Abstract
Context
Adrenal lesions are frequent among patients with sporadic neuroendocrine tumors (spNETs) or multiple endocrine neoplasia type 1 (MEN1). Armadillo repeat-containing 5 (ARMC5)-inactivating variants cause adrenal tumors and possibly other neoplasms.
Objective
The objective of this work is to investigate a large cohort spNETs or MEN1 patients for changes in the ARMC5 gene.
Patients and Methods
A total of 111 patients, 94 with spNET and 17 with MEN1, were screened for ARMC5 germline alterations. Thirty-six tumors (18 spNETs and 18 MEN1 related) were collected from 20 patients. Blood and tumor DNA samples were genotyped using Sanger sequencing and microsatellite markers for chromosomes. ARMC5 and MEN1 expression were assessed by immunohistochemistry.
Results
In 76 of 111 (68.4%) patients, we identified 16 different ARMC5 germline variants, 2 predicted as damaging. There were no differences in the prevalence of ARMC5 variants depending on the presence of MEN1-related adrenal lesions. Loss of heterozygosity (LOH) at chromosome 16p and ARMC5 germline variants were present together in 23 or 34 (67.6%) tumors; in 7 of 23 (30.4%) their presence led to biallelic inactivation of the ARMC5 gene. The latter was more prevalent in MEN1-related tumors than in spNETs (88.9% vs 38.9%; P = .005). LOH at the chromosome 16p (ARMC5) and 11q (MEN1) loci coexisted in 16/18 MEN1-related tumors, which also expressed lower ARMC5 (P = .02) and MEN1 (P = .01) proteins compared to peritumorous tissues.
Conclusion
Germline ARMC5 variants are common among spNET and MEN1 patients. ARMC5 haploinsufficiency or biallelic inactivation in spNETs and MEN1-related tumors suggests that ARMC5 may have a role in modifying the phenotype of patients with spNETs and/or MEN1 beyond its known role in macronodular adrenocortical hyperplasia.
Keywords: ARMC5 gene, germinal and somatic alterations, neuroendocrine tumors, multiple endocrine neoplasia type 1 (MEN1)
Primary macronodular adrenal hyperplasia (PMAH) is frequently caused by inactivation of the armadillo repeat-containing 5 (ARMC5) gene (1-4). Mutations both at the germline and somatic levels lead to complete abrogation of ARMC5 expression in tumor cells, consistent with the gene’s tumor-suppressor function. ARMC5-damaging variants are identified in up to 40% of patients with PMAH, with up to 50% of them being frameshift and/or nonsense mutations that cause premature termination of translation. A few of the missense variants that have been found in PMAH patients seem to also lead to the gene’s inactivation in vitro and are associated with additional ARMC5-inactivating changes in the patients’ adrenal tissues (5, 6). However, the biological effect of most missense variants of the ARMC5 gene (that are reported in international databases from germline or tumor studies) remains largely unknown.
Until now, 8 transcripts of the ARMC5 gene have been reported in databases (http://www.ensembl.org/index.html); 2 of these sequence isoforms, ARMC5-006 and ARMC5-007, are not protein-coding. Tissue-specific distribution of expression has been documented for 4 of the 6 remaining ARMC5 isoforms; only ARMC5-002 is present in all 46 normal human tissues, whereas thymus, pancreas, adrenal gland, adipose tissue, trachea, and lung appear to express all 4 isoforms. Finally, ARMC5-003 and ARMC5-001 appear to be more tissue specific (7).
Beyond PMAH, biallelic inactivation of the ARMC5 gene due to damaging germline and tumor-specific somatic mutations has been shown in meningioma, suggesting that ARMC5-associated oncogenesis may not be limited to the adrenal cortex (8). In general, the prevalence of adrenal masses varies depending on the source of data and selection of patients. Biological behavior for neuroendocrine tumors (NETs) ranges from an indolent to an aggressive clinical course with rapid cancer progression and death. The majority of these lesions are benign adrenal adenomas and their presence does not correlate with extra-adrenal metastases, while adrenal metastases were confirmed only in patients with G2 and G3 tumors. Subclinical cortisol hypersecretion was documented in 14% of patients with NET and coexistent adrenal tumor (9). The prevalence of adrenocortical masses among patients with multiple endocrine neoplasia type 1 (MEN1) appears to be as high as 20%, with bilateral involvement in 12.5% of these cases. In a retrospective study, the prevalence of adrenal masses in sporadic gastroenteropancreatic neuroendocrine tumors patients was 8.4% (9, 10).
In the present study, we speculated that ARMC5 variants may be frequent among patients with spNETs or MEN1, especially those with adrenal involvement. We sequenced both germline and tumor (when available) DNA from a large cohort of patients with spNETs or MEN1. The data suggest that ARMC5 may play a modifying role of the phenotype among patients with spNETs and/or MEN1, a role that is beyond its known association with PMAH.
Materials and Methods
Patients
A total of 111 unrelated patients with sporadic neuroendocrine tumors (spNETs) or NETs associated with genetically confirmed MEN1, regardless of the presence of known adrenal tumor(s), were recruited for the study. We then screened for the presence of adrenal involvement, and the latter was detected in 64.9% of all patients. A relatively small cohort of MEN1 cases and among them a high frequency of adrenal involvement was seen (12/17 [70.6%]). Patients with sporadic pituitary neuroendocrine tumors (pitNETs) were also included in the study (Table 1). Altogether these patients had 147 NETs and 6 adenocarcinomas, 2 breast cancers, 2 colorectal carcinomas, and 2 lung cancers (1 squamous and 1 adenocarcinoma). More than one-third, 22 of 62 (35.5%) patients with gastroenteropancreatic and lung NETs had locoregional disease or distant metastases and had good performance status score (Eastern Cooperative Oncology Group/World Health Organization Performance Status 0-1). Twenty-eight patients had paraneoplastic syndrome due to functional NET (see Table 1); 5 patients had carcinoid syndrome (history of flushing and/or diarrhea), 4 patients had Zollinger-Ellison syndrome due to gastrinoma, 3 patients had hypoglycemic syndrome caused by insulinoma, and in 10 ectopic Cushing syndrome was diagnosed (5 pancreatic and 5 lung NETs). Finally, we also saw 1 patient with ectopic acromegaly, 1 with ectopic production of calcitonin (both caused by lung NETs), 1 patient with duodenopancreatic somatostatinoma, and 3 patients had humoral hypercalcemia of malignancy (based on hypercalcemia and suppressed/undetectable parathyroid hormone). Somatostatin-receptor scintigraphy was positive in 61% of patients.
Table 1.
