Treatment of Autosomal Dominant Hypocalcemia Type 1 With the Calcilytic NPSP795 (SHP635)

Abstract

Autosomal dominant hypocalcemia type 1 (ADH1) is a rare form of hypoparathyroidism caused by heterozygous, gain-of-function mutations of the calcium-sensing receptor gene (CAR). Individuals are hypocalcemic with inappropriately low parathyroid hormone (PTH) secretion and relative hypercalciuria. Calcilytics are negative allosteric modulators of the extracellular calcium receptor (CaR) and therefore may have therapeutic benefits in ADH1. Five adults with ADH1 due to four distinct CAR mutations received escalating doses of the calcilytic compound NPSP795 (SHP635) on 3 consecutive days. Pharmacokinetics, pharmacodynamics, efficacy, and safety were assessed. Parallel in vitro testing with subject CaR mutations assessed the effects of NPSP795 on cytoplasmic calcium concentrations (Ca²⁺_i ), and ERK and p38^MAPK phosphorylation. These effects were correlated with clinical responses to administration of NPSP795. NPSP795 increased plasma PTH levels in a concentration-dependent manner up to 129% above baseline (p = 0.013) at the highest exposure levels. Fractional excretion of calcium (FECa) trended down but not significantly so. Blood ionized calcium levels remained stable during NPSP795 infusion despite fasting, no calcitriol supplementation, and little calcium supplementation. NPSP795 was generally safe and well-tolerated. There was significant variability in response clinically across genotypes. In vitro, all mutant CaRs were half-maximally activated (EC₅₀ ) at lower concentrations of extracellular calcium (Ca²⁺_o ) compared to wild-type (WT) CaR; NPSP795 exposure increased the EC₅₀ for all CaR activity readouts. However, the in vitro responses to NPSP795 did not correlate with any clinical parameters. NPSP795 increased plasma PTH levels in subjects with ADH1 in a dose-dependent manner, and thus, serves as proof-of-concept that calcilytics could be an effective treatment for ADH1. Albeit all mutations appear to be activating at the CaR, in vitro observations were not predictive of the in vivo phenotype or the response to calcilytics, suggesting that other parameters impact the response to the drug. © 2019 American Society for Bone and Mineral Research.

Keywords: CALCILYTIC; CALCIUM-SENSING RECEPTOR; HYPOCALCEMIA; HYPOPARATHYROIDISM; PTH.

Figure 1.. NPSP795 Clinical Study Design.

After an initial outpatient screening visit, subjects were admitted to the National Institutes of Health (NIH) Clinical Center. An i.v. dose of 5 mg/10min NPSP795 was administered on day 1, and subjects underwent serial sampling for pharmacokinetic/pharmacodynamic parameters. If subjects had an increase in ionized calcium to the normal range or an increase in intact parathyroid hormone (iPTH) to greater than the upper limit of normal (ULN), the same dose of NPSP795 would be administered on day 2. If subjects did not meet the ionized calcium or iPTH parameters, 15 mg/3.5h NPSP795 would be administered on day 2 along with serial sampling. The same criteria were used to determine the dose of NPSP795 on day 3. Subjects were discharged home on day 4.

Figure 2.. Pharmacokinetic and pharmacodynamic effects of NPSP795.

NPSP795 concentrations significantly increased after infusion as predicted from modeling studies (A-C) and resulted in a rapid and concentration-dependent increase in % change in PTH (D-F) and PTH (G-I). Significant effects on PTH were seen at the highest concentrations in NPSP795 during the 5 mg bolus infusion on day 1 (D) and with the 30 mg infusion on day 3 (F, I). The fractional excretion of calcium (FECa) (J-L) decreased by > 30% following the 15 mg/3.5h, and >40% 30 mg/3.5h doses but did not reach statistical significance. The ionized calcium levels (M-O) remained stable during NPSP795 infusions despite fasting and calcium and calcitriol doses being withheld. Statistical analysis was performed using mixed model repeat measures analysis.

Figure 3.. Clinical response to NPSP795 by CaR genotypes.

PTH area under the curve (AUC) varied widely across the different CAR mutations despite overall similar exposure to NPSP795 (B). The subject with E228K mutation in the CAR had little to no response to NPSP795 (A). Subjects 1 and 4 were first-degree cousins with the same A840V CAR mutation but had widely different responses to NPSP795 (C). Differences in response were not accounted for by differences in NPSP795 clearance, as subject 4 had higher NPSP795 exposure (D), but less of a response. Of note, subjects 3 and 5 both had E228 mutations, K and A, respectively, but markedly different responses. Subject 3 had little response (E), but robust NPSP795 blood levels (F).