Hypoparathyroidism is characterized by an inadequate synthesis of parathyroid hormone (PTH) in order to maintain normal plasma calcium levels. PTH is of major importance to calcium and bone homeostasis because it stimulates bone turnover and thereby facilitates the renewal and release of calcium from bone when needed. In addition, PTH increases renal tubular calcium reabsorption and promotes renal phosphate excretion. PTH also stimulates the renal 25-hydroxyvitamin D-1α-hydroxylase, thereby increasing synthesis of 1,25-dihydroxycholecalciferol [1,25(OH)2D3], which stimulates intestinal calcium absorption and tubular reabsorption of calcium.1 As a consequence, lack of PTH causes hypocalcemia owing to an increased renal calcium loss, decreased intestinal calcium absorption, and reduced ability to mobilize skeletal calcium. The dominating clinical symptoms are related to increased neuromuscular irritability.
Surgical removal of and damage to the parathyroid glands are the most common causes of hypoparathyroidism,3 but in rare instances the disease may be due to genetic2, 3 or autoimmune3, 4 disorder. Unfortunately, no reliable data are available on the epidemiology of hypoparathyroidism in terms of incidence, prevalence, comorbidity, treatment modalities, or mortality.
Currently, conventional therapy includes treatment with calcium and usually 1α-hydroxylated vitamin D metabolites to relieve symptoms caused by hypocalcemia.3, 5 Treatment focuses on maintaining stable or slightly reduced plasma calcium levels.3 The reasons for keeping plasma calcium levels low are (1) to minimize urinary calcium excretion and (2) to reduce the risk of hypercalcemia. However, not all patients are well regulated when treated with calcium and activated vitamin D analogues. Patients may experience symptoms of either hyper- and/or hypocalcemia causing discomfort. In accordance, indices of quality of life (QoL) have been reported to be significantly reduced in hypoparathyroidism on conventional treatment, including fatigue, reduced endurance, and a tendency to depression.3, 6, 7 Therefore, improved treatment options are warranted.
Hypoparathyroidism is the only major hormonal insufficiency state left that is usually not treated by replacing the missing hormone.3, 5 In recent years, a few studies have aimed to evaluate the feasibility of PTH-replacement therapy (PTH-RT) in hypoparathyroidism. The studies have used either PTH(1–34)3, 8–12 or intact PTH.13 Both the intact hormone PTH(1–84) and the truncated PTH(1–34) analogue have been shown to be able to stabilize plasma calcium8–13 and normalize plasma phosphate levels.8, 13 Furthermore, in some studies, PTH(1–34) treatment has been shown to reduce urinary calcium excretion.8, 10 Studies on the efficacy of PTH(1–34) have been randomized and controlled but open-labeled, whereas the study on PTH(1–84) was based on a case series. Some of the studies have included the effect of the treatment on BMD10, 12, 13 and biochemical markers of bone turnover,8–13 showing an increased bone turnover with either an unchanged10, 12 or increased BMD13 in response to treatment.
The aim of this study was to assess the feasibility of treating hypoparathyroidism patients with intact PTH(1–84) as an add-on therapy to conventional treatment in a randomized, placebo-controlled, double-blind setting as assessed by (1) indices of calcium-phosphate homeostasis and (2) effects on plasma 1,25(OH)2D3 levels, BMD, and biochemical markers of bone turnover.
Subjects and Methods
We considered men and women between 25 and 80 years of age eligible for study inclusion if they had been diagnosed with hypoparathyroidism. We defined this disorder as inappropriately low plasma PTH levels in the setting of hypocalcemia, necessitating continuous treatment with an active vitamin D analogue (ie, alphacalcidol or calcitriol) or high doses of ergocalciferol (vitamin D2) for at least 12 months. Participants were required to have plasma ionized calcium (Ca2+) and magnesium levels within or just below normal reference ranges. In order to ensure a replete vitamin D status in terms of plasma 25-hydroxyvitamn D [25(OH)D] levels, we only included patients with plasma 25(OH)D levels greater than 50 nmol/L or patients who reported regular use of vitamin D supplements (ie, ergocalciferol [D2] or cholecalciferol [D3]) in a daily dose of 10 µg (400 IU) or greater for a minimum of 3 months prior to study entry. We only included women of child-bearing age if they did not plan pregnancy and reported use of contraception. We excluded patients with severely impaired renal (plasma creatinine > 200 µmol/L) or hepatic function (plasma alanine aminotransferase [ALAT] > 100 U/L [normal range, women 10 to 45 U/L, men 10 to 70 U/L]) and/or alkaline phosphatase greater than 400 U/L (normal range 35 to 105 U/L). We also excluded patients who had been hospitalized owing to chronic drug or alcohol abuse. Other exclusion criteria consisted of a diagnosis of a malignant disease involving the skeleton, sarcoidosis, Paget disease of bone, and previous radiation therapy involving the skeleton. Finally, we excluded patients on treatment with digoxin, raloxifene, calcitonin, systemic corticosteroids (>5 mg/d), fluoride, lithium, PTH analogues, or anticonvulsants.
