Epoetin Dosing Regimens in Haemodialysis

Current European Best Practice Guidelines and the KDOQI Committee in the USA recommend preferential subcutaneous (SC) twice- to thrice-weekly epoetin administration. SC route significantly reduces epoetin requirements and therefore costs, compared to the intravenous (IV) route. Additionally, there is solid evidence showed once-weekly SC administration of epoetin beta to be equally efficient and well tolerated in hemodialyzed (HD) patients. Clinical studies suggest that the effects of epoetin alpha are sustained for more than one week, enabling less frequent schedules in pre-dialysis chronic kidney disease (CKD) patients. A recent European multicenter study in peritoneal dialysis (PD) patients confirmed that once-every-two-weeks SC epoetin beta is efficient and safe in the maintenance phase of anemia treatment for PD patients11. Thus, the optimum epoetin dosing regimen is still yet to be determined despite over a decade of clinical use.

All these data also suggest that the pharmacodynamic effects of epoetins could last longer than their plasma half-lives. The survival half-life of erythrocytes produced after exogenous epoetin administration is longer than that of erythrocytes produced in the absence of exogenous stimulation.

Apart of the academic interest, this issue has also direct implications on cost savings and increased patients' convenience. Furthermore, less frequent administration would reduce epidemiological hazards associated with populations at high risk of blood-borne viral infections such as hepatitis, particularly important in less developed countries, with a high prevalence of viral hepatitis infections (hepatitis B/C markers prevalences: 15.7%/45% in Romania versus 3%/13.5% in EuroDOPPS patients). While the potential benefits of less frequent dosing are clear, the efficacy and safety of such regimens must be fully investigated before adoption as standard treatment.

The present study investigates for the first time the equivalence of once-weekly versus once-every-two-weeks SC epoetin beta dosing regimens, in the maintenance phase of anemia therapy, in stable HD patients. The aim of the study is to assess whether once-every-other-week SC administration of the same total epoetin beta dose is as effective and as safe as SC once-weekly dosing.

In order to avoid a carry-over effect of the weekly schedule of epoetin treatment received prior to randomization, we defined a run-in period (weeks 1-12). Statistical analyses will be performed for the efficacy parameters determined during the assessment period only. The average for weeks 13-24 will be obtained by adding all values and dividing the sum by the number of time points for each patient during this period.

The two treatment schedules will be considered equivalent if the primary efficacy parameters will be simultaneously similar for both groups and in the predefined range of variation. Confidence intervals (CIs) will be used to compare groups. Since the target Hb in dialyzed patients is defined as 11g/dL (110 g/L) by the European Guidelines and as >10 g/dL (100 g/L) by the National Guidelines, with a recommended upper limit of 13 g/dL (130 g/L), the efficacy range for Hb in this study was predefined as 10-12 g/dL (100-120 g/L). The two treatment schedules will be considered to have similar efficacy if the mean Hb in Group 2w will not differ by more than ±0.5 g/dL (±5 g/L) compared to Group 1w during the assessment period. Once similar efficacy established, drug requirements will be compared calculating the ratio of the mean weekly epoetin doses in Group 2w/Group 1w. A range of 0.8 to 1.25 for the ratio is considered sufficient to define bioequivalence. Equivalence of drug usage in the two arms will be accepted if the whole 95% CI for this ratio will be within the above limits.

Lack of difference between group means does not imply similar distribution of treatment effects within each group. The individual hemoglobin change will be used to assess if response to treatment was similarly variable in the two arms. The change in Hb will be calculated for each patient as the difference between the mean Hb during the assessment period and the mean Hb during the baseline phase. The percentages of patients with target Hb and not requiring any increase in epoetin dose during the assessment period will be compared with chi-squared statistic.

The study is designed to detect a difference in Hb between the study groups of at least 0.5 g/dL (5 g/L), with a probability of 95% and a power of 0.9021, assuming a standard deviation of Hb in the HD population of 1.0 g/dL (10 g/L). A sample size of 85 patients is required in each arm of the trial. Since the study design includes epoetin doses variations in order to maintain target hemoglobin, it would have been expected that Hb would be constant and epoetin dose would vary. Therefore, the power calculations using epoetin doses as the outcome measure seem to be more appropriate. Considering a standard deviation of 39 IU/kg per week, a probability of 95% and a power of 90%, a sample size of 70 patients in each arm is required to detect a ± 20% difference in epoetin dose between groups. Because of an anticipated drop-out proportion of 15%, a minimum of 100 patients have to be enrolled in each group to satisfy both Hb and epoetin doses power conditions.