Targeting the hepcidin-ferroportin pathway in anaemia of chronic kidney disease

Abstract

Aims: Erythropoiesis-stimulating agents used to treat anaemia in patients with chronic kidney disease (CKD) have been associated with cardiovascular adverse events. Hepcidin production, controlled by bone morphogenic protein 6 (BMP6), regulates iron homeostasis via interactions with the iron transporter, ferroportin. High hepcidin levels are thought to contribute to increased iron sequestration and subsequent anaemia in CKD patients. To investigate alternative therapies to erythropoiesis-stimulating agents for CKD patients, monoclonal antibodies, LY3113593 and LY2928057, targeting BMP6 and ferroportin respectively, were tested in CKD patients.

Methods: Preclinical in vitro/vivo data and clinical data in healthy subjects and CKD patients were used to illustrate the translation of pharmacological properties of LY3113593 and LY2928057, highlighting the novelty of targeting these nodes within the hepcidin-ferroportin pathway.

Results: LY2928057 bound ferroportin and blocked interactions with hepcidin, allowing iron efflux, leading to increased serum iron and transferrin saturation levels and increased hepcidin in monkeys and humans. In CKD patients, LY2928057 led to slower haemoglobin decline and reduction in ferritin (compared to placebo). Serum iron increase was (mean [90% confidence interval]) 1.98 [1.46-2.68] and 1.36 [1.22-1.51] fold-relative to baseline following LY2928057 600 mg and LY311593 150 mg respectively in CKD patients. LY3113593 specifically blocked BMP6 binding to its receptor and produced increases in iron and transferrin saturation and decreases in hepcidin preclinically and clinically. In CKD patients, LY3113593 produced an increase in haemoglobin and reduction in ferritin (compared to placebo).

Conclusion: LY3113593 and LY2928057 pharmacological effects (serum iron and ferritin) were translated from preclinical-to-clinical development. Such interventions may lead to new CKD anaemia treatments.

Trial registration: ClinicalTrials.gov NCT01330953 NCT01991483 NCT02144285 NCT02604160.

Keywords: chronic kidney disease; immunoglobulins; transport.

Conflict of interest statement

D.R.W. is an employee and shareholder of Eli Lilly and Company and is the holder of the issued patent US 8183346, Anti‐Ferroportin 1 Monoclonal Antibodies and Uses Thereof. M.S. is a shareholder of Eli Lilly and Company and at the time this work was performed, was an employee of Eli Lilly and Company. P.B. is an employee and shareholder of Eli Lilly and Company. S.C. is an employee and shareholder of Eli Lilly and Company. P.H.B. is an employee and shareholder of Eli Lilly and Company. J.M. was a shareholder and employee of Eli Lilly and Company at the time this work was performed. T.M. is an employee and equity owner of Orlando Clinical Research Center that received a grant from Eli Lilly and Company and contracted to conduct the study protocol. B.D. has declared that no conflict of interest exists. G.L.D. is an employee and shareholder of Eli Lilly and Company, a shareholder of Amgen Inc. and is the holder of the issued patent WO 2009094551 A1 – Ferroportin Antibodies and Methods of Use licensed with Amgen Inc. with no knowing licensing or royalties. S.M.E.T. is an employee and shareholder of Eli Lilly and Company and holds the issued patents for US 8795665 and US 8980582. R.B. is an employee and shareholder of Eli Lilly and Company. D.L. is an employee and shareholder of Eli Lilly and Company and holds the issued patents for US 8183346 and US 8679497. J.B. is an employee of DaVita Clinical Research that received a grant from Eli Lilly and Company and Lilly contracted with DaVita Clinical Research to conduct the study protocol. Related specialties: Hematology, nephrology.

© 2019 The British Pharmacological Society.

Figures

Figure 1

Hepcidin regulation of iron homeostatsis and the ability of the ferroportin antibody (LY2928057) and BMP6 antibody (LY3113593) to control the hepcidin–ferroportin pathway. BMP, bone morphogenic protein; BMPR, BMP receptor; BMP‐RE, BMP receptor response element; HJV, haemojuvelin

Figure 2

Preclinical PD response and PK profile in cynomolgus monkeys (n = 3) following administration of LY3113593: A, Serum iron concentrations in cynomolgus monkeys following IV administration of LY3113593 at doses of 0.05 (red squares), 0.3 (blue triangles), 3 mg/kg (green nablas) and 3 mg/kg isotype control antibody (black circles). Data are the mean ± SD (n = 3); B, Serum hepcidin concentrations in cynomolgus monkeys following IV administration of LY3113593 at doses of 0.05 (red squares), 0.3 (blue triangles), 3 mg/kg (green nablas) and 3 mg/kg isotype control antibody (black circles). Data are the mean ± SD (n = 3); C, PK: serum concentrations of LY3113593 in cynomolgus monkeys following an IV bolus dose of 0.05 (red squares), 0.3 (blue triangles) and 3 mg/kg (green nablas). Data are the mean ± SD (n = 3/group). IV, intravenous; PD, pharmacodynamics; PK, pharmacokinetics; SD, standard deviation

