He has also guided Amgen (AMGN
) in increasing its research partnerships with small, innovative biotechs. The latest such deal for the Thousands Oaks (Calif.) company: a pact with Tularik, a San Francisco-based outfit, to develop cancer treatments. This complements Amgen's groundbreaking work in developing drugs to treat anemia, rheumatoid arthritis, and infections.
Perlmutter joined Amgen in 2001 from Merck Research Laboratories (MRK
), where he was an executive vice-president. He also serves on the board of the Institute for Systems Biology, a research group that brings together scientists from many different disciplines who work together to decode disease pathways. He recently spoke with BusinessWeek Correspondent Arlene Weintraub about the R&D process. Edited excerpts of their conversation follow:Q: What's the difference between contemporary biological research and systems biology?
A: I use the metaphor of a car. You can take it apart and look at the pieces. Perhaps you find something round and then discover it's a CD player. That's how contemporary biology works. You look at one component of a cell at a time.
In the systems approach, you look at the whole machine. You study the movement of the crank and the wheels, and you learn it's completely unrelated to the CD player. In the body, you look at all the genes, when they're expressed, and how that correlates with reactions in all the different parts of the body.Q: What's most challenging about the systems approach?
A: You need to work in teams. That's not easy for most scientists. Individual scientists view themselves as artists. They want to make some brilliant discovery and be celebrated. The idea of working with 200 other people is very unsatisfying. And different scientists don't speak a common language.Q: How is the biotech industry working to overcome that barrier?
A: At Amgen we work in teams of different types of scientists. But the systems approach is still only a narrow part of our R&D effort. We spend no more than 10% of the R&D budget on it. The other 90% is spent researching drugs the traditional way -- one gene target for one disease at a time.Q: Do you devote so little to the systems approach because the potential payoff is still unclear?
A: That's right. Ultimately, understanding how things work will help us make better drugs. But fundamentally our job is still to make drugs. Right now the industry is awash in potential drug targets. Should I invest in a major systems-biology effort, or should I look at drug targets we've already identified that might work and make some money? Of course I have to do the latter.Q: Will the dream of so-called personalized medicine ever come to fruition?
A: It would require a change in thinking that's very broad. It stands to reason that if we had real knowledge of diseases, we could develop tests to predict them and give drugs to people that would alter their prognoses. But then you have to ask: What level of probable disease will be significant enough to make people willing to take a $2 pill every day? That's a fundamental question that still needs to be answered. EDITED BY Edited by Patricia O'Connell