You might plausibly argue that ice cream parlors and bobby socks reached a peak back when Elvis was still the king, but drug discovery? Not a chance, which is why it's all the more alarming that the way we're developing new medications is stuck in some 1950s time warp. Consider that the number of new drugs that hit the market each year hasn't changed much in the last 60 years, says Dr. Garret FitzGerald, the dean for translational research at the University of Pennsylvania's Perlman School of Medicine. "The process of drug development has broken down," he says.
Drug development is an expensive, inefficient, drawn-out, failure-ridden process that hasn't kept up with new technologies, FitzGerald argues. The companies developing new medications seem myopic, focusing on a handful of diseases. Most of the drugs approved last year were for rare diseases. There's also a dearth of new drug candidates, and most of those that look promising often don't make it past clinical trials because they're found to be ineffectual or unsafe.
It takes years and billions of dollars to get a drug into your medicine cabinet, so it's no wonder companies charge small fortunes for drugs that may not work, or worse -- that turn out to be harmful.
This, FitzGerald says, is not only worrisome for public health, but as a business model for big and small pharma as well as biotech, it's just not sustainable. He recently gave a lecture about what's ailing this industry and offered up some remedies for fixing its problems at this year's meeting of the American Association for the Advancement of Science in Boston.
Wired caught up with FitzGerald to pick his brain on the topic.
__Wired: __Why should consumers especially care about what’s going on in the drug business right now?
__Garret FitzGerald: __If a drug gets approved, all that really counts for you is if it’s going to work in you. Everyone will need a drug at some point in our lives. The average person in their 60s and 70s is taking six or seven drugs. That probably reflects a combination of the fact that many don’t work very well. It’s a waste of [patients’] time and money if it’s not going to work in them. We’re moving toward having comprehensive genomic information, but right now we’re doing population pharmacology rather than personal pharmacology. That’s why you should care.
__Wired: __How has the process of drug discovery stagnated?
__FitzGerald: __We have a real problem with human capital. We have a lot of new technology, [but] we really need people who are familiar with preclinical science, contemporary technologies, and investigation of drug action in humans. Those types of people have really become like snowflakes in summer.
__Wired: __Has the process changed at all in the last few decades?
__FitzGerald: __It used to be that it was dominated by large, vertically integrated companies, like Merck or Pfizer, where everything would be done from soup to nuts within the company. We’re moving now to a more modular approach. Different parts of the process from discovery of a target all the way up to an approved drug are going to happen in different places, geographically but also in different sectors – academia, big pharma, biotech.
If you could mobilize the best people to collaborate to deliver an outcome, you could imagine it would be more efficient and perhaps cheaper. It turns out we actually have a proof of concept in the altruistic sector [for developing malaria drugs]. Because people perceive there’s no money to be made, they’ve been willing to collapse their intellectual property or at least their expectation. The question is can you export that into the for-profit sector?
__Wired: __So the patent frenzy is stifling innovation in drug development?
__FitzGerald: __The issue of intellectual property has been a bit of a dead man’s hand on the tiller in terms of industry’s [and academia’s] willingness to collaborate in an open fashion. There have been a variety of initiatives to expand the parts of the process where parties may be able to collaborate unconstrained by intellectual property.
Right now, the odds of a chemical becoming an approved drug is about 1 in 40,000. Those are very long odds, and yet the way our intellectual property is configured is that it’s vested in chemicals. That worked fine when the whole process was performed within a vertically integrated company because if that chemical became a drug, the company would own [it all]. If you move to a modular [system], you’re asking downstream collaborators to participate when all the intellectual property is vested with the chemist. Why should I play [for free] in a for-profit world?
__Wired: __What’s the fix?
__FitzGerald: __One of the ways we could go about this would be to utilize the clinical and translational science awards fostered by a new center in the NIH called NCATs [National Center for Advancing Translational Sciences]. NCATS has proven to be quite controversial. Some people have said [academia] has no place in drug development. It should be all in the private sector. Other people have said no, that the objective here is to provide the infrastructure, the education and the organization to allow investigators to collaborate in drug discovery and development.
People are also beginning to think of all sorts of ways about how money might be raised. As opposed to a Pfizer bankrolling the whole thing there might be other ways to raise money from the market.
I'm suggesting we model prospectively where the biggest hurdles [in discovery and development] actually are in the process of going from a chemical to an approved drug. We'd keep revising the model [along the way] so the model actually reflects what happened. We'd then assign the patent accordingly.
__Wired: __How will drug discovery and development evolve?
__FitzGerald: __The current approach is designed to detect very large signals of either benefit or risk. That approach is demanding and of limited value. The future of this business is going to be focused a lot more in trying to interpret the variability of drug response. Why is it that some people respond very well, while some people don’t respond at all? Can we predict the people that are likely to respond well and only expose them to the drugs? That’s the challenge at the moment.
