Dual β-lactam therapy against high-risk Pseudomonas aeruginosa isolates: a dynamic in-vitro infection model study integrating population genomics with quantitative systems pharmacology modelling and simulations.
Combination therapy with ceftolozane-tazobactam and meropenem may suppress bacterial resistance more effectively than monotherapy in treating Pseudomonas aeruginosa infections.
Where it sits
this study against the rest of the mazdutide (ibi362) corpusSummary and findings
This study investigated the response of extensively drug-resistant Pseudomonas aeruginosa isolates to β-lactam antibiotics using a hollow-fibre in-vitro infection model. The isolates were treated with ceftolozane-tazobactam and meropenem, both alone and in combination, to assess their effects on bacterial resistance. The findings indicated that combination therapy suppressed resistance more effectively than monotherapy.
Abstract
<h4>Background</h4>Pseudomonas aeruginosa has an extraordinary capacity for resistance emergence during treatment, even with newer antipseudomonals. There is a gap in understanding how resistance mechanisms affect the time-course of bacterial response to these newer agents. Traditional approaches for predicting pathogen response to an antibiotic do not apply to combination therapy. We aimed to develop a modelling framework to predict treatment response based on resistome information, using isolates of the worldwide-disseminated high-risk clone sequence type (ST) 235 and β-lactam antibiotics as the example.<h4>Methods</h4>In this hollow-fibre in-vitro infection study, we used three extensively drug-resistant ST235 clinical isolates from the national collection of the Clinical Microbiology Department of the Hospital Son Espases (Palma de Mallorca, Spain) that were hospital-acquired, were isolated following routine microbiological procedures from different patients between 2017 and 2022, were susceptible to ceftolozane-tazobactam, and had different levels of meropenem resistance. The selected isolates (ST235-05, ST235-09, and ST235-10) showed classical β-lactam resistance mechanisms pre-treatment. The isolates were investigated in 240-h dynamic hollow-fibre in-vitro infection models (HFIMs). The studies exposed the isolates to pharmacokinetic profiles of ceftolozane-tazobactam (simulating 1 g of ceftolozane and 0·5 g of tazobactam as a 3-h infusion every 8 h) and meropenem (simulating 6 g per day continuous infusion) as observed in hospitalised patients, as monotherapy and in combination. Treatment response was assessed through the quantification of the time-courses of viable total and resistant bacteria. Whole-genome sequencing identified the mechanisms of emerging resistance. A quantitative systems pharmacology (QSP) approach was used to model total and resistant bacterial counts and corresponding pharmacokinetic data from the HFIM. Monte Carlo simulations were used to predict treatment responses in 1000 virtual infected patients treated with ceftolozane-tazobactam and meropenem as monotherapies or in combination over 10 days.<h4>Findings</h4>In the HFIMs, each antibiotic alone amplified resistance by approximately 48 h for all isolates; that is, monotherapies resulted in a higher concentration of resistant bacteria compared with the control treatment at the respective time, except ceftolozane-tazobactam against ST235-10. Combination of ceftolozane-tazobactam and meropenem was synergistic (bacterial counts ≥2 log<sub>10</sub> colony forming units [CFU] per mL lower than the best performing monotherapy and initial inoculum) against all isolates and suppressed resistance. Against ST235-10, ceftolozane-tazobactam monotherapy reduced counts to less than 1 log<sub>10</sub> CFU per mL from 192 h onwards, whereas the combination reached less than 1 log<sub>10</sub> CFU per mL by 24 h. Across strains, population genomics confirmed monotherapy failures were associated with emerging resistance mechanisms (ceftolozane-tazobactam: ampC Ω-loop mutations; meropenem: ftsl mutation). The developed QSP model incorporated baseline resistance mechanisms and those emerging in resistant mutant subpopulations. The model explained and predicted the monotherapy failures involving amplification of these subpopulations, and synergistic killing and resistance suppression by the combination. Simulations using the model predicted bacterial regrowth above the initial inoculum for more than 90% of patients after 0 to approximately 3 days for meropenem monotherapy across all strains and for ceftolozane-tazobactam monotherapy against ST235-05 and ST235-09. For ceftolozane-tazobactam monotherapy against ST235-10, regrowth was predicted for approximately 30% of patients. In contrast, the simulations predicted sustained bacterial killing of at least 2 log<sub>10</sub> CFU per mL compared with the initial inoculum by the combination for more than 89% of patients across all strains.<h4>Interpretation</h4>To our knowledge, this model is the first to characterise and predict the time-course of responses of clinical isolates to antibiotics only by the resistance mechanisms present and their complex interplay, representing a step towards pathogen-specific, personalised medicine.<h4>Funding</h4>Australian National Health and Medical Research Council.
Background
This paper addresses the challenge of resistance emergence in Pseudomonas aeruginosa during antibiotic treatment, particularly with β-lactam antibiotics. Previous research has shown that traditional predictions of bacterial response do not apply to combination therapies. This study aims to develop a modeling framework that incorporates resistance mechanisms to predict treatment outcomes, which is crucial for improving therapeutic strategies against resistant infections.
Methods
The study utilized a hollow-fibre in-vitro infection model with three extensively drug-resistant ST235 clinical isolates. The isolates were subjected to pharmacokinetic profiles of ceftolozane-tazobactam and meropenem over a 240-hour period. Treatment response was measured through the quantification of viable total and resistant bacteria, and whole-genome sequencing was performed to identify resistance mechanisms.
Results
The primary endpoint indicated that monotherapy amplified resistance by approximately 48 h. Combination therapy was found to be synergistic, resulting in bacterial counts ≥2 log10 CFU per mL lower than the best performing monotherapy. The combination also suppressed resistance, with sustained bacterial killing predicted for more than 89% of patients across all strains.
Interpretation
The findings suggest that combination therapy may be more effective in suppressing resistance compared to monotherapy, aligning with previous literature on the benefits of combination treatments. However, the clinical significance of the effect size remains uncertain, and the study's reliance on in-vitro models limits the generalizability of the results. The implications for practice suggest a potential shift towards personalized medicine based on resistance mechanisms.
Key findings
- Monotherapy amplified resistance by approximately 48 h for all isolates.
- Combination therapy resulted in bacterial counts ≥2 log10 CFU per mL lower than the best performing monotherapy.
- Ceftolozane-tazobactam monotherapy reduced counts to less than 1 log10 CFU per mL from 192 h onwards against ST235-10.
- The combination reached less than 1 log10 CFU per mL by 24 h.
- Simulations predicted bacterial regrowth above the initial inoculum for more than 90% of patients after 0 to approximately 3 days for meropenem monotherapy.
Limitations
- In-vitro study may not fully replicate in vivo conditions.
- Limited number of clinical isolates used.
- Short duration of treatment in hollow-fibre model.
- No long-term follow-up data reported.