Pediatric Pheochromocytoma Workup

Perform initial workup of pheochromocytoma using the history, physical examination, laboratory, and diagnostic test findings. Indications for evaluation include the following:

• High blood pressure or recurrent hot flushes that are indicative of blood pressure peaks
• An adrenal mass
• Family history of multiple endocrine neoplasia type 2 (MEN 2) or von Hippel-Lindau disease

The cornerstone of pheochromocytoma diagnosis is the measurement of norepinephrine, epinephrine, and their catabolic products in the urine. Increasing evidence supports the measurement of metanephrines, in either plasma or urine, for detection of pheochromocytomas.

A complete blood count (CBC) and metabolic panel are indicated. On urinalysis, proteinuria may be found because of hypertension. Plasma catecholamines may be measured during a paroxysm. Tests for both suppression and stimulation of catecholamine release have been proposed.

When the diagnosis has been established, the tumor must be located to facilitate its surgical removal. Although larger tumors can usually be located easily with sonography, the smallest tumors may require CT scanning or MRI, particularly when located outside the adrenal area. Scintigraphy may be a valuable adjunct in the detection of extra-adrenal lesions.

Chest radiography can be used to evaluate for pulmonary edema.

Electrocardiography (ECG) findings may be abnormal, exhibiting changes such as left ventricular hypertrophy, tachycardia, and arrhythmias.

The standard method for confirming the diagnosis of pheochromocytomas is to measure the following urinary catecholamines and their metabolites in a 24-hour specimen:

• Epinephrine
• Norepinephrine
• Dopamine
• Metanephrine
• Homovanillic acid (HVA)
• Vanillylmandelic acid (VMA)

Creatinine levels should be determined for each 24-hour collection to assess the adequacy of the collection. If possible, the collection should be made while the patient is at rest and without recent exposure to radiographic contrast medium.

Foods and medications and foods that are known to interfere with the assay should be avoided. Foods that can increase urinary catecholamines include the following:

• Coffee
• Tea
• Bananas
• Chocolate
• Cocoa
• Citrus fruits
• Vanilla

The following drugs can increase catecholamine measurements:

• Caffeine
• Acetaminophen (Tylenol)
• Levodopa
• Lithium
• Aminophylline
• Chloral hydrate
• Clonidine
• Disulfiram
• Erythromycin
• Insulin
• Methenamine
• Methyldopa
• Nicotinic acid (large doses)
• Quinidine
• Tetracyclines
• Nitroglycerin

Drugs that can decrease catecholamine measurements include the following:

• Clonidine
• Disulfiram
• Guanethidine
• Imipramine
• Monoamine oxidase inhibitors (MAOIs)
• Phenothiazines
• Salicylates
• Reserpine

The major cause of false-positive catecholamine excretion results is administration of exogenous catecholamines, such as levodopa, methyldopa, and labetalol, which can elevate urine concentration for as long as 2 weeks.

Excessive stimulation of the sympathoadrenal system, such as those occurring in hypoglycemia, strenuous exertion, increased intracranial pressure, and clonidine withdrawal, may also increase catecholamine excretion enough to provide a false-positive result.

Urine should be acidified (pH < 3) and kept cold during and after the collection. The diagnostic yield is increased if the patient is symptomatic during the collection period.

A total urinary catecholamine excretion that exceeds 300 mcg/d is commonly found, provided that the patient is symptomatic or hypertensive at the time of the collection. Specific assays of epinephrine are frequently beneficial because excretion in excess of 50 mg/day suggests an adrenal lesion. In patients with benign pheochromocytoma, excretion levels of dopamine (DA) and DA metabolites, such as homovanillic acid (HVA), are usually normal. An increased level of urinary DA or HVA suggests malignancy.

In adults, the following conditions are also associated with an elevated dopamine (DA) level:

• Overcollection of urine
• Drug effects (eg, levodopa, methyldopa, clozapine, antidepressants, metoclopramide)
• Clinical effects (including those due to pheochromocytoma, carcinoid tumor, and pregnancy)

In children, high urine DA levels are found in the following conditions, addition to pheochromocytoma:

• Neuroblastoma
• Costello syndrome
• Leukemia
• Menkes disease
• Rhabdomyosarcoma of the bladder

Tumor size correlates with the ratio of free catecholamine metabolites in the urine. Patients with small pheochromocytomas tend to have low concentrations of catecholamines with high turnover and low urinary VMA-catecholamines ratios.