Clinical characteristics of patients with neuroendocrine tumors
| Characteristic | Patients with MEN1 syndrome | All patients with adrenal mass | All patients without adrenal mass |
|---|---|---|---|
| n/N, % | 17/111 (15.3) | 72(12 + 60)b/111 (64.9) | 39(5 + 34)b/111 (35.1) |
| BLT adrenal mass, n/N% | 10/12 (83.3) | 27/60 (45%) | NA |
| Age (range), y | 47.5 ± 14.3 (10-75) | 49.9 ± 14.3 (19-75) | 42.8 ± 14.9 (10-70) |
| Male sex, No., % | 3/17 (17.6) | 18/72 (25.0) | 8/39 (20.5) |
| NETs, n/N, % | 38/147 (25.8) | (38 + 53)b 91/147 (61.9) | 56/147 (38.1) |
| Functional NETs, n/N, %a | 5/28 (17.8) | (5 + 12)b 17/91 (18.7) | 11/56 (19.6) |
| Serum CgA, ng/mL | 203.3 ± 193.8 | 241.8 ± 225.9 | 233.9 ± 184.6 |
| Ki67, % (range) | 4.8 ± 4.7 (0.1-13.4) | 14.9 ± 20.9 (0.1-78.0) | 24.5 ± 35.3 (1.5-95.0) |
| pitNETs, n/N, % | 14/17 (82.3) | (14 + 17)b 31/60 (51.7) | 29/60 (48.3) |
| PRL-secreting | 6/14 (42.9) | (6 + 1)b 7/31 (22.6) | 3/29 (10.3) |
| ACTH-secreting | 1/14 (7.1) | (1 + 8)b 9/31 (29.0) | 18/29 (62.1) |
| TSH-secreting | 1/14 (7.1) | (1 + 0)b 1/31 (3.2) | 1/29 (3.5) |
| GH-secreting | 0/14 (0) | (0 + 3)b 3/31 (9.7) | 0/29 (0) |
| NFPA | 6/14 (42.9) | (6 + 6)b 12/31 (38.7) | 7/29 (24.1) |
| PHPT | 13/22 (47.2) | (13 + 14)b 17/22 (77.3) | 5/22 (22.7) |
| LNET | 3/23 (13.0) | (3 + 13)b 16/23 (69.6) | 7/23 (30.4) |
| PNET | 10/37 (27.0) | (10 + 16)b 26/37 (70.3) | 11/37 (29.7) |
| Ileal NET | 0/1 (0) | 1/1 (100.0) | 0/1 (0) |
| Appendiceal NET | 0/1 (0) | 1/1 (100.0) | 0/1 (0) |
| EAPG | 0/1 (0) | 0/1 (0.0) | 1/1 (100.0) |
| APG | 0/1 (0) | 0/1 (0) | 1/1 (100.0) |
| spMTC | 0/2 (0) | 0/2 (0) | 2/2 (100.0) |
| Nonendocrine tumors | 3/6 (50.0) | 2/6 (33.3) | 3/6 (66.7) |
Data are presented as mean ± SD.
Abbreviations: ACTH, adrenocorticotropin; APG, adrenal paraganglioma; BLT, bilateral, CgA, chromogranin A; EAPG, extra-adrenal paraganglioma; GH, growth hormone (somatotropin); LNET, lung neuroendocrine tumor; MEN1, multiple endocrine neoplasia type 1; NET, neuroendocrine tumor; NFPA, clinically nonfunctioning pituitary adenoma; PHPT, parathyroid adenoma/hyperplasia; pitNET, pituitary neuroendocrine tumor; PNET, pancreatic neuroendocrine tumor; PRL, prolactin; spMTC, sporadic medullary thyroid carcinoma; TSH, thyrotropin.
a Hypercalcemia due to primary hyperparathyroidism and endocrine syndromes caused by pitNETs are not counted.
b Sum of MEN1 patients and patients with sporadic NETs.
All of the patients underwent elective surgical resection or biopsy (during endoscopic/bronchoscopic procedures or ultrasound/computed tomography–guided biopsy) of primary tumors, except patients with prolactinomas. In addition to the light microscopy, diagnosis of NET was confirmed by immunostaining. Minimal immunohistochemical evaluation comprised assessment of reactions to synaptophysin, CD56, and chromogranin A. In all tumor tissues, mitotic and proliferative (Ki67) indexes were determined. Clinical characteristics of patients are presented in Table 1.
All patients underwent medical assessment in the Center for Neuroendocrine Tumors and Hereditary Cancer Syndromes of Clinic for Endocrinology, Diabetes and Metabolic Diseases, Medical School, University of Belgrade, between 2014 and 2018. Informed consent was obtained from all participants in the study, which was approved by the local institutional review board. DNA and certain tumor studies were conducted in a blinded manner at the National Institutes of Health laboratory of the authors after institutional approval of studies on unidentified samples.
DNA sequencing
Detection of germline alterations in the ARMC5 gene was carried out using genomic DNA in all 111 patients. In addition to patients with genetically confirmed MEN1 syndrome included in the study, we also searched for germline mutations in the MEN1 gene among patients with atypical syndrome consisting of adrenal and one typical MEN1 tumor: 17 patients with pituitary tumors, 4 patients with primary hyperparathyroidism, and 12 of 16 patients with pancreatic NET. The entire coding region (exons 2-10) plus flanking splice sites of the MEN1 gene were amplified by polymerase chain reaction (PCR) using specific primers. The set of primers for amplification of each exon were designed by Primer3 software and their sequences are available on request. Six exons and flanking intronic sequences of the ARMC5 gene were amplified by PCR using specific primers (Invitrogen) that were exploited in previous studies (1), and MyTaq Mix (Bioline). Patients were also sequenced for the coding sequences of the CDKN1B gene; no other genes were sequenced because there were no suggestive phenotypes. There were no variants of the CDKN1B gene in any of the patients studied (data not shown).
The PCR products were obtained from genomic and tumor DNA. Tissue samples that were collected during surgery included 26 well-differentiated (G1/G2) NETs, 14 from patients with sporadic and 12 from MEN1-related NETs. Also, from patients with MEN1 syndrome, 2 colorectal adenocarcinomas, 2 breast carcinomas, and 6 adrenocortical tumors were collected. Sequencing was performed in both directions by the Sanger method using the Big Dye Terminator v3.1 Ready Reaction Cycle Sequencing Kit (Thermo Fisher Scientific, Applied Biosystems) after purification with a GeneJET PCR Purification kit (Thermo Fisher Scientific). Samples were sequenced on an automated ABI PRISM 3130 Genetic Analyzer (Thermo Fisher Scientific, Applied Biosystems), and ABI DNA Sequencing Analysis Software v5.2 was used for analysis.
