« MYTH: Great Management is King | Sacred Cow Dung Home | MYTH: 10x "Best Case" Return is Compelling to Investors »
June 14, 2005
Target ID and Validation for Orphan Diseases
[ BioTech and Pharma VCs Beware: This article is yet further evidence of the Open-Source Drug Development Trend that is building rapidly and challenging the traditional pharmaceutical industry. Open-Source Drug Development will ultimately have a similar and equally devastating disintermediating effect on the pharmaceutical industry that Open-Source Software Development is currently having on the traditional software industry. — cgm]
Diseases such as malaria and dengue fever may not get the attention they deserve, but academic researchers and some companies are trying to change that
The Orphan Drug Act defines an orphan drug as one that "treats a disease affecting fewer than 200,000 people in the US, or which will not be profitable for a period of seven years" following approval. But that definition does not hold, says Joseph DeRisi, PhD, for diseases like tuberculosis and malaria. "Malaria affects 500 million people; there's somebody that dies every 15 seconds, 2.7 million deaths a year. That's not an orphan disease, it's an orphan cause," says DeRisi, a professor of biochemistry and biophysics at the University of California at San Francisco. DeRisi and colleagues analyzed gene expression in Plasmodium falciparum, the cause of a deadly form of human malaria, using DNA microarrays. They say their findings have uncovered a weakness that could be exploited to treat the disease.
Marlene Haffner, MD, MPH, director of the US Food and Drug Administration (FDA) Office of Orphan Products Development, says work on orphan diseases is an ongoing effort. "There's a lot happening all around the world in the arena of orphan drugs and orphan drug development. I just see more collaboration around the world which will result in better therapies for all patients with rare diseases. I think it's an idea whose time has come, and with the human genome, I think we're going to know more and more about rare diseases fairly quickly." But when it comes to diseases like malaria, tuberculosis, and dengue fever, diseases which most often plague poorer countries, the trends have been "rather bleak," says Paul Herrling, PhD, chairman of the Novartis Institute for Tropical Diseases (NITD), Singapore. That is something that institutes like NITD and researchers like DeRisi and Guy hope to change.
click the image to enlarge
The 48-hour transcriptome of intraerythrocytic development for Plasmodium falciparum. (Source: Joseph DeRisi)
They found the system was a highly cyclical one in which genes are expressed once per cycle "in sort of a just-in-time fashion." What is unique is that the transcriptome more closely resembles something such as late herpes viral gene expression than it does a typical eukaryote such as a fungus or a human cell. This fact could be exploited as an Achilles heel. The researchers believe the organism may not have the capacity to respond to its external environment to the degree that other organisms do, that it has become more streamlined over the course of evolution, more stripped down, more like a virus than a typical eukaryotic cell.
"That's very interesting, because any regulatory things disrupted midstream would cause catastrophic failure to the organism," DeRisi says. "If you find a validated, essential target and you can effectively make a compound for it, it argues that the parasite has very little in reserves to be able to combat it, except for mutation." Guy says malaria has a constant environment because it lives in the red blood cells of humans where there is no new protein production and very little metabolic activity. "It didn't need to be able to adapt, unlike a bacteria, where everything really circulates, so from our perspective it makes it more likely that a central pathway target will be critical. The only real option the organism has to evade it, to gain resistance, is to mutate, rather than to change a pathway or to change a response."
Guy's lab has also researched producing candidate compounds for several other orphan diseases. They are largely finished with a cystic fibrosis project that yielded validated leads. They are now running the chemistry portions of a target identification effort for prion disease, and they are also working on Trypanosoma brucei, the cause of African sleeping sickness. Guy says their labs both have grants from the National Institutes of Health for a target-oriented proposal. He adds that genomics work done in DeRisi's lab led to a collaborative effort to identify targets from genetic code which are substantially different from human gene products and then to systematically address those with small molecules. "We're really bringing all the standard drug discovery methodologies to bear."
But discovery is only the beginning. When it comes to later-stage work, things become more difficult. That requires a validated lead compound, something with efficacy in an animal model. Then researchers must go back to NIH to convince them to pay for good laboratory practices manufacture, preclinical work, and the start of coaching trials. Guy says he is moving at the end of the year to Saint Jude Children's Research Hospital, Memphis, Tenn., because they will allow him to set up a hospital-funded preclinical development pipeline for pediatric orphan diseases. "That enables me to really emphasize preclinical development, which is difficult to find, to be quite frank."