Conversely, large tumors tend to have high concentrations of catecholamines, low turnover rates, and high urinary VMA-catecholamine catecholamine ratios. Small tumors that store catecholamines well or metabolize a substantial amount of catecholamines within the tumor grow larger before becoming manifest.

Metanephrine is a metabolite of epinephrine. Free plasma metanephrine levels have been found to be a highly sensitive (100%) and specific (96.7%) measure, yielding a negative predictive value of 100%. Measurement of plasma normetanephrine and metanephrine are useful in screening for pheochromocytomas in patients with a familial predisposition to von Hippel-Lindau disease or MEN type 2.

Measurement of 24-hour urinary fractionated metanephrines using a tandem mass spectrometry assay appears to be an effective biochemical technique in the investigation of pheochromocytoma. Cut-offs for positivity with this study are defined as follows:

• Total metanephrines (sum of the metanephrine fractions) - 5163 nmol/d
• Normetanephrine fraction - 4001 nmol/d
• Metanephrine fraction - 1531 nmol/d

The diagnostic efficacy of 24-hour urinary fractionated metanephrines using tandem mass spectrometry with cut-offs for positivity is as follows:

• Normetanephrine fraction sensitivity is 87.3% and specificity is 95.0%
• Metanephrine fraction sensitivity is 56.9% and specificity is 95.0%
• Elevation of either normetanephrine or metanephrine fraction sensitivity is 97.1% and specificity is 91.1%

Areas under the receiver operating characteristic (ROC) curves are as follows:

• Total metanephrines - 0.991 (95% confidence index [CI], 0.958-0.996)
• Normetanephrine fraction - 0.972 (95% CI, 0.955-0.990)
• Metanephrine fraction - 0.8 (95% CI, 0.741-0.858)
• Metanephrine and normetanephrine fractions - 0.991 (95% CI, 0.985-0.996)

In a study of 22 patients with histologically proven pheochromocytoma, plasma free metanephrine level measurement has shown 100% sensitivity and 97.6% specificity. In comparison, sensitivity and specificity of other tests were as follows^[14] :

• Plasma catecholamine levels - 78.6% and 70.7%
• Urinary catecholamines - 78.6% and 87.8%
• Urinary metanephrines - 85.7% and 95.1%
• Urinary hydroxymethoxymandelic acid - 93% and 75.8%

All patients with pheochromocytoma had plasma free metanephrine levels at least 27% above the upper limit of the reference range. Only 3 other patients had concentrations of more than 50% above the upper limit of the reference range.^[14]

In a similar study of 159 outpatients tested in a tertiary referral center for pheochromocytoma over a 4-year period, the sensitivity of urinary free metanephrine level measurement was 100% (25 of 25 patients), and the specificity was 94% (116 of 123 patients).^[15] The sensitivity and specificity of other studies was as follows:

• Urinary catecholamines - 84% (21 of 25 patients) and 99% (127 of 129 patients)
• Urinary VMA - 72% (18 of 25 patients) and 96% (130 of 134 patients)
• Plasma catecholamines - 76% (16 of 21 patients) and 88% (66 of 75 patients)

Thus, metanephrines measured in either plasma or urine seemed to be the test of choice for detection of pheochromocytomas.

A CBC count is indicated when infection or abdominal pain is present.

Electrolytes, blood urea nitrogen (BUN), creatinine, and glucose determinations evaluate for lactic acidosis; renal failure secondary to hypertension, renal damage, or both; and hyperglycemia or hypoglycemia caused by the impaired insulin response. High levels of calcium may be present because of excess of parathyroid hormone (PTH).

For measurement of plasma catecholamines, patients must be in a basal and calm state. The measurement reflects only that single moment when the blood sample was obtained. Basal levels of more than 2 ng/dL support the diagnosis, whereas values of less than 0.5 ng/dL make the diagnosis unlikely.

