A phase 3b, multicenter, open‐label, single‐arm study of roxadustat (ASPEN): Operational learnings within United States dialysis organizations

Roxadustat is an oral hypoxia‐inducible factor prolyl hydroxylase inhibitor approved in several regions for the treatment of anemia of chronic kidney disease (CKD). ASPEN evaluated the efficacy, safety, and feasibility of roxadustat in patients with anemia of CKD in US dialysis organizations.


INTRODUCTION
2][3] The prevalence of anemia increases with CKD stage and disease severity, with 90% of patients with dialysis-dependent (DD)-CKD experiencing anemia. 4atients with anemia of CKD often experience symptoms such as fatigue, and a poor quality of life. 1 In addition, anemia is associated with an increased risk of all-cause mortality, cardiovascular mortality, major adverse cardiovascular events, hospitalization, and CKD progression. 1,4With increasing anemia severity, there is a heightened risk of these adverse outcomes. 1urrently, the standard of care treatment for anemia in patients with DD-CKD in the US primarily involves the use of erythropoiesis-stimulating agents (ESAs) and intravenous (IV) iron supplementation. 57][8] As these treatments are administered parenterally, they also require in-center administration.Thereby, there is a need for alternative treatments for anemia of CKD that are safe, effective, and can be administered in the home setting.
Roxadustat is an oral hypoxia-inducible factor prolyl hydroxylase inhibitor approved to treat anemia of CKD in patients with DD-CKD and nondialysis-dependent CKD in several regions, including China, Japan, the European Union, and the UK.0][11][12][13][14] However, the phase 3 roxadustat program did not include specific assessments of the operational characteristics of roxadustat use in hemodialysis organizations.Assessing how effective roxadustat is at maintaining Hb levels when administered in dialysis centers by dialysis nurses is important for its safe and effective use in the real world, particularly given its oral mode of administration and dose titration algorithm.
The ASPEN study aimed to evaluate the efficacy and safety of roxadustat in patients with anemia of CKD on hemodialysis, and to assess the feasibility of maintaining Hb levels with oral roxadustat in US dialysis organizations.

Study design
This open-label, single-arm study (NCT04484857) evaluated the efficacy and safety of roxadustat in maintaining Hb in patients with anemia of CKD receiving dialysis in 28 large, US community-based dialysis centers belonging to DaVita and US Renal Care.A 6-week screening period was followed by 24 weeks of treatment and posttreatment follow-up of 4 weeks (Figure 1).An optional treatment extension period up to 1 year after completion of the 24-week treatment period was offered to patients.
This study was conducted in accordance with the Declaration of Helsinki, the International Conference on Harmonisation E6 Guidance for Good Clinical Practice, and applicable local health and regulatory requirements.All patients provided written consent before screening.The final study protocol and informed consent form were approved by the applicable Independent Ethics Committee or Institutional Review Board for each site.

Patients
Patients were included in the study if all of the following criteria were met: ≥18 years of age and receiving chronic dialysis for kidney failure; a priori ESA user, that is, ESA use for ≥6 weeks prior to screening with a stable dose (as determined by the investigator) during the 4 weeks prior to initiation of roxadustat, or a non-ESA user a initiating anemia therapy, that is, <6 weeks of prior ESA use or no prior ESA use; a screening Hb level of 9.0-12.0g/dL (ESA user) or <10.0 g/dL (non-ESA user) based on central laboratory value (10 days prior to study treatment initiation); a functioning native arteriovenous (AV) fistula or graft with adequate flow or a permanent tunneled catheter; ferritin ≥50 ng/mL and transferrin saturation (TSAT) ≥10%; alanine aminotransferase and aspartate aminotransferase ≤3 times the upper limit of normal (ULN), and total bilirubin ≤1.5 times ULN; and estimated dry bodyweight 45.0-160.0kg.
If any of the following criteria were met, patients were excluded from the study: a red blood cell (RBC) transfusion or treatment with iron-chelating agents within the 4 weeks prior to enrollment; New York Heart Association Class III or IV chronic heart failure, uncontrolled hypertension (by the opinion of the investigator), or a history of a myocardial infarction, acute coronary syndrome, stroke, seizure, or a thrombotic/thromboembolic event (excluding vascular dialysis access stenosis/thrombosis), within the 12 weeks prior to enrollment; history of chronic liver disease or malignancy (except cancers determined to be cured or in remission for ≥2 years, curatively resected basal cell or squamous cell skin cancers, cervical cancer in situ, or resected colonic polyps); diagnosis or suspicion of renal cell carcinoma; use or anticipated use of dapsone or androgen in any dose amount, or acetaminophen or paracetamol (>2.0 g/day).Full inclusion/exclusion criteria are provided in the Supplementary material.

