Maternal Hypercalcemia Due to Failure of 1,25-Dihydroxyvitamin-D₃ Catabolism in a Patient With CYP24A1 Mutations

Abstract

Context:

Calcium metabolism changes in pregnancy and lactation to meet fetal needs, with increases in 1,25-dihydroxyvitamin D [1,25-(OH)₂D] during pregnancy playing an important role. However, these changes rarely cause maternal hypercalcemia. When maternal hypercalcemia occurs, further investigation is essential, and disorders of 1,25-(OH)₂D catabolism should be carefully considered in the differential diagnosis.

Case:

A patient with a childhood history of recurrent renal stone disease and hypercalciuria presented with recurrent hypercalcemia and elevated 1,25-(OH)₂D levels during pregnancy. Laboratory tests in the fourth pregnancy showed suppressed PTH, elevated 1,25-(OH)₂D, and high-normal 25-hydroxyvitamin D levels, suggesting disordered vitamin D metabolism. Analysis revealed low 24,25-dihydroxyvitamin D₃ and high 25-hydroxyvitamin D₃ levels, suggesting loss of function of CYP24A1 (25-hydroxyvitamin-D₃-24-hydroxylase). Gene sequencing confirmed that she was a compound heterozygote with the E143del and R396W mutations in CYP24A1.

Conclusions:

This case broadens presentations of CYP24A1 mutations and hypercalcemia in pregnancy. Furthermore, it illustrates that patients with CYP24A1 mutations can maintain normal calcium levels during the steady state but can develop hypercalcemia when challenged, such as in pregnancy when 1,25-(OH)₂D levels are physiologically elevated.

Significant changes in maternal calcium (Ca) metabolism occur during pregnancy and lactation to provide adequate Ca for fetal skeletal mineralization. During pregnancy, Ca intake, intestinal Ca absorption, and bone resorption increase, resulting in hypercalciuria (1). Enhanced 1,25-dihydroxyvitamin D [1,25-(OH)₂D] production by renal and placental 1α-hydroxylases plays an important role (1). Ionized Ca and albumin-corrected serum total Ca remain normal, whereas parathyroid hormone (PTH) levels become low or suppressed (1). As pregnancy progresses, PTH-related protein (PTHrp) production rises (1). During lactation, intestinal Ca absorption normalizes, and increased bone resorption and renal Ca reabsorption provide Ca for milk production in response to high PTHrP levels (1). Furthermore, 1,25-(OH)₂D levels normalize, and PTH remains low (1). Maternal hypercalcemia can be due to a broad range of disorders including primary hyperparathyroidism, familial hypocalciuric hypercalcemia, elevated PTHrP, and milk-alkali syndrome (1, 2). The recent appreciation that reduced 1,25-(OH)₂D catabolism, due to deficient CYP24A1 activity, can cause hypercalcemia raises this entity as another possible cause of maternal hypercalcemia.

Case

A 28-year-old pregnant female was referred for recurrent hypercalcemia in her fourth pregnancy. She had episodes of nephrolithiasis since age 6 years, complicated by urosepsis and requiring several lithotripsies. Laboratory testing and imaging revealed hypercalciuria (24 h urine calcium > 200 mg) and nephrocalcinosis, respectively. At age 15 years, she had normal serum total Ca (9.2 mg/dL; normal 8.7–10.1 mg/dL) and low PTH (8 ng/L; normal 18–73 ng/L). However, she was lost to follow-up until her pregnancies.

During her three prior pregnancies (Table 1), she developed gestational hypertension and hypercalcemia with a serum albumin-corrected total Ca (all subsequent serum calcium levels are corrected for albumin) of up to 12.3 mg/dL (normal 8.5–10.0 mg/dL) while taking a prenatal vitamin. Evaluation after her third pregnancy while taking prenatal vitamins again showed hypercalcemia (15 mg/dL), suppressed PTH (<3 pg/mL; normal 10–65 pg/mL), elevated 1,25-(OH)₂D (157 pg/mL; normal 18–72 pg/mL), and high-normal 25-hydroxyvitamin D (25-OH-D; 56 ng/mL; normal 30–100 ng/mL) (Table 1). Twenty-four-hour urine Ca was high (417 mg; normal <200 mg) and PTHrP was normal (18 pg/mL; normal 14–27 pg/mL). The chest x-ray and the tuberculin skin test were negative. Computed tomography of the abdomen demonstrated calcifications in the medullary regions in the right kidney and a few punctate calculi in the left collecting system. Serum Ca normalized to 9.9 mg/dL over the ensuing months and remained normal until her fourth pregnancy (Table 1).

