Abstract
Pheochromocytoma and paragangliomas (PPGLs) are rare neuroendocrine tumors carrying 25–40% pathogenic germline gene variants (PGVs). We evaluated clinical, laboratory, and germline molecular profile of 115 patients with pathologic (14 patients were relatives from 8 different families recruited for genetic survey) confirmed PPGL followed in our institution. Patients with classic MEN2A/MEN2B phenotypes and at-risk relatives underwent direct analysis of RET proto-oncogene, and the remaining had samples submitted to complete next-generation sequencing aiming 23 PPGL-related genes: ATM, ATR, CDKN2A, EGLN1, FH, HRAS, KIF1B, KMT2D, MAX, MDH2, MERTK, MET, NF1, PIK3CA, RET, SDHA, SDHAF2, SDHB, SDHC, SDHD, TMEM127, TP53, and VHL. We also developed a clinical judgment score (CJS) to determine the probability of patients having a potentially hereditary disease. The resulting genetic landscape showed that 67 patients (58.3%) had variants in at least one gene: 34 (50.7%) had exclusively pathogenic or likely pathogenic variants, 13 (19.4%) had pathogenic or likely pathogenic variants and variant of undetermined significance (VUS), and 20 (29.8%) carried only VUS. PGVs were found in RET (n = 18; 38.3%), VHL (n = 10; 21.3%), SDHB and NF1 (n = 8; 17% each), and MAX, SDHD, TMEM127, and TP53 (n = 1; 2.1% each). Direct genetic testing disclosed 91.3% sensitivity, 81.2% specificity, and 76.4% and 93.3% positive predictive value (PPV) and negative predictive values (NPV), respectively. The CJS to identify patients who would not benefit from genetic testing had 75% sensitivity, 96.4% specificity, and 60% and 98.2% PPV and NPV, respectively. In summary, the landscape of PPGL germline gene variants from 115 Brazilian patients resulted in slightly higher prevalent pathogenic and likely pathogenic variants, especially in the RET gene. We suggest a CJS to identify PPGL patients who would not require initial genetic evaluation, improving test specificity and reducing costs.
Introduction
Pheochromocytomas and paragangliomas (PPGLs) are rare neuroendocrine tumors that produce, store, and secrete catecholamines (Sutton et al. 1981, Karagiannis et al. 2007, Neumann et al. 2019, Al Subhi et al. 2022). The outdated general ‘rule of 10%’, which was formerly applied for genetic behavior for PPGL (Sutton et al. 1981, Bravo & Tagle 2003) also used to include inheritance. However, this rule no longer stands. Regarding genetic behavior, 25–40% of PPGL carriers have germline variants (Lenders et al. 2014, Garcia-Carbonero et al. 2021); for paragangliomas (PGLs) alone, there is a 50% chance of finding a pathogenic variant (PV) (Antonio et al. 2020), a rate greater than that for any other human neoplasm (Toledo et al. 2017).
Currently, nearly 70% of PPGLs result from germline or somatic PVs in a single driver gene. To date, there are more than 20 genes related to PPGLs, and more new genes are being discovered (7–10).
The genomic characteristics of PPGLs allow us to allocate them into three clusters according to their transcriptional signature (Buffet et al. 2017, Toledo et al. 2017, Jiang et al. 2020, Garcia-Carbonero et al. 2021, Cascon et al. 2023, Giminez-Roqueplo et al. 2023).
Cluster 1 includes genes related to the Krebs cycle. Characteristic genes are SDHA, SDHAF2, SDHB, SDHC, SDHD, FH,MDH2, GOT2, IDH1, SCLC25A11, EPAS1, and VHL. The presence of germline or somatic PVs in these genes leads directly or indirectly to deregulaton in HIF1a and HIF2a, generating pseudohypoxia and increasing angiogenesis and cell proliferation (Buffet et al. 2017, Toledo et al. 2017, Jiang et al. 2020, Garcia-Carbonero et al. 2021).
Cluster 2 has signature activation of MAP kinase signaling pathways. The following genes stand out: NF1, RET, HRAS, and TMEM127 (9–11).
Cluster 3 (identified by The Cancer Genome Atlas) has WnT pathways based on a transcriptional signature, which also increases the risk of metastatic PPGLs.
The aim of this study was to evaluate the germline molecular profile of a cohort of patients with PPGL from a single center in Sao Paulo, SP, Brazil.
Material and methods
Study design and participants
This prospective and retrospective study included 115 PPGL patients followed at the Endocrine Division, Universidade Federal de Sao Paulo (Unifesp), Brazil, who were evaluated due to the presence of (i) adrenal/extra-adrenal tumors and/or (ii) secondary arterial hypertension and/or (iii) because they had first-degree relatives with PPGLs. Fourteen patients were relatives from eight different families recruited for genetic survey.
Clinical, laboratory, and genetic data were collected between February 2000 and December 2019. The inclusion criterion was pathologic confirmation of PPGL.
The study was approved by the Ethics and Research Committee of Unifesp. All patients or their legal responsible signed an informed written consent form. Patients were distributed into two groups, named as ‘potentially hereditary’ (group 1) and as ‘apparently sporadic’ (group 2). The characteristics that defined the potentially hereditary cases are shown in Table 1. Patients presenting at least one of the characteristics suggestive of hereditary behavior entered group 1 (n = 91 patients) and those who did not present any of those characteristics comprised group 2 (n = 24 patients).
