Clinical Kidney Journal, 2018, 1–10 doi: 10.1093/ckj/sfy013 Original Article OR I G I N AL A R T I C L E Daprodustat for anemia: a 24-week, open-label, randomized controlled trial in participants with chronic kidney disease 1,8 2,3 1 2 Louis Holdstock , Borut Cizman , Amy M. Meadowcroft , Nandita Biswas , 4 5 6 7 Brendan M. Johnson , Delyth Jones , Sung Gyun Kim , Steven Zeig , 2 2 John J. Lepore and Alexander R. Cobitz Metabolic Pathways and Cardiovascular Therapeutic Area, GlaxoSmithKline, Research Triangle Park, NC, USA, Metabolic Pathways and Cardiovascular Therapeutic Area, GlaxoSmithKline, Collegeville, PA, USA, Renal Electrolyte and Hypertension Division, Perelman School of Medicine, University of Pennsylvania, 4 5 Philadelphia, PA, USA, Clinical Pharmacology, Roivant Sciences Inc., Durham, NC, USA, Clinical Statistics, 6 7 GlaxoSmithKline, Stevenage, UK, Hallym University Sacred Heart Hospital, Anyang, South Korea, Pines Clinical Research, Pembroke Pines, FL, USA and Present address: United Therapeutics Corporation, Research Triangle Park, NC, USA Correspondence and offprint requests to: Borut Cizman; E-mail: email@example.com ABSTRACT Background. This study assessed the short-term safety and efﬁcacy of daprodustat (an oral hypoxia-inducible factor-prolyl hydroxylase inhibitor) to achieve a target hemoglobin in patients with anemia of chronic kidney disease (CKD). Methods. Patients (n¼ 252) with Stages 3–5 CKD not receiving dialysis were enrolled in this 24-week, multicenter trial [hemoglobin entry criteria: 8–10 g/dL (Cohort 1) or 8–11 g/dL (Cohort 2) for recombinant human erythropoietin (rhEPO)-na¨ve ı participants; 9–10.5 g/dL (Cohort 1) or 9–11.5 g/dL (Cohort 2) for rhEPO users]. rhEPO-naı ¨ve participants were randomized 3:1 to daprodustat (1, 2 or 4 mg) or control (rhEPO per standard of care). rhEPO users were randomized 1:1 to daprodustat 2 mg or control. Study medication was titrated to maintain hemoglobin 9–10.5 g/dL (Cohort 1) or 10–11.5 g/dL (Cohort 2). Hemoglobin, iron metabolism markers and safety parameters were measured every 4 weeks. Results. Mean hemoglobin levels at Week 24 were 10.2 g/dL (Cohort 1) and 10.9 g/dL (Cohort 2) in the daprodustat group and 10.7 g/dL (Cohort 1) and 11.0 g/dL (Cohort 2) in the control group. Participants had hemoglobin levels within the target range a median of 82% and 66% of the time between Weeks 12 and 24 in the daprodustat and control groups, respectively. The adverse event proﬁle was consistent with clinical events in the CKD population. Conclusions. Daprodustat effectively maintained target hemoglobin over 24 weeks in CKD patients with anemia who were rhEPO naı ¨ve or had switched from existing rhEPO therapy. Received: 20.7.2017. Editorial decision: 9.1.2018 V C The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/ licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact firstname.lastname@example.org Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018 2| L. Holdstock et al. Keywords: chronic kidney disease, daprodustat, hypoxia-inducible factor, prolyl hydroxylase inhibitor, recombinant human erythropoietin Society of Nephrology-Chronic Kidney Disease Initiatives equation INTRODUCTION  for Japanese participants residing in Japan and the CKD- In patients with chronic kidney disease (CKD) not on dialysis, ther- Epidemiology Collaboration equation for Caucasian and apy for anemia with recombinant human erythropoietin (rhEPO) African-American populations. Participants who were rhEPO naı ¨ve and its analogs is often reserved for those with more advanced were not to have used rhEPO within 8 weeks before screening; CKD . Several large randomized trials [e.g. The Correction of those who were rhEPO users were required to be on stable doses of Hemoglobin and Outcomes in Renal Insufficiency (CHOIR) trial,  the same rhEPO for the 4 weeks before entering the trial. Inclusion The Cardiovascular Risk Reduction by Early Anemia Treatment and exclusion criteria, and predefined study medication stopping with Epoetin Beta (CREATE) trial  and Trial to Reduce criteria can be found in the Supplementary data, Item S1. Cardiovascular Events with Aranesp Therapy (Treat) trial ] have reported adverse cardiovascular outcomes and death when using rhEPO and its analogs to target hemoglobin levels>13 g/dL, which Study design led the US Food and Drug Administration to revise the safety This global study (ClinicalTrials.gov identifier NCT01977573) language in labeling for rhEPO and its analogs in 2007  and again was conducted from 17 December 2013 to 19 June 2015 at 84 in 2011 . These changes included limiting treatment initiation sites in 15 countries in adherence with the Declaration of in CKD patients not on dialysis to when the hemoglobin level is Helsinki, and was approved by the relevant institutional review <10 g/dL, and reducing or interrupting the dose of rhEPO if the boards or ethics committees. An internal GlaxoSmithKline hemoglobin level is >10 g/dL [5–7]. Safety Review Team reviewed blinded safety data instream and To improve treatment options over existing marketed rhEPO an independent data monitoring committee periodically re- and its analogs, new investigational agents including hypoxia- viewed the same safety data, but it was unblinded. inducible factor (HIF)-prolyl hydroxylase inhibitors (PHIs), such as The study consisted of a 4-week screening phase, a 24-week daprodustat, are being developed [8–13]. The mechanism of ac- treatment phase and a follow-up visit 4 weeks after complet- tion of these agents is described in detail elsewhere . Briefly, ing treatment. Participants who discontinued study medication PHIs act by inhibiting HIF-prolyl hydroxylase enzymes (PHD1, or withdrew early from the study attended an early withdrawal PHD2 and PHD3), leading to activation