Study treatment
Following screening, all patients received oral roxadustat three times a week for up to 24 weeks.Roxadustat was administered by a dialysis nurse in-center and was taken with or without food.Allowed and prohibited concomitant medications are outlined in the Supplementary material.
The starting dose of roxadustat and the dose titration algorithm were based on the phase 3 studies involving patients with kidney failure who were receiving chronic dialysis treatment.For ESA users, the initial roxadustat dose was based on the average prescribed ESA dose in the 4 (epoetin alfa and darbepoetin alfa) or 8 weeks (methoxy polyethylene glycol-epoetin beta) prior to the treatment period, and ranged from 70 mg/dose TIW to 150 mg/dose TIW (Table S1).If the converted initial dose exceeded the maximum dose of 3.0 mg/kg/dose, then the next lower dose step was chosen as the initial dose.Roxadustat treatment was started 2 days after stopping epoetin alfa, 1 week after stopping darbepoetin alfa, or 2 weeks after stopping methoxy polyethylene glycol-epoetin beta.Dose adjustments were allowed every 4 weeks, and roxadustat dose was titrated based on the patient's Hb level and rate of Hb change during the previous 4-week period (Table S2).If Hb rise was >2 g/dL within 4 weeks, dose was reduced.Dose was held if Hb was ≥13 g/dL.
Oral iron supplementation was unrestricted during the study.IV iron (five doses of 50 mg/dose) was allowed if patients had ferritin <100 ng/mL or TSAT <20%.If rapid correction of anemia was required or if deemed a medical necessity, RBC transfusion was permitted.Rescue ESA use was restricted to no more than one cycle (first use of ESA to achieve Hb >9 g/dL) during the treatment period and was stopped when Hb >9 g/dL was achieved, or after 4 weeks (whichever came first).If required, initiation of rescue ESA ≥2 days after the last dose of roxadustat was permitted if the patient's Hb was <8.5 g/dL on two consecutive measurements at least 5 days apart, with insufficient response to two or more dose increases, or the maximum dose of roxadustat had been reached, there was no suggestion of iron deficiency or bleeding as a cause of lack of response or rapid decline in Hb, and if the goal of rescue therapy would be to reduce the risk of alloimmunization in transplant-eligible patients and/or to reduce other RBC-transfusion-related risks.If there was further need for rescue in the same patient, non-ESA rescue therapy (RBC transfusion, in patients who received ESA rescue therapy but required further rescue) was used or the patient discontinued the study.
Permanent discontinuation of roxadustat was permitted due to patient or investigator decision, adverse events (AEs), noncompliance, or lack of follow-up.Any patient discontinuing study medication was followed up for 4 weeks post-discontinuation.

Endpoints
The primary efficacy endpoints were the proportion of patients with mean Hb ≥10 g/dL, averaged over weeks 16-24, and the mean Hb change from baseline to the average over weeks 16-24.Pre-specified subgroup analyses were performed by sex, age group, race, baseline Hb categories, baseline iron status, cardiovascular/cerebrovascular/ thromboembolic medical history, starting dose, participating dialysis organizations, and baseline ESA dose category.Exploratory endpoints included the proportion of patients with mean Hb 10.0-12.0g/dL from weeks 16 to 24; the proportion of patients with Hb >12 g/dL and > 13 g/dL at any time during the treatment period; the percentage of time with Hb 10.0-12.0g/dL during weeks 16-24 (among patients who had Hb 10.0-12.0g/ dL at any time during weeks 16-24); the time to first RBC transfusion; the proportion of patients with mean Hb ≥10 g/dL during weeks 0-8; the proportion of patients with mean Hb ≥10 g/dL, averaged over weeks 16-24 by baseline ESA use status; the effect of roxadustat on iron indices; dosing adherence; the starting dose of roxadustat compared to the dose at weeks 16-24 for patients converting from an ESA; and the mean number of dose adjustments during the study.
Safety and tolerability were assessed as the incidence of treatment-emergent adverse events (TEAEs) and treatment-emergent serious adverse events (TESAEs); time to a TEAE of vascular access thrombosis; the proportion of patients with a TEAE of vascular access thrombosis; the proportion of patients with a TEAE of vascular access thrombosis based on type of vascular access; clinically significant changes in laboratory parameters, including iron indices, from baseline; and vital signs.
AEs were coded to the Medical Dictionary for Regulatory Activities, version 24.0.An AE that started during the treatment period was considered a TEAE if it was not present prior to the first dose of study medication, or it was present prior to the first dose of study medication but increased in severity during the treatment.TEAEs were collected until week 24/end of treatment visit for patients continuing into the extension period, or until 4 weeks after the last dose for patients who did not participate in the extension period or who discontinued the study.
Arteriovenous (AV) fistula site complication, AV fistula site pseudoaneurysm, AV fistula thrombosis, AV graft thrombosis, vascular access complication, vascular access malfunction, and vascular access site thrombosis were used to calculate the time to vascular access thrombosis.For patients who developed vascular access thrombosis, the time at risk in weeks was calculated as: (first vascular access thrombosis start date À first dose date of study medication + 1)/7; patients without vascular access thrombosis were censored at last study visit and the time at risk was calculated as: (date of last study visit À first dose date of study medication + 1)/7.
The number of patient exposure years (PEY) for each patient was calculated as the (last dose date À first dose date + 1)/365.25.Incidence rate was calculated as 100 Â the number of patients with events/PEY.

