Genetic profile of a large Spanish cohort with hypercalcemia

Introduction The disorders in the metabolism of calcium can present with manifestations that strongly suggest their diagnosis; however, most of the time, the symptoms with which they are expressed are nonspecific or present only as a laboratory finding, usually hypercalcemia. Because many of these disorders have a genetic etiology, in the present study, we sequenced a selection of 55 genes encoding the principal proteins involved in the regulation of calcium metabolism. Methods A cohort of 79 patients with hypercalcemia were analyzed by next-generation sequencing. Results The 30% of our cohort presented one pathogenic or likely pathogenic variant in genes associated with hypercalcemia. We confirmed the clinical diagnosis of 17 patients with hypocalciuric hypercalcemia (pathogenic or likely pathogenic variants in the CASR and AP2S1 genes), one patient with neonatal hyperparathyroidism (homozygous pathogenic variant in the CASR gene), and another patient with infantile hypercalcemia (two pathogenic variants in compound heterozygous state in the CYP24A1 gene). However, we also found variants in genes associated with primary hyperparathyroidism (GCM2), renal hypophosphatemia with or without rickets (SLC34A1, SLC34A3, SLC9A3R1, VDR, and CYP27B1), DiGeorge syndrome (TBX1 and NEBL), and hypophosphatasia (ALPL). Our genetic study revealed 11 novel variants. Conclusions Our study demonstrates the importance of genetic analysis through massive sequencing to obtain a clinical diagnosis of certainty. The identification of patients with a genetic cause is important for the appropriate treatment and identification of family members at risk of the disease.


Introduction:
The disorders in the metabolism of calcium can present with manifestations that strongly suggest their diagnosis; however, most of the time, the symptoms with which they are expressed are nonspecific or present only as a laboratory finding, usually hypercalcemia.Because many of these disorders have a genetic etiology, in the present study, we sequenced a selection of 55 genes encoding the principal proteins involved in the regulation of calcium metabolism.
Methods: A cohort of 79 patients with hypercalcemia were analyzed by nextgeneration sequencing.
Results: The 30% of our cohort presented one pathogenic or likely pathogenic variant in genes associated with hypercalcemia.We confirmed the clinical diagnosis of 17 patients with hypocalciuric hypercalcemia (pathogenic or likely pathogenic variants in the CASR and AP2S1 genes), one patient with neonatal hyperparathyroidism (homozygous pathogenic variant in the CASR gene), and another patient with infantile hypercalcemia (two pathogenic variants in compound heterozygous state in the CYP24A1 gene).However, we also found variants in genes associated with primary hyperparathyroidism (GCM2), renal hypophosphatemia with or without rickets (SLC34A1, SLC34A3, SLC9A3R1, VDR, and CYP27B1), DiGeorge syndrome (TBX1 and NEBL), and hypophosphatasia (ALPL).Our genetic study revealed 11 novel variants.

Introduction
The third cause of consultation with the endocrinologist, after diabetes and thyroid diseases, is the disorders in the metabolism of calcium.A dynamic balance between intestinal absorption, bone resorption, and renal excretion maintains the calcium metabolism.Calcium is required for many intracellular functions (signal transmission and many enzymatic reactions) and extracellular functions (blood coagulation, muscle contraction, nerve conduction, hormone release, and mineralization of bone) and is regulated primarily by the actions of vitamin D and parathyroid hormone (PTH).Furthermore, calcitonin, a hormone that is produced and released by the C cells of the thyroid gland, opposes the actions of the PTH by decreasing calcium levels in blood (1).The calcium-sensing receptor (CaSR) plays a central role in the regulation of extracellular calcium homeostasis.Thus, in the presence of a high extracellular calcium concentration [Ca +2 ], the activation of CaSR inhibits the secretion of PTH, whereas the effect is reversed under low [Ca +2 ] (2).Moreover, hyperphosphatemia, hypomagnesemia, and the adrenergic action also contribute to PTH secretion.On the other hand, the 1,25-dihydroxyvitamin D3 inhibits PTH secretion.
The hypercalcemia can be an incidental finding, but, sometimes, patients have symptoms as lethargy, hypotonia, anorexia, weight loss, polyuria, polydipsia, vomiting, abdominal pain, and constipation.In long-term or severe cases, kidney failure, pancreatitis, arrhythmias, seizures, and psychiatric condition could be present.Because many of these disorders have a genetic etiology, in the present study, we sequenced a selection, carried out in 2017, of 55 genes encoding the principal proteins involved in the regulation of calcium metabolism to perform a genetic characterization of a cohort of 79 patients diagnosed with hypercalcemia.

