Universal Germline Panel Testing for Individuals With Pheochromocytoma and Paraganglioma Produces High Diagnostic Yield.

BACKGROUND: Practice guidelines to identify individuals with hereditary pheochromocytomas and paragangliomas (PPGLs) advocate for sequential gene testing strategy guided by specific clinical features and predate the routine use of multigene panel testing (MGPT).
OBJECTIVE: To describe results of MGPT for hereditary PPGL in a clinically and ancestrally diverse cohort.
SETTING: Commercial laboratory based in the United States.
METHODS: Clinical data and test results were retrospectively reviewed in 1727 individuals who had targeted MGPT from August 2013 through December 2019 because of a suspicion of hereditary PPGL.
RESULTS: Overall, 27.5% of individuals had a pathogenic or likely pathogenic variant (PV), 9.0% had a variant of uncertain significance, and 63.1% had a negative result. Most PVs were identified in SDHB (40.4%), followed by SDHD (21.1%), SDHA (10.1%), VHL (7.8%), SDHC (6.7%), RET (3.7%), and MAX (3.6%). PVs in FH, MEN1, NF1, SDHAF2, and TMEM127 collectively accounted for 6.5% of PVs. Clinical predictors of a PV included extra-adrenal location, early age of onset, multiple tumors, and positive family history of PPGL. Individuals with extra-adrenal PGL and a positive family history were the most likely to have a PV (85.9%). Restricting genetic testing to SDHB/C/D misses one-third (32.8%) of individuals with PVs.
CONCLUSION: Our data demonstrate a high diagnostic yield in individuals with and without established risk factors, a low inconclusive result rate, and a substantial contribution to diagnostic yield from rare genes. These findings support universal testing of all individuals with PPGL and the use of concurrent MGPT as the ideal platform.

Pheochromocytomas (PCCs) and paragangliomas (PGLs) (collectively PPGLs) are tumors of the autonomic nervous system derived from the adrenal medulla or the sympathetic or parasympathetic paraganglia, respectively. It has been estimated that 30% to 40% of individuals with PPGL harbor germline pathogenic gene variants (1–3). PPGLs are a genetically heterogeneous entity, with a growing number of genes known to contribute to hereditary predisposition. PPGLs can be associated with the succinate dehydrogenase (SDHx-related) hereditary paraganglioma syndromes (SDHA, SDHB, SDHC, SDHD, SDHAF2), TMEM127- and MAX-associated hereditary pheochromocytoma syndromes, Multiple Endocrine Neoplasia syndrome type 2 (RET), von Hippel-Lindau disease (VHL), and Neurofibromatosis type 1 (NF1). Associations with other genes such as EGLN, EPAS, KIF1B, MDH2, MEN1, and FH have been suggested, which may increase the number of established PPGL genes in the future (4–9). Identifying an underlying pathogenic variant in a patient with PPGL is important because it will further allow for the recommendation of surveillance protocols for associated tumors, which vary significantly between causative genes, and for the identification of at-risk family members (family cascade testing).

Much of what is known about the germline genetic causes of PPGLs comes from studies with inherent ascertainment and referral bias. Study cohorts often exclude individuals with features beyond PPGL that are suggestive of syndromic etiology, and subsequent genetic testing often excludes syndromic genes such as VHL, RET, FH, MEN1, and NF1, leading to underestimation of the contribution of these genes in reports of hereditary PPGLs (10–12). Studies may also enrich their sample using inclusion criteria such as early age of diagnosis, malignant tumors, multiple tumors, tumor location, and/or family history (13, 14). Practice guidelines advocating for a sequential gene testing algorithm driven by presence or absence of specific clinical features are based in part on the findings of these studies and predate the routine use of multigene panel testing (MGPT) (15). Here, we describe the results of MGPT for hereditary PPGL in a cohort of clinically and ancestrally diverse individuals, representative of current clinical genetic testing referrals, in an effort to better understand the contribution of germline predisposition to these rare neoplasms.