Clinical, laboratory, and imaging features suggestive of germline pathogenic variants.
| Potentially hereditary | |
|---|---|
| Associated diseases | Medullary thyroid carcinoma, primary hyperparathyroidism, neurofibromatosis, middle ear tumors, central nervous system hemangioblastoma, pancreatic neuroendocrine tumor, gastrointestinal stromal tumor (GIST), clear cell renal carcinoma, pituitary adenoma, Hirschsprung’s disease |
| Clinical, laboratory, and imaging features | Skin pigmentation (cafe-au-lait spots), precocious puberty, marfanoid habitus, mucosal neuromas, retinal angiomas, renal and/or pancreatic cysts, polycythemia, intestinal ganglioneuromatosis, megaesophagus, megacolon, and amyloidotic cutaneous lichen |
Additionally, we developed a clinical judgment score (CJS) (Table 2) to determine the probability of a patient having a potentially hereditary disease.
Clinical judgment score.
| Clinical score | Points |
|---|---|
| <20 years old | Mandatory genetic test |
| Any paraganglioma (PGL), regardless of age Classic clinical picture of NF1, VHL, MEN2A, MEN2B, or familial PGL syndrome | |
| 20–45 years of age | 3 |
| Family history of PPGL | 3 |
| Presence of bilateral masses | 2 |
| Presence of metastasis | 1 |
| Total | 9 |
We accessed the Fleury laboratory database of uncharacterized patients who underwent a genetic panel by next-generation sequencing (NGS) for PPGL from January 2018 through July 2022 and collected data from 105 unrelated patients. The Fleury Group operates in the four main regions of Brazil – Northeast, Midwest, Southeast, and South. Patients underwent the exam because they had a PPGL.
Data collection
The following data were collected from each patient: age; sex; family history of PPGL; presence of syndromic presentation; presence of comorbidities; use of antihypertensive medications; measurements of urinary and plasma metanephrines and catecholamines, vanillylmandelic acid (VMA), and chromogranin A; imaging findings of the abdomen, chest, and neck; pathologic results; and genetic testing (germline genetic panel by NGS for PPGL or Sanger genetic screening of the RET proto-oncogene, if the case had a typical MEN2A or 2B presentation). Somatic assessment was not performed at initial assessment and will not be discussed in this paper. The results of the genetic tests were classified as pathogenic, likely pathogenic, or variant of undetermined significance (VUS) according to the American College of Medical Genetics (ACMG) (Richards et al. 2015). Benign findings were not reported.
Genetic testing
Patients with the classic MEN2A or MEN2B phenotypes and their at-risk relatives underwent direct analysis of the RET proto-oncogene at the Laboratory of Molecular and Translational Endocrinology at Unifesp. DNA extraction and PCR amplification of genomic DNA were prepared from blood leukocytes according to standard protocols. Oligonucleotide primers for the amplification of different RET exons were designed at intronic sequences flanking exons 8, 10, 11, 13, 14, 15, and 16. PCRs were performed in a final volume of 25 µL containing 20 mM Tris–HCl (pH 8.4), 50 mM KCl, 1.5 mM MgCl2, 0.2 mM deoxynucleotide triphosphate, 1 U Taq polymerase, and 1 mM specific primers and using 100 or 200 ng of genomic DNA as input. Genomic DNA was denatured for 3 min at 94°C before 35 cycles of 1 min each at 94°C, at 65°C, and at 72°C, followed by 5 min at 72°C in a thermocycler (Corbett Research, Qiagen Ltd, Sydney, Australia). Following PCR, the amplicon sizes were analyzed on a 1.8% agarose gel, and the products were visualized by staining with GelREd® (Uniscience Ltd, Osasco, SP, Brazil). PCR products were purified with a commercial kit (Life Technologies, Inc.) and sequenced using an automated system employing fluorescent dye terminators (ABI Prism 3100, Applied Biosystems) (Lindsey et al. 2012).
Variant analyses were performed based on the medullary thyroid carcinoma guidelines by the American Thyroid Association (Wells et al. 2015, Thomas et al. 2019) and scored according to the evidence evaluated (prediction, literature, and information deposited in Arup, ClinVar, PubMed, and Varsome).
Patients who did not have the classic MEN2A or MEN2B phenotypes had their samples submitted to a genetic panel by NGS at Fleury Group laboratory with the complete sequencing of all coding region and adjacent flanking regions of 23 genes related to PPGL. The analysis included identification of pathogenic genetic variants and copy number variations by NGS.
DNA was extracted from peripheral blood leukocytes using the QIAsymphony DNA Mini Kit according to the manufacturer's recommendations: 200 ng of DNA were used for library preparation following the protocol for Sophia Genetics' Custom Bundle Solution. Capture-based target enrichment was performed on pools of up to eight samples using a custom panel for analysis of full-coding sequencing and flanking splice sites of 23 PPGL-related genes.
Sequencing was performed using a NextSeq® 500/550 Mid Output Kit v2 (300 cycles) on a NextSeq® platform (Illumina). The mapping of reads to the human reference genome GRCh37/hg19 and the calling of variants were performed by a custom bioinformatics pipeline, and the annotation and interpretation of variants were performed on the Sophia DDM® platform (Sophia Genetics, Boston, USA; https://www.sophiagenetics.com/).
Average coverage for targeted region was 722.5× of depth (minimum of 357× and maximum of 1788×). For the majority of the samples, we had ≥99.98% of the target region with minimum of 50× of depth of coverage (minimum 99.3% and maximum of 100%).
We used the scores from the association of the ACMG (American College of Medical Genetics and Genomics) and the ACGS (American Clinical Genomic Science) Best Practice Guidelines for Variant Classification 2019 for the analyses.
