IMI Programme Office: Why was a project like yours needed in the first place? Which challenge were you trying to address?
Gerry Davies: Tuberculosis (TB) is the single biggest killer amongst infectious diseases, and for quite some time there has been very little development of new drugs. One particular challenge is that not only do we not understand how the drugs ultimately kill the organisms in patients, we also have to use the drugs in combinations, often quite complex combinations of 4-7 drugs. That poses a huge challenge for drug developers who are used to developing one drug at a time. The other problem that we have is that there is a lack of validated predictive models for TB, either in the test the tube, or in animals, and so we’ve had to revisit that to try to understand how well those preclinical models work and to compare them with clinical data. That’s broadly speaking the challenge that we set out to address.
Justin Green: From a pharmaceutical perspective, all this is linked to the risk of failure happening very late in the drug development cycle. When you have good predictive models you can take compounds that you know are going to have a good chance of success into phase one clinical trials. Without this work that we’ve done in PreDiCT-TB, you wouldn’t have the ability to be more certain that there is a higher probability of success for those drug-like molecules as you take them into the clinic.
IMI Programme Office: What were the most important project achievements? Which ones are you most proud of as project coordinators?
Gerry Davies: The biggest single achievement has been that we found a way to quantitatively evaluate the contribution of different preclinical models in terms of how predictive they are in planning clinical trials. To do that, there were several things that we had to succeed in doing. One of them was creating a completely new database of preclinical data with many different preclinical models, from simple experiments in broth through to animal models. We compared that database with a complete re-evaluation of all the clinical trial data that’s been published. We also tried to obtain as much of the individual patient data from those trials as possible so that we would be able to compare the two things.
Perhaps most importantly, we developed new biomarkers, new tools which enable us to understand how the drugs work in those different preclinical systems. And we took a mathematical modelling approach to integrating that information, putting it together in a way that can be used by drug developers to predict what will happen in clinical trials. So we’ve produced a series of tools which drug developers could use to increase the probability of them choosing the right drug-like compounds and getting them through to phase 3 clinical trials successfully.
Justin Green: From my perspective, two things in addition to what Gerry said. We’re proud of the way in which everyone has collaborated openly and together in order to try and get through a massive bottleneck in TB. It’s been amazing to see people across the different disciplines work together, not just doctors and scientists but also engineers and then also industry with small and medium sized enterprises (SMEs) and academics. That’s not something that was a given at the beginning of a project. Secondly, actually seeing physical things that have come out of this project and seeing them being used in the company I work in. There are new things that exist because of this project.
IMI Programme Office: How are the tools and learnings developed in this project already being used in the industry and academia? And are they already speeding up the development of TB drugs?
Gerry Davies: The adoption of some those technologies and techniques has been strong, certainly among the partners in the consortium, and in some cases outside the consortium. To give an example, the molecular bacillary load assay that was developed by University of St. Andrews and University College London, has been used extensively by partners inside the projects and has been evaluated in a multi-centre study in Africa outside of the project. [This is a new method which offers a much faster way to diagnose tuberculosis via the ribosomal RNA, without the time-consuming process of culturing TB bacteria]. It has the potential to really harmonise preclinical and clinical development and give us much better predictions about the activity of drugs.
IMI Programme Office: It sounds like your project really achieved a lot. Would you say that it has been transformative?
Gerry Davies: It certainly changed people’s attitude towards drug development in TB. Some of the techniques that we brought in are simply best practices in modern drug development and we have applied them to a field in which perhaps that’s been lacking for decades. We can certainly say that some of the tools that we developed – some of the statistical modelling tools and some of the technologies in the lab – have the potential to be transformative.
IMI Programme Office: Has your project changed the standing of Europe when it comes to this field of research?
Gerry Davies: My view is that this has been a critical investment for European researchers and companies in this particular therapeutic area. It helps us to build collaborations within this sector outside Europe, particularly in the US, on an equal footing.
IMI Programme Office: Have the tools and learnings developed in the project contributed to the reduction of animal use in research?
Justin Green: We collaborated on optimising the design of experiments and making sure that we are using animals in the most effective way to get the best results. Thanks to PreDiCT-TB, we now have the ability to design the same experiments better, with less animals.
We have also tried very hard to look at which are the best animals to use for drug discovery and development work and really try to challenge whether particular species are actually giving us valuable information. That’s very important because it means you are not doing experiments that are giving you unhelpful information.
IMI Programme Office: What was the impact of this project on industry? How have you benefited overall from this project?
Justin Green: There a number of layers to that. The creation of a Europe-wide network that industry scientists could look towards for expertise, both within and beyond the project, is incredibly important. And then there are the more tangible benefits: the fact that we were able to take part in scientific research with a large number of scientists and access a large amount of data that is aligned with the kind of project ideals and end points that we’re interested in for our own efforts. Finally, as already mentioned, there are tangible things that now exist within certainly the company I work at, in terms of concrete tools, which we wouldn’t have been able to develop on our own.
IMI Programme Office: And what about the academic community – how did you benefit?
Gerry Davies: Some of the academic investigators who came into this project were experts in TB but had not worked in drug development before. This project has given them real experience and engagement with industry: an understanding of how their models and skills could be used to enable better and faster drug development. Those relationships have been a really important learning experience and have laid the foundation for a network of TB drug development expertise in Europe, which didn’t really exist before.
IMI Programme Office: You also had some SMEs in the project. How did they contribute to the project?
Gerry Davies: We had two SMEs in the project and they had different roles. As an example, Zf-Screens BV from the Netherlands were involved in the development of the zebrafish high-throughput screening model. They engaged very positively with developing a tool that enables them to screen very large numbers of drugs at lots of different concentrations, and combinations of drugs, very efficiently. They developed these methods rapidly during the project to the extent where they now have a tool that may become very valuable in the early phases of preclinical development.
IMI Programme Office: Would all this have been possible without the public-private collaboration brought by IMI?
Justin Green: Perhaps it could have been done, but it wouldn’t have been funded. We already have a large number of scientists, at GSK and in other companies, working on TB, but all of this was done in addition to what those companies were already planning and has given scrutiny to what they were previously doing. IMI is uniquely placed at that pre-competitive space where you’re going to try and really challenge the bottlenecks. This is definitely what this project has been able to do.
IMI Programme Office: What happens now to all the tools generated within the project? Will they be sustained and accessible to scientists outside the project?
Gerry Davies: We managed to find ways to make publicly available all of the key resources that were generated by the project. In particular, the databases of preclinical data and clinical data have permanent homes that are funded probably for at least the next decade. These data are a key resource for the wider TB drug development community which I hope will be added to and expanded in the future. On top of that, some of the technologies that we’ve talked about, such as the molecular bacillary load assay, have been provided in a way that any external scientific group can work with in a standardised and hopefully reproducible way. Finally, a lot of the code base for many of the modelling and simulation tools that we’ve developed will also be available through publicly accessible websites – so that’s there for people to take, modify and integrate into their drug development efforts as they see fit.
IMI Programme Office: Anything else to add?
Gerry Davies: It’s hard to imagine what other funder would have supported a project of this kind of ambition and scope in this particular area. What I am most proud of is the way that all of the scientists involved in the consortium, some of whom are very eminent, embraced that vision of drug development that we were putting forward and have made it a reality. Collectively they have done something that is very novel and hopefully will set a benchmark for the way that we think about evaluating drug development in tuberculosis in the future.