Methods

Data from individuals who underwent MGPT targeted for hereditary PPGL (PGLNext) at a commercial diagnostic laboratory (Ambry Genetics, Aliso Viejo, California) between August 2013 and December 2019 were retrospectively reviewed. Demographic and clinical information including sex, ethnicity, age, ordering provider specialty, tumor type, age of diagnosis, and family history were collected from a test requisition form and supporting clinical documents provided by the ordering clinician. This information was curated into a clinical database with corresponding genetic results for each individual. When clinical documents are provided in addition to the test requisition form, personal and family history are obtained first from the pathology report, followed by the clinic note, pedigree, and the test requisition form. Supporting clinical documents were provided in addition to the test requisition form in 43.3% (n = 780) of cases. To estimate the accuracy of clinical information provided on test requisition forms when it was the sole document provided, we compared tumor location (PGL vs PCC), age of PPGL diagnosis, and family history (± PPGL) in 100 patients who had a personal history of PPGL and had both a clinic note and completed test requisition form submitted.

Individuals underwent MGPT of 10 to 12 genes depending on test order date. From August 2013 through May 2015, 560 individuals had MGPT that included 10 genes (NF1, MAX, SDHA/B/C/D/AF2, RET, TMEM127, and VHL), and from May 2015 through December 2019, 1167 individuals had panel testing of 12 genes because of the addition of MEN1 and FH.

Genomic DNA was isolated from a blood, saliva, or fibroblast sample and then quantified (Nanodrop; Thermo Scientific, Pittsburgh, PA; or Infinite F200; Tecan, San Jose, CA). Sequence enrichment was performed by incorporating genomic DNA onto a microfluidics chip or into microdroplets with primer pairs or by a bait-capture methodology using long biotinylated oligonucleotide probes (RainDance Technologies, Billerica, MA; or Integrated DNA Technologies, San Diego, CA). This procedure was then followed by next-generation sequencing analysis (Illumina, San Diego, CA) of all coding exons in addition to least 5 bases into the 5′ and 3′ ends of all introns. Sanger sequencing was performed to confirm variant calls in regions missing or with insufficient read depth coverage, variants in regions complicated by pseudogene interference, and potentially homozygous variants. Gross deletion/duplication analysis for all 12 genes was also performed using a custom pipeline based on read depth from next-generation sequencing (NGS) data and/or targeted chromosomal microarray with confirmatory MLPA when applicable. Sequence analysis was based on the following NCBI reference sequences: FH: NM_000143.3; MAX: NM_002382.3; MEN1: NM_130799.2; NF1: NM_000267.3; RET: NM_020975.4; SDHA: NM_004168.2; SDHAF2: NM_017841.2; SDHB: NM_003000.2; SDHC: NM_003001.3; SDHD: NM_003002.2; TMEM127: NM_017849.3; and VHL: NM_000551.3.

Interpretation of sequence variations was performed according to the American College of Medical Genetics and Genomics guidelines (16). Variants were classified as pathogenic, likely pathogenic (together referred to as PV), variant of unknown significance (VUS), likely benign, or benign according to the Ambry 5-tier variant classification protocol (17). PVs were considered positive results. Cases with VUS in the absence of a PV were considered inconclusive. Cases with likely benign or benign findings in the absence of a PV and VUS finding were considered negative. Positive rates based on tumor location, number, age of diagnosis, and family history were compared using χ 2 or Fisher exact test.

Results

A total of 1727 individuals underwent MGPT for PPGL. Demographics and clinical characteristics are displayed in Table 1. Most individuals were female (59.0%) and Caucasian (58.8%); the index patient tested in a family (99.1%) and had a personal history of at least 1 PPGL (77.5%). For the purpose of this study, individuals with tumors that have been implicated in hereditary PPGL such as gastrointestinal stromal tumors (GISTs, n = 25) and renal cell carcinoma (RCC, n = 27), were not included in the “affected” group unless they also had PPGL. Among 274 individuals with clinical history provided and not reported to have PPGL, GIST, and/or RCC, 49 individuals had a first- or second-degree relative with PPGL and 92 additional individuals had clinical features associated with at least 1 of the genes on the panel (eg, medullary thyroid cancer [RET], neuroendocrine tumor [MEN1], schwannomas [NF1]), and 6 individuals were undergoing workup for possible PPGL. Indication for testing was not immediately apparent in 127 individuals.

Table 1.