We used Emedgene® software with all information relevant to the criteria and scored the reads according to the evidence evaluated (prediction, literature, and information deposited in PubMed, Arup, ClinVar, and Varsome).
The 23 genes analyzed were ATM, ATR, CDKN2A, EGLN1, FH, HRAS, KIF1B, KMT2D, MAX, MDH2, MERTK, MET, NF1, PIK3CA, RET, SDHA, SDHAF2, SDHB, SDHC, SDHD, TMEM127, TP53, and VHL.
The genetic analysis of 47 of the 105 controls (44.7%) was performed by the same platform Sophia DDM®, whereas the remaining 58 (55.3%) were investigated using the platform Twist Bioscience®; the average coverage for targeted region was 752.5× of depth (minimum of 411.75× and maximum of 1065×). For the latter 58 samples, we had 100% of the target region with minimum of 50× of depth of coverage (minimum 99.6% and maximum of 100%).
Hormonal testing
Hormonal evaluation for the biochemical diagnosis of PPGL was performed with 24-h urinary metanephrines, 24-h urinary catecholamines, 24-h urinary VMA, plasma metanephrines, and plasma catecholamines. As of 2014, and in accordance with the Endocrine Society's PPGL Guideline (Lenders et al. 2014), we only performed tests for plasma and/or 24-h urinary fractionated metanephrines and VMA, whereas urinary and plasma catecholamines were discontinued.
Statistical testing
All statistical analyses were performed using IBM SPSS Statistics for Windows, Version 25.0 (Released 2019; IBM Corp, Armonk, NY, USA). Descriptive analysis included absolute (n) and relative (%) frequencies and bar graphs of qualitative variables and summary measures (mean, standard deviation, median, minimum, and maximum). Inferential analysis included the association test using the chi-square or Fisher's exact test and the logistic regression-stepwise forward method and receiver operating characteristic (ROC) curve analysis to evaluate possible cutoff points for quantitative variables, complemented by the calculation of sensitivity and specificity. A significance level of 5% was adopted for all tests.
Results
Patients, age, clinical presentation, time of presentation
The median age of the patients was 44.5 years (range 14–79); 74 patients were women and 41 were men.
Ninety-one patients (79.1%) satisfied the criteria that suggested PPGL was potentially hereditary (group 1) and 24 (20.9%) did not (group 2). In groups 1 and 2, 63.4% and 66.6% of the patients were women.
The association of systemic arterial hypertension (SAH) + paroxysms and orthostatic hypotension was present in 16 (34%) of genetically confirmed cases, followed by SAH + paroxysms in 3 (6.4%), orthostatic hypotension in 4 (8.5%), isolated SAH in 6 (12.8%), SAH + orthostatic hypotension in 8 (17%), isolated paroxysms in 3 (6.4%), and orthostatic hypotension + paroxysms in 1 (2.1%). Six (12.8%) of the patients were normotensive.
Distributions of germline variants
The genetic landscape of the 115 PPGL patients submitted to the genomic study showed that 67 (58.3%) had a germline variant in at least 1 of the 23 PPGL-related genes studied: 34/67 (50.7%) had exclusively pathogenic or likely PVs, 13 (19.4%) pathogenic or likely PVs and VUS, and 20 (29.8%) carried only VUS. This group included germline variants in the following genes: RET (18; 38.3%), VHL (10; 21.3%), SDHB, and NF1 (8; 17% each) and MAX, SDHD, TMEM127, and TP53 (1; 2.1% each). The VUS group (n = 42 germline variants (GVs)) included GVs in the following genes: ATM (9; 21.4%), KMT2D (6; 14.2%), MERTK (5; 11.9%), SDHC, MET, RET, and FH (3; 7.1% each), SDHA and NF1 (2; 4.7% each), and ATR,CDKN2A, SDHAF2, KIF1B,MDH2, and MAX (1; 2.4% each).
Since 14 subjects from 115 PPGL patients were relatives from eight different families, we recalculated germline variants distribution exclusively from the 101 PPGL index cases. This information is presented in Figs. 1 and 2.
Genetic distribution of 101 PPGL patients (excluding 14 relatives) and 105 PPGL controls according to ACMG criteria.
Citation: Endocrine Oncology 3, 1; 10.1530/EO-22-0091
Distribution of 72 germline variants in 101 PPGL patients (excluding 14 relatives).
Citation: Endocrine Oncology 3, 1; 10.1530/EO-22-0091
A table with VUS cases is provided as a supplementary material (see section on supplementary materials given at the end of this article). We did not perform segregation or functional studies of VUS cases, but all patients who were followed up in the Adrenal Unit at Unifesp were reassessed regularly.
In the sample of 105 PPGL individuals from the Fleury Group lab, PVs were present in 22 (21%) individuals and VUS in 35 (33.3%), whereas 56 (53.3%) subjects did not have any variants (Fig. 1). The PV distribution was as follows: 5 (22.7%) in each SDHB, VHL, and RET, followed by 2 (9%) in each SDHD and MERTK, and 1 (5.7%) in each ATM, NF1, SDHA, and TMEM127. The VUS distribution was as follows: 8 (22.8%) ATM, 5 (14.3%) KMT2D, 4 (11.4%) in each MERTK and SDHA, 2 (5.7%) in each NF1, FH, RET, and CDKN2A, and 1 (2.8%) in each ATR, KIF1B, SDHC, TMEM127, SDHB, and MET.