Table 1.

Laboratory Values During Pregnancies and Postpartum

Date	Ca (8.5–10.1 mg/dL)a	Phos (2.5–4.6 mg/dL)	Cr (0.44–1.00 mg/dL)	Ionized Ca (1.2–1.4 mmol/L)	25-OH-D (30–100 ng/mL)	1,25-(OH)2-D (18–72 pg/mL)	PTH (10–65 pg/mL)	Presence of Renal Stones on Imaging?
December 1999	9.2		0.7	1.26			8	Yes
Pregnancy 1 (March 2010)	12.3		0.75					Unknown
Pregnancy 2 (February 2012)	12.3		0.74	1.43				Yes
Pregnancy 3 (January 2013)	11.7	2.8	0.64	1.52	64	127	<3	Unknown
2 wk postpartum (March 2013)b	15.0	3.2	0.73	2.13	56	157	<3	Yes
3 mo postpartum (June 2013)	9.9	3.3	0.82		27	33	5	Unknown
4 mo postpartum (July 2013)	9.0	2.5	0.91				31	Yes
6 mo postpartum (September 2013)	9.8	2.2	0.81		39	53	19	Yes
11 mo postpartum (February 2014)	9.2	3.1	0.75		14	74		Unknown
Pregnancy 4 (June 2014, 14 wk)	9.5	3.1	0.71	1.31	74	176	3	Unknown
Pregnancy 4 (July 2014, 18 wk)	10.6		0.77	1.51	41	211		Unknown
Pregnancy 4 (September 2014, 26 wk)	12.2	5.1	0.72	1.66	26	212	<3	Unknown
Pregnancy 4 (October 2014, 30 wk)	9.9				55	274		Unknown
Pregnancy 4 (December 2014, 37 wk)	10.6	4.8	0.72	1.38	14	181	<3	Unknown
6 wk postpartum (January 2015)	10.4	3.3	0.83		23	79	3	Unknown

Abbreviations: Cr, creatinine; Phos, phosphorus.

^aCa corrected for albumin in all measurements except no albumin available from December 1999.

^bNormal angiotensin-converting enzyme and TSH.

At 14 weeks' gestation in her fourth pregnancy, she had an elevated 1,25-(OH)₂D (176 pg/mL) with normal serum Ca (9.5 mg/dL) and suppressed PTH while off all supplements. Over the course of her pregnancy, serum Ca rose, peaking at 12.2 mg/dL with a 1,25-(OH)₂D level of 212 pg/mL at 26 weeks (Table 1). At that time, she had symptoms of marked fatigue and constipation but declined admission. The patient was advised to maintain good hydration and avoid dietary Ca, sunlight, and vitamins. At 27 weeks, serum Ca decreased (9.7 mg/dL); conservative measures were continued. At 30 weeks, her 24-hour urine Ca was high (378 mg) with serum Ca of 9.9 and 1,25-(OH)₂D level of 274 pg/mL (Table 1).

Hypercalcemia and high 1,25-(OH)₂D levels prompted consideration of disordered vitamin D metabolism. Vitamin D metabolite analysis (3) by liquid chromatography and tandem mass spectrometry revealed high 25-OH-D₃ (63.4 ng/mL, normal 20–50 ng/mL) and low 24,25-dihydroxyvitamin D₃ (24,25-(OH)₂D₃; 0.74 mg/mL), with a ratio of 25-OH-D₃ to 24,25-(OH)₂D₃ of 86, in the range (>80) observed in idiopathic infantile hypercalcemia (IIH) (Figure 1). This suggested reduced activity of CYP24A1. After written informed consent approved by the University of California, San Francisco, Committee on Human Research, CYP24A1 gene sequencing indicated an in-frame deletion (E143del) and missense mutation (R396W) (Supplemental Figure 1).

