To the Editor: Anemia is a critical complication of chronic kidney disease (CKD), causing high mortality and morbidity.[1] The standard treatment involves erythropoiesis-stimulating agents (ESAs) and iron supplementation. However, concurrent inflammation is always associated with a poor response to ESA therapy, requiring a higher ESA dosage to maintain hemoglobin (Hb) level.[1] Roxadustat, the first oral hypoxia-inducible factor prolyl-hydroxylase inhibitor (HIF-PHI), presents an effective alternative. It physiologically promotes endogenous erythropoietin production, improves iron metabolism, and demonstrates reduced susceptibility to inflammation.[1–3] HIF-PHIs are recommended by various guidelines and consensuses for treating renal anemia; with some suggesting HIF-PHI initiation at lower doses. This prospective, randomized, open-label, multicenter phase 4 study (No. NCT04059913) evaluated the efficacy and safety of a one-step lower dose, compared to the approved standard starting roxadustat dose, in Chinese dialysis patients with anemia, suggesting a potential for optimized dosing regimen. The study consisted of two treatment parts: Part 1 assessed lower vs. standard dose (weeks 1–20); Part 2 evaluated different dosing frequencies (weeks 21–36). Only the results of Part 1 are presented here. The study design was approved by national and institutional regulatory authorities and ethics committees at each study site. All patients provided written informed consent. The study was approved by Ethics Committee of Zhongda Hospital, Southeast University (No. 2019ZDSYLL017-P01). Eligible patients were adults aged 18–75 years with end-stage renal disease (ESRD) receiving dialysis. They were categorized into two groups: ESA-naive (hereafter "naive") group (patients without ESA within four weeks prior to day 1 and with Hb values ≥70–100 g/L); and ESA-treated (hereafter "converted") group (patients with stable ESA for at least six weeks prior to day 1 and with 90 g/L −10 g/L [Supplementary Table 10, https://links.lww.com/CM9/C587].Figure 1: Mean Hb CFB at select visits (FAS) in naïve patients (A) and in converted patients (B). CFB: Change from baseline; ESA: Erythropoiesis-stimulating agent; FAS: Full analysis set; Hb: Hemoglobin; LOCF: Last-observation-carried-forward. The line plot shows the mean ± standard error of Hb.Hb analysis by baseline CRP groups showed that the lower dose arm achieved a similar Hb increase and Hb response [Supplementary Tables 11 and 12, https://links.lww.com/CM9/C587] overall. After treatment, serum ferritin decreased, while total iron-binding capacity (TIBC) and transferrin increased. TSAT level changed to approximately 30% in both dose arms [Supplementary Table 13, https://links.lww.com/CM9/C587]. The AEs observed were typical for the CKD dialysis population and consistent with the known safety profile of roxadustat, with no unexpected safety signal observed. The proportion of patients experiencing ≥1 TEAE with lower compared to standard dose (82.5% [47/57] vs. 92.9% [52/56] in the naive group, 73.8% [76/103 ]vs. 80.4% [82/102] in the converted group) [Supplementary Table 14, https://links.lww.com/CM9/C587]. Overall, the proportion of patients experiencing ≥1 SAE was similar between dose groups: in the naive group, lower dose arm vs. standard dose was 21.1% [12/57] vs. 19.6% [11/56]; in the converted group, 16.5% [17/103] vs. 18.6% [19/102]. SAEs reported in >3 subjects were arteriovenous fistula thrombosis (1.9%, 6/318), arteriovenous fistula site complication (1.6%, 5/318), hypertension (1.6%, 5/318), arteriovenous fistula occlusion (1.3%, 4/318), and end-stage renal disease (1.3%, 4/318) [Supplementary Table 15, https://links.lww.com/CM9/C587]. One patient died prior to week 21 from respiratory failure considered by the investigator to be unlikely related to study drug or study procedure. Most of the commonly reported (≥10%) AEs occurred less frequently in the lower dose group of both naive and converted patients [Supplementary Table 16, https://links.lww.com/CM9/C587]. This study evaluated the efficacy and safety of different roxadustat starting doses for the treatment of anemia in Chinese dialysis patients. Efficacy analyses demonstrated comparable efficacy between the two roxadustat dose regimens. As anticipated, the rate of rise (ROR) of Hb in naive and treated patients during weeks 1–9 was more gradual using the lower dose regimen; thereafter, achieved Hb levels were similar to those with standard dose. The lower starting dose could effectively correct anemia at the recommended rate of 10–20 g/L Hb increase per month for ESA naive patients according to Chinese and international guidelines, while maintaining stable Hb levels for patients converted from ESA. The average weekly dose over 20 weeks was about 40–50 mg lower in the lower dose arm compared to the standard dose arm, indicating potential cost savings for healthcare systems. The smoother and slower Hb increase observed with the lower starting dose may be desirable for many patients. Inflammation plays a role in anemia and can lead to ESA hyporesponse. Despite inflammation affecting about 25% of patients in this study, both the lower and standard doses of roxadustat effectively raised Hb, reduced hepcidin, and improved iron parameters, achieving a steady-state TSAT of about 30%. Notably, both dose groups, irrespective of CRP, showed a gradual decrease in roxadustat dose between weeks 0–4 and weeks 17–20. The AEs observed were typical for the CKD dialysis population and generally consistent with the known safety profile of roxadustat. Both naive patients and converted patients had similar safety profiles for roxadustat across the two dose arms. Some limitations of this study include the absence of a formal predefined non-inferiority hypothesis test, a relatively small sample size, short treatment duration, and cautious interpretation of ad hoc biomarker analyses. In conclusion, a one-step lower starting roxadustat dose demonstrated a similar efficacy and safety profile for the treatment of anemia as the approved standard starting dose in Chinese CKD patients undergoing dialysis while potentially reducing drug exposure. Acknowledgment Mengmeng Qiao [FibroGen (China) Medical Technology Development Co., Ltd.] provided technical assistance. The authors thank the participating patients and their families, clinicians, and the investigators from the other 21 sites. Funding This study was sponsored by FibroGen (China) Medical Technology Development Co., Ltd. Conflicts of interest None.