Cohort description

Characteristic	Overall	Positive result (% of positive)	Negative result (% of negative)	Inconclusive result (% of inconclusive)
	n = 1727	n = 475	n = 1097	n = 155
Sex
Female	1019 (59.0%)	243 (51.2%)	680 (62.4%)	95 (61.2%)
Male	708 (41.0%)	232 (48.8%)	410 (37.6%)	60 (38.7%)
Ethnicity
African American	194 (11.2%)	35 (7.4%)	128 (11.7%)	30 (19.4%)
Ashkenazi Jewish	34 (2.0%)	4 (0.8%)	28 (2.6%)	2 (1.3%)
Asian	80 (4.6%)	10 (2.1%)	55 (5.1%)	15 (9.7%)
Caucasian	1016 (58.8%)	293 (61.7%)	649 (59.5%	69 (44.5%)
Hispanic	97 (5.6%)	33 (6.9%)	55 (5.1%)	9 (5.8%)
Other/unknown	306 (17.7%)	100 (21.1%)	175 (16.1%)	30 (19.4%)
Personal history
Affected	1338 (77.5%)	400 (84.2%)	810 (74.3%)	122 (78.7%)
Affected at least 1 PGL	649 (37.6%)	284 (59.8%)	693 (63.6%)	60 (38.7%)
Affected PCC only	689 (39.9%)	116 (24.4%)	117 (10.7%)	62 (40.0%)
Unaffected	127 (7.3%)	27 (5.7%)	92 (8.4%)	8 (5.2%)
No PPGL + other tumor	156 (9.0%)	27 (5.7%)	113 (1.4%)	15 (9.7%)
Not provided	106 (6.1%)	21 (4.4%)	75 (6.9%)	10 (6.5%)
Average age
All	46.8 (SD 17.0)	39.1 (SD 16.5)	50.0 (SD 16.3)	47.8 (SD 15.2)
Affected	46.9 (SD 17.0)	38.8 (SD 16.4)	50.5 (SD 16.0)	48.1 (SD 15.5)
Unaffected	46.7 (SD 17.1)	44.2 (SD 14.7)	48.6 (SD 17.2)	46.6 (SD 14.2)
Age at PPGL diagnosis	41.8 (SD 17.1)	32.2 (SD 15.0)	46.3 (16.4)	44.0 (SD 15.6)
Ordering provider specialty
Oncologya	739 (42.8%)
Geneticsb	249 (14.4%)
Surgery/surgical oncology	95 (5.5%)
General/family practice	46 (2.7%)
Endocrinology	43 (2.5%)
Pediatrics/pediatric oncology	43 (2.5%)
Other	48 (2.8%)
Not provided	464 (26.9%)

Abbreviations: PGL, paraganglioma; PPGL, pheochromocytoma and paraganglioma.

Includes medical oncology, hematology/oncology, gynecologic oncology, radiation oncology.

Includes medical genetics, clinical genetics, genetic counseling.

The average age at testing of all individuals at the time of testing was 46.8 years, which did not differ between affected and unaffected individuals (46.9 years and 46.7 years, respectively; P = 0.91). Oncology and genetics providers were responsible for most orders (78.2%; n = 988) among cases for which ordering provider was specified. In our evaluation of quality provided in the test requisition form, we found that tumor location, age at PPGL diagnosis, and family history of PPGL was provided with 97%, 99%, and 95% concordance between the test requisition and clinic note. Therefore, we estimate a high level of accuracy in clinical information submitted via test requisition form.

Overall, 27.5% of tested individuals had a positive result, 9.0% had an inconclusive result, and 63.5% had a negative result. Positive and inconclusive rate by gene is provided in Supplemental Figure 1, available on figshare.com (18). Positive and inconclusive rates were influenced by ethnicity, with positive results highest among Hispanics and Caucasians (34.0% and 28.8%, respectively) and inconclusive results highest among Asians and African Americans (18.8% and 15.5%, respectively) (Supplemental Figure 218). The positive rate also varied by clinical history, ranging from 19.3% in unaffected individuals to 29.9% in individuals affected with PPGL (Fig. 1). Pathogenic/likely pathogenic variants in SDHB accounted for the largest proportion of positive results (40.4% of positives), followed by SDHD (21.1%), SDHA (10.1%), VHL (7.8%), SDHC (6.7%), RET (3.7%), and MAX (3.6%). Pathogenic/likely pathogenic variants in FH, MEN1, NF1, SDHAF2, and TMEM127 collectively accounted for 6.5% of positive results. The observation of increased diagnostic yield by gene inclusion is illustrated in Fig. 2. Restricting genetic testing of hereditary PPGL to SDHB/C/D would lead to missed positives in more than one-third (32.8%) of individuals. The increase in yield is especially impactful in individuals diagnosed with PCC only, where 38.2% of positive results were in SDHB/C/D, 20.9% in VHL, 22.7% in MAX/TMEM127, and 18.1% in the remaining genes.

Figure 1.

