What is the best approach to neglected tropical disease drug discovery?
Matthew Todd discusses the importance of openness and collaboration in accelerating NTD drug discovery
Matthew Todd is Associate Professor at the School of Chemistry, The University of Sydney (Australia) and is founder and current leader of the Open Source Malaria (OSM) and Open Source TB (OSTB) consortia, in addition to co-founding the broader Open Source Pharma movement. In 2011 he was awarded a NSW Scientist of the Year award in the Emerging Research category for his work in open science and in 2012 the OSM consortium was awarded one of three Wellcome Trust/Google/PLoS Accelerating Science Awards. In this interview Todd discusses current drug discovery models towards the development of novel neglected tropical disease (NTD) therapies, if there is room for patent protection in NTD R&D and finally the benefits of open source science.
Those affected by neglected tropical diseases (NTDs) represent some of the poorest in the world. With zero margin for profit, how can we stimulate R&D towards new drugs, vaccines and diagnostics?
There’s no magic bullet. Where there is no prospect of cash profit, we are naive to look to the current incarnation of the pharmaceutical industry. We are already seeing major pushes from other sectors to stimulate investment - possible government-backed funds for antimicrobial resistance, major investments by philanthropic agencies, prize funds like The Longitude Prize, open screening platforms like CO-ADD or very significant resources leveraged from non-profit wings of the pharma industry. Everything we do will cost money (which we’ll need more of) and the key is to ensure that whatever money is spent is spent most efficiently and wisely. We’re not currently doing that well.
In 2015 the pharmaceutical company KaloBios made the news when it announced its intentions to register key Chagas drug, benznidazole, with the US FDA and sharply increase the price of this decades-old treatment. In general, do you believe that sufficient controls are in place to regulate the price of drugs for NTDs?
Controls make me uneasy, in that they will be subject to argument and vulnerable to lobbying. Market forces can be powerful - where very large agencies like WHO become a major client, then the price can be negotiated. In other cases controls may be unavoidable in case of deadlock. I would prefer a competition of models in how we discover and develop medicines. Raising the price of a drug, if it’s legal, makes complete sense financially in our current system. What’s lacking is a large-scale competing model that can step in and undercut such behavior.
Many argue that patent protection is needed in order to secure investment from pharmaceutical companies, in addition to their expertise and collaboration. Are you in agreement and if not, why?
I’d agree. We should not expect any investment from the pharmaceutical industry without the possibility of managing intellectual property, and this is clearly fine. For projects that do not protect intellectual property, one can nevertheless expect major inputs from the private sector in terms of expertise, compounds, assays and so on - we have been fortunate enough to receive these contributions from a large number of first rate scientists in pharma. More typical in-kind contributions from pharma tend to be monolithic projects, such as the screening of libraries vs. a given pathogen accompanied by the release of data. There remains huge capacity in the future for the most valuable resource - time. Specifically, the time of scientists in pharma to mentor not-for-profit projects. In some ways the legal sector is ahead of us, since there it is strongly encouraged that companies give a % of their resources to pro bono projects. One can easily imagine a pharma company in which scientists mentor students at colleges in the same city who are working on not-for-profit drug discovery. There are clear educational and public relations benefits there, as well as benefits to the progress of science.
In what ways do you think the patent system needs to be improved so that it both stimulates investment and the development of affordable NTD treatments, without stifling innovation?
The patent system works well in what it does - providing economic clarity and hence incentives to generate a return. Most patents, as documents, are masterful examples of obfuscation, and to improve them as objects we should require better language. Patent pooling looks to be working well for NTDs as a means of enabling faster research. My problem with patents is instead on the requirement for secrecy, in advance. Since we require that a patent is based on research that is not in the public domain, we poison the efficiency of basic research and collaboration, since we cannot talk to others about what we are doing. If we could retrospectively patent (if that is even meaningful) we could avoid this problem, and exert ownership rights (if we wished to) over something that was developed in public view, but such a thing does not exist.
To help coordinate NTD drug discovery efforts and reduce the duplication of research, recent years have seen the introduction of open source drug discovery platforms (e.g., Open Source Malaria) and data sharing frameworks. Have these models achieved success? What challenges are associated with discovery platforms of this type?
I’m biased. Since I started OSM with MMV a few years ago we have received an enormous number of inputs from people who are not funded by the core grant (which only supports one person). These inputs have come from all sectors, ranging from groups in big pharma through to school classes, from advice through to molecule synthesis and evaluation. Success will be easy to judge as the first papers start to emerge. The challenges remain partly psychological (worry of working transparently in the public domain, which involves revealing doubt and negative data) and partly technical (best way to record data so that they are machine-understandable, and the best way to discover collaborators). But the efficiency gains are clear, and the positive responses from a broad range of scientists have been overwhelming.
Many scientists may feel dissuaded from sharing their data online due to the perception that it might reduce their chances of publishing the data in a high-impact journal at a later date, in addition to the fact that they may wish protect their intellectual property. Is this an issue and, if so, how might scientists be encouraged to contribute their data to shared drug discovery platforms?
Protection of IP is a deal-breaker, sure. If you want to patent, forget about operating open source. A few years ago there was a problem with publishing research that was already in the public domain, and a few of the more traditional publishers still don’t like it, but I think everyone else has grown up a little. We first published a rigorously peer-reviewed research paper describing an open source biomedical project back in 2011. Sharing data is an excellent way to extract more value from those data, both in terms of inputs from observers as well as reuse. The public would be scandalized if they knew how much publicly-funded research never sees the light of day.
Do you think that, in general, the open source model is well suited to taking drugs all the way to registration?
Yes, but I can’t yet prove it. We started open source pharma as a broad movement precisely to look at this in more detail. We’ve run three meetings thus far and have a draft paper that details how open source practices would impact the traditional processes of bringing a drug to market. Tata Trusts just supported the formation of an Open Source Pharma Foundation in part to help advocate some of these ideas. It’s important to note that it’s perfectly possible to take a drug to market without any protection of IP - the important malaria medicine ASAQ/Coarsucam is an example and it looks like fexinidazole will be another. When we combine such projects with an open source research mechanism (where everyone can see everything that is taking place) in a project of sufficient size that we overcome expected rates of attrition, then we’ll have an empirical test of whether we can ever use open source methods to take a medicine all the way through to market. I’m confident we’ll see this work, and I’m impatient to make it happen.
Follow Matthew Todd on Twitter @mattoddchem