A Lifetime of Hypercalcemia and Hypercalciuria, Finally Explained

Abstract

Context:

Hypercalcemia, hypercalciuria, and recurrent nephrolithiasis are all common clinical problems. This case report illustrates a newly described but possibly not uncommon cause of this presenting complex.

Objective:

We report on a patient studied for over 30 years, with the diagnosis finally made with modern biochemical and genetic tools.

Design and Setting:

This study consists of a case report and review of literature conducted in a University Referral Center.

Patient and Intervention:

A single patient with hypercalcemia, hypercalciuria, and recurrent nephrolithiasis was treated with low-calcium diet, low vitamin D intake, prednisone, and ketoconazole.

Main Outcome Measure:

We measured the patient's clinical and biochemical response to interventions above.

Results:

Calcium absorption measured by dual isotope absorptiometry was elevated at 37.4%. Serum levels of 24,25-dihydroxyvitamin D were very low, as measured in two laboratories (0.62 ng/mL [normal, 3.49 ± 1.57], and 0.18 mg/mL). Genetic analysis of CYP24A1 revealed homozygous mutation E143del previously described. The patient's serum calcium and renal function improved markedly on treatment with ketoconazole but not with prednisone.

Conclusions:

Chronic hypercalcemia, hypercalciuria, and/or nephrolithiasis may be caused by mutations in CYP24A1 causing failure to metabolize 1,25-dihydroxyvitamin D.

Hypercalcemia, a common medical condition, is caused most often by primary hyperparathyroidism or malignancy. Hypercalcemia due to increased intestinal absorption of calcium is much less common, but is usually caused by abnormalities in ingestion or metabolism of vitamin D or its biologically active form, 1,25-dihydroxyvitamin D, and much less commonly by ingestion of large amounts of calcium together with absorbable alkali, the “milk alkali syndrome.” Although it was originally believed that the hypercalcemia of vitamin D intoxication might be mediated by the renal conversion of some proportion of the elevated level of 25-hydroxyvitamin D to 1,25-dihydroxyvitamin D (1), recent data from the CYP27B1-knockout mouse lacking the ability to make 1,25-dihydroxyvitamin D suggest that toxicity is due to the effects of very high 25-hydroxyvitamin D levels (2). The hypercalcemia of sarcoidosis, of some other granulomatous diseases, and of certain lymphomas is due to the accelerated 1α-hydroxylation of 25-hydroxyvitamin D by granulomatous macrophages or neoplastic lymphocytes (3, 4). A third mechanism of vitamin D-mediated hypercalcemia, only recently discovered, is a genetic mutation in the CYP24A1 enzyme that is responsible for the inactivation of 1,25-dihydroxyvitamin D by its metabolism to 1,24,25-dihydroxvitamin D (5). We describe a patient with chronic hypercalcemia mediated by inappropriately high levels of 1,25-dihydroxyvitamin D caused by an inactivating mutation in this gene. Although exceedingly rare, it is possible that with wider recognition of this etiology, more cases will be discovered.

