As I noted in last week’s post, I reconnected recently with a colleague who has an interest in accelerating the discovery of drugs for neglected diseases via a data-sharing platform. Michael Pollastri is an associate professor of chemistry at nearby Northeastern University and a ten-year veteran of the pharmaceutical industry, and I have mentioned him and his work on repurposing existing drugs to target parasites in previous posts (e.g., “A Rare Request, an Orphan Offer”). Mike is starting a web-based portal for sharing of data and ideas among medicinal chemists trying to develop new drugs for the neglected diseases (ND), which in this case, includes the thirteen WHO-listed neglected tropical diseases and the two better-known global diseases, tuberculosis and malaria. All are diseases of poverty, have low mortality but high morbidity with long periods of suffering and often a lifetime of disablement, and hence have a devastating effect on societies in the endemic areas (Kaiser Fnd Fact Sheet). Although there are drugs to treat most of these diseases, new drugs are needed to cure rather than treat, to overcome or prevent resistance, and to make widespread treatment easier, affordable, and less toxic. As for current efforts, the leading product development program, the Drugs for Neglected Disease initiative (DNDi) has programs for Human African Trypanosomiasis, Chagas, pediatric HIV, leishmaniasis, malaria, and filariasis, and the Gates Foundation has prioritized filariasis, onchocerciasis (river blindness), and soil-transmitted helminthic infections in a recent call for proposals in its Grand Challenges Round 10.
Mike’s reasons for the portal are sound. A major bottleneck in finding new drugs (not only for neglected disease but all disease) is finding lots and lots of “maybe drugs,” chemical structures that have biological activity in predictive assays and can be run through more assays and tinkered with to find non-toxic, bioactive, and potentially therapeutic versions that can be tested in humans. It is medicinal chemists that design and tinker with the structures to optimize their activities along the multiple parameters required by the desired drug. So, since the universe of potential structures is vast, the number of possible targets immense, and the design and testing cycle fast, Mike wants to see more sharing of ideas, advice, and results, especially negative ones, early in the process, before the years of data accumulation needed for a publication. He also sees the portal increasing opportunities to form collaborations; to share preferred assays, endpoints, and models; to bundle work to get better prices from contract research companies; to aggregate and organize data for extraction of insights; and to promote the drug profiles desired by organizations (and companies) looking for candidates to develop into products. Mike also notes that the product development programs like DNDi, Medicines for Malaria Ventures, and the TB Alliance seem to have only modest pipelines and few or no back-up compounds for their clinical leads and need more maybe drugs. Also getting industrial-strength collaborators to further test the maybes is getting easier now that a number of the major pharmaceutical companies are supporting ND drug discovery, e.g., by making their compound libraries available and participating in data- and resource-sharing efforts like Re:Search (see last week’s post) and GlaxoSmithKline’s Tres Cantos Open Lab Foundation.
As for models of collaborative ND drug discovery, Mike mentioned two to me: the successful search for a synthetic pathway to make a pure form of the anti-parasite drug, praziquantel (Woelfe et al. 2011) and the on-going OpenSourceMalaria project (OSDD- Malaria). As for the mechanics of the portal, he may utilize the software and considerable expertise of Collaborative Drug Discovery, Inc., a company in California I have mentioned previously (“Drug Development on the Cheap”) and which is the host of a public access “vault” of structure-activity data for ND research (CDD Public Access). One big bug in the ointment of Mike’s proposed portal is intellectual property rights and whether/how to protect them. He told me that access would be password-protected and limited to those who agree to confidentiality to assure that pre-patent-application disclosure did not occur (I think he may need a legal opinion on this). Also, some academic research administrators get antsy about their researchers disclosing pre-publication data, as UCLA did when GlaxoSmithKline requested proposals for funding recently (“The Good, the Bad, and the Ugly”). Other than the concern for obviating later patents, there is the problem of “confusion” of who invented what should any participant of the portal (really his/her institution or company) file an application. Theoretically, a patent may have value; practically, it will have none (one of the reasons these diseases are neglected). But for academic institutions that have over-staffed technology transfer offices and lots of money to throw at patents and for companies concerned that their own patents will be open to challenge, the problem needs to be addressed. As I noted last week, the WIPO open innovation program, Re:Search, requires participants to agree to grant no-cost, limited licenses to the other participants, but this solution generates a huge overhead that Mike’s effort can’t afford. I thought up three alternatives that kick the can down the road and require minimal bureaucratic involvement upfront. The participating parties could agree not to assert any resulting patents against each other or they could agree that everyone is prospectively granted an non-exclusive license under any subsequent licenses that result or everyone agrees everything goes into the public domain and there are no patents.
Not surprising, the other bug is money and right now, like many entrepreneurs, Mike is taking the crowd-funding route to raise $25K to support a pilot version of the portal. If you looking to increase your 2013 tax deductions, you should check out his site at Fight NTDs. Long term support of the what are likely to be modest operating costs could come through participant fees (may be from a big pharma who sees the value) or from grants from the government (unlikely) or foundations (more likely). One obvious source (for all things global health) is the Gates Foundation, but Mike says his approaches to Gates and two of the groups it funds, DNDi and BVGH/Re:Search, have not been fruitful. I think more probing of Gates is warranted since it is heavily invested in PATH that has an open source ND project (OneWorldHealth Share) and at least one of its program officers has written about the value of medicinal chemistry and data-sharing in drug discovery (Elliott 2012). Mike’s portal is too good an idea to flounder, but it needs money and a snappy name.