Pheochromocytoma

A number sign (#) is used with this entry because susceptibility to the development of isolated pheochromocytoma can be caused by germline mutation in several genes, including the TMEM127 gene (613403) on chromosome 2q11 and the MAX gene (154950) on chromosome 14q23.

Mutation in the KIF1B gene (605995) on chromosome 1p36 has been identified in 1 family. A mutation in the GDNF gene (600837) on chromosome 5p may modify susceptibility to pheochromocytoma.

Pheochromocytomas most commonly occur as part of several syndromes, and mutations in the genes that cause these syndromes have been identified in patients who manifest only pheochromocytoma. These include von Hippel-Lindau syndrome (VHL; 193300), caused by mutation in the VHL gene (608537), and multiple endocrine neoplasia types IIA (MEN2A; 171400) and IIB (MEN2B; 162300), which are caused by mutations in the RET gene (164761). Pheochromocytomas also occur with paraganglioma type 1 (PGL1; 168000), type 2 (PGL2; 601650), type 3 (PGL3; 605373), type 4 (PGL4; 115310), and type 5 (PGL5; 614165), which are caused by mutations in the SDHD (602690), SDHAF2 (613019), SDHC (602413), SDHB (185470), and SDHA (600857) genes, respectively. Pheochromocytomas have less commonly been observed in neurofibromatosis I (NF1; 162200), which is caused by mutation in the gene encoding neurofibromin-1 (613113).

In addition, somatic mutation in several of the genes involved in familial disease, including NF1, VHL, RET, and MAX, have been identified in tumor tissue from patients with sporadic pheochromocytoma (Welander et al., 2012; Burnichon et al., 2012).

For associations pending confirmation, see MOLECULAR GENETICS.

Pheochromocytomas are catecholamine-secreting tumors that usually arise within the adrenal medulla. Approximately 10% arise in extraadrenal sympathetic ganglia, and are referred to as 'paragangliomas.' Approximately 10% are malignant, and approximately 10% are hereditary (Maher and Eng, 2002; Dluhy, 2002).

Bolande (1974) introduced the concept and designation of the neurocristopathies, and identified 'simple,' including pheochromocytoma and medullary carcinoma of the thyroid, and 'complex' neurocristopathies and neurocristopathic syndromes, including NF1 and MEN2.

Knudson and Strong (1972) applied Knudson's 2-mutation theory to pheochromocytoma (see discussion in 180200) and concluded that it fits.

Maher and Eng (2002) reviewed the clinical entities and genes associated with pheochromocytoma.

Familial pheochromocytoma was first reported by Calkins and Howard (1947).

Hadorn (1963) reported a German family in which 3 sibs had adrenal tumors consistent with pheochromocytomas. A brother and sister suffered from tachycardia, sweating, hypertension, and albuminuria. The sister had advanced hypertensive retinopathy and the brother had congestive heart failure. At autopsy, the sister showed cerebral hemorrhage and bilateral adrenocortical tumors. A surviving sib developed similar symptoms. The Regitine test was strongly positive, the urine contained large amounts of norepinephrine, and pneumoperitoneum demonstrated an enlarged right adrenal which contained adrenal and paraganglion tissue.

Engelman et al. (1968) noted that familial pheochromocytoma is usually bilateral and the patients are likely to show resistance to the vasopressor effects of tyramine.

Swinton et al. (1972) reported a family in which 4 members, including a father and son, had pheochromocytomas. They pointed out that associated hypercalcemia may be due to secretion of a calcitonin-like substance; hypercalcemia could be corrected by adrenalectomy.

Kaufman and Franklin (1979) reported a family with 7 documented and other possible cases of pheochromocytoma.

Ohno et al. (1982) observed pheochromocytoma in 2 sisters whose father also had pheochromocytoma. One of the sisters had aniridia and her pheochromocytoma was malignant.

