Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
Title Page THE WINSTON CHURCHILL MEMORIAL TRUST OF AUSTRALIA
Report by
- Stuart Ralph - 2012 Churchill Fellow
The Dorothea Sandars Churchill Fellowship to Investigating Open Source strategies for developing drugs against neglected parasitic diseases
I understand that the Churchill Trust may publish this Report, either in hard copy or on the internet or both, and consent to such publication. I indemnify the Churchill Trust against any loss, costs or damages it may suffer arising out of any claim or proceedings made against the Trust in respect of or arising out of the publication of any Report submitted to the Trust and which the Trust places on a website for access over the internet. I also warrant that my Final Report is original and does not infringe the copyright of any person, or contain anything which is, or the incorporation of which into the Final Report is, actionable for defamation, a breach of any privacy law or obligation, breach of confidence, contempt of court, passing-off or contravention of any other private right or of any law.
Signed
Stuart A. Ralph
Dated March 28, 2013
Page 1 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
Contents: Title Page ...................................................................................................................................... 1 Introduction ................................................................................................................................. 3 Executive Summary .................................................................................................................... 4 Programme ................................................................................................................................... 5 Main Body .................................................................................................................................... 6 Obstacles to malaria drug and vaccine development ......................................................... 6 A background to open source drug discovery .................................................................... 8 My fellowship and lessons learnt .......................................................................................... 9 Challenges to overcome in the open source drug discovery model ............................... 14 Conclusions And Recommendations ................................................................................... 18 References .................................................................................................................................. 20
Page 2 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
Introduction I travelled to India to investigate a new method for developing drugs based on open source philosophy and methodology. This approach is an alternative to the very expensive traditional closed model based on market driven motivation and carried out by big pharmaceutical companies. Open source drug discovery uses collaborative research and development in an open framework, with data, expertise and techniques shared without patent protection. India has recently initiated the world’s largest program for open source drug discovery, and I visited the lead principal investigator at the Central Drug Research Institute in Lucknow, Uttar Pradesh, as well as dozens of laboratories in four other major Indian cities that participate in the open source drug discovery project. I am deeply indebted to my Host in Lucknow, Dr Saman Habib, who not only provided inspirational mentorship in open source drug discovery, but was a patient guide and friend, helping my family and I to live so enjoyably in Lucknow. Sincere thanks also Dr Amit Misra, who so generously introduced me to pharmacology and toxicology, and for even more sobering and enlightening discussions on the plight of India’s poor. In Delhi, I was kindly hosted by Dr Asif Mohammed at the ICGEB, and by Dr Pushkar Sharma at the National Institute of Immunology, I am grateful for their kind assistance and warm welcome. In Mumbai, I was kindly hosted by Dr Gotam Jarori at the TIFR, by Dr Swati Patankar at the IIT Bombay, and by Dr. Abhay Chowdhary at the Haffkine Institute. In Pune, thanks go to my friend Dhanasekaran Shanmugam, who not only provided fantastic introductions into the world of synthetic and medicinal chemical chemistry, but very helpfully assisted in arranging our later travels in South India. In Bangalore, I am grateful for the very warm reception of Utpal Tatu, and his generosity in welcoming us into his home to celebrate the dawning of the 2013 New Year, I thank colleagues at The University of Melbourne, who assisted in supervising my research laboratory while I was away, and my staff and students for their tolerance of my lengthy absence from the office. My sincere thanks go to the late Dorothea Sandars and the Churchill Fellowship trust for the financial support that made this trip possible. I am enormously grateful for this amazing opportunity, and will endeavour to ensure that I make the most of the fortunate chance to learn so much about malaria and drug discovery in India, and share the lessons learnt from this research visit.
Page 3 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
Executive Summary Stuart A Ralph Australian Research Council Future Fellow and Senior Lecturer Department of Biochemistry and Molecular Biology, The University of Melbourne Phone +61 3 8344 2284 Project title: Investigating Open Source strategies for developing drugs against neglected parasitic diseases Between September 2012 and January 2013 I travelled to India to investigate an initiative designed to produce affordable drugs for neglected diseases. This network, called the Open Source Drug Discovery Initiative (OSDD), was established in 2008 and currently focuses on two diseases, tuberculosis and malaria, which inflict a heavy disease burden on India. Australia and India both have strong malaria research and there is ample opportunity for expanded interaction and collaboration between researchers in these two countries. Because Australia has a small pharmaceutical industry, collaboration with the Indian open source drug discovery network offers a key opportunity for Australian researchers pursuing the early stages drug discovery to interact with Indian scientists working on mid and later stages of drug development. Highlights of my research visit included the opportunity to learn more about the practical application of open source drug discovery from Dr Saman Habib at the Central Drug Research Institute, Lucknow, and the numerous interactions with research scientists and students who were so motivated to generate therapies for India’s important infectious diseases. The collaboration of these scientists and their dedication to so important a goal was inspirational. My interaction with the many participants in the OSDD project confirmed for me that this approach to drug discovery has the potential to deliver quality research in an effective and affordable manner. The obvious failures of traditional profit-driven drug discovery means that we must either rework those systems to better target the most important international infectious diseases, or come up with alternative strategies. Open source drug discovery is a promising option that capitalises on the existing extended network of academic research scientists who possess skills spanning much of the drug discovery pipeline, and who are eager to work together to generate drugs for neglected diseases. Australia should embrace open source drug discovery as one of several strategies to translate basic research into therapeutic products for important diseases that effect Australia and the world. We should expand our interactions with Indian research through open source networks, which not only provide a perfect framework for international collaborative research, but also allows us to reach out to one of world’s fastest growing research communities. This can be implemented by expanded funding for the Australian open source drug discovery for malaria project to interact more closely with the Indian OSDD project, by improved funding for bilateral projects, and by Australian malaria scientists participating in the Indian OSDD using existing funded projects. I will continue to pursue close links with Indian research networks, to advocate the use of Open source networks as a framework for this research, and I will assist in