Figure 1.

Vitamin D metabolite analysis. The figure depicts chromatograms from liquid chromatography and tandem mass spectrometry analysis of DMEQ-TAD derivatives of vitamin D metabolites studied in this patient. Chromatograms show specific daughter fragments in panel A at mass to charge ratio (m/z) 746→468 from 25-OH-D₃ and in panel B at m/z 762→468 from 24,25-(OH)₂D₃. Each vitamin D metabolite produces two adduct peaks when derivatized with DMEQ-TAD, and in panel A for 25-OH-D₃, these are at 3.3 and 3.8 minutes; and in panel B for 24,25-(OH)₂D₃, these are at 1.5 and 2.2 minutes. There are small peaks of 1,25-(OH)₂D₃ at 2.34 and 2.70 minutes because it has the same molecular mass and fragmentation pattern as 24,25-(OH)₂D₃ (3). A, Measurement of serum 25-OH-D₃ in patient (designated as G4P3) and unaffected female control. Control 25-OH-D₃ level was 30.2 ng/mL, whereas the affected patient has a 25-OH-D₃ level of 63.4 ng/mL (normal range for 25-OH-D₃ 20–50 ng/mL). The quantitation is based on the comparison of the observed peaks at m/z 746→468 with the peaks at m/z 749→471 from the d₃-25-OH-D₃ internal standard (3). B, Measurement of 24,25-(OH)₂-D₃ in patient (G4P3) and unaffected control. The control has a 24,25-(OH)₂-D₃ level of 3.0 ng/mL, and the patient has a 24,25-(OH)₂-D₃ level of 0.74 ng/mL (normal range for 24,25-(OH)₂D₃ 1.5–5 ng/mL). The quantitation is based on the comparison of the observed peaks at m/z 762→468 with the peaks at m/z 768→468 from the d₆-24,25-(OH)₂D₃ internal standard (3). The normal range for the ratio of 25-OH-D₃ to 24,25-(OH)₂D₃ is 5–25 (3). In the control here, it is 10, whereas the ratio in the patient is 86, consistent with the ratio of greater than 80 seen in other patients with proven IIH due to mutations of CYP24A1. This suggests that the patient has a disorder of vitamin D degradation.

Serum Ca levels monitored throughout the remaining weeks of pregnancy were 9.9–10.6 mg/dL (Table 1). The baby was delivered at 37 weeks due to preeclampsia. Serum Ca level in the infant, drawn 7 hours postnatally, was slightly low (8.8 mg/dL; normal 9–10.9 mg/dL) with an ionized Ca of 1.17 mmol/L (normal 1–1.50 mmol/L) and high normal phosphate (6.4 mg/dL; normal 3.9–7.7 mg/dL), suggestive of suppressed neonatal parathyroid function. PTH and vitamin D metabolites were not obtained. Although the infant is an obligate carrier of one mutation in CYP24A1, perturbations in Ca would not be expected because the presence of only one of these mutations has not been reported to affect Ca levels. Placental staining for CYP27B1 (1-α-hydroxylase) was similar to control placenta (data not shown). Six weeks postpartum, while lactating, the patient's serum Ca was 10.4 mg/dL with 1,25-(OH)₂D of 79 pg/mL. The newborn was doing well clinically at 5 months of age.

Discussion

1,25-(OH)₂D levels increase during pregnancy due to enhanced production by renal and placental 1α-hydroxylases, which can result in absorptive hypercalciuria. Gertner et al (4) studied 16 healthy pregnant women and determined average serum 1,25-(OH)₂D levels and urinary calcium levels. While Ca remained normal (average 8.6–9.3 mg/dL), 1,25-(OH)₂D levels during all trimesters averaged 94–118 pg/mL with 24-hour urinary calcium levels of 247–316 mg (4). Postpartum, mean serum Ca levels were normal, while mean serum 1,25-(OH)₂D and urinary calcium levels decreased to 51 pg/mL and 91 mg/d, respectively (4). Therefore, serum 1,25-(OH)₂D and 24-hour urinary calcium levels during pregnancy and postpartum in our patient were significantly higher than previous reports, suggesting disordered vitamin D metabolism. Suspected causes include the overproduction or reduced metabolism of 1,25-(OH)₂D, with reduced CYP24A1 activity being a possible explanation.