Statistical analysis
A sample size of 300 patients was deemed sufficient to produce a 95% confidence interval (CI), with a width of 11.5%, assuming the proportion of patients with Hb ≥10 g/dL over weeks 16-24 was ≥80%.
The full analysis set (FAS) population comprised all patients who were enrolled in the study and provided baseline Hb data and at least one post-baseline Hb value.All efficacy endpoints were assessed in the FAS population.The safety analysis set comprised all enrolled patients who received at least one dose of roxadustat.All safety endpoints were assessed in the safety population.
A multiple imputation method was used for the primary efficacy endpoint analyses.Missing Hb data in the FAS population were imputed relying on nonmissing data from all patients, using the Monte Carlo Markov Chain imputation model.
The proportion of patients with mean Hb ≥10 g/dL in the first 8 weeks after conversion or initiation of treatment, and the proportion of patients with mean Hb level ≥10 g/dL (averaged from weeks 16 to 24), based on baseline ESA use status, were calculated using the multiple imputation method.Rubin's method was thereafter used to combine the estimates of the proportion of responders (defined as patients with mean Hb ≥10 g/dL after conducting multiple imputation, averaged over weeks 16-24)  and standard errors.The proportion of patients with mean Hb 10.0-12.0g/dL from weeks 16 to 24, proportion of patients with Hb >12 g/dL, proportion of patients with Hb >13 g/dL, and percentage of time with Hb 10.0-12.0g/dL from weeks 16 to 24 were calculated using observed Hb values.Time to first RBC transfusion was analyzed using the Kaplan-Meier method.
All analyses were performed using SAS ® , version 9.1.3or higher.

Primary efficacy endpoints
The proportion of patients with mean Hb ≥10 g/dL during weeks 16-24 was 83.7% (95% CI 78.9-88.6),compared with 78.7% at baseline.Two patients were incorrectly classified as non-ESA users at baseline as they had a history of ESA use >6 weeks prior, and thus the true number of non-ESA users was six instead of eight.At baseline, five of six (83.3%) non-ESA users and 215 of 274 (78.5%)ESA users had mean Hb ≥10 g/dL.During weeks 16-24, a total of five of six (83.3%) non-ESA users and 231 of 276 (83.7%)ESA users achieved a mean Hb ≥10 g/dL.
In the subgroup analysis, the proportions of patients with mean Hb ≥10 g/dL during weeks 16-24 within the subgroups were consistent with the results for the overall population (Figure S1).The proportion of patients with mean Hb ≥10 g/dL during weeks 16-24 was 79.6% (95% CI 71.2-87.9;n = 90/113) in Black or African American patients and 86.5% (95% CI 80.7-92.4;n = 148/169) in patients not Black or African American.
The mean (SD) change in Hb from baseline to the average over weeks 16-24 was 0.2 (1.0) g/dL.In the Patients who participated in the post-treatment follow-up or completed the EOS visit refers to the patients who completed the main 24-week treatment or permanently discontinued from the main 24-week treatment.Patients who participated in the extension treatment period skipped the post-treatment follow-up and EOS visit, and only treatment-emergent serious AEs were collected.AE, adverse event; EOS, end of study; FAS, full analysis set.majority of patients, mean Hb values were stabilized and maintained at the target Hb level (10.0-12.0g/dL) during the 24-week treatment period (Figure 3).
During the treatment period, 12 patients (4.3%) received an RBC transfusion (Figure S2), with an incidence rate of 10.4 per 100 PEY.The proportion of patients requiring RBC transfusion or ESA rescue from week 5 to week 24 or end of treatment was 7.7% (n = 21; 95% CI 4.9-11.6).