Regarding age distribution, none of the patients were in the first decade of life. For patients in the second decade of life, 50% had PV in the RET gene, 25% in VHL, and 25% in SDHB genes. For patients in the third decade of life, the incidence of PV was highest for the VHL and SDHB genes (28.6% each), followed by RET, MAX, and NF1 genes (14.2% each). For patients in the fourth decade of life, the incidence of PV was highest for the RET gene (36.3%), followed by VHL and SDHB (18.2% each) and the SDHD, TMEM127, and NF1 genes (9.1% each). For patients in the fifth decade of life, PV in the RET gene predominated (61.5%), followed by VHL and NF1 (15.4%, each) and the SDHB gene (7.7%). For patients in the sixth decade of life, PV in the NF1 gene predominated (66.6%), followed by the RET and VHL genes (16.7% each). For patients in the seventh decade of life, 50% had PV in SDHB and 50% in VHL genes.
Table 3 shows the general characteristics of the PPGL cohort and Table 4 shows the genetic (B) characteristics of the PPGL cohort.
General characteristics of the genetic PPGL cohort.
| Family | Positive cases | Age at diagnosis (years) | PPGLa (with metastasisg) | Hormonal profileb | Imagingc | Years of follow-up | |
|---|---|---|---|---|---|---|---|
| With PPGL | Without PPGL | ||||||
| 01 | 1 | – | 22 | Unilateral Pheo | NMN + MN | CT | 14 |
| 3 | 45 | Bilateral Pheo | MRI + MIBG | ||||
| 19 | |||||||
| 14 | |||||||
| 02 | 3 | – | 44 | Unilateral Pheo | NMN + MN | MRI + MIBG | 12 |
| 40 | |||||||
| 32 | |||||||
| 2 | 44 | Bilateral Pheo | |||||
| 52 | |||||||
| 03 | 2 | – | 49 | Unilateral Pheo | NMN + MN | CT + MIBG | 14 |
| 18 | |||||||
| 04 | 1 | – | 50 | Unilateral Pheo | NMN + MN | MRI + MIBG | 10 |
| 05 | 1 | – | 35 | Unilateral Pheo | NMN + MN | MRI + MIBG | 10 |
| 06 | 1 | – | 17 | Unilateral Pheo | MN | MRI + MIBG | 10 |
| 07 | 1 | 1 | 42 | Bilateral Pheo | NMN + MN | MRI + MIBG | 11 |
| 08 | 1 | – | 45 | Unilateral Pheo | MN | MRI + MIBG | 12 |
| 09 | 1 | – | 34 | Bilateral Pheo | MN | MRI + MIBG | 3f |
| 10 | 1 | 1 | 37 | Bilateral Pheo | NMN + MN | MRI + MIBG | 5 |
| 11 | 3 | – | 45 | Bilateral Pheo | NMN | MRI + MIBG | 15 |
| 40 | |||||||
| 18 | |||||||
| 12 | 1 | 5 | 38 | Unilateral Pheo | Nonfunct.f | MRI + MIBG | 4 |
| 13 | 1 | – | 21 | Abdominal PGL # | NMN | MRI + MIBG | 17 |
| 1 | 17 | 13 | |||||
| 14 | 1 | – | 48 | Unilateral Pheo | NMN | MRI + MIBG | 14 |
| 15 | 1 | – | 23 | Unilateral Pheo | NMN | MRI + MIBG | 15 |
| 16 | 1 | 1 | 33 | Bilateral Pheo | NMN | MRI + MIBG | 13 |
| 17 | 1 | 1 | 38 | Bilateral Pheo | NMN | MRI + MIBG | 7 |
| 18 | 1 | 2 | 62 | Neck PGL# | Nonfunct. | MRI + MIBG | 15 |
| 19 | 1 | – | 50 | Unilateral Pheo | NMN | MRI + MIBG | 14f |
| 20 | 1 | 8 | 17 | Abdominal PGL# | NMN | MRI + MIBG | 5 |
| 1 | 39 | Neck PGL | Nonfunct. | 7 | |||
| 21 | 1 | – | 23 | Abdominal PGL # | NMN | MRI + MIBG | 12 |
| 22 | 1 | – | 28 | Abdominal PGL | NMN | MRI + MIBG | 12 |
| 23 | 1 | – | 32 | Abdominal PGL | NMN | MRI + MIBG | 13 |
| 24 | 1 | – | 17 | Neck PGL | Nonfunct. | MRI + MIBG | 4 |
| 25 | 1 | – | 59 | Unilateral Pheo | NMN + MN | MRI + MIBG | 14 |
| 26 | 1 | – | 33 | Unilateral Pheo # | NMN + MN | MRI + MIBG | 24f |
| 27 | 1 | – | 52 | Unilateral Pheo | Nonfunct. | MRI + MIBG | 13 |
| 28 | 1 | – | 31 | Unilateral Pheo | NMN + MN | MRI + MIBG | 13 |
| 1 | 53 | Bilateral Pheo | Nonfunct. | ||||
| 29 | 2 | – | 49 | Unilateral Pheo | NMN + MN | MRI + MIBG | 11f/15 |
| 29 | |||||||
| 30 | 1 | – | 60 | Unilateral Pheo | NMN + MN | MRI + MIBG | 15 |
| 31 | 1 | 1 | 39 | Neck PGL | Nonfunct. | MRI + MIBG | 8 |
| 32 | 1 | – | 22 | Abdominal PGL | NMN + MN | MRI | 12 |
| 33d | 1 | 1 | 32 | Unilateral Pheo | NMN + MN | CT + MIBG | 21 |
| 1 | |||||||
| Total | 47 | 22 | Median= 35 years (14–62) | Unilateral = 21 Bilateral = 15 Neck PGL = 4 Abdominal PGL = 7 #with metastasis= 6 |
NMN+MN= 23 NMN= 14 MN= 3 Nonfunct. = 7 |
MRI + MIBG= 41 CT + MIBG=4 CT=1 MRI= 1 |
Median = 13 years (3–24) |
aPPGL, pheochromocytoma (= Pheo; adrenal)/paraganglioma (= PGL; extra-adrenal).