Figure 3

Haemoglobin levels in a rat anaemia of chronic disease model. Inflammation is induced on day 1 (dashed line) with PG‐APS and rats are subsequently treated with either 10 mg/kg HuIgG4 antibody control (black circles), 10 mg/kg HuIgG4 antibody control with 3 μg/kg darbepoeitin‐α (red squares), 10 mg/kg LY3113593 (blue triangles) or 10 mg/kg LY3113593 with 3 μg/kg darbepoeitin‐α (green nablas) weekly beginning on day 8. Carats indicate treatment dosing time points after inflammation is induced. Asterisks indicate statistical significance of 10 mg/kg LY3113593 with 3 μg/kg darbepoeitin‐α vs the 10 mg/kg HuIgG4 antibody control with 3 μg/kg darbepoeitin‐α treatment group by 2‐way ANOVA with Dunnett's test. P < .05*, <.005**, <.0005***, <.0001****. Data are the mean ± standard error of the mean (n = 6/group). Hgb, haemoglobin; PG‐APS, peptidoglycan‐polysaccharide

Figure 4

A, LY3113593 concentration (mean ± SD) vs time following single IV administration of LY3113593 1.5 mg (open circles), 5 mg (black circles), 15 mg (open triangles), 50 mg (black up triangles) and 150 mg (black down triangles) in healthy subjects and 150 mg (red circles) in patients with chronic kidney disease. The green line indicates the LY3113503 concentration at which hepcidin expression is inhibited by 50% (the model predicted value from preclinical data). B, Hepcidin (expressed as ratio relative to predose baseline value [mean ± SD]) vs time following single administration (at time 0) of LY3113593 150 mg (red triangle) and placebo (black circles) to patients with chronic kidney disease. The black horizontal lines indicate a ratio of 0.5 and 1. C, Iron (expressed as ratio relative to predose baseline value, [mean ± SD]) vs time following administration of LY3113593 150 mg (red triangles) and placebo (black circles) to patients with chronic kidney disease. The black horizontal lines indicate a ratio of 1 and 1.5. D, Ferritin (expressed as ratio relative to predose baseline value, [mean ± SD]) vs time following single administration (at time 0) of LY3113593 150 mg (red triangles) and placebo (black circles) to patients with chronic kidney disease. The black horizontal line indicates a ratio of 1. E, haemoglobin (expressed as absolute change relative to predose baseline value [mean ± SD]) vs time following administration of LY3113593 150 mg (red triangles) and placebo (black circles) to patients with chronic kidney disease. The black horizontal line indicates no change and a 1 g/dL increase in haemoglobin relative to baseline. SD, standard deviation

Figure 5

A, Serum iron concentrations in cynomolgus monkeys following i.v administration of LY2928057 at doses of 0.3 (black circles), 1 (red squares), 3 (blue triangles), 10 (green nablas) and 30 mg/kg (orange diamonds). Data are the mean ± SD (n = 4). B, Serum hepcidin concentrations in cynomolgus monkeys following intravenous administration of LY2928057 at doses of 0.3 (black circles), 1 (red squares), 3 (blue triangles), 10 (green nablas) and 30 mg/kg (orange diamonds). Data are the mean ± SD (n = 4). C, Serum concentrations of antiferroportin monoclonal antibody LY2928057 in cynomolgus monkeys following an IV bolus dose of 0.3 (black circles), 1 (red squares), 3 (blue triangles), 10 (green nablas) and 30 mg/kg (orange diamonds). Data are the mean ± SD (n = 4/group). SD, standard deviation

Figure 6

Haemoglobin represented as absolute change from baseline (mean ± SD) vs time after administration of LY2928057; 300 mg (blue line), 600 mg (green line), 1000 mg (red line) or placebo (black line) every 2 weeks (dosing at time 0, 14 and 28 day) to patients with chronic kidney disease. SD, standard deviation

Figure 7

A, LY2928057 PK concentration (mean ± SD) vs time following single i.v administration of LY2928057 30 mg (black circles), 100 mg (open circles), 300 mg (black triangles), 1000 mg (open triangles) in healthy subjects and 300 mg (red open squares), 600 mg (red triangles) and 1000 mg (red squares) in patients with chronic kidney disease. The green line indicates the LY2928057 concentration at which hepcidin activity on ferroportin receptor is inhibited by 50%. B, Serum iron concentration (mean ± SD) vs time after administration of LY2928057 300 mg (blue line), 600 mg (green line), 1000 mg (red line) or placebo (black line) dosing every 2 weeks (at time 0, 14 and 28 days) to patients with chronic kidney disease. C, hepcidin concentration (mean ± SD) vs time after administration of LY2928057 300 mg (blue line), 600 mg (green line), 1000 mg (red line) or placebo (black line) every 2 weeks (at time 0, 14 and 28 days) to patients with chronic kidney disease. D, Ferritin levels represented as ratio relative to baseline predose value (mean ± SD) at the end of the 6th week treatment (time 42 days) after multiple administration of LY2928057 (at 0, 14 and 28 days) to patients with chronic kidney disease. The baseline value of ferritin for each dose group is also indicated (Figure 7D for 600 and 1000 mg, the 90% confidence of the mean ratio [ferritin at week 6:ferritin baseline] is below 1: 0.65–0.83 for 600 mg and 0.47–1.00 for 1000 mg. This indicates that ferritin levels are significantly decreased at those LY2928057 dose levels). SD, standard deviation