The Concept

General introduction of the Building Bridges concept

Presenter Talk: Barry Hardy

Key points from discussion

Past projects in which we all have been engaged in generated many methods and research results. The organization of data is a challenge, as is making the results actionable. Modelling results are difficult to share and transfer into a more practical way. Also the regulators are acting today on procedures based on old technologies.

Goals:

Define use cases how e.g. clinicians with patients could use the existing data for the benefit of their current or future patients. The question is about the clinicians needs and to define how the data could be organized and visualized. This also includes the workflow of data that would be helpful for clinicians. We should define something simpler, than what is currently done as standard practice. Define a picture based on the scientific data / modelling that could help a clinician.

Presentation: Introduction Perspective

General introduction on the usage of stem cells

Presenter Talk: Tomo Saric

Key points from discussion

It is important to emphasize the differences between adult stem cells (such as mesenchymal stromal cells, bone marrow derived stem cells) and pluripotent stem cells (iPSC, ESC) in terms of their uses and translational potential.

There is an opportunity to develop clinically relevant human cell-based disease models based on the increased knowledge and insight we gain from research on patient-specific induced pluripotent stem cells (iPSC) and their derivatives, such as cardiomyocytes, neurons, hepatocytes and other tissues specific cells affected by the disease.

Today's insight from these models can reduce time for preclinical drug development, increase the safety of future drugs and lead to application of patient targeted pharmacotherapies for a number of disorders in different medical disciplines. Based on this information, safety related therapy improvement could be aimed at.

In contrast to the potential of human iPSC-based in vitro disease models, the use of iPSC-derivatives for regeneration of damaged tissue that occurs for example as a consequence of infarction or neurodegeneration, will further depend of preclinical research results and overcoming of obstacles that currently prevent broad use of cell-based therapies in the clinic.

Some of these obstacles are the lack of consistency and reproducibility in production of pure therapeutic cell populations at high clinically relevant yields, low survival rate and functional integration of cells after transplantation, risk of teratoma formation from contaminating pluripotent stem cells, and risk of immune rejection of cells carrying antigens foreign to the recipient’s immune system.

Despite a few clinical stage I trials in which iPSC-based therapeutic strategies are being investigated for their safety and efficacy in spinal cord injury and macular degeneration, broad clinical translation of this therapeutics is still not possible without concerted preclinical multidisciplinary research.

Use case idea:

Using stem cell biology, in-vitro disease modelling to better understand the pathophysiology of genetic diseases, develop new and more efficient personalized drugs, and to identify drugs that may exert toxic side-effects in patients early in the process of drug development to protect patients as well as to reduce the drug attrition rate and costs associated with this. (EBiSC meets OpenTox)

Presentation: Stem cells and their applications

Use of stem cells to model Alzheimer's Disease and other dementias

Presenter Talk: Selina Wray

Key points from discussion

Selina referred to a special opportunity for stem cells as they can be developed into special neurons that are critical in the disease, be it Alzheimer or Parkinson's or other neurological diseases.

Neurons can be now developed in a dish. Such in-vitro models have been shown react similarly and therefore to complement the in-vivo system that is used. This can also contribute to the 3R initiative. Organ on a chip can be developed from the same basis. Detecting pathologies is possible, but they might be part of the normal aging process. The challenge is to identify the early disease state versus a state of normal aging process.

Accelerating of the nurturing of the neurons through organ on a chip is currently investigated. There are various approaches and studies to do that.

The 1946 cohort in UK offers an exciting opportunity to investigate some of the questions of our interest. Especially the 500 sub set for which a richer information base has been collected.

The area of visualization might be of key interest as well. Safety assessment could benefit very much from good image analysis that would extract features from the pictures that would be actionable by the physicians.

Could one detect an early neuronal loss that is pathological? The changes could be an early warning signal for onset of Alzheimer. Today, one does not yet know how to analyse the MRI pictures.

Presentation: Stem cell models of dementia

Identifying adverse effects of substances by mode-of-action analysis

Presenter Talk: Costanza Rovida

Key points from discussion

Selina referred to a special opportunity for stem cells as they can be developed into special neurons that are critical in the disease, be it Alzheimer or Parkinson's or other neurological diseases.

Today, we need reliable results and reproducible test methods to test hundreds of substances. Cells should work as expected - question of quality.

Quantitative in-vitro and in-vivo extrapolation (QIVIV) in chemical exposure situations are challenging, because they are mainly unknown.

Drugs can interact with different targets. Pathway idea could help us to verify the outcomes that we want and also those we do not want. With a bioinformatics approach some adverse outcomes could be predicted through variations in the cell biology.

Specific expressions of genes also play a role and they have to be considered. This must be also integrated in the studies. The background knowledge regarding the drugs effect already exists. The message to the clinician will be at the end: this drug should be prescribed to this kind of person. This would be a start for a more rational diagnoses and treatment e.g. in inflammatory conditions. Read-across approach will be key for the future work, which is already on going.

