Development of a 'humanised' model for renal cancer

Immune checkpoint inhibitors (ICPIs) are a type of cancer treatment that boosts the body's immune system to fight cancer. ICPIs are revolutionising cancer treatment as they can lead to years of control of cancers in up to 50% of patients. However, unfortunately less than 50% of kidney cancer patients benefit from treatment with these costly drugs (1year treatment up to £90,000), yet up to 80% of patients experience side effects relating to 'over-activation' of their immune system which can attack their normal organs. Therefore, there is an urgent need to accurately predict which patients will benefit from treatment whilst providing alternative treatment options for patients who do not.

Normally such studies can be performed using laboratory models to prevent patients being unnecessarily exposed to sub-optimal drugs (and combinations) and their side effects, but models for testing ICPIs in kidney cancer are limited as they lack immune cells that work like a human immune system. However, a new genetically engineered mouse has been developed (called a NSG MHC I/II knockout (KO) mouse) which can grow a human immune system when immune cells from patients' blood (called peripheral blood mononuclear cells; PBMCs) are injected into the mouse, to generate a so-called 'humanised' mouse.

The goal of my project is to develop a new 'humanised' mouse model for kidney cancer.

The specific objectives are:
1) To generate 'humanised' mice by injecting PBMCs from kidney cancer patients into NSG MHC I/II KO mice, plus transplant RCC and normal kidney tissue from patients having surgery to remove their kidney cancer to generate a new 'humanised' mouse model for RCC
2) To test the new model to determine the extent to which it reflects patient responses by comparing immune and cancer responses in kidney cancer patients vs 'humanised' mice treated with ICPIs
3) To provide preliminary (pilot) data for a future large grant (called a Clinician Scientist grant)

To do this I will inject PBMCs from kidney cancer patients into the new strain of mice to create 'humanised' mice. I will also implant kidney and normal kidney tissue (obtained when those patients have surgery to remove their kidney cancer) into the mice. This will create the new 'humanised' model for kidney cancer which has both the immune system and kidney cancer that the patient has (acting like an 'avatar'). We will do this for up to 30 patients as some work will be needed to optimise the procedures to make sure that the PBMCs, cancer and normal kidney cells grow properly in the mice. I will then test the model and see how well it mirrors patient responses to ICPI treatment. The mice will be treated with ICPI for 4 weeks and their immune, tumour and normal tissue responses will be compared to the same responses in kidney cancer patients who receive the same drugs in the clinic as part of their normal treatment.

This data will form essential pilot data for my future Clinician Scientist application which will aim to fully assess and validate the model with arrange of different drugs for a longer time period. It could then be used to identify the most effective novel therapies and combinations of different drugs (with the least side effects) for kidney cancer patients which can then be tested in clinical trials, rather than testing all treatments and combinations in patients as currently happens. This would save significant time and cost. It could also identify markers from blood or tumours that could predict which patients are likely to respond or have very serious side effects to new treatments so those patients who would not benefit from a particular treatment could be spared treatment or side effects and offered an alternative treatment. The model will also be useful to many academic and commercial researchers worldwide who can use it to streamline drug development (as above) but also to investigate how the immune system interacts with kidney cancer (and vice versa).

Unfortunately, pre-clinical development of novel drugs for renal cancer (RCC) is significantly restricted by a lack of pre-clinical models with a functional human immune response. Therefore, the development of a preclinical model of RCC with a competent 'human' immune system is a critical unmet need for RCC, especially given the plethora of immune checkpoint inhibitors (ICPIs) targeting the immune system which may be of benefit to RCC patients.

