No increased ophthalmic risk with roxadustat treatment in DD CKD-related anaemia

Roxadustat treatment for anaemia in patients with dialysis-dependent chronic kidney disease (DD CKD) does not come with an increased risk of ophthalmic abnormalities, according to a phase III trial conducted in Japan.

Adults (age ≥20 years) with CKD and anaemia who had been undergoing stable haemodialysis thrice weekly (TIW) for >12 weeks at 58 sites in Japan were randomized 1:1 to receive oral roxadustat TIW (n=150; mean age 64.6 years, 67.3 percent male) or intravenous injections of darbepoetin alfa once weekly (QW; n=152; mean age 64.9 years, 71.1 percent male) for up to 24 weeks. The doses of the medications were titrated to maintain haemoglobin levels of 10–12 g/dL.

Fifty-four percent of patients in each group had diabetes mellitus (DM). Ninety-two and 93.4 percent of roxadustat and darbepoetin alfa recipients, respectively, had a history of hypertension. Patients in the roxadustat and darbepoetin alfa groups were on haemodialysis for a mean 59.45 and 56.13 months, respectively, and had anaemia for a mean 68.37 and 58.71 months, respectively. Compared with those on darbepoetin alfa, patients in the roxadustat group had a non-significantly higher rate of previous or concurrent retinal vascular disorder* (41.3 percent vs 37.5 percent) and history of DM and retinal vascular disease (32.7 percent vs 27.6 percent).

Overall, new or worsening retinal haemorrhage occurred at a comparable rate between roxadustat and darbepoetin alfa recipients (32.4 percent vs 36.6 percent). This finding was consistent at week 12 (23.5 percent vs 25.7 percent), week 24 (26.5 percent vs 27.0 percent), and at treatment end (23.9 percent vs 29.0 percent). [ERA-EDTA 2020, abstract MO002]

“There was no clinically meaningful change in the number of retinal haemorrhages in either group [mean change from baseline to end of treatment, 0.2 and 0.3 in the roxadustat and darbepoetin alfa groups, respectively],” said Dr Yasir Sepah from the Nguyen Eye Lab, Stanford School of Medicine, Stanford, California, US, who presented the results at ERA-EDTA 2020.

During treatment, new or worsening retinal haemorrhages occurred in 19.1 and 25.0 percent of roxadustat and darbepoetin alfa recipients, respectively, who had no retinal haemorrhage at baseline, and in 58.3 and 59.2 percent, respectively, who had ≥1 retinal haemorrhage at baseline.

Among patients without DM at baseline, new or worsening retinal haemorrhages occurred in 21.1 and 29.5 percent of roxadustat and darbepoetin alfa recipients, respectively, during treatment, and in 13.3 and 23.2 percent at treatment end. Among patients with DM, new or worsening retinal haemorrhages occurred at a comparable rate between roxadustat and darbepoetin alfa recipients (51.9 percent vs 50.0 percent [during treatment] and 42.3 percent vs 40.0 percent [treatment end]). 

According to Sepah, non-clinical data has suggested that angiogenesis may be associated with an increased risk of certain retinal pathologies. “Angiogenesis may be mediated via vascular endothelial growth factor, which can be induced by HIF-1α,” he said.

“During the 24-week treatment period, these data suggest that DD CKD patients who were treated with [the HIF-PHI**] roxadustat were not at an increased risk of ophthalmic abnormalities, compared with patients treated with darbepoetin alfa,” he concluded.