Multiplex ligation-dependent probe amplification analysis
The copy number of the MEN1 gene was tested in all 36 tumor samples using the SALSA multiplex ligation-dependent probe amplification (MLPA) P017-C1 MEN1 kit as described in the manufacturer’s general protocol for detection and quantification of DNA sequences (MLPA General Protocol, MRC-Holland). PCR products were run on an ABI PRISM 3130 Genetic Analyzer with (LIZ) GS 500 size standard. Data were visualized with GeneMapper Software, version 3.7 (Applied Biosystems). Coffalyser.net software (MRC-Holland) was used for fragment analysis and comparative analysis of MLPA samples. DNA samples obtained from healthy control individuals and a negative control (no DNA control) were included in the MLPA analysis. Probe ratios less than 0.6 or greater than 1.3 were considered as cutoff values for heterozygous deletion or amplification, respectively.
Loss-of-heterozygosity analysis
Loss-of-heterozygosity (LOH) analysis for chromosome 16p was performed using 10 microsatellite markers (D16S521, D16S3024, D16S3134, D16S3082, D16S475, D16S3065, D16S3027, D16S423, D16S3135, and D16S3108), as previously described (1). LOH analysis for chromosome 11q was performed using 9 microsatellite markers (D11S480, D11S1883, D11S599, PYGM, D11S4936, D11S449, D11S4907, D11S4908, and D11S913), as described elsewhere (11). PCRs were performed using fluorescently labeled primers (Applied Biosystems) and MyTaq Mix (Bioline). PCR products were analyzed on an ABI PRISM 3130 Genetic Analyzer by Gene Mapper software, version 3.7. The allelic ratios were determined for each marker in the leukocyte and in the tumor tissues. The normalized ratio was obtained by dividing the allelic ratios of the leukocyte and that of the tumor tissue. The threshold for allelic imbalance (AI) was defined as 40% reduction of one allele, in accordance with an AI factor of 1.7 or greater or 0.59 or less; values between were discarded completely from the analysis.
In silico analysis
The predicted effects of all variants in the ARMC5 and MEN1 genes were assessed using Polyphen-2 (http://genetics.bwh.harvard.edu/pph2/), SIFT (http://sift.jcvi.org/www/SIFT_enst_submit.html), and Mutation Taster (http://www.mutationtaster.org) platforms.
Immunohistochemical staining of tumor tissues for armadillo repeat-containing 5 and menin proteins
Eleven tumor samples, 8 NETs, and 3 adenocarcinomas (2 breast cancers and 1 rectal carcinoma) obtained from patients with genetically confirmed MEN1 syndrome underwent immunohistochemistry for the expression of ARMC5 and menin proteins. Fixation time was 24 to 36 hours. Formalin-fixed paraffin-embedded tissue in 4-µm-thick sections were cut, dewaxed, rehydrated, and boiled in 10 mM citrate buffer of pH 6.0, using the Thermo Scientific PT Module (at 98°C for 20 minutes). Endogenous peroxidase activity was blocked with 3% hydrogen peroxide for 10 minutes. Subsequently, slides were incubated for 10 minutes in 1% bovine serum albumin followed by primary antibody incubation with a rabbit polyclonal anti-ARMC5 antibody (NBP1-94024; Novus Biologicals, LLC; 1:20) or with mouse monoclonal antimenin antibody (EPR3986, Abcam; 1:100) for 1 hour at room temperature. Primary antibody binding was detected by using an avidin-biotin-peroxidase kit (UltraVision LP Detection System: HRP Polymer; Thermo Scientific) according to the manufacturer’s protocol.
Quantification of ARMC5 and menin expression was performed by a scoring system previously described (12). Briefly, to generate the score number (range, 0-9), the percentage of cells stained at each intensity level (0 = 0%, 1 ≤ 30%, 2 = 30%-60%, and 3 > 60%) was multiplied by the weighted intensity of staining (eg, 1, 2, or 3; where 0 is no staining, 1 is weak staining, 2 is moderate staining, and 3 is strong staining).
Statistics
Results were recorded as frequencies and percentages and mean ± SD as appropriate. Depending on the variable type used in comparisons, differences among the groups were determined using the ?2 test with the Fisher exact test, and t test for independent samples or nonparametric tests as required. To calculate the relative risk with 95% CI for detection of LOH on chromosome 16p related to germline alterations in the ARMC5 gene, we used a 2 × 2 contingency table. Results were considered significant when P was less than .05.
Results
Armadillo repeat-containing 5 alterations in patients with sporadic neuroendocrine tumors and multiple endocrine neoplasia type 1
Compared to the reference genome (GRCh37), 16 variants in the ARMC5 gene were identified in 76 of 111 (68.4%) patients. Seven of these alterations are frequent in the general population, 8 are rare variants with a minor allele frequency (MAF) of less than 1%, and 1 alteration was found that has not been reported in public databases (Table 2). All rare variants, 2 synonymous and 6 missense single nucleotide changes, were located within the coding gene region (see Table 2). According to American College of Medical Genetics and Genomics standards and guidelines, 2 of these rare gene variants were classified as pathogenic (c.1223A > G p.Q408R) and likely pathogenic (c.1676C > T p.P559L), and both of them were detected in patients with mutations in the MEN1 gene (c.941delG p.R314Rfs*54 and c.494TG > A p.C165 Y) and classical MEN1 syndrome (primary hyperparathyroidism, pituitary tumor, and pancreatic NET). Three ARMC5 gene variants were classified as likely benign (c.1641G > A p.A547A, c.1831C > T p.R611W, and c.2045G > A p.R682Q) and 3 as benign (c.968G > C p.G323A, c.1499C > T p.A500V, and c.2256C > T p.F752F) (see Table 2).
Table 2.
Armadillo repeat-containing 5 germline alterations in the cohort of patients with sporadic and multiple endocrine neoplasia type 1–related neuroendocrine tumors
| ARMC5 cDNA | ARMC5 protein | Gene | rs(dbSNB) | MAF of ARMC5 gene variants | |
|---|---|---|---|---|---|
| Variation | Variant | Region | Studied patients | GnomAD | |
| c.41T > A | p.F14Y | Exon 1 | rs151069962 | A = 0.03603 (8/222) | A = 0.04858 |
| c.475 + 58A > G | – | Intron 1 | rs9926717 | G = 0.02702 (6/222) | G = 0.29571 |
| c.508A > G | p.I170V | Exon 2 | rs35923277 | G = 0.04504 (10/222) | G = 0.03638 |
| c.968G > C | p.G323A | Exon 3 | rs35461188 | C = 0.00450 (1/222) | C = 0.00107 |
| c.1007A > Ga | p.D336G | Exon 3 | rs1039442244 | G = 0.00450 (1/222) | – |
| c.1223A > Gb | p.Q408R | Exon 3 | rs141923065 | G = 0.00450 (1/222) | G = 0.00310 |
| c.1499C > T | p.A500V | Exon 4 | rs536986858 | T = 0.00450 (1/222) | T = 0.00007 |
| c.1520C > T | p.P507L | Exon 4 | rs142376949 | T = 0.02702 (6/222) | T = 0.02426 |
| c.1641G > Ad | p.A547A | Exon 4 | rs61732352 | A = 0.00900 (1/222) | A = 0.00488 |
| c.1676C > Tc | p.P559L | Exon 4 | rs200115942 | T = 0.00450 (1/222) | T = 0.00098 |
| c.1831C > Td | p.R611W | Exon 4 | rs369074065 | T = 0.00450 (1/222) | T = 0.00002 |
| c.1842C > G | p.L614L | Exon 4 | rs55800131 | G = 0.06756 (15/222) | G = 0.06575 |
| c.2114C > Te | p.A705V | Intron 4 | rs11150624 | T = 0.20720 (46/222) | T = 0.40469 |
| c.2045G > Ad | p.R682Q | Exon 6 | rs372459614 | A = 0.00450 (2/222) | A = 0.00006 |
| c.2256C > T | p.F752F | Exon 6 | rs200335852 | T = 0.00450 (1/222) | T = 0.00146 |
| *234_*238dupGGCCT | – | 3′UTR | rs142544346 | –(12/222) | – |
Abbreviations: ARMC5 complementary DNA (cDNA) nucleotide variation RefSeq NM_001105247.1; ARMC5 protein RefSeq NP_001098717.1. Biological impact of presented gene variants is in accordance with obtained American College of Medical Genetics and Genomics categories (13): PVS1, null variant (nonsense, frameshift, canonical ± 1 or 2 splice sites, initiation codon, single or multiexon deletion) in a gene where loss of function is a known mechanism of disease; PP3, multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc); BS1 to BS4, strong evidence of benign impact; BP1 to BP7, supporting evidence of benign impact.
