FC 071ENHANCED DEGRADATION OF 3-HYDROXY-3-METHYLGLUTARYL COENZYME A REDUCTASE (HMGCR) MAY CONTRIBUTE TO THE LOWERING OF LDL CHOLESTEROL SEEN WITH ROXADUSTAT IN CHRONIC KIDNEY DISEASE PATIENTS WITH ANEMIA

Background and Aims Roxadustat is an oral hypoxia-inducible factor prolyl hydroxylase (HIF-PH) inhibitor approved in China and Japan for the treatment of anemia in patients with chronic kidney disease (CKD). In Phase 3 clinical studies, roxadustat was shown to increase hemoglobin levels in both dialysis-dependent (DD) and non-dialysis-dependent (NDD) CKD patients with anemia. This increase in hemoglobin and the accompanying decrease in hepcidin, a hormone that sequesters iron in intracellular stores, is consistent with the known role that HIF plays in the regulation of erythropoiesis. The HIF pathway may also influence cholesterol metabolism; at high altitude, total and low-density lipoprotein cholesterol (LDL-C) decrease in healthy individuals. The cholesterol biosynthesis pathway is well characterized and 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) has long been recognized as the rate-limiting enzyme. Multiple feedback mechanisms are known to regulate HMGCR activity, one of which involves degradation of HMGCR via interaction with insulin-induced genes 1 and 2 (INSIG1 and INSIG2), two closely related endoplasmic reticulum membrane proteins. Here, we report clinical trial data showing a reduction in LDL-C in patients with anemia of CKD who were treated with roxadustat, and describe the results of investigation into the mechanism of this cholesterol-lowering effect. Method We studied the effect of roxadustat on LDL-C levels in human subjects and examined the potential underlying mechanisms via cell culture experiments. Clinical data were pooled from six pivotal phase 3 studies, three in patients with NDD-CKD and three in patients with DD-CKD, including those with incident dialysis (ID; on dialysis <4 months at randomization). Mean changes from baseline (CFB) in LDL-C (regardless of statin use) averaged over weeks 12–28 were analyzed using a mixed ANCOVA model of repeated measures. We investigated the effect of roxadustat on HMGCR activity in a cell-free enzyme assay and on HMGCR protein levels in cultured Hep3B cells. We explored involvement of the HIF pathway in regulation of HMGCR protein levels in Hep3B cells by means of HIF-1α/HIF-2α small interfering (si)RNA knockdown. Protein expression levels were assessed by SDS-PAGE and Western blot analysis or electrochemiluminescent immunoassays. Gene expression levels were evaluated by real-time quantitative reverse-transcriptase polymerase chain reaction (RT-PCR). Results In patients with NDD-CKD, there was a 17.2% reduction in LDL-C averaged over weeks 12–28 in the roxadustat group (n=1994) vs. a 1.4% increase in the placebo group (n=1430) (least-squares mean [LSM] [SE] treatment difference = -19.83 mg/dL [1.186], p<0.0001) (Table). In patients with DD-CKD, LDL-C was reduced by 18.5% in the roxadustat group (n=1650) vs. 1.7% in the epoetin alfa group (n=1741) (LSM treatment difference = -15.80 mg/dL, p<0.0001) (Table). In the ID-DD patients, LDL-C dropped by 21.5% in the roxadustat group (n=680) vs. 4.6% in the epoetin alfa group (n=691) (LSM treatment difference = -17.50 mg/dL, p<0.0001) (Table). After confirming that roxadustat does not directly inhibit HMGCR enzyme activity, we found that the increase in HMGCR protein levels caused by cholesterol depletion in cells was suppressed by roxadustat. However, roxadustat had no effect on HMGCR mRNA expression, indicating that the suppression of HMGCR protein levels by roxadustat was the result of an effect on protein turnover. Subsequent siRNA studies revealed that the regulation of HMGCR protein levels by roxadustat was HIF dependent and mediated via upregulation of INSIG2. Conclusion These data show that treatment with roxadustat vs. placebo or epoetin alfa lowered LDL-C in patients with NDD-CKD and DD-CKD, respectively, and highlight a potential HIF-dependent mechanism for this cholesterol-lowering effect.