of HIF-responsive genes (EW) visit, an EW follow-up visit 4 weeks later, phone assess- that regulate the tissue response to hypoxia. One key activity is ments aligned with the remaining key study visits and an EW induction of erythropoiesis through direct activation of the EPO final visit at Week 24. Study investigators or staff were respon- gene. In addition, HIF activation results in the induction of a num- sible for monitoring and recording adverse events (AEs). AEs ber of genes directly or indirectly involved in iron uptake, mobil- were collected from the start of study treatment and until the ization and transport, resulting in decreased hepcidin production, follow-up contact. (Additional information on the study design as well as vascular endothelial growth factor (VEGF) . Potential and study assessments performed, including laboratory assess- advantages of daprodustat and other PHIs over rhEPO and its ana- ments, safety reporting, echocardiology and ophthalmology logs in the treatment of anemia of CKD include: (i) raising hemo- examinations, is included in the Supplementary data, Item S2; globin without exposing the patient to the supraphysiologic EPO the study design is shown in Supplementary data, Figure S1.) levels and increases in blood pressure (BP) associated with intra- Participants were assigned to one of two groups based on venous (IV) rhEPO and its analogs; (ii) improving iron availability previous exposure to rhEPO: those who were currently receiving for erythropoiesis, which may reduce the need for iron replace- rhEPO (rhEPO users) and those who were not (rhEPO naı ¨ve). ment therapy; and (iii) avoiding the need for parenteral injection. Hemoglobin entry criteria and target values were modified dur- The dose response for daprodustat was previously defined ing the study to more closely align with anemia guidelines and in rhEPO-naı ¨ve patients with CKD . Here, we report the re- clinical practices outside the USA (Table 1). Separate random- sults of a 24-week clinical trial that assessed the ability of dap- ization lists were generated for the two groups (rhEPO naı ¨ve and rodustat to achieve and maintain hemoglobin levels within a rhEPO users) by a GlaxoSmithKline statistician using the prespecified target range in patients with CKD Stages 3, 4 or 5, GlaxoSmithKline randomization system RandAll. The stratifica- both as an initiation therapy in rhEPO-naı ¨ve participants and as tion variables (region and baseline hemoglobin range; a switch therapy for participants already receiving rhEPO. Supplementary data, Table S1 in Item S2) were used as blocking factors. Block size of four was used. Participants were assigned MATERIALS AND METHODS a randomization number by an interactive voice/web response system. Participants who were rhEPO naı ¨ve were randomized Study population 3:1 to once-daily daprodustat (1, 2 or 4 mg daily depending on Informed consent was obtained from all participants. Eligible par- baseline hemoglobin) or control, and rhEPO users were random- ticipants had CKD Stages 3, 4 or 5 as defined by the Japanese ized 1:1 to daprodustat 2 mg daily or control. Starting doses of Table 1. Hemoglobin entry criteria Cohort 1 Cohort 2 (ex-USA) Hemoglobin entry Hemoglobin target range Hemoglobin entry Hemoglobin target range rhEPO-naı ¨ve participants 8–10 g/dL 9–10.5 g/L 8–11 g/dL 10–11.5 g/dL rhEPO users 9–10.5 g/dL 9–10.5 g/dL 9–11.5 g/dL 10–11.5 g/dL rhEPO, recombinant human erythropoietin. Cohort included all countries until Cohort 2 was added; thereafter it applied to the USA only. Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018 Daprodustat for anemia of CKD | 3 daprodustat were based on dose-response modeling from previ- replete throughout the study, an iron protocol was instituted ous studies. Control participants received rhEPO (epoetins or from Week 4 onwards (Supplementary data, Table S2 in Item their biosimilars or darbepoetin) as necessary per standard of S3). The complete study objectives and endpoints can be found care as determined by the investigator. Because of the different in Table 2. routes of administration, rhEPO dosing data were standardized in terms of epoetin IV dosing (reported in IU/kg/week) (see Statistical analyses Supplementary data, Item S2). The dose of daprodustat was kept constant for the first 4 The planned sample size for this study was driven by exposure weeks, after which the dose could be adjusted based on a pre- requirements to achieve 100 evaluable participants treated specified dose adjustment algorithm to achieve and/or main- with daprodustat, where a participant was considered evaluable tain hemoglobin within the target range. Dose adjustments if they had on-treatment data at Week 24. (See Supplementary were made automatically by the interactive voice/web response data, Item S4 for additional justification.) system so the participant, investigator, site staff and study The intent-to-treat (ITT) population in the daprodustat team were blinded to the dose being administered during the groups comprised all randomized participants who received at study. Those in the control group could receive rhEPO therapy, least one dose of study treatment and had a baseline and at with doses adjusted as determined by the investigator with the least one postbaseline on-therapy hemoglobin assessment. For intention of treating to the same hemoglobin target range as the control groups, the ITT population comprised all random- the daprodustat group. To ensure participants remained iron ized participants who had a baseline and at least one Table 2. Study objectives and endpoints Objectives Endpoints Primary Characterize the ability of daprodustat to achieve Hemoglobin concentration at Week 24 mean hemoglobin response within the target range Secondary Characterize the ability of