positions 46, 207, and 230 (Table 2). Most SDHB-positive patients belonged to the sporadic presentation group (18 of 21 patients, 85.7%), and this group was the largest group within the sporadic presentation patients who had mutations (18 of 30 patients, 62.1%; Table 3). Tumors in the SDHB mutation carriers were frequently malignant (15 of 21 patients, 71.4%). Among the 52 patients with a malignant tumor, 18 (34.6%) harbored a mutation, which was primarily located on the SDHB gene (15 of 18 patients, 83.3%). Twenty-one patients died from their malignant disease (10 patients had a mutation and 11 had no mutations), among whom nine had an SDHB mutation. In the whole series, there was a strong association between malignancy and SDHB mutation (15 [83.3%] of 18 patients were SDHB mutation–positive carriers and 37 [12.5%] of 296 patients were SDHB mutation–negative carriers; P < .0001). We found that the SDHB-mutated tumors were significantly larger than the non–SDHB-mutated tumors (P < .0006) and were more often extra-adrenal (17 of 21 patients, 81%; P < .0001; Table 4).
DISCUSSION
Until 2002, 10% of the total amount of pheo were considered to be hereditary.8 However, a large cohort of 271 patients, in which 24% of patients had sporadic pheo and no family history, challenged that notion. That study suggested that routine analysis for mutations on RET, VHL, SDHD, and SDHB genes in patients with apparently sporadic presentation of pheo should be considered as the clinical standard of care.2,9 In the present series, our results are similar to those obtained by the Freiburg-Warsaw-Columbus Pheochromocytoma Group, with 27.4% of germline mutations being identified in 314 patients or 24.2% of patients (72 of 301 patients) if we exclude, as was done in the Freiburg-Warsaw-Columbus study, the patients with neurofibromatosis type 1. However, 65.9% of patients with mutations (56 of 86 patients) clearly had a family disease clinically diagnosed by a specific family history and/or a syndromic presentation. In the 258 patients with an apparently sporadic presentation, we identified only 11.6% of patients having mutations, which is a percentage equivalent to our previous but more limited study.3 The use of direct sequencing method as a gold standard may limit the percentage of detected mutations. In particular, the amplification of the normal allele in heterozygous patients may not detect rare mutations such as large deletions and chromosomal rearrangements.10,11 Thus, the real number of patients with germline mutations may be slightly higher. Finally, this study supports our previous data of approximately 10% of hereditary pheo in apparently sporadic pheo.
Our findings highlight the importance of scrupulous work-up, both medically (cutaneous examination, assay of plasma thyrocalcitonin level, retinal examination, abdominal computed tomography scan, and head and neck magnetic resonance imaging scan) and genetically (asking questions about family medical history, drawing the family pedigree, giving information on the new findings concerning the genetics of pheo, proposing genetic testing, and receiving informed consent). This work-up should allow the detection of patients with clinical and paraclinical evidence to inherited syndromic pheo or pgl before genetic testing, which can then be targeted to one specific gene depending on the lesion (for example, RET for medullary thyroid carcinoma, VHL for hemangioblastoma, and SDHD/SDHB for head and neck pgl). As previously reported, we observed that a young age at diagnosis and extra-adrenal or bilateral tumors are predictive factors for a germline mutation.12,13 Moreover, we have demonstrated in this study that these parameters allow one to choose the genes to be first tested, such as VHL first in children with pheo, SDHB and SDHD for extra-adrenal tumors, and VHL and then RET for bilateral tumors. The sporadic presentation patients with a germline mutation are significantly younger than patients with no mutations, and no patients having a germline mutation were older than 58 years of age. However, the number of patients with mutations in our series was too small (< 25 patients per gene) to determine an age limit for genetic testing. So far, pheo has not been reported in the few published cases of patients with an SDHC mutation, and our data suggest that screening of this gene is not indicated.
The data from the systematic sequencing of all five pheo-pgl susceptibility genes in 314 patients suffering from pheo or functional pgl suggest that complete screening is useless. We propose the following strategy to detect the inherited forms and to reduce the delay and the cost of genetic analysis (Fig 1). The indications for genetic testing should be first targeted on the knowledge of the family history and syndromic presentation, notably for the RET, VHL, and SDHD genes. In the apparently sporadic form, the screening of the SDHB, VHL, and then the SDHD genes should be systematically carried out. In case of a malignant phenotype, the molecular analysis should first concentrate on the SDHB gene. For bilateral tumors, the VHL gene should be the first to be screened, and the SDHD and SDHB genes should be screened in patients with extra-adrenal tumors. If we had used that strategy, we would have detected 84 (98.8%) of 85 of the mutations in our cohort. Obviously, more options could also be suggested. For bilateral tumors, the VHL gene should be tested first in case of noradrenergic secretion, and the RET gene should be tested first in case of adrenergic secretion. For malignant tumors, the SDHB gene should be tested first, followed by the VHL gene. For extra-adrenal tumors, the SDHD and SDHB genes should be analyzed first, followed by the VHL gene. Finally, our results indicate that identifying a mutation in one causative gene would stop the genetic screening because none of our 314 patients harbored two mutations in two different genes.