In 34 sporadic and 7 familial instances of pheochromocytoma, Khosla et al. (1991) found evidence of loss of heterozygosity (LOH) at multiple sites: 1p in 42%, 3p in 16%, 17p in 24%, and 22q in 31%. They also noted a correlation between LOH on 1p and urinary excretion of metanephrine by these patients (p = 0.02). LOH on 1p, 3p, and 17p also appeared to be associated with increased tumor volume. They suggested that tumor formation and/or progression in pheochromocytoma might involve multiple genes, analogous with the model proposed for colon carcinoma (Fearon and Vogelstein, 1990). The findings of Moley et al. (1992) suggested that LOH on 1p is particularly frequent in pheochromocytoma, being found by them in all 9 pheochromocytomas in MEN2A and MEN2B, in 2 of 7 sporadic pheochromocytomas, and in 1 of 2 pheochromocytomas in von Hippel-Lindau patients.

Linkage to Chromosome 2q11

Dahia et al. (2005) reported a family of Brazilian-Portuguese descent in which 6 sibs had pheochromocytoma. Multipoint parametric linkage analysis revealed identical lod scores of 2.97 for chromosome 2cen and 16p13 loci. A 2-locus parametric linkage analysis produced a maximum lod score of 5.16 under a double-recessive multiplicative model, suggesting that both loci are required to develop the disease. Allele-specific LOH was detected only at the chromosome 2 locus in all tumors from this family, consistent with a tumor suppressing gene. High density LOH mapping with SNP-based array identified a total of 18 of 62 unrelated pheochromocytomas with LOH within the chromosome 2 region, which further narrowed down the locus to less than 2 cM. Dahia et al. (2005) interpreted their results as consistent with double-recessive digenic inheritance being responsible for the disease phenotype in this family. Qin et al. (2010) provided follow-up of the family reported by Dahia et al. (2005), and reported another affected individual. Reanalysis indicated that transmission pattern was consistent with autosomal dominant inheritance with reduced penetrance. The identification of more affected families allowed refinement of linkage to a 19.62-Mb region on chromosome 2q11.

Among a total of 130 patients with 185 pheochromocytomas, Neumann et al. (1993) determined that 43 had von Hippel-Lindau disease, 24 had MEN2, and 63 had sporadic tumors. The patients with familial pheochromocytoma were younger, had multifocal localization much more often, and had cancer more frequently; however, the frequency of extraadrenal tumors was lower in the familial cases. Pheochromocytoma was the only clinical manifestation of their syndrome in 38% of carriers of von Hippel-Lindau disease and 24% of carriers of MEN2. Neumann et al. (1993) concluded that all patients with pheochromocytoma should be screened for MEN2 and von Hippel-Lindau disease, and that all patients in families with MEN2 or von Hippel-Lindau disease should be screened for pheochromocytoma, even if they are asymptomatic.

Eisenhofer et al. (1999) found that measurements of plasma normetanephrine and metanephrine were useful in screening for pheochromocytoma in patients with a familial disposition to these tumors. Both a high sensitivity (97%) and a high specificity (96%) were found. (Normetanephrine and metanephrine are the respective O-methylated metabolites of norepinephrine and epinephrine.)

Pacak et al. (2001) reported 2 novel approaches for localization of pheochromocytoma in a patient in whom conventional imaging modalities failed to show the tumor. First, they showed that measurements of plasma free metanephrines coupled with vena caval sampling were useful for localizing occult pheochromocytoma. Second, they showed that positron emission tomographic scanning using the imaging agent 6-[18F]fluorodopamine as a substrate for the norepinephrine transporter offered a highly effective method for tumor localization. The authors concluded that these novel approaches may be of value in difficult cases, in which biochemical and clinical evidence of pheochromocytoma is compelling, yet conventional imaging modalities fail to locate the tumor.

Sawka et al. (2003) compared the diagnostic efficacy of fractionated plasma metanephrine measurements to measurements of 24-hour urinary total metanephrines and catecholamines in outpatients tested for pheochromocytoma at Mayo Clinic Rochester. The sensitivity of fractionated plasma metanephrines was 97% compared with a sensitivity of 90% for urinary total metanephrines and catecholamines (p = 0.63). The specificity of fractionated plasma metanephrines was 85% compared with 98% for urinary measurements. An adrenal pheochromocytoma was missed by urinary testing in 2 patients with familial syndromes and 1 asymptomatic patient with an incidentally discovered adrenal mass. An extraadrenal paraganglioma was missed by plasma testing in 1 patient. The authors concluded that measurements of 24-hour urinary total metanephrines and catecholamines yield fewer false-positive results, an attribute preferred for testing low-risk patients, but that fractionated plasma metanephrine measurements may be preferred in high-risk patients with familial endocrine syndromes.

