Primary hyperparathyroidism (PHPT) is a frequently encountered endocrine disorder with an increasing incidence worldwide (1, 2). PHPT is usually diagnosed by routine laboratory tests in asymptomatic patients or patients with normocalcemic PHPT (2, 3). The disease results from a single parathyroid adenoma (PTA) in 80% of cases and from multiple glandular adenomas or hyperplasia (MGH) in 20% of cases; parathyroid carcinoma (PTC) is rare, with an incidence of less than 1% (4). Surgery is the only curative treatment (2, 5, 6). Germline mutations may be identified in up to 10% of cases (3), warranting genetic evaluation in young patients below 30 years of age with PHPT, or patients with a familial history of isolated PHPT (FIHP), or presenting clinical and/or laboratory features suggesting a syndromic disease (2): familial hypocalciuric hypercalcemia (FHH) subtypes due to mutation in the calcium-sensing receptor (CASR), the guanine nucleotide-binding protein subunit alpha11 (GNA11) or the adaptor protein complex-2 subunit sigma (AP2S1), or tumoral syndromes such as Multiple Endocrine Neoplasia (MEN) type 1, 2A, and 4, PHPT-Jaw Tumor Syndrome associated with various genetic mutations.

Most cases of FIHP have no genetic diagnosis, but in some cases, variants in the GCM2 encoding for the chorion-specific transcription factor GCM2 protein may be found (3, 7), although their penetrance and pathogenetic significance remain unclear (3, 8, 9). The GCM2 gene (OMIM #603716) localizes to 6p24.2, encodes a key transcription factor for 506 amino acids in parathyroid embryogenesis and development, which remains expressed at the adult stage (10). It has been shown that gcm2-null mice develop parathyroid hypoplasia and hypoparathyroidism (11), and that homozygous and heterozygous germline inactivating GCM2 pathogenic variants in humans lead to hypoparathyroidism (12, 13). Other variants are associated with hyperparathyroidism (14, 15), with an overall higher frequency of GCM2 variants in sporadic and familial PHPT than in the general population (8, 16). In 2017, Guan et al. (17) reported that the GCM2 p.Tyr394Ser rs142287570 variant was found in 41% and 27% of Ashkenazi Jews (AJ) with FIHP kindreds and sporadic PHPT in their referral cohort, respectively, and at a lower prevalence in other European origin individuals.

The aim of the present study was to evaluate the presence of the GCM2 p.Tyr394Ser variant in Israeli patients being evaluated for potential or definite familial PHPT, and to describe the clinical and laboratory presentation as well as the management of patients with the GCM2 p.Tyr394Ser variant.

Patients diagnosed with PHPT with no clinical or laboratory findings suggestive of MEN1, MEN2A, or Jaw-Tumor Syndrome were referred to our specialized center for genetic evaluation. The inclusion criteria for genetic study enrollment were suspected FHH based on low urinary fraction excretion of calcium in the context of elevated blood calcium and/or PTH levels, with or without familial history of hyperparathyroidism, patients with a diagnosis of familial PHPT based on biochemical tests and a first-degree familial history of hyperparathyroidism, young age (<30 years old), and failed parathyroidectomy with persistent disease.

Following informed consent, DNA was extracted from the leukocytes and stored for future evaluation. Until 2017, Sanger sequencing using adapted primers was performed first only for CASR (NM_000388.4) (18) and then completed after 2013 (19, 20) for the hotspot AP2S1 pathogenic rs397514499 variant (p.Arg15Leu) and GNA11 (NM_002067.5), as previously described (21). In this cohort of patients with an initial normal genetic screening, retrospective focused analysis of the GCM2 (NM_004752.4) p.Tyr394Ser was performed by Sanger sequencing. Patients addressed for molecular studies since 2018 were evaluated using a genetic panel for PHPT, including CASR, AP2S1, GNA11, CDC73, MEN1, CDKN1A, CDKN1B, CDKN2B, CDKN2C, and RET. The test was performed using the xGen Exome Research Panel IDT-V2 kit in combination with xGen Human mtDNA Research Panel v1.0 on NovaSeq 6000. Fast Q data were computer-analyzed. The changes were compared to those in the human genome (hg19). Common changes (MAF >1% in gnomAD) were removed, as were non-coding synonymous changes (except for intronic changes that were less than eight bases away from a splice site) or synonymous changes that were more than three bases distant from splice sites. Quantitative change analysis was performed using the CNV-finder (software written in-house) and CNV KIT software (PMID: 27100738).