Abbreviations: ARMC5, armadillo repeat-containing 5 gene; dbSNP, Single-Nucleotide Polymorphism Database; ID, identifying number given only for carriers of pathogenic and likely pathogenic ARMC5 variants; MAF, minor allele frequency; GnomAD, The Genome Aggregation Database https://gnomad.broadinstitute.org/; rs, reference SNP; UTR, untranslated region.
a New benign alteration.
b Pathogenic; ID P80.
c Likely pathogenic; ID P113.
d Likely benign.
e Variant observed only in the isoform NM_024742.2.
Distribution of rare alterations was not significant between patients with (5/50) and without (4/26) adrenal involvement (P = .48). Bilateral adrenal gland involvement was found in 2 patients with likely benign ARMC5 variants (c.1641G > A p.A547A and c.2045G > A p.R682Q) and 1 with a benign gene alteration (c.1499C > T p.A500V); 2 patients with alterations in exon 4 of ARMC5 had MEN1 syndrome. We documented unilateral adrenal mass in 2 patients, 1 with pathogenic (c.1223A > G p.Q408R) and 1 with the benign gene variant (c.968G > C p.G323A). Among 4 patients with rare ARMC5 genetic alterations without adrenal involvement, 1 was a carrier of a likely pathogenic gene variant (c.1676C > T p.P559L) and 3 carried likely benign alterations (c.1831C > T p.R611W and c.2045G > A p.R682Q) (see Table 2).
Altogether, 76 germline alterations in the ARMC5 gene were detected in 111 patients (see Table 2), being similarly distributed among patients with 50 of 72 (69.4%) and without 26 of 39 (66.7%) adrenoglandular involvement (P = .53). There was no significant difference in the frequency of germline alterations in the ARMC5 gene between patients with MEN1, 10 of 17 (58.8%), and patients with spNETs, 66 of 94 (70.2%) or between patients with spNETs 41 of 60 (66.7%) and without adrenoglandular involvement 25 of 34 (73.5%) (P = .29). However, significant number of ARMC5 alteration carriers had an adrenal tumor or tumors. Nine of 10 patients with MEN1 had adrenoglandular involvement and 4 of them were bilateral; in 41 of 66 (62.1%) carriers with spNETs, adrenal tumor with bilateral involvement was found in 16 of 41 (39%) cases. We also found 4 somatic, previously unreported, alterations in the ARMC5 gene in MEN1-related tumors (3 NETs, 1 adrenocortical adenoma, and 1 rectal carcinoma). Two of them are predicted as possibly damaging: c.2302G > C p.A768P and c.2666T > A p.L889G (Table 3). Moreover, a novel somatic MEN1 gene missense variant of uncertain significance, c.445G > A p.G149S, was detected in 1 of 3 pancreatic NETs from the same patient (Table 4). No somatic ARMC5 or MEN1 variants were found in the spNET patients.
Table 3.
Unreported somatic alterations in armadillo repeat-containing 5 gene in multiple endocrine neoplasia type 1–related tumors
| Patient ID | Sex | Tumor | Somatic alterations in ARMC5 genea | In silico modeling | ||||
|---|---|---|---|---|---|---|---|---|
| cDNA | Protein | Gene region | Mutation Tasterb | PolyPhen-2c | SIFTd | |||
| P49 | F | R-AcA | c.2302G > C | p.A768P | Exon 6 | –(0.767) | +(0.852) | +(0.030) |
| PNET | c.2302G > C | p.A768P | ||||||
| P80 | F | nPHPT | c.2666T > A | p.L889G | Exon 6 | –(0.840) | +(0.964) | +(0.010) |
| P97 | F | RC | c.1843C > G | p.H615D | Exon 4 | –(0.051) | –(0.003) | –(0.370) |
| LNET | c.2647C > T | p.H883Y | Exon 6 | –(1.000) | –(0.000) | –(0.070) | ||
Gene variant is predicted to be neutral by the bioinformatics tool (–) or probably damaging/damaging (+).
Abbreviations: ARMC5, armadillo repeat-containing 5 gene; F, female; LNET, lung neuroendocrine tumor; nPHPT, nodular hyperplasia of parathyroids; PNET, pancreatic neuroendocrine tumor; R-AcA, right-sided adrenocortical adenoma; RC, rectal carcinoma.
a ARMC5 complementary DNA (cDNA) nucleotide variation RefSeq NM_001105247.1
b Score (in parentheses) is the probability of the prediction; a value close to 1 indicates a high security of the prediction.
c Score (in parentheses) is the probability that the substitution is damaging; a value close to 1 indicates higher possibility.
d Score (in parentheses) is the probability that the substitution is tolerated; values less than 0.05 are considered intolerant.
Table 4.
