Efficacy and safety of hypoxia-inducible factor prolyl hydroxylase inhibitors in patients with chronic kidney disease: meta-analysis of phase 3 randomized controlled trials

Abstract

Background: Hypoxia-inducible factor prolyl hydroxylase inhibitors (HIF-PHIs) are new therapeutic agents for anaemia in chronic kidney disease (CKD). We evaluated by meta-analysis and meta-regression the efficacy and safety of HIF-PHIs in patients with CKD-related anaemia.

Methods: We selected phase 3 randomized clinical trials (RCTs) comparing HIF-PHIs and erythropoiesis-stimulating agents (ESAs) in dialysis and non-dialysis patients. Efficacy outcomes were the changes from baseline of haemoglobin, iron parameters (hepcidin, serum iron, TIBC, TSAT, ferritin) and intravenous iron dose; as safety outcomes we considered cancer, adjudicated major adverse cardiovascular events (MACE), MACE+ (MACE plus hospitalization for hearth failure or unstable angina or thromboembolic event), thrombotic events (deep vein thrombosis, pulmonary embolism), arterovenous fistula (AVF) thrombosis and death.

Results: We included 26 RCTs with 24 387 patients. Random effect meta-analysis of the unstandardized mean difference between HIF-PHIs and ESAs showed a significant change in haemoglobin levels from baseline of 0.10 g/dL (95% CI 0.02 to 0.17). Meta-regression analysis showed a significantly higher haemoglobin change for HIF-PHIs in younger patients and versus short-acting ESA (0.21 g/dL, 95% CI 0.12 to 0.29 versus -0.01, 95% CI -0.09 to 0.07 in studies using long-acting ESA, P < .001). No significant effect on heterogeneity was found for type of HIF-PHIs. In comparison with ESAs, HIF-PHIs induced a significant decline in hepcidin and ferritin and a significant increase in serum iron and TIBC, while TSAT did not change; intravenous iron dose was lower with HIF-PHI (-3.1 mg/week, 95% CI -5.6 to -0.6, P = .020). Rate ratio of cancer (0.93, 95% CI 0.76 to 1.13), MACE (1.00, 95% CI 0.94 to 1.07), MACE+ (1.01, 95% CI 0.95 to 1.06), thrombotic events (1.08, 95% CI 0.84 to 1.38), AVF thrombosis (1.02, 95% CI 0.93 to 1.13) and death (1.02, 95% CI 0.95 to 1.13) did not differ between HIF-PHIs and ESAs.

Conclusions: HIF-PHIs at the doses selected for the comparisons are effective in correcting anaemia in comparison with ESA therapy with a significant impact on iron metabolism without notable difference among various agents. No safety signals emerge with use of HIF-PHIs.

Keywords: CKD; HIF-PHI; anemia; haemoglobin; meta-analysis.

Graphical Abstract

Figure 1:

Flow chart of selected studies.

Figure 2:

Random effect meta-analysis of unstandardized mean difference in change of Hb levels from baseline (A) and Hb target achievement (B) between HIF-PHI and ESA comparator.

Figure 3:

Meta-regression of effects of baseline age on change in Hb levels from baseline between HIF-PHI and ESA comparator.

Figure 4:

Random effect meta-analysis of unstandardized mean difference from baseline between HIF-PHI and ESA comparator in change of serum hepcidin (A), serum iron (B), TIBC (C), TSAT (D), serum ferritin (E) and intravenous iron dose (F).

Figure 5:

Random effect meta-analysis of rate ratio for cancer (A), MACE (B), MACE+ (C), deep venous thrombosis/pulmonary embolism (D), arteriovenous thrombosis (E) and all-cause death (F) between HIF stabilizers and ESA comparator.