Ethics statement
The study was approved by the Ethics Committee for Clinical Research of Euskadi (CEIC-E, code E20/31).A written informed consent was obtained from the patients, as well as their participating relatives and minors' legal guardian for the publication of any potentially identifiable images or data included in this article.Patients and their participating relatives provided a written informed consent for the genetic study.The research was carried out in accordance with the Declaration of Helsinki on human experimentation of the World Medical Association.

Patients
A cohort of 79 patients with hypercalcemia from 79 different unrelated families was included in this study.Patients were referred to our laboratory between 2003 and 2023.The main reason for this study was the genetic analysis of patients clinically diagnosed with familial hypocalciuric hypercalcemia (66 patients).Moreover, we included 13 patients who had neonatal or familial hypercalcemia.Patients with primary hyperparathyroidism were excluded.Firstdegree family members (mother and father) in 18 families with a genetic variant were analyzed.Clinical diagnoses were made by adult and pediatric endocrinologists or nephrologists working in 25 different hospitals.The molecular analysis was done in the Molecular Genetic Laboratory at Biobizkaia Health Research Institute, Barakaldo, Spain.

Gene selection and DNA analysis
A selection of 55 genes identified as potentially associated with disorders of calcium metabolism was screened for pathogenic variants (Table 1).For next-generation sequencing (NGS), a targeted panel was designed using the computer tool Ion AmpliSeq Designer (Thermo Fisher Scientific, Waltham, Massachusetts, USA) with an expected coverage of 98%.This panel included the exon regions and flanking intronic (at least 40 bases) sequences of the 55 genes.
Genomic DNA was extracted from peripheral blood leukocytes using the MagPurix instrument (Zinexts Life Science Corp., New Taipei City, Taiwan, R.O.C.).DNA purity and concentration were then determined using Qubit 2.0 fluorometer (Thermo Fisher Scientific).Library preparation was done using the Ion AmpliSeq Library Kit v2.0 (Thermo Fisher Scientific) according to the manufacturer's instructions.Samples were then sequenced using the Ion GeneStudio S5 System (Thermo Fisher Scientific).Base calling, read filtering, alignment to the reference human genome GRCh37/hg19, and variant calling were done using Ion Torrent Suite and Ion Reporter Software (Thermo Fisher Scientific).Not appropriately covered amplicons (<20×) and candidate variants were assessed by Sanger sequencing after polymerase chain reaction, sequenced with fluorescent dideoxynucleotides (BigDye Terminator v3.1 Cycle Sequencing Kit, Life Technologies, Grand Island, NY, USA), and loaded onto an ABI3130xl Genetic Analyzer (Thermo Fisher Scientific).
In order to confirm the deletion detected by NGS in the TBX1 gene, a commercially available MLPA (Multiplex Ligationdependent Probe Amplification) kit, SALSA MLPA Probemix P250 DiGeorge (MRC Holland, Amsterdam, The Netherlands), was used.
Novel DNA variants were named according to the Human Genome Variation Society guidelines (www.hgvs.org)and classified according to ACMG-AMP (American College of Medical Genetics and Genomics and the Association for Molecular Pathology) guidelines (3).

