Chenguang Wang Ph.D. (1), Klas Petersson Ph.D. (2), Dorina van der Mey Ph.D.(1), Silvia Lentini M.D.(1), Dirk Garmann Ph.D.(1), Takahiko Tanigawa Ph.D.(3), Ashraf Yassen Ph.D.(1)
(1) Bayer AG, Wuppertal, Germany, (2) qPharmetra LLC, Stockholm, Sweden, (3) Bayer Yakuhin Ltd, Osaka, Japan
Introduction: Anaemia is a common complication in patients with chronic kidney disease (CKD) and is in CKD patients mostly caused by a deficiency in the erythropoietin (EPO) synthesis [1]. The subcutaneously administration of recombinant EPO or erythropoietin stimulating agents (ESAs) is the standard-of-care (SoC) treatment for renal anaemic patients. As a potential alternative for the treatment of renal anaemia in CKD patients, an oral administered selective hypoxia-inducible factor -prolyl hydroxylase (HIF-PH) inhibitor, molidustat (BAY853934), was developed at Bayer AG.
Objectives: The objectives of this population pharmacokinetic (popPK) analysis were (1) to characterise the pharmacokinetic of molidustat in non-dialysis (ND) and haemodialysis (HD) patients with CKD anaemia and associated inter-patient variability based on phase II studies (2) to explore the influence of haemodialysis procedure on PK (3) to identify covariates that explain variability with particular interests on the difference between ND and HD patients and between Japanese (JP) and non-Japanese (non-JP) patients and (4) to provide rationale of dosing design for phase III studies in Japan from a PK perspective.
Methods: Three multi-centre phase II clinical studies [2] from the daily oral treatment increasing endogenous erythropoietin (DIALOGUE) program were included in this analysis. DIALOGUE1 was a placebo-controlled, fixed-dose trial (25, 50, and 75 mg once daily; 25 and 50 mg twice daily) in ND CKD patients naïve to ESA. DIALOGUE2 and DIALOGUE4 were open-label, variable-dose trials, in which patients on darbepoetin or epoetin treatment were randomly switched to molidustat or maintained in the original ESA treatment for 16 weeks. Starting dose ranged between 25–75 and 25–150 mg daily in DIAGLOGUE2 and 4, respectively, and could be titrated to maintain haemoglobin levels within predefined target ranges. The primary end point in all trials was the change in haemoglobin level between baseline and the mean value from the last 4 weeks of the treatment period. In DIALOGUE1, PK samples were collected at Pre-dose (DAY15, 29, 57, 85 and 113), 1 hour post dose (DAY1, 57 and 113), 2-3 hours post-dose (DAY1, 57 and 113), 8-12 hours post-dose (DAY1, 43 and 113). In DIALOGUE2 and 4, PK samples were collected at Pre-dose (DAY15, 29, 57, 85 and 113), 1-3 hours post-dose (DAY1, 15, 29, 57, 85 and 113), 6-12 hours post-dose (DAY1, 43, 71 and 113). Additionally, PK samples were collected at the start and end of dialysis on DAY1 and DAY113 in DIALOGUE4. Plasma concentrations of molidustat were measured by LC-MSMS with LLOQ of 0.2 µg/L. The popPK analysis was performed with NONMEM 7.2. Bootstrap based forest plot [3] was used to investigate the covariate impact on AUCSS and Cmax,SS.
Results: Molidustat PK was linear in the dose range of 15 to 200 mg among 296 patients and was best described by a 1st-order absorption, two-compartment distribution model with parameters of absorption rate (ka), clearance (CL/F), volume of distribution at steady state (Vss/F), fraction of Vc/F to the Vss/F (Frac Vc/Vss) and inter-compartmental clearances (Q/F). Inter-individual variability (IIV) in the PK parameters were identified for ka, CL/F and Vss/F. The estimates of population parameters (inter-individual variabilities) were ka =1.52 [1/h](123%) ; CL/F=35 [L/h](37.7%); Vss/F=338 [L](69.1%); Frac Vc/Vss =0.272; Q/F=26.8 [L/h]. Weight, eGFR, total bilirubin (TBIL) and age were identified covariate for CL/F. Weight and HD patient were identified covariates for Vss/F. Influence of ongoing dialysis procedure or ethnicity was not found on PK parameters. Identified covariate models were:
CL/F=35*(WT/71)0.834*(eGFR/12.85)0.253*(TBIL/0.35)-0.229*(AGE/66)-0.342
VSS/F=338*(1+HD*1.44)*(WT/71)1.29
None of identified covariate at extreme values (5th and 95th percentiles of typical values for continuous covariates) resulted in changes in exposure of more than 50% relative to a typical patient.
Conclusions: PopPK analysis revealed no differences between PK in non-Japanese and Japanese patients. In addition, the clinical relevance for starting dose selection based on identified PK covariates was low, given that individual dose titration was performed with the aim to reduce variability in Hb response.
References:
[1] Jodie L. Babitt et al,. J Am Soc Nephrol 23: 1631–1634, 2012.
[2] C. Macdougall. et al,. CJASN Jan 2019, 14 (1) 28-39
[3] Li, J., et al,. Br J Clin Pharmacol. 2017 Aug; 83(8):1723-1733
Reference: PAGE () Abstr 9364 [www.page-meeting.org/?abstract=9364]
Poster: Drug/Disease Modelling - Other Topics