Overall test results for all individuals tested. Percentage of positive, inconclusive, and negative results in all individuals, affected individuals with PPGL, and individuals, and unaffected individuals without PPGL.

Figure 2.

Increased yield by gene and personal history. Proportion of PVs in PPGL predisposition genes in all individuals with positive results regardless of personal history, individuals affected with either PGL or PCC, individuals affected with at least 1 PGL (with or without PCC), individuals affected with PGL only (without PCC), and individuals affected with PCC only (without PGL).

When using clinical characteristics for results stratification, extra-adrenal location, age of diagnosis before 45 years, multiple tumors, and positive family history were associated with increased likelihood of positive results (Table 2). Affected individuals with a family history of PPGL were the most likely to have a PV (70.6% of individuals with PCC + family history; 85.9% of individuals with PGL + family history). The positive rate in nearly all clinical subgroups even without predictors of a PV remained > 10%, including individuals with a single tumor (PCC = 16.7%; PGL = 46.7%) and those without a family history (PCC and negative family history = 15.8%; PGL and negative family history = 43.7%). Individuals with only PCCs diagnosed ≥ 45 years were the least likely to have a positive result (6.9%). Nearly one-half (46.5%) of individuals with head and neck PGLs had a PV, and 34.0% of those with abdominal or thoracic PGLs had a PV (Supplemental Table 1)(18).

Table 2.

Clinical predictors of positive results

Clinical feature	Total	Number positive (%)	OR	95% CI	P value
At least 1 PGL ± PCC	588	284 (48.3%)	4.1	3.1-5.2	<0.001
PCC Only	629	118 (18.8%)
PCC only Dx < 45 y	341	96 (28.2%)	5.3	3.0-9.3	<0.001
PCC only Dx ≥ 45 y	232	16 (6.9%)
Multiple PCC	30	18 (60.0%)	7.5	3.5-16.0	<0.001
Single PCC	599	100 (16.7%)
PCC + Fam Hx PPGL	34	24 (70.6%)	12.8	5.9-27.7	<0.001
PCC – Fam Hx PPGL	596	94 (15.8%)
PGL Dx < 45 y	315	203 (64.4%)	5.5	3.8-8.0	<0.001
PGL Dx ≥ 45 y	233	58 (24.9%)
PGL + PCC or PGL	57	36 (63.2%)	2.0	1.1-3.6	0.02
Single PGL	531	248 (46.7%)
PGL + Fam Hx PPGL	64	55 (85.9%)	7.9	3.8-16.3	<0.001
PGL—Fam Hx PPGL	524	229 (43.7%)

Abbreviations: Dx, diagnosis; Fam Hx, family history (1st-, 2nd-, 3rd-degree relative); PCC, pheochromocytoma; PGL, paraganglioma; PPGL, pheochromocytoma and paraganglioma.

Clinical phenotype in relation to the underlying germline variants were largely in accordance with prior studies. Clinical details for individuals with pathogenic variants in genes associated with syndromic features (FH, MEN1, NF1, RET, and VHL), and genes with limited available published phenotype data (FH, MAX, MEN1, SDHA, SDHAF2, and TMEM127) are provided in Tables 3 and 4, respectively. Among 65 individuals with a positive finding in a gene typically associated with syndromic features for whom clinical information was provided, fewer than one-half (41.5%) had syndromic features beyond PPGL that were consistent with their molecular diagnosis (Table 3). Of note, 76.9% of these individuals had supportive clinical documents provided in addition to the test requisition form. Individuals with FH, MEN1, and TMEM127 PVs exclusively had adrenal PCCs; individuals with SDHAF2 PVs exclusively had PGLs; and individuals with MAX and SDHA had a combination of PCC and PGL (Table 4). Interestingly, 3 SDHC PV carriers had tumors below the diaphragm, which is a less appreciated phenotype (19). Further details regarding tumor spectrum by gene is provided in Supplemental Table 118.

Table 3.

Syndromic positives with characteristic features other than PPGL

Gene	Total positive	No. (%) with features outside PPGL
FH	5	1 (20)
MEN1	4	4 (100)
NF1	7	5 (71.4)
RET	18	7 (38.9)
VHL	37	10 (27.0)
All	65	27 (41.5)

Abbreviation: PPGL, pheochromocytoma and paraganglioma.

Table 4.