We identified potential participants by hospital discharge codes and recruited patients from outpatient clinics in the western part of Denmark. We screened 170 patients and included 62 between June 25, 2008, and October 8, 2009. All patients provided written informed consent. We performed the study in accordance with the Declaration of Helsinki II and the guidance on Good Clinical Practice (GCP). The GCP Unit at the University Hospital of Aarhus, Denmark, monitored the study. The Danish Data Protection Agency was notified about the study. The Ethical Committee of Central Denmark (No. M20080040) and the Danish National Board of Health approved the study (EudraCT No. 2008-000606-36, Protocol No. 84421383; ClinicalTrials.gov No. NCT00730210).
The study was an investigator-initiated double-blind, randomized, placebo-controlled parallel-group trial comparing the effect of adding PTH(1–84) or placebo to conventional treatment with calcium and alphacalcidol/calcitriol/ergocalciferol for 6 months.
Patients were randomly assigned to receive a daily self-administrated subcutaneous injection in the femoral region with either PTH(1–84) (Preotact) in a fixed dose of 100 µg or similar placebo. PTH(1–84) and placebo were provided free of charge by Nycomed A/S (Roskilde, Denmark). Randomization was performed by the hospital pharmacy at Aarhus University Hospital using a computer-generated randomization code. A restricted block randomization procedure with serial entry in permuted blocks was used. Four individuals were included in each block; two were randomly allocated to placebo, whereas the other two received PTH. As investigators, we were unaware of the number of subjects in each block and permutations within blocks during the trial. According to a predefined schedule, daily doses of vitamin D analogues and calcium supplements were downtitrated by 50% and 25%, respectively, if participants developed hypercalcemia (plasma Ca2+ > 1.40 mmol/L) or increased 24-hour urinary calcium excretion (>7.5 mmol). We unblinded the randomization code when all data were on file. At end of study (before unblinding), we asked our patients whether they believed that they had received treatment with PTH(1–84) or placebo.
Clinic visits and measurements
After randomization, participants attended our outpatient clinic at weeks 1, 2, 3, 4, 6, 8, 12, 16, 20, and 24. At all clinic visits, we recorded adverse events by questioning participants in a general, nondirective manner and drew fasting blood samples (24 hours after the last injection of study drugs). The day before each visit, participants collected a 24-hour urine sample.
At baseline and week 24, the patients completed a questionnaire regarding dietary calcium intake, use of concomitant drugs, use of vitamin and mineral supplements, medical history, previous fractures, and symptoms of hypocalcemia. Total daily calcium intake was assessed according to reported dietary intake of milk, cheese, milk products, and use of calcium supplements.14
BMD at the forearm, spine (L1–L4), hip, and whole body was measured by dual-energy X-ray absorptiometry (DXA) using the same Hologic Discovery scanner (Hologic, Inc., Waltham MA, USA) at baseline and at week 24. The coefficient of variation (CV) was 1.5% for the lumbar spine BMD, 2.1% for the femoral neck, and 1.9% at the ultradistal radius, with a high long-term stability of less than 2% per year.15, 16
We measured plasma and urinary levels of calcium, creatinine, albumin, phosphate, and magnesium by standard laboratory methods. To reduce analytical variation, we analyzed intact PTH, vitamin D metabolites, and biochemical markers of bone turnover obtained at all time points for each patient in a single batch. Blood samples were kept frozen at −80 °C and urine samples at −20 °C until the time of analysis. We calculated the renal clearance of creatinine (ClCr) from 24-hour urinary creatinine and plasma creatinine and computed the tubular reabsorption of calcium (TRCa, %) from plasma calcium, creatinine clearance, and renal calcium excretion.17 We estimated the renal maximal tubular reabsorption of phosphate per liter of glomerular filtrate (TmP/GFR, mmol/L) according to Bijvoet and colleagues.18
We measured plasma intact PTH using a second-generation electrochemiluminescent immunoassay (ECLIA) on an automated instrument (Cobas 6000, Roche Diagnostics, GmbH, Mannheim, Germany). The lower limit of detection of the assay is 0.127 pmol/L, with a total imprecision (CV, %) of 3.3% and 2.7% at PTH levels of 3.7 and 26.6 pmol/L, respectively. We analyzed plasma 25(OH)D levels by isotope dilution liquid chromatography–tandem mass spectrometry (LC-MS/MS) according to a method adapted from Maunsell and colleagues19 and described earlier in detail.20 The method quantifies both 25(OH)D2 and 25(OH)D3. Calibrators traceable to NIST SRM 972 (Chromsystems, Münich, Germany) were used. CV values (%) for 25(OH)D3 were 6.4% and 9.1% at levels of 66.5 and 21.1 nmol/L, and for 25(OH)D2 the CV values were 8.8% and 9.4% at levels of 41.2 and 25.3 nmol/L, respectively. We determined plasma 1,25-dihydroxyvitamin D [1,25(OH)2D3] levels by a radioimmunoassay (Gamma-B 1,25-Dihydroxy Vitamin D, Immunodiagnostic Systems [IDS], Ltd., Boldon, England). The CV (%) was between 6.8% and 14.0% at plasma levels in the range of 16 to 220 pmol/L.