Access to tools, expertise
Unlike Guy and other academic researchers, researchers at NITB can draw on the vast pool of tools and expertise developed by Novartis over the years, says Herrling. "Being part of the Novartis family, we have full access to the entire Novartis library . . . We can screen all of our pathogen targets, and we would then select and optimize just a few of those which show effects, and thereby not contaminate the rest of the library and keep it for other targets that other people are interested in screening." They could also build on the technologies the company used to develop targets for Alzheimer's disease and diabetes and apply them to diseases like dengue fever and tuberculosis. Dengue fever is a viral disease, so NITD is now applying work Novartis has already done on AIDS and other viral diseases in which the number of targets are relatively limited. NITD is also working on the growing problem of the resistance that Mycobacterium tuberculosis has developed to two of the drugs most commonly used to treat it, isoniazid and rifampicin.
The entire M. tuberculosis genome was sequenced about five years ago, but the function of a significant portion of the genes is unknown. Herrling says work has been slowed because M. tuberculosis does not have the Dicer enzyme, which is crucial for highly efficient small interfering RNA (siRNA) methods. "The work has been slow, and we are currently searching for methods where, in a high-throughput manner, we can try to determine gene function in the tuberculosis bacterium."
Herrling says a promising tuberculosis lead has come in the form of peptide deformylase inhibitors, a new class of antibiotics that Novartis is collaborating on with Vicuron Pharmaceuticals Inc., King of Prussia, Pa. This class of antibiotics targets a protein essential for bacterial growth and provides the basis for selective activity against a number of bacteria. It has also demonstrated promising in vitro and in vivo activities against bacteria resistant to widely used penicillin, cephalosporin, macrolide, and quinolone antibiotics. "This was a target we developed for upper respiratory disease infections for the developed world and now, using bioinformatics, we saw that Mycobacterium tuberculosis has the same enzyme, so we could build on all the work that has been done in the commercial labs for our tropical disease efforts." Some of the same targets can be detected at the bioinformatics level in some of the parasites responsible for Chagas disease, says Herrling. "Because you have these bioinformatics tools, these things as sequences, you can very quickly try to see and select targets that are shared by more than one pathogen. Then possibly with one drug, you could affect several of them. That's certainly a trend we are following very vigorously."
While they don't have the funding that large companies like Novartis can supply, Guy says the tools and techniques they use at UCSF are comparable and include high-throughput screening, high-throughput chemistry, structure-based design, and analytical chemistry. "I think we're actually very, very similar to modern medicinal chem. In fact, many of my colleagues in industry are surprised at how similar, generally, the way we run our operation is to what they do." DeRisi says his lab takes a different approach, because there are no basic malaria biology programs at large pharma companies to model themselves on. "We take a functional genomics approach. We take basic biology and biochemical approaches, along with a heavy emphasis on bioinformatics and phylogenomics to try and identify pathways that are interesting to look at. Then we feed them into the chemistry unit."
Guy says those approaches have resulted in "major" areas of synergy between the labs. "We've been involved for quite some time in cheminformatics like library design, extraction of data from screening results, rapid evolution of libraries and materials. So as we've started to work closer and closer together, we've brought those two informatics approaches into unity." They are now identifying unknown gene products that look promising, comparing them to a known gene product in other organisms, extracting ligands known to bind to that gene product, and using those as screening leads. "We try to move quickly and intelligently because we're resource limited. It drives us toward doing things as efficiently as possible."
DeRisi and Guy believe orphan diseases will continue to garner more attention and advances, and that an open-source drug development effort could also play a vital role. Under this model, research volunteers could use a variety of computer software, hardware and databases to search for new protein targets, find chemicals to attack known targets, and post data. This, proponents believe, would help contain the costs of drug discovery, development, and manufacture. "I would say we're still actively engaged in working out how to do the proper strategies to use open source, which speeds things greatly, without causing any risk to the industrial partners. But I think there's a clear understanding now in the funding agencies, in the potential industrial partners, and developing in the academic community, that this needs to get worked out, and it will get worked out in a way that's profitable for everybody."
© 2005 Reed Business Information a division of Reed Elsevier Inc. All rights reserved.
Posted by cmayaud at 03:23 PM | Permalink| Comments (0)
Del.icio.us Tagging | Digg This | Posted to Business Strategy | Online Business Networking | Open Source | Pharmaceutical Industry | Political Economy | TRENDS | Venture Capital Process | eHEALTH
Post a comment
Thanks for signing in, . Now you can comment. (sign out)(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)