Somatic alterations in armadillo repeat-containing 5 and multiple endocrine neoplasia type 1 genes in sporadic neuroendocrine and multiple endocrine neoplasia type 1–related tumors
| Patient ID | Germline variants cDNA/protein | Tumor | LOH (microsatellites) | Somatic gene variants cDNA/protein | |||
|---|---|---|---|---|---|---|---|
| ARMC5 a | MEN1 b | Chr. 16p | Chr. 11q | ARMC5 | MEN1 | ||
| MEN1-related NETs (n = 12/18) | |||||||
| P49 | c.1520C > T p.P507L | c.1363_1372del10 | nPHPT | D16S3135 | ND | ND | ND |
| p.V455* | |||||||
| PNET | D16S3024 | D11S4907 | c.2302G > Cd p.A768P | ND | |||
| PYGM | |||||||
| R-AcA | D16S3082 | D11S4907 | c.2302G > Cd p.A768P | ND | |||
| D11S599 | |||||||
| D11S480 | |||||||
| D11S1883 | |||||||
| PGYM | |||||||
| D11S4907 | |||||||
| L-AcA | ND | D11S599 | ND | ND | |||
| D11S480 | |||||||
| D11S1883 | |||||||
| PGYM | |||||||
| P80 | c.1223A > G p.Q408R | c.941delG | nPHP | D16S3082 | D11S4936 | c.2666T > Ad p.L889Q | ND |
| p.R314Rfs*54 | D16S475 | ||||||
| TPNET1 | D16S3082 | D11S480 | ND | c.445G > Ae | |||
| D11S449 | p.G149S | ||||||
| D11S4907 | |||||||
| D11S4908 | |||||||
| D11S913 | |||||||
| PNET2 | D16S3082 | D11S4907 | c.1842C > G, p.L614L | ND | |||
| D11S480 | |||||||
| PYGM | |||||||
| D11S4936 | |||||||
| PNET3 | D16S3082 | D11S4907 | c.1842C > G p.L614L | ND | |||
| D16S475 | D11S480 | ||||||
| D11S449 | |||||||
| D11S4936 | |||||||
| MAH | D16S3082 | D11S1883 | ND | ND | |||
| D11S449 | |||||||
| D11S4936 | |||||||
| P96 | c.1842C > G p.L614L | c.1547_1548insC | hPHPT | D16S3065 | D11S48 | ND | ND |
| K517Efs*14 | |||||||
| D16S3024 | D11S1883 | ||||||
| D16S3134 | D11S4936 | ||||||
| D16S475 | |||||||
| BDC | D16S521 | D11S480 | ND | ND | |||
| D16S3082 | D11S1883 | ||||||
| D16S3134 | D11S4936 | ||||||
| P97 | c.41T > A p.F14Y | c.1063C > G | hPHPT | D16S3065 | D11S449 | ND | ND |
| p.R355W | |||||||
| LNET | D16S3065 | D11S449 | c.2647C > Td p.H883Y | ND | |||
| DCIS | D16S3065 | D11S449 | ND | ND | |||
| CRC | D16S3065 | D11S449 | c.1843C > Gd p.H615D | MLPA+ | |||
| D16S314 | PYGM | ||||||
| P224 | c.2045G > A p.R862Q | c.563C > Te p.P188L | pitNETg | D16S3027 | ND | c.2114C > Tc p.A705V | ND |
| rs199706698 | D16S3065 | ||||||
| D16S3108 | |||||||
| D16S3134 | |||||||
| P445 | c.41T > A p.F14Y | c.563C > Te p.P188L | EAPG | D16S3027 | D11S4907 | ND | ND |
| D11S599 | |||||||
| rs199706698 | pitNETn | ND | D11S4908 | c.*234_*238dupGGCC | ND | ||
| Sporadic NETs (n = 14/18) | |||||||
| P33 | c.508A > G p. I170V | WT | LNET | D16S3108 | ND | ND | ND |
| D16S3134 | |||||||
| MAH | D16S3108 | ND | ND | ND | |||
| P55 | c.508A > G p.I170V | WT | pitNETc | D16S3134 | ND | ND | ND |
| D16S521 | |||||||
| P68 | WT | WT | Ileal | ND | ND | ND | ND |
| P74 | c.1842C > G p.L614L | WT | pitNETc | ND | ND | ND | ND |
| P90 | c.41T > A p.F14Y | WT | pitNETc | ND | ND | c.2114C > Tc p.A705V | ND |
| P92 | c.*234_*238dupGGCCT | WT | pitNETc | ND | ND | ND | ND |
| P125 | c.2114C > Tc p.A705V | WT | pitNETg | D16S3027 | ND | ND | ND |
| D16S3134 | |||||||
| D16S475 | |||||||
| P150 | c.475 + 58A > G | WT | PNET | ND | ND | ND | ND |
| L-AcA | ND | ND | ND | ND | |||
| P178 | c.41T > A p.F14Y | WT | LNET | D16S3108 | ND | c.2114C > Tc p.A705V | ND |
| P184 | c.2114C > Tc p.A705V; | WT | pitNETn | ND | ND | ND | ND |
| c.*234_*238dupGGCCT | CRC | ND | ND | ND | ND | ||
| c.1842C > G p.L614L | R-AcA | ND | ND | ND | ND | ||
| P187 | c.1831C > T p.R611W; c.1842C > G p.L614L |
WT | pitNETn | D16S3135 | ND | ND | ND |
| P206 | c.2114C > Tc p.A705V | WT | pitNETc | ND | ND | ND | ND |
| P295 | WT | WT | pitNETn | ND | ND | ND | ND |
| P359 | c.2114C > Tc p.A705V | WT | pitNETn | D16S3024 | ND | ND | ND |
Abbreviations: ARMC5, armadillo repeat-containing 5; BDC breast ductal carcinoma; Chr., chromosome; CRC, colorectal carcinoma; DCIS, breast ductal carcinoma in situ; EAPG, extra-adrenal paraganglioma; hPHPT, hyperplasia of parathyroids; LNET, lung neuroendocrine tumor; LOH, loss of heterozygosity; MAH, macronodular hyperplasia of adrenal cortex; MEN1, multiple endocrine neoplasia type 1; ND, not detected; NETs, neuroendocrine tumors; nPHPT, nodular hyperplasia of parathyroids; P, patient; pitNET, pituitary neuroendocrine tumor (n, nonfunctioning; g, gonadotroph; PNET, pancreatic neuroendocrine tumor; c, corticotroph); RC, rectal carcinoma; R/L-AcA right- or left-sided adrenocortical adenoma; rs, reference single-nucleotide variation (SNV); WT, wild-type.
a ARMC5 complementary DNA (cDNA) nucleotide variation RefSeq NM_001105247.1; ARMC5 protein RefSeq NP_001098717.1.
b MEN1 cDNA nucleotide variation RefSeq NM_130799.2.
c Gene variant found only in the ARMC5 isoform NM_024742.2.
d Unreported ARMC5 gene variants detected in tumor tissues.
e Variant of uncertain significance.