Page 4 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
facilitating links between young Australian and Indian scientists working in neglected disease research. Programme Lucknow: Central Drug Research Institute Host: Dr Saman Habib, Chair of Division of Molecular and Structural Biology, Lead Principal investigator of Indian Open Source Drug Discovery Initiative for Malaria Oct 1 – Dec 31, 2012 Lucknow: King George's Medical University Host Dr Vimala Venkatesh, Department of Microbiology Oct 2012 Bombay – Tata Institute for Fundamental Research Host: Professor Gotam Jarori, Department of Biological Sciences Oct 29 2012 Bombay – Indian Institue of technology Bombay Host: A/Prof Swati Patankar, School of Biosciences and Bioengineering Oct 30 2012 Bombay – Haffkine Institute Host: Dr Dr. Abhay Chowdhary, Director Oct 31 2012 Pune - National Chemical Laboratories Host: Dr Dhanasekaran Shanmugam Oct 31-Nov 4 New Delhi – National Institute of Immunology Host: Dr Pushkjar Sharma Nov 4 New Delhi – International Centre for Genetic Engineering and Biotechnology Hosts: Dr Asif Mohammad, Pawan Malhotra, Renu Tuteja, Chief investigators, Mammalian Biology: Malaria Group, Nov 4-5 New Delhi – Council of Scientific and Industrial Research Host: Zakir Thomas, Project Director, Open Source Drug Discovery Nov 5 Bangalore, Indian Institute of Science Host: Professor Utpal Tatu, Department of Biochemistry Dec 31 2012 -Jan 1 2013
Page 5 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
Main Body Obstacles to malaria drug and vaccine development Infectious diseases exert an immense burden on human health, with enormous impacts on human societies and economies. This burden is disproportionately borne by the poorest and least politically powerful people in the world. Two of the most significant infectious diseases are tuberculosis (caused by the bacterium Mycobacterium tuberculosis) and malaria (caused by Plasmodium parasites). Each of these diseases kills more that 1 million people worldwide each year, but relatively few of these deaths occur in rich, economically developed nations. In the case of tuberculosis, an effective but imperfect vaccine has existed for more than a century, but for malaria, no vaccine exists. Drugs are a major part of our toolkit to control these infectious diseases. An important consequence of the concentration of burden in less economically developed nations is that individuals who suffer from these diseases have little money to spend on therapeutic interventions necessary to prevent or treat these diseases. Major financial obstacles exist to the use of drugs to control neglected diseases. The first is that where satisfactory drugs exists, they may be out of financial reach of those who are affected, and in some cases there is insufficient funding to provide and distribute these drugs to those in need. The second obstacle is that the lack of a market for drugs for such diseases means that there is little immediate financial incentive for pharmaceutical companies to invest in the development of drugs for neglected diseases. There is no short-term financial reason for pharmaceutical companies to spend money on a market that is not viable, with little prospect of any financial return. The consequences of this lack of financial incentive are clear; pharmaceutical and biotechnology companies contributed only 13.8% of the global funding dedicated to neglected disease research and development between 2007 and 2011 (Moran, Guzman et al. 2012). These failures of markets have led to a massive shortfall in the development of drugs - only 1% of the drugs developed between 1975 and 1999 were directed against the neglected diseases that account for more than 11% of global disease burden (Trouiller, Olliaro et al. 2002). The small investment role played by industrial groups in neglected disease R&D means that nearly all funding comes from public and philanthropic sources. Australia plays an important role in the support of research and development for neglected diseases. The Australian NHMRC is the 12th largest global funder of neglected disease R&D
Page 6 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
and in malaria, an area of specific research strength for Australian science, the Australian NHMRC was the 9th largest global funder of R&D (Moran, Guzman et al. 2012). Significant additional funding for R & D for neglected diseases including malaria is contributed through the Australian Research Council and via Department of Industry, Innovation, Science, Research and Tertiary Education research grants. Australian funding for malaria control is also provided via AusAid and contributions to international agencies. Although global public and philanthropic funding for neglected disease research and development is considerable (around $2.5 billion per year) the means by which new therapies might be developed through these R&D activities is not always clear. In particular, strategies to translate fundamental research from academic laboratories into drug and vaccine development programs are fragmentary, and at times conflicting. A number of strategies exist to provide incentives drug discovery in neglected diseases. These may be classified as either push and pull factors. Push factors include direct government and philanthropic support for neglected disease R & D, or tax breaks for pharmaceutical or biotech firms involved in this work. Pull factors include initiatives such as the Orphan Drug schemes operated by the Australian Therapeutic Goods Administration or the USA’s Food and Drug Administration that make regulatory approval easier and cheaper for drugs that are not commercially viable. An additional pull incentive is the priority review voucher scheme in place in the USA, whereby companies that obtain approval for a drug for a neglected disease are awarded a voucher that can be redeemed for priority review of a separate drug for a non-neglected disease. (Ridley, Grabowski et al. 2006; Ridley and Sanchez 2010) The most important pull incentive for neglected diseases is the Advance Purchase Commitments, through which governments or large philanthropic bodies, (particularly the Bill and Melinda Gates Foundation) provide a guaranteed future market - equivalent in value to a blockbuster drug in some cases - for any company that produces a particular vaccine or drug product (Towse and Kettler 2005). APCs are yet to deliver any big success stories, and have been criticized as being an unrealistic solution that is unlikely to encourage many pharmaceutical companies to initiate large scale R & D in neglected tropical diseases (Light 2005; Löfgren 2005). Even if pull factors eventually entice more pharmaceutical or biotech firms into drug and vaccine development, nearly all of the fundamental research in developing novel chemical entities still takes place almost exclusively in academic research environments and small research firms. If APCs incentivize big pharma to take compounds into development pipelines, those compounds are very likely to have been identified, validated and improved by academic research groups. A common and increasingly successful route for drug development in neglected disease research is through Publicprivate-partnerships that bring together academic research programs with small pharmaceutical firms, contract research organisations, or with larger drug development companies (Nwaka and Ridley 2003). These appear to offer considerably better value for money than direct funding or APCs for big pharma (Moran 2005).
