In 2022, the European commission complemented the new Regulation (EU) 2019/6 on veterinary medicinal products by several articles aiming to further promote prudent and responsible use of antibiotics. Some antibiotics were banned for all veterinary usages, while others were strictly limited to specific indications. Veterinarians will also be strongly encouraged to use the few antibiotic classes classified in category D by the EMA (European Medicines Agency), i.e. category with the lower impact for human health including penicillins, aminopenicillins, tetracyclines and TMP/sulfonamides. The relative use of these antibiotics should increase in the coming years across Europe, especially if their efficacy is not compromised compared to more critical antibiotics. However, many of these drugs were approved in the 1980s, and at that time, the dossiers for dose selection did not have the same requirements in terms of pharmacodynamic (PD)/pharmacokinetic (PK) data to justify the dose selection. Registered dosing regimens for these old antibiotics are likely to be suboptimal. Indeed, EMA considers that the posology of old products may require critical evaluation, and published a reflection paper in 2021 on dose optimization of established veterinary antibiotics by PK/PD approach (1). PK/PD assessments are based on the MIC of the target pathogen and the antibiotic concentrations in the biophase. Therefore, in some cases, new PK and PD studies must be performed to obtain all the necessary data. Drugs for which pharmacokinetic studies have been published with very inaccurate concentrations determined by microbiological testing are an example, as is the lack of MIC distribution for certain veterinary pathogen/antibiotic pairs. The PK/PD indices (AUC/MIC, Cmax/MIC and T > MIC) can then be calculated from the PK and PD data and compared with their target values. Again, if target values are not available, relevant values can be obtained from in vitro time-kill studies. Finally, the relationship between dose and probability of achieving the target values (PTA) can be modeled and the expected efficacy of current dosing regimens predicted for each pathogen/antibiotic pair in each animal species. If low efficacy is predicted, two options can be considered depending on regulatory and clinical constraints: adjusting doses or restricting use to more susceptible bacterial strains or species. Such an approach will allow a responsible and prudent use of AMD in veterinary medicine and will benefit human and environmental health by avoiding inefficacious treatments and by limiting the recourse to second-line antibiotics more critical for human health. Reference: 1. EMA/CVMP/849775/2017