Clinical effect of roxadustat vs. erythropoietin in non-dialysis CKD with diabetes: a single center propensity score matching analysis.

Abstract

Renal anemia is a common complication of chronic kidney disease. Currently, recombinant human erythropoietin and roxadustat are the main treatments. In China, diabetic kidney disease is the primary cause of chronic kidney disease. However, high-quality evidence on the efficacy of roxadustat in patients with non-dialysis-dependent chronic kidney disease and diabetes mellitus is scarce. This study aimed to assess the clinical effect of roxadustat in such patients. Patients with non-dialysis-dependent anemia and diabetes mellitus who received roxadustat or recombinant human erythropoietin for ≥ 4weeks were enrolled. We compared baseline characteristics, including age, gender, hypertension, and hemoglobin level, and then employed a 1:3 ratio propensity score matching. The primary efficacy outcomes were changes in hemoglobin levels. After propensity score matching, 212 patients were analyzed, including the roxadustat (n = 53) and recombinant human erythropoietin (n = 159) groups. Baseline characteristics were comparable, including hemoglobin level, estimated glomerular filtration rate, and glycated hemoglobin A1c (p > 0.05). After 4, 12, and 24weeks of treatment, the median hemoglobin levels in the roxadustat group were 97.5g/L, 104g/L, and 106.5g/L, respectively, significantly surpassing the corresponding levels in the recombinant human erythropoietin group at 91g/L, 94.5g/L, and 94.5g/L (p = 0.002, p = 0.025, p = 0.006, respectively). Additionally, subgroup analysis demonstrated better treatment efficacy of roxadustat patients with elevated high-sensitivity C-reactive protein and low albumin levels. In Chinese patients with anemia and diabetes not on dialysis, roxadustat efficiently and rapidly improved and maintained hemoglobin levels unaffected by elevated high-sensitivity C-reactive protein and low albumin levels.