Pheochromocytomas and paragangliomas (PPGLs) are catecholamine-producing tumors that arise from chromaffin cells of the adrenal gland and extra-adrenal tissue, pheochromocytomas (PCC) and paragangliomas (PGL), respectively. Pediatric PPGLs have important differences compared to those in adults, a consequence of the more frequent genetic predisposition seen in the younger population (80% in pediatric patients compared to 40% in adults, see Table 1 ) (2, 4). In children with PPGL, disease is more likely to be extra-adrenal (30-60%), bilateral adrenal (12-48% of pediatric vs. 7-14% of adult patients), recurrent (13-38%), and multifocal (17-62%) (2, 4–8). The most common underlying genetic causes of pediatric PPGL are von Hippel-Lindau syndrome (VHL, 27-51%), mutations in succinate dehydrogenase subunit genes (collectively referred to as SDHx, 13-39% in SDHB and 8-10% in SDHD), multiple endocrine neoplasia type 2 (MEN-2, 0.6-10%), and neurofibromatosis type 1 (NF-1, 1-3%) (2, 4, 5, 7–11).

The underlying pathologic mechanisms that give rise to PPGL are heterogeneous and have been organized into clusters of genes by their effect on different intracellular pathways and gene expression profiling, including pseudohypoxia (cluster 1), kinase signaling (cluster 2), and Wnt signaling (cluster 3) (10, 12–15). The significance of this molecular taxonomy is not only descriptive but also helps to define different aspects of chromaffin cell metabolism that have direct clinical applications, such as the prototypical examples of VHL (cluster 1) and MEN-2 (cluster 2).

PCCs in patients with MEN-2 produce either only epinephrine or both norepinephrine and epinephrine (and their metabolites normetanephrine and metanephrine, respectively; collectively termed metanephrines) and have higher catecholamine levels due to increased expression of tyrosine hydroxylase, the rate-limiting enzyme in catecholamine synthesis (12). PCCs in VHL patients produce norepinephrine almost exclusively, related to decreased expression of the enzyme phenylethanolamine N-methyltransferase (PNMT) which converts norepinephrine to epinephrine (12). Furthermore, it was shown that expression of proteins involved in catecholamine secretion are differentially regulated between MEN-2 and VHL, as VHL-related PCCs show decreased expression of some secretory components, resulting in a less coordinated secretion system and therefore, continuous secretion of catecholamines in sharp contrast to paroxysmal release of catecholamines by MEN-2 PCCs (13). One study reported that, on average, norepinephrine-secreting PPGLs stored 1,760,000 picograms (pg) of norepinephrine/gram (g) tissue with 53% released each day, in contrast to epinephrine-secreting PPGLs that contained 3,801,000 pg of epinephrine/g tissue with only 5% released daily (16). Norepinephrine has a higher affinity for α₁-adrenergic receptors, and epinephrine has a higher affinity for β₁-adrenergic receptors, which broadly translates into more of an increased risk for hypertension in those with tumors that produce predominantly norepinephrine and more risk for tachycardia and arrhythmias in those tumors that produce predominantly epinephrine (17).

Taken together, these molecular and biochemical features are foundational for understanding the different clinical features in these patients, with paroxysmal hypertension occurring in the context of the adrenergic biochemical phenotype (increased epinephrine and metanephrine) of MEN-2 as compared to sustained hypertension with the noradrenergic biochemical phenotype (increased norepinephrine and normetanephrine) of VHL. Additionally, decreased expression of dopamine β-hydroxylase or tyrosine hydroxylase can result in a dopaminergic or biochemically silent (non-secreting) phenotype, respectively, which may both be seen from mutations in cluster 1 genes (18). Pediatric PPGL is more often due to pathogenic variants in cluster 1 genes (particularly VHL, SDHx, and EPAS1; at least 57-80% of patients with pediatric PPGL when somatic variants are included) and is thus more likely to have a non-adrenergic (noradrenergic, dopaminergic, or non-secreting) biochemical phenotype (93% of pediatric patients without increased plasma metanephrine vs. 57% in adults, see Table 1 ) (2, 4, 5, 7).