Page 7 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
Several not-for profit organisations, including Medicines for Malaria venture, and the Drugs for Neglected Disease initiative have shepherded development and registration of new drugs or formulations in the past few years (Anthony, Burrows et al. 2012; Wells, Diap et al. 2013). In some cases PPPs have advanced chemical entities with intellectual property protection (von Seidlein, Jaffar et al. 1997), in others, drugs have been brought to market stage without any intellectual property being claimed (Lacaze, Kauss et al. 2011), and in at least one case, patent applications were initiated and then abandoned, but nevertheless brought to market (though later withdrawn) (Luzzatto 2010). Decisions about the handling of intellectual property (IP) play an important and controversial role in neglected disease drug R & D. In general, the lack of commercially markets for neglected diseases mean that IP for the treatment of these diseases alone has little commercial value, although such IP may be valuable for licensing purposes for other, more viable markets. As indicated above, several approaches to IP have resulted in the successful registration of drugs for neglected diseases. In this report I do not advocate a one-size-fits-all approach to the handling of IP in neglected disease R & D. A background to open source drug discovery One approach that has growing appeal to many in the field of neglected disease is the open source drug discovery movement. This movement is based on several premises; • that market based incentives for drug discovery have failed neglected disease R&D. • that open sharing of data, techniques, ideas, reagents, infrastructure and expertise will lead to accelerated R & D among motivated groups compared to closed systems. • that traditional IP protections in general and patenting in particular are of little value in neglected disease research because of the lack of commercial market for therapeutic products, so protection of these generates unjustifiable obstacles to the free sharing of data and expertise that would otherwise accelerate R & D. Although the description “open source” is used to mean different things in different spheres, the philosophy generally describes collaborative projects whereby data is shared freely to create some product which is itself freely redistributed. Innovations derived from this process may be used for other purposes, as long as they are freely distributed themselves. Licensing systems differ, but open source systems do not necessarily restrict individuals or companies from selling an aggregate product (such as a drug) that contains innovations derived from the open source project. Several open source drug discovery initiatives now exist, a large proportion of which concentrate on drug discovery for neglected disease. These include the Tropical Diseases Initiative (Maurer, Rai et al. 2004), (Orti, Carbajo et al. 2009), and The Synaptic Leap project (www.thesynapticleap.org) – both are collaborations based predominantly on computational investigations of neglected disease drug targets and drugs. Two recent open source initiatives in Australia and India have combined computational initiatives in drug discovery with substantial laboratory based investigations of drugs and drug targets. I am a participant in both projects. To my knowledge these are the most active and extensive networks for the development of drugs for neglected diseases
Page 8 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
explicitly using open source approaches. The smaller of these initiatives is the Open Source Drug Discovery project initiated by Dr Matthew Todd at the University of Sydney (http://openwetware.org/wiki/Open_Source_Drug_Discovery) This is a collaboration between bioinformaticians, biochemists, parasitologists, medicinal and synthetic chemists, and is financially supported by a Linkage grant from the Australian Research Council and the Geneva based NGO, Medicines for Malaria Venture. A much larger and better endowed project is the Indian Open Source Drug Discovery initiative, which I will discuss more fully below. My fellowship and lessons learnt The Dorothea Sandars Churchill Fellowship supported my travel to and within India, to investigate open source methods to pursue drugs for neglected diseases. In addition to the period supported by the Dorothea Sandars Churchill Fellowship I prolonged my stay to learn more about malaria research and the impact of malaria on India. I travelled with my wife and two daughters (4 and 7), the eldest of whom went to a local school for 3 months during this period while we were in Lucknow. My first and longest stop in India was in Lucknow, the capital of Uttar Pradesh. Uttar Pradesh, like most of northern Indian, has a peak of vector borne diseases during the monsoon, and we timed our travel to coincide with the retreat of the monsoon. My hosts in Lucknow were Dr Saman Habib, a molecular biologist and parasitologist and Dr Amit Misra, a pharmacologist and tuberculosis researcher, both researchers at the Central Drug Research Institute. Dr Habib is a lead principal investigator on the head of Open Source Drug Discovery Initiative for Malaria (OSDDm). This initiative is a sector of the wider OSDD consortium, the largest open source drug discovery initiative in the world. This consortium is led by the Indian Council of Scientific and Industrial research, an autonomous, government-funded body which is the largest R & D organisation in India. The OSDD has a total budget of approximately USD$46 million. OSDD was established in 2008 with the goals of providing affordable healthcare through collaborative research and development for neglected diseases like tuberculosis, malaria, HIV, and leishmaniasis. The initial phase of OSDD, launched in 2008 was focused on tuberculosis (Bhardwaj, Scaria et al. 2011), and is now well established, with tangible research and community engagement outcomes (Singh 2008; Ardal and Rottingen 2012). A second phase launched in 2012 targeting malaria (OSDDm), and Dr Habib has been a spearhead of this section of OSDD. In Lucknow, I worked with Dr Habib as well as discussing widely with her colleagues, and students involved on the OSDD project, to gain a better understanding of the strong points and challenges of this initiative. OSDD is managed through a central web portal called sysborg (http://sysborg2.osdd.net/) that integrates the efforts of all the OSDD participants. The goal is that an initial phase, involving open sharing of ideas, and exchange of expertise and reagents will lead to prioritization of candidate drug targets and inhibitors. Compounds that are potential drug leads will be synthesized where necessary or supplied when already existing. Natural product collections will also be
Page 9 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