daprodustat to achieve Percentage of time within, below and above target range, be- hemoglobin within the target range tween Weeks 12 and 24 Number (%) of participants with hemoglobin in the target range at Week 24 Number (%) of participants reaching predeﬁned hemoglobin stopping criteria Characterize the effect of daprodustat on measures Change from baseline in hepcidin, ferritin, transferrin, TSAT, of iron metabolism and utilization, on indices of total iron, TIBC and CHr at Week 24 hematopoiesis, EPO and on VEGF Change from baseline in hematocrit and RBCs RBCs, reticulocyte number at Week 24 Maximum observed change from baseline in EPO Maximum observed change from baseline in VEGF Characterize the steady-state population PK of dap- Population plasma PK parameters of daprodustat and rodustat and metabolites metabolites Evaluate the daprodustat dose adjustment scheme Number, frequency and timing of dose adjustments Total cumulative and ﬁnal dose Number of hemoglobin excursions, hemoglobin cycles and dose cycles Number (%) of participants with at least one hemoglobin excursion Number (%) of participants with at least one hemoglobin cycle Number (%) of participants with at least one dose cycle Time that dose is held because hemoglobin exceeded the upper limit Number (%) of participants receiving additional therapies of blood transfusions, IV iron or rhEPO Safety Assess the safety and tolerability of daprodustat Incidence and severity of AEs and SAEs as reasons for discon- tinuation of study treatment, including discontinuation for safety-related reasons (e.g. prespeciﬁed stopping criteria or AE), absolute values and changes from baseline in laboratory parameters, estimated sPAP, LVEF, ophthalmology assess- ments and vital signs Preliminary assessment of MACE (deﬁned as all-cause mor- tality, myocardial infarction or stroke) and other CV events AE, adverse event; CHr, reticulocyte hemoglobin content; CV, cardiovascular; IV, intravenous; LVEF, left ventricular ejection fraction; MACE, major adverse cardiovascu- lar events; PK, pharmacokinetics; RBC, red blood cell; rhEPO, recombinant human erythropoietin; sPAP, systolic pulmonary artery pressure; TIBC, total iron-binding capacity; TSAT, transferrin saturation; VEGF, vascular endothelial growth factor. Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018 4| L. Holdstock et al. postbaseline on-therapy hemoglobin assessment. The safety randomized, 250 (>99%) were included in the safety population population in the daprodustat groups comprised all participants and 235 (93%) in the ITT population, with 222 (88%) completing who received at least one dose of study treatment, whereas it the study (i.e. participants who completed the Week 24 visit re- comprised all randomized participants in the control groups. gardless of whether they remained on study treatment) [148 For the primary analysis, mean hemoglobin concentration at (86%) in the daprodustat group and 74 (93%) in the control Week 24 was measured and corresponding 95% confidence inter- group]. vals were calculated. As part of the secondary analysis, percent- The most common reasons for premature withdrawal were age of time hemoglobin values were within, above and below the an AE in the daprodustat group (5%) and withdrawal by partici- target range between Weeks 12 and 24 (inclusive) was also calcu- pant in the control group (3%). lated. Efficacy endpoints were summarized in the ITT population Similar proportions of participants (20% randomized to dap- for each rhEPO group (naı ¨ve and user) and the two treatment rodustat; 18% randomized to control) discontinued study treat- groups (daprodustat and control) using summary statistics. In ment prematurely. The most common reasons for treatment particular, hemoglobin was summarized by rhEPO group, the two discontinuation were an AE in the daprodustat group (10%) and cohorts (based on hemoglobin entry criteria) and treatment reaching the protocol-defined stopping criteria in the control group. The data for the control groups served as a reference only; group (8%) [included renal transplant, increased systolic pul- no formal comparisons with daprodustat were made. monary artery pressure (sPAP) of 20 mmHg, drop of left ven- tricular ejection fraction (LVEF) 10% from baseline and <50% and blood transfusion]. Participants that finished the entire RESULTS treatment as well as those with an abbreviated treatment schedule were included in the ITT population. Study population Overall, baseline characteristics were well balanced between Two hundred fifty-two participants were randomized to either the combined daprodustat and control groups. The majority of daprodustat (n¼ 172) or control (n¼ 80) (Figure 1). Of those participants were female (59%), with a mean body mass index FIGURE 1: Study ﬂow diagram. One participant withdrew from the study because of an AE; however, this participant’s AE was actually a protocol-deﬁned stopping cri- terion and is presented here in the stopping criteria category rather than the AE category. Completed all study visits or abbreviated schedule if study medication was discontinued. Fifteen participants discontinued study medication because of AEs; however, ﬁve of these participants had AEs that were actually protocol-deﬁned stop- ping criteria and are presented here in the stopping criteria category rather than the AE category. One participant discontinued study medication because of an AE; however, this participant’s AE was actually a protocol-deﬁned stopping criterion and is presented here in the stopping criteria category rather than the AE category. Protocol-deﬁned reasons for discontinuation of study medication included hemoglobin <7.5 g/dL, renal transplant, increased sPAP of 20 mmHg, drop of LVEF 10% from baseline and <50% and blood transfusion. The complete list of stopping criteria is included in Supplementary data, Item S1. IP: investigational product; PI: princi- pal investigator. Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018 Daprodustat for anemia of CKD | 5 Table 3. Participant characteristics (ITT population) rhEPO naı ¨ve rhEPO user Total Daprodustat Control Daprodustat Control Daprodustat Control (n¼ 123) (n¼ 43) (n¼ 33) (n¼ 36) (n¼ 156) (n¼ 79) Age, years, mean (SD) 67.6 (12.21) 64.3 (14.22) 62.0 (14.06) 66.7 (12.89) 66.5 (12.78) 65.4 (13.6) Sex, female, n (%) 75 (61) 23 (53) 17 (52) 23 (64) 92 (59) 46 (58) Weight (kg), mean (SD) 73.2 (19.64) 77.1 (16.03) 77.1 (23.49) 71.7 (18.60) 74.1 (20.49) 74.6 (17.35) BMI (kg/m ), mean (SD) 27.7 (6.9) 28.4 (5.4) 28.1 (7.1) 27.3 (6.4) 27.8 (6.9) 27.9 (5.9) Hemoglobin criteria n (%) Cohort 1 52 (42) 18 (42) 21 (64) 17 (47) 73 (47) 35 (44) Cohort 2 77 (58) 25 (58) 12 (36) 19 (53) 83 (53) 44 (56) Baseline hemoglobin level (g/dL), mean (SD) 9.9 (0.80) 9.9 (0.73) 10.2 (0.62) 10.2 (0.57) 10.0 (0.77) 10.1 (0.67) Type of rhEPO, n (%) Epoetin SC 0 0 14 (42) 5 (14) 14 (42) 5 (14) Darbepoetin 0 0 19 (58) 30 (86) 19 (58) 30 (86) Geographical region, n (%) Japan 17 (14) 6 (14) 9 (27) 9 (25) 26 (17) 15 (19) North America 35 (28) 6 (14) 6 (18) 7 (19) 41 (26) 13 (16) Russia 15 (12) 10 (23) 9 (27) 2 (6) 24 (15) 12 (15) Rest of world 56 (46) 21 (49) 9 (27) 18 (50) 65 (42) 39 (49) Stage of CKD, n (%) 2 0 0 0 1 (3) 0 1 (1) 3a 4 (3) 1 (2) 1 (3) 1 (3) 5 (3) 2 (3) 3b 23 (19) 6 (14) 3 (9) 3 (8) 26 (17) 9 (11) 4 58 (47) 17 (40) 17 (52) 15 (42) 75 (48) 32 (41) 5 38 (31) 19 (44) 12 (36) 16 (44) 50 (32) 35 (44) Baseline eGFR (mL/min/1.73 m ), mean (SD) 21.3 (10.69) 19.4 (10.9) 17.9 (9.17) 18.8 (11.97) 20.6 (10. 45) 19.2 (11.3) Cardiovascular risk factors, n (%) n ¼ 134 n ¼ 45 n ¼ 36 n ¼ 35 n ¼ 170 n ¼ 80 Any 29 (22) 12 (27) 9 (25) 6 (17) 38 (22) 18 (23) Angina pectoris 7 (5) 5 (11) 0 3 (9) 7 (4) 8 (10) Hyperlipidemia 4 (3) 2 (4) 1 (3) 0 5 (3) 2 (3) Hypertension 2 (1) 0 1 (3) 0 3 (2) 0 Diabetes 0 0 2 (6) 0 2 (1) 0 BMI, body mass index; CKD, chronic kidney disease; eGFR, estimated glomerular ﬁltration rate; rhEPO, recombinant human erythropoietin; SC, subcutaneous; SD, standard deviation. Hemoglobin entry criteria for rhEPO-naı ¨ve participants were 8–10 g/dL in Cohort 1 and 8–11 g/dL in Cohort 2. Criteria in rhEPO users were 9.0–10.5 g/dL in Cohort 1 and 9.0–11.5 g/dL in Cohort 2. Baseline hemoglobin levels for Cohort 1 and Cohort 2 combined. Safety population. (BMI) of 27.8 kg/m and mean age of 66.1 years (Table 3). Mean group versus 10.7 g/dL (Cohort 1) and 11.0 g/dL (Cohort 2) in the baseline hemoglobin was 10.0 g/dL and most participants had control group. Stage 4 (46%) or Stage 5 (36%) CKD. This represented an increase from baseline in rhEPO-naı ¨ve participants within the daprodustat and control groups for Cohorts 1 and 2 (Figure 2). In rhEPO users, hemoglobin at 24 Exposure to study medication weeks was similar to baseline levels in both daprodustat co- In the rhEPO-naı ¨ve group (n¼ 179), 134 participants received a horts and increased in both control cohorts. median daprodustat dose of 1 mg at Week 20. Of the rhEPO users (n¼ 71), 36 received a median daprodustat dose of 2 mg at Secondary analyses. Participants had hemoglobin levels within Week 20. Forty-five rhEPO-naı ¨ve participants in the control the target range a median of 82% and 66% of the time and above group (rhEPO per investigator discretion) received a median the target range a median of 0% and 18% of the time between rhEPO dose, standardized to IV epoetin, of 16.5 IU/kg/week at Weeks 12 and 24 in the daprodustat and control groups, respect- Week 20. The remaining 35 rhEPO users in the control arm ively (Table 4). Similar results were seen in a sensitivity analysis received 51.3 IU/kg/week at Week 20 (Supplementary data, performed in all ITT participants where a participant on dapro- Table S3 in Item S5). One study participant in the control arm dustat was considered as having hemoglobin below target range was incorrectly randomized based on prior rhEPO exposure, once he/she discontinued randomized study medication. which accounts for the difference in the safety/drug exposure However, for participants in the control arm, the hemoglobin populations as compared with the randomized population. measurement was used. If the hemoglobin measurement was missing, we assumed it to be within the hemoglobin target range (Supplementary data, Table S4 in Item S6). Seventy-four Hemoglobin response percent and 59% of participants in the combined daprodustat Primary analysis. Mean hemoglobin levels at Week 24 were and combined control groups, respectively, had hemoglobin lev- 10.2 g/dL (Cohort 1) and 10.9 g/dL (Cohort 2) in the daprodustat els within the target hemoglobin range at Week 24 (Table 4). Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018 6| L. Holdstock et al. AC BD FIGURE 2: Mean and 95% conﬁdence interval for hemoglobin over time by original (A) and amended (B) hemoglobin criteria for the rhEPO-naı ¨ve group and original (C) and amended (D) hemoglobin criteria for the rhEPO user group. For Cohort 1, the hemoglobin entry criterion was 8–10 g/dL in the rhEPO-naı ¨ve group and 9–10.5 g/dL in rhEPO users, and the hemoglobin target range was 9–10.5 g/dL. For Cohort 2, the hemoglobin entry criterion was 8–11 g/dL in the rhEPO-naı ¨ve group and 9–11.5 g/dL in rhEPO users, and the hemoglobin target range was 10–11.5 g/dL. Baseline values were calculated as the average of measurements at Week-4 and Day 1. BL, baseline; rhEPO, recombinant human erythropoietin. Table 4. Hemoglobin time in or above target range (ITT population) rhEPO naı ¨ve rhEPO user Total Daprodustat Control Daprodustat Control Daprodustat Control (n¼ 112) (n¼ 38) (n¼ 30) (n¼ 32) (n¼ 142) (n¼ 70) Percentage