About 10-20% of PPGLs are diagnosed in pediatric patients (19), therefore, it is important for clinicians who care for pediatric patients to be aware of this treatable cause of hypertension, not only to alleviate the symptomatic burden of catecholamine excess, but also to make a timely diagnosis due to the risk of metastatic disease and consequent morbidity and mortality. Among adults under the age of 35 years with PPGL, many similarities to pediatric PPGL are seen (including hereditary, noradrenergic, and multifocal disease), and bilateral tumors were seen even more often than in children or in adults over the age of 35 years, suggesting that these tumors may have evaded clinical detection and were in fact present earlier in childhood (2), thus highlighting the need for clinicians to be aware of the possibility of PPGL in children and adolescents. Retrospective analysis from the Department of Defense Serum Repository found that in adults, biochemical evidence of elevated plasma metanephrines was seen at a median of 6.6 years (elevation above the upper reference limit [URL]) and 4.1 years (3 times above the URL) prior to diagnosis of PCC (20). Therefore, it is reasonable to suspect that pediatric PPGLs may go undetected for many years before diagnosis, putting these patients at risk not only for continued tumor growth and metastatic disease but also catastrophic complications if stored catecholamines in a clinically occult PPGL are suddenly released when provoked by surgery, induction of anesthesia, or as an adverse effect of medication.

While isolated PPGLs are often definitively treated by surgical removal, metastatic disease requires additional treatment modalities, such as chemotherapy, radiotherapy, targeted molecular therapies, or ablative therapies (21). Metastatic disease, defined by the presence of chromaffin tumor cells in tissues without chromaffin cells, is a major aspect of PPGL care and is itself largely genetically determined (1). Metastatic disease can occur in 12% of pediatric patients, in general, but may be up to 70% among SDHB mutation carriers, the second most commonly mutated gene in pediatric PPGL after VHL (19, 22, 23). The sites of involvement are most often bone, lymph nodes, liver, and lungs (though primary PGLs have been described in liver and lungs), which can result in significant impairment in quality of life and prognosis (24, 25). Metastases may be present at the time of initial diagnosis (synchronous) or later (metachronous or non-synchronous); in pediatric PPGL, metastases are less often synchronous as compared to adults, underscoring the importance of long-term surveillance for any child diagnosed with PPGL (2). In addition to those patients who harbor a mutation in SDHB, other established risk factors for metastatic disease include tumor size ≥ 5 cm, extra-adrenal PGL, dopaminergic phenotype (plasma 3-methoxytyramine higher than three times the URL), and a Ki-67 index > 3% (3, 26).

Finally, metastatic disease has direct implications in the approach to and management of pediatric PPGL, including imaging and treatment. Advances in functional imaging modalities with positron emission tomography/computed tomography (PET/CT) scans using different radionuclides provide insight into how patients from particular genetic backgrounds can receive an increasingly personalized approach to management of these challenging tumors. And while metastatic PPGL is incurable, knowledge of the molecular pathogenesis can direct treatments in some cases, informed by the cellular pathways disrupted by the genetic predisposition. For these reasons, the genetic background may guide management of the pediatric PPGL patient, and genetic testing is essential to providing optimal care for these patients.

PPGLs are estimated to occur at an incidence of 0.57 per 100,000 person-years, of which, up to 20% of PPGLs are diagnosed in pediatric patients, at an average age of 11 years (19, 22, 23, 27). Most pediatric patients are symptomatic (around 90%), and hypertension is the most common presentation of pediatric PPGL (64-93%), which causes pediatric hypertension in up to 1% of cases (7, 8, 22). Therefore, one should suspect PPGL in pediatric patients with sustained hypertension, with headache (39-95%), diaphoresis (90%), palpitations (53%), and signs/symptoms of mass effect or as an incidental mass (30%) ( Table 1 ) (8, 11, 22).

Pediatric hypertension is defined as an auscultatory blood pressure > 130/80 mmHg for children ages 13 years and above or ≥ 95^th percentile for age, sex, and height for children ages 1 to 12 years, on more than 3 occasions, according to the most recent Clinical Practice Guideline from the American Academy of Pediatrics (28). Pediatric hypertension can be further complicated by hypertensive emergency in patients with catecholamine excess, such as retinopathy, resulting in visual disturbances, and hypertensive encephalopathy, resulting in seizures and disorientation (22, 29). Hypertrophic and dilated cardiomyopathy can arise as sequelae of hypertension in pediatric patients with PPGL; among pediatric patients with dilated cardiomyopathy, surgical excision of their tumors led to resolution of hypertension and improved cardiac function (8, 30).