bHormonal profile: MN, metanephrines; NMN, normetanephrines.
cImaging: CT, computerized tomography, MIBG, metaIodo-benzyl-guanidine; MRI, magnetic resonance imaging.
dIndex case and one relative (without PPGL) with PV in the TMEM127 and TP53 genes and one relative with PV in the TMEM127 gene only.
eMAX, SDHD, and TMEM127/TP53.
fDeath.
gMetastasis
Nonfunct., nonfunctioning.
Genetic characteristics of the PPGL cohort.
| Family | Patients with PPGL | Gene | Familial syndromes | Other tumors | Nucleotide (NT) | Protein |
|---|---|---|---|---|---|---|
| 01 | 4 | RET | MEN2A | MTC / PHP | c.1900T>G | p. Cys634Gly |
| 02 | 5 | RET | MEN2A | MTC / PHP | c.1900T>G | p. Cys634Gly |
| 03 | 2 | RET | MEN2A | MTC | c.1901G>A | p. Cys634Tyr |
| 04 | 1 | RET | MEN2A | MTC | c.1901G>A | p. Cys634Gly |
| 05 | 1 | RET | MEN2A | MTC | c.1900T>C | p. Cys634Arg |
| 06 | 1 | RET | MEN2B | MTC | c.2753T>C | p. Met918Thr |
| 07 | 1 | RET | MEN2A | MTC | c.1859G>C | p. Cys620Ser |
| 08 | 1 | RET | MEN2B | MTC | c.2753T>C | p. Met918Thr |
| 09 | 1 | RET | MEN2B | MTC | c.2753T>C | p. Met918Thr |
| 10 | 1 | RET | MEN2A | MTC/PHP | c.1900T>C | p. Cys634Arg |
| 11 | 3 | VHL | VHL2A | Pancreatic NET (1)/retinal angioma (2)/cerebellum HB (2) | c.496G>T | p. Val166Phe |
| 12 | 1 | VHL | VHL2C | c.256C>T | p. Pro86Ser | |
| 13 | 2 | VHL | VHL2C | c.467A>G | p. Tyr156Cys | |
| 14 | 1 | VHL | VHL2B | Pancreatic NET/CCRC | c.233A>G | p. Asn78Ser |
| 15 | 1 | VHL | VHL2C | c.239G>T | p. Ser80Ile | |
| 16 | 1 | VHL | VHL2B | Pancreatic NET+retinal angioma+cerebellum HB+CCRC | c.499C>T | p. Arg167Trp |
| 17 | 1 | VHL | VHL2A | Cervical HB | c.293A>G | p. Tyr98Cys |
| 18 | 1 | SDHB | PGL4 | PHP | c.293G>A | p. Cys98Tyr |
| 19 | 1 | SDHB | PGL4 | c.591del | p. Ser198Alafs*22 | |
| 20 | 2 | SDHB | PGL4 | GIST (1) | c.137G>T | p. Arg46Leu |
| 21 | 1 | SDHB | PGL4 | Del involving exon 1, identified by CNV | ||
| 22 | 1 | SDHB | PGL4 | c.3G>A | p. Met1? | |
| 23 | 1 | SDHB | PGL4 | Del involving exon 1, identified by CNV | ||
| 24 | 1 | SDHB | PGL4 | c.724C>A | p. Arg242Ser | |
| 25 | 1 | NF1 | NF1 | c.1260+1G>A | p. (?) | |
| 26 | 1 | NF1 | NF1 | c.5487_5490 dup | p. Gly1831Profs*11 | |
| 27 | 1 | NF1 | NF1 | c.4537C>T | p. Arg1513* | |
| 28 | 2 | NF1 | NF1 | c.1527+4_1527+7 del (Del of 4 NT at position +4 (intron 13) | Affects a NT from the splicing donor site that possibly alters the processing of messenger RNA by skipping exon 13 | |
| 29 | 2 | NF1 | NF1 | c.6999 +2 T>G | p. (?) | |
| 30 | 1 | NF1 | NF1 | c.7248_7256del | p. Tyr2417_Ala2419del | |
| 31 | 1 | SDHD | PGL1 | c.196A>G | p. Met66Val | |
| 32 | 1 | MAX | Del involving exons 1 through 4*, identified by CNV | |||
| 33** | 1 | TMEM127 / TP53 | -/Li-Fraumeni | Breast Ca+pancreatic Ca | c.117_120 delc.1010 G>A | p. Ile41Argfs*39 p. Arg337His |
| Total | 47 | RET= 18 VHL= 10 SDHB= 8 NF1= 8 Other$= 3 |
MEN2A=15 MEN2B= 3 VHL2A=4 VHL2B=2 VHL2C=4 PGL4=8 PGL1=1 |
|||
(1), one affected patient; (2), two affected patients.
CCRC, clear cell renal carcinoma; HB, hemangioblastoma; MTC, medullary thyroid carcinoma; PHP, primary hyperparathyroidism.