Tests of plasma fractionated metanephrines levels, on which the initial diagnosis of pheochromocytoma relies, have a high false positive rate due to the disease's rarity. After a study to evaluate 3 approaches to distinguish between true-positive and false-positive tests, Algeciras-Schimnich et al. (2008) recommended that unless plasma fractionated metanephrines levels are elevated more than 4-fold above the upper limit of normal, patients with a positive plasma fractionated metanephrines test should be evaluated with urine fractionated metanephrines and serum/plasma CGA assays before being subjected to imaging or invasive diagnostic tests.

Neumann et al. (2001) stated that germline mutations in the VHL gene and in the SDHD gene together account for 15 to 20% of all nonfamilial presentations of pheochromocytoma. Neumann et al. (2002) identified germline mutations in 66 (24%) of 271 patients who presented with nonsyndromic pheochromocytoma and without a family history of disease. Eleven patients (4%) had 7 different germline mutations in the SDHD gene (see, e.g., 602690.0002; 602690.0004; 602690.0025; 602690.0026). Twelve patients (4%) had 9 different germline mutations in the SDHB gene (see, e.g., 185470.0004-185470.0006; 185470.0008; 185470.0009). Thirteen patients (5%) had 7 different germline mutations in the RET gene (see, e.g., 164761.0003-164761.0006; 164761.0011; 164761.0012; 164761.0034). Thirty patients (11%) had 22 different mutations in the VHL gene (see, e.g., 608537.0003; 608537.0014; 608537.0026). Clinically, the presence of a germline mutation was associated with younger age, multifocal tumors, and extraadrenal tumors. However, among the 66 patients who were positive for mutations, only 21 had multifocal pheochromocytoma. In 23 (35%), the tumor presented after the age of 30 years, and in 17 (8%) after the age of 40. Neumann et al. (2002) concluded that since almost one-fourth of patients with apparently sporadic pheochromocytoma may be carriers of mutations, routine analysis for mutations in the 4 genes studied is indicated to identify pheochromocytoma-associated syndromes that would otherwise be missed. Sixty-one (92%) of the 66 patients had no associated signs and symptoms of a syndrome at the time of presentation.

Mutation in the VHL Gene

In affected members of 2 unrelated kindreds with pheochromocytoma with no clinical evidence of VHL disease, Crossey et al. (1995) identified 2 missense mutations in the VHL gene (V84L; 608537.0025 and R238W; 608537.0003).

In 4 of 48 sporadic pheochromocytomas, Eng et al. (1995) identified mutations in the VHL gene. Two mutations were somatic and 2 were germline.

Woodward et al. (1997) identified germline missense mutations in the VHL gene in 3 of 8 kindreds with familial pheochromocytoma. A germline VHL mutation was also characterized in 1 of 2 patients with bilateral pheochromocytoma. No mutations were identified in the VHL or RET genes in 6 patients with multiple extraadrenal pheochromocytoma or adrenal pheochromocytoma with a family history of neuroectodermal tumors.

Brauch et al. (1997) found VHL mutations in 2 (3%) of 62 German patients with pheochromocytoma without a history of hereditary disease; No mutations were detected in the RET gene. Bar et al. (1997) found that 1 of 27 sporadic patients with pheochromocytoma had a VHL germline mutation; none had a RET mutation. Both groups concluded that sporadic pheochromocytomas are rarely associated with germline mutations in either of these genes.

Van der Harst et al. (1998) identified a mutation in the VHL gene (R64P; 608537.0015) in an uncle and his nephew with pheochromocytoma. Mutations in the VHL gene were identified in 4 other unrelated patients with pheochromocytomas (see, e.g., L63P, 608537.0016). In total, 6 (8.8%) of 68 patients with pheochromocytomas had germline mutations in the VHL gene.

