Development of an erythropoietin prescription simulator to improve abilities for the prescription of erythropoietin stimulating agents: is it feasible?

Abstract

Background: The increasing use of erythropoietins with long half-lives and the tendency to lengthen the administration interval to monthly injections call for raising awareness on the pharmacokinetics and risks of new erythropoietin stimulating agents (ESA). Their pharmacodynamic complexity and individual variability limit the possibility of attaining comprehensive clinical experience. In order to help physicians acquiring prescription abilities, we have built a prescription computer model to be used both as a simulator and education tool.

Methods: The pharmacokinetic computer model was developed using Visual Basic on Excel and tested with 3 different ESA half-lives (24, 48 and 138 hours) and 2 administration intervals (weekly vs. monthly). Two groups of 25 nephrologists were exposed to the six randomised combinations of half-life and administration interval. They were asked to achieve and maintain, as precisely as possible, the haemoglobin target of 11-12 g/dL in a simulated naïve patient. Each simulation was repeated twice, with or without randomly generated bleeding episodes.

Results: The simulation using an ESA with a half-life of 138 hours, administered monthly, compared to the other combinations of half-lives and administration intervals, showed an overshooting tendency (percentages of Hb values > 13 g/dL 15.8 ± 18.3 vs. 6.9 ± 12.2; P < 0.01), which was quickly corrected with experience. The prescription ability appeared to be optimal with a 24 hour half-life and weekly administration (ability score indexing values in the target 1.52 ± 0.70 vs. 1.24 ± 0.37; P < 0.05). The monthly prescription interval, as suggested in the literature, was accompanied by less therapeutic adjustments (4.9 ± 2.2 vs. 8.2 ± 4.9; P < 0.001); a direct correlation between haemoglobin variability and number of therapy modifications was found (P < 0.01).

Conclusions: Computer-based simulations can be a useful tool for improving ESA prescription abilities among nephrologists by raising awareness about the pharmacokinetic characteristics of the various ESAs and recognizing the factors that influence haemoglobin variability.

Figure 1

Haemoglobin course; simulation A (before CERA's marketing). Hb as a function of the weeks elapsed since the start of the simulation for the 6 combinations of half-lives and administration intervals (half-life in h: 24, 48 and 138; administration interval, weekly (W) or monthly (M); 24W solid line with black triangles, 48W solid line with black squares, 138W solid line with black diamonds, 24M dotted line with white triangles, 48M dotted line with white squares, 138M dotted line with white diamonds). The randomly-assigned bleeding phase between the two equilibration exercises is not represented; the second equilibration phase is synchronised. N = 25.

Figure 2

Haemoglobin course in simulation B (after CERA's marketing). Hb as a function of the weeks elapsed since the start of the simulation for the 6 combinations of half-lives and administration intervals (half-life in h: 24, 48 and 138; administration interval, weekly (W) or monthly (M); 24W solid line with black triangles, 48W solid line with black squares, 138W solid line with black diamonds, 24M dotted line with white triangles, 48M dotted line with white squares, 138M dotted line with white diamonds). To better evaluate the equilibration phase no bleeding episodes were introduced. N = 25.

Figure 3

Monthly haemoglobin course comparing monthly and weekly administration intervals; simulation B. Monthly Hb as a function of the weeks elapsed since the start of the simulation for the 2 administration intervals: monthly M (dotted line with white squares) or weekly W (solid line with black squares). N = 25

Figure 4

Weekly haemoglobin course comparing monthly and weekly administration intervals; simulation B. Weekly Hb as a function of the weeks elapsed since the start of the simulation for the 2 administration intervals: monthly M (dotted line with black squares) or weekly W (solid line with black squares). The simulator user was aware of the monthly values only. N = 25.

Figure 5

Haemoglobin variability in simulation B. Hb variability expressed as standard deviation (SD) and delta Hb (absolute value of the difference between consecutive measurements) comparing the 2 administration intervals: monthly M (black columns) and weekly W (white columns). The difference between columns in "Delta Hb" is significant; P < 0.01. N = 25.

Figure 6

Epoetin dose reduction, equilibration vs. maintenance in simulation B. Error in determining the maintenance dose estimated by calculating the percentage difference between the mean erythropoietin dose used in the equilibration phase and that used in the maintenance phase (half-life of 24, 48 and 138 h; monthly M and weekly W administration). N = 25.

Figure 7

Haemoglobin variability and epoetin dose adjustments in simulation B. Correlation between the number of therapy modifications and Hb variability (expressed as delta Hb) in the maintenance phase. N = 25.

Figure 8

RBC lifespan variability in simulation B. RBC lifespan variability (expressed as standard deviation) as a function of erythropoietin half-life and administration interval (half-life in hours, 24, 48 and 138; administration interval, weekly W or monthly M). N = 25.