Impact of antibiotic pharmacokinetics in urine on recurrent bacteriuria following treatment of complicated urinary tract infections

ABSTRACT The clinical relevance of bacteriuria following antibiotic treatment of complicated urinary tract infections in clinical trials remains controversial. We evaluated the impact of urine pharmacokinetics on the timing of recurrent bacteriuria in a recently completed trial that compared oral tebipenem pivoxil hydrobromide to intravenous ertapenem. The urinary clearance and urine dwell time of ertapenem were prolonged relative to tebipenem and were associated with a temporal difference in the repopulation of bladder urine with bacteria following treatment, potentially confounding the assessment of efficacy.

The concentration-time profile in urine for TBP following oral administration of a 600 mg dose of TBP-PI-HBr was determined in healthy adult volunteers (38).The concentration-time profile in urine for ETP was modeled based on renal elimination of a 1,000 mg intravenous dose over time, as previously reported (39)(40)(41).Due to high protein binding, renal filtration and urinary elimination of ETP are considerably delayed as compared to TBP (Fig. 1).PK simulations demonstrated that urine concentrations of TBP fell below 0.12 µg/mL in virtually all patients within 24-36 hours.In contrast, ETP urine concentrations above 0.12 µg/mL were maintained in approximately 10% of the patients for 5 days and in some patients for more than a week after the last dose.
To further investigate the possible impact of the longer interval required for ETP to drop to subinhibitory urine levels in patients on the occurrence of bacteriuria at TOC assessment, Monte Carlo simulations were performed vs the MIC distribution of the most common Enterobacterales uropathogens (Fig. 2 and 3).The probability of target attainment (PTA) was evaluated by PK-PD population simulation methods.For each treatment regimen, 30,000 virtual subjects were generated.PK parameters were scaled by allometry.Urine concentrations were calculated based on the fraction excreted during the interval of interest (defined as 6 hours) and a constant assumed rate of urine production of 1 mL/kg/h.The fraction of time above MIC was calculated by subject-sam pled MIC distribution across all subjects and summarized by drug, pathogen, and MIC.Theoretical PK-PD targets of maintaining a urine concentration of TBP or ETP above a given MIC for 10% or 30% of the 7 days following the cessation of antibiotic therapy were interrogated.These theoretical targets were chosen to provide a quantitative compari son of the proportion of patients with inhibitory urine concentrations of TBP or ETP following the EOT.For a target of 10% urine time above MIC (T > MIC), based on the urine concentration-time profile, ETP demonstrated PTA > 99% at 0.12 µg/mL, the MIC 90 for the Enterobacterales pathogens isolated at baseline.In contrast, TBP demonstrated a parallel PTA of ~30% for the 7 days following the EOT.For the theoretical target of maintaining urine concentrations above the MIC 90 for 30% of the 7 days following the cessation of antibiotic therapy, ETP PTA was >70% compared with PTA of 0% for TBP.
It is plausible that repopulation of the bladder urine following treatment of a cUTI requires that antibiotic concentrations in urine drop to subinhibitory levels, allowing for persistent pathogens to resume planktonic growth (18,19,24,25).Although urine PD requirements for suppression of bacteriuria have not been defined, the longer dwell time of ETP in urine after the last dose may contribute to the larger difference in microbiologic response seen at the TOC as compared to the EOT or LFU in this clinical trial.These hypothesis-generating data are being tested in an ongoing cUTI clinical trial comparing oral TBP-PI-HBr to intravenously administered imipenem-cilastatin, both with comparable urine PK (Fig. 1).

FIG 2
FIG 2 PTA in urine vs Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis in the week following cessation of treatment with TBP-PI-HBr 600 mg q8h and ETP 1,000 mg q24h.A target for PTA of 10% of the week above MIC was used.Pathogen frequency and MIC distributions from references 21, 31, with no correlation assumed.

FIG 3
FIG 3 PTA in urine for Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis in the week following cessation of treatment with TBP-PI-HBr 600 mg q8h and ETP 1,000 mg q24h.A target for PTA of 30% of the week above MIC was used.Pathogen frequency and MIC distributions from references 21, 31, with no correlation assumed.