Case Report

This 73-year-old university professor, born in Greece to nonconsanguineous parents, suffered “leg weakness” as a child that resolved as he grew older. First symptomatic of kidney stones in his early 20s, he was admitted to Columbia University Medical Center a decade later with recurrent right renal calculi. The serum calcium was elevated, between 10.7 and 11.5 mg/dL, and phosphorus (4.0 mg/dL), blood urea nitrogen (21.5 mg/dL), and 24-hour urine calcium (269 mg) were all normal. Because reliable measurements of PTH were not available at this time, and thus did not yet have diagnostic utility, the patient underwent neck exploration and 3 years later mediastinal exploration in an effort to find and excise abnormal parathyroid tissue. Histological findings of these two operations included only two normal parathyroid glands, remnants of the thymus, and chronic lymphadenitis. Hypercalcemia persisted, and virtually all PTH levels before and after these two operations were at or below the lower limit of the assay, with serum calcium levels persistently in the range of 11.0 to 13.2 mg/dL. Further evaluation at age 44 confirmed persistent hypercalcemia (11.7 mg/dL) and the following: phosphorus, 2.7 mg/dL; creatinine, 1.7 mg/dL; 24-hour urine calcium (three determinations), 508, 288, and 360 mg (normal [nl], <300 mg); with 24-hour urine hydroxyproline, 19 and 18 mg/d (nl, 14.9–43.7 mg/d). Angiotensin converting enzyme level was normal, and the Kveim test was negative. Serum 1,25-dihydroxyvitamin D was 38 pg/mL. Treatment with oral prednisone 30 mg/d for 1 week did not reduce the serum calcium. A percutaneous, transiliac bone biopsy with quantitative static and dynamic histomorphometry showed normal values of trabecular bone volume, cortical width, percentage osteoid, osteoid width, extent of tetracycline uptake, percentage resorption surface, and osteoclast presence. At age 53, his ionized calcium level was 5.92 mEq/L (nl, <5.3 mEq/L); with osteocalcin, 13.4 ng/mL (nl, 8–52 ng/mL); bone-specific alkaline phosphatase activity, 9.5 μg/L (nl, 5.9–27.8 μg/L); 24-hour urine N-telopeptide, 22.8 nmol BCE/mmol creatinine (nl, <86); 25-hydroxyvitamin D, 25 ng/mL; and 1,25-dihyroxyvitamin D, 74 pg/mL (nl, 15–60 pg/mL). A PTHrP level was 0.4 pmol/L (nl, <1.3 pmol/L). A bone mineral density test showed T-scores of +0.6 in L1–4, +1.1 in total hip, and +2.9 in the 1/3 radius, each measurement having changed by +0.6%, −0.3%, and +10.5%, respectively, in comparison with 8 years earlier. A bone density measurement at age 73 showed values only slightly lower.

At age 72, he was hospitalized with fatigue and confusion after his dose of hydrochlorothiazide was increased from 25 to 100 mg/d for worsening hypertension. The serum calcium had risen to 15.8 mg/dL, creatinine to 4.3 mg/dL, and phosphorus to 3.6 mg/dL. The 25-hydroxyvitamin D level was 20 ng/mL, and the 1,25-dihydroxyvitamin D was 48 pg/mL. With iv saline hydration and prednisone (50 mg/d), the serum calcium fell to 11.5 mg/dL. Two weeks later, he experienced low-grade fever and poorly localized pains in his back and shoulders. His erythrocyte sedimentation rate test was 110 mm/h. For a presumptive diagnosis of polymyalgia rheumatica, he was started on prednisone 20 mg/d. The 1,25-dihydroxyvitamin D was 75 pg/mL. Symptoms did not improve, and the serum calcium rose again to 15 mg/dL. Prednisone was increased to 60 mg/d, and he was started on ketoconazole 200 mg twice a day. The prednisone was rapidly tapered to 10 mg/d, and after 3 weeks on ketoconazole the serum calcium fell to 10.3 mg/dL. In 3 months, his calcium had fallen to 10.2 mg/dL, with 1,25-dihydroxyvitamin D of 44 pg/mL. Ten months later, his calcium level was 10.6 mg/dL, creatinine was 2.62 mg/dL, and 1,25-dihydroxyvitamin D was 25 pg/mL. The patient has no siblings or children and knows of no difficulties with calcium or renal stones in his parents, although a deceased maternal cousin was said to have had a “calcium problem.”