Tachycardia and dysrhythmias, by contrast, are seen more often in adults, which may be related to the relatively higher proportion of cluster 2 mutations in adults resulting in stimulation of cardiac β₁-adrenergic receptors by epinephrine (22). Stimulation of β-adrenergic receptors by epinephrine in cluster 2 mutations results in hepatic glycogenolysis and gluconeogenesis, which may account for the higher proportion of adults as opposed to children who present with elevated fasting glucose levels (22, 31).

In addition to the signs and symptoms noted above, other findings related to catecholamine excess may be non-specific and include pallor, orthostatic hypotension and syncope, shortness of breath, abdominal pain, nausea, vomiting, constipation, diarrhea, hyperglycemia, polyuria and polydipsia, anxiety, behavioral symptoms, worsening performance in school, and ADHD (9, 30, 32). Interestingly, among pediatric patients harboring an SDHB mutation, sweating was significantly associated with earlier development of metastatic disease (p = 0.0073), and for those without metastases at initial diagnosis, patients with tumor pain developed metastases at an earlier interval than those without tumor pain (p = 0.0088) (23).

Presenting signs and symptoms may also be related to mass effect of the growing tumor, and patients may present with a palpable abdominal mass on physical examination (30). Abdominal masses may present with abdominal pain and distension or back pain, and bladder masses may result in hematuria or symptoms with voiding (11, 32). Parasympathetic head and neck PGLs (HNPGLs) would not be expected to produce catecholamines, and the clinical presentation may be related to mass effect on cranial nerves and other local structures, resulting in hearing loss, tinnitus, hoarseness, dysphagia, cough, pain, or feeling of fullness in the neck (32). As in adults, PPGLs may also be found as an incidental mass on imaging studies performed for other indications (8, 11). For patients with clinical findings of or genetic predisposition to PPGL, biochemical evaluation is indicated.

The biosynthesis of catecholamines, metanephrines, and 3-methoxytyramine (3-MT, the O-methylated metabolite of dopamine), as derivatives of tyrosine, is shown in Figure 1 . Plasma free metanephrines should be measured, as this highly sensitive test can help to rule out PPGL in children, as in adults (33). Eisenhofer et al. found that the optimal combination of diagnostic sensitivity (97.9%) and specificity (94.2%) was achieved by using age-nonspecific URLs for metanephrine (446 picomoles/liter [pmol/L]) and 3-methoxytyramine (107 pmol/L) with age-specific URLs for normetanephrine, increasing from age 5 years (542 pmol/L) to age 65 years (1092 pmol/L) as modeled by the equation URL_NMN = (2.07 × 10^-3 × age ³) + 545 (pmol/L) ( Table 2 ) (34). Although plasma fractionated metanephrines are the test of choice in children as in adults, urinary free metanephrines may be considered without much loss of sensitivity (97.9% for plasma free metanephrines and 93.4% for urinary free metanephrines), which may be preferable to avoid venipuncture in some children (35).

Dopamine-producing PPGLs show elevations of 3-MT; thus, plasma 3-MT should be used to increase the diagnostic sensitivity in PGLs, especially for HNPGLs (22.1% to 50%) using a cut-off of 0.1 nanomoles/liter (nmol/L) (36). Urinary dopamine reflects renal metabolism and is not helpful to identify dopaminergic PPGL (36). Although testing for 3-MT is of more limited clinical availability than that of metanephrines, it is helpful to assess risk of metastatic disease, as plasma elevations above 0.2 nmol/L had a 57% sensitivity and 85% specificity for metastatic disease (37).

Chromogranin A is a biomarker of neuroendocrine tumors, including PPGL, and has been shown to increase the sensitivity of diagnosis when used in conjunction with metanephrines, especially in patients with SDHB-related PCC or sympathetic PGL, though its performance was not as good in patients with HNPGL (38). As its production is independent of the catecholamine metabolism pathway, it is a helpful adjunct to detect non-secreting tumors.