of time spent with hemoglobin in or above target range between Weeks 12 and 24 Within the target range, median 81 47 94 88 82 66 Above the target range median 033 0 5 0 18 Participants with hemoglobin in or above target range at Week 24 Within the target range, n (%) 78 (74) 20 (56) 22 (73) 19 (63) 100 (74) 39 (59) Above the target range, n (%) 23 (22) 14 (39) 5 (17) 9 (30) 28 (21) 23 (35) rhEPO, recombinant human erythropoietin. Target range was 9–10.5 g/dL in the original and 10–11.5 g/dL in the amended criteria. The minimum and maximum for all values were 0 and 100, respectively. The changes in hemoglobin were associated with expected Table 5. Maximum observed plasma EPO levels (IU/L), plasma VEGF changes in red blood cell (RBC) count, hematocrit and absolute level (ng/L) and change from baseline (ITT population) reticulocyte count (Supplementary data, Table S5 in Item S7). Combined Combined Three participants on daprodustat (two rhEPO naı ¨ve and one daprodustat control rhEPO user) discontinued treatment because of the protocol- (n¼ 156) (n¼ 78) defined hemoglobin stopping criteria (hemoglobin<7.5 g/dL). In the rhEPO-naı ¨ve group, six participants (4%) on daprodustat Plasma EPO levels (IU/L) and four participants (5%) in the control group had hemoglobin Baseline, median (min, max) 10.5 (2.5, 100.3) 11.4 (2.5, 61.5) 13 g/dL over the course of the 24-week treatment period. Maximum observed EPO 16.1 (4.2, 109.3) 19.2 (4.9, 620.6) Among the rhEPO users, no participants on daprodustat and one level, median (min, max) participant (3%) in the control group had hemoglobin 13 g/dL. Plasma VEGF level (ng/L) Baseline, median (min, max) 65.8 (17.9, 467.9) 68.7 (22.9, 1118.5) For simplicity, data for parameters other than hemoglobin are Maximum observed VEGF, 105.8 (33.1, 1205.3) 94.5 (25.5, 472.7) presented for all participants on daprodustat (combined daprodu- median (min, max) stat group) and all control participants (combined control group). rhEPO, recombinant human erythropoietin; VEGF, vascular endothelial growth Erythropoietin and vascular endothelial growth factor factor. levels Daprodustat measurements predose and postdose up to 15 h after dosing; control included a single sample during visit (predose if scheduled to receive rhEPO at visit). Plasma EPO levels remained near baseline in the combined dap- Baseline is the last predose value. rodustat group and the combined control group (single sample c Daprodustat measurements predose and postdose up to 15 h after dosing; control collected during the study visit, predose if scheduled to receive included a single sample during visit (predose if scheduled to receive rhEPO at visit). rhEPO) (Table 5). Baseline plasma VEGF values were similar in the combined Iron use, metabolism and utilization daprodustat and control groups (Table 5). VEGF levels were highly variable at all time points, with no clear signal for a During the treatment period, 67% of participants received oral change in any treatment group. iron and 10% of participants received IV iron. Participants in the Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018 Daprodustat for anemia of CKD | 7 FIGURE 3: Geometric mean and 95% conﬁdence interval for hepcidin (mg/L) by visit [intent-to-treat (ITT) population]. Table 6. AEs and frequency of MACE and component endpoints in The most frequently reported AEs (5%) in the combined dapro- the combined daprodustat or combined control group (safety dustat group were nasopharyngitis, diarrhea and nausea. One population) participant (<1%) discontinued study treatment/withdrew from the study due to nausea. Combined Combined On-therapy serious AEs (SAEs) were reported in 26 partici- daprodustat control pants (15%) in the combined daprodustat group and 13 partici- (n¼ 170) (n¼ 80) pants (16%) in the combined control group. The only SAE AEs occurring in 5% of participants reported in >1% of participants in either treatment group was Any event, n (%) 119 (70) 54 (68) worsening CKD [chronic renal failure-MedDRA (Medical Nasopharyngitis 20 (12) 3 (4) Dictionary for Regulatory Activities Terminology)], which was Diarrhea 10 (6) 6 (8) reported in three participants (2%) in the combined daprodu- Nausea 8 (5) 1 (1) stat group and three participants (4%) in the combined control Arthralgia 3 (2) 4 (5) group. Cardiovascular events, including major adverse cardio- Pruritus 0 4 (5) vascular events (MACEs) and component endpoints, are sum- Frequency of MACE and component endpoints marized in Table 6. Other preidentified AEs of interest are Composite endpoints, n (%) described in the Supplementary data, in Item S9. Any MACE 5 (3) 1 (1) b Mean (SD) estimated glomerular filtration rate (eGFR) at Any MACEþ 7 (4) 4 (5) c baseline was 20.6 (10.5) mL/min/1.73 m in the combined dapro- Components of composite endpoints, n (%) dustat group and 19.2 (11.3) mL/min/1.73 m in the combined All-cause mortality 4 (2) 1 (1) control group and did not change by end of study [0.2 mL/min/ MI (fatal or nonfatal) 1 (<1) 0 2 2 1.73 m (daprodustat) and 1.0 mL/min/1.73 m (control)]. Two Stroke (fatal or nonfatal) 0 0 participants (1%) on daprodustat and three (4%) on control Hospitalization due to heart failure 4 (2) 4 (5) started dialysis during the study. MACE, major adverse cardiovascular event; MI, myocardial infarction. Seven participants (4%) on daprodustat and one (1%) in the MACE is deﬁned as a ﬁrst occurrence of all-cause mortality, a nonfatal MI or a control group had a blood transfusion during the study. All nonfatal stroke. transfusions occurred during hospitalization for various med- MACEþ is deﬁned as a ﬁrst occurrence of MACE or hospitalization for heart ical conditions (e.g. sepsis, congestive heart failure exacerba- failure. c tion, non-GI bleeding); in six of seven participants, these Includes all