Learning to deliver: Challenges in NTD drug research and development

Graeme BilbeDr Graeme Bilbe, Research & Development Director, DNDi
[September 2014]

DNDi and other not-for-profit drug research and development (R&D) organizations have explored new R&D pathways to develop and deliver safe, adapted, and affordable treatments for neglected diseases. Most of these new pathways have shown signs of success, while several have faced challenges that span across the drug discovery and development processes.

With currently 30 projects from screening of molecules to implementation of treatments, DNDi’s core focus has been, and is, primarily on the three kinetoplastid diseases (leishmaniasis, human African trypanosomiasis, and Chagas disease) and while preparing the completion of our malaria projects, DNDi took on filarial diseases and paediatric HIV projects. While initial successes were based on the short-term strategies of repurposing, reformulating, or combining existing drugs in new regimens, the pipeline now includes 13 entirely new chemical entities, one of which is currently in Phase II/III testing. The robustness of the portfolio compared to a decade ago, has been the result of solid partnerships worldwide with public and private organizations, both bilateral partnerships with DNDi or within research consortia and platforms. With the momentum created through these alliances, DNDi’s overall objective is to deliver 11 to 13 treatments by 2018. The goal is reachable, but there are a few challenges that illustrate the work and scientific barriers to overcome in building such a portfolio for neglected diseases.

New compounds for neglected diseases require new approaches
In discovery, to achieve relatively rapid R&D success, and here we mean the identification of potential and promising new compounds, and exert a real impact on treatment of neglected diseases, open access to valuable compound libraries from pharmaceutical, biotechnology, or academic partners is vital. It requires strong alliances, willingness of all parties to share not only molecules but also the know-how and data behind them, and a shared vision of testing new approaches, including innovative intellectual property (IP) licensing practices. One example of this is the SCYX-7158 compound against human African trypanosomiasis, which is now completing Phase I trials. While DNDi has managed to bring new chemical entities into its portfolio, it should be said that accessing larger collections of greater diversity than has been the case to date is necessary to accelerate the discovery of quality validated ‘hits’. New computational tools which learn from previous screens are helping us to be more intelligent in selection of new libraries.

Gaining greater understanding of how to predict whether a drug will cure
So how do we actually predict whether a drug will ‘behave’ in the same way in vitro, in vivo, in healthy volunteers, and in sick patients? Indeed a great challenge is that we still have a lot to learn about the predictive models we are using to determine which drug candidates will be effective. There is also the challenge of reaching what we call a ‘sterile cure’, meaning that a drug may seem to cure a disease, but can sometimes leave even virtually undetectable parasites behind. In some cases, this can lead to relapse, other disease manifestations, or asymptomatic disease reservoirs. So in many cases, at DNDi we have studied the models themselves as we study the molecules, and we have learned a lot along the way, sometimes retrospectively.

For example, the E1224 proof-of-concept trial, as well as another trial with posaconazole, led the Chagas research community to further question and investigate the models used to predict efficacy of a compound. These studies not only clarified the need for and potential of better regimens with existing drugs (i.e. benznidazole in shorter courses or in combination with other drugs), they also accentuated the importance of identifying useful markers of therapeutic response such as those we are testing now (‘biomarkers’) and provided a tool to challenge predictive models. A positive outcome of the E1224 trial is that we have been able to review and refine both in vitro and in vivo models so that they now better predict clinical outcome, which should help us to increase the odds of getting entirely new compounds into, and past, clinical testing.

Each disease poses its own difficulties in building a pipeline
Because we know that drugs can fail during the drug development process, keeping a solid pipeline of ‘backups’ progressing steadily in the research and translation phases is of paramount importance, a real game of checks and balances in the pipeline. For example, for sleeping sickness we have an improved therapy already reaching patients, a new chemical entity (NCE) in Phase II/III and another NCE slated for Phase II soon, with backups as well in research phase. Chagas disease, on the other hand, is more challenging, as we have described above in terms of predictability of getting a complete or ‘sterile’ cure, and we thus have a gap in the early clinical stage of the pipeline. We do, however, have promising projects at the proof-of-concept trial stage.

Leishmaniasis, notably the visceral form, has shown extreme diversity in treatment response depending on geographical location – even within one region, but especially among continents. Treatments that work for visceral leishmaniasis on the Indian subcontinent, for example, show significantly lower levels of efficacy in East Africa and Latin America, emphasizing the need to better understand disease strains and treatment response. Although we have made great strides in developing combinations of current drugs with efficacious and much improved tolerability profiles in Asia and East Africa, current therapies still suffer the drawbacks of painful injections and potential side effects. We do have a better pipeline today for this disease with one oral molecule currently in a proof-of-concept study in Sudan and another one slated for Phase I next year and a few well-characterized lead series in late-stage optimization, both at DNDi and with partners.

It is possible to conduct trials in remote settings
Conducting clinical trials in resource-limited settings in accordance with international Good Clinical Practice (GCP) standards, has proven to be challenging, requiring investment in the rehabilitation of infrastructure, provision of equipment, capacity strengthening and continuous training. Nonetheless, the conduct of GCP clinical trials even in very remote settings has become a reality. For example, the current Phase II/III trial on fexinidazole for human African trypanosomiasis in the Democratic Republic of the Congo (DRC), where several sites were refurbished and for which more than 250 professionals were trained, shows that conducting a complex pivotal trial, while essential to drug development, is also beneficial overall to the hospitals and staff: a true mutual benefit. The challenges are there, but they can be overcome.

Clinical trials can also be designed to support policy changes, especially when endemic countries are involved from the outset and stay involved through to implementation and access.

Even when a treatment is delivered, challenges of implementation and access remain
Indeed, there is a great need to ensure that implementation studies are carried out to inform and accompany the policy changes required to ensure patient access to treatments. Several additional challenges await even those treatments that have proven safe and effective in clinical trials, including the need for regulatory capacity strengthening and harmonization among countries where neglected diseases are endemic, and the need to avoid fragility of production, notably when there is only one producer of a drug. Indeed, while under certain circumstances drug donations can be appropriate and be utilized to support specific elimination targets, it is important that with a vision of sustainability, the price of a treatment remains affordable but also allows a minimal margin of profit to ensure that supply is continuous.

For sure, R&D for neglected diseases still has room to ‘learn to deliver’, but requires a more conducive environment in which priorities are determined by public health leaders, R&D development is coordinated in such a way that duplication of research is reduced or stopped, and sustainable funding is secured.

Dr. Graeme Bilbe, Research & Development Director, DNDi