Iron indices
Baseline reticulocyte Hb content and serum iron levels were maintained through week 24 (Figure 4).
During the treatment period, mean serum ferritin and TSAT values decreased slightly from baseline.However, these changes were not deemed clinically significant.

Safety
During the treatment period, 180 (63.6%) patients reported TEAEs (Table 2).A small proportion of TEAEs were considered related or possibly related to the study drug (18 [6.4%]), and the majority of TEAEs were grade 1 or 2 in severity.
The proportion of patients who experienced TEAEs of special interest, including myocardial infarction, stroke, seizures, deep vein thrombosis, pulmonary embolism, hypertension, infection, malignancy, and sepsis, are listed in Table S3.
Except for two patients who had markedly elevated bilirubin levels (≥2 Â ULN; patient narratives provided in Supplementary material), potentially clinically significant laboratory abnormalities were consistent with the study population of patients with CKD (Table S5).Over the study treatment period, vital signs remained relatively constant, with minimal changes from baseline in mean diastolic and systolic blood pressure and mean arterial pressure.

DISCUSSION
This study enhances the understanding of how oral roxadustat can be incorporated into hemodialysis organizations.By weeks 16-24, the majority of patients maintained Hb ≥10 g/dL, and this proportion was numerically greater than at baseline.This suggests that patients were successfully treated with roxadustat as demonstrated by maintaining Hb and treatment adherence.
Patient demographics and baseline characteristics, including the proportion of Black/African American patients, were generally consistent with US patients receiving in-center hemodialysis. 15,16In a subgroup analysis, the proportion of patients with Hb ≥10 g/dL during weeks 16-24 was similar between Black/African a Patients reported with a preferred term of "COVID-19 pneumonia" were different patients than those reported with preferred term of "acute respiratory failure".An AE (classified by preferred term) that started during the treatment period was considered a TEAE if it was not present prior to the first dose of study medication, or it was present prior to the first dose of study medication but increased in severity during the treatment.The TEAE investigation period was extended up to 28 days after the last roxadustat dose for patients who did not participate in the extension treatment, or cut to week 24/end of treatment visit when patients were continuing into extension treatment.If a patient had more than one TEAE coded to the same preferred term, the patient was counted only once for that preferred term.Similarly, if a patient had more than one TEAE within a system organ class, the patient was counted only once in that system organ class.PEY for each patient = (last dose date À first dose date + 1)/365.25.Incidence rate = 100 Â number of patients with events/PEY.
American patients and non-Black/African American patients.
During the study, the main reason for dose increases and decreases was protocol-mandated dose adjustment, with similar proportions of patients with dose increases and dose decreases.At DaVita sites, roxadustat dose adjustments were managed by dialysis nursing staff instead of study staff, further supporting the operational feasibility of oral roxadustat in an in-center, communitybased, hemodialysis clinic setting.The maximum initial dose for patients converting from ESA in this study was 150 mg/dose TIW, which is lower than the maximum starting dose in the ROCKIES Phase III trial in patients with dialysis-dependent CKD (200 mg/dose TIW).A lower dose was selected in our trial because in the global Phase III program the dosage occasionally led to a higher than desired increase in Hb.
No new safety signals were identified during the study or during the extension period.However, a longer follow-up period, a larger patient population, and a control group are required to fully assess safety in this population.][19] We highlight that this study occurred during the first wave of the COVID-19 pandemic in 2020.Despite this, study enrollment was successful.The incidence of COVID-19 was comparable to incidence rates in patients on dialysis reported in the literature and reflect that the study was performed prior to widespread vaccination availability in patients who frequently visited healthcare facilities. 20ne limitation of this study is the small number of non-ESA users included in the study; however, this may be expected given the patient population, as the use of ESAs is recommended in patients on dialysis to avoid Hb <9 g/dL. 5Our study did not capture perceptions of patients or the effect on staff workload; as such our measure of feasibility was restricted to the ability to maintain Hb levels and the incidence of AEs.Furthermore, the single-arm, unblinded design of this study limits any direct comparisons between roxadustat and ESAs in this patient population.][12] In conclusion, this study demonstrated that patients with anemia of CKD in large communitybased dialysis organizations could be effectively treated with roxadustat.

