By Nancy Weil, IDG News Service
May 15, 2004 | Keynote speakers shared their vision of the impact technology advances are having on biomedical research, new medicines, and patient care, while emphasizing the challenges of transforming tons of data into life-saving therapeutics.
‘Give Science the Opportunity to Fail,’ Infinity Chief Says
To succeed, pharmaceutical companies have got to focus more than anything else on their drug pipelines and products, said Infinity Pharmaceuticals CEO Steven Holtzman during his keynote speech.
Assessing the slim odds for most biotech startups, Holtzman said companies must balance the constant need for investors with the fact that it takes a long time to start making money, he said.
“It’s all about the products,” Holtzman said, urging companies not to fear failure, and to empower scientists to explore. Companies must “give the science the opportunity to fail often enough to succeed,” he said.
The failure rate for companies and products is high, and there is no shortage of reasons for that, Holtzman said. Developing new treatments costs a lot of money, though every potential drug target and every drug in a company’s pipeline working its way through clinical trials has “a value in the marketplace,” he said.
IT plays a key but incremental role in drug discovery, Holtzman said. Pharmaceutical companies need robust IT that allows “facile integration” of rapidly changing data and organizational transparency so information is shared, he said.
Successful pharmaceutical companies are those that can develop drugs more quickly, more cheaply, and with fewer failures than major pharmas like Pfizer and Merck, he said. That also means startups and small companies have to avoid selling product rights as a way to bring in capital, Holtzman said, emphasizing how crucial it is to hold onto what actually is likely to make money down the road.
He outlined the “big brother strategy” of partnerships with larger companies willing to make a substantial investment in return for a percentage of ownership in a smaller pharmaceuticals vendor. Those kind of deals can be the lifeblood of a small company, keeping it going long enough to give it a shot at taking products to market and turning a profit.
Reiterating that a company’s main focus has to be on developing better medicines that help peoples’ lives, Holtzman said: “If you focus on that and you’re smart and you’re lucky, you have a slim chance of success.”
Don’t Fear the Data Flood, Mayo Doctor Says
Doctors shouldn’t fear the vast amounts of data available to them, said Peter Elkin, director of the Laboratory of Biomedical Informatics in the Department of Internal Medicine at The Mayo Clinic’s medical school, during his keynote at the Health-IT World Conference + Expo. As a doctor, Elkin said he has an obligation to know about the genetic basis of the diseases he treats and should be able to respond to patients who show up armed with information from the Internet.
Elkin gave a broad overview of different types of technology, such as microarrays, gene profiling, and gene expression, as a way to show how bioinformatics is intersecting with daily work done in hospitals and doctors’ offices.
Over time, doctors will be able to compare information on individual patients with large databases about other people to come up with better diagnoses and treatment options, Elkin said. For now, however, information is available from myriad sources in various medical databases without effective ways to compare and analyze data.
This can be daunting to doctors, particularly given that the majority of those in clinical practice work in small offices with high patient loads and little time to stay up with what seems like an overwhelming amount of new research.
Although the opposite seems true at times, Elkin said, “things are not moving that rapidly” when it comes to discoveries that can be translated into clinical practice. What’s more, doctors need only stay abreast of the diseases they actually treat, he said.
Even so, the move toward personalized healthcare is well under way. That means “we’re going to need excellent health-IT systems” that doctors can use in their daily medical practices, he said.
Breakthroughs in communications, science, and technology -- from cave paintings to the manipulation of genetic information -- have all depended on information and how that information is gathered, recorded, and used, said Mike Ruettgers, executive chairman of storage-technology company EMC, in his keynote talk.
At EMC, the healthcare focus is on enabling customers to seamlessly and securely control the flow of electronic data, he said in the speech, which emphasized information life-cycle management. In healthcare, that means short-term and long-term storage of information, including such things as X-rays, and contending with regulations that require records be kept for specified periods of time. Ruettgers showed a slick video testimonial from John Halamka, CIO of CareGroup Healthcare System and also CIO of Harvard Medical School, to drive home his point.
As storage needs of customers have changed, EMC has responded by acquiring companies that fill gaps in its product and service lines, including buying Documentum and Legato Systems. Ruettgers later told Bio-IT World that EMC has its eye on a "small list" of gaps in its software portfolio and services, and expects to fill those through acquisitions.
Whitehead Institute for Biomedical Research director Susan Lindquist
's love for her work was palpable during her keynote presentation, entitled “From Mad Cows to Bio-Electronic Circuits.”
"The research we do at the Whitehead is quite amazing," Lindquist said. From a personal perspective, it is a "mind-blowing experience" to participate in pioneering biomedical research that helps humankind, she said.
Research at Whitehead is important for early detection of diseases, combating bioterrorism, making industrial processes cleaner, helping to contend with poverty and hunger, and dealing with emerging illnesses like the outbreak of avian flu in Asia, she said.
"The most extraordinary thing about the research is the sheer pace of it," Lindquist said, outlining a variety of ongoing research projects at the institute. Lindquist is a pioneer in stress response and protein-folding research and has focused much of her work on baker's yeast as a model organism.
Sometimes, It’s All About the RAM
Here’s how Mark Boguski, senior director of the Allen Brain Atlas Project, summed up the IT challenges faced by scientists: “At times it just comes down to having enough random access memory.”
The three-year project Boguski heads up to map the mouse brain will by its end have generated 1 petabyte of data, so he’s talking about serious RAM, serious storage needs, and serious high-throughput capabilities – and as he pointed out in his speech, some of it is available now, but a lot of it is not.
Despite the technological challenges, he does not despair. Those involved in the Brain Atlas Project at the Allen Institute for Brain Science in Seattle went into their work with the belief they would use the hardware and software available to them, augmenting it with extensions and in-house development.
Meanwhile, the 26 staff members, including scientists and IT specialists, cope with the fact that they are limited to some extent by existing technologies, and they work around those problems with “informatics tricks” and collaboration with IT vendors.
Boguski didn’t discuss any software they are developing themselves. Eventually, the project is meant to include a technology infrastructure that will be publicly available for geneticists, biologists, and other scientists to use.
Although the technology is expected to be an important outgrowth of the project, its overall purpose is to create a 3-D molecular map of the mouse brain that scientists hope to apply to human brain studies someday. The mapping involves taking images of slices of mouse brain. But each section of mouse brain that is made into an image takes up 50M bytes, so hundreds of terabytes are produced each quarter of the project.
“This, I think, challenges the capabilities of existing networks and storage today,” Boguski said.
Although he did not say which vendors the project is collaborating with, he did say that talks have been under way with IBM, Microsoft, and SGI about forthcoming technologies and products that could help.
Boguski offered a vivid display of one challenge scientists face, showing a slide of a dissected mouse brain. “You can tell that this brain has been sliced up and then put back together like a sandwich,” he said, noting that not all of the pieces matched up that well, and that type of issue is something the scientists are grappling with because the images have to be clean and clear.