screened for anti-parasite activity. Synthesis of compounds identified through in silico screens will be conducted in medicinal chemistry laboratories, but also by Masters students enrolled in chemistry degrees, supervised by skilled chemists. This seems to be a highly innovative aspect of the project that not only uses the resources of students who might otherwise be learning techniques using trivial examples, but additionally may motivate these students to become involved in further research in neglected diseases. Compounds have already been synthesized via this means for the tuberculosis section of the OSDD project, and these students were reportedly highly motivated to be participating in such a worthwhile endeavour. The applicability of this innovation to an Australian context is unclear. India’s volume of Masters degree students in synthetic chemistry laboratories is not mirrored in Australia, although such chemical syntheses might also be appropriate for Honours students training in medicinal or organic chemistry laboratories. Highly motivated undergraduate students at the University of Sydney are involved in the OSDD project in Dr Todd’s research group, which may serve as a successful template for future such participation. Candidate protein targets for drugs will also be characterised and validated by the OSDD network. Where these targets meet appropriate criteria they will be scaled into highthroughput assays and tested against the available compounds described above. Where appropriate lead molecules are identified, they will be subject to toxicology testing, pharmacological testing and further medicinal chemistry refinement, before being subject to clinical trials. These later stages will involve researchers from research institutions and hospitals, but where expertise or large infrastructure items for specific steps is lacking, these tasks will be outsourced to contract research organisations (CROs). I was interested to note that such CROs play a significantly more central role in Indian biomedical research than in Australia, and were regarded by the Indian scientists I spoke to as appropriate for routine tasks but not for the establishment of assays or procedures. These CROs appear to be particularly useful when a procedure requires an expensive infrastructure A schematic generated by OSDD, describing the item that is not available conceptual model for collaborative development of a in an academic research drug for tuberculosis. setting. The selection of initial R & D projects and the prioritization of molecules and targets that will be further pursued will be decided by networks of scientists, advised by open peer review. The sysborg webtool allows any user to propose initial ideas for further feedback, and then allows investigators to submit proposals for work packages to be funded by OSDD. These are open to comments and questions by any other OSDD participant (non
Page 10 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
anonymously, unlike regular grant peer review), who also provide a score, and the lead investigator can then modify the grant, ask for further specific feedback, or invite participation in the work package by other researchers. For example, a molecular cell biologist might ask for assistance by interested experts in structural biology to assist with structural determination of a protein. A grant committee can solicit further expert review if necessary, and then makes decisions about which projects are worthy of funding and at what budget level. A previous study of the OSDD system pointed out that there was some lack of transparency in this final stage of review (Ardal and Rottingen 2012), although openness in review is certainly not a regular feature of research funding decision making. A potential drawback of the open system of review is that colleagues might not wish to criticise a project, or more specifically to question the capacity of an investigator to conduct a particular project when the applicant can identify them. I did observe generic reviews praising the value of a submitted project without any detail about why the project was praiseworthy, but equally, there are respectful comments posted that point out shortcomings or oversights in proposals. In any case, the final decision-making group has sufficient expertise to discern the platitudes from the measured recommendations or criticisms. Indeed my observations from reading these reviews, and from participating in teleconferences with multiple participants openly appraising research, is that the diverse source of feedback made possible by open review provides highly valuable input that is not possible within a traditional review system. A noteworthy aspect of the network is that some elements of the OSDD development pipeline are flexible, whereas other elements are determined in advance. For example, any participant can propose to conduct a genetic screen for a drug target in their own lab, but high throughput screening is to be conducted in specified centres with known quality control measures. The CDRI for instance, has established ex vivo and animal models for testing anti-malarial compounds, and they will be a central testing site when such assays are required. I felt that the committed involvement of established centres that will conduct some of the “heavy lifting” is a particularly important part of the OSDD structure. Another structural strength is the availability of multiple nodes throughout India able to provide pre-existing expertise in
The CDRI Macaque monkey colony. This facility allows researchers to test drug responses in monkey models of malaria. The availability of many such screening facilities is a strength of the OSDD network
Page 11 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
most of the important steps of drug target characterisation. This includes bioinformatic expertise to identify drug targets, molecular biology laboratories to genetically validate targets, biochemistry and structural biology labs to characterise protein targets, organic and medicinal chemistry laboratories to synthesise and optimise compound series, multiple animal model laboratories to test the effects of compounds in rodent and monkey malaria models, as well as toxicology, pharmacology and drug delivery labs for preclinical testing. Though participants do not need to be CSIR funded laboratories to participate, the existence of a network of CSIR institutes such as CDRI, ready to commit to involvement in OSDD, is a distinct advantage of the Indian OSDD that may might not be easily reproduced if similar networks were to be replicated in other countries. One highlight of my stay in CDRI was being able to observe an integrated drug development network, with expertise in so many of the stages of drug development. The co-location of so many scientists with complementary skills allowed (at least in theory) for very easy collaboration between researchers on the same site, involved in multiple steps of drug development pipelines. I spoke with bioinformaticians, structural biologists, biochemists, toxicologists and pharmacologists, all working on related projects in neglected disease drug development. At the time of my stay, the CDRI was in the process of moving from an complex of buildings called the Chattar Manzil, a palace built by the 18th Century architect Claude Martin, and later home to the Nawabs of Awadh, to a newly purpose built campus on the outskirts of Lucknow. The Chattar Manzil is wonderfully atmospheric, and has a rich heritage, but it is The faded glory of the CDRI campus at Chattar rather ill suited to the modern Manzil – most activities have now been moved to a demands on a drug research new campus in Lucknow. institute, and the move is met with mixed emotions by the staff and students. Travel to interacting nodes of the Indian OSDDm initiative In addition to discussion of OSDD with the lead investigators at the CDRI, I travelled to headquarters of the OSDD at the CSIR in New Delhi, as well as to research institutes in Mumbai, Pune, New Delhi and Bangalore that are participants in the OSDD network. These campuses were very physically different from Australian research institutes – all had housing on campus for staff and students, potentially enhancing the community cohesion of these institutes, and allowing the institute members to live in affordable