Care must be taken to reduce false-positive biochemical laboratory results. This can be achieved by avoiding circumstances that provoke an adrenergic response at the time of collection. For this reason, the patient should be placed in a supine position with venipuncture performed 20-30 minutes before biochemical labs are obtained – as both non-supine position and venipuncture may lead to catecholamine release (32). Medications that can artifactually increase plasma or urinary fractionated metanephrines include acetaminophen, α-methyldopa, tricyclic antidepressants, monoamine oxidase inhibitors, sympathomimetics, catecholamine reuptake inhibitors, mesalamine/sulfasalazine, phenoxybenzamine, levodopa, anesthetics, neuromuscular blockers, antiemetics, linezolid, peptide hormones, steroids, cocaine, and opioids (17, 39). Glucocorticoids potentiate catecholamine biosynthetic enzymes and should be avoided in patients with PPGL due to risk of causing PCC crisis (40). Therefore, when able, these medications should be held before biochemical labs are obtained.

Cellular consequences of cluster 1 mutations converge on hypoxia-inducible factor 2α (HIF-2α), a transcription factor that dimerizes with HIF-1β and mediates downstream effects such as blocking the glucocorticoid-induced expression of PNMT (41). Gain-of-function mutations in EPAS1 (encoding for HIF-2α) contribute to aberrant neuroendocrine-to-mesenchymal transition which gives rise to PPGLs and contributes to their metastatic behavior (41). Biochemically, this translates to the observation that PNMT and other biosynthetic enzymes are down-regulated in patients with cluster 1 mutations, such that the biochemical phenotype is either noradrenergic, dopaminergic, or non-secreting, but not expected to be adrenergic.

Finally, the metabolic role of succinate dehydrogenase in the tricarboxylic acid (TCA) cycle and electron transport chain is related to diagnostic imaging and may impact treatment. Mutations in TCA cycle genes (cluster 1A) result in decreased function of these metabolic enzymes and increase dependency on glycolysis for energy production (15). Radiologically, this correlates with higher uptake of ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) on PET imaging for tumors of a cluster 1 genetic background when compared with cluster 2 (42). In addition to its role in the conversion of succinate to fumarate in the TCA cycle, succinate dehydrogenase also plays a crucial role as complex II in the mitochondrial electron transport chain, using reducing equivalents of FADH₂ to ultimately generate ATP (43, 44). Defects in SDH subunit genes (i.e., SDHx) may increase the cell’s dependence on NADH oxidation through complex I, resulting in increased availability of NAD⁺ to be used by poly (ADP-ribose) polymerase (PARP) in base excision repair of DNA (43). Therefore, SDHx mutations increase chemoresistance to genotoxic agents by stimulating DNA repair mechanisms via PARP. For these reasons, PARP inhibition may deprive these tumors of this protective mechanism and play a role in personalized treatment tailored by genetics (43).

Anatomic imaging is essential to the diagnosis for any patient being evaluated for PPGL and is typically performed by CT or MRI. In pediatric patients, high signal-intensity T2-weighted MRI is preferable over CT to reduce radiation exposure and may be better suited to identify extra-adrenal tumors, invasion into the spinal canal, and involvement of major vessels (2, 9, 11, 65). Children may not tolerate MRI as well as adults and may require sedation, so clinicians should be mindful regarding the choice of sedative so as not to cause precipitous catecholamine release; medications such as opioids (except for fentanyl) can cause histamine release, which may stimulate catecholamine secretion (66).

Anatomical imaging is used for tumor localization and detection of metastases after positive biochemical testing but also for screening, as non-functional tumors may be negative on biochemical evaluation, such as in patients with SDHx mutations or those with parasympathetic PGLs arising in the head and neck (48). For follow-up imaging in the setting of metastases, CT and MRI can be complementary, as CT is better for visualizing lung lesions, and MRI is better for liver lesions (26).

