The recent news from Biogen-Idec of the success of their experimental drug BG-12 for relapsing-remitting multiple sclerosis is in many ways remarkable. Although BG-12 is not approved yet, here are seven things about it which challenge many of the widely held beliefs about how we should best discover drugs today.
1. History: BG-12 is more commonly called dimethyl fumarate (DMF), known since the early days of chemistry as attested by its relatively low Chemical Abstracts registry number of 624-49-7. Synthesis of fumaric acid was first reported by Lassaigne as early as 1819, so it is nearly 200 years later that the use in multiple sclerosis is nearing approval. How can it be that today’s pharmaceutical industry, with its high throughput screening and computer modelling, cannot have identified this important use before now?
2. Chemistry: dimethyl fumarate has a molecular formula of C6H8O4, and a molecular weight of 144. That turns out to be the same as another remarkably simple drug, the anti-epileptic valproic acid. It has no chiral centres, or even any significant 3-D structure (excepting the methyl ester groups). For the medicinal chemist, it is also quite unusual for a drug not to contain a nitrogen atom, although DMF is not unique in this regard. So overall, while it may comply with Lipinski’s rules, it is not immediately drug-like to me.
3. Toxicity: DMF was used for a long while as a mould inhibitor, for incorporation in leather items such as sofas during storage. However, it is also an allergic sensitizer at very low concentrations (down to 1 ppm), producing extensive, pronounced eczema that is difficult to treat. In 2007, sixty Finnish users of leather sofas that had been manufactured with DMF in them were given serious rashes, and as a consequence the importation of products containing DMF has been banned in the European Union since 2009.
4. Previous uses: DMF (and other fumarate esters) have been proposed for the treatment of psoriasis for decades, and DMF itself has been approved in Germany for this use since 1994. Both psoriasis and multiple sclerosis share immune and inflammatory mechanisms in their pathophysiology. This not the same as saying that the use in multiple sclerosis is obvious based on the prior, known use in psoriasis, but nevertheless the two indications are related, and some therapies (e.g. steroids) are used in both.
5. Patents: as an old molecule, DMF itself has no composition of matter patent protection. This is unusual (although not unique) for a product of a large pharmaceutical/biotech company. The use of DMF for the treatment of psoriasis has been the subject of various previous patents (e.g. EP0312697) which have now expired. The use for the treatment of multiple sclerosis is the subject of various US patents such as US Pat. 6436992 and 7320999 (originally to Fumagen AG, but licensed by Biogen-Idec). There are other filings for the use of DMF in mitochondrial diseases (US6858750), atherosclerosis (US20100324327) and for controlled release formulations of DMF (EP1951206). Thus the commercial exclusivity available to Biogen-Idec as the developers of this may be bracketed by these other patents; moreover, there is potential in Germany for generic products of dimethyl fumarate to arrive when the patents protecting the psoriasis use expire, and be prescribed off-label for MS. It is also possible that the availability of this simple chemical through the internet will test the regulatory and legal system in the US in a way that we have not previously seen. It is in my view different from imports of products for erectile dysfunction, since these are primarily for occasional recreational use, whereas multiple sclerosis is a chronic and serious disease with a vocal patient advocacy network.
6. Mechanism of action: I have heard it said that the development of compounds without a well-understood mechanistic target is ‘bad science’. There are many such compounds used as pharmaceuticals, so this may represent an extreme or even naive view. Various hypotheses have been advanced for DMF’s mechanism of action, including modulation of the intracellular redox system resulting in consequential effects on nuclear factor kappa B, TNF-alpha, interleukin (IL)-8 and IL-1b. From then, we see effects on anti-inflammatory cytokines such as IL-10 and IL-1RA; decreased expression of adhesion molecules such as ICAM-1, VCAM-1 and E-selectin; stimulation of TH1 and down-regulation of TH2 T-cell subtypes; and induction of apoptosis in antigen-presenting cells and induction in anti-inflammatory stress protein heme oxygenase 1.
But no precise mechanistic target: so is this a deficiency in the regulatory package that Biogen-Idec will submit to the FDA; or if not, how will it address this question?
Edit: the mechanism of action has now been suggested to involve the Nrf2 pathway
7. Commercial: although DMF is an old, well-known compound with limited patent protection, it seems the commercial outlook is bright. Analysts are forecasting sales up to $1 billion or more, and Biogen-Idec are proposing to price it in line with other agents for MS (perhaps like Tysabri, which is another Biogen-Idec product). That figure equates to $100 million for each atom in BG-12 that is not hydrogen.
So there you have it, seven things that makes BG-12 one of the most unusual drugs in the pharmacopeia, assuming it is approved for medicinal use. It is also an example of drug repurposing, but there are plenty of those!