Abstract Background and Aims Anemia is a common complication in hemodialysis (HD) patients. Its treatment with erythropoiesis-stimulating agents (ESAs) is challenging due to a nonlinear dose-response relationship and time delays between ESA administration and hemoglobin (Hgb) response. Anemia treatment protocols are frequently used in clinical settings. However, high variability of patient-specific disease characteristics complicate attainment and maintenance of desired Hgb levels. We developed a novel fully personalized ESA dose recommendation tool and present clinical results of a multi-center, randomized controlled trial (RCT) using this software. Method We conducted an RCT in adult HD patients in six dialysis facilities in the US. Patients were randomized 1:1 and treated with our personalized ESA dose recommendation tool for twenty-six weeks (intervention group) or continued to be treated using an anemia protocol that was used as part of standard of care in those clinics (control group). The recommendation tool utilized a physiology-based model of anemia to estimate patient-specific physiological key characteristics, such as red blood cell lifespan, to create a patient-individual model from recent routine clinical data (gender, height, weight, Hgb measurements, and ESA treatment). These key characteristics and model-based outcome predictions were used to generate patient-specific ESA dose recommendations to stabilize Hgb levels in a target window of 10–11 g/dL. Dose recommendations were disseminated to the anemia managers of patients enrolled in the study for evaluation and further decision-making (Figure 1). This procedure was repeated biweekly with updated clinical data. Results Ninety-six patients were enrolled in the RCT. Patients were included in the statistical analysis when they remained in the clinical study for at least 30 days (n = 45 control group, n = 46 intervention group). We evaluated outcome measures showing efficacy and efficiency of the treatment in achieving target Hgb levels. Hgb-related outcomes were significantly different between the two study groups, with an improved Hgb control in the intervention arm, manifesting in a reduction of the Hgb distance to target by more than 30%. Epoetin-beta utilization in the intervention group was decreased by over 20%, while iron-related parameters showed no difference between the two arms (Table 1). Acceptance rate of dose recommendations was high; roughly 95% of the recommendations were accepted and implemented by the clinical staff. Conclusion A therapy software for personalized anemia management was developed for use with epoetin beta. The model-based ESA dose recommendation tool was evaluated in a clinical RCT in HD patients. Hgb control improved significantly in the group using the novel software tool, while ESA usage decreased, thus providing more efficient anemia management for the individual patient while reducing epoetin-beta drug exposure.