events. occurred at a hemoglobin level>7.5 g/dL. LVEF, estimated sPAP and baseline values for other echo- combined daprodustat group received a mean total of 38.1 g of cardiographic parameters were well balanced in all groups, as oral elemental iron in the study (0.16 g/day) and participants in were the absolute changes from baseline in on-treatment as- the control group received a mean total of 28.0 g (0.13 g/day). sessments (Supplementary data, Table S7 in Item S9). Six The time course of change in hepcidin is shown in Figure 3, (4%) participants in the combined daprodustat group and four and the baseline and change from baseline at Week 24 in meas- (5%) in the combined control group had LVEF values <50% ures of iron metabolism are shown in Supplementary data, at any time during the study. Thirteen (9%) and five (7%) Table S6 and Supplementary data, Figure S2 in Item S8. participants in the combined daprodustat and control groups, respectively, had an absolute decrease from baseline in LVEF of >10%. One (<1%) participant in the daprodustat group and Safety one (2%) in the control group had an estimated sPAP value that increased>20 mmHg from baseline at any time during the The incidence of any on-therapy AEs was 70% in the combined study. daprodustat and 68% in the combined control groups (Table 6). Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018 8| L. Holdstock et al. Table 7. Summary of SBP and DBP at Week 24 (safety population) lower doses of daprodustat were required to maintain hemoglo- bin in target ranges. It is conceivable that efficacy of distinct Combined HIF-PHI investigational products differs and that higher dapro- daprodustat Control dustat doses than needed to achieve target hemoglobin are (n¼ 170) (n¼ 80) required for increased hepcidin suppression. Participants were considered iron replete at baseline based SBP (mmHg) on the currently acceptable ferritin and transferrin saturation Baseline entry criteria and, although rates of IV iron supplementation in n 150 74 Mean (SD) 138.0 (18.5) 136.4 (16.5) the study were very low (10%), 67% of participants received Change from baseline at Week 24 oral iron supplementation, albeit at low doses. Though HIF-PHIs n 137 66 are hypothesized to be iron sparing, this trial, as well as other Mean (SD) 0.4 (16.9) 0.1 (15.3) published CKD trials [11–13], have failed to show a reduction in DBP (mmHg) iron usage; it is likely that trials of longer duration are required Baseline to demonstrate a difference. n 150 74 Treatment with daprodustat for up to 24 weeks demon- Mean (SD) 73.0 (12.6) 72.1 (10.6) strated an AE profile consistent with the comorbidities typical Change from baseline at Week 24 of the CKD population . No new safety signals based on the n 137 66 collective safety data were identified for daprodustat. Mean (SD) 0.1 (10.5) 0.3 (9.9) Seven (4%) participants in the daprodustat group and one (1%) in the control group had a blood transfusion, with the most DBP, diastolic BP; SBP, systolic BP. a frequent reason for transfusions cited by investigators being Baseline is the last predose value. underlying chronic anemia followed by active bleeding (not gastrointestinal). It is possible that transfusions were infrequent in the control group because rhEPO in these participants could No adverse trends were noted for clinical laboratory values, be titrated frequently at the investigator’s discretion. In the dap- electrocardiogram (ECG) values or BP; data for the latter are rodustat group, adjustments were only allowed per the prespe- shown in Table 7. Furthermore, no changes were seen in visual cified dose adjustment algorithm every 4 weeks. acuity or intraocular pressure based on the protocol-specified Although the study was relatively small and too short in dur- ophthalmology examinations (data not shown). ation to determine MACE frequency and events were not for- mally adjudicated, the overall rate of MACE in the study was lower than anticipated based on the historical data [2, 4, 25]. DISCUSSION A long-term cardiovascular outcome study is ongoing to conclu- sively assess these events. To our knowledge, this is the largest (>250 participants random- There were no noteworthy changes in the measured echo- ized) and longest (24-week treatment duration) study reported cardiographic parameters for either of the treatment groups. to date of a HIF-PHI in the anemic CKD population not yet on This is in contrast to a 24-week study of daprodustat in patients dialysis. Results demonstrate that daprodustat can increase on hemodialysis where there was an imbalance in the number and maintain hemoglobin over 24 weeks within a narrow pre- of participants with an increase from baseline in estimated specified target range, most notably evidenced by the percent- sPAP of >20 mmHg in the daprodustat group relative to control age of time hemoglobin was in the target range between Weeks due to a number of potential confounders . Additionally, 12 and 24 for the rhEPO-naı ¨ve participants (81%) and rhEPO dedicated preclinical data (GlaxoSmithKline, unpublished data) users (94%). and a clinical study in healthy volunteers  have demon- Consistent with its mechanism of action and similar to pre- strated no effect of daprodustat on peak right ventricular pres- vious reports,  the effect of daprodustat on hemoglobin sure or sPAP, respectively, even in the setting of a hypoxic occurred with minimal changes in plasma EPO levels that were challenge. similar in magnitude to endogenous EPO levels seen during ex- No adverse trends were apparent in either treatment group posure to high altitudes in healthy individuals [17, 18]. If higher following review of the laboratory data, vital signs, ECGs and EPO levels contribute to the cardiovascular risk associated with ophthalmology examinations. This included no change in BP for rhEPO,  then daprodustat may provide a safer cardiovascular the combined daprodustat and control groups. Contrary to the risk profile. known risk of hypertension associated with rhEPOs and its ana- Daprodustat and control had minimal, if any, effect on VEGF, logs, [7, 28] the lack of effect on BP in the combined control one of the target genes of HIF, consistent with prior observa- group could be due to the overall smaller doses of rhEPO in this tions at similar  and higher daprodustat doses (10–100 mg) study and adequate BP control in the rhEPO user group.  