Potential next step would be to provide clinical evidence for verifying the human evidence for the alternative approaches. Another option would be combination and comparison of the various models.

The US project (ToxCast) could provide justification for the read-across approach, i.e. to understand the similarities among the chemicals.

Presentation: Can we replace animal testing by extrapolating research results from stem cells?

Computational Challenges of Personalised Medicine

Presenter Talk: Maria Chatzou

Key points from discussion

Computational performance is key to the translational medicine approach. High performance computers are in operation today, however the software technology is still catching up.
Same algorithms on different operating systems produce different results. Standardization is surprisingly a big challenge. Regulators are just realizing that they have to talk to scientists.

Connection between phenotype and genotype in regards of how a drug is working in different people. It would be nice if in the package insert would be written that if you have this or that genetic profile this drug has this probability to work in you. On which is this dependent? It could be genetics, environmental aspects, gender, age and others. There are reasons behind this fact.
Heart attack drugs are mainly tested in men. So there is also this bias in the data.

If we have access to patient classification, we could reproduce the group results in-vitro. Painkillers responsiveness is one interesting field. If we could understand how the drugs work (differences in metabolism and absorption), we could understand which drugs could be used and in which sequence.

A clear goal is to clarify why one drug is affecting different people differently. We have to find a means to find out why this is the case. If we find this methods, we will also know why this is the case.
Reprograming liver cells which were gained from the stem cells and see how they respond could be a way forward.
A good argument could be to show that the human models are more responsive than the animal models. This could help us to convince more people.

Stem cell banking and possible usages and ethics aspects

Presenter Talk: Glyn Stacey

Key points from discussion

Communication with the public is key for establishing successful research programs. This has not been done very well at times in the past and has lacked researcher commitment.

Changes in the genes do not have to be associated with some serious condition. Genes may show response to environmental conditions. In some cases we do not know what has caused the change.

In stem cells one can observe so called «spontaneous differentiation».
It is also challenging to create an homogenous cell culture and thus we are faced with scaling issues. Variation in cell metrics is significant issue between labs e.g. population doublings, confluency. People in various labs could produce different cells although they use apparently the same procedure.

Summary of challenges:

Currently there is little standardisation as there is also little real work on reference materials which may need to be aimed at analytical methods rather than the cell therapy products which are complex and may well require bespoke controls for each individual product.
The storage of data is also not trivial. We have to deal with contamination of data in databases such as microbial sequences in human genetic data. Metrics on images are also not yet well defined.
Establishing functional assays that would tell you something about the potency of a stem cell therapy product but this may not always be predictive of efficacy due to individual variation.

Making sense of the genetic changes between the cells and their association with particular diseases.
Ethical issues regarding dealing with tissues and also with regulation of patient specific information.

Regulatory acceptance of new methods

Presenter Talk: Rex FitzGerald

Key points from discussion

Data relevance: How do we clarify what the test is actually measuring? There is a major shift from animal testing to evidence based in-vitro testing. Looking for pathway analysis based framework that has been developing over the last 15 years.

OECD has adopted this framework: Using pathway analysis to define the relevance of test data. This also includes data on stem cells. The question is how to interpret the data.

Quality in in-vitro testing is still at its beginning to be defined. Today there is lack of guidance on this.

So there is some guidance for the patients we are looking at, but the stem cells data is very unstable, there is lot that is potentially skewing the results. This has to be addressed to increase the reliability of the stem cells.

There are several peculiar test systems, which might or might not be relevant to the actual human disease.

Evidence based toxicology - one has to start with the question that one tries to answer. AOP framework - to embrace this concept is actually quite difficult. One start with a chemical initiating an event at a target, triggering successive downstream key events which lead to an adverse outcome are being triggered, e.g. lung cancer or skin sensitisation.

To link the cause and effect is the main difficulty. A key event can be shared by various AOPs. There could be more than 20 chemicals, stress, and temperature variations leading to it. Key events are shared between various toxicity pathways, shared between species and life stages. This will be a question of ontologies.

Regulators are currently struggling with the question how to weight the results from the AOP results. They look for reliable and relevant data. The tic box approach is not anymore valid. Today, they have to look at in-vitro data and make a decision without any help of animal testing data - this is the new toxicology.

It is good if you can tie a key event with omics. There are single points, that can be graphically depicted, but there is no knowledge yet. It is unclear how these things are connected. In-vitro systems are one tool in the whole systemic biology picture and should not be oversold.

We need examples for evidence: non-animal assays that can predict an adverse event. Pharma companies are currently doing this. US have put massive effort behind these questions. They are asking and answering these questions. Also they have the critical mass and they are delivering. As Europe is still fighting fragmentation, US are many years ahead.

Next big issue is if we can include question of potency in in-vitro experiments. How do we validate it? A Givaudan researchers have proven, that in-vitro results can be more predictive of human skin sensitization than mouse studies.

Presentation: Regulatory acceptance of new methods