We measured plasma osteocalcin, C-terminal telopeptide of type 1 collagen (β-CrossLaps, CTX), and N-terminal propeptide of procollagen 1 (P1NP) by ECLIA using an automated instrument (Cobas 6000 Immunoassay Analyzer, Roche Diagnostics, GmbH). We used antibodies recognizing both intact (1–49) and N-mid-osteocalcin(1–43).21 The CV values (%) for CTX were 2.97% and 1.51% at plasma levels of 0.32 and 2.77 µg/L, respectively. The CV values (%) for P1NP were 1.64% and 2.11% at plasma levels of 65.84 and 748.5 µg/L, respectively. The CV values (%) for osteocalcin were 0.97% and 1.06% at plasma levels of 25.3 and 84.1 µg/L, respectively. We measured plasma bone-specific alkaline phosphatase (BSAP) levels by an immunoassay (METRA BAP EIA Kit, Quidel Corporation, San Diego, CA, USA). The CV values (%) were 4.2% and 6.7% at concentrations of 15 and 65 U/L, respectively. The renal excretion of cross-linked N-terminal telopeptide of type 1 collagen (NTX) was quantified by ELISA (Osteomark, OSTEX/Wampole, Princeton, NJ, USA) using an automated instrument (Vitros ECI, Ortho Clinical Diagnostics, Amersham, UK).22 The CV (%) was less than 10% at 415 and 1328 nmol/L. We expressed values relative to creatinine excretion as nanomoles of bone collagen equivalents (nmol BCE) per millimoles of creatinine.
We analyzed data using an intention-to-treat (ITT) approach consisting of all randomized subjects. We assessed differences between study groups using Fisher's exact test for categorical variables and a two-sample t test or Mann-Whitney U test for continuous variables as appropriate. We used Cohen's kappa score to assess the degree of agreement between treatment allocation and patients' own beliefs on the treatment they had received. We studied serial changes using analysis of variance for repeated measurements (RM-ANOVA). We checked the assumptions for repeated-measures ANOVA by Mauchly's test of sphericity, and accordingly, we adjusted the degrees of freedom (Huynh-Feldt epsilon). In case of a significant between-group difference by RM-ANOVA, we analyzed differences between groups at each time point of measurements by a posteriori analysis using a two-sample test. In case of missing data, we imputed missing postbaseline data as the median value of the two observations on each side of the missing value. Less than 5% of data were missing. Correlations between variables were tested by bivariate correlation analysis (Pearson's correlation [r] or Spearman's rho [ρ] as appropriate). We performed all statistical tests as between-group comparisons on percentage changes from baseline values except for Z-scores, where absolute changes were calculated. We report results as mean ± SD or median with ranges (min; max) unless otherwise stated. A value of p < 0.05 was considered statistically significant. We used PASW Statistics 18 (IBM, formerly SPSS, Chicago, IL, USA) for the statistical analyses.
Table 1 shows baseline characteristics of the 62 included patients with hypoparathyroidism. The mean age of the included patients was 52 years (range 31 to 78 years), and 86% were females. A total of 58 (94%) had postoperative hypoparathyroidism. The four patients with idiopathic hypoparathyroidism all tested negative for autosomal dominant hypocalcemia (ADH) caused by a gain-of-function mutation in the CaSR gene. Two patients were treated with high-dose oral ergocalciferol droplets, one patient received calcitriol (Rocaltrol, Roche Pharma AG, Grenzach-Wyhlen, Germany), and the remaining 59 patients were on treatment with alphacalcidol (Etalpha, Leo Pharmaceutical, Leo Pharma Nordic, Malmö, Sweden). At time of inclusion, all patients also were on treatment with a daily calcium supplement (Table 1). A total of 52 (84%) patients received levothyroxine substitution. Table 2 gives the measured biochemical variables at baseline with normal reference ranges. None of the patients had hypercalcemia, whereas 42 (68%) patients had slightly reduced plasma Ca2+ levels. Thirty-three (53%) patients had subnormal PTH levels (<1.6 pmol/L), and 25 (40%) patients had levels within the lower third of the reference range. The average 24-hour urinary calcium excretion was above normal. Plasma phosphate and 24-hour urinary phosphate excretion were within the reference range. Twelve of the patients (19%) had slightly elevated plasma creatinine levels (>90 µmol/L), and 11 (18%) patients had mildly reduced ClCr (<1.17 mL/s). The two patients treated with oral ergocalciferol droplets both had very high plasma 25(OH)D levels. All patients were clinically euthyroid, but 19 (31%) had intentionally slightly reduced TSH levels (<0.3 × 10−3 IU/L) owing to a prior diagnosis of thyroid cancer (Table 2).
Table 1. Baseline Characteristics of All Included Patients With Hypoparathyroidism and Stratified by Treatment Allocation
All (n = 62)
PTH group (n = 32)
Note: Patients were randomized to PTH(1–84) 100 µg/d or similar placebo. Number of subjects (n) with percentages within group (%) or median with range (minimum–maximum).
Duration of disease, years
Table 2. Baseline Biochemical Characteristics of Included Patients With Hypoparathyroidism
All (n = 62)
Median with range (minimum–maximum).