Page 12 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
accommodation in what would otherwise be unaffordable areas. The most striking example of these was the TATA Institute for Fundamental Research (TIFR), with fantastic shore-front accommodation in one of the more expensive precincts of Mumbai. This inner city campus was a distinct contrast with the equally pleasant campus at the Indian Institute for Technology, Bombay, Powai, on the outskirts of Mumbai, which is on the edge of thick jungle, and where students and locals encounter crocodiles and leopards. These visits resulted in very fruitful discussions, with helpful discussions on OSDD. At each of these institutes I gave presentations on my work relevant to Open Source Drug discovery for malaria, and established collaborations that will be conducted through Open Source methods.
The Powai Lake on the edge of the IIT-Bombay campus, home to crocodiles
The seaside campus of TIFR, Bombay, looking out to the billionaires’ home of Malabar hill.
A recurring theme in these discussions was research scientists’ dismay at the failure of traditional pathways for drug development to deliver much needed drugs for the serious infectious diseases of India. A near universal opinion was that standard intellectual property protections were an obstacle rather than an attractant to progress of research projects. Some of the scientists interviewed had sought patent protection for their discoveries in neglected disease. In these cases patent applications were in response to pressure from institute research offices that believed that this would help commercialise their research outcomes, in opposition to the scientists views on the matter. This is an important aspect of open source drug discovery – scientists are generally uncomfortable with restrictions on their ability to share data and reagents, and are motivated when working in collaborative projects where their research is enhanced by synergistic interactions. This is perhaps even more apparent in neglected disease research, where disillusionment with the failure of intellectual property as an incentive to produce therapeutic products is so widespread. Scientists I spoke to had no professed desire to make any personal financial gains from their research, and asserted their strong aspirations that their research should benefit poor populations in need of medicine. Of course, scientists’ desire to work in such open source models is not in itself proof of the practical superiority of this model over traditional, closed IP protection models. However
Page 13 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
the strong motivation of scientists working under such open source systems is itself a factor that strengthens the efficacy of open source research. Challenges to overcome in the open source drug discovery model A consistent challenge to collaboration through open source models was the necessity of web-infrastructure that facilitates such transparent collaboration. The Sysborg tool that is used to organise research and collaboration by the Indian OSDD is well supported and provides extensive support for different forms of data sharing. This does require considerable manual maintenance and administration. The sysborg site is not entirely open – visitors must register first (a process taking several days) before they can participate, which no doubt inhibits some spontaneous interactions. Although this restriction to registered users may be necessary for some applications, my feeling is that wherever possible barriers to interaction by outsiders to OSDD projects should be eliminated. In addition to some frustrations with registration, technology for sharing the relevant data types necessary for a biomedical open source collaboration is still imperfect. Such collaborations require web infrastructure that allows the sharing of chemical structures, searching compounds by molecular similarity, and easy uploading of compound information. Such collaborations should also use computational tools that support web-based laboratory workbooks, archiving of important meta-data relevant to chemical and biological datasets, and social media type facilities to allow communication by participants. Multiple tools exist that attempt to support these aims – none are perfect, causing some annoyance among the OSDD participants I spoke to in India, as well as among scientist I have discussed this with in Argentina, USA, UK and Australia. In addition to honing these specific tools to support collaboration, more needs to be done to create an internet landing place that supports collaboration between different OSDD projects. Such a unified site would allow scientists to share data and skills between different projects, to identify areas of redundancy and overlap, as well as allowing nonresearch based participants to identify which OSDD projects they might usefully contribute to. This gap has been previously identified by a recent analysis of open source projects (Masum and Harris 2011), and is acknowledged by participating scientists. An additional gap that has not yet been adequately met by open source drug discovery networks is the lack of a unified clearing house for data on compounds that might serve as anti-malarial drug leads. Many small-scale screening efforts have already been conducted and published, presenting promising compounds that have not been pursued (not necessarily because of limitations in the compound properties). In addition, several large-scale anti-malarial screens have recently been published, providing researchers with tens of thousand of chemical starting points for anti-malarial drug discovery. Some initiatives have attempted to assemble and describe these data (Aguero, Al-Lazikani et al. 2008; Hohman, Gregory et al. 2009; Gaulton, Bellis et al. 2012). None completely satisfy the ideal desires of an open source project to provide maximum information to researchers on a comprehensive compound library while simultaneously allowing researchers to post information about what they are working on. Significant work to improve such tools is needed.
