Untreated, hypoparathyroidism (hypoPT) is a state of hypocalcemia with inappropriately low plasma parathyroid hormone (PTH) levels and hyperphosphatemia. PTH administration normalizes plasma calcium and phosphate levels and reduces the doses of calcium and active vitamin D analogues needed. To develop an evidence-based clinical algorithm to monitor hypoPT patients treated with recombinant human PTH (rhPTH[1-84]) injected subcutaneously in the thigh, we performed a 24-hour monitoring study of pharmacokinetic and pharmacodynamic effects in a group of 38 patients who had completed a 6-month randomized study on effects of treatment with a fixed rhPTH(1-84) dose of 100 µg/d or similar placebo as an add-on to conventional treatment. PTH levels rose immediately, reaching a median peak level of 26.5 (interquartile range [IQR], 20.1–42.5) pmol/L 15 minutes following injection. Thereafter, levels gradually decreased until reaching predosing levels after 16 hours, with a plasma half-life of 2.2 (1.7–2.5) hours. rhPTH(1-84) changed the diurnal rhythms of ionized calcium levels and 1,25-dihydroxyvitamin D (1,25[OH]2D) levels, with rising levels following injection. Ionized calcium peaked after 7.0 (5.0–10.0) hours. Asymptomatic hypercalcemia was present in 71% of the rhPTH(1-84)-treated patients. Compared with placebo, 24-hour urinary calcium, phosphate, and magnesium did not change, although the diurnal variation in renal excretion rates changed significantly in response to treatment. In conclusion, as a safety precaution, we recommend occasionally measuring calcium levels at approximately 7 hours after administration in order to reveal episodes of hypercalcemia. A 100-µg daily dose of rhPTH(1-84) appears to be too high in some patients, suggesting a need for a device allowing for individual dose adjustments. © 2013 American Society for Bone and Mineral Research.
Hypoparathyroidism (hypoPT) is a disease in which the synthesis of parathyroid hormone (PTH) is insufficient to maintain normal plasma calcium levels. Conventional treatment consists of active vitamin D and calcium supplements. Previous studies have demonstrated that PTH treatment is also a therapeutic option, because preinjection serum calcium levels can be maintained within the physiological range in patients with hypoPT by treatment with truncated recombinant human PTH (rhPTH[1-34])[2-6] or rhPTH(1-84)[7, 8] as either a substitution for or as an add-on therapy to conventional treatment.
The pharmacodynamic (PD) effects of rhPTH(1-34) on diurnal variations in indices of calcium homeostasis in patients with hypoPT have been investigated in several studies by Winer and colleagues[2, 3, 5] and others, showing a peak in plasma calcium concentrations 4 to 8 hours after subcutaneous (sc) injection into the extremities. A similar effect on plasma calcium levels has been reported in healthy subjects and in patients with osteoporosis treated with rhPTH(1-84) injected sc into the abdomen.[10-12] However, in hypoPT, rhPTH(1-84) is preferably injected sc into the thigh in order to obtain a slower absorption, and thereby a prolonged duration of action. So far, no data have been published on pharmacokinetic (PK) and PD effects of treatment with rhPTH(1-84) in patients with hypoPT.
Recently, we completed a trial in which 62 patients with hypoPT were randomized to 6 months of treatment with either 100 µg of rhPTH(1-84) or similar placebo administered as a once-per-day sc injection into the thigh. We chose this site because we expected a slower rate of absorption from the thigh compared to the abdomen or the arm, thereby maintaining therapeutic drug concentrations for a prolonged period.[13, 14] Plasma calcium levels, measured 24 hours after the last injection, increased significantly in response to the rhPTH(1-84) treatment, and 19% of the plasma calcium measurements during the 6 months of treatment were above the upper limit of the reference range. As a result of these PD effects of rhPTH(1-84), we hypothesized that high plasma calcium levels may occur with an even higher frequency in the hours following injection than at 24 hours after the last injection. In order to develop an evidence-based clinical algorithm to monitor patients with hypoPT on treatment with rhPTH(1-84) injected sc in the thigh, we performed a substudy on PK/PD effects. In this substudy we included 38 of our study subjects who were admitted to a full 24-hour inpatient stay at our hospital at the end of the 6-month treatment period. The results from this substudy are reported here, including data on the PK profile of PTH and PD data on plasma and urinary levels of calcium, phosphate, magnesium, and plasma 1,25-dihydroxyvitamin D (1,25[OH]2D). The study also included assessment of potential effects of rhPTH(1-84) treatment on cardiovascular health (safety) as determined by electrocardiograph (ECG) recordings and measurement of blood pressure. Moreover, in the present work we included data on the effect of the 6 months of treatment on plasma and urinary magnesium, because these data have not previously been reported.