A specific review of the GCM2 p.Tyr394Ser variant in the PHPT genetic panel was performed in patients addressed for evaluation since 2018. The presence of the GCM2 p.Tyr394Ser variant was assessed using our local genome database, which included all genome tests performed at our facility for any medical reason. Ethnicity was assessed using a questionnaire and/or anamnesic assessment. Specifically, Ashkenazi origin was self-reported.

Overall, 42 probands from distinct kindreds were addressed for genetic evaluation of PHPT at our center since 2009 ( Figure 1 ). Two patients had no genetic evaluation because informed consent was not obtained. A total of 31 patients underwent an evaluation before and eight after 2018, respectively, and one patient was specifically addressed to test the GCM2 p.Tyr394Ser variant as she was of Ashkenazi origin with FIHP without any syndromic feature (see Figure 1 describing the chart of patients’ selection).

We identified 11 families (11/42, 26.2%) with a relevant genetic variant ( Figure 1 ): one with a CDKN2B mutation, two with hotspot AP2S1 p.Arg15Leu, and eight with CASR mutations.

In those evaluated before 2018, 23 patients from different kindreds remained without a genetic diagnosis, but one patient did not sign an informed consent form to conserve the DNA sample to perform additional specific genetic tests related to hyperparathyroidism. Sanger sequencing of the GCM2 p.Tyr394Ser variant was performed on 22 patients who provided informed consent.

Among the eight patients addressed since the beginning of 2018 who had an appropriate genetic panel for PHPT, three had a positive genetic variant ( Figure 1 ). In the remaining five patients, an analysis of the GCM2 p.Tyr394Ser variant was performed. Overall, 28 Israeli patients with FIHP, young age, or failed surgery underwent a targeted evaluation of the GCM2 p.Tyr394Ser variant ( Figure 1 ). Of these, 15/28 (53.5%) or 25/42 (59.5%) of the full original cohort comprised of different kindreds, were from Ashkenazi origin, while 13/28 (46.4%) and 17/42 (40.5%) of the full cohort were from non-Ashkenazi backgrounds. Among the tested individuals, 3/28 (10.7%) patients, or 3/42 (7.1%) of the fully tested and untested cohorts, were found to be heterozygous for the GCM2 p.Tyr394Ser variant. Overall, among the 15 Ashkenazi patients with clinical reasons to diagnose or suspect a familial state of PHPT and who agreed to our testing protocol, the prevalence of the GCM2 p.Tyr394Ser variant was 3/15 (20%), which represented 3/25 (12%) of all the patients from Ashkenazi origin, two of which were identified within the context of the retrospective analysis, and one individual was enrolled specifically for screening for the GCM2 p.Tyr394Ser ( Figure 1 , Table 1 , Cases 1–3). Four other patients who were identified with FHH and a disease-causing pathogenic variant in CASR or AP2S1 were also tested for the GCM2 p.Tyr394Ser variant to identify a possible double hit. This was not observed in any of these patients ( Table 1 ).