Alterations in the multiple endocrine neoplasia type 1 gene
We found 16 different variants of MEN1 in 17 patients with clinically diagnosed MEN1. Two of these alterations were intronic, 5′-UTR variant c.-6G > A (rs768088337) and a splice donor variant c.783 + 1G > A (rs794728652), both with an MAF of less than 1% according to the GnomAD database. All other MEN1 gene alterations were located within the coding region of the gene. Eight of them were single nucleotide changes resulting in a missense substitution in the protein. One of the variants, c.658T > G p.W220G, has not been reported before; 5 are known variants: c.125G > A p.G42N, c.416A > G p.H139R, c.494G > C p.C165Y, c.949C > T p.H317Y, and c.1063C > T p.R355W, and 2 patients had the same nonsynonymous change, c.563C > T p.P88L with an MAF of less than 1%. Two nonsense mutations, a new one caused by deletion c.1363_1372delGTGCGCATAG p.V455*, and one registered in public databases, c.1579C > T p.R527*, were detected. The patient with the nonregistered variant had classical clinical presentation of MEN1 syndrome, primary hyperparathyroidism (PHPT), pitNET, and multiple pancreatic NETs (PNETs). We found 4 unregistered frameshift variants in public databases caused by deletion or insertion with stop codon in the new reading frame. Two of them, c.941delG p.R314Rfs*54 and c.984_990delCTGTCGC p.H328Qfs*38, were associated with PHPT, pitNET, and multiple PNETs; 2 patients had a c.1547_1548insC p.K517Efs*14 mutation with PHPT and pitNET. One patient had only PHPT caused by a c.1602_1618delGGCTCAGGTGCCAGCAC p.A535Hfs*16 mutation. All germline alterations found in the MEN1 gene were predicted as damaging, according to our in silico analysis, including the 5′ UTR variant.
Loss of heterozygosity at chromosome 16p and 11q regions in sporadic neuroendocrine tumors and multiple endocrine neoplasia type 1–related tumors
LOH studies were performed in 36 tumor tissue samples available from 20 patients, 14 obtained from patients with spNETs and 6 from patients with MEN1 syndrome. Altogether, of 36 available tumors, 26 (72.2%) were NETs, 6 (16.6%) were adrenocortical tumors, and we analyzed 4 concomitant adenocarcinomas (2 colorectal and 2 breast) obtained from patients with genetically confirmed MEN1 (see Table 4). All but 2 of 36 tumors were obtained from carriers of the germline ARMC5 alteration. Neither of 2 tumors provided from wild-type ARMC5 carriers with spNETs had LOH, whereas AIs at the chromosomal 16p region co-occurred with germline ARMC5 alterations in 23 of 34 (67.6%) tumors. In 11 of 34 (32.4%) tumors we could not verify the presence of AI.
Hemizygosity was detected at the chromosome 16p region in 16 of 18 (88.9%) tumors obtained from 6 patients with concurrent germinal alterations in the MEN1 gene (4 hyperplastic parathyroid glands, 4 pancreatic NETs, 1 lung NET, 1 extra-adrenal paraganglioma, 1 gonadotroph pituitary adenoma, 1 adrenocortical adenoma, 1 adrenal nodule from a patient with hyperplasia, 2 breast cancers, and 1 colorectal carcinoma). Coexistence of LOH at the chromosome 11q region was demonstrated in 14 of 16 (87.5%) tumors (see Table 4). In spNETs coexistence of AIs at the chromosome 16p region with a germinal ARMC5 alteration was found in 7 of 18 (38.9%) tumors from 14 patients (2 lung NETs, 4 pituitary adenomas, and 1 nodule characterized as MAH) (see Table 4).
The majority of ARMC5 germline variants associated with LOH in these tumors were nondamaging (see Table 4). One rare ARMC5 gene variant, c.1831C > T p.R611W, predicted to likely be benign was found in a patient with sporadic pituitary adenoma, whereas the other 2, pathogenic c.1223A > G p.Q408R and likely benign, c.2045G > A p.R862Q, were co-present with germline alterations in the MEN1 gene, c.941delG p.R314Rfs*54 and c.563C > T p.P188L (see Table 4). Inactivation of both ARMC5 alleles due to simultaneous presence of a predicted to be damaging germline gene variant and LOH at the chromosome 16p region was documented in 7 of 23 (30.4%). Six of these tumors were obtained from 2 patients with MEN1, and 5 of them concomitantly displayed biallelic inactivation of the MEN1 gene (germline mutation plus LOH) (see Table 4). The occurrence of LOH in the tumor tissue was independent of the germline ARMC5 variant by Fisher exact test (P = .13). On the other hand, the probability of LOH at the chromosomal 16p region was increased among carriers of an inactivating MEN1 variant with a risk ratio of 2.28; 0.95 CI, 1.25 to 4.17 (P = .005).
Armadillo repeat-containing 5 protein and menin expression in multiple endocrine neoplasia type 1–related tumors
Immunoreactivity of the ARMC5 protein was detected in all peritumor as well as tumor tissues (Table 5). In agreement with the documented haploinsufficiency of the ARMC5 gene in analyzed tissues, the cytoplasmic immunoreactivity score for the ARMC5 protein was significantly lower in tumoral compared to peritumoral tissues (3.1 ± 1.8 vs 5.2 ± 1.4, P = .02) (see Table 5). Similarly, MEN1 was underexpressed in tumors compared to peritumor controls (4.0 ± 3.3 vs 7.7 ± 1.8, P = .01). It is important to note that one patient (ID P80) carried pathogenic germline variants in the ARMC5 c.1223A > G p.Q408R and MEN1 c.941delG p.R314Rfs*54 genes, but also had several NETs and an adrenocortical tumor. LOH at the chromosome 16p and 11q regions were documented in all tumors, including hyperplastic parathyroid tissue (see Table 4). Although expression of the ARMC5 protein is undetectable in a hyperplastic nodule of the adrenal cortex (see Table 5 and Fig. 1; row 6), it was present in 3 PNETs within the range of very low level of expression to the half-normal (see Table 5 and Fig. 1; lines 3, 4, and 7). In hyperplastic parathyroid tissue the expression of ARMC5 was similar to the surrounding peritumor tissue.
Table 5.
Armadillo repeat-containing 5 and MENIN immunostaining in tumors with allelic imbalance at chromosome 16p in multiple endocrine neoplasia type 1–related tumors
| Patient ID | Tumor | ARMC5 expression (cytoplasm) | MENIN expression (nucleus) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Tumor tissue | Peritumoral tissue | Tumor tissue | Peritumoral tissue | ||||||
| INT | +Cell, % | INT | +Cell, % | INT | +Cell, % | INT | +Cell, % | ||
| P49 | hPHPTa | 1 | 80 | 1 | 100 | 2 | 60 | 3 | 100 |
| P49 | PNET | 1 | 80 | 1 | 100 | 1 | 40 | 2 | 100 |
| P80 | PNET | 1 | 15 | 2 | 100 | 1 | 3 | 2 | 100 |
| P80 | PNET | 1 | 30 | 2 | 100 | 1 | 10 | 2 | 100 |
| P80 | nPHPT | 2 | 90 | 2 | 100 | 2 | 20 | 3 | 100 |
| P80 | MAH | 0 | 0 | 1 | 100 | 2 | 20 | 3 | 100 |
| P80 | PNET | 2 | 60 | 2 | 100 | 3 | 70 | 3 | 100 |
| P97 | DCIS | 2 | 50 | 2 | 100 | 3 | 90 | 3 | 100 |
| P97 | LNET | 1 | 30 | 2 | 100 | 1 | 80 | 3 | 100 |
| P97 | RC | 1 | 80 | 2 | 100 | 1 | 50 | 2 | 100 |
| P97 | hPHPT | 2 | 70 | 2 | 100 | 3 | 90 | 3 | 100 |
Patient ID P80 carries pathogenic germline alterations in ARMC5 gene, c.1223A > G p.Q408R, and MEN1 gene c.941delG p.R314Rfs*54.