Patients and Methods
As previously detailed, our main study included 62 patients (9 men and 53 women) aged 31 to 78 years with chronic hypoPT, who were enrolled in an investigator-initiated, double-blind, randomized, placebo-controlled, parallel-group study on the effects of rhPTH(1–84) treatment. The patients received either 100 µg rhPTH(1–84) (Preotact; Nycomed, Zürich, Switzerland) or identical placebo administered as an sc injection in the thigh once per day for 6 months. The study drug was given as an add-on to the patients' conventional treatment with oral calcium and active vitamin D. The doses of calcium and active vitamin D were reduced as necessary following a predefined down-titration scheme. Inclusion and exclusion criteria have been reported in detail. At the end of the 6-month treatment trial, 39 patients consented to participate in our substudy on diurnal variations and were admitted to a 24-hour inpatient stay at our hospital. The present study was an extension of the 6-month trial and measurements on diurnal variations were accordingly performed on patients who had been exposed to PTH for 6 months. The patients were maintained in their respective treatment groups and were not rerandomized. They went on to the 24-hour study after completing the 6-month trial with no drug holiday between, and the double-blind design was maintained. The study was not unblinded until the last patient had completed both the 6-month trial and the 24-hour study. One of the patients was excluded post hoc due to severe concerns about the patient's compliance with the study drug. 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. (Public Clinical Trial Registration: http://clinicaltrials.gov/show/NCT00730210. Treatment of Hypoparathyroidism With Subcutaneous PTH (1-84) Injections: Effects on Muscle Function and Quality of Life. European Union Clinical Trials Register: https://www.clinicaltrialsregister.eu/ctr-search/search?query=2008-000606-36 EudraCT Number: 2008-000606-36 Sponsor Protocol Number: 84421383.)
Patients were hospitalized in the morning (approximately 8:00 a.m.) and baseline assessments (t = 0) were performed prior to injection of the study drug, with the participants having fasted since midnight. At t = 0 we measured plasma levels of 25-hydroxyvitamin D (25[OH]D), 1,25(OH)2D, ionized calcium (Ca2+), phosphate, creatinine, magnesium, and PTH, in addition to blood pressure and heart rate. An ECG recording was also performed. Immediately after the baseline measurements, the patients injected themselves with 100 µg of rhPTH(1-84) or placebo sc in the thigh under supervision of the investigator. Subsequently, blood samples were drawn at the following time points after injection of study drugs: 15 and 30 minutes, then 1, 1.5, 2, 3, 4, 5, 6, 7, 8, 10, 12, 16, 20, and 24 hours. Urine was collected in six separate collections for the time intervals of 0 to 2, 2 to 4, 4 to 6, 6 to 8, 8 to 16, and 16 to 24 hours. ECG recordings and blood pressure measurements were repeated after 1 and 10 hours. Investigations were performed on 12 separate days, some days 1 subject was studied and on other days as many as 6 subjects were studied. Standardized meals consisting of the same ingredients were served on each of the study days. No restrictions were applied on the amount of food consumed by the individual participants.
The patients took their calcium and active vitamin D supplements as usual, either once a day (at breakfast time) or divided into two times a day (at breakfast and dinner time) or three times a day (at breakfast, lunch, and dinner time).
For blood sampling, an indwelling cannula was placed in the medial cubital vein. Before every sampling, about 1 mL blood was drawn and discarded and the cannula was flushed with saline after each collection. No heparin was used to flush the catheter. Following sampling, blood was processed by centrifugation and divided into aliquots, which were stored at −80°C within 1 hour. Similarly, urine samples were stored at −20°C until analyzed.
We measured plasma and urinary levels of calcium, creatinine, albumin, phosphate, and magnesium by standard laboratory methods. Plasma Ca2+ levels were adjusted to a pH value of 7.4. To reduce analytical variation, we analyzed PTH and 1,25(OH)2D levels for each patient in a single batch.
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 (coefficient of variation [CV]) of 3.3% and 2.7% at PTH(1-84) levels of 3.7 pmol/L 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 colleagues and described in detail. The method quantifies both 25(OH)D2 and 25(OH)D3. In this paper, total vitamin D levels are reported. Calibrators traceable to National Institute of Standards and Technology (NIST) Standard Reference Materials (SRM) 972 (Vitamin D in Human Serum; Chromsystems, Munich, Germany) were used. CVs for 25(OH)D3 were 6.4% and 9.1% at levels of 66.5 and 21.1 nmol/L, respectively; the CV values for 25(OH)D2 were 8.8% and 9.4% at levels of 41.2 and 25.3 nmol/L, respectively. We determined plasma 1,25(OH)2D levels by radioimmunoassay (Gamma-B 1,25-Dihydroxy Vitamin D; Immunodiagnostic Systems (IDS) Ltd, Boldon, UK). The CV was between 6.8% and 14.0% at plasma levels in the range of 16 to 220 pmol/L.
ECG and blood pressure
ECG recordings, pulse, and blood pressure (BP) measurements were performed after 10 minutes of rest. BP was measured in millimeters of mercury (mmHg) and heart rate in beats per minute (bpm) using the same electronic device (Microlife BPA100 plus; Microlife, Widnau, Switzerland) at each measurement. ECG recordings were performed as a 12-lead surface ECG (Schiller Cardiovit AT-101; Schiller, Baar, Switzerland) and results were analyzed by a cardiologist blinded to the randomization code. The PR interval, which starts at the beginning of the P wave and ends at the beginning of the QRS complex, was read from lead V2 and the rest were read from lead II. The QT interval is measured from the beginning of the QRS complex to the end of the T wave and because of it's inverse relationship to heart rate, the measured QT intervals were corrected for heart rate using Fridericia's correction:
where the RR interval is the interval between two successive R waves, which under normal circumstances equals heart rate.