CYP24A1 metabolizes 25-OH-D and 1,25-(OH)₂D, thus controlling 1,25-(OH)₂D levels (5). 24-Hydroxylated metabolites of vitamin D have uncertain functions in vivo, but CYP24A1 activity is tightly controlled (5). CYP24A1 expression is regulated by the following: 1,25-(OH)₂D, which induces its expression; PTH, which suppresses renal expression and induces osteoblastic expression; and fibroblast growth factor-23, which induces renal expression (5). CYP24A1 knockout mice have impaired catabolism and delayed elimination of 1,25-(OH)₂D₃ and elevated ratios of 25-OH-D₃ to 24,25-(OH)₂D₃ (3, 6, 7). These findings demonstrate the critical role of CYP24A1 in the degradation of vitamin D.

The effect of mutations in CYP24A1 in humans first came to clinical attention in Great Britain in the 1950s when milk products started to be fortified with vitamin D. This resulted in an epidemic of non-PTH-mediated hypercalcemia among infants and was eventually termed idiopathic infantile hypercalcemia (IIH). This temporal relationship suggested an increased sensitivity to vitamin D in these infants (8). Many years later, this increased sensitivity to vitamin D was explained when Schlingmann et al (8) studied a cohort of familial IIH cases and identified CYP24A1 mutations in 10 children. Six children had classic IIH, whereas four had suspected vitamin D intoxication (8). These children presented with several symptoms including failure to thrive, polyuria, lethargy, and hypotonia between approximately 1 and 11 months of age (8). Laboratory tests revealed hypercalcemia, low PTH levels, normal to high 25-OH-D and 1,25-(OH)₂D levels, and hypercalciuria (8). The CYP24A1 mutations identified in these children affected highly conserved residues of functional importance and hence impacted vitamin D degradation, resulting in hypercalcemia (8). The mutations found in our patient (E143del and R396W), a compound heterozygote, were also identified in homozygotes and compound heterozygotes in this cohort. These CYP24A1 mutations led to severely reduced enzymatic activity in vitro (8).

Since that report, additional cases and mutations have been reported (9). Patients with CYP24A1 mutations present at all ages and with different manifestations (9). Infants tend to present with IIH, whereas children and adults present with nephrolithiasis, hypercalciuria, and nephrocalcinosis. Affected patients are homozygotes or compound heterozygotes with one report of an affected heterozygous patient with the L409S mutation, who developed hypercalcemia and nephrolithiasis while on vitamin D (3700 IU daily) (9). These cases did not present as hypercalcemia during pregnancy, but Dinour et al (10) described such a case with CYP24A1 mutations with hypercalciuria and nephrolithiasis similar to our patient.

Our patient and the above-mentioned case (10) illustrate a newly appreciated important cause of maternal hypercalcemia, broadening the spectrum of presentations of CYP24A1 mutations. Interestingly, our patient initially presented classically, with childhood nephrolithiasis and hypercalciuria, and was diagnosed with absorptive hypercalciuria. She did not return to clinical attention until she developed hypercalcemia during pregnancy. She demonstrates that women with CYP24A1 mutations can maintain normal Ca homeostasis at steady state but that hypercalcemia can develop when the system is challenged, as happens in pregnancy, by physiologically elevated 1,25-(OH)₂D levels and Ca and vitamin D supplements.

Conclusion

These new insights into the pathophysiology and presentation of CYP24A1 mutations have further broadened our understanding of disorders of vitamin D degradation. In addition to considering the more traditional causes of maternal hypercalcemia, physicians should expand their differential to include CYP24A1 mutations, especially in women with a history of nephrolithiasis, hypercalciuria, and/or nephrocalcinosis. This is very important given the increasing use of Ca and vitamin D supplements and given that this disorder can be greatly exacerbated during pregnancy.