Introduction
Anemia is a common and clinically significant complication of chronic kidney disease (CKD), particularly in its advanced stages [1–3]. It primarily results from reduced endogenous erythropoietin synthesis due to progressive nephron loss and is associated with increased cardiovascular morbidity, accelerated renal function decline, and impaired health-related quality of life [2,4,5]. Therapeutic correction of anemia using erythropoiesis-stimulating agents (ESAs) has been a standard management approach for patients with non-dialysis CKD (ND-CKD). Darbepoetin alfa (DARB) is a commonly used long-acting ESA due to its extended half-life and subcutaneous bioavailability [6–10].
Although the benefits of ESA therapy are well established, previous large-scale clinical trials have raised concerns regarding the safety of high-dose ESA use and targeting elevated hemoglobin (Hb) [2,11,12]. Elevated Hb levels exceeding 13 g/dL have been associated with increased risks of stroke, vascular access thrombosis, and cardiovascular events [11–15]. In response, recent clinical guidelines, including those from Kidney Disease: Improving Global Outcomes (KDIGO), recommend conservative Hb targets (<11.5 g/dL) and individualized treatment plans that minimize ESA exposure while achieving clinically meaningful anemia correction [2,11,16].
Given the evolving recommendations, extended-interval ESA dosing has garnered interest in mitigating risks associated with Hb overcorrection and improving treatment adherence. The pharmacokinetic profile of DARB, with a longer serum half-life than that of short-acting ESAs, renders it amenable to dosing intervals of up to 4 weeks [9,17]. Although biweekly (Q2W) administration remains the conventional standard, emerging evidence suggests that once-every-4-week (Q4W) dosing may be clinically feasible and sufficient to maintain target Hb levels in selected patient populations [8,9,18,19].
Moreover, in national healthcare systems where ESA use is regulated by reimbursement policies, treatment strategies must align with cost-efficiency and policy compliance [20]. Due to its lower acquisition cost, DARB is often favored over continuous erythropoietin receptor activators (CERAs). However, concerns regarding its shorter half-life remain. If Q4W DARB administration proves as effective and safe as more frequent dosing, it may offer a cost-effective and operationally efficient alternative for anemia management in patients with ND-CKD [21]. Previous studies have evaluated monthly DARB regimens in both dialysis and ND-CKD cohorts [9,17]. However, many of these studies were largely observational, lacked standardized statistical comparisons, and were conducted predominantly outside Asia.
Accordingly, we conducted a prospective, multicenter, phase 4 study to ascertain whether patients with ND-CKD who had achieved stable Hb levels through biweekly DARB administration could be safely and effectively transitioned to a once-monthly regimen. In addition to assessing non-inferiority in anemia control, the effects of extended-interval dosing on iron metabolism, dosing stability, and treatment-related adverse events were characterized. By reflecting real-world clinical practice and policy constraints, the results can guide individualized anemia management and inform reimbursement strategies for extended-interval ESA therapy in South Korea.
Methods
Participants and study design
This prospective, open-label, multicenter, single-arm phase 4 clinical trial was conducted in South Korea to evaluate the efficacy and safety of once-monthly DARB administration in patients with ND-CKD and anemia. Patient enrollment was performed at three centers: Yongin Severance Hospital, Yongin; Wonju College of Medicine, Wonju; and Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul. The study included a 12-week Q2W dosing run-in period followed by a 12-week Q4W dosing evaluation period (Fig. 1).
The study protocol was approved by the Institutional Review Board of the Gangnam Severance Hospital, Yonsei University College of Medicine (No. 3-2019-0358). It was conducted per the guidelines of the Helsinki Declaration of 1975, as revised in 2000. The study was registered at ClinicalTrials.gov (NCT07025460).
During the run-in period, the patients received subcutaneous DARB every 2 weeks. Patients whose Hb levels were maintained above 9.5 g/dL and who demonstrated stable dosing with <25% variation during this period were enrolled in the evaluation period. The inclusion criteria were as follows: age ≥19 years, estimated glomerular filtration rate ≤45 mL/min/1.73 m2, mean Hb concentration ≤11.5 g/dL, serum ferritin level ≥100 µg/L, transferrin saturation (TSAT) >20%, and normal vitamin B12 and folate levels. The exclusion criteria included: anemia from causes other than CKD; history of or anticipated kidney transplantation; uncontrolled hypertension; recent cardiovascular events, such as myocardial infarction or hospitalization for heart failure within the previous 12 weeks; hematological disorders or active infections, malignancies, recent major surgery, red blood cell transfusion within 8 weeks before enrollment; and the use of other investigational drugs within 30 days.
During the evaluation period, participants received DARB Q4W via subcutaneous injection, with the initial dose based on the total amount administered during the previous month. Dose adjustments were based on Hb levels: the dose was increased by 25% if Hb was <10.0 g/dL, maintained if Hb was between 10.0 and 11.0 g/dL, decreased by 25% if Hb was between 11.1 and 12.0 g/dL, and withheld if Hb was >12.0 g/dL, in which case treatment was resumed at a reduced dose once Hb levels declined to ≤12.0 g/dL. Iron supplementation was managed individually by each trial center to maintain ferritin ≥100 μg/L and TSAT >20%, consistent with protocol guidelines but without a unified regimen.
Laboratory assessments were performed at baseline and weeks 13, 17, 21, and 25. During these visits, approximately 20 mL of blood was collected to evaluate complete blood count, serum iron, TSAT, and ferritin levels. Initially, 77 participants were estimated for enrollment as registered on ClinicalTrials.gov. However, only 65 subjects entered Q2W DARB run-in period, and 40 met predefined criteria to proceed to Q4W evaluation and final analysis. A total of 65 patients were screened across all sites during the enrollment period. Among them, 21 patients did not proceed to the evaluation period due to screening failure, including withdrawal of informed consent, ineligibility based on Hb levels (<9.5 or >11.5 g/dL), and >25% variation in DARB dosing during the run-in period. Four of the 44 patients who entered the evaluation period were excluded due to follow-up loss or consent withdrawal. Thus, 40 patients successfully completed the study per the protocol and were included in the final analysis. To ensure transparency, the study description and enrollment information on ClinicalTrials.gov were updated to reflect the final study design and analyzed population.
Clinical outcomes
Primary and secondary clinical outcomes were defined explicitly to ensure clarity and transparency in accordance with the study objectives. The primary outcome was the absolute change in Hb levels during Q2W and Q4W dosing periods, assessed using a non-inferiority test with a predefined margin. Secondary outcomes included response frequency regarding Hb levels within, above, or below target ranges, as well as longitudinal changes in weekly DARB dose and iron status markers. The target Hb range was defined as >10.0 g/dL and ≤11.0 g/dL.
Statistical analysis
Baseline demographic and clinical characteristics are presented as mean ± standard deviation for continuous variables and frequencies (percentages) for categorical variables. The final analysis included only patients who completed the study protocol through Week 25 without major protocol violations (per-protocol population).
For the primary outcome, the absolute change in Hb during the Q2W (Weeks 1 and 2) and Q4W (Weeks 13–25) periods was compared using a non-inferiority test. Non-inferiority was determined if the upper limit of the 95% confidence interval (CI) for the mean difference in Hb change was less than the predefined 0.2 g/dL margin, based on previous literature and clinical criteria.
The McNemar test was used to compare response frequencies (within, below, and above the target Hb range) at Visits 13 and 25.
A linear mixed model was employed to evaluate longitudinal changes in Hb levels and weekly DARB doses across the five visits. Mean profile plots were constructed to visualize these trends, and post-hoc analyses were conducted to examine pairwise visit differences.
Generalized estimating equations (GEEs) were used to analyze time-related changes in the proportion of patients within outcome categories across the five visits. The outcomes included: 1) Hb within the target range (<10.0 to ≤11.0 g/dL) or 2) Hb exceeding the target range (>11.0 g/dL). If the overall time effect was statistically significant (p < 0.05), post-hoc comparisons were performed using Week 13 as the reference, with Bonferroni correction applied to adjust for multiple comparisons.
All statistical analyses were performed using SAS version 9.4 (SAS Institute).
Results
Baseline characteristics
A total of 40 patients with ND-CKD adhered to the inclusion criteria and were enrolled in the study. The mean age was 71.1 ± 11.4 years, and 55.3% were male. The prevalence of diabetes mellitus was 45.0%, hypertension 92.5%, coronary artery disease 32.5%, and diabetic retinopathy 7.5%. None of the patients had a history of stroke, and 77.5% were non-smokers. No adverse events or serious adverse events were reported during the study period. Blood pressure and other safety parameters remained stable throughout DARB treatment. Detailed demographic and clinical characteristics are summarized in Supplementary Table 1 (available online).
Change in hemoglobin and erythropoiesis-stimulating agent dose over time
A linear mixed-effects model demonstrated significant time-dependent changes in Hb and DARB dose (Table 1, Fig. 2). During the Q2W run-in period, the mean Hb increased from 10.0 ± 0.8 g/dL at Week 1 to 10.8 ± 0.7 g/dL at Week 13 (p < 0.001). After switching to Q4W dosing, Hb progressively declined to 10.6 ± 0.8 g/dL at Week 17 (p = 0.09 vs. Week 13), 10.4 ± 0.7 g/dL at Week 21 (p = 0.002), and 10.2 ± 0.8 g/dL at Week 25 (p < 0.001), yielding an overall time effect of p < 0.001. Nevertheless, the proportion of patients with Hb ≥10.0 g/dL remained high throughout the Q4W evaluation period—87.5% at Week 17, 82.5% at Week 21, and 75.0% at Week 25—with 10–11.0 g/dL peaking at Week 21 (72.5%) (Supplementary Table 2, available online).
The weekly DARB dose decreased from 16.4 ± 9.5 µg at Week 1 to 11.4 ± 3.7 µg at Week 13 (p < 0.001) and then increased during the Q4W evaluation period to 11.5 ± 4.3 µg at Week 17, 12.4 ± 6.7 µg at Week 21, and 13.1 ± 7.1 µg at Week 25; the overall time effect was significant (p = 0.003). Meanwhile, the difference between Week 13 and Week 25 did not reach significance (p = 0.06). Post-hoc pairwise comparisons confirmed significant within-period changes for Hb and DARB during the Q2W run-in period, while in the Q4W evaluation period, only Hb differed significantly across visits (Supplementary Table 3, available online).
Non-inferiority analysis
The absolute change in Hb values between the start and end of each treatment period was 0.867 ± 0.735 g/dL for Q2W and 0.728 ± 0.619 g/dL for Q4W. The mean difference between groups was –0.140 g/dL (95% CI, –0.434 to 0.154), which falls below the predefined non-inferiority margin of 0.2 g/dL. This confirmed the non-inferiority of Q4W administration (p = 0.03) (Fig. 3).
To address concerns regarding sample size adequacy (n = 40), a post-hoc power analysis was conducted. Based on the observed variability in Hb changes and the predefined non-inferiority margin of 0.2 g/dL, the sample size provided 87.2% statistical power to detect non-inferiority. These findings support the robustness of the non-inferiority conclusion for Q4W DARB administration compared to Q2W dosing.
Response frequency
Table 2 summarizes the response frequencies and rates for the Hb range during the Q2W run-in period versus the Q4W evaluation period. During the Q2W period, 57.5% of patients maintained Hb levels within the target range of 10–11.0 g/dL, whereas 35.0% exceeded 11.0 g/dL, and 7.5% remained below 10.0 g/dL. In the Q4W period, 50.0% of patients were within the target range, 15.0% exceeded it, and 35.0% were below 10.0 g/dL. The difference in the proportion of patients exceeding 11.0 g/dL was statistically significant (p = 0.02), whereas no statistically significant differences were observed in the proportion of patients within the target range (p = 0.51).
When applying Bonferroni correction for multiple comparisons, the initially significant difference in the proportion of patients with Hb below 10 g/dL between Q4W and Q2W groups (p = 0.03) became nonsignificant (p = 0.10). This suggests that while numerically higher, the increase in sub-target Hb levels during the Q4W period does not reach statistical significance after correction. Importantly, this increase is balanced by a significant reduction in Hb overshoot in the Q4W group, indicating effective and safe anemia control. Thus, these data support the clinical viability of the Q4W dosing regimen despite the observed fluctuations in Hb response rates (Supplementary Table 4, available online).
Generalized estimating equation analysis
The GEE model was applied to evaluate the pattern of response categories. The proportion of patients within the target Hb range (10.0–11.0 g/dL) increased from 47.5% at Week 1 to 72.5% at Week 21 before declining to 50.0% at Week 25. This trend did not reach statistical significance (time p-value = 0.22) (Table 3).
In contrast, the proportion of patients exceeding the upper Hb threshold (>11.0 g/dL) showed a significant time-dependent decrease (p = 0.001), peaking at Week 13 (35.0%) and falling to 10.0% at Week 21. Post-hoc analysis indicated that Week 1 significantly differed from Week 13 (p = 0.003), which differed from Visits 4 (p = 0.003) and 5 (p = 0.02). These findings suggest a reduction in overshooting Hb levels after transitioning to Q4W (Table 3).
Discussion
This prospective, multicenter phase 4 study focused on a real-world ND-CKD cohort stabilized on Q2W dosing to determine whether Q4W dosing could maintain Hb within clinically acceptable ranges without increasing Hb overshoot risk. The study confirmed non-inferiority in Hb control, reduced Hb overshoot, and stability in dosing and iron parameters over time. Conducted within the framework of national reimbursement policies, this trial provides clinically actionable evidence supporting extended-interval ESA use in real-world outpatient settings.
Anemia is a prevalent and clinically significant complication of CKD, particularly in patients with advanced renal dysfunction [1,4,22–24]. Effective anemia management in CKD aims to correct Hb levels and avoid risks associated with overcorrection, including cardiovascular complications and thromboembolic events [1,2,10–13,15,25,26]. ESAs, such as DARB, have been widely used to raise Hb levels and reduce transfusion requirements [6–9,17-19,21]. However, evidence from large-scale clinical trials and meta-analyses underscores the importance of balancing anemia correction with safety, particularly in light of adverse outcomes associated with high ESA doses and excessive Hb targets [1,2,11,13,15].
The KDIGO guidelines recommend a conservative approach to ESA use, favoring the lowest possible ESA dose to maintain Hb above 10 g/dL while discouraging routine escalation beyond 11.5 g/day [2,27–34]. These recommendations represent a paradigm shift toward minimizing ESA exposure while preserving clinical benefits. Extended-interval ESA administration is gaining attention to mitigate Hb overshoot risk, optimize resource utilization, and enhance patient convenience. DARB, owing to its prolonged serum half-life, is particularly amenable to these regimens.
The findings from this study demonstrated that Q4W DARB administration was non-inferior to Q2W regarding Hb stability. These results are consistent with those of previous studies, including the EXTEND trial [35–38], which reported long-term Hb stability in patients maintained on Q2W or Q4W DARB. In the current study, over 75% of patients maintained Hb ≥10 g/dL during the Q4W period, indicating effective anemia control.
A notable finding was the marked reduction in Hb overshoot during Q4W therapy compared to Q2W dosing. This finding holds critical clinical and policy-making relevance in Korea, where ESA dosing is subject to strict regulatory oversight by the Health Insurance Review and Assessment Service [20]. Under current guidelines, ESA therapy must be dose-reduced or suspended if Hb exceeds 11.0 g/dL, making overshoot avoidance a practical necessity [2,29,30,34]. The findings of this study indicate that Q4W administration could serve as an effective risk mitigation strategy in routine clinical practice while maintaining treatment efficacy. However, this reduction in Hb overshoot was accompanied by an increase in the proportion of patients with Hb levels below 10.0 g/dL during the Q4W dosing period, suggesting a potential risk of under-treatment. Despite this, a majority of patients maintained Hb within the target range throughout the study. These findings highlight the importance of individualized dosing and careful monitoring when implementing extended-interval dosing strategies. Future studies should aim to optimize dosing regimens to balance the risks of both overshoot and under correction, ensuring safe and effective anemia management in patients with ND-CKD.
GEE application confirmed significant time-dependent changes in response patterns. Specifically, a meaningful temporal reduction was observed in Hb overshoot (p = 0.001), while post-hoc pairwise comparisons identified statistically significant differences between Week 13 and subsequent visits, particularly Weeks 21 and 25. These results reinforce the temporal stability and safety of Q4W DARB, supporting its appropriateness for maintenance therapy in patients with ND-CKD.