Topographic distribution and metastases
In the confirmed hereditary patients, 34% had bilateral PPGL, 34% had right adrenal PPGL, 15% had retroperitoneal PGL, 8.5% had left adrenal PPGL and 8.5% had neck PGL. Among the bilateral PPGL cases, we identified germline PVs in RET (56.2%), VHL (37.5%), and NF1 (6.3%) genes (Table 3).
Six patients (12.7%) with germline PV had metastatic PPGL: four had PV in SDHB, one in VHL, and one in NF1 genes. The most frequent sites of metastasis were lymph nodes (n = 5), bone (n = 5), liver (n = 4), blood vessels (n = 4), and lungs (n = 3). Three patients with metastatic PPGL (two with PV in SDHB and one with PV in NF1) died because of tumor progression. Table 3 also shows individual imaging procedures for PPGL diagnosis and time of follow-up for each patient.
Clinical judgment score
Using the criteria for direct genetic testing, we found a sensitivity of 91.3%, specificity of 81.2%, positive predictive value (PPV) of 76.4%, negative predictive value (NPV) of 93.3%, and accuracy of 85.2%.
When the CJS was used to identify patients who did not benefit from the genetic test using a cutoff of ≥4 points, we found a sensitivity of 75%, specificity of 96.4%, PPV of 60%, NPV of 98.2%, and accuracy of 95%.
This clinical score developed from our cohort of patients was useful to exclude those who would not benefit from genetic testing.
The following characteristics were statistically significant: age, bilateral masses, classic PPGL-related disease phenotypes, and family history of PPGL. Metastasis and topography (adrenal or extra-adrenal) were not statistically significant. On logistic regression, the association with the greatest statistical power (to determine whether to perform the genetic test) was age ≤45 (P = 0.016 with odds ratio (OR) of 5.9 (1.4–25)), classic PPGL-related disease phenotypes (P < 0.001 with OR of 189 (34.4–1,038.3)), and paraganglioma (P < 0.001 with OR of 8.3 (1.0–39.3)). Family history, bilateral masses, and metastasis were not statistically significant in the logistic regression model.
The age cutoff of 45.5 years disclosed a sensitivity of 70.2% and a specificity of 54.4% (area under the curve of 0.694; P < 0.001) for the diagnosis of potentially genetic PPGL, as per ROC curve analysis.
Correlation of germline variants and metanephrines
Hormone production (urine and/or plasma metanephrines and normetanephrines) correlated with germline PV, as shown in Table 3. Note that cases with PVs in the RET, NF1, MAX, and TMEM127 genes had a predominant adrenergic profile, whereas cases with PVs in VHL and SDHB had an exclusive noradrenergic profile.
Discussion
The incidence of certain affected genes in PPGL is higher according to the population studied. The Chinese population, for instance, has fewer germline variants than Europeans. SDHx genes predominate among the European population (Bausch et al. 2017, Jiang et al. 2020).
Our study evaluated the profile of germline PPGL gene variants from 115 patients from a single reference center in São Paulo, Brazil. This population sample originated in part from an outpatient clinic that is specialized in MEN; this may have produced some reference bias, resulting in a higher PV frequency of the RET gene; on the other hand, it is also possible that the prevalence of RET gene PV in the Brazilian population is actually higher, as was also observed with the control sample from the Fleury Group databases, which encompasses patients from the four major regions of Brazil. Some reports have demonstrated that the RET PV profile may vary according to geographical area (Pena et al. 2009, Pena et al. 2011, Maia et al. 2014, Bausch et al. 2017, Cunha et al. 2017, Maciel et al. 2019, 2021). Further studies should be carried out to clarify this issue. Neumann et al. highlighted a greater incidence of PV in the VHL gene in the first decade of life, which decreases in the third decade of life but remains one of the most important genes found. From the second decade of life onward, the incidence of PV in the genes of the SDHx complex began to increase, assuming a greater proportion in cases of PGL and metastasis. The incidence of RET gene variants is important around the third decade of life but is less than those of the VHL and SDHx genes, highlighting the characteristic of bilateral disease and virtual absence of metastases. NF1 gene variants stand out in the fourth and fifth decades of life. (Neumann et al. 2019).
The distribution in our population sample was different: RET was the most prevalent gene with PV in the second, fourth, and fifth decades of life, whereas NF1 was the most prevalent gene with PV in the sixth decade of life. The SDHB and VHL genes were present in all decades of life.
The pathogenic or likely PVs found in this study were comparable to those reported in the medical literature. Rare variants, however, were found in the SDHB, VHL, and NF1 genes, such as SDHB: c.137G>T in exon 2, c.293G>A in exon 4, and c.724C>A in exon 7; VHL: c.467A>G and c.496G>T, both in exon 3; and NF1: c.5487_5490dup resulting in frameshift in exon 37, c.1527+4_1527+7del in intron 13, and c.6999+2T>G in intron 46. We also found a unique association of PV in the TMEM127 and TP53 genes (R337H) in a 47-year-old patient who developed bilateral metastatic breast cancer, meningioma of the CNS, and pancreatic cancer; her 20-year-old daughter is a carrier of the same variants who did not develop PPGL thus far but faced bilateral breast cancer and meningioma of the CNS and her 5-year-old granddaughter carries only the TMEM127 gene variant and currently has no clinical or laboratory evidence of PPGL.
The patient with PV c.5487_5490dup in the NF1 gene developed bone, liver, and lung metastases at 22 years after PPGL resection and died after developing severe febrile neutropenia following therapy with 131I-mIBG; to our knowledge, this is the first NF1-associated PPGL case reported in which metastasis developed only 20 years later.
