Towards precision medicine: Therapeutic drug monitoring–guided dosing of vancomycin and β-lactam antibiotics to maximize effectiveness and minimize toxicity

Abstract

<jats:title>Abstract</jats:title> <jats:sec> <jats:title>Purpose</jats:title> <jats:p>The goal of this review is to explore the role of antimicrobial therapeutic drug monitoring (TDM), especially in critically ill, obese, and older adults, with a specific focus on β-lactams and vancomycin.</jats:p> </jats:sec> <jats:sec> <jats:title>Summary</jats:title> <jats:p>The continued rise of antimicrobial resistance prompts the need to optimize antimicrobial dosing. The aim of TDM is to individualize antimicrobial dosing to achieve antibiotic exposures associated with improved patient outcomes. Initially, TDM was developed to minimize adverse effects during use of narrow therapeutic index agents. Today, patient and organism complexity are expanding the need for precision dosing through TDM services. Alterations of pharmacokinetics and pharmacodynamics (PK/PD) in the critically ill, obese, and older adult populations, in conjunction with declining organism susceptibility, complicate attainment of therapeutic targets. Over the last decade, antimicrobial TDM has expanded with the emergence of literature supporting β-lactam TDM and a shift from monitoring vancomycin trough concentrations to monitoring of the ratio of area under the concentration (AUC) curve to minimum inhibitory concentration (MIC). PK/PD experts should be at the forefront of implementing precision dosing practices.</jats:p> </jats:sec> <jats:sec> <jats:title>Conclusion</jats:title> <jats:p>Precision dosing through TDM is expanding and is especially important in populations with altered PK/PD, including critically ill, obese, and older adults. Due to wide PK/PD variability in these populations, TDM is vital to maximize antimicrobial effectiveness and decrease adverse event rates. However, there is still a need for studies connecting TDM to patient outcomes. Providing patient-specific care through β-lactam TDM and transitioning to vancomycin AUC/MIC monitoring may be challenging, but with experts at the forefront of this initiative, PK-based optimization of antimicrobial therapy can be achieved.</jats:p> </jats:sec>

Evaluation of a Pilot Vancomycin Precision Dosing Advisory Service on Target Exposure Attainment Using an Interrupted Time Series Analysis

“…However, previous work has established that the precision and bias for AUC estimation is within clinically acceptable limits. 38,39 Furthermore, while it is acknowledged that this surrogate measure requires further confirmation, particularly with respect to the upper toxicity threshold, 40 this study demonstrated a significant reduction in the length of vancomycin therapy following Service implementation, which may reflect a decreased time to infection resolution. Of note, this change occurred in the context of a simultaneous 27% relative reduction in the proportion of AUC 24 values exceeding the toxicity threshold.…”

From Therapeutic Drug Monitoring to Model‐Informed Precision Dosing for Antibiotics

“…The importance of TDM in maximizing antimicrobial effectiveness and decreasing adverse events has been emphasized also for additional populations with altered PK, e.g., obese, elderly, pregnant, or burns patients, as well as patients with difficult-to-treat bone infections. 76,86…”

Continuous Learning in Model‐Informed Precision Dosing: A Case Study in Pediatric Dosing of Vancomycin

“…Still, there are a number of other drugs well-positioned for MIPD that could benefit from the continuous learning strategy presented here. 24,25 There appears to be limited gains in predictive performance from fitting a continuous learning model on > 200 patients, with the models fit on 350 patients and 200 patients performing similarly in prediction error and bias. Consistent with this finding, the models from the literature trained on at least 200 patients (Avedissian: 250, Colin: 2,554, and Kloprogge: 616) generally showed better predictive performance than those of models fit on fewer than 200 patients (Lamarre: 78, and Le: 138).…”

“…For these applications, continuous learning may need to be expanded to include a systematic covariate selection process. Still, there are a number of other drugs well‐positioned for MIPD that could benefit from the continuous learning strategy presented here 24,25 …”