In addition to Hb trends, ESA dose adjustments and iron metabolism were assessed. Weekly DARB doses remained stable after the initial adjustment required for Q4W conversion, with no evidence of increasing requirements over time. The modest increase in weekly DARB dose observed during the Q4W evaluation period likely reflects the transition from the run-in period during the Q2W period—when Hb was actively stabilized and ESA doses titrated—to a maintenance phase during the Q4W dosing, where dosing requirements increased modestly to sustain target Hb levels. This increase should not be interpreted as regimen failure or inefficiency, but as a natural dosing adjustment from correction to maintenance. This suggests that extended-interval DARB can maintain erythropoietic activity without dose intensification once anemia correction is achieved. Furthermore, iron indices, including serum ferritin, TSAT, and total iron-binding capacity, remained within acceptable ranges throughout the study, aligning with KDIGO recommendations emphasizing iron sufficiency and avoiding functional iron deficiency [1,2,10].
The current study’s findings underscore the practical advantages of Q4W regimens in real-world outpatient settings. Reduced injection frequency can alleviate treatment burden, particularly in older adults or patients with impaired mobility, potentially improving adherence and quality of life. From a healthcare system perspective, Q4W ESA dosing can improve clinical workflow efficiency, reduce nursing workload, and support resource-optimized anemia management. These advantages have been echoed in previous studies, including the AFFIRM study [39] and long-term observational cohorts such as the EXTEND study [35,38].
This study provides comparative insights into other ESA formulations and strategies. Previous studies have evaluated the efficacy of long-acting ESAs such as CERA and switching protocols between agents; however, direct comparisons remained limited [21,39]. This study focused exclusively on a structured transition from Q2W to Q4W DARB within a single, stable population under national reimbursement policy constraints. This approach improves internal validity and supports its translational relevance. Moreover, to ensure a broad and clinically relevant study population, we set the Hb inclusion cutoff at <11.5 g/dL, encompassing mild to moderate anemia consistent with 2025 KDIGO guidelines recommending individualized ESA initiation. Although the Korean National Health Insurance Service reimbursement criteria set thresholds at Hb <10 g/dL or Hct <33%, clinical practice allows flexibility to manage patients slightly above these limits to prevent anemia progression. Our cohort included ESA-naïve and ESA-experienced patients, enhancing real-world applicability. Observed dose increases after switching to Q4W dosing reflect individualized titration to maintain targets rather than regimen inefficiency. However, this study has some limitations. While our use of GEE allowed for adjustment of temporal trends and accounted for repeated measurements, it cannot fully eliminate potential confounding arising from the lack of a concurrent control group, which is an inherent limitation of single-arm trials that reduces internal validity and increases susceptibility to temporal bias. Although rigorous multiple comparison corrections were applied to mitigate statistical concerns, residual confounding remains possible, warranting caution in causal interpretation. Furthermore, the discrepancy between the initially registered estimated enrollment and the analyzed sample size was due to real-world feasibility challenges and strict eligibility criteria, resulting in a reduced sample size which, along with the single-arm, non-randomized design, limits generalizability. Therefore, our results should be interpreted as supportive, hypothesis-generating evidence specific to ESA-stable patients with ND-CKD rather than definitive proof of equivalence between dosing regimens. Additionally, the relatively short evaluation period (12 weeks) is insufficient to capture long-term clinical outcomes, such as cardiovascular events, hospitalization rates, or progression to dialysis. Finally, the study enrolled predominantly ESA-responsive patients, which may limit generalizability to individuals with inflammatory states, high comorbidity burden, or ESA hyporesponsiveness [1,2,40].
Despite these limitations, this study offers clinically meaningful evidence supporting Q4W DARB as a viable alternative for stable patients with ND-CKD. The findings align with modern therapeutic principles emphasizing individualized care, safety, and cost-effectiveness. In Korea, where clinical decisions are closely aligned with reimbursement policy, extended-interval ESA administration strategies such as Q4W DARB may facilitate better adherence, fewer dose interruptions, and improved anemia management.
In conclusion, transitioning from Q2W to Q4W DARB administration is a feasible and safe approach for maintaining Hb levels in patients with ND-CKD who are stable on ESA therapy. The regimen effectively mitigates the risk of Hb overshoot, supports dose stability, and aligns with contemporary guidelines. These findings support the adoption of Q4W ESA therapy as a practical, patient-centered strategy for anemia management in ND-CKD. Future randomized controlled studies with longer follow-up periods and more diverse patient populations are warranted to confirm these results and expand their applicability across broader clinical contexts.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement table 1
Print