25-Hydroxyvitamin D (nmol/L)
1,25-Dihydroxyvitamin D (pmol/L)
TSH, all (103 IU/L)
Calcium, total (mmol/L)
Calcium, ionized (mmol/L)
Creatinine clearance (mL/s)
Calcium (mmol/24 h)
Phosphate (mmol/24 h)
Magnesium (mmol/24 h)
The subjects were randomized successfully with respect to baseline characteristics (Table 1) and biochemistry (data not shown; all p values for comparisons between groups > 0.15). Fifty-seven patients completed the entire 24 weeks of study, and a further 2 patients completed 17 and 19 weeks of treatment, in whom end-of-study measurements were collected before study drugs were stopped.
Supplementation with active vitamin D and oral calcium during study
Among patients randomized to PTH treatment, the need for active vitamin D (ie, alphacalcidol/calcitriol) and calcium supplements to maintain plasma calcium levels decreased significantly throughout the 6 months of treatment (Table 3). The two patients receiving ergocalciferol were both in the placebo groups and had no change in their vitamin D dosage. Treatment with active vitamin D and calcium was stopped completely in 7 patients in the PTH arm compared with none in the placebo arm (p < 0.01). After stopping calcium supplements and active vitamin D, 5 patients needed their daily PTH dose reduced to less than 100 µg/d; that is, at end of study, they received PTH injections either every second day (n = 1), every third day (n = 1), or five times a week (n = 3). The median dose of alphacalcidol/calcitriol decreased in the PTH arm from baseline to end of study by 50% (range 0% to −100%), whereas the dose did not change (median change 0%, range −20% to +200%) in the placebo arm (p < 0.001). Furthermore, use of oral calcium supplements was stopped in 15 patients in the PTH arm compared with none in the placebo arm (p < 0.01). At the same time, the dietary intake of calcium decreased by 11% (range −19% to 0%) in the PTH arm in contrast to no change (range −10% to 10%) in the placebo arm (p = 0.02). The reduction in active vitamin D and calcium supplementation was independent of the ethiology of hypoparathyroidism (data not shown).
Table 3. Changes in Active Vitamin D and Calcium Supplementations During Study in Hypoparathyroid Patients Randomized to 100 µg/d of PTH(1–84) or Placebo as Add-on Therapy for 6 Months
PTH (n = 32)
Placebo (n = 30)
Change in median (range) dose in percent of baseline supplementation.
Change in median (range) intake in percent of baseline intake.
Changes in calcium-phosphate homeostasis and renal function
Throughout the 24 weeks of treatment, plasma Ca2+ levels were increased in the PTH arm compared with the placebo arm (p < 0.01) despite the preplanned reduction in calcium and active vitamin D supplements (Fig. 1). However, the zenith for plasma calcium was observed around weeks 8 to 12, followed by a tendency to a decrease (p = 0.07) until the end of the study. During the study, 19% of Ca2+ measurements were above the upper limit of the reference range. Among those with hypercalcemia, we found an average degree of hypercalcemia at 2.85 mmol/L (range 2.56 to 3.81 mmol/L) for total calcium and 1.48 mmol/L (range 1.33 to 1.97 mmol/L) for Ca2+ levels. In total, 96% of the hypercalcemic episodes were in the PTH arm and only 4% of in the placebo group. Hypercalcemia was ascertained in 60% of blood samples in patients receiving PTH alone after stopping active vitamin D and calcium supplements compared with 29% of those still needing supplements despite PTH treatment (p = 0.11). Figure 2 shows changes in plasma Ca2+ stratified by treatment allocation and need for calcium and active vitamin D supplementation at the end of the study. Those who only received the standard dose of PTH of 100 µg/d or less at the end of the study had higher plasma Ca2+ levels during the study than those receiving PTH and active vitamin D (p < 0.05) and those receiving both active vitamin D and calcium (p < 0.01). In the PTH treatment arm, the occurrence of hypercalcemia, defined as elevated plasma Ca2+ levels at two or more time points, was not associated with a detoriation in renal function, as measured by plasma creatinine (p = 0.69) or ClCr (p = 0.16) compared with those without hypercalcemia (data not shown).
Compared with placebo, PTH increased renal calcium excretion (p < 0.01; Fig. 1). Post hoc analysis revealed that the difference between groups was significant between weeks 2 and 8. From week 12 and until the end of the study, groups did not differ. Overall, plasma Ca2+ and 24-hour urinary calcium correlated significantly at weeks 1, 2, 3, 6, 8, and 16 (r between 0.28 and 0.40, p < 0.03 for all correlations). The daily dose of calcium from supplements did not correlate with the degree of hypercalciuria at any time point (p > 0.10 for all correlations). At the end of the study, hypercalcemia was present in 3 patients (10%) in the placebo arm and in 6 patients (21%) in the PTH arm (p = 0.48). In comparison, hypocalcemia was observed in 16 (53%) patients in the placebo arm and in 8 (29%) patients in the PTH arm (p = 0.07).
During the study, plasma phosphate levels decreased in the PTH group (p < 0.05), with significant between-group differences at weeks 3 and 8 (Fig. 1). However, the plasma calcium-phosphate product did not differ between groups (p = 0.55; data not shown).
Changes in calcitropic hormones
There was no difference between groups in plasma 25(OH)D levels (Fig. 3). However, plasma 1,25(OH)2D3 increased rapidly in the PTH group and remained elevated throughout the study (p < 0.01) despite the marked reduction in dose of active vitamin D. Plasma PTH levels are not reported because blood was drawn at its nadir level before the next injection.