Using comparative genomic hybridization, Hering et al. (2006) found that 10 (72%) of 14 pediatric pheochromocytoma tumors had a combinatorial loss of chromatin from chromosome 3p and 11p, resulting from either a total loss of chromosomes 3 and 11 (6 patients) or confined deletions of the 3p and 11p arms (4 patients). All of these patients had mutations in the VHL gene. The findings suggested that mutations in the VHL gene select for combinatorial deletions of 3p and 11p. Of the 4 remaining patients, 2 had familial syndromes (NF1 and PGL1, respectively) and 2 had unknown etiology. Hering et al. (2006) concluded that true sporadic pheochromocytoma is rare in childhood and that affected children should be screened for a predisposing gene.

Mutation in the RET Gene

In 5 of 48 apparently sporadic pheochromocytomas, Eng et al. (1995) identified mutations in the RET gene. Of these, 1 was a germline mutation (C634G; 164761.0003) and another was a somatic mutation (M918T; 164761.0013).

Mutation in the SDHD Gene

In tumor tissue from a patient with sporadic pheochromocytoma, Gimm et al. (2000) identified a mutation in the SDHD gene (P81L; 602690.0003). Flanking markers also showed loss of heterozygosity.

Mutation in the KIF1B Gene

In a pheochromocytoma tumor sample and in germline DNA from the corresponding patient, Schlisio et al. (2008) identified a mutation in the KIF1B gene (S1481N; 605995.0005). The proband was a 28-year-old female who presented at 17 months of age with a neuroblastoma and in adulthood developed a mature ganglioneuroma and bilateral pheochromocytoma. Her paternal grandfather harbored the mutant S1481N allele and also developed bilateral pheochromocytoma. Functional studies in primary rat sympathetic neurons revealed that induction of apoptosis was impaired with the S1481N KIF1B variant compared to wildtype.

Mutation in the TMEM127 Gene

Qin et al. (2010) identified 7 different heterozygous mutations in the TMEM127 gene (see, e.g., 613403.0001-613403.0004) in 7 unrelated probands with pheochromocytoma. Six of the mutations were truncating mutations, consistent with a loss of function. All tumors examined showed loss of heterozygosity at the TMEM127 locus, suggesting a classic mechanism of the 2-hit model of tumor suppressor inactivation. Four of the probands had a family history of pheochromocytoma. The average age of onset was 45.3 years, all tumors arose from the adrenal medulla, and they were bilateral in about half of cases. Overall, mutations were found in about 30% of familial cases and 3% of sporadic cases. Microarray-based expression profiling showed that the transcription signature of TMEM127-mutant tumors was increased in kinase receptor signals, similar to pheochromocytomas due to NF1 (162200) and RET (164761) mutations. This was in contrast to the expression profiles of pheochromocytomas with mutations in the VHL (608537), SDHB (185470) or SDHD (602690) genes, which were uniquely enriched in transcripts involved in response to hypoxia.

Mutation in the MAX Gene

Using exome sequencing in 3 unrelated families with bilateral pheochromocytoma, Comino-Mendez et al. (2011) identified 3 different heterozygous germline mutations in the MAX gene (154950.0001-154950.0003) that segregated with the disease. A follow-up study of 59 patients with pheochromocytoma identified 5 additional mutations (see, e.g., 154950.0004-154950.0005). Studies of tumor tissue showed a lack of full-length MAX protein and loss of heterozygosity (LOH) of the MAX allele, which resulted from paternal uniparental disomy (UPD) and loss of the maternal allele. This LOH constituted the somatic second-hit of the Knudson hypothesis. The paternal origin of the mutated allele detected in 6 families suggested preferential paternal transmission of the disease (p = 0.031). In addition, 2 children who inherited the mutation from their mother and 2 obligate carriers from another family did not develop tumors, further supporting this theory. Eight of 12 cases had bilateral tumors, and 3 of 8 probands had metastases at diagnosis. Overall, the findings indicated that MAX acts as a classic tumor suppressor gene.

