« Prev Next »
The idea for using cultured cells to find and develop useful drugs is not new. In fact, this type of test tube experiments have been going on for decades, and features prominently in cancer research. But more recently, there has been a steady, but still tentative, shift in test tube disease-modeling. It may someday be possible to grow diseases in a dish that were previously impossible because they require an unusual amount of TLC to grow. Not only that, it may be possible to customize these cellular models for specific patients and even for that patient's organs. How is this possible? You may thank pluripotent cells and reprogramming technology. It now appears that not only can they provide us with spare parts, but these spare parts may also serve as bioactive sieves to help with drug discovery and development.
I've mainly focused on the regeneration side of things at this point, but this newer development cannot be ignored. Most biomedical researchers know that drug development is expensive, and the failure rate is high because so much of the initial work is done in non-human models. The meter starts running, and the bills run up fast, as soon as a drug discovery program is in place. Keep in mind that this is just testing new chemicals (concocted in-house, or from a shared resource) to see if they do anything biologically. If you stop to think about it, this is already troublesome because the type of biological system a researcher chooses to use (this may include whole animals, cells in a dish, or purified proteins) can bias the readout in a major way. For instance, a chemical may appear to be useful in cultured cell, but be completely useless or harmful in a whole organism; and what works in a mouse may not work at all in people. Worse than that, sometimes those side effects are not found until (and in rare instances after) the clinical trials, when millions have already been spent to painstakingly study the drug's effects in non-human models. The crux of the problem is that none of pre-clinical studies that are done in a test tube or a model animal can provide the big picture. This is because none of these models are specific enough. It's like trying to sift for gold dust using a colander with giant holes. Sure, you'll find the gold, but it'll be contaminated with a lot of junk too. Well, what if we use a living model system that is not only specific to a particular disease, but can also be tailored to a particular patient.
This is where pluripotent cells may really be handy. There are several reasons for this. First, one may account for individual genetic differences without the bias introduced by stock models. Most of the live models (either cells or whole animals) we have today are simply what we're able to keep alive and house easily, not what's most suitable for studying a particular disease. If we switch to a pluripotent cell-based system, and if we can have enough samples from enough people to represent a population, then we can conceivably sift out more of the junk and end up with a more potent (and maybe even safer) panel of candidates for further development. Second, current cellular reprogramming technologies are capable of immortalizing previously hard-to-grow diseases, thus providing an endless supply of model disease cells for research. This development will eliminate the restriction imposed by limited starting material and may advance understanding of some diseases. This may also be profitable because it would mean a broader range of potential druggable targets, or new applications for older drugs. Third, the same reprogramming technologies will also make it possible to study how an individual's genomic signature may influence drug metabolism in different organs. Reprogrammed cells or stem cells from an individual can be manipulated to develop into different cell types, which will enable drug evaluation in many different cell types from the same person and perhaps allow researchers to more readily identify potential side effects or toxicity in several organs. In short, pluripotent cell-based drug screening may be the closest we can get to testing new drugs in patients without actually doing it.
Of course, it's not all sunshine and roses. For starters, it's not a good model for diseases that are unrelated to genetic mutations, or for observing diseases that develop over time, or for tracking the complex disease-causing interactions between different organs. However, this pluripotent cell-based sieve can help make drug discovery more precise, and maybe help to evaluate their safety and utility. So, unless we can put aside human rights and inject people with un-vetted drugs, this may be the closest thing we have to a truly humanized disease model.
Image credit: FitzGerald GA. Anticipating change in drug development: the emerging era of translational medicine and therapeutics. Nat Rev Drug Discov 4, 815-818 (2005). http://www.nature.com/nrd/journal/v4/n10/fig_tab/nrd1849_F3.html
Further reading:
http://www.forbes.com/sites/johnfarrell/2012/06/01/screening-new-drugs-with-stem-cells/
http://www.nature.com/news/stem-cells-take-root-in-drug-development-1.10713
