2024-25 Project (Baker & Standing & Lonsdale)
Determining the pharmacokinetic-pharmacodynamic index of commonly used antimicrobials and exploring emergence of drug resistance using non-linear mixed-effects modelling techniques
Professor Emma Baker at SGUL
Professor Joe Standing at UCL
Dr Dagan Lonsdale at SGUL
This PhD is designed to provide an exciting opportunity for someone with an interest in data science, laboratory work and a passion for improving the outcome of patients with infection.
Our research group is interested in answering the question – what is the right dose of antibiotics to give people to maximise the chance of curing infection and minimise the opportunity for antibiotic resistant bacteria to emerge? Your project will combine state of the art laboratory work with pharmaceutical industry standard data analytic techniques to answer this question.
Historically, laboratory studies to investigate the best doses of drugs to use have utilised animal models (usually mice). Whilst these studies are considered the best representation of human infections in which to investigate drug dosing (dose-range experiments), they have limitations. The animals must be rendered without an immune system (neutropenic), they must have a high dose of bacteria injected into a muscle or wound (which may not grow) and they have to have high drug doses given (the concentration-time profile of which may not mimic humans).
Our collaboration has established novel methodology to grow bacteria in closed circuit systems in which we can closely control drug concentration and make it mimic the rise and fall of antibiotic concentration seen in human blood after taking a dose of a drug. These models allow us to take rich data about bacterial growth and the emergence of resistance – something you can’t do in the animial models as you can only test the bacteria after killing the animal. Alongside this laboratory work we have a rich data-lake of clinical drug concentration (pharmacokinetic) studies.
We are looking for someone with an interest in data science to investigate the relationship between drug and bacteria. You will learn how to model drug concentration and bacterial growth using pharmaceutical industry standard techniques (non-linear mixed-effects modelling) and combine the knowledge gained from the laboratory experiments and clinical studies to simulate the effect of different dosing regimens in humans. We hope this will allow us to either confirm that the current doses we use are acceptable or provide information to design clinical trials of optimal doses.
You will have the opportunity to undertake a period in the laboratory learning how to run the hollow fibre infection models so you can expand the number of drugs which we have data available for. You will learn how to use industry standard software (NONMEM, R, nlmixr) that will provide a foundation for a career in data science in the pharmaceutical sector. We have a dynamic and vibrant team that will welcome you and help you on your journey.
Project Key Words
Pharmacometrics, infection, antimicrobial pharmacokinetics-pharmacodynamics, hollow fibre infection model
MRC LID Themes
- Global Health = Yes
- Health Data Science = Yes
- Infectious Disease = Yes
- Translational and Implementation Research = Yes
MRC Core Skills
- Quantitative skills = Yes
- Interdisciplinary skills = Yes
- Whole organism physiology = No
Skills we expect a student to develop/acquire whilst pursuing this project
The student will develop knowledge and skills in non-linear mixed-effects modelling. They can expect to be proficient in one or more coding languages by the completion of the PhD (NONMEM, R, Python). They will gain skills in scientific communication (written and oral) and will have laboratory skills in running hollow fibre infection models (tissue culture, drug preparation). It is anticipated that the student will drive the direction of the modelling and simulation and in so doing acquire skills in independent scientific thought and project development.
Which route/s is this project available for?
- 1+4 = No
- +4 = Yes
Is this project available for full-time study? Yes
Is this project available for part-time study? Yes
Particular prior educational requirements for a student undertaking this project
- SGUL’s standard institutional eligibility criteria for doctoral study.
- This project would suit a student with a background in pharmacology and/or statistics at undergraduate level with some basic experience of modelling. Masters level training in these disciplines would be desirable. Equivalent experience through non-traditional routes of training will be considered by the project team.