as well as with other HIF-PHIs . This is also consistent There are several limitations in this study that need to be con- with preclinical work demonstrating that VEGF expression is sidered. Amending the hemoglobin entry criteria during study less sensitive to HIF activation than is EPO expression . conduct necessitated analyzing changes in hemoglobin in the Measures of iron metabolism changed in a manner con- context of different entry criteria and target ranges. This, how- sistent with data from previous studies with several PHI agents ever, did not affect the interpretation of the hemoglobin time-in- [9–13]. Hepcidin levels, in general, are suppressed by erythropoi- range data where daprodustat was shown to achieve and main- esis, due to increased iron utilization and decreased available tain hemoglobin within a relatively narrow target range. Given stores [21–23]. In the combined daprodustat group, hepcidin lev- the global nature of the study and differing practices in treating els decreased by 17% while hemoglobin increased, suggesting anemia, the control group received rhEPO per standard of care as that this effect could be secondary to erythropoiesis. While the necessary, with the added instruction to treat to the hemoglobin decrease in hepcidin is not as prominent as reported in the ear- target range. This lack of standardization as to how and when lier dose-ranging studies with daprodustat [9, 10] in this study, Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018 Daprodustat for anemia of CKD | 9 rhEPO was utilized may have contributed to fewer participants submission. All authors agree to be accountable for all as- with hemoglobin levels within the target range and more above pects of the work in ensuring that questions related to the the target range, which may have been consistent with regional accuracy or integrity of any part of the work are appropri- standard of care. This was not the case with daprodustat, which ately investigated and resolved. was dosed based on a prespecified dose adjustment algorithm. Additionally, the tight eligibility criteria limit the generalizability of the findings, and the degree of noncompletion may complicate CONFLICT OF INTEREST STATEMENT interpretation of the results. B.C., A.M.M., N.B., D.J., J.J.L. and A.R.C. are employees of, and Finally, the open-label design could potentially influence hold stock in, GlaxoSmithKline (GSK). D.J., J.J.L. and A.R.C. how AEs are reported by participants or recorded by study staff. also own stock options in GSK. L.H. and B.M.J. are former For example, more AEs of nausea were reported in the daprodu- stat group, which could have been influenced by daprodustat employees of GSK, and hold stock options in GSK. S.G.K. is being an investigational medication, or due to an oral route of an employee of Hallym University, and receives research daprodustat administration. Furthermore, cardiovascular funding from C.J. Healthcare and Fibrogen. S.Z. has no con- events were not formally adjudicated, even though overall ﬂicts to declare. This manuscript is not under consideration number of MACEs was low. Although this is the largest study to for publication elsewhere. The results presented in this date, the sample size and 6-month duration is not long enough paper have not been published previously in whole or part, to draw conclusions regarding the long-term safety of daprodu- except in abstract form. Some of the data included in the stat, particularly regarding cardiovascular risk and cancer. present manuscript were presented at the American Society In conclusion, in this 24-week study in anemic CKD patients, of Nephrology meeting in Chicago, IL, 15–20 November 2016. daprodustat effectively achieved target hemoglobin levels in rhEPO-naı ¨ve participants and maintained target hemoglobin levels in rhEPO users. Daprodustat was well tolerated with an REFERENCES AE profile consistent with the CKD population. These data sup- port continued development of daprodustat to treat anemia of 1. Bonomini M, Del Vecchio L, Sirolli V et al. New treatment CKD, regardless of whether patients are rhEPO naı ¨ve or already approaches for the anemia of CKD. Am J Kidney Dis 2016; 67: taking rhEPO. This includes an ongoing large-scale cardiovascu- 133–142 lar outcomes trial (ASCEND-ND) in this population to assess the 2. Singh AK, Szczech L, Tang KL et al. Correction of anemia with safety and efficacy of daprodustat (NCT02876835). epoetin alfa in chronic kidney disease. N Engl J Med 2006; 355: 2085–2098 3. Dru ¨ eke TB, Locatelli F, Clyne N et al. Normalization of hemo- SUPPLEMENTARY DATA globin level in patients with chronic kidney disease and an- Supplementary data are available at ckj online. emia. N Engl J Med 2006; 355: 2071–2084 4. Pfeffer MA, Burdmann EA, Chen C-Y et al. A trial of darbepoe- tin alfa in type 2 diabetes and chronic kidney disease. N Engl ACKNOWLEDGEMENTS J Med 2009; 361: 2019–2032 The authors thank the study participants, investigators and 5. Jenkins J. Statement regarding erythropoiesis-stimulating site staff (see ClinicalTrials.gov NCT01977573 for site list). The agents (ESA) before the Committee on Ways and Means authors acknowledge the following GlaxoSmithKline (GSK) Subcommittee