Page 14 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
In some cases, web-tools that underlie open source projects also need to support crowdsourcing campaigns that seek participation or input from large numbers of non-skilled participants. These might be game-style modelling of drug-protein interactions, or may be distributed computing initiatives that conduct computer-based drug modelling. Distributed computing initiatives are most well known from the famous SETI@home project that uses the spare processor time of many networked individuals to search for extra-terrestrial radio signals. Such initiatives are now being used in neglected disease drug discovery projects, and include Go Fight against Malaria project (gofightagainstmalaria.scripps.edu) and the Discovering Dengue Drugs – Together project (Tomlinson, Malmstrom et al. 2009). Within the OSDD project, two major computational accomplishments so far have been the improved annotation of the genome of Mycobacterium tuberculosis (Bhardwaj, Bhartiya et al. 2009), and an improved map of M. tuberculosis protein-protein interactions achieved through manual curation by participants of varying expertise and crowd-sourcing (Vashisht, Mondal et al. 2012). These crowd-sourcing projects allow some difficult computational projects to be conducted more affordably, but also provide points for important community engagement with students and interested laypeople. One discussion point in different research groups and at the CSIR headquarters was the place of tangible incentives for participation in OSDD. In some cases, participating students have been awarded prizes, including laptop computers, as a reward for their participation in OSDD. In the case of the laptop computers, the award was made with no prior advertisement that there would be prizes for outstanding participation, though in some other instances small prizes have been offered in advance, for example to students who make the best video promoting OSDD activities. The Australian OSDD network has also discussed the possibility of offering awards for participation in specific activities. A general recommendation from OSDD members was that prizes to participants were appropriate in some circumstances, but that long term commitment was only likely to be motivated by genuine desire to further the project and its goals, or desire to gain relevant research skills, rather than to win an award. A related theme that emerged from conversations with network members was the need for concrete advances in the research being conducted in order to maintain motivation for participation. Open Source projects where no apparent progress is being made quickly lose attractiveness to outsiders, so it is important that the key participants provide regular progress updates. It is not sufficient to ask volunteers to devote computational efforts to design new compounds to test – it has to be explicit that once those are designed, there is a lab ready to purchase or synthesise those compounds, and personnel ready to conduct biological assays on those compounds. Future participation of volunteers is dependent on the results of those experiments then being clearly fed back to collaborators. This lab based research of course requires funding to maintain – whether it be in the form of research groups that are funded for other projects but which can expand to work on OSDD projects without incurring extra costs, or groups directly funded to work on OSDD research. One way of leveraging existing funding is for research projects to submit work established outside the OSDD sphere into the environment of the OSDD network, so that existing research is enhanced through collaborations with other OSDD participants.
Page 15 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
One major challenge for open source drug discovery projects is that while early stages of compound and target validation may be relatively affordable, and in some cases may be absorbed in to an academic laboratory’s existing budget without need for an additional funding, late stage drug discovery is unavoidably expensive. Clinical trials in particular are very costly, running from tens to hundreds of million dollars, and require specific expertise. The cost of privately run clinical trials in India is approximately 40% of the cost in Australia (KPMG 2012), but the cost is still substantial. THE CSIR says that the cost of running clinical trials that come out of the OSDD projects will be borne mainly by government funding and carried out in public hospitals. Cost savings achieved by the OSDD through carrying out collaborative research at existing, publicly funded organisations have been substantial and demonstrable for some specific tasks in the early stages of drug development (Ardal and Rottingen 2012), However, it is not obvious whether OSDD methodologies will result in similar cost reductions for clinical trials. In an Australian setting, compounds that progress to the point of clinical trials could potentially be integrated into the Indian OSDD program for further development, or could be conducted in collaboration with PPPs such as Medicines for Malaria venture. Lessons from working in a malaria-endemic country In addition to learning much about anti-malarial drug discovery from discussion with researchers, I gained considerable awareness of some of the issues in neglected disease research from working in a malaria-endemic country for more than 3 months, This awareness of vector-borne disease was heightened by our having our two young daughters travelling with us, and facing the presence of serious disease in our living environment. Although we were deliberately working and travelling at a time of year of
Mosquito fogging, Hindu temple, Chennai, Tamil Nadu
Poster asking people to look out for mosquitos to combat disease, Fort Cochin, Kerala
Page 16 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
low vector borne disease transmission, we did travel to some regions with considerable dengue and malaria transmission. Vector control is a major aspect of malaria prevention. In many areas inadequate funding for mosquito surveillance and spraying leads to frequent disease outbreaks. Public messages encourage members of the people to be help eradicate mosquitos as a vector for disease, although there is little infrastructure to support such programs. In the wet season particularly, mosquitos, and thus malaria infections, are particularly prevalent. My anecdotal discussions with locals as well as expert diagnostic microbiologists, physicians and malaria biologists indicated that pathology services are relatively affordable, and many people seek malaria diagnosis for a fever, before purchasing malaria medication. Others self-diagnose and purchase antimalaria drugs from local pharmacists, without prescription. I was able to readily obtain, separately, doxycycline and proguanil, two established and cheap malaria drugs. Proguanil was suggested by the pharmacist as a monotherapy, although use of proguanil in isolation without being combined with other drugs should not generally be advised because such usage leads to accelerated drug resistance by parasites. Although genuine drugs are generally readily available and affordable for middle class incomes, counterfeit drugs are also a major problem. These counterfeit drugs lead to deaths both through failure to cure the disease, as well as due to toxins in some counterfeit products (Dash, Valecha et al. 2008) (Cockburn, Newton et al. 2005). Some counterfeits also contain lower than advertised amounts of the relevant drug compounds. Prescription of single malaria drugs for monotherapy or prophylaxis, as well as treatment with sub-curative levels of drug are factors contributing to the emergence of and selection for drug resistant parasites.
