20 Predicting treatment response to vancomycin using bacterial DNA load as a pharmacodynamic marker in premature and very low birth weight neonates: a population PKPD study

Introduction LOS has a high risk of morbidity and mortality among premature and VLBW newborns. Whilst positive blood cultures are the gold standard for the diagnosis and subsequent treatment of sepsis, this is time-consuming and results in suboptimal antibiotic treatment regimens. The objective of the present study was to investigate whether treatment response to vancomycin could be quantified using BDL based on RT-qPCR. Methods VLWB and premature neonates with suspected late-onset sepsis were included in a single-centre, observational study. Serial blood samples were collected for measurement of BDL and vancomycin concentration (t=0, 1, 2, 4, 8, 12, 24 and 48h). BDL were measured with RT-qPCR, whereas vancomycin concentrations were measured using LC-MS. A population PKPD model was developed with NONMEM software. Results 28 patients with LOS that were treated with vancomycin were included. A total of 94 vancomycin concentrations and 103 BDLs levels were available. A one- compartment model with PMA and serum creatinine was used to describe vancomycin PK. In 12 patients there was no decrease in BDL over time. Close inspection of the clinical records explained the underlying mechanism of the lack of effect. In 16 patients time profiles of BDL were described with a PD turnover model. The relationship between vancomycin concentration and the increase in first-order BDL elimination was described with a linear effect model. The slope of this model increased with rising PMA. Conclusions BDLs determined through RT-qPCR could be predicted with the population PKPD model. Our findings demonstrate that using RT-qPCR, treatment response to vancomycin may be evaluated as early as 4 hours after treatment initiation, allowing early assessment of efficacy of vancomycin in LOS.