New Computer Model Could Greatly Reduce R&D Spending
A new technique to map out alternate uses for existing drugs has the potential to lower R&D costs, increade profits and improve patient care.
New methods to determine alternative used for drugs already on the market could cut drug development costs and cut the time it takes to make treatment available to patients, according to a report in ACS' Journal of Medicinal Chemistry.
Truism of the day: it’s expensive to bring new drugs to market. To be precise, the cost of bringing a single new drug to market has far surpassed the $1 billion dollar barrier. That is a ludicrous figure; it is on the scale of government budgets. And, lest we forget, we’re talking about a single drug here. Start multiplying that in your head against all the drugs out there. The mind boggles – my distinctly un-mathematical one all but breaks down.
And this is before we even think about the number of drugs that don’t even get to the pharmacies, with the most recent high-profile failure being Pfizer and Johnson & Johnson’s Alzheimer’s medicine. While R&D spending on this drug has not been released, we all know that it will already have been substantial, just to get it to this stage.
The benefits of investing more energy into finding alternate uses for already approved drugs are clear, it increases the number of available treatments for patients and provides the pharma industry new profit-making opportunities without the sky-high development costs. But with current methods dogged by inaccuracy amongst other disadvantages, it hasn’t always been the best use of R&D spending to invest in this area.
The comprehensive new computer method devised by Sivanesan Dakshanamurthy’s team looks set to change that. The system is called "Train-Match-Fit-Streamline" (TMFS) and it uses 11 factors to quickly pair likely drugs and diseases.
The innovative new system maps new drug–target interaction space and also predicts new uses. It does this by combining shape, topology, and chemical signatures, including docking score and functional contact points of the ligand; and it has been shown to predict potential drug–target interactions with remarkable accuracy.
The method has already been used to find a new potential use for Celebrex: the popular prescription medicine for pain and inflammation has a chemical signature and architecture that could be successfully used to treat a virulent form of cancer. The have also found evidence that a hookworm medicine could be redeployed against cancer, by exploiting its ability to cut off blood supply. To have already found two potential new cancer treatments is indeed encouraging.
These are early days, but the initial results are undoubtedly positive. For patients who currently have a rare disease, there is often little hope, since the R&D needed to develop drugs is so high that the pharma industry simply can’t afford to invest such huge sums of money for a drug that will be needed by so few. This new computer model represents a life-line – a real life-line – for those patients. And even patients with less rare diseases will benefit: they could get the best ‘new’ drugs that most effectively treat their disease far more quickly than they would if the drug had to be created from scratch.
And finally, to end on a slightly self-interested note, this is, of course, positive for pharma: it is cheaper. Much cheaper.