Results
Thirty percent of our cohort (24 out of the 79 index cases) presented one pathogenic or likely pathogenic variant in genes associated with hypercalcemia.In total, we found 15 pathogenic variants, nine likely pathogenic variants, and 12 variants of uncertain significance (10 variants in patients with hypocalciuric hypercalcemia and two variants in patients with hypercalcemia).Importantly, our genetic study revealed 11 variants not described so far (Table 2).
We confirmed the initial clinical diagnosis (biochemical data are shown in Table 3) of familial hypocalciuric hypercalcemia in 25% of the patients (17 out of the 66 patients had pathogenic or likely pathogenic variants in the CASR and AP2S1 genes).Moreover, we confirmed the initial clinical diagnosis of neonatal hyperparathyroidism in one patient (index case CA0110 had a homozygous pathogenic variant in the CASR gene) and infantile hypercalcemia in another patient (index case CA0139 had two pathogenic variants in compound heterozygous state in the CYP24A1 gene).
As expected, most patients had suspected variants in the CASR gene (15 patients) or AP2S1 gene (four patients) that have been associated with hypocalciuric hypercalcemia type 1 (Online Mendelian Inheritance in Man (OMIM), #145980) and hypocalciuric hypercalcemia type 3 (OMIM, #600740), respectively (biochemical data are shown in Table 3).
Two index cases (CA0143 and CA0102) had variants of uncertain significance in the GCM2 gene, and gain-of function mutations in the GCM2 gene cause autosomal dominant hyperparathyroidism type 4 (OMIM, #617343), a disorder characterized by hypercalcemia and elevated PTH secretion by parathyroid glands.These patients had hypercalcemia but not hyperparathyroidism.One of the limitations of our study is the lack of functional studies.Therefore, we cannot assure that the hypercalcemia observed is due to these variants.
Furthermore, many patients had variants in genes usually associated with hypophosphatemia.Thus, five patients had suspected variants in sodium/phosphate cotransporters.Three families (CA0112, CA0065, and ME0136) had variants in the SLC34A3 gene, two previously described as pathogenic and one of uncertain significance.This gene encodes the sodium/phosphate cotransporter 2C, and pathogenic variants in this gene have been associated with hypophosphatemic rickets with hypercalciuria (OMIM, #241530).Moreover, index cases CA0086 and SOR0257 had a pathogenic variant (p.Val408Glu) and a likely pathogenic variant (p.Val91_Ala97del), respectively, in the SLC34A1 gene (sodium/phosphate cotransporter 2A).Pathogenic variants in the SLC34A1 gene have been associated with infantile hypercalcemia type 2 (OMIM, #616963) and nephrolithiasis/osteoporosis hypophosphatemic type 1 (OMIM, #612286).Another two patients (CA0048 and CA0104) had variants of uncertain significance in the SLC9A3R1 gene.This gene encoded a sodium/hydrogen exchanger regulatory cofactor (NHERF1).Pathogenic variants in the SLC9A3R1 gene have been associated with autosomal dominant nephrolithiasis/ osteoporosis hypophosphatemic type 2 (OMIM, #612287).
On the other hand, we found variants in genes associated with vitamin D metabolism; one index case (CA0139) had two pathogenic variants in compound heterozygous (p.[Leu148Pro];[Arg396Trp]) in the CYP24A1 gene that encodes the vitamin D 24-hydroxylase; in the same gene, index case CA0029 had a variant of uncertain significance (p.Arg157Trp) in heterozygous state.Pathogenic variants in the CYP24A1 gene have been associated with infantile hypercalcemia type 1 (OMIM, #143880).Finally, in one patient (CA0159), we found two variants of uncertain significance in the VDR gene (vitamin D receptor).Unfortunately, we have not been able to check if the variants are on different alleles.Pathogenic variants in the VDR gene have been associated with autosomal recessive vitamin D dependent rickets type 2A (OMIM, #277440).
In addition, we found suspected variants in other five genes not associated with the phenotype (TBX1, ALPL, NEBL, CLDN19, and CYP27B1) (Table 2).Some of these variants were found in heterozygous state (CLDN19 and CYP27B1), in diseases described previously as recessively inherited.
Regarding patients without a genetic diagnosis, more studies should be carried out, for example, exome or genome sequencing, because they presented phenotypic characteristics similar to patients with a genetic variant.