Further clinical details for select genes

Gene	PCC without PGL (bilateral)	PGL ± PCC (PGL location)	Unaffected
FH	4 (0)	0	1
MAX	14 (6)	1 (1NP)	2
MEN1	1 (0)	0	0
SDHA	6 (0)	35 (10 Ab, 4 HN, 2 Th, 19 NP)	6
SDHAF2	0	3 (2 HN, 1 NP)	1
TMEM127	11 (1)	0	0

Abbreviations: Ab, abdominal; HN, head and neck; Th, thoracic; NP, not provided; PCC, pheochromocytoma; PGL, paraganglioma.

Eleven of the 25 (44.0%%) individuals with GIST had a pathogenic variant (6 SDHA, 3 SDHB, 2 NF1), 4 of whom also had a personal history of PPGL. Of the 7 individuals with isolated GIST and a pathogenic variant (5 SDHA, 2 SDHB), abnormal SDHA or SDHB immunohistochemistry staining was noted. Nine of the 27 individuals with RCC had a pathogenic variant (5 SDHB, 1 SDHA, 1 SDHC, 1 FH, 1 VHL); however, most of these individuals also had a personal history of PPGL (77.8%; 7 of 9 individuals).

Discussion

Here, we report the outcomes of genetic testing for hereditary PPGL using MGPT in a commercial diagnostic laboratory cohort. Our findings are consistent with previously reported trends, such as the overall positive rate among affected individuals and the preponderance of SDHx pathogenic variants. We also confirmed extra-adrenal location, early age of onset, multiple tumors, and positive family history as predictors of a positive result, as previously reported (20, 21). However, the positive rate is > 10% in nearly all subgroups, even without these risk factors. For colorectal and breast cancer, a priori risks of 2.5% and 5% have been recommended as a threshold for performing germline genetic testing, respectively (22, 23). Therefore, although these mutation likelihood estimates may be helpful for risk assessment during pretest counseling, especially in regions where resources for genetic testing are limited, the observed positive rate in all patients with PPGL justifies testing.

The high proportion of heritable PPGL is reflected in the overall positive rate of 27.5% and relatively low inconclusive rate of 9.0%. A major limitation with concurrent MGPT is the high inconclusive rate compared with positive rate. However, when compared with published positive and inconclusive rates observed in MGPTs targeted for other cancer indications, MGPT for PPGL performs especially well. For example, it has been previously reported that a breast cancer-focused panel yielded a 7.4% positive and 19.8% inconclusive rate; a renal cancer-focused panel yielded a 6.1% positive and 18.4% inconclusive rate; and a pan-cancer panel yielded a 9.6% positive and 23.5% inconclusive rate (24, 25).

With regard to demographic factors, our data echo the underrepresentation of minority groups receiving genetic testing reported throughout the field. This poses the obvious concerns of limited access to increased screening and risk reducing interventions, but also hinders the ability to accurately assess alterations in minority groups, as demonstrated by the inconclusive rates by ethnicity (Supplemental Figure 2)(18). Demographics detailing the varieties of ordering provider specialty reflect the multidisciplinary care involved in the management of PPGL. For orders that had specialty provided, only 2.5% were received from endocrinologists compared with 42.8% from oncology providers and 14.4%% from genetics practice. Further study on factors influencing ordering behaviors may identify opportunities to increase the number of patients identified by endocrinologists, who are able to be involved in the continuum of care throughout life.

Clinical features of 89 individuals in this study with pathogenic variants in FH, MAX, MEN1, SDHA, SDHAF2, and TMEM127 described here add to the limited body of literature pertaining to phenotype associated with these genes. Observations of tumor types described previously in these genes were largely reproduced in the current study. Both individuals with a pathogenic variant in SDHAF2 that had tumor location provided had head-and-neck PGL (26) and all individuals with positive results in TMEM127 had exclusively adrenal PCC (27). Nearly all individuals with a positive result in MAX were affected with PCC (28); however, 1 individual had a composite paraganglioma-ganglioneuroma, which, to our knowledge, has not been previously reported in MAX carriers. This individual underwent testing because of his personal history, in addition to clinical suspicion of neurofibromatosis type 1 in his sibling, highlighting the genetic heterogeneity of PPGL and the importance of molecular confirmation of this disease. All affected individuals with a pathogenic variant in FH had a PCC, but only 1 had a personal or family history of features consistent with hereditary leiomyomatosis and RCC. In contrast, all individuals with MEN1 pathogenic variants had a clinical diagnosis of multiple neoplasia syndrome type 1 (MEN1), and a MGPT was presumably ordered to rule out pathogenic variants in other genes associated with PPGL as well. It is not yet known what proportion of individuals with hereditary leiomyomatosis and RCC or MEN1 present with PPGL as the primary and/or only feature of the syndrome; however, our findings suggest this would be a rare occurrence.