Other useful information
- Potential CASE conversion? = No
PROJECT IN MORE DETAIL
Scientific description of this research project
Mortality from sepsis in adults, children and neonates remains high (25-30%) and reducing death from infection remains a global priority. The cornerstone of drug therapy in infection are antibiotics. Clinical research has prioritised optimising antibiotic dosing, but it is often challenging to know what the right dose regimen to use is, in part because the relationship between drug concentration and antibacterial effect is difficult to determine in human trials.
This project aims to use state of the art mathematical modelling techniques to describe the relationship between antimicrobial drug concentration, its antibacterial action (bacterial stasis/killing) and the evolution of antimicrobial resistant bacterial strains. The PhD candidate will use pharmacokinetic-pharmacodynamic models derived from in-vivo laboratory experiments and known drug pharmacokinetics in humans to simulate optimal dosing strategies for antimicrobials that can be used to confirm the validity of existing licensed dosing strategies or inform clinical trials for new ones.
Project objectives and techniques to be used:
Using non-linear mixed-effects modelling techniques
1. Determine pharmacokinetic-pharmacodynamic targets that maximise drug efficacy for meropenem, amikacin, ciprofloxacin, piperacillin-tazobactam and fosfomycin using the hollow-fibre infection model
2. Describe the relationship between exposure and emergence of resistant bacteria for these drugs (mutant selection window)
3. Using targets from objectives 1&2 and in-vivo human pharmacokinetic data, simulate clinical trials to predict optimal dosing strategies
Use intersectional analytical methodology to explore the impact of social identities (e.g. race, gender) on pharmacokinetics
Using wet-laboratory methodology
4. Undertake hollow fibre infection model work to determine the dose-response relationship for co-amoxiclav (placement)
Historically, pre-clinical dose-ranging experiments that were used to determine antimicrobial dosing in humans are often animal (mouse) models. However, there are a number of limitations to these studies – drug half-life is very short in animals and fails to accurately mimic human pharmacokinetics, some organisms do not grow well and chemotherapy used to render animals neutropenic may confound drug response. In clinical settings, it is challenging to determine whether the dose-response relationships established in animal models are mirrored in humans. This is because in clinical studies we do not know the timing of infection, we cannot control the pathogen, or its susceptibility and patient comorbidity (including critical illness) confound studies
Recently, the hollow fibre infection model has been developed.This in vitro system allows similar dose-ranging studies to be carried out by growing bacteria in a closed system that can have drug flowing through it. Drug concentration can therefore closely mirror the concentration-time profile in humans and dynamic sampling of the bacteria allows investigation of bacterial properties at multiple time points in an experiment (instead of one time point in the animal models). Whilst researchers have often used it to simulate clinical dosing regimens, another application is to mimic dose fractionation studies thereby replacing animal models. (J Antimicrob Chemother. 2022 Dec 23;78(1):8-20)
Confirmed availability of databases/specialist materials: Data on dose fractionation experiments including colony forming unit counts on normal and drug containing plates at multiple time points have been collected from the hollow fibre infection model for the study antibiotics. Our group has a large data-lake from several antimicrobial pharmacokinetic studies in neonatal, child and adult populations (J Antimicrob Chemother. 2023 Sep 5;78(9):2148-2161. doi: 10.1093/jac/dkad196 & J Antimicrob Chemother. 2020 Dec 1;75(12):3625-3634).
The UCL hollow fibre laboratory (Professor Standing) will host the student for a three-month period during which time the PhD candidate will learn how to do wet-lab hollow-fibre experiments themselves.
Risk for the project and mitigation: Data is available and the track record of the supervisory team is strong. The largest skill gap for a potential candidate is likely to be knowledge in non-linear mixed-effects modelling techniques. The team run courses aimed at people in various stages of their training and we anticipate the risk will be mitigated by using materials from these as well as offering attendance to the candidate.
(Relevant preprints and/or open access articles)
Additional information from the supervisory team
- The supervisory team has provided a recording for prospective applicants who are interested in their project. This recording should be watched before any discussions begin with the supervisory team.