Discussion
In this study, we examined the presence of genetic alterations in genes related to calcium metabolism in a cohort of 79 patients with hypercalcemia.We found pathogenic or likely pathogenic variants in 30% of patients in our cohort (24 out of 79 patients) that could explain the phenotype observed in those patients.Moreover, we have genetically confirmed the clinical diagnosis given by the clinician in the 24% of our cohort (17 patients with familial hypocalciuric hypercalcemia, one patient with neonatal hyperparathyroidism, and another patient with infantile hypercalcemia).Our complete genetic study revealed 15 pathogenic variants, nine likely pathogenic variants, and 12 of uncertain significance variants.Importantly, our genetic study revealed 11 novel variants.
We confirmed the clinical diagnosis of hypocalciuric hypercalcemia in 17 patients.Our high diagnostic yield respect others cohorts analyzed (20) may be due to a better clinical and biochemical characterization of the patients, to be restrict in the selection of the patients and the division in two groups (in one group, all the patients have hypercalcemia and hypocalciuria).Importantly, one of these patients, index case CA0022, had a de novo pathogenic variant in the CASR gene in mosaic (variant found in a 20% of reads), assuming that, it is possible that a minimal amount of defective protein is enough to develop the disease.To our knowledge, familial hypocalciuric hypercalcemia due to mosaicism in CASR has not been described before.Moreover, we found a pathogenic variant (p.Arg648*) in the CASR gene in another patient (index case CA0109) who had hypercalcemia but with hypercalciuria.It has been previously described that, in some cases, urinary calcium levels are normal or even high.Furthermore, the biochemical profile varies considerably and this variability is thought to be mutation dependent (21).
On the other hand, in four patients initially diagnosed with familial hypocalciuric hypercalcemia, we found suspected variants in other genes (TBX1, NEBL, and ALPL) not associated with the phenotype studied, and the definitive diagnosis could be changed.Thus, index case CA0080 had hypercalcemia, parathyroid hyperplasia, and hypocalciuria with a family history of hypocalciuria (mother and brother).Moreover, she presented with autoimmune hypothyroidism.In this patient, we found a pathogenic deletion in the DiGeorge syndrome chromosome region 22q11.2(TBX1 gene included).It is difficult to assess how this deletion is influencing the patient's phenotype because DiGeorge syndrome comprises hypocalcemia and hypoparathyroidism with parathyroid hypoplasia (OMIM, #188400).Furthermore, her mother and brother with hypocalciuria do not have the deletion.It is important to note that some patients with the deletion have minimal clinical expression.Thus, it has been described that 10% to 25% of parents of patients with DiGeorge exhibit the deletion and have no symptoms (22), and hypocalcemia has been described in up to 80% of adults with a 22q11.2deletion sometime during their lifetime (23).The patient could have another alteration because the hypocalciuria detected is inherited.Moreover, environmental factors such as hydration and sodium intake may explain why a patient with DiGeorge syndrome might have hypercalcemia.On the other hand, patient CA0080 presented an autoimmune disease, and this disease is observed in most age groups with this large deletion in the chromosome region 22q11.2(24).In another study, the authors described the phenotypic features of 78 adults with this deletion, and 20.5% presented with hypothyroidism (25).Importantly, another index case with hypocalciuric hypercalcemia (CA0068) of our cohort had a nonsense variant of uncertain significance (p.Try89*) in the NEBL gene.The NEBL gene has been found deleted in two patients with DiGeorge syndrome type 2, who showed cardiac defects, but not in two patients with the more distal deletion, which is associated with hypoparathyroism, deafness, and renal dysplasia (26).