Abbreviations: ARMC5, armadillo repeat-containing 5; DCIS, breast ductal carcinoma in situ; hPHPT, hyperplasia of parathyroids; ID, identification; INT, intensity of staining; LNET, lung neuroendocrine tumor; MAH, macronodular hyperplasia of adrenal cortex; nPHPT, nodular hyperplasia of parathyroids; PNET, pancreatic neuroendocrine tumor; RC, rectal carcinoma.
a Tumor without loss of heterozygosity at chromosome 11q.
Figure 1.
Armadillo repeat-containing 5 (ARMC5) expression in tumor tissues. Cytoplasmic ARMC5 expression in tumor tissue with documented loss of heterozygosity (LOH) at chromosome 16p (column A with magnification of 12.5×, and column B with magnification of 200×). ARMC5 expression in peritumoral tissues (column C; 200×). Nuclear coexpression of MENIN in tumors (column D; 200×). All tumors but nodular parathyroid hyperplasia (nPHPT) displayed LOH at chromosome 11q locus. The genetics of these tumors are presented in Table 3 (ID P49, P80, and P97). Nodular parathyroid hyperplasia delineated by arrowheads in rows 1 and 5; pancreatic neuroendocrine tumor (PNET) delineated with arrowheads in rows 2, 3, 4, and 7; macronodular adrenocortical hyperplasia (MAH), row 6 (arrowheads); ductal breast carcinoma in situ (DCIS), (arrowheads in row 8); lung neuroendocrine tumor (LNET), (arrowheads, row 9); colorectal carcinoma (CRC), (row 10); parathyroid hyperplasia (hPHPT) in row 11.
Discussion
In this study, we report frequent germline alterations in the ARMC5 gene in patients with spNETs and genetically confirmed MEN1; we also show common allelic losses of the chromosome 16p ARMC5 locus. Contrary to what motivated us to engage in this investigation, the presence of an ARMC5 variant did not depend on adrenoglandular involvement. On the other hand, ARMC5 was frequently altered, more frequently than expected from the population data. All detected alterations in the ARMC5 gene (except for c.1007A > G p.D336G) were known variants. Two of 16 different ARMC5 variants that were identified (c.1223A > G p.Q408R and c.1676C > T p.P559L) were predicted to be probably damaging or damaging missense gene variants by in silico analysis. In addition, LOH for the chromosome 16p ARMC5 locus was more frequent among MEN1-related tumors (88.9%) than spNETs (38.9%), and ARMC5 immunostaining (albeit decreased) was not completely absent and this is of unclear significance.
Are our findings consistent with ARMC5’s role as a tumor suppressor gene and the Knudson hypothesis (14)? Haploinsufficiency for ARMC5 (which was found in most of our tumors) could contribute to a number of other tumorigenic pathways; it is also possible that additional somatic ARMC5 defects were present but not detected, as is the case in about 10% of examined tumors in other settings (8, 15). This is not necessarily in contrast with extensive genetic variability of the second ARMC5 allele, such as has been observed in benign adrenocortical nodules obtained from a single patient with PMAH and a single germline pathogenic mutation (16). Germline alterations in the ARMC5 gene confer susceptibility for subsequent somatic events as demonstrated for some other neoplasms (17, 18). Our results provide molecular evidence for possible effects of ARMC5 as a tumor suppressor gene in other endocrine tissues beyond the adrenal cortex (5). It is important to recognize that the number of spNETs and genetically confirmed MEN1 patients presenting with ARMC5 damaging variants is limited. Thus, additional cohorts should be investigated to conclude definitively the role of ARMC5 as a modifier of the phenotype in patients with spNETs or MEN1.
In addition, haploinsufficiency both for the MEN1 and ARMC5 genes in the tissues where we found them occurring concurrently may further enhance their effects on genomic integrity (19, 20). The frequency of LOH for chromosomes 16p and 11q was the same; each one was found in 16 of 18 (88.9%) tumors. Two predicted to be damaging germline variants in the ARMC5 gene (c.1223A > G p.Q408R and c.2045G > A p.R862Q) were also accompanied by LOH for chromosome 16p in 6 tumors, whereas inactivation of both MEN1 alleles was confirmed in 16 tumor tissues. Coexistence of biallelic inactivation of both the MEN1 and ARMC5 genes was demonstrated in 5 tumor tissues. The relative risk of detecting chromosomal 16p losses increased 2-fold when a germline mutation in the MEN1 gene was present. Indeed, chromosome 16p losses have been seen before in approximately 25% of MEN1-related duodenopancreatic NETs (21, 22). Interestingly, in our cases, we did not find complete chromosomal 11 or 16 absence, suggesting that LOH for 11q and 16p, respectively, was the result of locus-specific changes, rather than mitotic nondisjunction (23). Moreover, the variable expression of ARMC5 in different MEN1-related tumors from the same patient (ID P80) with confirmed LOH at chromosomes 16p and 11q and established biallelic inactivation of ARMC5 in an adrenocortical nodule might point to tissue-specific effects.
The main limitation of this study was the small number of tumor samples obtained from noncarriers of ARMC5 germline alterations. This deficiency did not allow us to draw firm conclusions about the relationship between ARMC5 germline variants and the susceptibility for somatic events that we documented. Larger series of patients are needed to confirm the role of ARMC5 germline variants in development of particular NET types.
In summary, germline ARMC5 alterations among patients with spNETs or MEN1 are frequent. Chromosomal 16p losses in tumors obtained from these patients are also common and occur with predicted to be damaging as well as nondamaging ARMC5 germline variants. It is possible that in this study, the prevalence of adrenal tumors was overestimated because 19 patients had adrenocorticotropin-dependent Cushing syndrome, and chronic adrenocorticotropin excess leads to bilateral adrenal hyperplasia. Nevertheless, despite its limitations, this study does suggest that ARMC5 variants may play a modifier role in the phenotype of patients with spNETs or MEN1; clearly, larger cohorts need to be studied and the ARMC5 variants identified here need to be studied functionally in NET tissue-derived cell lines for their functional effects and/or in cells with menin deficiency.
