Cancer Models Forum

Posted by Repositive, February 2019

How will drug development need to change to bring personalised cancer treatments to patients?

Key takeaways from our webinar - Cancer, genomics and personalised medicine: Modelling oncology’s future

Researchers have made big strides to advance new cancer therapies to the clinic and have doubled the cancer survival rate in the UK from 24%-50% in the last 40 years. With the discovery that the immune system can be harnessed to fight tumours, there has been an explosion of new immunotherapy treatments heading into the clinic over recent years, which have made an important contribution to this success.

One of the main challenges oncology now faces is tumour drug resistance, where a tumour acquires mechanisms over time through its evolution to overcome a treatment’s mode of action. As a result, a tumour no longer responds to the initial therapy, which usually results in patient mortality. Therefore, to further improve survival rates, cancer drug development is accelerating towards an era of precision medicine where therapies are tailored to the specific genomic and mutational profiles of an individual’s tumour to help overcome any resistance mechanisms a tumour has acquired. In addition, through regularly assessing a patient’s response to a treatment through liquid tumour biopsies, clinicians hope to detect signs of resistance developing earlier and change patients to another suitable therapy.

But how will the traditional drug development process need to adapt to develop personalised cancer treatments that physicians and patients are confident will be both safe and effective? It’s a question that is raising discussion amongst the translational oncology community, so we invited three industry professionals to share their opinions in a recent webinar.

Our webinar panel included: Professor Gordon McVie (Clinical Research Adviser, FIRC Institute of Molecular Oncology, Milan and Founding Editor of, Dr Stefan Jellbauer (Technical Liaison for Translational Applications, Mitra Biotech) and Danyi Wen (President & CEO, Shanghai LIDE Biotech)

View the full webinar recording

Here are our key takeaways from the panel discussion:

Better preclinical cancer models are needed for more accurately predicting patient response

43% of our webinar viewers agree that the biggest challenge currently facing personalised cancer medicine is the need for more physiologically representative models that enable us to more accurately predict patient response during preclinical drug development. As Dr Stefan Jellbauer highlighted, we not only need truly translational cancer models, but we also need to understand which biomarkers to measure in order to accurately predict patient response in the clinic.

Over the last few years, there have been a number of innovations developed that are already being used by biotechs to improve the predictability of preclinical cancer models. New technologies to the market include:

  • CANScript™, a preclinical platform where drug efficacy is assessed on fresh tumour samples that are cultured in a physiologically representative tumour microenvironment. By utilising machine learning, CANScript™ can convert a series of endpoints from a multi-dimensional phenotypic analysis into an M-score, which provides a more accurate indication of clinical response.

  • MiniPDX Pharmacodynamics Research Platform, where a ‘mini’ tumour xenograft can be created in a much short timeframe compared to traditional PDX models by simply inoculating immunocompromised mice with tumours cells using a specialised capsule. Cancer drugs are then administered to the mice systemically for seven days, after which the tumour cells in the capsule are removed and monitored using a variety of multi-omics analyses to assess treatment efficacy.

But what other technologies are on the horizon for further enhancing the reliability of preclinical cancer studies? Artificial intelligence is generating a lot of excitement as it is set to rapidly advance drug development by enabling researchers to draw more complex insights from the integration of multiple, large datasets (such as multi-omics analyses across thousands of patient tumours). Not only this but 3D organoids are showing promising results for recapitulating tumours ex vivo, including more accurately modelling tumour heterogeneity and the tumour microenvironment, as well as assessing drug efficacy. With the ability to quickly derive organoids from only a small tumour biopsy sample, they also hold great potential for accelerating patient-specific drug testing to the clinic.

Co-clinical trials will play a valuable role in stratifying cancer patients for the development of new therapies

Thanks to the recent technological advances, we’ve been able to draw valuable insights into the molecular and genomic characteristics of different types of cancer. However, difficulties in accurately modelling patient response during preclinical development, have delayed translation of these insights into new, effective treatments. To help overcome this, researchers introduced the co-clinical trial approach, where preclinical tests using trial participants’ tumour tissue are conducted alongside early phase clinical trials (usually Phase I/II). By assessing drug responses using assays or preclinical models which can be run faster and in parallel to clinical trials, the results can be used to directly inform whether a drug candidate advances to later phase clinical trials in a much shorter timeframe, saving drug developers billions of dollars by preventing failed Phase III trials.

One particularly valuable application of co-clinical trials is helping researchers to understand possible tumour drug resistance mechanisms in preclinical models. These results can be used to stratify patients for clinical trials based on whether their tumour harbours these mutations and may ultimately lead to mechanisms for identifying patients most likely to respond to individual therapies – increasing the chance of treatment success.

During the webinar Danyi Wen highlighted how co-clinical trials are already providing powerful insights for cancer drug development. In an interesting case study, a customer of LIDE Biotech was interested to see if a breast cancer drug would work for gastric cancer patients. However, they didn’t know if the target gene was also a driver for gastric cancer. Utilising LIDE Biotech’s innovative MiniPDX Platform, the company was able to assess what percentage of gastric cancer patients would be responders to the drug, helping them to decide if it would be suitable to advance into clinical trials for this indication.

Combination therapies hold the most potential for overcoming tumour drug resistance

As Professor Gordon McVie stated, chronic myeloid leukeamia is a rare example of a cancer that will respond to the first line of therapy for a significant period of time without developing resistance. Unfortunately, despite our best efforts this simple treatment strategy hasn’t been easy to replicate across other cancer types. So, will monotherapies have a place in the future of cancer therapy? To overcome tumour drug resistance, Professor McVie explained that we need to look at every component of a signalling pathway, not just individual genes. By considering pathways as a whole and the complex interplay of all the elements, we can develop combination therapies to knock out multiple key proteins simultaneously, making it intrinsically harder for a tumour to develop resistance. It looks like it could be time for monotherapies to step aside and let combination treatments take to the spotlight over the coming years.

However, the current approach for selecting combination therapies to advance into clinical trials isn’t particularly well-informed and, as a result, multiple different combinations are typically tested. In preclinical development a treatment combination is usually only compared against each of the compounds individually and if it demonstrates greater efficacy than the individual drugs, then the combination is considered successful and progressed into clinical trials. However, for researchers to make more informed decisions about whether a drug combination will show promising results in clinical trials, it is important to understand the degree of synergy between the compounds. To assess this, combinations should also to be tested against other reported treatment combinations. By investing in high-throughput screening and running side-by-side comparisons in the preclinical stages, drug developers could use the results to select the best combination to move into clinical trials.


As we enter an era of precision cancer medicine, the traditional drug development process needs to change to ensure new treatment combinations tailored to the mutational profile of patients’ tumours are accelerated to the clinic. From our thought-provoking panel discussion, it’s clear that we not only need to develop better preclinical cancer models, but we should look to use these models alongside Phase I clinical trials to make more informed decisions and dictate the direction of later phase clinical trials. However, it’s not only the next generation of new translational models and technologies that will be important - cancer drug developers also need to have access to the best models for testing, which has previously been difficult due to model inventories being held separately by different providers. But with the creation of global cancer model directories combining preclinical models from CROs around the world, such as the Cancer Models Platform, biopharma researchers can now more easily source the preclinical models they need, boosting us towards precision cancer medicine.

To keep up-to-date with the latest developments and hot topics in translational oncology, sign up to our mailing list!

Sign up to mailing list button