Open Science approaches to the treatment of a rare childhood cancer
The pharmaceutical industry has historically undertaken a consistent strategy for the discovery and development of new drugs for a range of different diseases, involving a competitive approach in which organizations maintain confidentiality of all information supported by patents to protect their inventions. This can result in duplication of efforts and the repetition of failures as different organizations are unaware of failed activities elsewhere. The not-for-profit virtual company, M4K Pharma, is pioneering a new experimental Open Science approach to the drug discovery process in which all available information is shared across the wider scientific community. Using this approach and focusing on diseases with an urgent unmet medical need (such as diffuse intrinsic pontine glioma (DIPG), the first disease currently being investigated), M4KPharma believes that the identification of a candidate drug will be achieved much faster by pooling available knowledge of both successes and failures along the way.
M4K Pharma is a wholly owned subsidiary of the Agora Open Science Trust and is led by Owen Roberts (CEO) and Aled Edwards (Director). The wider M4K project team comprises a number of different organizations, with the key drivers being the Ontario Institute of Cancer Research (OICR) in Canada (project management, chemistry and ADME), Structural Genomics Consortium (SGC) in Oxford, U.K. (cellular screening and crystallography), and Charles River Early Discovery in Harlow and Cambridge, in the U.K. (chemistry and drug discovery advice). The main project activities are currently funded by grants alongside provision of in-kind services, including the chemistry provided by Charles River Early Discovery in the UK as part of their Corporate Social Responsibility scheme and compound screening provided by Reaction Biology. In addition, the M4K group also works with a number of academic collaborators with expertise in the relevant field.
The team holds monthly project meetings, which are open to the Scientific Community to attend and are recorded and subsequently posted on YouTube with links from the company’s website. All the key information generated on the project is immediately available on the M4KPharma website, and the SGC team operates an open lab book policy. With all compound structural information being publicly available, no patents will be filed, and collaborators are free to share their results and are positively encouraged to publish their work. This means that other organizations not involved in the project will be unable to file patents and new chemical matter will not be owned by anyone. The open science approach allows for free and open discussion and also encourages organizations not directly involved in the project to provide support and new insight, as have scientists from Boehringer Ingelheim, AbbVie and Bayer. Once a potential candidate drug has been identified, the intention will be to use market exclusivities provided by the regulatory bodies to ensure that following approval, generic companies cannot immediately enter the market place. Given the length of time it can take to get a drug to market, these post approval exclusivities typically last for at least as long as any filed patent, especially in the case of orphan drugs.
The first project to which we are applying this approach is to discover inhibitors of activin A receptor type 1 (ACVR1, also known as activin receptor-like kinase 2 - ALK2) for the treatment of DIPG, a rare, aggressive and uniformly fatal childhood brain tumor. Mutations in the protein are known to be involved in around 25-30 percent of cases of the disease, and there are currently no known treatments. The diffuse nature of the tumor and the location in the brain stem make surgery impossible, and radiotherapy can only provide palliative relief. Hence, there is a dire unmet medical need for new treatments but in general most pharmaceutical companies cannot commit resources to something that will only treat at most a handful of cases each year; this is where an organization like M4KPharma can help using the Open Science approach.
Fibrodysplasia ossificans progressive (FOP) is a disease also known to be the result of mutations in the ACVR1 protein—in that case, somatic mutations—and there are a range of non-selective inhibitors of ALK2 under investigation for the treatment of this debilitating disease. However, most of these structural types are not brain penetrant and hence of no utility in the treatment of DIPG. Thus, it was important for M4K Pharma to either identify a starting point that was brain penetrant or to have a clear strategy to improve the brain penetration of known inhibitors. The first approach has led the team to identify a lead series, which started from the moderately active, but orally available and brain penetrant ALK2 inhibitor LDN-214117.1 Modifications to this structure led rapidly to the identification of the early lead compound, M4K2009. This compound has an excellent in-vitro profile, with low nanomolar activity on ALK2 and good selectivity over the related isoform ALK5, which is required to avoid any potential cardiotoxicity. A crystal structure of M4K2009 has recently been solved at SGC (Figure 1) with a good resolution confirming the anticipated binding mode in the ATP binding pocket of the protein.
Figure 1: Structure of M4K2009 and crystal structure bound in the ATP binding pocket of ALK2 protein to 1.4Å
Further profiling of this early lead has shown it to have a good in-vitro ADME profile, with high metabolic stability and good permeability with no efflux. This has translated into a good in-vivo profile, with excellent oral bioavailability and good brain penetration. Further in-vivo evaluation has shown that the compound has good linearity in dose escalation studies and is tolerated in mice up to 100mg/kg once a day for five days with no deaths, body weight loss or gross pathology. Further analogues have been prepared that have similar or improved potency, selectivity and DMPK properties.
To confirm functional target engagement in an animal disease model, we evaluated compound MK2009 in a mouse model of FOP at the University of Pennsylvania (Dr. Shore lab). This model was used due to its ability to measure inhibition of ALK2-mediated signaling in a disease relevant system. The objective was to evaluate the ability of M4K2009 to inhibit Acvr1R206H induced heterotopic ossification. Following dosing at 100mg/kg per day for 2 weeks in the R206H mutant mice, a significant reduction in the formation of the abnormal bone growth typical of the disease state was observed compared to the vehicle treatment. No adverse effects were observed from this study and evaluation of additional doses of M4K2009 as well as evaluation of other compounds is currently under consideration. As the mutations present in DIPG may not be exclusively the same as those in FOP, we are now considering how to test activity against other mutated forms of the target in vivo.
As a backup to the extensive work carried out on the lead series, we have also considered all the literature around ALK2 inhibitors that has been developed for FOP. As mentioned previously, many of these series consist of compounds that have properties that are not suitable for brain penetration. We have prepared samples from a number of these series to confirm the activity in our assay at Reaction Biology Corp. The main focus of our efforts has been around the series described in the BluePrint patent2. This provides compounds that are highly potent against ALK2 with excellent selectivity over ALK5, as demonstrated by BLU-782, presumed to be a compound from this patent, which has recently entered Phase 1 as a potential treatment for FOP. However, the one compound evaluated in vivo shows poor cell permeability and lacks brain penetration. We are now using this series as a starting point and using a multi-parameter optimization approach to identify a new series with the desired properties to enhance brain penetration.
In conclusion, identification of the early lead has allowed the project to progress through to the current stage in less than 18 months. Adoption of the Open Science approach to drug discovery has allowed input into our project from a wide variety of different organizations, from academics to large pharma, providing help and guidance to progress the project through to a proof-of-concept animal disease model. Progression through the next stages of the project remain to be validated within the Open Science framework so that we can reach the market place a soon as possible with a drug for these children who currently face no future.