Acknowledgments
Financial Support: This work was supported by the Ministry of Science of Republic of Serbia (Grants III41009 and 175033) and, in part, by the Intramural Program of the Eunice Kennedy Shriver National Institute of Child Health & Human Development, National Institutes of Health, Bethesda, Maryland, USA (Grant Z01-HD008920-01).
Glossary
Abbreviations
- AI
allelic imbalance
- ARMC5
armadillo repeat-containing 5
- LOH
loss of heterozygosity
- MAF
minor allele frequency
- MEN1
multiple endocrine neoplasia type 1
- MLPA
multiplex ligation-dependent probe amplification
- NET
neuroendocrine tumor
- PHPT
primary hyperparathyroidism
- pitNET
pituitary neuroendocrine tumor
- PMAH
primary macronodular adrenal hyperplasia
- PNET
pancreatic neuroendocrine tumor
- spNET
sporadic neuroendocrine tumor
Additional Information
Disclosure Summary: C.A.S. holds patents on the function of the PRKAR1A, PDE11A, and GPR101 genes and related issues; his laboratory has also received research funding on the GPR101 gene, abnormal growth hormone secretion, and its treatment by Pfizer, Inc; F.R.F. holds a patent on the GPR101 gene and/or its function. The other authors have nothing to disclose.
Data Availability
All data generated or analyzed during this study are included in this published article or in the data repositories listed in “References.”
References
- 1. Assié G, Libé R, Espiard S, et al. ARMC5 mutations in macronodular adrenal hyperplasia with Cushing’s syndrome. N Engl J Med. 2013;369(22):2105-2114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Faucz FR, Zilbermint M, Lodish MB, et al. Macronodular adrenal hyperplasia due to mutations in an armadillo repeat containing 5 (ARMC5) gene: a clinical and genetic investigation. J Clin Endocrinol Metab. 2014;99(6):E1113-E1119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Alencar GA, Lerario AM, Nishi MY, et al. ARMC5 mutations are a frequent cause of primary macronodular adrenal hyperplasia. J Clin Endocrinol Metab. 2014;99(8):E1501-E1509. [DOI] [PubMed] [Google Scholar]
- 4. Gagliardi L, Schreiber AW, Hahn CN, et al. ARMC5 mutations are common in familial bilateral macronodular adrenal hyperplasia. J Clin Endocrinol Metab. 2014;99(9):E1784-E1792. [DOI] [PubMed] [Google Scholar]
- 5. Assié G, Libé R, Espiard S, et al. ARMC5 mutations in macronodular adrenal hyperplasia with Cushing’s syndrome. N Engl J Med. 2013;369:2105-2114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Espiard S, Drougat L, Libé R, et al. ARMC5 mutations in a large cohort of primary macronodular adrenal hyperplasia: clinical and functional consequences. J Clin Endocrinol Metab. 2015;100(6):E926-E935. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Berthon A, Faucz F, Bertherat J, Stratakis CA. Analysis of ARMC5 expression in human tissues. Mol Cell Endocrinol. 2017;441:140-145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Elbelt U, Trovato A, Kloth M, et al. Molecular and clinical evidence for an ARMC5 tumor syndrome: concurrent inactivating germline and somatic mutations are associated with both primary macronodular adrenal hyperplasia and meningioma. J Clin Endocrinol Metab. 2015;100(1):E119-E128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kanakis G, Kamp K, Tsiveriotis K, et al. The prevalence and relevance of adrenal masses in patients with sporadic gastroenteropancreatic neuroendocrine tumours (GEP-NET). Clin Endocrinol (Oxf). 2013;78(6):950-956. [DOI] [PubMed] [Google Scholar]
- 10. Gatta-Cherifi B, Chabre O, Murat A, et al. Adrenal involvement in MEN1. Analysis of 715 cases from the Groupe d’etude des Tumeurs Endocrines database. Eur J Endocrinol. 2012;166(2):269-279. [DOI] [PubMed] [Google Scholar]
- 11. Manickam P, Guru SC, Debelenko LV, et al. Eighteen new polymorphic markers in the multiple endocrine neoplasia type 1 (MEN1) region. Hum Genet. 1997;101(1):102-108. [DOI] [PubMed] [Google Scholar]
- 12. Klein M, Vignaud JM, Hennequin V, et al. Increased expression of the vascular endothelial growth factor is a pejorative prognosis marker in papillary thyroid carcinoma. J Clin Endocrinol Metab. 2001;86(2):656-658. [DOI] [PubMed] [Google Scholar]
- 13. Richards S, Aziz N, Bale S, et al. 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. Genet Med. 2015;17:405-424. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Knudson AG Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci U S A. 1971;68(4):820-823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Berthon A, Faucz FR, Espiard S, Drougat L, Bertherat J, Stratakis CA. Age-dependent effects of Armc5 haploinsufficiency on adrenocortical function. Hum Mol Genet. 2017;26(18):3495-3507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Correa R, Zilbermint M, Berthon A, et al. The ARMC5 gene shows extensive genetic variance in primary macronodular adrenocortical hyperplasia. Eur J Endocrinol. 2015;173(4):435-440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Lu Y, Ek WE, Whiteman D, et al. ; Q-MEGA and AMFS Investigators; ANECS-SEARCH; UKOPS-SEARCH; BEACON Consortium . Most common ‘sporadic’ cancers have a significant germline genetic component. Hum Mol Genet. 2014;23(22):6112-6118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. García-Nieto PE, Schwartz EK, King DA, et al. Carcinogen susceptibility is regulated by genome architecture and predicts cancer mutagenesis. EMBO J. 2017;36(19):2829-2843. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Fang M, Xia F, Mahalingam M, Virbasius CM, Wajapeyee N, Green MR. MEN1 is a melanoma tumor suppressor that preserves genomic integrity by stimulating transcription of genes that promote homologous recombination-directed DNA repair. Mol Cell Biol. 2013;33(13):2635-2647. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Hu Y, Lao L, Mao J, et al. Armc5 deletion causes developmental defects and compromises T-cell immune responses. Nat Commun. 2017;8:13834. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Kytölä S, Mäkinen MJ, Kähkönen M, Teh BT, Leisti J, Salmela P. Comparative genomic hybridization studies in tumours from a patient with multiple endocrine neoplasia type 1. Eur J Endocrinol. 1998;139(2):202-206. [DOI] [PubMed] [Google Scholar]
- 22. Hessman O, Skogseid B, Westin G, Akerström G. Multiple allelic deletions and intratumoral genetic heterogeneity in MEN1 pancreatic tumors. J Clin Endocrinol Metab. 2001;86(3):1355-1361. [DOI] [PubMed] [Google Scholar]
- 23. Thiagalingam S, Laken S, Willson JK, et al. Mechanisms underlying losses of heterozygosity in human colorectal cancers. Proc Natl Acad Sci U S A. 2001;98(5):2698-2702. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data generated or analyzed during this study are included in this published article or in the data repositories listed in “References.”