Functional imaging provides insight into the molecular features of PPGLs by use of different radiopharmaceuticals, which is especially informative in metastatic disease. “Theranostics” refers to this unification of therapeutic and diagnostic utility with the same radiopharmaceutical platform, for example using ¹²³I-labeled metaiodobenzylguanidine (MIBG) scintigraphy to identify PPGLs which can then be treated by ¹³¹I-MIBG (67). MIBG is an analog of norepinephrine and enters tumor cells via the norepinephrine transporter (42, 68). Medications such as over-the-counter cough and cold treatments, calcium channel blockers, labetalol, and tricyclic antidepressants can interfere with MIBG uptake and should be avoided before treatment. Special preparation is also required to reduce accumulation of radioiodine in the thyroid (68). While highly specific, MIBG may not be as sensitive as other functional imaging modalities, particularly in hereditary PPGL, and especially in those with SDHx mutations. Additionally, sensitivity is higher for PCC than PGL (88% vs 67%, respectively), and in metastatic PPGL, per-lesion sensitivity is < 60% (42). Despite these limitations in the pediatric population with metastatic PPGL, MIBG may still have a useful theranostic role in patients where ¹²³I-MIBG avidity is observed in all lesions.

The other relevant application of the theranostic paradigm is for tumors expressing somatostatin receptors – especially type 2 (SSTR2) – that bind to somatostatin analogs (SSAs) linked to a chelator (DOTA) for the radionuclide. Typically, ⁶⁸Ga- or ⁶⁴Cu-DOTA-SSA is used for imaging, and ⁹⁰Y- and ¹⁷⁷Lu-DOTA-SSAs deliver the radiation dose to the sites of tumors, with the latter referred to as peptide receptor radionuclide therapy (PRRT) (42, 69). ⁶⁸Ga-DOTATATE is commonly used since DOTATATE binds preferentially to SSTR2 over other somatostatin receptors, as opposed to DOTATOC or DOTANOC (42). ⁶⁸Ga-DOTATATE is particularly useful for cluster 1A mutations (especially SDHx), HNPGLs, metastatic PPGL, and pediatric PPGL (42). Among nine pediatric patients with SDHx mutations, both ¹⁸F-FDG and ⁶⁸Ga-DOTATATE detected lesions in all of the patients but ⁶⁸Ga-DOTATATE had superior sensitivity on a per-lesion basis (94% compared to 79% on ¹⁸F-FDG) and was also more sensitive than anatomic imaging by CT or MRI with contrast enhancement (74%). ⁶⁸Ga-DOTATATE was especially superior to CT/MRI for mediastinal lesion detection and superior to ¹⁸F-FDG for detection of adrenal and liver lesions, though ¹⁸F-FDG and CT/MRI outperformed ⁶⁸Ga-DOTATATE for detection of other abdominal lesions ( Figure 2 ) (70). These data offer a compelling case for performing ⁶⁸Ga-DOTATATE PET/CT or ⁶⁸Ga-DOTATATE PET/MRI with contrast enhancement to detect such abdominal lesions. Currently both functional and anatomic imaging are required for staging and for assessing treatment response in pediatric metastatic PPGL patients and, therefore, simultaneous PET/MRI may be considered in this cohort due to decreased radiation exposure, fewer instances requiring sedation or general anesthesia, fewer appointments, and simultaneous imaging with two advanced diagnostic imaging techniques (whole-body PET and MRI) (71). Recommended imaging modalities for different types of PPGL based on genetic and clinical features are summarized in Table 3 . A 2015 meta-analysis by Han et al. pooled patients of unknown genetic background, finding a significantly superior detection rate of ⁶⁸Ga-DOTA-conjugated somatostatin receptor-targeting peptide (⁶⁸Ga-DOTA-SST) PET (93%) as compared to other functional imaging modalities (72). Therefore, in cases where a genetic etiology is not known, ⁶⁸Ga-DOTATATE PET may be performed.

¹⁸F-fluorodihydroxyphenylalanine (¹⁸F-FDOPA) enters cells via the large neutral amino acid transporter (LAT-1) and enters into the catecholamine synthesis pathway (42). Preparation for ¹⁸F-FDOPA PET/CT involves fasting and administration of carbidopa to block decarboxylation of DOPA to dopamine, improving uptake in target tissue. There are no known interfering medications (42). ¹⁸F-FDOPA PET/CT is particularly helpful for imaging PPGLs in clusters 1B (VHL and PZS) and 2 (MEN-2 and NF-1) and has an advantage over MIBG in that normal adrenal tissue has lower uptake, increasing the sensitivity to detect nonmetastatic PCC (94% in patients of known genetic background and up to 100% in patients with apparently sporadic nonmetastatic PCC); indeed, the detection of metastatic PPGL may also vary on the basis of the genetic background with greater sensitivity in non-SDHx PPGL (42, 73).

