CASE: Extrapolation of haematopoiesis dynamics following cytotoxic insult to personalise paediatric drug development
Joe Standing, Nigel Klein, James Yates, AstraZeneca
UCL: Dr Joe Standing, Prof Nigel Klein
AstraZeneca: Dr James Yates, Dr Amy Cheung
Clinical collaborators: Prof Paul Veys, Dr Amos Burke
This is an interdisciplinary PhD in mathematical and statistical modelling applied to clinical datasets and will be of interest to graduates in biology/pharmacy/medicine wishing to enhance their quantitative skills, or graduates from quantitative disciplines such as mathematics, physics or statistics wishing to underake applied research in the biological sciences. Training opportunities to fill knowlegde gaps will be offered in the first year. The PhD is a collaboration between the UCL Great Ormond Street Institute of Child Health (where the student will primarily be based) and AstraZeneca in Cambridge (who will offer a 3-month placement each year), with extensive datasets available from both parties.
Haematopoiesis relates to the development of three important lineages: Lymphocytes, myelocytes and erythroid cells. This project will study the longitudinal effects of drug and stem cell transplant on this system using mathematical modelling to investigate two key stumbling blocks in drug development of agents affecting the haematiopoietic system: Ascertaining human expected effect from pre-clinical animal data; and predicting haematopoietic drug effects in children.
Mechanistic nonlinear mixed effects models will be developed to describe the decline and recovery the haematopoietic system in preclinical and clinical data. Our preliminary work indicates the traditional (Friberg) model is too simplistic for the proposed purpose, and we will explore published extensions and possibly derive new ones. Thrombocytopenia models will be extended to handle exogenous platelet supplementation and no satisfactory mechanistic model has yet been proposed to describe the bi-phasic early lymphocyte recovery. All models on stem cell transplant data will also need to be extended to account for the receipt of exogenous “dosing”. Once the models are developed, multivariable covariate analysis will be used to investigate age-related and species-related effects whilst attempting to delineate factors such as drug and other treatment effects, and disease effects aiming to explain inter-individual variability. A proposed time line is given below (subject to change depending on how the project develops. Funding for travelling to Cambridge from London for the AstraZeneca placements is available):
Months 1-6: Spent at UCL learning nonlinear mixed effects modelling, becoming familiar with and cleaning the data, fitting individual models of neutrophils, platelets and lymphocytes.
Months 7-9: First AZ placement, setting up access to AZ systems (laptop, secure login to data analytics platform) becoming familiar with preclinical and clinical datasets. Two-week placement with at Addenbrookes to become familiar with data. Continue modelling GOSH data under supervision of AZ collaborators.
Months 9-18: Complete GOSH data model, apply to Addenbrookes data, write/defend PhD upgrade report
Months 19-21: Second AZ placement. Begin pre-clinical-clinical translation modelling.
Months 22-24: Continue pre-clinical-clinical translation model under UCL supervision.
Months 25-27: Third AZ placement, finalise pre-clinical-clinical translation model.
Months 28-36: Using the developed models to explore one or more of the following: inclusion of pharmacokinetics, lymphocyte subset analysis, optimising granulocyte colony stimulating factor (G-CSF) dosing and timing.
Hoare RL, Veys P, Klein N, Callard R, Standing JF. Predicting CD4 T-Cell Reconstitution Following Pediatric Hematopoietic Stem Cell Transplantation. Clin Pharmacol Ther. 2017 Aug;102(2):349-357.
Prendergast AJ, Szubert AJ, Berejena C, Pimundu G, Pala P, Shonhai A, Musiime V, Bwakura-
Dangarembizi M, Poulsom H, Hunter P, Musoke P, Kihembo M, Munderi P, Gibb DM, Spyer M, Walker AS, Klein N; ARROW Trial Team. Baseline Inflammatory Biomarkers Identify Subgroups of HIV-Infected African Children With Differing Responses to Antiretroviral Therapy. J Infect Dis. 2016 Jul 15;214(2):226-36.
Standing JF. Understanding and applying pharmacometric modelling and simulation in clinical practice and research. Br J Clin Pharmacol. 2017 Feb;83(2):247-254.
James W. T. Yates, Phillippa Dudley, Jane Cheng, Celina D’Cruz, Barry R. Davies. Validation
of a predictive modeling approach to demonstrate the relative efficacy of three different
schedules of the AKT inhibitor AZD5363. Cancer Chemotherapy and Pharmacology (2015).