The phenotype of our patients with the GCM2 p.Tyr394Ser variant presented with asymptomatic mild to moderate hypercalcemia, and are low urinary calcium excretion fraction was below 1% or below 2% and may thus overlap with the phenotype observed in FHH. However, in these cases, a classic mutation in CASR should be suspected first, as these mutations seem more frequent than GCM2 variants in this context. Indeed, in our cohort, based on the same inclusion criteria for genetic screening as utilized above, we found most frequently a CASR variant: 8/42 (19%) index-cases from non-related kindreds ( Figure 1 ), whereas we found GCM2 p.Tyr394Ser variant in 3/42 (7.1%) and hotspot AP2S1 p.Arg15Leu variant in 2/42 (4.7%) addressed patients. We also found one case of CDKN2B mutation (1/42, 2.3%). We found no case of the GNA11 variant, which is known to be rare (22). We also observed that more than one PTA or glandular hyperplasia was found in patients with the GCM2 p.Tyr394Ser variant. However, in contrast to cases with FHH due to pathogenic variants in CASR, AP2S1, or GNA11, surgery may cure the disease (Case 3) unless hyperplastic glands or other parathyroid adenomas are missed during the first surgery in cases of MGH (Case 2). Our clinical findings expand and tend to contrast with a message from previously published descriptions of more severe disease associated with GCM2 variants; for example, a previous genotype–phenotype correlation study in 18 patients with GCM2 variants compared to 457 patients with sporadic PHPT suggested higher preoperative levels of PTH, higher rate of multiglandular parathyroid disease (78% versus 14.3%), a lower cure rate (86% versus 99%), and a higher risk of PTC (5% versus 0%) (23) in patients with GCM2 variants; another study described rare cases of PTC in the context of GCM2 variants (24). Finally, patients with GCM2 variants and FIHP as compared to ones without GCM2 variants had significantly larger resected parathyroid glands (15). We showed that hypercalcemia could be successfully controlled with cinacalcet in one patient (Case 1) with the GCM2 p.Tyr394Ser rs142287570 variant, and that resection of more than one gland may be necessary to cure the disease as multiple parathyroid glands can be involved, otherwise leading to failed surgery.

In this study of a highly selected cohort of Israeli patients with PHPT addressed for genetic evaluation, we found a heterozygous GCM2 p.Tyr394Ser variant in 3/28 (10.7%) patients. This represents 3/42 (7.1%) of the full cohort with different kindreds. Among the AJ patients, 3/15 (20%) or 3/25 (12%) were included in the full cohort. Based on these findings, a recent abstract described the presence of the GCM2 p.Tyr394Ser variant in 2/40 (5%) self-reported AJ patients with PHPT, including one familial case of PHPT, whereas it was found in 3/200 (1.5%) healthy AJ controls (25).

This shows an over-representation of the evaluated variant compared to the general Israeli population, as evaluated in the presence of the variant in our local genome database (141/15,407, 0.98%). However, this prevalence would be higher if calculated among Ashkenazi Jews tested only (approximately 1.8%). Data available from gnomAD v.4.0.0 (https://gnomad.broadinstitute.org/variant/6-10874335-T-G) show an allele frequency of the variant of 1.26% in AJ (26). These observations, especially in an Israeli population where AJ patients with FIHP can occasionally (and apparently more than in the general population) carry the GCM2 p.Tyr394Ser variant, warrant further and larger studies to evaluate its penetrance and whether screening for its presence might have a useful role in clinical management. Moreover, the exact pathophysiology explaining how the GCM2 p.Y394S variant acts as a proto-oncogene leading to familial PHPT needs to be further studied, as a mouse knock-in model inserting this variant using CRISPR/Cas9 gene editing technology did not reveal any serum calcium and PTH changes, nor histological, parathyroid gland size, or cell proliferation changes in genetically modified mice (27).

Our report has several limitations in addition to the selection bias in the way our patients were recruited, which focused on patients highly suspected of genetic cause for PHPT: First, it included only a small number of patients with GCM2 p.Tyr394Ser rs142287570 variant; second, as only one specific GCM2 variant was sequenced in 23 of the included patients (22 from a retrospective analysis of samples obtained before 2018 and one prospectively screened, Figure 1 ), other GCM2 variants may have been missed, as well as other disease-causing mutations in other genes. Finally, the genetic evaluation we performed could not allow haplotype analysis of the patients with the p.Tyr394Ser rs142287570 to classify the variant as a founder mutation in AJ with PHPT.

In summary, in contrast with previous observations, the phenotype associated with the GCM2 p.Tyr394Ser variant in AJ with FIHP may be mild, sometimes mimicking FHH. The clinical spectrum and clinical utility of screening for this variant warrant further investigation.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

This study involving humans were approved by Hadassah Local Institutional Review Committee. The Study was conducted in accordance with the local legislation and institutional requirements. Written Informed Consent for participation to this study was provided by all participants or by the participants' legal guardians/next of kin.

AS: Conceptualization, Formal Analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing – original draft, Writing – review & editing. SS: Data curation, Investigation, Writing – review & editing. RP: Data curation, Investigation, Writing – review & editing. HM: Data curation, Investigation, Writing – review & editing. RD: Data curation, Investigation, Writing – review & editing. VM: Data curation, Validation, Writing – original draft, Writing – review & editing.