Changes in renal function and handling of calcium and phosphate
Plasma creatinine levels (p = 0.11) and creatinine clearance (p = 0.17) did not differ between groups during study (data not shown). Neither did treatment cause changes in TRCa (p = 0.13) or TmP/GFR (p = 0.38; Fig. 2).
Changes in BMD and bone markers
Baseline BMD and Z-scores are shown in Table 4. The Z-scores were significantly increased above 0 at the whole body (p < 0.001), the spine (p < 0.001), the total hip (p < 0.001), the femoral neck (p < 0.001), the ultradistal forearm (p < 0.01), and the distal 1/3 forearm (p < 0.01) but not at the middle (p = 0.93) or total forearm (p = 0.41). Following PTH(1–84) treatment, BMD decreased significantly at the whole body (p < 0.05), the spine (p < 0.05), the hip (p < 0.01), and the femoral neck (p < 0.01) but not at the forearm (Table 4).
Table 4. Baseline BMD (g/cm2) With Z-Scores in All Included Patients With Hypoparathyroidism
Baseline (n = 62)
Changes between baseline and week 24
PTH(n = 29)
Placebo (n = 30)
Note: Changes in BMD are given in percentage, and changes in Z-scores are given in absolute values. Changes between baseline and week 24 are stratified by treatment allocation. Median with range (minimum–maximum).
p < 0.001 compared with norm-based age- and gender-matched controls.
p < 0.01 compared with norm-based age- and gender-matched controls.
At baseline, subnormal levels were observed in 5% of participants for plasma CTX (< 0.03 µg/L), 69% for NTX (<17 nmol BCE/mmol creatinine), 21% for osteocalcin (<10 µg/L), 32% for P1NP (20 µg/L), and 2% for BSAP (<10 U/L; data not shown). Following PTH treatment, all bone markers increased significantly (p < 0.01 for all markers) throughout the study, with a tendency to level off for CTX and P1NP at the end of the study (Fig. 4). Inspection of the curves revealed a tendency toward an increase in formative bone markers before the increase in resorptive markers. In the total group, changes in whole-body BMD correlated inversely with changes in renal NTX/creatinine excretion (r = −0.35, p < 0.05), serum BASP (r = −0.29, p < 0.05), osteocalcin (r = −0.31, p < 0.05), and P1NP (r = −0.20, p < 0.05). Similarly, changes in spine, total-hip, and femoral neck BMD correlated inversely with changes in plasma CTX, BASP, osteocalcin, and P1NP (p < 0.05 to 0.01; data not shown).
Side effects and dropouts
All patients randomized to placebo and 27 of the 32 patients randomized to the PTH completed the entire 24 weeks of treatment. One patient never started PTH treatment, and two dropped out following 1 and 23 weeks of treatment for personal reasons. Further, two patients dropped out following 17 and 19 weeks of treatment, one owing to nausea that persisted after the end of the study and one owing to dizziness and a feeling of variations in plasma calcium levels.
In the placebo and PTH groups, a total of 74 and 139 adverse events were reported, respectively. One or more adverse events were reported by 30 (97%) of the patients who started PTH-RT and by 27 (90%) of the patients in the placebo group. Stratification by type of adverse event showed that only nausea occurred with a significantly higher frequency in the PTH group than in the placebo group (Table 5).
Table 5. Side Effects and Serious Adverse Events (SAEs) in Hypoparathyroid Patients Randomized to Either 100 µg/d PTH(1–84) or Placebo for 6 Months
PTH (n = 32)
Placebo (n = 30)
Note: Number of patients (% of patients in the treatment group).
Types of infections. A total of 33 events in 27 patients.
Hypocalcemia (n = 1), hypercalcemia (n = 1), left shoulder surgery (n = 1), erysipelas (n = 1), and arterial occlusion in the right femoral artery (n = 1).
Psychosis (n = 1), surgery for goiter (n = 1), and anaphylactic reaction (n = 1).
The most common complaints were infections, musculoskeletal aches, paresthesias, and headache followed by cardiovascular complaints. In the PTH group, 11 patients had a total of 17 episodes of symptomatic hypercalcemia, whereas only one episode occurred in the placebo group.
In total, 8 significant adverse events were recorded, among which 5 occurred in the PTH group and 3 in the placebo-group. There was no difference in the nature of the significant adverse events (Table 5). One significant adverse event was due to hospitalization because of hypercalcemia, and another was due to hospitalization because of hypocalcemia, whereas the remaining six was due to other reasons not judged to be related to treatment (Table 5).
At the end of the study 79% of the patients (23 of 29 responders) in the PTH group reported that they believed they had been on treatment with PTH. Among the patients in the placebo group, 57% (16 of 28 responders) believed that they had been treated with placebo, and 43% thought that they had received PTH (Cohen's kappa score 0.366, p < 0.01).
Our study showed that by adding PTH(1–84) to conventional treatment with calcium and active vitamin D analogues, plasma calcium and phosphate levels are maintained within the normal physiologic range, and the need for calcium supplements and active vitamin D analogues is reduced after a run-in period. Moreover, our study showed that in patients with hypoparathyroidism on conventional treatment, bone turnover is very low, and BMD is increased. PTH-RT increased bone turnover with a concomitant decrease in BMD, which may point toward a normalization of bone metabolism during long-term treatment.