There is a strong genetic determinism in PPGL. Clinical features at the initial diagnosis of PPGL may suggest germline variants, allowing for grouping of such cases as potentially hereditary, especially in the presence of typical phenotypic conditions, as mentioned before. It is generally agreed that all PPGLs must be submitted to genetic evaluation (Jiang et al. 2020, Garcia-Carbonero et al. 2021, Cascon et al. 2023, Gimenez-Roqueplo et al. 2023); however, when the availability of genetic NGS panels is greatly limited, as in Brazil, we suggest the use of a CJS to define which individuals should be initially exempt from genetic testing while maintaining a permanent follow-up, with the need for genetic testing being reassessed periodically. Although our CJS had excellent specificity, NPV, and accuracy, it must be validated prospectively. On the other hand, the association of the presence of genetic variants with age <45 years, the presence of PPGL-related genetic syndromes, and topography (adrenal/extra-adrenal) strongly indicates that patients with such characteristics should undergo genetic testing.
The clinical characteristics employed to direct NGS research in our patients were sufficient to guarantee helpful coverage for the main related genes, achieving good sensitivity with acceptable specificity and useful NPV.
As per the ACMG, 33% of our cohort of 101 index cases was classified as pathogenic/likely pathogenic, 30% as VUS, and 37% as probably sporadic. Therefore, 63% carry some germline variant, which is slightly higher than other published series. All VUS cases are periodically reassessed to determine whether they will change category; thus far, however, all such cases in this study remain VUS, a percentage that agrees with the literature.
Although ATM gene variants were significantly in our cohort, there is no current evidence between ATM variants and the presence of PPGL, making this finding of unknown clinical significance.
When compared to the control group (patients from the Fleury Lab database), we noticed a higher rate of PV. We did not have access to clinical or imaging characteristics of the Fleury Group of patients, once they were uncharacterized, but it is possible that they were not properly selected for genetic testing; conceivably, the use of the CJS would have reduced the number of sporadic cases. We believe that use of the CJS would make genetic testing more judicious and cost-effective for public and private health systems.
In summary, we describe the landscape of PPGL germline gene variants from a large sample of Brazilian patients in which pathogenic and likely PVs were highly prevalent, especially in the RET gene. We also suggest a CJS to identify PPGL patients who would not require initial genetic evaluation, improving test specificity and reducing costs.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EO-22-0091.
Declaration of interest
There is no conflict of interest that could be perceived as prejudicing the impartiality of this research.
Funding
This project has been partially supported by the Fleury Lab Group (grant numbers NP-335 and NP- 334). This work did not receive any specific grant from any funding agency in the public.
Acknowledgements
The authors thank Erica S. Souza de Araujo and Pamela M. Lima do Livramento (Fleury Group) for their assistance in analyzing patient samples and José de Sá for the analysis of the database. The authors are grateful to colleagues from Santa Casa de São Paulo School of Medical Sciences for referring patients for the study and to Erika Tiemi Fukunaga and Ting Hui Ching from the Statistics Service for their helpful advice.
References
Al Subhi AR, Boyle V & & Elston MS 2022 Systematic review: incidence of pheochromocytoma and paraganglioma over 70 years. Journal of the Endocrine Society 6 bvac105. (https://doi.org/10.1210/jendso/bvac105)
Antonio K, Valdez MMN, Mercado-Assis L, Taïeb D & & Pacak K 2020 Pheochromocytoma/paraganglioma: recent updates in genetics, biochemistry, immunohistochemistry, metabolomics, imaging, and therapeutic options. Gland Surgery 9 105–123. (https://doi.org/10.21037/gs.2019.10.25)
Bausch B, Schiavi F, Ni Y, Welander J, Patocs A, Ngeow J, Wellner U, Malinoc A, Taschin E, Barbon G, et al.2017 European-American-Asian pheochromocytoma-paraganglioma registry study group. Clinical characterization of the pheochromocytoma and paraganglioma susceptibility genes SDHA, TMEM127, MAX and SDHAF2 for gene-informed prevention. JAMA Oncology 3 1204–1212. (https://doi.org/10.1001/jamaoncol.2017.0223)
Bravo EL & & Tagle R 2003 Pheochromocytoma: state-of-the-art and future prospects. Endocrine Reviews 24 539–553. (https://doi.org/10.1210/er.2002-0013)
Buffet A, Burnichon N, Favier J & & Gimenez-Roqueplo A-P 2017 An overview of 20 years of genetic studies in pheochromocytoma and paraganglioma. Best Practice and Research in Clinical Endocrinology and Metabolism 34 1–14.