My wife, Sybil, purchasing the antimalarial drug doxycycline from our local chemist. Lucknow, Uttar Pradesh. No prescriptions are needed, but some pharmacists may sell counterfeit or low quality drugs.
Although these are well known issues, my own experiences seeking malaria drugs, reinforced the pernicious problems encountered when purchasing malaria drugs. Although the problems of non-universal reach of healthcare and counterfeit drugs would not be entirely eradicated by the availability of very cheap genuine malaria drugs, they reinforced for me the importance of affordability and quality in antimalarial drugs. Open Source drug discovery approaches are specifically focused on these issues.
Page 17 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
Conclusions And Recommendations My exposure to the Indian OSDD initiative was a very scientifically enriching experience. The initiative has many inspirational features, and it was particularly encouraging to see the high motivation levels of student wanting to be part of a positive R & D collaboration to achieve important outcomes for healthcare. The segment of the OSDD network focused on tuberculosis has already produced some important outcomes, and these were demonstrably achieved more quickly and affordably than via traditional R & D mechanisms. There is much to be gained from such methods, and the abject failure of traditional methods of drug discovery to deliver for neglected diseases mean that new approaches are needed. As well as OSDD, several alternative paradigms to traditional market-motivated R & D are being trialled, including new incentives for commercial players to engage in vaccine and drug development and new Public-Private partnerships (PPPs) to shepherd drugs through interactions between academia and small and large pharmaceutical firms. While PPPs in particular appear to be a highly promising model, open source drug discovery approaches have made considerable advances and deserve expanded support to test whether they are a plausible mechanism to produce new drugs. I will present seminars at upcoming parasitology conferences disseminating this information, and will communicate these messages in academic seminars at University departments. Several steps are important to build on the success of existing open source initiatives, and these could help improve links to drug development pathways for the extensive Australian malaria research community. First is a continued funding commitment to the established OSDD network created by Dr Matthew Todd at the university of Sydney. I have a vested interest in this network in that I contribute to their screening activities, but I receive no direct or indirect funding from this project. I will be an advocate for extension of the funding of this project and will assist with grant writing for renewal of this project, and in seeking funding beyond the existing sponsors. Second is an enhanced interaction between the Australian and Indian OSDD networks. While Australian malaria research could clearly benefit from the expertise of a large network working on similar problems, there are also some areas of malaria research (eg parasite transfection, parasite metabolomics) in which Australia has technological advantages. There are opportunities for fruitful collaborations in areas of bioinformatic prioritisation of protein targets, genetic validation of drug targets, sharing of existing priority compounds, synthesis of analogues of promising inhibitors, and screening of compounds in bioassays that are only available in one of the countries. Collaboration between the Indian OSDD will provide mutual benefits for progression of the drug development pipeline, and transfers of expertise between India and Australia will enhance the research profile of both countries. Dr Todd and I are participants in both networks, and I will advocate closer ties between the two networks. I will lead the submission of a collaborative grant to the Australian-Indian Strategic Research Fund to
Page 18 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
support collaborative research between the two networks, and I will also seek similar funding from the Bill and Melinda Gates Foundation. To make sure that scientific dialogue continues between Australian and Indian scientists involved in OSDD, I will encourage researchers in the Australian OSDD network to apply for an Australian Academy of Science-funded Australia-India Fellowship to travel to Indian OSDD labs, and will search for opportunities to send students to India to participate in OSDD projects. I will also seek to organise a workshop, attached to an upcoming malaria conference, to bring together Australian and Indian OSDD researchers. Third is the development of improved informatic infrastructure. Better tools are needed to support the utilisation of the open chemical data now available for malaria and a unifying website is necessary to better integrate the multiple open source initiatives, crowd sourcing efforts and distributed computing projects focused on neglected diseases. The TDRtargets project on which I am a lead investigator is one candidate to provide better tools to investigate chemical screening data, and we are working on improved access and annotation of these data types. The Chembl database has also emerged as a provider of quality access to public drug screening data, and I will communicate with the Chembl researchers to investigate mechanisms for incorporating researchers’ manual annotations and small scale malaria screening information into these databases. The provision of a central landing place to steer volunteers to relevant open source projects and to provide communication between OSDD projects is not easily implemented, and is complicated by fragmented existed websites with unclear identities. Dr Todd has identified this as an important issue for the Australian OSDD, and I will work with him to establish a site that brings together data from international OSDD projects.