Further study is needed on phenotypic and molecular aspects among individuals with other tumors in the hereditary PPGL spectrum, such as GIST and RCC. Although the numbers in this study are small, results suggest individuals with GIST and RCC are most likely to have pathogenic variants in SDHA and SDHB, respectively. However, these findings may be confounded by the fact that many individuals with tumors beyond PPGL pursue testing for other predisposition genes rather than a targeted PPGL panel and therefore may not be well represented in our cohort.

Several groups have proposed a phenotype driven algorithm for sequential PPGL genetic testing. Clinical features such as tumor location, malignancy, and specific biochemical expression often dictate the sequence of testing (13, 29, 30). These tiered approaches focus on SDHB, SDHC, and SDHD, with consideration for VHL, MAX, and RET testing in certain circumstances. SDHA, SDHAF2, NF1, FH, and MEN1 are excluded from some testing algorithms altogether, and in others, testing for syndromic genes is only considered in individuals with suggestive features outside of PPGL. However, as shown in Fig. 2, diagnostic yield improves incrementally with the addition of genes considered to be rare or most often associated with syndromic features. Bausch et al reported that 6% of PPGL patients were found to carry a pathogenic alteration in the MAX, SDHAF2, or TMEM127 genes, similar to our finding of an 6.9% positive rate in these genes combined (31). Furthermore, pathogenic variants in genes associated with syndromic conditions make up 15.0% of positives and fewer than one-half of individuals with pathogenic alterations in syndromic genes presented with personal or family history other than PPGL (Table 4), though this may be due in part to incomplete clinical history collection and/or omitted clinical data provided to our laboratory.

In addition, the use of malignancy as a requisite for testing can be counterproductive when results of genetic testing are used to determine the risk for malignancy, as in the case of individuals found to carry SDHB or MAX pathogenic variants, and in many cases, malignancy may not be known for years after a diagnosis is made. In addition, evidence continues to emerge regarding the malignant potential for other genes, such as SDHA (32). Furthermore, accurate characterization of biochemical phenotype can be influenced by collection method, coexistence of multiple PPGLs, types of biochemical tests used, and drug interactions (33). In these ways, using clinical features to guide genetic testing decision making may serve as an impediment rather than an aid in identifying individuals with hereditary PPGL. In our sample of positive cases, information regarding malignancy and biochemical expression was only provided in 27.5% and 30.4% of cases, respectively, although we cannot conclude if this is due to unreliability or absence of the information at the time genetic testing was ordered, or if the information was simply omitted from the clinical details provided to our laboratory.

Although the Endocrine Society provides practice guidelines outlining a stepwise testing algorithm, the group acknowledges that broader concurrent panel testing may be more feasible with the adoption of NGS (13). Since then, several groups have proposed concurrent MGPT as the preferred testing method and more data have emerged about the genomic heterogeneity of PPGL (34–36). The NGS in PPGL Study Group developed a consensus statement based on expert opinion regarding appropriate testing platforms, variant interpretation, and genes targeted. Recommendations include use of NGS panel testing options consisting of a basic (11 genes), extended (15 genes), or comprehensive (27 genes) panel (36). Clinical and genetic overlap amongst hereditary PPGL make such recommendations necessary, and advances in genetic testing technology have now made them practical as well. Guidance from expert panels and robust evaluation of the clinical validity of PPGL predisposition genes is essential when designing MGPT to maximize the clinical utility and reduce limitations of such testing.

Genetic testing for hereditary PPGL can inform surveillance and treatment recommendations; therefore, prompt identification of at-risk individuals is imperative. Our data, in which positive rate is consistently high in individuals with or without characterized risk factors, support recent assertions that genetic testing should be performed in all individuals with PPGLs regardless of age of onset, tumor type, functional status, metastatic disease, syndromic features, or family history. Moreover, the high ratio of positive to uncertain results, high proportion of individuals with syndromic positives presenting with isolated PPGL, and substantial contribution to diagnostic yield from rare genes when included in concurrent testing all support the use of concurrent MGPT as the preferred testing method. Considering the technological advances over recent years and the results of this study, concurrent testing of PPGL genes is the most efficient approach to identify underlying germline variants and recognize at-risk individuals.