Fully humanised 3D vascular perfused model for breast cancer modelling and therapeutic screening

Breast cancer is the most common cancer affecting women. Scientists use a range of models to study this in the laboratory. This includes growing cells on plastic dishes or using animal models to mimic how breast cancer behaves. Neither model is ideal; growing cells on plastic is very different from how cells exist in the human body, while the animals used to mimic breast cancer are usually mice, which are very different from humans. We wish to implement a closer-to-patient model of breast cancer. This still uses cells growing in the laboratory, but these are growing in a way that is more similar to how they are in the human body. The cells grow in a three-dimensional environment, Currently, this three-dimensional environment contains animal-derived substances. In this project we wish to change the three-dimensional environment to make it completely free of any animal products. We can do this using a substance called a hydrogel. We can add different substances to the hydrogel to make it the same as the environment in which breast cancer cells grow in the body. Once we have achieved this we will make this environment closer still to how cancer cells are in the body. In the body cancer cells have a blood supply which allows them to receive a continuous flow of nutrients required to keep them alive. This is how drugs designed to kill cancer cells are also delivered. This is hard to achieve in the laboratory. We have overcome this by using an artificial blood substitute which we will introduce into our three-dimensional environment. This will produce a breast cancer model which has many similarities to how breast cancer cells grow in the human body, which we will be able to use routinely in our own labs. We want as many scientists as possible to know about this model and be able to use this, so we will organise training events where scientists from all over the UK can receive practical hands-on training in our technology so that are confident to start using this in their own labs. Finally, as university academics we are responsible for training the next generation of scientists so will introduced this technology into our teaching through lectures and practical sessions.

Our technology has developed from two independently funded NC3Rs projects, uniting the Merry (developer) and Speirs (end user) groups. Merry has pioneered the development of self-assembling peptide hydrogels. These provide a fully tuneable matrix, devoid of any animal components, for disease modelling and therapeutic screening. Speirs is using a novel, synthetically vascularised, Organ-on-a-Chip screening platform, PerfusionPal, developed through NC3Rs CRACK IT by collaborator Vukasinovic and advanced though a current NC3Rs studentship to Speirs, to meet demands for physiologically closer in vitro platforms to model human tissues. The PerfusionPal model will be advanced by introduction of peptide hydrogels. Technological development and skills transfer will be driven by a postdoc. The hydrogel system will be incorporated into PerfusionPal and its effects tested experimentally, in perfused/unperfused conditions, against the current standard (Matrigel). We will identify optimal synthetic matrix components required to functionalise the 'naked' hydrogel, applying methods developed during Merry's current NC3Rs-funded project to generate gels with the preferred combination of physical and biochemical attributes to support the luminal A breast cancer model. The postdoc will visit Merry's lab to learn how to make hydrogels and apply this technology in Aberdeen. Cell viability assays will be performed, followed by live/dead Calcein AM/PI imaging. Further imaging will assess cell distribution throughout the hydrogel and identify changes in cell organisation due to the components selected to functionalise the hydrogel. A PerfusionPal platform, currently used by Speirs NC3Rs student in Leeds, will be relocated to Nottingham when the studentship ends (9/19). The system will be tested at both centres to ensure reproducibility. A free hands-on training workshop will introduce and promote uptake of the technology to the wider scientific community, with bursaries provided for ECRs.

The project will focus on the reduction and partial replacement of mouse models for the most common subtype of breast cancer, Luminal A, delivering a humanised 3D vascular perfused breast cancer model which will provide researchers with an alternative model for studying breast cancer biology and in therapeutic drug screening. This technology has developed from two independently funded NC3Rs projects, to meet demands from scientists for physiologically closer in vitro platforms to model human breast cancer. We are aware of this demand through workshops we have run in the past aimed at introducing 3D culture models to the breast cancer research community. Our project aims to replace costly and time-consuming breast cancer PDX models. There are profound limitations in the ability of these to mimic human-specific features of breast cancer. This includes general differences between human and animal physiology and metabolism, compromised tumour microenvironment with human cells frequently replaced by mouse cells, and favouring establishment from more aggressive breast cancers. Discrepancies between metastatic patterns in PDX compared with those of the patient from which they originate are also apparent. Furthermore, take rates of 21% in SCID/Beige and 19% in NSG mice, with stable, transplantable PDX generated from just 2-18%, have been reported (https://www.ncbi.nlm.nih.gov/pubmed/23737486). Excluding animals needed to maintain PDX through serial transplantation, this single study used >1416 mice. Regardless of these limitations, PDX use is showing no signs of slowing; a PubMed search (25/1/19) showed an exponential rise in publications using breast cancer PDX models, from 1 (2012) to 53 (2018). PDX are now widely accepted as the model of choice for translational research in both academic and commercial sectors and with the continued increase in numbers of publications which use breast PDX, rodent use will continue to increase. Taking the publication used above as an example I each of the 53 studies used the same numbers of animals (1416), this would equate to 75 000 animals. If our model was adopted by just 20% of academic labs, this trend could be reversed significantly. While our figures exclude animals used in commercial sectors, recent data from Crown Biosciences, evaluating the numbers required in the design, analysis and application of PDX in mouse clinical trials, indicate 26127 mice from 2883 different treatments, an average of 9 mice per treatment group https://www.biorxiv.org/content/biorxiv/early/2018/09/24/425256.full.pdf). If used in these types of preclinical drug screens, the model may help identify those agents more likely to have clinical impact to take forward in vivo and ultimately, clinical trials, reducing attrition through acting as a first pass screen to minimise the numbers of rodents used. Importantly, the technology used to create this model is already commercially available which removes a potential barrier to its implementation by others. Our combined technology offers an effective, alternative, humanised model, with good potential for reduction and partial replacement of mouse PDX models. We expect to see uptake of this model, not only in our own Institutes, but also more widely, changing the way in which scientists think about designing and implementing experiments. By holding a hands-on training workshop, we will introduce the technology to the community to encourage its uptake particularly by early career researchers. Finally, as university academics, we can incorporate this thinking into our undergraduate lectures/tutorials, and offer undergraduate projects using this technology, embedding the 3Rs concept in the minds of next generation at an early stage of their higher education, providing a 3Rs legacy.