on Health, US House of Respresentatives. employees for their assistance with study management and Government Publishing Ofﬁce. https://www.gpo.gov/fdsys/pkg/ critical review: Deborah Kelly, MD, and Douglas Wicks, MPH, CHRG-110hhrg49981/pdf/CHRG-110hhrg49981.pdf. Published 26 June 2007 (20 December 2017, date last accessed) CMPP. Authors meet the authorship criteria set forth by the 6. FDA Drug Safety Communication: Modiﬁed dosing recom- International Committee for Medical Journal Editors. Medical mendations to improve the safe use of erythropoiesis- editorial support (Nancy Price, PhD, and Gautam Bijur, PhD) stimulating agents (ESAs) in chronic kidney disease. US Food and graphic services were provided by AOI Communications, and Drug Administration. https://www.fda.gov/Drugs/Drug L.P., and were funded by GSK. B.M.J. was previously afﬁliated Safety/ucm259639.htm. Published 24 June 2011 (20 with Clinical Pharmacology Modeling and Simulation, December 2017, date last accessed) GlaxoSmithKline, Research Triangle Park, NC, USA. 7. Procrit [package insert].Horsham,PA: Janssen Products, LP;2013 8. Johnson BM, Stier BA, Caltabiano S. Effect of food and gemﬁbro- FUNDING zil on the pharmacokinetics of the novel prolyl hydroxylase in- hibitor GSK1278863. Clin Pharmacol Drug Dev 2014; 3: 109–117 Funding for this study (clinicaltrials.gov: NCT01977573; 9. Brigandi RA, Johnson B, Oei C et al. A novel hypoxia- EudraCT: 2013-002681-39; GSK registry: PHI113747) was pro- inducible factor-prolyl hydroxylase inhibitor (GSK1278863) vided by GlaxoSmithKline. for anemia in CKD: a 28-day, phase 2A randomized trial. Am J Kidney Dis 2016; 67: 861–871 10. Holdstock L, Meadowcroft AM, Maier R et al. Four-week stud- AUTHORS’ CONTRIBUTIONS ies of oral hypoxia-inducible factor-prolyl hydroxylase in- Concept and study was designed by L.H., B.C., A.M.M., N.B., hibitor GSK1278863 for treatment of anemia. J Am Soc Nephrol B.M.J., D.J., J.J.L. and A.R.C.; S.G.K. and S.Z. were in charge of 2016; 27: 1234–1244 data acquisition; statistical analysis was conduted by N.B. 11. Pergola PE, Spinowitz BS, Hartman CS et al. Vadadustat, a and D.J. All authors reviewed the analysis and interpreted novel oral HIF stabilizer, provides effective anemia treat- the data; L.H. wrote the initial draft and all authors contrib- ment in nondialysis-dependent chronic kidney disease. uted critically and approved of the ﬁnal manuscript for Kidney Int 2016; 90: 1115–1122 Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018 10 | L. Holdstock et al. stimulates erythropoietin production without hypertensive 12. Besarab A, Provenzano R, Hertel J et al. Randomized placebo- controlled dose-ranging and pharmacodynamics study of effects. PLoS One 2014; 9: e111838 roxadustat (FG-4592) to treat anemia in nondialysis- 21. Ashby DR, Gale DP, Busbridge M et al. Plasma hepcidin levels dependent chronic kidney disease (NDD-CKD) patients. are elevated but responsive to erythropoietin therapy in Nephrol Dial Transplant 2015; 30: 1665–1673 renal disease. Kidney Int 2009; 75: 976–981 13. Provenzano R, Besarab A, Sun CH et al. Oral hypoxia- 22. Pinto JP, Ribeiro S, Pontes H et al. Erythropoietin mediates inducible factor prolyl hydroxylase inhibitor roxadustat (FG- hepcidin expression in hepatocytes through EPOR signaling 4592) for the treatment of anemia in patients with CKD. Clin J and regulation of C/EBPalpha. Blood 2008; 111: 5727–5733 Am Soc Nephrol 2016; 11: 982–991 23. Sasaki Y, Noguchi-Sasaki M, Yasuno H et al. Erythropoietin 14. Lenihan CR, Winkelmayer WC. The dawning of a new day in stimulation decreases hepcidin expression through hemato- CKD anemia care? J Am Soc Nephrol 2016; 27: 968–970 poietic activity on bone marrow cells in mice. Int J Hematol 15. Imai E, Yasuda Y, Makino H. Japan association of chronic 2012; 96: 692–700 kidney disease initiatives (J-CKDI). Jpn Med Assoc J 2011; 54: 24. Tonelli M, Wiebe N, Guthrie B et al. Comorbidity as a driver of 403–405 adverse outcomes in people with chronic kidney disease. Kidney Int 2015; 88: 859–866 16. Levey AS, Stevens LA, Schmid CH et al. A new equation to esti- mate glomerular ﬁltration rate. Ann Intern Med 2009; 150: 604–612 25. United States Renal Data System. USRDS 2013 Annual Data 17. Milledge JS, Cotes PM. Serum erythropoietin in humans at Report. Atlas of Chronic Kidney Disease and End-Stage Renal high altitude and its relation to plasma renin. J Appl Physiol Disease in the United States. Bethesda, MD: National Institutes 1985; 59: 360–364 of Health, National Institute of Diabetes and Digestive and 18. Piperno A, Galimberti S, Mariani R et al. Modulation of hepci- Kidney Diseases, 2013 din production during hypoxia-induced erythropoiesis in 26. Meadowcroft AM, Cizman B, Holdstock L et al. Daprodustat humans in vivo: data from the HIGHCARE project. Blood for anemia: a 24-week, open-label, randomized, controlled 2011; 117: 2953–2959 trial in participants on hemodialysis. Clin Kidney J 2018 19. McCullough PA, Barnhart HX, Inrig JK et al. Cardiovascular 27. Demopoulos L, Haws T, Mahar K et al. Lack of correlation be- toxicity of epoetin-alfa in patients with chronic kidney dis- tween PK exposure and changes in pulmonary artery systolic ease. Am J Nephrol 2013; 37: 549–558 pressure in healthy volunteers on the prolyl hydroxylase 20. Flamme I, Oehme F, Ellinghaus P et al. Mimicking hypoxia to inhibitor, GSK1278863. Pharmacotherapy 2014; 34: e222 treat anemia: HIF-stabilizer BAY 85-3934 (Molidustat) 28. Aranesp [package insert]. Thousand Oaks, CA: Amgen Inc., 2017 Downloaded from https://academic.oup.com/ckj/advance-article-abstract/doi/10.1093/ckj/sfy013/4925438 by Ed 'DeepDyve' Gillespie user on 12 July 2018
Clinical Kidney Journal – Oxford University Press
Published: Mar 9, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera