Anemia is a common problem in advanced chronic kidney disease (CKD) especially in maintenance hemodialysis patients. It is a multifactorial disease and is caused by decreased renal erythropoietin production, decreased erythrocyte survival, chronic inflammation, iron deficiency and repeated dialysis-related bleeding [2].

 Hepcidin also gets activated during inflammation, which promotes iron binding to storage proteins or the reduction of intestinal iron absorption and thus results in a functional iron-deficiency, even if the storage iron is sufficient [11]. ESAs are still important in the treatment of renal anemia, but an important fraction of the patient population will develop ESA hyporesponsiveness. For a large group of hemodialysis patients, about 12.5% were ESA hyporesponsive with lower hemoglobin level, higher ESA dosage and higher mortality compared to patients with ESA response [4]. Therefore, the resistance to ESA is a therapeutic hurdle and is associated with higher risk of clinical treatment failures [5]. Increased dose of ESA alone may not be enough to prevent the development of resistance and can lead to exposure to high doses for an extended period [8]. Oral hypoxia-inducable factor prolyl-hydroxylase inhibitor roxadustat stimulates endogenous erythropoietin (EPO) levels, which induces erythropoiesis, and iron mobilization [3]. Randomized trials have demonstrated that roxadustat is able to keep hemoglobin levels stable or raise it, to an extent similar to that of epoetin alfa, in dialysis-dependent CKD patients; pooled phase III data have been shown to support the efficacy of roxadustat in large dialysis populations [7,15]. A critically important dialysis trial also showed a positive impact on transferrin, the availability of serum iron and hepcidin versus epoetin alfa [6]. Inflammation and functional iron restriction are often seen in conjunction in ESA-resistant anemia, thus these properties may be particularly relevant to this disease. There are meta-analytical data showing roxadustat has beneficial effects on hemoglobin and some measures of iron use in dialysed CKD [10]. The response to roxadustat on hemoglobin is also relatively spared at different levels of baseline inflammation, as indicated by recent pooled analyses [14]. This offers a biological justification for evaluating the effect of low dose roxadustat when anemic despite high exposure to ESAs.

There are some encouraging, but limited, early combination evidence. A study by Xu et al. found that the addition of reduced-dose roxadustat to the current erythropoietin therapy improved hemoglobin levels in ED-HD patients [12]. A multicenter retrospective cohort study also showed a more rapid hemoglobin response when using low dose roxadustat with recombinant human erythropoietin versus use of either one agent alone [9]. Recently, roxadustat has been shown to be beneficial in ESA-hyporesponsive hemodialysis patients, as an alternative to ongoing ESA therapy [13]. The current KDIGO guideline, however, recommends not to routinely use ESAs and HIF-prolyl hydroxylase inhibitors together due to a lack of sufficiently powered randomized safety data [1].

Objective

To evaluate the efficacy and safety of full-dose erythropoietin combined with low-dose roxadustat in hemodialysis patients with erythropoietin-resistant anemia.

This retrospective observational study was conducted at Department of Nephrology HBS Medical and Dental College Islamabad from July 2025 to July 2026 and included 185 adult patients with end-stage kidney disease receiving maintenance hemodialysis who had erythropoietin-resistant anemia. Patients aged ≥18 years who had been receiving maintenance hemodialysis for at least three months, had persistent hemoglobin levels <10 g/dL despite adequate or high-dose erythropoietin therapy for at least four weeks, and had subsequently received low-dose roxadustat (20–50 mg) in addition to ongoing erythropoietin therapy were included. Erythropoietin resistance was defined as failure to achieve an adequate hemoglobin response despite appropriate erythropoietin treatment after correction of reversible causes of anemia. Patients with active bleeding, recent blood transfusion, hematological disorders, severe vitamin B12 or folate deficiency, active infection, severe liver disease, uncontrolled hypertension, recent myocardial infarction or stroke, pregnancy, incomplete records, or inadequate follow-up were excluded. Data Collection After approval from the Institutional Ethical Review Committee, eligible patients were identified from nephrology and hemodialysis unit records using a non-probability consecutive sampling technique. Data regarding age, gender, comorbidities, duration and frequency of hemodialysis, vascular access, erythropoietin dose, roxadustat dose, and relevant laboratory parameters were recorded. Hemoglobin, hematocrit, serum ferritin, transferrin saturation, serum creatinine, albumin, and C-reactive protein were documented at baseline and during follow-up. Hemoglobin levels were assessed at baseline and approximately 2, 4, 8, and 12 weeks after initiation of adjunctive roxadustat. The primary outcome was the change in hemoglobin from baseline to 12 weeks. Treatment response was defined as an increase in hemoglobin of ≥1.0 g/dL and/or achievement of hemoglobin ≥10.0 g/dL without blood transfusion. Adverse events and treatment discontinuation were also recorded. Statistical Analysis Data were analyzed using SPSS version 26.0. Continuous variables were expressed as mean±standard deviation or median with interquartile range, while categorical variables were presented as frequencies and percentages. Normality was assessed using the Shapiro–Wilk test. Baseline and post-treatment hemoglobin levels were compared using the paired-samples t-test or Wilcoxon signed-rank test, as appropriate. Changes in hemoglobin across follow-up intervals were assessed using repeated-measures ANOVA or the Friedman test. Associations between treatment response and clinical or laboratory variables were assessed using the chi-square test, Fisher's exact test, independent-samples t-test, or Mann–Whitney U test, as appropriate. Multivariable binary logistic regression analysis was performed to identify independent predictors of treatment response, and adjusted odds ratios with 95% confidence intervals were reported. A p-value ≤0.05 was considered statistically significant.