Page 19 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
References Aguero, F., B. Al-Lazikani, et al. (2008). "Genomic-scale prioritization of drug targets: the TDR Targets database." Nat Rev Drug Discov 7(11): 900-907. Anthony, M. P., J. N. Burrows, et al. (2012). "The global pipeline of new medicines for the control and elimination of malaria." Malaria journal 11: 316. Ardal, C. and J. A. Rottingen (2012). "Open source drug discovery in practice: a case study." PLoS neglected tropical diseases 6(9): e1827. Bhardwaj, A., D. Bhartiya, et al. (2009). "TBrowse: an integrative genomics map of Mycobacterium tuberculosis." Tuberculosis 89(5): 386-387. Bhardwaj, A., V. Scaria, et al. (2011). "Open source drug discovery--a new paradigm of collaborative research in tuberculosis drug development." Tuberculosis 91(5): 479486. Cockburn, R., P. N. Newton, et al. (2005). "The global threat of counterfeit drugs: why industry and governments must communicate the dangers." PLoS medicine 2(4): e100. Dash, A. P., N. Valecha, et al. (2008). "Malaria in India: challenges and opportunities." Journal of biosciences 33(4): 583-592. Gaulton, A., L. J. Bellis, et al. (2012). "ChEMBL: a large-scale bioactivity database for drug discovery." Nucleic Acids Research 40(Database issue): D1100-1107. Hohman, M., K. Gregory, et al. (2009). "Novel web-based tools combining chemistry informatics, biology and social networks for drug discovery." Drug Discovery Today 14(5-6): 261-270. KPMG (2012). Competitive Alternatives – KPMG's Guide to International Business Location Costs. Lacaze, C., T. Kauss, et al. (2011). "The initial pharmaceutical development of an artesunate/amodiaquine oral formulation for the treatment of malaria: a publicprivate partnership." Malaria journal 10: 142. Light, D. W. (2005). "Making practical markets for vaccines. Why I decided that the Center for Global Development Report, Making Markets for Vaccines, offers poor advice to government and foundation leaders." PLoS medicine 2(10): e271. Löfgren, H. (2005). "Purchase commitments: Big business bias or solution to the ‘neglected diseases’ dilemma?" Australian Review of Public Affairs. Luzzatto, L. (2010). "The rise and fall of the antimalarial Lapdap: a lesson in pharmacogenetics." Lancet 376(9742): 739-741. Masum, H. and R. Harris (2011). Open Source for Neglected Diseases: Magic Bullet or Mirage? Washington, DC, Results for Development Institute. Maurer, S. M., A. Rai, et al. (2004). "Finding cures for tropical diseases: is open source an answer?" PLoS medicine 1(3): e56. Moran, M. (2005). "A breakthrough in R&D for neglected diseases: new ways to get the drugs we need." PLoS medicine 2(9): e302. Moran, M., J. Guzman, et al. (2012). NEGLECTED DISEASE RESEARCH AND DEVELOPMENT: A FIVE YEAR REVIEW. Annual G-FINDER Survey Report, Policy Cures. 5.
Page 20 of 21
Stuart Ralph, Dorothea Sandars Churchill Fellowship
Open Source Drug discovery for neglected diseases
Nwaka, S. and R. G. Ridley (2003). "Virtual drug discovery and development for neglected diseases through public-private partnerships." Nature Reviews Drug Discovery 2(11): 919-928. Orti, L., R. J. Carbajo, et al. (2009). "A kernel for open source drug discovery in tropical diseases." PLoS neglected tropical diseases 3(4): e418. Ridley, D. B., H. G. Grabowski, et al. (2006). "Developing drugs for developing countries." Health Affairs 25(2): 313-324. Ridley, D. B. and A. C. Sanchez (2010). "Introduction of European priority review vouchers to encourage development of new medicines for neglected diseases." Lancet 376(9744): 922-927. Singh, S. (2008). "India takes an open source approach to drug discovery." Cell 133(2): 201-203. Tomlinson, S. M., R. D. Malmstrom, et al. (2009). "New approaches to structure-based discovery of dengue protease inhibitors." Infectious disorders drug targets 9(3): 327-343. Towse, A. and H. Kettler (2005). "Advance price or purchase commitments to create markets for treatments for diseases of poverty: lessons from three policies." Bulletin of The World Health Organization 83(4): 301-307. Trouiller, P., P. Olliaro, et al. (2002). "Drug development for neglected diseases: a deficient market and a public-health policy failure." Lancet 359(9324): 2188-2194. Vashisht, R., A. K. Mondal, et al. (2012). "Crowd sourcing a new paradigm for interactome driven drug target identification in Mycobacterium tuberculosis." PLoS ONE 7(7): e39808. von Seidlein, L., S. Jaffar, et al. (1997). "Treatment of African children with uncomplicated falciparum malaria with a new antimalarial drug, CGP 56697." The Journal of infectious diseases 176(4): 1113-1116. Wells, S., G. Diap, et al. (2013). "The story of artesunate--mefloquine (ASMQ), innovative partnerships in drug development: case study." Malaria journal 12(1): 68.
Page 21 of 21