Suppression and stimulation tests

Suppression tests (eg, phentolamine, clonidine) and stimulation tests (eg, glucagon, histamine, metoclopramide) have both been proposed for improving diagnostic accuracy of plasma catecholamine measurement. Stimulation tests are dangerous; administer with extreme caution.

Consider a glucagon stimulation test if basal plasma catecholamine values are from 0.5-1 ng/dL. Patients demonstrate a significant rise in plasma catecholamine levels within minutes of glucagon administration; however, this can lead to severe hypertension.

When the values are 1-2 ng/dL, a clonidine suppression test, which is highly sensitive and specific, is indicated. Mildly elevated levels of catecholamines in healthy individuals are suppressed by a dose of clonidine. The clonidine suppression test (0.3 mg) involves the collection of plasma free normetanephrine before and after oral administration of clonidine.

When the diagnosis of pheochromocytoma has been established, the tumor must be located to facilitate its surgical removal. Although larger tumors can usually be located easily with ultrasonography, the smallest tumors may require CT scanning or MRI, particularly when located outside the adrenal area. (See the images below.)

The average diameter of a pheochromocytoma is approximately 5 cm at the time of diagnosis, and the sensitivity of CT scanning or MRI approaches 100%, although the specificity is lower. The major drawback of CT scanning is its relatively poor tissue discrimination, particularly in the perinephric zone. A head CT scan is indicated if abnormal neurologic examination findings are noted.

Scintigraphy with radiolabeled iodine-131 (¹³¹ I) or¹²³ I-metaiodobenzylguanidine (MIBG) allows whole-body exploration. Owing to its high specificity (97%), this morphologic study seems to be a valuable adjunct in the detection of extra-adrenal lesions.

The main limitation of MIBG scintigraphy is its slightly lower sensitivity (adrenal, 84%; extra-adrenal, 64%) than MRI (adrenal, 97%; extra-adrenal, 88%) or CT scanning (adrenal, 94%; extra-adrenal, 64%). However, despite the lower sensitivity, MIBG scanning offers the greatest specificity, and tumors seen on these images are almost certainly pheochromocytomas. If the MIBG scanning results are positive in a child, consider a diagnosis of neuroblastoma until proven differently.

In patients with adrenal incidentalomas (ie, asymptomatic adrenal masses discovered incidentally during high-resolution imaging procedures, such as CT scanning or MRI), MIBG scintigraphy has been used in combination with measurement of serum chromogranin-A (CgA) to rule in or rule out pheochromocytoma.^[16]

CgA is a member of the granin family contained in secretory vesicles of chromaffin adrenal cells. Serum CgA levels increase in patients with pheochromocytoma and other diseases of the chromaffin system. Giovanella et al reported that patients with adrenal incidentalomas larger than 20 mm without clinical or biochemical signs and a positive¹²³ I-MIBG finding had a confirmed pheochromocytoma in all cases.^[16] Serum levels of CgA were significantly higher in patients with pheochromocytoma. Giovanella et al suggest that a negative CgA assay result may obviate MIBG imaging.

When a pheochromocytoma is suspected biochemically and is not visualized on standard studies, consider the possibility of an intrathoracic or intracranial tumor. MRI is probably the best study for detecting extra-abdominal tumors, although MIBG scanning may help reveal the general location of the tumor. CT scans or MRI can be focused in that region to localize the tumor.

MIBG may be used to seek extra-adrenal tumors or to provide additional diagnostic information about adrenal masses found with conventional techniques.

Positron emission tomography (PET) with radiopharmaceuticals provides functional imaging. It is designed to show substrate precursor uptake, cellular metabolism, or receptor binding in neoplasms with CT as a single modality; hybrid PET/CT directly correlates function and anatomy.

PET scanning uses 3,4-dihydroxy-6-(18)F-fluoro-phenylalanine ([18]F-FDOPA), an amino acid transporter substrate, as an independent marker for detection of benign and malignant pheochromocytomas. It identifies localized pheochromocytoma with a sensitivity of 84.6%, a specificity of 100%, and an accuracy of 92%.^{[17, 18, 19]}

Zelinka et al recommended that bone scintigraphy be used in the staging of patients with malignant pheochromocytoma and paraganglioma, particularly in patients with SDHB mutations. In patients without SDHB mutations, the modality with the best sensitivity for bone metastases was (18)F-FDOPA PET scanning.^[20]

Arteriography and selective venous sampling are almost never indicated. However, they may be helpful in patients predisposed to multiple tumors or when clinical and biochemical evidence is consistent with pheochromocytoma but unsupported by other imaging modalities.