¹⁸F-FDG is a radiolabeled form of glucose and is taken up by glucose transporters (particularly GLUT-1) and enters into the glycolytic pathway (42). Preparation for ¹⁸F-FDG involves fasting, controlling hyperglycemia, and avoidance of ambient cold temperature to prevent artifacts from glucose metabolism in thermogenic brown fat (42, 74). The sensitivity of ¹⁸F-FDG in detection of metastatic PPGL is superior to MIBG (usually > 80%), particularly in SDHx patients, with sensitivity in metastatic disease for SDHx patients ranging from 83-92% compared to 62% for non-SDHx (42). Furthermore, while there are no studies of direct comparisons, GIST appears to be better appreciated on ¹⁸F-FDG compared to ⁶⁸Ga-DOTATATE, and therefore, some SDHx patients would benefit from ¹⁸F-FDG in addition to ⁶⁸Ga-DOTATATE PET/CT scan when GIST is suspected (75, 76).

After biochemical studies and tumor localization by imaging, definitive treatment of PPGL should be addressed. In addition to surgery of the primary tumor, when appropriate, patients with metastatic disease require other treatment modalities to address disease at sites of metastasis, such as chemotherapy, radiotherapy, ablative therapy, or other targeted therapies, such as somatostatin analogs or small molecule inhibitors (21). Clinical trials for pediatric patients with advanced/metastatic PPGL that are recruiting, active, or approved are listed in Table 4 .

The first medication clinicians should consider in pediatric PPGL is an α₁-adrenergic antagonist, such as the non-competitive long-acting phenoxybenzamine or competitive short-acting drugs, typically doxazosin. Alpha-adrenergic antagonists help to reduce the symptoms associated with catecholamine excess, improve hypertension, and prepare patients for therapeutic interventions, therefore α-adrenergic antagonists should be started in all PPGL patients (1, 26). Blood pressure and heart rate should be monitored, and β-adrenergic blockade should be considered in patients who have persistent tachycardia but only after starting α-adrenergic blockade for at least 2 to 3 days to avoid the effect of unopposed α-adrenergic stimulation (16). To prevent complications related to catecholamine release during treatment, patients should start α-adrenergic blockade 7 to 14 days before surgery or before non-surgical treatments, such as chemotherapy or radiotherapy, as these may all result in catastrophic surges of stored catecholamines (26). Competitive inhibition of tyrosine hydroxylase by α-methyl-L-tyrosine (metyrosine, or Demser^®) results in decreased synthesis of all downstream catecholamines and can be helpful for patients with highly elevated levels of catecholamines to improve hemodynamic stability before and after surgery but must be used cautiously as it is contraindicated in patients with severe depression and suicidal ideation (17, 26). Calcium channel blockers, especially amlodipine, can also be helpful in controlling hypertension in pediatric patients with PPGL (77).

Surgical interventions should be considered even for patients with metastatic disease, to reduce the tumor burden contributing to catecholamine excess and to improve radiopharmaceutical uptake in remaining metastatic lesions (26). Genetic testing preoperatively can be helpful to inform the surgical approach, as cortical-sparing adrenalectomy is preferred in younger patients to reduce the risk of adrenal insufficiency (and requirement for exogenous steroids) and does not seem to increase the risk of recurrence for patients with mutations in VHL, RET, or NF1, but in patients with higher risk for metastatic disease, such as those with SDHB mutations, more extensive resection may be needed (21).

