Anne Ravix1, Yann Thoma2, Annie Cathignol2, Paul Thoueille3, Catia Marzolini3, Eric Giannoni4, Thierry Buclin3, Chantal Csajka1,5,6, Kim Dao3, Monia Guidi1,3,5
1Centre for Research and Innovation in Clinical Pharmaceutical Sciences, Lausanne University Hospital and University of Lausanne , 2School of Engineering and Management Vaud, HES-SO University of Applied Sciences and Arts Western Switzerland, 3Service of Clinical Pharmacology, Lausanne University Hospital and University of Lausanne, 4Service of Neonatology, Lausanne University Hospital and University of Lausanne, 5Institute of Pharmaceutical Sciences of Western Switzerland, University of Geneva, University of Lausanne, 6School of Pharmaceutical Sciences, University of Geneva, University of Lausanne
Objectives: Vancomycin is a glycopeptide antibiotic with a narrow therapeutic index, widely used to treat severe Gram-positive infections in neonates. Its efficacy best correlates with the daily area under the curve (AUC24) to minimum inhibitory concentration (MIC) ratio, which is challenging to estimate directly in clinical practice due to the need for multiple blood samples. Consequently, trough concentration (Ctrough) is often used as a target proxy, despite being a suboptimal indicator. Model-informed precision dosing (MIPD) could provide a relevant alternative by estimating AUC24 based on both one or two concentration measurements and a validated pre-existing population pharmacokinetic (popPK) model. This proof of concept study aimed to establish the usefulness of Bayesian therapeutic drug monitoring (TDM) based on AUC24 for vancomycin in virtual neonates. Methods: Three dosing adjustment strategies were evaluated in a virtual cohort of neonates: -Strategy 1 (Control) – No dose adjustment: The initial dosing regimen remained unchanged, regardless of whether or not the therapeutic target was achieved. -Strategy 2 – Standard TDM: The dose was adjusted based on the trough level using a proportional rule of three to reach the Ctrough target of 10-20 mg/L, replicating the historical method commonly used in most hospitals. -Strategy 3 – Bayesian TDM: Dosage regimen optimization was performed based on two samples using the Tucuxi MIPD software to achieve the recommended AUC24/MIC target [1]. PK parameters and drug concentrations were simulated for 1,000 virtual neonates using the popPK model developed by Dao et al. (NONMEM), with population characteristics derived from the observed distributions, i.e. median post-menstrual age (PMA): 31 weeks (range: 24-42), bodyweight: 1.4 kg (0.4-4.3), serum creatinine: 54 µmol/L (16-217) [2]. Four initial dosing regimens were tested: the manufacturer’s recommendation (15 mg/kg every 8 h), PMA-based dosage proposed by the neonatal formulary NNF7, the SwissPedDose recommendation (loading dose of 25 mg/kg followed by NNF7 recommendations), and the optimal regimen recently identified in El Hassani study (loading dose of 15 mg/kg followed by 11 mg/kg every 8 h) [3-6]. For Strategy 2, a Ctrough was simulated before the 4th dose, while for Strategy 3, two samples per patient were generated after the first dose: one at 4 hours and one at trough. The dosing adjustment criterion was assessed for each patient, with Strategy 2 targeting a Ctrough between 10-20 mg/L while Strategy 3 aimed for an AUC24/MIC between 360-600 h with a MIC fixed at 1 mg/L. After administration of the initial dosing regimen, the regimen was modified (or not) according to the 3 strategies. A control Ctrough, reflecting clinical practice, was then simulated after three additional modified doses, considering the overall dosing history, for a second TDM assessment. This process was repeated until the respective therapeutic target was reached, with a maximum of four successive adjustment cycles. The three scenarios were compared by assessing the cumulative percentage of patients with an AUC24/MIC within the target therapeutic range. Results: Strategy 1 resulted in a low proportion of patients achieving the therapeutic target, ranging from 8% with the manufacturer’s regimen to 57% with the NNF7 regimen. Patients were often overexposed, ranging from 33% of patients with the NNF7 regimen to 92% with the manufacturer’s regimen. Strategy 2 allowed 86 and 91% of patients to achieve an AUC24/MIC between 360-600 h by the second TDM cycle for the manufacturer and El Hassani regimens, respectively (both q8h). However, it proved ineffective for regimens with longer dosing intervals (q12-24h, NNF7 regimens), with only 35% of patients reaching the target even after four TDM cycles and frequent overexposure. For strategy 3, all dosing adaptations according to their respective guidelines reaches over 76% of patients within therapeutic range after one TDM cycle, 86-95% after two cycles and 97-98% after 4 cycles. Conclusions: TDM is an important tool to optimize vancomycin exposure in neonates and increase the proportion of patients in the therapeutic target. Ctrough-based adjustments can be readily achieved but correlate suboptimally with an AUC24/MIC within the desired target. In contrast, our results demonstrate the superiority of an MIPD software tool such as Tucuxi to optimize exposure, regardless of the initial dosing regimen.
1. Dubovitskaya A, Buclin T, Schumacher M, Aberer K, Thoma Y, editors. Tucuxi – An intelligent system for personalized medicine: from individualization of treatments to research databases and back. 2017: ACM. 2. Dao K, Guidi M, André P, Giannoni E, Basterrechea S, Zhao W, et al. Optimisation of vancomycin exposure in neonates based on the best level of evidence. Pharmacological Research. 2020;154:104278. 3. El Hassani M, Blouin M, Marsot A. A simulation study to assess the influence of population pharmacokinetic model selection on initial dosing recommendations of vancomycin in neonates. British Journal of Clinical Pharmacology. 2024. 4. SwissPedDose. Vancomycin (IV). General use (neonatology). [updated 01/14/2025; Date Accessed: 02/13/2025]. Available from: https://db.swisspeddose.ch/. 5. Ainsworth S. Neonatal Formulary: Drug Use in Pregnancy and the First Year of Life. Vancomycin. seventh ed2015. 6. SwissMedicInfo. Vancomycin Sandoz® i.v. [updated 11/2023; Date Accessed: 02/13/2025]. Available from: https://www.swissmedicinfo.ch/.
Reference: PAGE 33 (2025) Abstr 11641 [www.page-meeting.org/?abstract=11641]
Poster: Methodology - Other topics