Cascón A, Calsina B, Monteagudo M, Mellid S, Díaz-Talavera A, Currás-Freixes M & & Robledo M 2023 Genetic bases of pheochromocytoma and paraganglioma. Journal of Molecular Endocrinology 70 e220167. (https://doi.org/10.1530/JME-22-0167)
Cunha LL, Lindsey SC, França MIC, Sarika L, Papathoma A, Kunii IS, Cerutti JM, Dias-da-Silva MR, Alevizaki M & & Maciel RMB 2017 Evidence for the founder effect of RET533 as the common Greek and Brazilian ancestor spreading multiple endocrine neoplasia 2A. European Journal of Endocrinology 176 515–519. (https://doi.org/10.1530/EJE-16-1021)
Garcia-Carbonero R, Matute Teresa F, Mercader-Cidoncha E, Mitjavila-Casanovas M, Robledo M, Tena I, Alvarez-Escola C, Arístegui M, Bella-Cueto MR, Ferrer-Albiach C, et al.2021 Multidisciplinary practice guidelines for the diagnosis, genetic counseling and treatment of pheochromocytomas and paragangliomas. Clinical and Translational Oncology 23 1995–2019. (https://doi.org/10.1007/s12094-021-02622-9)
Gimenez-Roqueplo AP, Robledo M & & Dahia PLM 2023 Update on the genetics of paragangliomas. Endocrine-Related Cancer 30 e220373. (https://doi.org/10.1530/ERC-22-0373)
Jiang J, Zhang J, Pang Y, Bechmann N, Li M, Monteagudo M, Calsina B, Gimenez-Roqueplo AP, Nölting S, Beuschlein F, et al.2020 Sino-European differences in the genetic landscape and clinical presentation of pheochromocytoma and paraganglioma. Journal of Clinical Endocrinology and Metabolism 105 3295–3307. (https://doi.org/10.1210/clinem/dgaa502)
Karagiannis A, Mikhailidis DP, Athyros VG & & Harsoulis F 2007 Pheochromocytoma: an update on genetics and management. Endocrine-Related Cancer 14 935–956. (https://doi.org/10.1677/ERC-07-0142)
Lenders JWM, Duh QY, Eisenhofer G, Gimenez-Roqueplo AP, Grebe SKG, Murad MH, Naruse M, Pacak K, Young WF & Endocrine Society 2014 Pheochromocytoma and paraganglioma: an Endocrine Society clinical practice guideline. Journal of Clinical Endocrinology and Metabolism 99 1915–1942. (https://doi.org/10.1210/jc.2014-1498)
Lindsey SC, Kunii IS, Germano-Neto F, Sittoni MY, Camacho CP, Valente FOF, Yang JH, Signorini PS, Delcelo R, Cerutti JM, et al.2012 Extended RET gene analysis in patients with apparently sporadic medullary thyroid cancer: clinical benefits and cost. Hormones and Cancer 3 181–186. (https://doi.org/10.1007/s12672-012-0109-7)
Maciel RMB, Camacho CP, Assumpção LVM, Bufalo NE, Carvalho AL, de Carvalho GA, Castroneves LA, de Castro FM, Ceolin L, Cerutti JM, et al.2019 Genotype and phenotype landscape of MEN2 in 554 medullary thyroid cancer patients: the BrasMEN study. Endocrine Connections 8 289–298. (https://doi.org/10.1530/EC-18-0506)
Maciel RMB & & Maia AL 2021 GLOBAL ENDOCRINOLOGY: geographical variation in the profile of RET variants in patients with medullary thyroid cancer: a comprehensive review. European Journal of Endocrinology 186 R15–R30. (https://doi.org/10.1530/EJE-21-0753)
Maia AL, Siqueira DR, Kulcsar MA, Tincani AJ, Mazeto GM & & Maciel LM 2014 Diagnosis, treatment, and follow-up of medullary thyroid carcinoma: recommendations by the Thyroid Department of the Brazilian Society of Endocrinology and Metabolism. Arquivos Brasileiros de Endocrinologia e Metabologia 58 667–700. (https://doi.org/10.1590/0004-2730000003427)
Neumann HPH, Young WF Jr & & Eng C 2019 Pheochromocytoma and paraganglioma. New England Journal of Medicine 381 552–565. (https://doi.org/10.1056/NEJMra1806651)
NGS in PPGL (NGSnPPGL) Study Group, Toledo RA, , Burnichon N, , Cascon A, , Benn DE, , Bayley JP, , Welander J, , Tops CM, , Firth H, , Dwight T, et al.2017 Consensus Statement on next-generation-sequencing-based diagnostic testing of hereditary phaeochromocytomas and paragangliomas. Nature Reviews. Endocrinology 13 233–247. (https://doi.org/10.1038/nrendo.2016.185)
Pena SD, Bastos-Rodrigues L, Pimenta JR & & Bydlowski SP 2009 DNA tests probe the genomic ancestry of Brazilians. Brazilian Journal of Medical and Biological Research 42 870–876. (https://doi.org/10.1590/s0100-879x2009005000026)
Pena SD, Di Pietro G, Fuchshuber-Moraes M, Genro JP, Hutz MH, Kehdy Fde S, Kohlrausch F, Magno LA, Montenegro RC, Moraes MO, et al.2011 The genomic ancestry of individuals from different geographical regions of Brazil is more uniform than expected. PLoS One 6 e17063. (https://doi.org/10.1371/journal.pone.0017063)
Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E, Spector E, et al.2015 Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genetics in Medicine 17 405–424. (https://doi.org/10.1038/gim.2015.30)
Sutton MG, Sheps SG & & Lie JT 1981 Prevalence of clinically unsuspected pheochromocytoma. Review of a 50-year autopsy series. Mayo Clinic Proceedings 56 354–360. (https://doi.org/10.1016/S0022-5347(1753807-0)
Thomas CM, Asa SL, Ezzat S, Sawka AM & & Goldstein D 2019 Diagnosis and pathologic characteristics of medullary thyroid carcinoma-review of current guidelines. Current Oncology 26 338–344. (https://doi.org/10.3747/co.26.5539)
Wells SA Jr, Asa SL, Dralle H, Elisei R, Evans DB, Gagel RF, Lee N, Machens A, Moley JF, Pacini F, et al.2015 Revised American Thyroid Association guidelines for the management of medullary thyroid carcinoma. Thyroid 25 567–610. (https://doi.org/10.1089/thy.2014.0335)