The approach to the patient with metastatic disease depends on the extent of disease and the rate of progression. For patients with very extensive disease or rapid progression, chemotherapy should be tried first, with first-line therapy consisting of cyclophosphamide, vincristine, and dacarbazine (CVD); temozolomide as monotherapy may also be considered as an alternative or a second-line agent in these cases (78). The impact of SDH deficiency on chemoresistance, as previously mentioned, suggests that PARP inhibitors such as olaparib may be useful in combination with temozolomide, and a clinical trial is currently recruiting to evaluate this combination in adults (NCT04394858). Hypoxia-inducible factors increase expression of growth factors such as VEGF, platelet-derived growth factor (PDGF), and transforming growth factor α (TGFα), which bind to their respective receptor tyrosine kinases (53). Therefore, additional therapies to consider include tyrosine kinase inhibitors with anti-VEGF activity, such as sunitinib (53). As a last resort, additional approaches to consider include immunotherapy with checkpoint inhibitors (e.g., pembrolizumab or combination nivolumab-ipilimumab) in patients with cluster 1 mutations (consequent to impaired immune recognition related to pseudohypoxia) or mTORC1 inhibitors (e.g., everolimus) in patients with cluster 2 mutations (mTORC1 regulates cellular activity downstream of kinase signaling pathways), but additional studies are needed in children with metastatic PPGL (78).

For patients with less rapid progression or smaller tumor burden, systemic radiotherapy can be particularly useful as part of a personalized approach when lesions have avidity on imaging scans for the corresponding diagnostic radionuclides. For patients with DOTA-SSA-avid lesions, as is often seen with cluster 1 mutations, PRRT with ¹⁷⁷Lu-DOTA-SSA (“hot SSA”) or SSAs that do not carry radiopharmaceuticals (referred to as “cold SSAs”) may also be helpful, especially for SDHx patients with DOTATATE-avid lesions, and additional clinical trials are evaluating these in children ( Table 4 ). Conventional or high-specific activity (HSA) ¹³¹I-MIBG (referred to as Ultratrace™, or Azedra^®) may be particularly useful for patients with cluster 2 mutations who are more likely to have adrenal disease but requires positive findings on diagnostic ¹²³I-MIBG scan before use (69). When considering systemic radiotherapy, clinicians should ideally perform both ¹²³I-MIBG and ⁶⁸Ga-DOTATATE scans to determine which radiopharmaceutical best shows all the metastatic lesions in order to guide which radiotherapy should be used to treat the patient (79).

Locoregional approaches may be helpful for certain aspects of metastatic disease. External radiation therapy can be useful for head and neck primary disease due to a more favorable risk profile as compared to surgical intervention and for rapidly-growing metastatic tumors, such as in bone, to provide symptomatic relief (1, 26, 78). Treatment of a primary HNPGL by external beam radiation therapy, therefore, may be helpful for those with SDHD mutations. Bone metastases are also treated with bisphosphonates or denosumab (3, 26). Interventional procedures such as radiofrequency ablation, cryoablation, or ethanol injection can be used in cases for treatment of a single metastatic lesion or oligo-metastases (3, 26). Furthermore, patients with inoperable primary tumors can be considered for potential treatment with a cold-SSA if lesions are ⁶⁸Ga-DOTATATE-avid, as reported in an adult SDHB patient with pterygopalatine fossa PGL, who was successfully controlled for 36 months with octreotide (80).

Belzutifan is one of the newer targeted therapies that specifically inhibits HIF-2α and has been approved to treat VHL-associated tumors (81). Given the central role that HIF-2α plays in hypoxia signaling, this therapy could be incredibly impactful for patients with cluster 1 mutations. A recent case report demonstrated the efficacy of belzutifan in treating multiple PGLs in an adolescent patient with PZS, with improvement in biochemical (normetanephrine and chromogranin A) and hematologic (erythropoietin and hemoglobin) markers and reduced tumor size on imaging. Belzutifan was well-tolerated in this patient with minimal side effects, most notably, anemia that did not require transfusion (81).

While many studies examining prognosis, disease progression, and outcomes in PPGL are focused on adults or include both adults and children, it seems that the most important prognostic factor for disease-free survival (DFS) is extent of surgical resection of the primary tumor (for pediatric PPGL: 45.6% for complete resection vs. 24.1% for incomplete resection, p < 0.001) (7, 21, 30). Redlich et al. described other significant differences in 10-year DFS in pediatric patients with PPGL, including the presence of metastatic disease (29.6% vs. 43.5% in those without metastatic disease, p = 0.014), and PGL (36.6% vs. 47.8% in PCC, p = 0.039), whereas the presence of an SDHB mutation was associated with lower DFS but did not reach statistical significance (45.1% in non- SDHB vs. 24.4% in SDHB, p = 0.063) (7). Outcomes in pediatric patients with SDHB mutations have been well-described by Jochmanova et al., and specific outcomes are also discussed above (23).