Methods

Assays for all routine laboratory data, including PTH, 25-hydroxyvitamin D, 1,25-dihydroxyvitamin D, and PTHrP, were done in a variety of different clinical laboratories. Calcium absorption studies were performed using a dual stable isotope technique (6). The initial assay for 24,25-dihydroxyvitamin D was performed at the Mayo Clinic (Rochester, MN) under Dr Rajiv Kumar (7). The second assay for 24,25-dihydroxyvitamin D was performed at the laboratory of Dr Glenville Jones at Queen's University (Kingston, Ontario, Canada). Serum 25-hydroxyvitamin D and 24,25-dihydroxyvitamin D were measured in 100-μL serum samples by liquid chromatography tandem mass spectrometry with a Waters Xevo TQ-S instrument after DMEQ-TAD derivitization. Deuterated standards were used to quantitate the vitamin D metabolites. Results are expressed as a ratio of 25-hydroxyvitamin D3 to 24,25-dihydroxyvitamin D3. Genetic analysis was performed at the laboratory of Dr Karl Peter Schlingmann at Wilhelms-University (Munster, Germany) (5). Written informed consent was given by the patient for the genetic analysis.

Results

True fractional calcium absorption in this patient was 37.4%, with published values for normal males of 21.9 ± 1.7% to 27.4 ± 1.8% (8). The serum concentration of 24,25-dihyroxyvitamin D measured initially before the patient's episode of severe hypercalcemia was very low at 0.62 ng/mL (nl, 3.49 ± 1.57 ng/mL when 25-hydroxyvitamin D levels are 15–50 ng/mL) at a time when the patient's serum calcium was 11.4 mg/dL and 25-hydroxyvitamin D was 23 ng/mL. Serum level of 24,25-dihydroxyvitamin D measured subsequently was confirmed to be low at 0.18 ng/mL. The ratio of 25-hydroxyvitamin D to 24,25-dihydroxyvitamin D on this specimen was 169, a value that represents the highest ratio ever measured in that laboratory. At that time, the serum 25-hydroxyvitamin D was 30 ng/mL, and calcium was 12.1 mg/dL. Genetic analysis of CYP24A1 revealed a homozygous mutation E143del previously described by Schlingmann et al (5), Streeten et al (9), and Dauber et al (10).

Discussion

In the 1950s in the United Kingdom, an epidemic of infantile hypercalcemia, hypercalciuria, and nephrocalcinosis was observed after the introduction of vitamin D recommendations up to 4000 IU/d for these young children (11). The diagnosis of idiopathic infantile hypercalcemia (IIH) was coined, and the epidemic subsided when recommended doses of vitamin D were reduced to the previous level, 500 IU/d. Because most infants tolerated the higher levels without difficulty, some observers felt that the children who became ill had an unusual sensitivity to vitamin D (12), although the mechanism was unclear and occasional sporadic cases of IIH continued to be observed over the coming years. With the explosion of knowledge in vitamin D metabolism in the 1970s, it became clear that both 1,25-dihydoxyvitamin D, the active form of vitamin D, and its precursor, 25-hydroxyvitamin D, were primarily metabolized by CYP24A1 (24-hydroxylase), an enzyme activity that resulted in a five-step process to the side chain-truncated, inactive product, calcitroic acid (13). A CYP24A1-null mouse model was developed by St-Arnaud et al (14), exhibiting a phenotype with hypercalcemia and nephrocalcinosis and resulting in 50% lethality at weaning. Survivors lacking CYP24A1 demonstrated very delayed clearance of 1,25-dihydroxyvitmin D from the blood, liver, and kidney; reduced expression of 25-hydroxyvitamin D 1-α hydroxylase; higher levels of 25-hydroxyvitamin D in blood and other tissues; and no formation of 24,25-dihydroxyvitamin D. Keratinocytes from these mice failed to synthesize calcitroic acid from 1,25-dihydroxyvitamin D (15).