If a dominant tumor is identified and another smaller lesion is seen (particularly in the contralateral adrenal gland) that does not appear typical for pheochromocytoma on MRI or MIBG scanning, selective venous sampling may resolve the issue.

The intra-arterial administration of radiographic contrast media releases catecholamines; if pheochromocytoma is suspected, perform arteriography only in patients who have received adrenergic blocking agents. However, radiopaque contrast media can be safely administered intravenously.

Genetic testing should be performed in patients with pheochromocytoma who have the following^[21] :

• A family history of pheochromocytoma or paraganglioma syndrome
• Age less than 35 years
• Multifocal tumors
• Bilateral or multicentric adrenal pheochromocytomas
• Sympathetic paragangliomas, especially multiple tumors

Those characteristics are highly associated with the presence of a germline mutation. Because several genes are involved in the genetics of pheochromocytoma and paraganglioma, prioritizing the gene (or genes) to be tested first using simple clinical information can reduce the efforts and costs of this analysis.^[22]

Jimenez et al suggested that the age for screening of sporadic pheochromocytoma should be reduced to patients younger than 20 years.^[23] Their recommendation was based on a study done by Neumann et al, which found that hereditary disease (70% of cases are VHL mutations) is usually seen in young patients.^[24]

Furthermore, in patients older than 50 years, the probability of having any genetic mutation is less than 1.3%. Thus, genetic testing should be done in young adults, focusing on ruling out VHL mutations first, followed by mutations in MEN2, SDHB, and SDHD.

The clinicians offering and performing the genetic testing should provide or make available adequate counselling and preventive treatment. Considering collaborating with referral centers as well as research groups would help coordinate the management of these families.

Researchers are studying the use of gene expression profiling to distinguish between metastatic and benign pheochromocytomas, as this differentiation cannot be reliably done by histopathologically.^{[25, 26, 27]} Malignant potential appears to be largely characterized by a less-differentiated pattern of gene expression.^[27]

Suh et al identified 10 differentially expressed genes that had high diagnostic accuracy; 5 of these genes in combination had an area under the receiver operating characteristic (ROC) curve of 0.96 for distinguishing benign versus malignant tumors.^[26]

Characteristically, pheochromocytomas have 2 populations of cells that can be observed with microscopy and distinguished with immunohistochemistry: the cells in the Zellballen stain positive for chromogranin, neurospecific enolase markers for cells of neuronal derivation, or both, whereas the sustentacular cells stain positive for S-100 protein (a marker for cells of schwannian derivation).

With electron microscopy, the cells in the Zellballen show neuronal features with abundant ectoplasmatic processes that contain dense-core neurosecretory granules.

Benign and malignant pheochromocytomas cannot reliably be distinguished with microscopy examination. Features that have been suggested to correlate with malignancy behavior include degree of necrosis, nuclear pleomorphism, mitotic rate, capsular invasion, and vascular invasion; none of these has proven to reliably indicate malignancy. Only the presence of metastatic disease is an absolute indicator of malignancy.

Malignant pheochromocytomas occur in 10-15% of cases. Histologic discrimination from benign cases is unreliable. Possible tissue markers for malignancy have been characterized, including different angiogenetic factors, including cyclooxygenase-2, secretogranin II-derived peptide, and heat-shock protein 90 (hsp90), which are mutations in the SDHB gene that encode succinate dehydrogenase subunit B.

Expression of human telomerase reverse transcriptase (hTERT) with concomitant high telomerase activity or high Ki-67 immunoreactivity apparently identifies invasive behavior of the tumor with notable specificity.

Altered states of hsp90 and hTERT may not only be useful for the classification of pheochromocytomas as malignant but could also hold therapeutic potential because cancer cells might be especially dependent on hsp90 to ensure correct folding and function of the large quantities of mutated and overexpressed oncoproteins.