This review highlights important features to consider in pediatric patients with PPGL, with an emphasis on metastatic disease. Pediatric PPGL is overwhelmingly due to an underlying genetic predisposition, especially from cluster 1 genes associated with pseudohypoxia. The central role of hypoxia signaling by HIF-2α in these patients results in a pro-metastatic phenotype and disruption of normal development of chromaffin cell precursors that can result in increased susceptibility to PPGL in multiple tissues. Indeed, the biochemical and secretory features could be described as more primitive or underdeveloped in pediatric patients with cluster 1 mutations predisposing them to PPGL. All pediatric PPGL patients require lifelong surveillance and monitoring, even those with cluster 2 mutations who may not present with PPGL until their teenage years. To the extent possible, these rare and challenging patients should be managed at centers of expertise with multidisciplinary teams that can address the different aspects of care.

Rather than viewing the pediatric patient with a genetic mutation as having an immutable risk factor for PPGL, the knowledge of this risk can guide clinicians to develop an appropriate screening and management plan with their patients and families, especially since the penetrance of PPGL is incomplete for all of the susceptibility genes (26). Genetic counseling can play an important supportive role in addition to contributing valuable information about risk of disease. Reducing anxiety by dispelling misconceptions about genetic risks as well as emphasizing preventive strategies and forming a screening plan to help reduce anxiety and empower patients and families is a key part of the approach to genetic counseling as shown in a study of patients with SDHx mutations. Out of 164 patients, only 2 (1.2%) had increased anxiety in response to genetic testing results, and only 1 (0.6%) did not proceed with any preventive measures nor any screening for her positive children (82).

Though not as prevalent a cause of metastatic disease as cluster 1 mutations, additional studies of targeted therapies for patients with cluster 2 mutations are needed in the pediatric population as our increased understanding of cellular pathways has generated specific therapeutic targets. Mweempwa et al. demonstrated efficacy of a selective RET inhibitor (selpercatinib) in metastatic PCC due to an activating gene fusion of RET-SEPTIN9 in an adult with recurrent PCC with biopsy-confirmed metastatic disease to the lungs and liver with negative urinary metanephrines but highly elevated chromogranin A (33,710 µg/L). After 12 weeks of therapy with selpercatinib, chromogranin A had decreased to 598 µg/L, and pulmonary and hepatic tumors demonstrated reduction in size on CT (46% in the sum of diameters of lesions), raising the possibility that this targeted therapy may also be useful for pediatric patients with activating RET mutations; however additional studies are needed (83). Similarly, targeted therapy directed by the genetic background may also inform the approach to patients with NF-1. Selumetinib selectively inhibits MEK in the RAS-MAPK pathway, and in children with inoperable plexiform neurofibromas, treatment with selumetinib resulted in a reduction in tumor volume (median reduction of 27.9%) and progression-free survival of 84% (compared to 15% in natural history controls) at 3 years from treatment initiation. MEK inhibitors have also shown responses in other NF1-related tumors (e.g., optic pathway glioma), suggesting that PPGLs driven by NF1 mutations might also be considered for this targeted therapy (84).

PPGLs have been described as the tumors that are the most highly genetically determined of any human tumor, and in children, around 80% have a germline predisposition. When accounting for those with additional somatic changes (e.g., EPAS1 gain-of-function mutations), an even higher proportion is realized. Germline genetic testing to determine an underlying cause of PPGL is mandatory, especially in children with metastatic disease, but somatic tumor genetic testing has clear benefits as well and may identify new genes and targetable mechanisms that can personalize the care of these vulnerable patients.

MK drafted and revised the initial manuscript, figures, and tables. MN aided in conceptualizing and contributing to the manuscript. He also revised the manuscript. AJ aided in conceptualizing, contributing, and revising the manuscript. KP conceived, conceptualized, critically reviewed the manuscript for intellectual content, and aided in revision of the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported, by the Intramural Research Program of the National Institutes of Health, Eunice Kennedy Shriver National Institute of Child Health and Human Development.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.