In 2010, Nguyen et al (16) reported 20 infants with hypercalcemia, hypercalciuria, and low levels of PTH. Serum levels of 1,25-dihydroxyvitamin D were high-normal or frankly elevated, and levels of 25-hydroxyvitamin D were normal. Analysis of the promoter sequences of four genes important in vitamin D metabolism in 17 patients and 12 family members showed only an excess of Klotho minor alleles in the hypercalcemic patients when compared with 195 control subjects. Evaluation of 24-hydroxylase activity in the fibroblasts of one patient showed detectable 24,25-dihydroxyvitamin D under basal conditions, but no increase after exposure to 1,25-dihydroxyvitamin D. Patients treated with ketoconazole, a broad-spectrum cytochrome P450 inhibitor known to block synthesis of 1,25-dihydroxyvitamin D (17, 18), normalized their serum calcium levels sooner than patients not treated. The following year, Schlingmann et al (5) reported six patients with IIH from four families and four young patients with vitamin D intoxication after receiving a single oral dose of ergocalciferol 600 000 IU. Analysis of three genes (CYP27B1, FGF23, and KL) important in calcium metabolism resulted in discovery of no functionally meaningful mutations, but analysis of CYP24A1 revealed five different mutations present in either homozygous or compound heterozygous form, which resulted in alterations in amino acid residues that are highly conserved in vitamin D-metabolizing enzymes. Recreation of the mutations in the wild-type human gene and in vitro transfection of the mutant genes into V79–4 host cells resulted in stable or transient expression of CYP24A1, with no measurable expression of 24-hydroxylase activity in all but one construct, which expressed only 5.3% of wild type (5). In 2012, Tebben et al (19) reported a family in which affected members had hypercalcemia, hypercalciuria, nephrolithiasis, and elevated levels of 1,25-dihydroxyvitamin D. Sequencing of the CYP24A1 gene carried out in the proband and seven family members in three generations showed two canonical splice junction mutations in the CYP24A1 gene. Analysis of family members showed a phenotype associated with one or both mutations, suggesting autosomal dominant transmission with partial penetrance of the trait. Treatment with ketoconazole 200 mg every 8 hours for 2 months led to normalization of serum and urine calcium and serum 1,25-dihydroxyvitamin D, and previously decreased concentrations of PTH increased into the normal range (19). Streeten et al (9) reported a 47-year-old man who presented with nephrolithiasis at age 19 but was discovered at age 39 to be hypercalcemic and hypercalciuric with suppressed PTH, elevated levels of 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D, and low levels of 24,25-dihydroxyvitamin D. Sequencing of CYP24A1 revealed the E143del mutation also present in our patient (9).

The patient described in this report appears to represent the longest course of chronic hypercalcemia and hypercalciuria due to a mutation in CYP24A1. This syndrome appears to be compatible with a long and healthy life, although with obvious high risk for calcium urolithiasis and nephrocalcinosis. His presentation is different from that of the adult patient reported by Tebben et al (19) in that his bone mineral density is well above average and rising with age, possibly representing a lifetime of exposure of skeletal tissues to low levels of PTH (20) and high levels of calcium and 1,25-dihydroxyvitamin D. His serum calcium rose rapidly in response to an escalation of his dose of hydrochlorothiazide, a medication known to reduce renal calcium excretion (21), suggesting that patients with this syndrome are at risk from exposure to thiazide diuretics as well as to calcium supplements and vitamin D. They may also be at special risk from exposure to other medical conditions or medications known to reduce renal calcium excretion (eg, renal failure) or to cause hypercalcemia through accelerated bone resorption. His elevated serum calcium did not respond to glucocorticoids, unlike many patients with other causes of hypercalcemia associated with elevated levels of 1,25-dihydroxyvitamin D, but it did appear to respond to the administration of ketoconazole in a modest dose, presumably because of the effect of ketoconazole on the metabolism of 1,25-dihydroxyvitamin D (18, 22). A great many patients with calcium nephrolithiasis appear to absorb and excrete dietary calcium at abnormally high rates and have serum 1,25-dihydroxyvitamin D levels in the normal or mildly elevated range (23, 24). It remains to be determined whether inborn or acquired abnormalities of vitamin D metabolism play a part in their predilection to stone formation. This case also illustrates that a mutation in this gene should be considered in the context of long-standing, unexplained hypercalcemia associated with normal or elevated 1,25-dihydroxyvitamin D and kidney stones.