The Fridericia's correction is a formula that takes into account the physiologic shortening of the QT interval which occurs as the heart rate increases, permitting comparison of the QT interval across a range of rates. QTcF is a corrected QT interval that would be observed at a heart rate of 1 cycle per second.
Because some of our patients had low but measurable endogenous PTH secretion (although insufficient to maintain a physiological calcium homeostasis), we performed PK analyses on baseline (t = 0) adjusted data; ie, plasma PTH levels below the level measured at t = 0 were omitted from the analyses. For each study subject, total area under the plasma drug concentration-time curve (AUC0–t) was calculated for plasma PTH concentrations using trapezoid integration, and clearance (CL) was calculated by dividing the dose by AUC0–t. For each study subject, plasma PTH concentrations were log-transformed and plotted against time. Using linear regression, a straight line was plotted and the plasma half-life (T½) was determined. Because the study drug was injected sc, it took a certain time (Tmax) before maximal plasma concentrations (Cmax) were obtained. Accordingly, T½ values were calculated from the time of the highest plasma PTH concentration; ie, if PTH concentrations measured at t = 0.25 hours or t = 0.5 hours were lower than the concentration measured at t = 1 hour, the initial measurements were omitted from this calculation.
Serial changes in measured biochemical indices were calculated using analysis of variance for repeated measurements (RM-ANOVA), with treatment group as the independent variable. Assumptions for analyses were checked by Mauchly's test of sphericity, and if required the degrees of freedom (Huynh-Feldt epsilon) were adjusted. Statistical inferences were based on tests of time versus group interaction measures. In case of a significant between groups difference by RM-ANOVA, differences between groups at each time point of measurements were analyzed by a posteriori analysis using a two-sample test. For urinary variables, individual median excretion per hour was calculated for each collection interval.
Correlations between variables were tested by bivariate correlation analysis (Spearman's rho, ρ).
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 report results as mean ± SD or median with interquartile range (IQR, 25% to 75%) unless otherwise stated; p < 0.05 was considered statistically significant. We used IBM SPSS Statistics version19 (IBM, New York, NY, USA) for the statistical analyses.
In this substudy, the 38 patients included had received treatment with either rhPTH(1-84) (n = 21) or placebo (n = 17) for 30.0 ± 4.6 weeks. Characteristics of studied subjects are shown in Table 1. In brief, 37 patients had postsurgical hypoPT; 1 patient had idiopathic hypoPT, but had tested negative for activating mutations in the calcium-sensing receptor (CaSR) gene. The median time span from diagnosis to entering the 24-hour study was 6 (IQR, 1.5–33.5) years.
|PTH group (n = 21)||Placebo group (n = 17)||p|
|Gender, n (%)||0.82|
|Male||4 (19)||2 (12)|
|Female||17 (81)||15 (88)|
|Age (years), median (range)||57 (37–75)||47 (32–68)||0.20|
|BMI (kg/m2), median (range)||28.2 (19.4–41.5)||29.6 (17.7–44.2)||0.44|
|Etiology, n (%)||0.08|
|Atoxic goiter||12 (57)||4 (24)|
|Toxic goiter||4 (19)||7 (41)|
|Cancer||3 (14)||3 (18)|
|PHPT||1 (5)||3 (18)|
|Idiopathic||1 (5)||0 (0)|
|Duration of disease (years), median (range)||7 (3–34)||6 (2–32)||0.59|
|Alfacalcidol, n (%)||17 (81)||16 (94)|
|Dose (µg/d)||0.75 (0.0; 2.5)||1.75 (0; 4)||<0.01|
|Ergocalciferol, n (%)||0 (0)||1 (6)|
|Dose (µg/d)||(–)||0 (0; 2143)||0.27|
|Cholecalciferol D3, n (%)||12 (57)||17 (100)|
|Dose (µg/d)||5 (0; 38)||20 (0; 50)||<0.05|
|Calcium, n (%)||10 (48)||17 (100)|
|Dose (g/d)||0 (0; 3300)||1500 (400; 3000)||<0.001|
|Magnesium, n (%)||1 (5)||4 (24)|
|Dose (g/d)||0 (0; 360)||0 (0; 1000)||0.09|
|Thiazides, n (%)||1 (5)||1 (6)||0.70|
|Levothyroxine, n (%)||16 (76)||14 (82)||0.48|
|Systolic (mmHg)||143 (128; 151)||135 (121; 144)||0.28|
|Diastolic (mmHg)||86 (76; 88)||78 (72; 82)||0.19|
|Heart rate (beats per minute)||69 (56; 78)||58 (53; 64)||0.01|
|PR interval (ms)||161 (134; 181)||175 (165; 185)||0.07|
|QTcF interval (ms)||421 (402; 444)||412 (402; 442)||0.37|
|Normal, n (%)||15 (71)||14 (82)||0.48|
At initiation of the 6-month study, the group of patients randomized to rhPTH(1-84) treatment did not differ from the group randomized to placebo. On the day of study, the participants in the active treatment group used a significantly lower daily dose of vitamin D analogues and had a lower intake of calcium from supplements (Table 1). Table 2 shows the biochemical profile at t = 0 on the day of study.