Calcium and phosphate homeostasis
We found elevated levels of plasma Ca2+ during the first 5 months of the study in the PTH treatment arm compared with the placebo arm. Since we administered PTH treatment as an add-on therapy, this is to be expected because we only downtitrated the daily dose of calcium and active vitamin D if hypercalcemia developed. Importantly, plasma Ca2+ levels were within the normal physiologic range at the end of the study in all patients whether they were treated with PTH alone, PTH and active vitamin D, or PTH, active vitamin D, and calcium in combination.
The renal calcium excretion increased during the initial 8 weeks of PTH treatment probably owing to an increased filtered calcium load caused by the raised plasma calcium levels. The increased load is easily explained by an enhanced mobilization of calcium from the skeleton, as evidenced by the rise in bone-turnover markers and the decrease in BMD and an increased intestinal calcium absorption caused by the increase in plasma 1,25(OH)2D3. There were no meaningful changes in TRCa. This may be explained by the synergistic effects of PTH and 1,25(OH)2D3 that increase TRCa1 and the increased filtered load that decreases TRCa through an effect on the CaSR in the renal tubule cells.1, 23, 24
The PTH(1–84) treatment caused an early decrease in plasma phosphate compared with placebo without significant changes in 24-hour urinary phosphate excretion or TmP/GFR. Other studies have reported no changes in plasma or urine phosphate following PTH(1–34)10, 12 or a reduction in plasma phosphate following PTH(1–84) compared with baseline.13 It is well established that TmP/GFR is the main determinant of plasma phosphate.18 In accordance, post hoc analysis in this study showed that TmP/GFR was significantly reduced at weeks 1, 2, 3, 8, and 12 (p < 0.05). The lack of overall significance may be explained in part by the large variance in TmP/GFR, which is calculated based on four independent biochemical measurements. However, the patients also were in an unstable condition during titration with an increased renal calcium load, which may affect the renal phosphate handling through the CaSR.24
The main regulator of the renal phosphate handling is PTH and fibroblast growth factor 23 (FGF23) produced in the osteocytes that both increase phosphaturia and lower plasma phosphate.25–27 The skeletal production of FGF23 is upregulated by 1,25(OH)2D3 in a negative-feedback loop.25, 27 In this study PTH treatment increased the renal production of 1,25(OH)2D, which therefore could lead to higher FGF23 levels. In this context, it would be of interest to measure FGF23 levels in our patients. The normal 24-hour phosphate excretion may be explained by a tradeoff between a reduced renal filtered load because of lowered serum phosphate and a reduced tubular reabsorption of phosphate.
In this study, plasma 1,25(OH)2D3 increased without significant overall changes in plasma 25(OH)D levels. This is in accordance with the established effect of PTH(1–84) on the renal 25-hydroxyvitamin D-1-α-hydroxylase.28, 29 It is noteworthy, however, that the rise in 1,25(OH)2D3 was observed despite a marked reduction in the oral supplementation with active vitamin D. Our findings are in contrast to a previous report showing unchanged 1,25(OH)2D3 following PTH(1–84) 100 µg every other day.13 Post hoc analysis of the individual time points in our study revealed a significant decrease in plasma 25(OH)D at all time points after week 12 (p < 0.05). This could be related to the reduction in calcium supplementation because vitamin D3 was included in the calcium carbonate tablets. Another explanation would be that the increase in 1,25(OH)2D3 accelerates the degradation of 25(OH)D by enhancing 24-hydroxylase activity.30
A proportion of our patients had reduced bone turnover at baseline. Following PTH treatment, our study showed a rapid increase in all measured bone markers, as also observed in previous studies.8–13, 31 PTH has two established major effects on the skeleton. Chronically raised PTH levels, as in primary hyperparathyroidism, may enhance bone turnover with an early increase in bone resorption followed by a delayed increase in bone formation, resulting in a variable decreased bone mass and BMD.32 However, in patients with osteoporosis, daily injections of PTH(1–34) or PTH(1–84) have anabolic effects with an increase in bone formation later, followed by an increase in resorption.33 This anabolic window33 between formation and resorption leads to an increase in bone mass except for the forearm.34, 35 In this study, we observed no clear-cut time separation between the rise in anabolic and resorptive bone markers. The immediate anabolic effect of PTH(1–84) in hypoparathyroidism is in accordance with histomorphometric findings showing an increase in labeled surfaces, mineral appositional rate (MAR), and bone-formation rate (BFR) already 3 months after treatment start at cancellous, endocortical, and intracortical surfaces.31 One important safety issue regarding the increase in bone turnover is whether the remodeling stays increased and continuous to reduce BMD. However, in the 3-year controlled study by Winer and colleagues10 on PTH(1–34) treatment in hypoparathyroidism, the authors observed a decrease in bone markers after 2.5 years of treatment. This suggests that the decrease in BMD will be limited over time. However, further studies on the mechanism of bone loss during PTH(1–84) treatment and on the outcome of prolonged treatment are necessary.