Among patients with hypocalciuric hypercalcemia, two index cases (CA0121 and CA0128) had novel variants in the ALPL gene (one large duplication and one missense, respectively).The ALPL gene is known to cause autosomal recessive infantile hypophosphatasia (OMIM, #241500) and dominant or recessive adult odontohypophosphatasia (OMIM, #146300).However, index case CA0121 had high levels of alkaline phosphatase (655 IU/L; normal range, 44 IU/L to 147 IU/L).Therefore, as far as we know, this duplication could be the first variant reported in the ALPL gene with a gain of function effect and associated with high levels of alkaline phosphatase.He had the whole-gene ALPL duplicated, and this is the first whole-gene duplication detected in literature.On the other hand, patient CA0128 had osteopenia and weakness of the femoral head, clinical characteristic of hypophosphatasia.Unfortunately, we do not have the alkaline phosphatase value of patient CA0128.Administration of drugs that inhibit bone resorption is contraindicated in patients with hypophosphatasia.She was treated with vitamin D, which could have caused hypercalcemia in this patient.Further studies aimed at the functional characterization of these variants will be of help in defining the hypothesized pathogenic roles of these two variants in the ALPL gene.
Seven patients diagnosed with hypercalcemia had a variant in heterozygous state in genes associated with hypophosphatemia (SLC34A3, SLC34A1, and SLC9A3R1 genes).Minor symptoms such as mild hypophosphatemia and bone demineralization over time or kidney stones have been described in patients carrying heterozygous variants in these genes (27).Probably, in these patients, hypercalcemia may be justified because the renal phosphate wasting activates the secretion of calcitriol, which, in turn, increases calcium intestinal absorption (28).Moreover, we identified a heterozygous variant in the CYP24A1 gene in one patient (CA0029) with hypocalciuric hypercalcemia.Figueres et al. found the p.(Arg157Trp) variant in compound heterozygous state with a second potentially causative variant in two patients with hypercalcemia (11).On the other hand, allele frequency is greater than expected for disorder (0.33% of alleles in individuals of European descent in Genome Aggregation Database).Due to the absence of functional analysis the clinical significance of this variant is uncertain.Although an autosomal dominant disease inheritance has been proposed (29), we cannot exclude a second undetected variant in this patient.We recommended studying the genes associated with genetic rickets (SLC34A1, SLC34A3, and SLC9A3R1) in those patients with hypercalcemia of suspected genetic cause but without a confirmatory genetic diagnosis, because they could have a heterozygous pathogenic variant.
Finally, one patient (CA0159) had two variants of uncertain significance in the VDR gene.Although, patient CA0159 had hypophosphatemia and high PTH levels, it does not present the typical characteristics of the disease such as rickets, alopecia, cutaneous cysts, or hypocalcemia.Therefore, these two variants are probably located on the same allele or have no clinical relevance.
In conclusion, our study shows the utility of NGS in unraveling the genetic origin of disorders in the calcium and phosphorus metabolism and reveled interesting findings that demonstrate the importance of genetic analysis through massive sequencing (panels, exome, and genome) to obtain a clinical diagnosis of certainty.The identification of patients with a genetic cause is important for the appropriate treatment and identification of family members at risk of the disease.

TABLE 1
Genes involved in the regulation of calcium metabolism included in the panel for analysis.

TABLE 2
Molecular results in patients with a genetic variant.
(3)iants marked in bold have not been reported to date.*First-degreefamilymembers (mother and father) in 18 families were analyzed.+Classificationaccordingto ACMG-AMP guidelines(3).ACMG criteria: PVS, very strong evidence of pathogenicity; PS, strong evidence of pathogenicity; PM, moderate evidence of pathogenicity; PP, supporting evidence of pathogenicity.

TABLE 3
Laboratory findings of index cases with a genetic variant.