Pooled Efficacy and Cardiovascular Safety Results of Roxadustat Compared with Epoetin Alfa in the Treatment of Anemia in Chronic Kidney Disease Patients on Dialysis

Introduction

Anemia affects > 90% of patients with chronic kidney disease (CKD) who require dialysis (Stauffer ME, Fan T. PLoS ONE 2014;9(1):e84943; Goodkin DA, et al. JASN 2011;22:358-365). The current standard of care for anemia of CKD is erythropoiesis-stimulating agents (ESAs) with or without IV iron supplementation and red blood cell (RBC) transfusion. In 2011, the FDA issued a warning that ESAs are associated with high risk of death and cardiovascular (CV) events (all-cause mortality, myocardial infarction, and stroke), especially at high doses. Hyporesponsiveness to ESAs is common due to CKD-related inflammation, requiring higher compensatory doses. Due to the increased risk of death and CV events, some patients with anemia of CKD are not receiving optimal doses of ESAs and RBC transfusions have increased.

According to the US Renal Data Systems 2018 report (United States Renal Data System. National Institute of Diabetes and Digestive and Kidney Diseases; 2018.), first-year survival rates among incident dialysis patients (ie, on dialysis for ≤ 4 months) was 78%. Thus, there is an urgent unmet need for better anemia treatments for this vulnerable group of patients.

Roxadustat is an oral hypoxia–inducible factor prolyl hydroxylase inhibitor that regulates erythropoiesis and iron metabolism. This pooled Phase 3 analysis examines efficacy and cardiovascular (CV) endpoints for roxadustat vs epoetin alfa in dialysis patients.

Methods

Three open-label Phase 3 studies (FGCL-4592-063, FGCL-4592-064, and D5740C00002) compared roxadustat with epoetin alfa (EPO); data were pooled for these analyses. Patients requiring dialysis and anemia treatment were randomly assigned to treatment with roxadustat or EPO (1:1). Death, myocardial infarction, and stroke events (MACE), and heart failure or unstable angina requiring hospitalization events (MACE+) were adjudicated. Studies were event-driven. Efficacy endpoints included Hb change from baseline (CFB) and transfusion avoidance. Time to first MACE and MACE+ compared CV safety of roxadustat with epoetin alfa. Analyses were conducted for all dialysis and incident dialysis patients (enrolled within 4 months of beginning dialysis).

Results

Overall, 3,917 patients were randomly assigned to roxadustat (n = 1,960) or EPO (n = 1,957). The average patient age was 54.3 years for roxadustat vs 55.1 years for EPO. Both arms had similar mean baseline Hb levels (9.63 vs 9.67 g/dL). Subjects received peritoneal dialysis (9.3% and 9.8%, roxadustat and EPO groups, respectively) or hemodialysis. The primary endpoint for the US FDA analysis, mean Hb CFB over weeks 28-52, was met for each individual study (both noninferior and superior) and was 1.21 in the roxadustat vs 0.95 g/dL in the EPO arms (difference 0.26 g/dL; 95% CI 0.20, 0.33) in the pooled analysis, also noninferior and superior (p < 0.0001). The roxadustat group received fewer transfusions than the EPO group (9.5% vs 12.8%, respectively; HR = 0.82; 95% CI 0.679, 0.997). Comparing roxadustat with EPO, hazard ratio (HR) for MACE was 0.96 (95% CI 0.82, 1.13) and for MACE+ it was 0.86 (0.74, 0.98) (p = 0.028) among all dialysis patients. Of 1,526 incident patients, the HRs of MACE and MACE+ were lower in the roxadustat compared with EPO arms, 0.70 (95% CI 0.51, 0.96; p = 0.029) and 0.66 (95% CI 0.50, 0.89; p = 0.005).

Conclusions

Roxadustat was noninferior and superior in increasing Hb compared with EPO in 3 Phase 3 trials. It also significantly reduced MACE risk in incident dialysis patients and MACE+ risk in both the all-dialysis patients and incident dialysis patients groups. Additionally, roxadustat reduced RBC transfusion compared with EPO in the all-dialysis pooled analysis.