At baseline, our patients had an increased BMD at all measured sites except the total forearm and the middle forearm. This is in agreement with several cross-sectional studies10, 36, 37 and is probably caused by the low bone turnover, which increases mean bone age and decreases the remodeling space,36, 38 and structural changes with an increase in bone volume, trabecular width, and cortical width.36 Following treatment, BMD decreased significantly at all sites except the forearm. This decrease in BMD can be explained by the increase in bone turnover, which again expands the remodeling space and a decreased mean bone age, although specific structural changes have not been excluded. Our observation time is too short to predict the long-term effects on the skeleton of PTH(1–84) treatment. Nevertheless, our findings of a decreased BMD are in contrast to previous findings of an unchanged or even increased BMD in response to PTH-RT in hypoparathyroidism.10, 12, 13
Comparisons of our study with previous studies on PTH-RT in hypoparathyroidism
There may be several reasons for the different findings between studies. First, the initial studies by Winer and colleagues8–12 all were conducted with PTH(1–34). Although no head-to-head study has been performed, this truncated PTH molecule may have a shorter plasma half-life than PTH(1–84) (around 1 hour versus 1.5 hours), may peak earlier in plasma (0.5 hour versus 1 to 2 hours), and may result in a faster peak in plasma calcium (4 to 6 hours versus 6 to 8 hours). Plasma calcium also returns to predose levels after 16 to 24 hours with PTH(1–34) versus 24 hours with PTH(1–84).39, 40 In accordance, two studies9, 11 investigating the effect of PTH(1–34) once a day versus twice a day found a better calcium regulation in the twice-a-day treatment arm and more episodes of hypocalcemia in the once-a-day arm. The finding that PTH(1–84) can be used as every-second-day treatment also supports the idea that this compound has a more prolonged biologic effect.13 However, it should be stressed that there are no major pharmacokinetic or pharmacodynamic studies on PTH(1–84) in hypoparathyroidism. Moreover, PTH(1–84) in our study was given in the thigh, which would tend to decrease absorption compared with injection in the abdominal subcutis.
Second, the design of previous controlled studies on PTH(1–34) treatment in hypoparathyroidism8–12 differed from our design, using either an individual once-a-day PTH(1–34) dosing regimen without calcium supplementation8 or a twice-a-day PTH(1–34) dosing regimen with10 or without12 oral calcium. In all studies, the dose of PTH(1–34) had been adjusted in such a manner that none of the patients needed concomitant treatment with calcitriol. Initial titration was performed during a 1- to 2-week inpatient period followed by an outpatient period during which subsequent small dose adjustments were performed. Compared with our study, this dose titration regime resulted in fewer episodes of hypercalcemia during the initial part of the study and in some8, 10 but not all12 of the studies a decreased urinary calcium excretion. Moreover, BMD was unchanged in response to 3 years of treatment with PTH(1–34).10, 12
In contrast to the studies by Winer and colleagues,8–12 we did not have access to different doses of PTH(1–84), and accordingly, we administered PTH(1–84) as an add-on treatment in a fixed daily dose. A similar approach with a fixed dose of PTH(1–84) added to conventional treatment was used in the cohort study by Rubin and colleagues,13 although PTH(1–84) was administered only every second day in this study. In both studies, PTH(1–84) treatment significantly reduced the need for calcium and alphacalcidol/calcitriol while maintaining plasma calcium levels reasonable and within the normal range at the study end. Both studies had preplanned but slightly different schedules for downtitration of calcium and active vitamin D. In both studies, episodes of hypercalcemia occurred with a relatively high frequency during the first 5 months, during which calcium and active vitamin D were downtitrated if hypercalcemia developed. Neither of the studies found a reduced urinary calcium excretion in response to treatment. However, in contrast to our study, Rubin and colleagues13 reported a significantly increased BMD in the spine with unchanged BMD at the hip and a decreased BMD at the forearm. Several reasons may explain the discrepant results. It may be that PTH(1–84) affects bone metabolism differentially if administered once a day as compared with every second day. If so, the more frequent injection scheme, as used by us, may cause a more catabolic effect on the skeleton than an intermittent dosing regimen.32, 34, 35 Furthermore, the inclusion of a control group in our study could affect the relative outcome because BMD values in this group tended to increase during study, as also observed by Winer and colleagues.10 Finally, studies differ in terms of characteristics of included patients. Patients with autosomal dominant hypocalcemia (ADH) were allowed to be included in previous studies.8–13 ADH is caused by a gain of function in the CaSR gene. PTH secretion is downregulated because of the increased sensitivity to Ca2+, leading to a new steady state characterized by hypocalcemia and low PTH levels. However, the mutation also influences renal calcium handling41 and may affect skeletal metabolism.42
Treatment with PTH(1–84) resulted in episodes of hypercalcemia during titration. However, compared with treatment with active vitamin D metabolites, hypercalcaemic symptoms reportedly were less pronounced and of shorter duration. Renal function remained stable, and there was no difference between groups in the occurrence of serious adverse effects. We found a significantly higher number of adverse effects in the PTH(1–84) treatment group, which mostly consisted of hypercalcemic symptoms (ie, headache, thirst, dizziness, and muskoskeletal complaints). This was to be expected during the titration period. However, only 5 of the 8 subjects in the PTH(1–84) group with nausea had hypercalcemia at the time of the complaint.
Strengths and limitations to the study
The major strength of our study is that it is the first randomized, placebo-controlled study on the effects of exogenous PTH(1–84) in hypoparathyroidism. Also, our study is the largest so fare performed and the first double-blinded study on PTH-RT in hypoparathyroidism. Although most patients in the active treatment group correctly guessed that they received treatment with PTH(1–84), the importance of performing the study in a double-blind manner is emphasized by the fact that half the patients in the placebo group thought that they also were on active treatment. Our study lasted only for 24 weeks, which is too short to ascertain long-term variations in calcium homeostasis, bone turnover, and bone mass. During the study, we were hampered by the fact that we had only one standard dose of PTH for injection of 100 µg. In 5 patients, we had to reduce the PTH dose to less than 100 µg/d. Furthermore, 8 patients received active vitamin D and 15 received active vitamin D and calcium at study end together with the standard dose of PTH. In comparison with a previous open-label cohort study on PTH(1–84) treatment in only hypoparathyroidism that used every-second-day treatment,13 we aimed at giving daily doses. However, our findings underscore the importance of having PTH injection pens with different doses in order to individualize the treatment. At present, an ongoing study (REPLACE) is focusing on defining individual replacement doses of PTH(1–84) (clinicaltrial.gov No. NCT00732615).
During the study we used a predefined titration scheme for reduction in calcium and active vitamin D supplementation that allowed for reduction only in case plasma Ca2+ increased above 1.40 mmol/L. In view of the frequent occurrence of hypercalcemia, especially in the first part of the treatment period, this scheme probably was too conservative. However, we chose to do so in order to avoid episodes of symptomatic hypocalcemia during downtitration of conventional therapy because many patients find such symptoms to be very unpleasant.
Side effects were reported more frequently in the PTH group than in the placebo group. Although only nausea was reported significantly more often in the PTH than in the placebo group, this does not exclude that the frequency of other side effects also differed between treatment groups. Sample size may not have been large enough to detect significant differences in the occurrence of adverse events. The relatively high frequency of side effects may be related in part to elevated plasma Ca2+ levels. Long-term large studies are warranted to determine long-term safety, including whether the high frequency of adverse events persists after normalization of plasma Ca2+ levels.
Combining our results with the findings from previous studies, it seems that PTH-RT may increase, decrease, or cause no effects on BMD. Apparently, the BMD response is highly sensitive to the dosing regimen used and may as well differ according to PTH analogue tested. However, in all studies performed so far, bone turnover has increased in response to treatment, indicating a reversal of the abnormal low bone turnover in hypoparathyroidism. This may point toward a more physiologic normal bone metabolism, and the demonstrated bone loss may be seen as self-limited, but this remains an open issue.
Overall, our study demonstrated that PTH(1–84)-RT administered as an add-on to conventional treatment reduces the need for calcium and active vitamin D while maintaining plasma calcium levels within the normal physiologic range. Moreover, PTH replacement reduced the elevated plasma phosphate levels, enhanced renal 1,25(OH)2D3 production, and maintained unchanged calcium phosphate product levels. Regarding the increased number of side effects and hypercalcemic episodes, long-term treatment trials with a more sensitive dosing regimen than ours are needed to establish safety before PTH-RT may be recommended as standard treatment in hypoparathyroidism.
Leif Mosekilde is a principal investigator on the REPLACE study initiated by NPS Pharmaceuticals. All other authors state that they have no conflicts of interest.
We are indebted to the patients who participated in this study and made it possible. The technical assistance of Tove Stenum, Lisbeth Flyvbjerg, Christina Wiegers, Lene Sørensen, and Helle Thøgersen is greatly appreciated. Also, we would like to thank the staff at the pharmacy of Aarhus University Hospital for helping us with the study medication. We thanks Nycomed for supplying the study drugs [PTH(1–84) and similar placebo). The Danish Council for Independent Research in Medical Science, the Novo Nordic Foundation, and the Central Denmark Region Foundation are acknowledged for financial support
Hypoparathyroid Study Group: Besides of the authors of this article, the following study group has contributed to the project: Helle Brockstedt, Silkeborg Regional Hospital, Silkeborg, Denmark; Lis Stilgren, Odense University Hospital, Svendborg, Denmark; Kjeld Hasselström, Regional Hospital West Jutland, Herning, Denmark; Jeppe Gram, Regional Hospital South Western Jutland, Esbjerg, Denmark; Henning Kaspersen Nielsen, Regional Hospital of Randers, Randers, Denmark; and Charlotte Ejersted, Hospital Lillebaelt, Fredericia, Denmark.
Authors' roles: Study design: Rejnmark, Sikjaer, and Mosekilde. Study conduct: Sikjaer, Rolighed, and Rejnmark. Data collection: Sikjaer, Rolighed, Rejnmark, and the Hypoparathyroid Study Group. Data analysis: Sikjaer, Rejnmark, Heickendorff, and Mosekilde. Data interpretation: Sikjaer, Rolighed, Rejnmark, Heickendorff, and Mosekilde. Drafting manuscript: Sikjaer, Rejnmark, Rolighed, and Mosekilde. Revising manuscript content: Sikjaer, Rolighed, Rejnmark, Heickendorff, and Mosekilde. All authors approved the final version of the manuscript. Sikjaer takes responsibility for the integrity of the data analysis.