Somatic Mutations

Using genomewide copy number analysis to study several genes known to be associated with pheochromocytomas, Welander et al. (2012) found that 35 (83%) of 42 samples had an altered copy number of at least 1 of the genes involved in familial pheochromocytoma. Eleven (26%) of the tumors had loss of 1 copy of NF1, and sequencing showed that 10 of the 11 carried a somatic truncating mutation in the NF1 gene. Loss of NF1 was associated with low mRNA expression in the tumors. Most tumors displayed loss of the normal allele, but in 2 cases there was no sign of loss of heterozygosity, although mRNA expression was clearly reduced. Frequent copy number variation in sporadic tumors was also observed for the VHL, SDHD, SDHAF2, and KIF1B genes. The findings suggested that the NF1 gene constitutes a common target of somatic mutations in sporadic pheochromocytomas.

By direct sequencing of the NF1 gene, Burnichon et al. (2012) identified a somatic inactivating NF1 mutation in 25 (41%) of 61 pheochromocytomas, which was associated with loss of the wildtype allele in 21 (84%) of the 25 cases. Gene expression signature of NF1-related tumors highlighted the downregulation of NF1 and the major overexpression of SOX9 (608160). Among a second set of 11 tumors, 2 sporadic tumors carried somatic mutations in NF1 as well as in another susceptibility gene. These findings suggested that NF1 loss of function is a frequent event in the tumorigenesis of sporadic pheochromocytoma.

Modifier Genes

In 1 of 28 sporadic pheochromocytomas, Woodward et al. (1997) identified a mutation in the glial cell line-derived neurotrophic factor gene (GDNF; R93W; 600837.0001), which is a natural ligand for RET. The mutation was present in both germline and tumor tissue. The findings suggested that variation at the GDNF locus may modify pheochromocytoma susceptibility.

Associations Pending Confirmation

For discussion of a possible association between pheochromocytoma and variation in the SDHAF2 gene, see 613019.0002.

The first description of pheochromocytoma is attributed to Felix Frankel (Manger, 2006). The 1886 publication described an 18-year-old woman named Minna Roll, a resident of Wittenweier (near the country town of Lahr) in Germany, who had died in 1884 (Frankel, 1886). She was treated and died at the University Hospital of Freiburg. At autopsy, bilateral adrenal tumors were found. The patient had paroxysms of palpitations, dizziness, headache, and reduction of visual acuity. Signs of hypertension included classic features of stage IV hypertensive retinopathy on retinal examination. Frankel and his colleagues in pathology reported what they called bilateral adrenal sarcoma and angiosarcoma. Remarkably, Frankel considered that abnormal quantities of a substance normally present in the blood might be released in an unregulated manner to the circulation, resulting in 'irritation' of the blood vessels and parenchyma of other organs. Thus, he postulated the endocrine nature and function of the adrenal medulla. Living relatives of Minna Roll were identified in the Black Forest region. Three of them were found to have pheochromocytoma plus medullary thyroid carcinoma, 3 others had pheochromocytoma, and 1 had isolated medullary thyroid carcinoma. In a grandnephew of the proband, a cys634-to-trp missense mutation of the RET gene (164761.0053) was found. In the gross autopsy, a 'goiter' was described in the proband's thyroid, which was not histologically pursued. Given the medullary thyroid carcinoma and pheochromocytoma in this family and the RET mutation, the patient clearly had multiple endocrine neoplasia type 2 (171400).

Fairchild et al. (1979) described a 29-year-old woman who had neuroblastoma (256700) during infancy, developed an extraadrenal pheochromocytoma at age 16 years, with subsequent hepatic recurrence, and was found to have multifocal renal cell carcinoma (144700). Although renal cell carcinoma and pheochromocytoma are combined in the von Hippel-Lindau syndrome, there was no other evidence for VHL in this patient. Schimke et al. (2010) reported 2 sibs of the patient reported by Fairchild et al. (1979) who developed paraspinal paragangliomas in adulthood, and a cousin of these sibs who died of metastatic renal cell carcinoma and had a history of a benign paraaortic PGL. Genetic analysis identified a heterozygous mutation in the SDHB gene (V140F; 185470.0016), consistent with paragangliomas-4 (PGL4; 115310). There were 2 unaffected family members, suggesting decreased penetrance or a 'leaky' mutation. Schimke et al. (2010) noted the importance of family history in elucidating the etiology of this inherited disorder.