ACKNOWLEDGMENTS
The authors thank the patients, their families, and all investigators involved in this study.Medical writing support was provided by Jessica Gorrill, MSc, and editorial support was provided by Sharmin Saleque, MSc, both of Core Medica, London, UK, supported by AstraZeneca according to Good Publication Practice guidelines (https://www.acpjournals.org/doi/10.7326/M22-1460).The sponsor was involved in the study design, collection, analysis, interpretation of data, and data checking of information provided in the manuscript.However, ultimate responsibility for opinions, conclusions, and data interpretation lies with the authors.Roxadustat is being developed in collaboration between FibroGen, Astellas, and AstraZeneca.

FUNDING INFORMATION
This study was supported by FibroGen Inc. (NCT04484857; grant number not applicable).

CONFLICT OF INTEREST STATEMENT
Yemmie Oluwatosin and Kerry Cooper are former employees of AstraZeneca.Jacqueline Nolen and Lixia Zhu are employees and shareholders of FibroGen Inc. Steven Fishbane has done research and consulting for AstraZeneca, FibroGen, Akebia, and GSK.Mark Vishnepolsky has no conflicts to report.Amy Young is an employee of DaVita Clinical Research.

DATA AVAILABILITY STATEMENT
FibroGen Inc., is committed to data sharing and to furthering medical research and patient care.Based on scientific merit, requests from qualified external researchers for anonymized patient-level and study-level clinical trial data (including redacted clinical study reports) for medicines and indications approved in the United States and Europe will be considered after the respective primary study is accepted for publication.All data provided are anonymized to respect the privacy of patients who have participated in the trial in line with applicable laws and regulations.

Up to 6 weeks 0
F I G U R E 1 Study design.ESA, erythropoiesis-stimulating agent; Hb, hemoglobin; HD, hemodialysis; TESAE, treatment-emergent serious adverse event; TIW, three times a week.[Color figure can be viewed at wileyonlinelibrary.com] Patient demographics and baseline characteristics.
T A B L E 1Note: Full analysis set unless otherwise specified.Abbreviations: AV, arteriovenous; ESA, erythropoiesis-stimulating agent; Hb, hemoglobin; IQR, interquartile range; SD, standard deviation; TSAT, transferrin saturation.aBaseline Hb was defined as the mean of available central laboratory values prior to first dose of study medication, including the pre-dose values collected on day 1. b Safety population.ROXADUSTAT USE IN DIALYSIS ORGANIZATIONS Mean hemoglobin (g/dL) at each visit during the study up to week 24.Full analysis set.Intermittent missing Hb data were imputed relying on nonmissing data from all patients using the Monte Carlo Markov Chain imputation model, using baseline Hb, and available nonmissing Hb data for each scheduled week.Hb values were censored for 4 weeks if RBC was administered during the treatment period.Baseline Hb was defined as the mean of available central laboratory Hb values prior to first dose of study medication, including the pre-dose Hb value collected on day 1.Hb, hemoglobin; RBC, red blood cell; SD, standard deviation.[Color figure can be viewed at wileyonlinelibrary.com] F I G U R E 4 Iron indices: Change from baseline in (a) reticulocyte Hb content, (b) ferritin, (c) TSAT, and (d) iron over time.Full analysis set.Baselines for reticulocyte count, TSAT, ferritin, and iron were defined as the last available value obtained on or before day 1, prior to the first dose of study medication.CHr, reticulocyte hemoglobin content; Hb, hemoglobin; SD, standard deviation; TSAT, transferrin saturation.[Color figure can be viewed at wileyonlinelibrary.com] T A B L E 2 Overview of treatment-emergent adverse events (safety population).Note: Safety population.An AE (classified by preferred term) that started during the treatment period was considered a TEAE if it was not present prior to the first dose of study medication, or it was present prior to the first dose of study medication but increased in severity during the treatment.The TEAE investigation period was extended up to 28 days after the last roxadustat dose for patients who did not participate in the extension treatment, or cut to week 24/end of treatment visit when patients were continuing into extension treatment.If a patient had more than one TEAE coded to the same preferred term, the patient was counted only once for that preferred term.Similarly, if a patient had more than one TEAE within a system organ class, the patient was counted only once in that system organ class.PEY for each patient = (last dose date À first dose date + 1)/365.25.EAEs reported in ≥2% patients (safety population).