|Reference range||PTH group (n = 21)||Placebo group (n = 17)||p|
|Calcium, total (mmol/L)||2.00–2.55||2.28 (2.10; 2.39)||2.21 (2.05; 2.29)||0.21|
|Calcium, ionized (mmol/L)||1.18–1.32||1.18 (1.13; 1.29)||1.20 (1.14; 1.25)||0.78|
|Phosphate (mmol/L)||0.76–1.41||1.08 (0.99; 1.23)||1.18 (1.12; 1.24)||0.10|
|Magnesium (mmol/L)||0.70–1.10||0.81 (0.73; 0.86)||0.89 (0.84; 0.91)||<0.01|
|PTH (pmol/L)||1.6–6.9||1.07 (0.83; 1.49)||1.73 (1.27; 2.57)||0.02|
|25-hydroxyvitamin D (nmol/L)||50–160||65 (61; 83)||90 (78; 118)||<0.01|
|1,25-dihydroxyvitaminD (pmol/L)||60–180||128 (103; 194)||103 (78; 116)||0.01|
|Creatinine (µmol/L)||45–90||73 (67; 81)||75 (69; 81)||0.62|
As shown in Fig. 1, the median PTH concentration reached a maximum concentration (Cmax) of 26.5 pmol/L after a median Tmax of 15 minutes after the injection (Table 3). However, by visual inspection of the individual time-concentration curves, a biphasic pattern with two peaks in plasma PTH concentrations was evident in most subjects (76%). Within this group, a first peak with a median Cmax(1) of 26.5 (IQR, 20.7–39.1) pmol/L was evident after a Tmax(1) of 15 (IQR, 15–30) minutes, whereas a second peak with a Cmax(2) of 18.4 (IQR, 13.4–25.4) pmol/L occurred at a Tmax(2) of 120 (IQR, 90–120) minutes.
|PTH group (n = 21)||Placebo group (n = 17)||p|
|T½ (hours)||2.2 (1.7; 2.5)||–||–|
|Tmax (minutes)||15 (15; 90)||–||–|
|Cmax (pmol/L)||26.53 (20.08; 42.46)||–||–|
|PTH/kg body weight (µg/kg)||1.18 (0.96; 1.34)||–||–|
|AUC0–t (pmol*h/L)||334 (240; 451)||–||–|
|PTH clearance rate (L/h)||32 (24; 44)||–||–|
|Ca2+ AUC0–24h (mmol*h/L)||1853 (1804; 2032)||1733 (1685; 1795)||0.003|
|Ca2+ Cmax (mmol/L)||1.37 (1.32; 1.49)||1.25 (1.24; 1.28)||<0.001|
|Ca2+ Tmax (hours)||7.0 (5.0; 19.0)||12.0 (6.0; 17.0)||<0.05|
|Phosphate Cmin (mmol/L)||0.90 (0.83; 1.00)||1.05 (0.97; 1.12)||<0.01|
|Phosphate Tmin (hours)||3.0 (2.0; 24.0)||2.0 (1.5; 3.0)||0.08|
|1,25(OH)2D Cmax (mmol/L)||221 (168; 288)||128 (100; 151)||<0.01|
|1,25(OH)2D Tmax (h)||10.0 (6.5; 10.0)||5.0 (1.0; 12.0)||0.19|
|Calcium (mmol/24 h)||9.2 (5.2;11.5)||9.6 (7.7; 13.5)||0.26|
|Calcium Emax (mmol/h)||0.6 (0.4; 0.8)||0.7 (0.5; 0.9)||0.13|
|Calcium Tmax (h)||5.0 (1.0; 12.0)||5.0 (3.0; 7.0)||0.76|
|Calcium Emin (mmol/h)||0.2 (0.1; 0.3)||0.3 (0.2; 0.3)||0.02|
|Calcium Tmin (h)||3.0 (3.1; 6.0)||12.0 (4.0; 20.0)||0.05|
|Phosphate (mmol/24 h)||31.6 (22.2; 38.7)||23.1 (17.9; 34.2)||0.11|
|Phosphate Emax (mmol/h)||3.1 (2.6; 3.6)||1.7 (1.3; 2.4)||<0.01|
|Phosphate Tmax (h)||5.0 (3.0; 5.0)||5.0 (1.0; 7.0)||0.78|
|Phosphate Emin (mmol/h)||0.3 (0.2; 0.6)||0.6 (0.3; 0.8)||0.06|
|Phosphate Tmin (h)||20.0 (20.0; 20.0)||5.0 (2.0; 20.0)||<0.001|
PTH showed a plasma T½ of median 2.2 (IQR, 1.7–2.5) hours (Table 3). PTH values had returned to predose levels after approximately 16 hours. The dose of rhPTH(1-84) per kilograms of body weight did not correlate with PTH AUC0–t (ρ = −0.155, p = 0.50), PTH clearance (ρ = −0.155, p = 0.50), Cmax (ρ = 0.073, p = 0.75), or Tmax (ρ = 0.141, p = 0.54). None of the aforementioned indices correlated significantly with dose of rhPTH(1-84) divided by body mass index (data not shown). We did find a correlation between PTH clearance and plasma creatinine levels (ρ = −0.465, p = 0.03), but not between PTH clearance and creatinine clearance (ρ = 0.294, p = 0.20).
PD responses to rhPTH(1-84)
At study start, plasma calcium concentrations did not differ between the two groups (Table 2). During the 24-hour study, the diurnal variation in plasma Ca2+ differed between the groups, with significantly higher Ca2+ levels in the rhPTH(1-84) group from 2 to 12 hours after injection (Fig. 2). Plasma Ca2+ levels had returned to baseline values after approximately 24 hours. In accordance, we found a significantly larger plasma Ca2+ AUC0–24h in the rhPTH(1-84)-treated patients compared with the placebo group (Table 3). The patients in the rhPTH(1-84) group also had significantly higher plasma Ca2+ Cmax and earlier Ca2+ Tmax than the placebo group (Table 3). We measured plasma Ca2+ levels above the upper limit of the reference range in 15 out of the 21 patients (71%) receiving rhPTH(1-84) within the 24 hours, with a median plasma calcium value of 1.41 (IQR, 1.36–1.48) mmol/L during hypercalcemic periods. Hypercalcemia was found 24 hours after the last rhPTH(1-84) injection at time point t = 0 hours and t = 24 hours in 10% and 24% of the rhPTH(1-84)-treated participants, respectively. During the 24-hour study, the highest plasma Ca2+ value measured in the rhPTH(1-84) treatment arm was 1.59 mmol/L. The median plasma Ca2+ measured in the rhPTH(1-84)-treated patients who developed hypercalcemia was 1.30 (IQR, 1.26–1.41) mmol/L compared with 1.21 (IQR, 1.12–1.23) mmol/L in the patients who did not develop hypercalcemia (p = 0.001). There were no significant difference in supplementation dosage between the rhPTH(1-84)-treated patients who became hypercalcemic and those who did not. The hypercalcemic patients received a median of 2.5 (IQR, 2.0–2.75) µg active vitamin D compared to 1.75 (IQR, 1.0–3.0) µg in the nonhypercalcemic group (p = 0.15), and the hypercalcemic patients received a median of 1400 (IQR, 800–2000) g calcium supplements compared to 1500 (IQR, 800–2000) g in the nonhypercalcemic (p = 0.19). Among the 17 patients in the placebo group, 2 (12%) had one or more plasma Ca2+ level measurements above the upper level of the reference interval. The highest Ca2+ level measured in patients randomized to placebo was 1.46 mmol/L. There were no differences either in the supplementation dosage between the hypercalcemic and nonhypercalcemic patients in the placebo group (data not shown). The number of measurements with hypercalcemia correlated significantly with plasma PTH AUC0–t (ρ = 0.614, p < 0.01), but not with plasma PTH Cmax or PTH dose/kg. Neither weight, body mass index (BMI), gender, nor age was associated with hypercalcemia in any of the two treatment groups (data not shown).
Overall, the median 24-hour renal calcium excretion did not differ between the rhPTH(1-84) and placebo group (Table 3); although rhPTH(1-84) treatment changed the diurnal variation of renal calcium excretion, leading to a significantly lower excretion from 2 to 8 hours after injection (Fig. 2).
Plasma Ca2+ (mmol/L) correlated with urine calcium excretion (mmol/h) in the rhPTH(1-84)-treated patients at all measurement points from t = 3 hours to t = 16 hours (ρ = 0.61–0.77, with p < 0.001 to p = 0.02).
Neither baseline plasma phosphate (Table 2), 24-hour renal phosphate excretion (Table 3), or diurnal variation in plasma phosphate levels during the 24-hour study (Fig. 2) differed significantly between groups. However, rhPTH(1-84) treatment significantly changed the diurnal variation in renal phosphate excretion, with an increased excretion during the first 8 hours of study followed by a decreased excretion during the last 8 hours (Fig. 2). In accordance, the maximal urine phosphate excretion (Emax) differed between the groups, with the highest value in the rhPTH(1-84)-treated group (Table 3). Furthermore, plasma phosphate Cmin was significantly lower in the rhPTH(1-84) group compared with the placebo group (Table 3).
At study start, plasma levels of 1,25(OH)2D were significantly higher in the rhPTH(1-84) group compared with the placebo group (Table 2). During the 24-hour study, variations in plasma 1,25(OH)2D differed significantly between groups, with higher levels in the rhPTH(1-84)-treated group at most time points, including at the end of the 24-hour study (Fig. 3). In accordance, Cmax was nearly twice as high in the rhPTH(1-84) group compared with the placebo group (Table 3). In the placebo group, the time-concentration curve was almost flat.
During the 6 months of treatment (n = 62 study subjects), plasma magnesium levels were borderline significantly lower in the rhPTH(1-84) group compared to the placebo-group (Fig. 4A), although urinary magnesium (mmol per 24 hours) did not differ between groups (Fig. 4B). Similarly, in the 24-hour substudy (n = 38 participants), plasma magnesium levels were significantly lower in the rhPTH(1-84) group compared with the placebo group throughout the study period (Table 2, Fig. 4C). In the individual patients, plasma magnesium levels fluctuated only slightly during the 24-hour study period, without any clear diurnal variation. Accordingly, an unambiguous Cmax and Cmin could not be defined in the individual patients. However, inspection of the concentration versus time curve (Fig. 4C) showed a tendency toward a decrease within hours of administration of PTH(1-84), followed by rising levels until the mid-afternoon, with a subsequent decrease until the next morning. Test for group versus time interaction showed a significant difference in the diurnal variations of plasma magnesium levels between groups (p = 0.004). The median 24-hour urinary magnesium excretion did not differ between groups (p = 0.39), but there was a significant difference between groups in variation over the 24 hours, with a lower excretion in the rhPTH(1-84) group from 2 to 8 hours after injection (Fig. 4D).
Correlations between studied indices
AUC0–t for plasma PTH correlated with AUC0–24h of plasma Ca2+ (ρ = 0.539, p < 0.05) and 1,25(OH)2D (ρ = 0.550, p = 0.001), as well as with Cmax of plasma Ca2+ (ρ = 0.614, p < 0.01) and 1,25(OH)2D (ρ = 0.60, p < 0.001). There were no correlations between PTH AUC0–t and AUC0–24h or Cmin of plasma phosphate. Cmax of PTH correlated with Cmax (ρ = 0.57, p < 0.001) and AUC0–24 (ρ = 0.52, p < 0.01) of 1,25(OH)2D, but not with the Cmax or AUC0–24h of Ca2+ levels, nor with AUC0–24h or Cmin of plasma phosphate. The dose of rhPTH(1-84) per kg body weight correlated with Cmin of plasma phosphate (ρ = 0.516, p = 0.02) and inversely with Tmax of plasma Ca2+ (ρ = −0.538, p = 0.01), but not with any of the other studied PD indices.
ECG and blood pressure
Heart rate and BP did not differ between groups at the start or at the end of the 6-month study (data not shown). On the day of the 24-hour study, BP did not differ between groups, but heart rate was significantly higher in the rhPTH(1-84) group compared with the placebo group at 0, 1, and 10 hours (Table 4). In accordance with the higher heart rate, ECG recordings showed a significantly lower RR-interval at the three measurement time points (Table 4). There was a significant upward excursion in heart rate from t = 0 hours to t = 1 hour within the rhPTH(1-84) group (p = 0.007). Plasma PTH concentration at t = 1 hour correlated positively with heart rate (ρ = 0.45, p = 0.005). No correlation was found between plasma PTH and blood pressure or between plasma Ca2+ and the cardiovascular indices.
|Time (hours)||PTH group (n = 21)||Placebo group (n = 17)||p|
|Systolic (mmHg)||0||143 (128; 151)||135 (121: 144)||0.28|
|1||130 (124; 140)||132 (119; 144)||0.48|
|10||133 (125; 150)||124 (111; 135)||0.09|
|Diastolic (mmHg)||0||86 (76; 88)||78 (72; 82)||0.19|
|1||80 (69; 88)||73 (71; 81)||0.43|
|10||80 (72; 90)||72 (66; 80)||0.06|
|Heart rate (beats per minute)||0||69 (56; 78)||58 (53; 64)||0.01|
|1||75 (72; 89)||60 (57; 73)||0.01|
|10||70 (65; 87)||61 (54; 68)||0.01|
|PR interval (ms)||0||161 (134; 181)||175 (165; 185)||0.07|
|1||155 (140; 172)||172 (150; 190)||0.06|
|10||159 (149; 182)||170 (158; 183)||0.14|
|RR interval (ms)||0||837 (748; 1009)||1054 (986; 1166)||0.001|
|1||795 (678; 870)||997 (842; 1060)||0.002|
|10||848 (671; 937)||997 (926; 1076)||0.01|
|QTcF (ms)||0||421 (402; 444)||412 (402; 442)||0.37|
|1||418 (405; 430)||425 (419; 439)||0.78|
|10||410 (388; 424)||420 (406; 441)||0.11|
|Normal, n (%)||15 (71)||14 (82)||0.48|
|Abnormal, n (%)||6 (29)||3 (18)|
PTH levels at t = 1 hour or the rise in PTH levels from t = 0 hour to t = 1 hour did not correlate with changes between t = 0 hours and t = 1 hour in heart rate, systolic BP, or diastolic BP.
No other pathological changes were observed as determined by PR or QTcF length or in the overall interpretation of the ECG recordings (data not shown).
In a randomized controlled design, we investigated PK and PD effects of rhPTH(1-84) treatment in patients with hypoPT. Our study showed PK characteristics similar to the findings in healthy individuals.[10-12] An important result with clinical relevance is our finding of peak plasma calcium levels 6 to 8 hours after rhPTH(1-84) injection, indicating that surveillance of plasma calcium levels may be insufficient if only preinjection nadir levels are assessed because intermittent episodes of hypercalcemia may be overlooked. Actually, within our group of patients randomized to rhPTH(1-84) treatment, hypercalcemia occurred in 71% of the patients, whereas only 12% of the patients on conventional therapy (and placebo PTH) developed episodes of hypercalcemia during the 24 hours of monitoring. In further trials, effects of different doses and dosing regimens (eg, two or three times a day) should be investigated in order to determine whether a more frequent dosing regimen, with lower doses per injection, may reduce the number of episodes of hypercalcemia, as was observed with rhPTH(1-34) in the studies by Winer and colleagues.[3, 5, 18]
The plasma concentration curve of PTH was in most patients biphasic, showing an initial peak at the first measurement 15 minutes after injection and a second peak after 2 hours. Similar results have been reported by other investigators. This is most likely due to the subcutaneous administration. The first peak may represents rhPTH(1-84) going straight into the circulation, whereas the second peak may reflect rhPTH(1-84) transported by the lymphatic system into the circulation.[19-21] This is supported by results on plasma levels of PTH and calcium following four different dosing methods of rhPTH(1-84) administration of a fixed dose of 100 µg in postmenopausal women. In that study, intravenous administration of PTH showed a single peak concentration curve, with a rapid increase followed by a steep decrease, whereas a biphasic pattern was found in response to subcutaneous administration in either the thigh or the abdomen.
The ionized calcium levels measured in terms of either AUC0–24h or peak concentration (Cmax) were both positively correlated to plasma PTH AUC0–t, but did not correlate with peak PTH concentrations. This indicates that a lower but more sustained exposure to circulating PTH is more efficient in maintaining calcium homeostasis than a higher concentration with a shorter duration, such as when rhPTH is injected into the abdomen. Therefore, our findings support using the thigh as the injection site, instead of the abdomen, because of the slower absorption rate in the thigh. Nevertheless, following a single injection of rhPTH(1-84) in postmenopausal women, the mean increase in plasma calcium levels has been shown to be more pronounced with an injection into the thigh as compared with an injection into the abdomen. Whether this also applies during long-term treatment of patients with hypoPT is currently unknown. If so, this may suggest the use of smaller doses given several times a day in order to lower the total dose needed.
After 24 hours, plasma levels of ionized calcium, phosphate, and 1,25(OH)2D were back to baseline, and the effect of the PTH injection was no longer evident. This suggests that daily injections are preferred instead of injections every other day. If a rhPTH(1-84) dose of 100 µg is too large for a patient, it would be more favorable to fractionate into a smaller daily dose than to introduce drug-free days. Our findings of a correlation between plasma PTH levels and 1,25(OH)2D levels are in accordance with the well-known stimulatory effects of PTH on the renal 1α-hydroxylase,[22, 23] thereby reducing the needs for exogenously administered active vitamin D analogues.
Calcium and phosphate homeostasis
The study was not designed to investigate the exact physiology of PTH, calcium, and phosphate, but it is noteworthy that we found a correlation between plasma Ca2+ and urinary excretion of calcium. This may suggest that the effect of rhPTH(1-84) on the tubular reabsorption of calcium (TRCa) is overridden by the effect of an increased filtered load. The increased plasma Ca2+ filtered load might also decrease the TRCa through an effect on the calcium sensing receptor in the renal tubule cells.
Our data also suggest that an administration time difference in the rhPTH(1-84) induced calcium and phosphate response. Administration of rhPTH(1-84) caused phosphate Tmin at 3 hours and calcium Tmax at 7 hours. This might imply that PTH acts more rapidly at the renal tubule to lower serum phosphate, and less rapidly at the skeleton and renal 1α-hydroxylase to increase serum calcium.
At all time points of measurement, plasma magnesium levels in the PTH group were below the levels of the placebo group, including at start of the 24-hour study (t = 0). This indicates a carryover effect from the previous 6 months of daily rhPTH(1-84) injections. Importantly, plasma magnesium levels fell within hours of administration of rhPTH(1-84) and stayed significantly below the placebo-treated patients. This was not a result of increased renal excretion because urinary magnesium levels in the placebo group were above the levels of the PTH group 2 to 8 hours after injection. Therefore, it seems most likely that the decreased plasma magnesium levels during PTH treatment were due to a redistribution of magnesium between the extracellular and intracellular compartments.
Effects of PTH on cardiovascular indices
Although our 6-month treatment study showed no significant effects on BP or heart rate, patients allocated to rhPTH(1-84) treatment had a significantly higher heart rate at all three time points of measurement on the day of the 24-hour study. Heart rate correlated significantly with plasma PTH levels at t = 1 hour; ie, when the ECG recording was performed at the highest plasma PTH levels. Because our study did not show significant correlations between plasma magnesium or Ca2+ levels and heart rate, it seems that the increased heart rate most likely is attributable to a direct effect of PTH. This is in accordance with the reported chronotropic action of PTH on cardiac tissue,[24-28] and in agreement with findings from previous clinical studies showing a dose-dependently increased pulse rate in response to stepwise increased iv doses of rhPTH(1-34).
Strengths and limitations to the study
Randomization was balanced within the group of patients who were accepted to participate in the current substudy, with no difference in demographics, etiology, or time since diagnosis. Our PK analyses were based on the assumption of a single-compartment model, which according to a visual inspection of the time versus plasma concentration curves seems acceptable, despite a small secondary increase in some individuals. It should be noted that the secondary increase is most likely not due to the distribution of study drugs between different compartments, but rather explained by differences in rate of absorption according to whether PTH(1-84) was absorbed directly into the blood or transferred through the lymphatic system after injection into the subcutis.[19-21] Therefore, we applied the one-compartment model because the assumptions for a two-compartment model were not strictly fulfilled. Also, endogenously synthesized PTH may in some patients have contributed to the measured plasma PTH levels. We circumvented this by excluding levels below the first measurement. However, in our PD analyses, we chose to perform analyses on the biochemical concentrations actually measured; ie, we did not adjust for baseline levels. Although such an adjustment may have resulted in data showing a more pronounced PD response to rhPTH(1-84) treatment, we believe that the analyses performed on actually measured indices is of more clinical relevance than adjusted values. Because alfacalcidol is used in the treatment of hypoPT, it should be recognized that the measured level of 1,25(OH)2D is a composite measure of endogenously synthesized and exogenously administered active vitamin D. As a result of the design of our study, the participants in the placebo group received more alfacalcidol than those randomized to the rhPTH(1-84) group. We do not believe that this affected our PD analyses to any major degree because the time-concentration curve for plasma 1,25(OH)2D concentrations was almost flat and showed no signs of increased levels following dosing of alfacalcidol at the time of breakfast, lunch, and/or dinner.
As assessed 24 hours after the last injection with rhPTH(1-84), our findings of hypercalcemia in 10% to 24% of our participants corresponds well with the values of 12% of participants with hypercalcemia in the Parathyroid Hormone and Alendronate (PaTH) study,[30, 31] 12.5% in the PReOtact in Postmenopausal OStEoporosis (PROPOSE) study, and 28% in the Treatment of Osteoporosis with Parathyroid Hormone Study Group (TOP) study. Further studies should of course be performed with repeated measurement in the same individuals in order to assess the generalizability of our findings and intraindividual day-to-day variations, as well as effects of different dosing regimens.
We found that rhPTH(1-84) was able to maintain plasma calcium within the normal range throughout the 24 hours despite a reduction or removal of oral calcium and active vitamin D supplements in patients with hypoPT. However, using the present fixed-PTH dosing scheme, hypercalcemia was more frequent in the rhPTH(1-84) treatment group compared to the placebo group throughout the day, and therefore we conclude that 100 µg/d of rhPTH(1-84) seems to be too high a dose in some patients.
We suggest that these patients may benefit from a smaller dose of rhPTH(1-84) than 100 µg or even a twice-a-day dosing scheme. It would be preferable to administer rhPTH(1-84) in a device that allows individual dosing.
It is noteworthy that even though we found a high prevalence of hypercalcemic episodes in the patients treated with rhPTH(1-84), they were all asymptomatic and there was no deterioration of renal function during the total 6-month treatment period. Presumably, PTH-induced hypercalcemia is less harmful than hypercalcemia triggered by other mechanisms; eg, vitamin D intoxication. This may be due to the fact that vitamin D intoxication causes elevated calcium and phosphate levels, whereas hypercalcemia during treatment with rhPTH is not accompanied by high plasma levels of phosphate. From a clinical point of view this consideration is supported by the often-benign course of asymptomatic primary hyperparathyroidism.
PTH replacement appears to be a safe alternative treatment for hypoparathyroidism, but for a more optimized treatment it is advisable to measure plasma calcium levels approximately 7 hours after administration instead of only nadir levels. This may allow for further adjustments in daily doses of supplements.
LM was a principal investigator on the REPLACE study initiated by NPS Pharmaceuticals. All authors other state that they have no conflicts of interest.
This work was supported by the Danish Council for Independent Research in Medical Science, the Novo Nordic Foundation, the Central Denmark Region Foundation, and NPS Pharmaceuticals. NPS Pharmaceuticals provided financial support for the biochemical analyses. We thank Nycomed Pharmaceuticals, Denmark, for supplying study drugs (rhPTH(1-84) and similar placebo) free of charge. We are indebted to the patients who participated in this study and made it possible. The study could not have been completed without the help from Charlotte Ebsen Paaskesen, Melina Gade, Tove Stenum, Lisbeth Flyvbjerg, and Christina Wiegers. The study was investigator-initiated and the authors had full access to all data.
Authors' roles: Study design: TS, LM, and L Rejnmark. Study conduct: TS and L Rejnmark. Data collection: TS, L Rolighed, SGK, AKA, and L Rejnmark. Data analysis: TS, AKA, LM, and L Rejnmark. Data interpretation: TS, LM, and L Rejnmark. Drafting manuscript: TS, LM, and L Rejnmark. Revising manuscript content: TS, L Rolighed, SGK, AKA, LM, and L Rejnmark. Approving final version of manuscript: TS, L Rolighed, SGK, AKA, LM, and L Rejnmark. TS takes responsibility for the integrity of the data analysis. All authors approved the final version of the manuscript.