Finally, this study also confirms that the presence of the SDHB mutation is a risk factor for malignancy and a factor of poor prognosis. Indeed, an SDHB mutation was found in most of the malignant patients (15 of 18 patients) in whom a positive genetic testing was obtained. Furthermore, among the 10 patients who died in the follow-up, nine had an SDHB mutation. We previously reported eight patients with an SDHB mutation; five of six patients with a complete follow-up had a malignant pheo.3 Several other similar isolated cases have been published.14,15 The European-American Paraganglioma Study Group reported 11 malignant pgls (34.3%) among 32 SDHB mutation carriers.16 The high percentage of malignancy in SDHB-mutated patients (71.4%) in our series could be explained by a difference in the clinical evaluations and follow-up of the patients as well as by a difference in the sources of recruitment between the different centers. We also found that, at diagnosis, the diameter of the tumor was larger in SDHB-positive patients compared with SDHB-negative patients. This may be explained by the malignant and more aggressive tumoral phenotype and/or by the delay in diagnosis of patients with an SDHB mutation and apparently sporadic presentation. The SDHB gene, like the VHL or SDHD genes, is a tumor-suppressor gene. There have been several suggestions to explain malignancy in SDHB-related malignant tumors, such as a loss of heterozygosity at 1p chromosome or activation of the hypoxia-angiogenic pathway. However, the molecular and cellular mechanisms linking SDHB mutations and tumorigenesis remain unknown. Further fundamental and therapeutic research is needed to discover new therapeutic targets and to determine curative and preventive protocols for malignant pheo.17 Altogether, these findings highlight the importance of earlier SDHB gene testing for every patient diagnosed with pheo or functional pgl. Physicians should be aware of the malignant potential of SDHB-pheo, which should be operated on as early as possible to prevent metastasis and should benefit from a strict follow-up. In conclusion, this study emphasizes the importance of genetic testing for all patients with a pheo or functional pgl. We recommend the genetic testing of SDHB and VHL for each patient with an apparently sporadic tumor and, if these tests are negative, screening of the SDHD gene. This should allow screening for most of the inherited pheo or functional pgl and the detection of SDHB-positive carriers at a high risk of malignancy.
Authors’ Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Steven Perkins and Claire Dray-Gallier for editing the English text. We also thank Valérie Nau, Isabelle Roncelin, and Valérie Boccio for technical assistance. We are grateful to the following clinicians and members of the Paraganglioma Network for their support: Patrice Rodien, Angers; Vincent Darrouzet, Cyril Goizet, and Antoine Tabarin, Bordeaux; Béatrice Hamon, Chambery; Philippe Thieblot and Christine Francannet, Clermont-Ferrand; Alexis Bozorg Grayeli and Olivier Sterkers, Clichy; Clément Abbou, Créteil; Pierre Goudet, Laurence Faivre, and Jean-Michel Petit, Dijon; Sébastien Schmerber, Grenoble; Catherine Cardot-Bauters, Marie Nocaudie, and Jean-Louis Wemeau, Lille; Philippe Chanson and Jacques Young, Le Kremlin-Bicêtre; Fran?oise Borson-Chazot, Alain Calender, Jean-Christian Pignat, and Jean-Louis Peix, Lyon; Bernard Conte-Devolx, Anne Barlier, Jean-Pierre Lavieille, Michel Zanaret, Fran?ois Eisinger, and Olivier Mundler, Marseille; Isabelle Coupier, Pascal Pujol, Jean Ribstein, Alain Bonafé, and Marc Makeieff, Montpellier; Jean-Claude Carel, Catherine Adamsbaum, Florence Tenenbaum, Cécile Badoual, Daniel Brasnu, Pierre Bonfils, Stéphane Hans, Anne Hernigou, Philippe Halimi, Anne-Sophie Jehenne, Marc Froissart, Michèle Duet, Franck Zinzindohoue, Khadija Lahlou-Laforet, Philippe Herman, and Patrice Tran Ba Huy, Paris; Dominique Gaillard, Reims; Gérard Chabrier, Jean-Louis Schlienger, Olivier Caron, Jean-Marc Limacher, and Dominique Stephan, Strasbourg; Frédéric Chabolle, Suresnes; Patrick Calvas, Jacques Amar, Béatrice Duly-Bouhanick, Thierry Battiston, and Philippe Caron, Toulouse; Patrice Beutter, Pierre Lecomte, and Alain Robier, Tours; and Agnès Chompret, Jean Lumbroso, and Sophie Leboulleux, Villejuif, France.
The acknowledgment is included in the full-text version of this article, available online at www.jco.org. It is not included in the PDF (via Adobe? Acrobat Reader?) version.
NOTES
Supported by the Cortico et Medullosurrenale: les Tumeurs Endocrines network, with the support of Projet Hospitalier de Recherche Clinique Grant No. AOM02068 and Grants from L’Institut National de la Santé et de la Recherche Médicale (INSERM) and the Ministère Délégué à la Recherche et aux Nouvelles Technologies; the Paraglioma network, with the support of Groupement d’Intèrêt Scientifique Institut des Maladies Rares; and Groupe des Tumeurs Endocrines, with the support of Ministère de la Santé et de la Protection Sociale. L.A. holds a fellowship from the Société Fran?aise d’Hypertension Artérielle, and A.-P.G.-R. holds a Contrat d’Interface pour Hospitaliers from INSERM.
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Bryant J, Farmer J, Kessler LJ, et al: Pheochromocytoma: The expanding genetic differential diagnosis. J Natl Cancer Inst 95:1196-1204, 2003
Neumann HP, Bausch B, McWhinney SR, et al: Germ-line mutations in nonsyndromic pheochromocytoma. N Engl J Med 346:1459-1466, 2002
Gimenez-Roqueplo AP, Favier J, Rustin P, et al: Mutations in the SDHB gene are associated with extraadrenal and/or malignant phaeochromocytomas. Cancer Res 63:5615-5621, 2003
Neurofibromatosis: Conference statement—National Institutes of Health Consensus Development Conference. Arch Neurol 45:575-578, 1988
Gimenez-Roqueplo AP, Favier J, Rustin P, et al: The R22X mutation of the SDHD gene in hereditary paraganglioma abolishes the enzymatic activity of complex II in the mitochondrial respiratory chain and activates the hypoxia pathway. Am J Hum Genet 69:1186-1197, 2001
Gimenez-Roqueplo AP, Favier J, Rustin P, et al: Functional consequences of a SDHB gene mutation in an apparently sporadic pheochromocytoma. J Clin Endocrinol Metab 87:4771-4774, 2002
Niemann S, Muller U: Mutations in SDHC cause autosomal dominant paraganglioma, type 3. Nat Genet 26:268-270, 2000
Bravo EL, Gifford RW Jr: Current concepts: Pheochromocytoma—Diagnosis, localization and management. N Engl J Med 311:1298-1303, 1984
Dluhy RG: Pheochromocytoma: Death of an axiom. N Engl J Med 346:1486-1488, 2002
Baysal BE, Willett-Brozick JE, Filho PA, et al: An Alu-mediated partial SDHC deletion causes familial and sporadic paraganglioma. J Med Genet 41:703-709, 2004
McWhinney SR, Pilarski RT, Forrester SR, et al: Large germline deletions of mitochondrial complex II subunits SDHB and SDHD in hereditary paraganglioma. J Clin Endocrinol Metab 89:5694-5699, 2004
Bravo EL, Tagle R: Pheochromocytoma: State-of-the-art and future prospects. Endocr Rev 24:539-553, 2003
Amar L, Servais A, Gimenez-Roqueplo AP, et al: Year of diagnosis, features at presentation, and risk of recurrence in patients with pheochromocytoma or secreting paraganglioma. J Clin Endocrinol Metab 90:2110-2116, 2005
Young AL, Baysal BE, Deb A, et al: Familial malignant catecholamine-secreting paraganglioma with prolonged survival associated with mutation in the succinate dehydrogenase B gene. J Clin Endocrinol Metab 87:4101-4105, 2002
Maier-Woelfle M, Brandle M, Komminoth P, et al: A novel succinate dehydrogenase subunit B gene mutation, H132P, causes familial malignant sympathetic extraadrenal paragangliomas. J Clin Endocrinol Metab 89:362-367, 2004
Neumann HP, Pawlu C, Peczkowska M, et al: Distinct clinical features of paraganglioma syndromes associated with SDHB and SDHD gene mutations. JAMA 292:943-951, 2004
Eisenhofer G, Bornstein SR, Brouwers FM, et al: Malignant pheochromocytoma: Current status and initiatives for future progress. Endocr Relat Cancer 11:423-436, 2004(Laurence Amar, Jér?me Ber)
Département de Génétique and Unité d’Hypertension Artérielle, H?pital Européen Georges Pompidou, and Service des Maladies Endocrininiennes et Mètaboliques, H?pital Cochin, Assistance Publique-H?pitaux de Paris
L’Institut National de la Santé et de la Recherche Médicale (INSERM) U36, Collège de France
INSERM U567
Centre National de la Recherche Scientifique Unitè Mixte de Recherche 8104, Institut Cochin
Service de Cancérologie Endocrinienne and Service de Génétique, Institut Gustave Roussy, Villejuif, France
Service de Mèdecine Interne et Endocrinologie H?pital Henri Mondor, Créteil, France
Unité d’Endocrinologie, Centre Hospitalier Universitaire (CHU) and INSERM Equipe Mixte 0105 Grenoble, France
Service de Medecine Interne et Hypertension Artérielle, CHU Rangueil
INSERM U558, 31059 Toulouse, France
Service d’Endocrinologie, CHU de Reims, Reims
Laboratoire de Génétique, H?pital Edouard Herriot, Lyon
Service d’Endocrinologie, H?pital de l’H?tel-Dieu, CHU de Nantes, Nantes
Service d’Endocrinologie, CHU de Marseille, Marseille
Génétique Oncologique Ecole Pratique des Hautes Etudes-Unitè Mixte de Recherche 8125, Faculté de Médecine Paris-Sud
Service d’Urologie, H?pital de Bicêtre, Le Kremlin-Bicêtre
Service d’Endocrinologie, CHU d’Angers, Angers
Service d’Endocrinologie, H?pital de l’Archet, CHU de Nice, Nice, France
ABSTRACT
PURPOSE: To assess the yield and the clinical value of systematic screening of susceptibility genes for patients with pheochromocytoma (pheo) or functional paraganglioma (pgl).
PATIENTS AND METHODS: We studied 314 patients with a pheo or a functional pgl, including 56 patients having a family history and/or a syndromic presentation and 258 patients having an apparently sporadic presentation. Clinical data and blood samples were collected, and all five major pheo-pgl susceptibility genes (RET, VHL, SDHB, SDHD, and SDHC) were screened. Neurofibromatosis type 1 was diagnosed from phenotypic criteria.
RESULTS: We have identified 86 patients (27.4%) with a hereditary tumor. Among the 56 patients with a family/syndromic presentation, 13 have had neurofibromatosis type 1, and germline mutations on the VHL, RET, SDHD, and SDHB genes were present in 16, 15, nine, and three patients, respectively. Among the 258 patients with an apparently sporadic presentation, 30 (11.6%) had a germline mutation (18 patients on SDHB, nine patients on VHL, two patients on SDHD, and one patient on RET). Mutation carriers were younger and more frequently had bilateral or extra-adrenal tumors. In patients with an SDHB mutation, the tumors were larger, more frequently extra-adrenal, and malignant.
CONCLUSION: Genetic testing oriented by family/sporadic presentation should be proposed to all patients with pheo or functional pgl. We suggest an algorithm that would allow the confirmation of suspected inherited disease as well as the diagnosis of unexpected inherited disease.
INTRODUCTION
Pheochromocytomas (pheo) are catecholamine-secreting adrenal medulla tumors, and functional paragangliomas (pgl) are catecholamine-secreting extra-adrenal tumors most often located in the sympathetic ganglias. In recent years, significant progress has been made in understanding the genetic determinism of these tumors. The following four inherited disorders have so far been identified: multiple endocrine neoplasia type 2, von Hippel-Lindau disease, neurofibromatosis type 1, and hereditary pheo-pgl; these syndromes are caused by mutations in the RET, VHL, NF1, and SDH (SDHD, SDHB, and SDHC) genes, respectively.1 It has been recently shown that a germline mutation may be present in 12% to 24% of apparently sporadic pheo. The Freiburg-Warsaw-Columbus Pheochromocytoma Group genotyped 271 patients with a nonsyndromic pheo and identified mutations in 66 patients (24%); 30 patients had mutations on VHL, 13 had mutations on RET, 11 had mutations on SDHD, and 12 had mutations on SDHB.2 Our group genotyped an initial series of 84 patients with an apparently sporadic pheo. These patients had been observed for an average duration of 8.8 ± 5.7 years, and a germline mutation was identified in 12% of the patients principally on the SDHB and VHL genes. A germline mutation on SDHB was associated with a high risk of recurrence or malignancy, showing the importance of SDHB genetic testing.3 Here, we have extended our recruitment and analyzed the SDHB, VHL, SDHD, RET, and SDHC genes in a series of 314 patients recruited nationally.
PATIENTS AND METHODS
In this study, 314 patients (140 men and 174 women) with a pheo or a functional pgl were recruited from several different French clinical centers participating in several clinical research networks (Cortico et Medullosurrènale: les Tumeurs Endocrines, ParaganGlioma Network, Groupe des Tumeurs Endocrines). The whole series contains the 84 patients from the COMETE network who were previously reported.3 The diagnosis of pheo or functional pgl was based on the presence of an adrenal or an extra-adrenal thoracoabdominal tumor associated with a high level of total urinary metanephrines. For every patient, the diagnosis has been confirmed by histologic examination after surgery except for three patients having an extensive malignant metastatic dissemination who could not be operated on. Malignancy was defined by the presence of extra-paraganglionic metastases, including lung, liver, bone, and lymph nodes. For each patient, a family history and clinical features included in hereditary pheo syndromes were investigated, as previously described.3 The diagnosis of neurofibromatosis type 1 was based on phenotypic criteria.4 As described elsewhere3 and in accordance with the national ethical rules, a written informed consent was obtained from every patient before drawing blood peripheral sample and DNA extraction. For each patient, we directly sequenced all the coding exons of SDHB, SDHD, SDHC, and VHL and exons 10, 11, 13, 14, 15, and 16 of RET, as previously described.5-7
The statistical analysis was performed with Staview version 5.0 (SAS Institute Inc, Cary, NC) statistical software. We used the Fisher’s exact test for small samples, the standard 2 test for larger groups, and the analysis of variance test to compare more than two variables. P < .05 was considered statistically significant.
RESULTS
Clinical Characteristics
In this series of 314 patients with pheo or functional pgl, 256 patients (81.5%) had an adrenal tumor (pheo) exclusively (41 of these patients had a bilateral pheo), 50 patients (15.9%) had a functional extra-adrenal pgl exclusively, and eight patients (2.5%) had both pheo and functional pgl. Among the 314 patients, 52 (16.6%) had a malignant form. We found a positive family history and/or a syndromic presentation in 56 patients (17.8%); 13 of these patients had a family history of pheo, nine patients had a phenotype of von Hippel-Lindau disease, 14 patients had multiple endocrine neoplasia type 2, and seven patients had hereditary pgl. Thirteen patients were diagnosed as having neurofibromatosis type 1 from clinical criteria. The other 258 patients (82.2%) had an apparently sporadic presentation. Patients with a family history and/or a syndromic presentation were younger at diagnosis than sporadic presentation patients (30.2 ± 13.1 v 43.6 ± 15.6 years of age, respectively; P < .0001) and more often had bilateral tumors (55.4% v 3.9%, respectively; P < .0001; Table 1).
Genetic Data According to the Presentation
We identified a heterozygous germline mutation in 73 patients and diagnosed neurofibromatosis type 1 in 13 patients. Thus, 86 (27.4%) of the 314 patients had a hereditary pheo or functional pgl. Among the 86 patients with a hereditary pheo or functional pgl, 56 (65.1%) had a syndromic presentation and/or a positive family history, whereas 30 (34.9%) had apparently sporadic presentation. Among the 258 sporadic presentation patients, 30 (11.6%) had a germline mutation. Most mutations (27 of 30 mutations, 90%) were identified on the SDHB (n = 18, 60%) or VHL (n = 9, 30%) genes. We did not find any mutation on the SDHC gene. Despite the fact that all patients were systematically screened for the entire set of genes, we found only one causative mutation for each patient. The type and the number of identified germline mutations according to the presentation are listed in Tables 2 and 3, respectively.
Genetic Data According to the Clinical Characteristics
Patients with a mutation were younger at time of diagnosis than patients without a mutation (P < .0001; Table 4). The younger patients were affected by the von Hippel-Lindau disease. Extra-adrenal tumors were present in 58 patients. Among them, 28 patients (48.3%) had a mutation located on the SDHB (17 of 28 patients, 60.7%), SDHD (seven of 28 patients, 25.0%), or VHL (four of 28 patients, 14.3%) genes. The tumors were more often bilateral in patients with a mutation than in patients without a mutation (36 of 86 patients, 41.9% v five of 228 patients, 2.2%; P < .0001). Most patients with bilateral tumors harbored a mutation (36 of 41 patients, 87.8%). These mutations were mainly on the VHL (17 of 36 patients, 47.2%) and RET (11 of 36 patients, 30.5%) genes. In the subgroup of 52 patients with a malignant tumor, 18 patients had a mutation, with 15 of the mutations (83.3%) on the SDHB gene. These data suggest that earlier onset, bilateral, extra-adrenal, and malignant tumors indicate the presence of a genetic disease.
NF1 Patients
All patients with neurofibromatosis type 1 had classic cutaneous lesions (café au lait pigmented spots) and neurofibromas. The average age of the 13 patients when diagnosed with pheo (40.1 years) was higher than the average age observed in the other patients with a hereditary disease. Every patient had an adrenal medullary tumor, with five patients having bilateral disease and one having a malignant pheo (Table 4).
RET Mutation Carriers
Most mutations were located in codon 634 (Table 2). The only patient with an apparently sporadic presentation harbored a mutation in exon 13 (Lys790Phe). Among the 16 patients with a RET mutation, 15 (93.8%) had a personal or family history of medullary thyroid cancer (Table 3). Among these patients, the tumors were always adrenal and often bilateral (11 of 15 patients, 73%; Table 4).
SDHD Mutation Carriers
We found nine different mutations that were nonsense or frameshift mutations (Table 2). An SDHD mutation was identified in 11 patients, among whom nine (81.8%) had a family history on the father’s side of the family (six of nine patients, 66.7%) and/or an associated head and neck pgl (seven of nine patients, 77.7%; Table 3). The majority of these patients (seven of 11 patients, 63.6%) had extra-adrenal tumors (Table 4).
VHL Mutation Carriers
Twelve different mutations were identified in exons 1 and 3 of the VHL gene (Table 2). They were missense, frameshift, and splicing mutations. We observed two different mutations in codon 98 of exon 1 (Tyr98His and Tyr98Cys) and frequent mutated codons and recurrent mutations in exon 3 (Tyr156Cys, Arg161Gln, Arg167Trp, and Arg167Gln). The 25 patients with a VHL mutation included 16 patients (64%) with family history and/or a syndromic presentation and nine patients (36%) with sporadic presentation, representing 30% (nine of 30 patients) of the patients with an apparently sporadic presentation who had mutations (Table 3). At diagnosis, a positive family history was found in 10 patients, and other syndromic lesions (renal tumors and hemangioblastomas) were found in nine patients. The tumors were bilateral in 17 patients (70.8%; Table 4). All nine sporadic presentation patients had a mutation located in exon 3 of the VHL gene. In the Tyr156Cys mutation carriers, we did not find any other syndromic von Hippel-Lindau disease lesions, except pheo or pgl.
SDHB Mutation Carriers
We found 15 different mutations with recurrent mutated codons at positions 46, 207, and 230 (Table 2). Most SDHB-positive patients belonged to the sporadic presentation group (18 of 21 patients, 85.7%), and this group was the largest group within the sporadic presentation patients who had mutations (18 of 30 patients, 62.1%; Table 3). Tumors in the SDHB mutation carriers were frequently malignant (15 of 21 patients, 71.4%). Among the 52 patients with a malignant tumor, 18 (34.6%) harbored a mutation, which was primarily located on the SDHB gene (15 of 18 patients, 83.3%). Twenty-one patients died from their malignant disease (10 patients had a mutation and 11 had no mutations), among whom nine had an SDHB mutation. In the whole series, there was a strong association between malignancy and SDHB mutation (15 [83.3%] of 18 patients were SDHB mutation–positive carriers and 37 [12.5%] of 296 patients were SDHB mutation–negative carriers; P < .0001). We found that the SDHB-mutated tumors were significantly larger than the non–SDHB-mutated tumors (P < .0006) and were more often extra-adrenal (17 of 21 patients, 81%; P < .0001; Table 4).
DISCUSSION
Until 2002, 10% of the total amount of pheo were considered to be hereditary.8 However, a large cohort of 271 patients, in which 24% of patients had sporadic pheo and no family history, challenged that notion. That study suggested that routine analysis for mutations on RET, VHL, SDHD, and SDHB genes in patients with apparently sporadic presentation of pheo should be considered as the clinical standard of care.2,9 In the present series, our results are similar to those obtained by the Freiburg-Warsaw-Columbus Pheochromocytoma Group, with 27.4% of germline mutations being identified in 314 patients or 24.2% of patients (72 of 301 patients) if we exclude, as was done in the Freiburg-Warsaw-Columbus study, the patients with neurofibromatosis type 1. However, 65.9% of patients with mutations (56 of 86 patients) clearly had a family disease clinically diagnosed by a specific family history and/or a syndromic presentation. In the 258 patients with an apparently sporadic presentation, we identified only 11.6% of patients having mutations, which is a percentage equivalent to our previous but more limited study.3 The use of direct sequencing method as a gold standard may limit the percentage of detected mutations. In particular, the amplification of the normal allele in heterozygous patients may not detect rare mutations such as large deletions and chromosomal rearrangements.10,11 Thus, the real number of patients with germline mutations may be slightly higher. Finally, this study supports our previous data of approximately 10% of hereditary pheo in apparently sporadic pheo.
Our findings highlight the importance of scrupulous work-up, both medically (cutaneous examination, assay of plasma thyrocalcitonin level, retinal examination, abdominal computed tomography scan, and head and neck magnetic resonance imaging scan) and genetically (asking questions about family medical history, drawing the family pedigree, giving information on the new findings concerning the genetics of pheo, proposing genetic testing, and receiving informed consent). This work-up should allow the detection of patients with clinical and paraclinical evidence to inherited syndromic pheo or pgl before genetic testing, which can then be targeted to one specific gene depending on the lesion (for example, RET for medullary thyroid carcinoma, VHL for hemangioblastoma, and SDHD/SDHB for head and neck pgl). As previously reported, we observed that a young age at diagnosis and extra-adrenal or bilateral tumors are predictive factors for a germline mutation.12,13 Moreover, we have demonstrated in this study that these parameters allow one to choose the genes to be first tested, such as VHL first in children with pheo, SDHB and SDHD for extra-adrenal tumors, and VHL and then RET for bilateral tumors. The sporadic presentation patients with a germline mutation are significantly younger than patients with no mutations, and no patients having a germline mutation were older than 58 years of age. However, the number of patients with mutations in our series was too small (< 25 patients per gene) to determine an age limit for genetic testing. So far, pheo has not been reported in the few published cases of patients with an SDHC mutation, and our data suggest that screening of this gene is not indicated.
The data from the systematic sequencing of all five pheo-pgl susceptibility genes in 314 patients suffering from pheo or functional pgl suggest that complete screening is useless. We propose the following strategy to detect the inherited forms and to reduce the delay and the cost of genetic analysis (Fig 1). The indications for genetic testing should be first targeted on the knowledge of the family history and syndromic presentation, notably for the RET, VHL, and SDHD genes. In the apparently sporadic form, the screening of the SDHB, VHL, and then the SDHD genes should be systematically carried out. In case of a malignant phenotype, the molecular analysis should first concentrate on the SDHB gene. For bilateral tumors, the VHL gene should be the first to be screened, and the SDHD and SDHB genes should be screened in patients with extra-adrenal tumors. If we had used that strategy, we would have detected 84 (98.8%) of 85 of the mutations in our cohort. Obviously, more options could also be suggested. For bilateral tumors, the VHL gene should be tested first in case of noradrenergic secretion, and the RET gene should be tested first in case of adrenergic secretion. For malignant tumors, the SDHB gene should be tested first, followed by the VHL gene. For extra-adrenal tumors, the SDHD and SDHB genes should be analyzed first, followed by the VHL gene. Finally, our results indicate that identifying a mutation in one causative gene would stop the genetic screening because none of our 314 patients harbored two mutations in two different genes.
Finally, this study also confirms that the presence of the SDHB mutation is a risk factor for malignancy and a factor of poor prognosis. Indeed, an SDHB mutation was found in most of the malignant patients (15 of 18 patients) in whom a positive genetic testing was obtained. Furthermore, among the 10 patients who died in the follow-up, nine had an SDHB mutation. We previously reported eight patients with an SDHB mutation; five of six patients with a complete follow-up had a malignant pheo.3 Several other similar isolated cases have been published.14,15 The European-American Paraganglioma Study Group reported 11 malignant pgls (34.3%) among 32 SDHB mutation carriers.16 The high percentage of malignancy in SDHB-mutated patients (71.4%) in our series could be explained by a difference in the clinical evaluations and follow-up of the patients as well as by a difference in the sources of recruitment between the different centers. We also found that, at diagnosis, the diameter of the tumor was larger in SDHB-positive patients compared with SDHB-negative patients. This may be explained by the malignant and more aggressive tumoral phenotype and/or by the delay in diagnosis of patients with an SDHB mutation and apparently sporadic presentation. The SDHB gene, like the VHL or SDHD genes, is a tumor-suppressor gene. There have been several suggestions to explain malignancy in SDHB-related malignant tumors, such as a loss of heterozygosity at 1p chromosome or activation of the hypoxia-angiogenic pathway. However, the molecular and cellular mechanisms linking SDHB mutations and tumorigenesis remain unknown. Further fundamental and therapeutic research is needed to discover new therapeutic targets and to determine curative and preventive protocols for malignant pheo.17 Altogether, these findings highlight the importance of earlier SDHB gene testing for every patient diagnosed with pheo or functional pgl. Physicians should be aware of the malignant potential of SDHB-pheo, which should be operated on as early as possible to prevent metastasis and should benefit from a strict follow-up. In conclusion, this study emphasizes the importance of genetic testing for all patients with a pheo or functional pgl. We recommend the genetic testing of SDHB and VHL for each patient with an apparently sporadic tumor and, if these tests are negative, screening of the SDHD gene. This should allow screening for most of the inherited pheo or functional pgl and the detection of SDHB-positive carriers at a high risk of malignancy.
Authors’ Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Steven Perkins and Claire Dray-Gallier for editing the English text. We also thank Valérie Nau, Isabelle Roncelin, and Valérie Boccio for technical assistance. We are grateful to the following clinicians and members of the Paraganglioma Network for their support: Patrice Rodien, Angers; Vincent Darrouzet, Cyril Goizet, and Antoine Tabarin, Bordeaux; Béatrice Hamon, Chambery; Philippe Thieblot and Christine Francannet, Clermont-Ferrand; Alexis Bozorg Grayeli and Olivier Sterkers, Clichy; Clément Abbou, Créteil; Pierre Goudet, Laurence Faivre, and Jean-Michel Petit, Dijon; Sébastien Schmerber, Grenoble; Catherine Cardot-Bauters, Marie Nocaudie, and Jean-Louis Wemeau, Lille; Philippe Chanson and Jacques Young, Le Kremlin-Bicêtre; Fran?oise Borson-Chazot, Alain Calender, Jean-Christian Pignat, and Jean-Louis Peix, Lyon; Bernard Conte-Devolx, Anne Barlier, Jean-Pierre Lavieille, Michel Zanaret, Fran?ois Eisinger, and Olivier Mundler, Marseille; Isabelle Coupier, Pascal Pujol, Jean Ribstein, Alain Bonafé, and Marc Makeieff, Montpellier; Jean-Claude Carel, Catherine Adamsbaum, Florence Tenenbaum, Cécile Badoual, Daniel Brasnu, Pierre Bonfils, Stéphane Hans, Anne Hernigou, Philippe Halimi, Anne-Sophie Jehenne, Marc Froissart, Michèle Duet, Franck Zinzindohoue, Khadija Lahlou-Laforet, Philippe Herman, and Patrice Tran Ba Huy, Paris; Dominique Gaillard, Reims; Gérard Chabrier, Jean-Louis Schlienger, Olivier Caron, Jean-Marc Limacher, and Dominique Stephan, Strasbourg; Frédéric Chabolle, Suresnes; Patrick Calvas, Jacques Amar, Béatrice Duly-Bouhanick, Thierry Battiston, and Philippe Caron, Toulouse; Patrice Beutter, Pierre Lecomte, and Alain Robier, Tours; and Agnès Chompret, Jean Lumbroso, and Sophie Leboulleux, Villejuif, France.
The acknowledgment is included in the full-text version of this article, available online at www.jco.org. It is not included in the PDF (via Adobe? Acrobat Reader?) version.
NOTES
Supported by the Cortico et Medullosurrenale: les Tumeurs Endocrines network, with the support of Projet Hospitalier de Recherche Clinique Grant No. AOM02068 and Grants from L’Institut National de la Santé et de la Recherche Médicale (INSERM) and the Ministère Délégué à la Recherche et aux Nouvelles Technologies; the Paraglioma network, with the support of Groupement d’Intèrêt Scientifique Institut des Maladies Rares; and Groupe des Tumeurs Endocrines, with the support of Ministère de la Santé et de la Protection Sociale. L.A. holds a fellowship from the Société Fran?aise d’Hypertension Artérielle, and A.-P.G.-R. holds a Contrat d’Interface pour Hospitaliers from INSERM.
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Bryant J, Farmer J, Kessler LJ, et al: Pheochromocytoma: The expanding genetic differential diagnosis. J Natl Cancer Inst 95:1196-1204, 2003
Neumann HP, Bausch B, McWhinney SR, et al: Germ-line mutations in nonsyndromic pheochromocytoma. N Engl J Med 346:1459-1466, 2002
Gimenez-Roqueplo AP, Favier J, Rustin P, et al: Mutations in the SDHB gene are associated with extraadrenal and/or malignant phaeochromocytomas. Cancer Res 63:5615-5621, 2003
Neurofibromatosis: Conference statement—National Institutes of Health Consensus Development Conference. Arch Neurol 45:575-578, 1988
Gimenez-Roqueplo AP, Favier J, Rustin P, et al: The R22X mutation of the SDHD gene in hereditary paraganglioma abolishes the enzymatic activity of complex II in the mitochondrial respiratory chain and activates the hypoxia pathway. Am J Hum Genet 69:1186-1197, 2001
Gimenez-Roqueplo AP, Favier J, Rustin P, et al: Functional consequences of a SDHB gene mutation in an apparently sporadic pheochromocytoma. J Clin Endocrinol Metab 87:4771-4774, 2002
Niemann S, Muller U: Mutations in SDHC cause autosomal dominant paraganglioma, type 3. Nat Genet 26:268-270, 2000
Bravo EL, Gifford RW Jr: Current concepts: Pheochromocytoma—Diagnosis, localization and management. N Engl J Med 311:1298-1303, 1984
Dluhy RG: Pheochromocytoma: Death of an axiom. N Engl J Med 346:1486-1488, 2002
Baysal BE, Willett-Brozick JE, Filho PA, et al: An Alu-mediated partial SDHC deletion causes familial and sporadic paraganglioma. J Med Genet 41:703-709, 2004
McWhinney SR, Pilarski RT, Forrester SR, et al: Large germline deletions of mitochondrial complex II subunits SDHB and SDHD in hereditary paraganglioma. J Clin Endocrinol Metab 89:5694-5699, 2004
Bravo EL, Tagle R: Pheochromocytoma: State-of-the-art and future prospects. Endocr Rev 24:539-553, 2003
Amar L, Servais A, Gimenez-Roqueplo AP, et al: Year of diagnosis, features at presentation, and risk of recurrence in patients with pheochromocytoma or secreting paraganglioma. J Clin Endocrinol Metab 90:2110-2116, 2005
Young AL, Baysal BE, Deb A, et al: Familial malignant catecholamine-secreting paraganglioma with prolonged survival associated with mutation in the succinate dehydrogenase B gene. J Clin Endocrinol Metab 87:4101-4105, 2002
Maier-Woelfle M, Brandle M, Komminoth P, et al: A novel succinate dehydrogenase subunit B gene mutation, H132P, causes familial malignant sympathetic extraadrenal paragangliomas. J Clin Endocrinol Metab 89:362-367, 2004
Neumann HP, Pawlu C, Peczkowska M, et al: Distinct clinical features of paraganglioma syndromes associated with SDHB and SDHD gene mutations. JAMA 292:943-951, 2004
Eisenhofer G, Bornstein SR, Brouwers FM, et al: Malignant pheochromocytoma: Current status and initiatives for future progress. Endocr Relat Cancer 11:423-436, 2004(Laurence Amar, Jér?me Ber)