Estimation of MIC breakpoints for various intravenous dosage regimens of mecillinam

Abstract number: P1458

Jensen K.S., Henriksen A.S., Frimodt-Møller N.

Objectives: Mecillinam (i.v.) is used in Denmark for treatment of complicated, even bacteraemic, urinary tract infections (UTI). Apart from being highly active against E. coli with MIC values of 0.125–0.5 mg/l, it also retains activity against b-lactamase producing, even many ESBL-producing, E. coli, where the MIC has increased to 1–2 mg/l. Recently, the dose vial has been changed to 1000 mg. We therefore wanted to analyze a number of dosage regimens related to the new vial size with respect to relevant MIC levels including ESBL.

Methods: Data for mecillinam concentrations after intravenous bolus injection of a 400, 800, and 1200 mg dose, respectively, in healthy volunteers were on file at LEO Pharma. 11 measurements of serum concentration within 8 hours from the injection were recorded in 6 individuals for the 800 mg as well as the 1200 mg dose. For the 400 mg dose, 9 measurements within 6 hours were recorded in 9 individuals. The NPAG program was used to fit a linear two-compartment model to each of the three data sets. Monte Carlo simulations of various dosage regimens were then made with the compartment model, and results were obtained on MIC breakpoints with respect to a Time>MIC of 40 and 50%, respectively, for several values of the probability of target attainment (PTA = 50, 90, 95%). The free fraction of mecillinam in serum was assumed to be 0.9.

Results: After the parameter estimation process, the compartment model was able to reproduce the measured serum concentrations with good precision. Using the population PK parameter distributions derived from data on the 800 and 1200 mg dose, we simulated the mecillinam concentration in serum when receiving a 1000 mg dose TID. For a target Time>MIC of 40% (PTA=95%), the calculated MIC breakpoint was 2.0 mg/l. Based on the PK parameter values obtained from the data on a 400 mg dose, another dosage regimen, 500 mg QID, was simulated for which a MIC breakpoint of 1.15 mg/l was found (for a Time>MIC of 40%, PTA=95%).

Conclusion: According to the simulation results, a dose of 500 mg QID i.v. would suffice for UTI caused by non-ESBL producing E. coli. For ESBL-producing E. coli, if clinical data allow, the dose should be increased to 1000 mg TID. The high dose is still manageable since mecillinam has low toxicity; a total dose of 60 mg/kg is tolerated in mature humans. The results suggest that a clinical breakpoint of 1 mg/l should be recommended for mecillinam.