Recent advances have shown that knocking out HLA I and II from in an immunocompromised preclinical model (lacking T-, B-, NK-cells, and HLAI/II) allows the development of a functional 'humanised' immune system using peripheral blood mononuclear cells (PBMCs) from patients. This translational project aims to use this novel preclinical model to develop a novel 'humanised' preclinical model for RCC. The specific objectives (in order of priority) are:

1) To generate a novel 'humanised' preclinical model of RCC using with allogenic and autologous PBMCs, renal cancer and normal kidney tissue from RCC patients
2) To determine the extent to which this 'humanised' model reflects patient responses to immune checkpoint inhibitors (ICPIs)
3) To provide pilot data for a future Clinician Scientist application

RCC and normal kidney tissue from RCC patients will be implanted under the kidney capsule and subcutaneously in the 'humanised' preclinical model above. Preliminary validation of the model will then be achieved by comparing blood, RCC, normal tissue, and toxicities in the 'humanised' model vs RCC patients receiving ICPIs (at baseline and 4 weeks) using mass cytometry (CyTOF), Luminex based multiplex cytokine assays, and immunohistochemistry.

The model would be useful to identify the most efficacious (and least toxic) novel immune-directed therapies for RCC patients for clinical trials and will be useful to researchers for drug development and investigating the interactions of RCC with the human immune system.

The proposed model would represent a significant step forward in the field by providing the first immune-competent model for renal cancer (RCC). The project would mainly benefit the following groups: researchers (academic and commercial), and patients.

Researchers (both academic and commercial) within the Saeb-Parsy Lab, our Cambridge-based collaborative renal cancer network (Cambridge Renal Cancer (CamRenCan); a group of over 40 clinical and basic science researchers with an interest in RCC), in addition to those worldwide will benefit from the development of a new RCC model with a competent 'humanised' immune system in 3 main ways.

Firstly, this will be the first model of its kind in the RCC field and, once validated (within 3 years), the model will be available as a 'tool' for use in collaborative projects via The Cambridge In Vivo Assessment platform. Examples of potential uses include: performing pre-clinical testing of novel drugs or interventions for efficacy and toxicity; to investigate the effects of the immune system on RCC or vice versa and how manipulation of either can affect RCC growth, the immune system, toxicities, or the well-being of the host; and/or to identify immune and/or tumour-related biomarkers. It is expected that the model will generate significant interest from commercial entities (biotechnology and pharmaceutical companies). The Seab-Parsy Lab and CamRenCam already have established commercial collaborations for a range of projects from basic science (with joint PhD studentships and postdoctoral fellowships) through to joint investigator led, industry funded, clinical trials, with full associated support from The University of Cambridge which will allow the model to reach its full potential in both the academic and commercial sectors.

Secondly, tissue samples generated during development and testing of the model will be available via The Cambridge Biorepository for Translational Medicine (within 6 months). Whole blood, plasma, PBMCs, tumour, normal kidney and mouse tissue will be available and could be used in a range of projects. Eg to validate basic science findings in human (or mouse) samples, or to investigate expression of proteins or pathways of interest in clinical or mouse tissue. Tissue samples are already planned for use in various local Clinician Scientist, postdoctoral and PhD projects as listed in the 'Academic Beneficiaries' section.

Thirdly, all methodology used in this project will be fully published in peer-reviewed journals (within 3 years) thus will be available to any researcher or commercial entity. This knowledge will be critical for other scientists developing similar models for other cancer types or in other disease areas where 'humanised' models are of significant interest. Mr Saeb-Parsy (Collaborative partner for this project) leads the UK Humanised Preclinical Models Group (including organising annual conferences) where this knowledge can be disseminated via the group's networks of basic, clinical and commercial collaborations.

Finally, patients will benefit as development of this immune-competent model will enable in depth testing of drugs, including being able to identify the best drugs and combinations in the preclinical setting such that only the most efficacious and least toxic drugs are selected for clinical trials. This will save time and mean that patients can get access to the best drugs more quickly than without preclinical testing in this model. Currently, comprehensive immunological assessment can not be achieved until drugs enter clinical testing, potentially exposing patients to suboptimal drugs or combinations and unexpected toxicities. Patients will also benefit from early development of biomarkers able to select which patients are likely to respond, or are at greater risk of toxicity, such that alternative options can be offered which are more likely to work, or follow up regimes can be tailored to the risk of developing toxicity.