The mean age of the 185 patients was 54.8±12.6 years; 112 (60.5%) were male and 73 (39.5%) female. Mean BMI was 25.7±4.3 kg/m² and mean dialysis duration was 4.2±2.3 years. Hypertension was present in 149 (80.5%), diabetes in 82 (44.3%), and cardiovascular disease in 47 (25.4%). Most patients, 153 (82.7%), had an arteriovenous fistula. The mean baseline erythropoietin dose was 12,480±2,960 IU/week; 103 (55.7%) received roxadustat 20–30 mg and 82 (44.3%) received 40–50 mg.

Table 1. Baseline demographic and clinical characteristics of patients (n=185)

Hemoglobin increased from 8.1±0.8 g/dL at baseline to 10.4±1.1 g/dL at week 12 (p<0.001), while hematocrit rose from 25.2±3.1% to 32.3±3.9% (p<0.001). Ferritin decreased from 468±186 to 416±161 ng/mL (p=0.018), whereas transferrin saturation increased from 27.6±7.4% to 31.8±7.8% (p<0.001). CRP declined from 8.9±5.3 to 6.6±4.3 mg/L (p=0.012), and erythropoietin requirement decreased from 12,480±2,960 to 9,420±2,610 IU/week (p<0.001). Albumin remained largely unchanged.

Table 2. Changes in hematological and biochemical parameters following combination therapy

Overall, 134 (72.4%) patients responded to treatment, with 139 (75.1%) achieving a hemoglobin rise ≥1.0 g/dL and 131 (70.8%) reaching hemoglobin ≥10.0 g/dL. Erythropoietin requirements decreased in 96 (51.9%) patients, while 9 (4.9%) required transfusion. Adverse events occurred in 29 (15.7%), most commonly hypertension in 17 (9.2%) and gastrointestinal symptoms in 8 (4.3%); treatment discontinuation occurred in 7 (3.8%).

Table 3. Treatment response and safety outcomes after 12 weeks (n=185)

Lower treatment response was independently associated with age ≥60 years (aOR 0.49, p=0.028), diabetes (aOR 0.46, p=0.019), dialysis duration ≥5 years (aOR 0.52, p=0.047), elevated CRP (aOR 0.37, p=0.006), and albumin <3.5 g/dL (aOR 0.41, p=0.013). Roxadustat 40–50 mg was associated with a greater likelihood of response (aOR 2.74, p=0.008), while male gender, baseline hemoglobin <8 g/dL, and TSAT <30% were not significant.

Table 4. Factors associated with treatment response to erythropoietin plus low-dose roxadustat

