Abstract
The lack of biomimetic and tractable preclinical in vitro models in Acute Lymphoblastic Leukaemia (ALL) hinders drug development and drug testing. There is still a heavy reliance on patient derived xenograft models in immune-compromised mice, which are gold standard preclinical models in leukaemia research. These models have a slow data turnaround time which can vary between 1-6 months, so are not beneficial for clinical and pharmaceutical applications. Developing a predictive preclinical model needs to capture the complexities of the leukemic bone marrow niche including cellular interactions; the extracellular matrix (ECM); and the biomechanical properties. These aspects are established in driving the disease and influencing treatment response. Such models are especially relevant today given Food and Drug Administration’s (FDA) recent announcement that it will no longer require all drugs to be tested on animals prior to in-human trials. Therefore, this project provides a key opportunity in the development of predictive human-cell-based preclinical models with optimal and clinical translatability.Here development of an animal component-free prototype hydrogel formulation (VAL) is described, comprising of key representatives of ECM components of the human bone marrow in defined formulations. This hydrogel is compatible as a bioink for 3D bioprinting within a Reactive Jet Impingement (ReJI) 3D bioprinting platform. iPSC- derived mesenchymal stem cells (iMSC) are engineered and characterised before incorporation into a iMSC-VAL bone marrow organoid. These models have the ability to support long term co-culture of patient-derived leukaemia cells and they exhibit optimal predictivity to clinically applicable compounds.
The PDX-ALL-iMSC-VAL models have been applied to study mechanisms of treatment resistance including treatment-induced quiescence and tunnelling nanotubules (TNTs). Here, TNTs demonstrate a clear mechanism of microenvironment-conferred treatment response, with TNT formation observed under treatment pressure, and inhibition of these TNTs via addition of CDH2 inhibitor ADH- 1 to the treatment regimen.
The model production is also shown to be adaptable towards 3D bioprinting processes, with extensive optimisation reported here in order to efficiently produce PDX-ALL-iMSC-VAL models in a high throughput and reproducible manner. Such pipelines are shown to produce patient-specific drug response data to a triple drug combination matrix.
Generation of such tractable models in a high throughput, and defined manner lends itself to wider uptake and furthermore demonstrates the flexibility of the model to generate patient-specific data. This results in the potential for use in personalised medicine strategies.
| Date of Award | 26 Jun 2025 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Deepali Pal (Supervisor) & Ruchi Shukla (Supervisor) |
Keywords
- Organoid Modelling
- 3D Bioprinting
- Paediatric Leukaemia and Leukaemia Niche
- Drug Screening
- Induced-Pluripotent Stem Cells
Cite this
- Standard