In the present study, the use of low-dose roxadustat with background erythropoietin therapy was correlated with significant improvement in anemia in erythropoietin-resistant hemodialysis patients. Mean hemoglobin increased from 8.1±0.8 g/dL at baseline to 10.4±1.1 g/dL after 12 weeks, while hematocrit increased from 25.2±3.1% to 32.3±3.9%. In previous trials of erythropoietin-resistant hemodialysis (ESRD) patients, a combination of low-dose roxadustat and erythropoietin led to a significant increase in hemoglobin from approximately 7.77±1.12 g/dL to 9.20±0.84 g/dL after 12 weeks of treatment [16]. The differences in baseline characteristics, roxadustat dosing and erythropoietin exposure may explain the greater improvement seen in the present study. In total, 134 (72.4%) patients in the present study did reach the predefined treatment response, while 139 (75.1%) patients did improve their hemoglobin level by at least 1.0 g/dL and 131 (70.8%) patients did reach a hemoglobin level ≥10 g/dL. The data from the previous one-and-a-half years of testing of low-dose roxadustat and erythropoietin in renal anemia who have resistant anemia had a similarly high response rate, around 72%. The combination therapy also yielded a quicker hemoglobin response compared to roxadustat or erythropoietin alone, as shown in previous studies of patients with moderate anemia undergoing hemodialysis, where the median response time was around 20 days [17]. The observations confirm the possibility of complementary action of both treatments on erythropoietic activity. There were positive changes in the parameters related to iron status with improvement in anemia. Serum ferritin decreased from 468±186 to 416±161 ng/mL, while transferrin saturation increased from 27.6±7.4% to 31.8±7.8%. The same has been shown in the past: Ferritin levels of 171.05±54.75 ng/mL were reduced to 140.83±42.03 ng/mL with combination therapy [18]. The evidence from meta-analysis has also demonstrated a positive effect on serum iron, transferrin, total iron binding capacity and iron utilisation with roxadustat therapy. These changes could be due to enhanced mobilization and utilization of stored iron for erythropoiesis, not necessarily to iron depletions. Another significant observation was that erythropoietin requirement was decreased from 12,480±2,960 IU/week to 9,420±2,610 IU/week at 12 weeks. During follow up, 96 (51.9%) of the patients needed a reduction in the dose of the erythropoietin. A similar study found that the inclusion of low doses of roxadustat boosts hemoglobin response and still maintains low doses of erythropoietin [19]. This is clinically significant in ESA resistant patients, as it is suggested that increasing doses of ESA may be of limited benefit. Treatment responses seemed to be affected by inflammatory and nutritional status. A higher level of CRP was independently linked to poor response (aOR 0.37; p=0.006), and serum albumin <3.5 g/dL was associated with poor response (aOR 0.41; p=0.013). CRP decreased from 8.9±5.3 to 6.6±4.3 mg/L during follow-up. Previous studies have shown that roxadustat was effective at maintaining hemoglobin levels in patients with higher inflammatory activity and could lead to greater decreases in CRP than traditional ESA therapy [20]. Meta-analysis has also shown that the hemoglobin response is relatively maintained in patients with high inflammatory markers. However, ongoing inflammation can result in ongoing impairment of erythropoiesis and contribute to some of the variability seen in response to therapy in individual patients. Older age ≥60 years (aOR 0.49; p=0.028), diabetes mellitus (aOR 0.46; p=0.019), and dialysis duration ≥5 years (aOR 0.52; p=0.047) were also poor predictors of treatment response. These factors might constitute increased burden of comorbidities, chronic inflammation, nutritional deficits, and increased duration of exposure to dialysis associated factors associated with ESA hyporesponsiveness. Patients receiving roxadustat 40–50 mg were more likely to be treatment responders (aOR 2.74, p=0.008) and there may be a dose-response relationship in the low-dose range. This should be viewed with caution, however, due to treatment dose differences from retrospective studies that may be due to baseline disease severity and physician preference. Generally, combination therapy was well tolerated over the 12-weeks treatment period, and there were adverse events (AEs) reported in 29 (15.7%) patients. The most frequent events were hypertension (17, 9.2%), gastrointestinal symptoms (8, 4.3%), liver enzyme elevation (5, 2.7%), vascular access thrombosis (4, 2.2%) and thromboembolic events (2, 1.1%). Seven (3.8%) patients dropped out of treatment. Previous studies have also demonstrated no significant differences between low dose combination therapy and individual therapy in the number of overall adverse events. Likewise, in pooled analysis of dialysis-dependent CKD patients, there were no significant differences in overall treatment emergent adverse events between roxadustat and ESA therapy. However, it is important to consider the cardiovascular and thrombotic safety also, especially when two erythropoiesis-promoting drugs are used together. There were a number of limitations in this study. Retrospective design and limited to one center decreased the ability to control for potential confounding variables and might have introduced limited generalizability. The long-term efficacy, cardiovascular safety, and sustained hemoglobin response as a result of this relatively short follow-up could not be assessed. Moreover, the lack of a control arm posed challenges in assessing the independent effect of roxadustat.

It is concluded that the addition of low-dose roxadustat to ongoing full-dose erythropoietin was associated with significant improvement in hemoglobin levels and reduced erythropoietin requirements in hemodialysis patients with erythropoietin-resistant anemia. Most patients achieved a favorable treatment response with acceptable tolerability. However, larger prospective controlled studies with longer follow-up are required to confirm the long-term efficacy and safety of this combination.