Common to Both Academia and Industry

The Challenge of Discovery

Perry Molinoff recognizes the distinctions between basic and applied science, between academic and industrial research, and between the preclinical and clinical realities of drug development. But he generally discusses these categories in fluid, practical terms, having throughout his career crossed the lines of distinction that have sometimes been rather heavily drawn among pharmacologists. As a third-year medical student at Harvard, he decided “to take a year off” to conduct laboratory research. After receiving his MD and pursuing further clinical and postdoctoral work, he enjoyed an academic career that included fourteen years as the A.N. Richards Professor and Chair of Pharmacology at the University of Pennsylvania School of Medicine. He has just completed six years as Vice President of Neuroscience and Genitourinary Drug Discovery for Bristol-Myers Squibb and will soon return to teaching, in the Departments of Psychiatry and Pharmacology at Yale University. Referring to himself as either pharmacologist or neuroscientist, depending on context, he has made fundamental discoveries in receptor biology, has overseen the discovery and development of drugs and their subsequent clinical trials, and has mentored a host of pharmacologists and neuroscientists who themselves have established careers in industry and academia. The pursuit of discovery as its own reward emerges as a theme that has marked his professional life (and is perhaps reflected also in the images displayed in his office of the Himalayan mountains, photographed by Molinoff himself from the Everest base camp last year).

MI: Your career has given you perspectives of both academia and industry. But you started out by pursuing academia. How was it that you first chose to be an academic, as opposed to an industrial scientist?

PM: I was finishing my postdoc with Julius Axelrod (Nobel laureate, 1970) in 1970, and I was about to start looking for a job. A friend of mine, Lincoln Potter, who was a Research Associate with Bernard Katz (Nobel laureate, 1970) at University College in London and had been in Axelrod's lab before me, came through and said, “Instead of taking a job, why don't you come to England for a couple of years?” It seemed like a nice idea and so I spent almost two years in England, working on α-bungarotoxin and cholinergic receptors. We published the first paper in which this toxin was used to label nicotinic receptors in vitro.

MI: That kind of made a big splash, didn't it?

PM: It made a terriffic splash. It was funny: We submitted our work to Nature.1 During the ensuing weeks we received no acknowledgment from the editorial office, nor did we ever receive any reviews from the Editor. About three weeks after we had submitted the paper, I was on a London bus riding to work and read in the London Times a discussion of our paper, which is how I found out that our manuscript had been received—and accepted! We had never seen proofs, we never received a letter of acceptance. Nothing.

MI: Did you know that your manuscript was that hot?

PM: Well, yes. I mean, people knew that receptors existed, and people were trying to figure out a way to use biochemical techniques to study them. Neuroreceptors at that time were largely defined operationally. There were a few papers out using tritium-labeled ligands, which may or may not have shown a detectable signal—it was at that borderline level.

But we didn't come up with the idea of using bungarotoxin on our own. Bungarotoxin comes from the Formosan krait Bungarus multicinctus. Lee and Chang, working in Formosa, had already shown that they could label the crude venom and that components of the venom would bind to the motor endplate region of the innervated diaphragm. They then showed that if they denervated the muscle, which causes a spread of sensitivity, the labeled toxin bound diffusely. So there was already data that said this approach was going to work—we didn't come up with it. What Link did was to purify the active toxin from the crude venom. And then we labeled the toxin and used it in vitro to study the receptor.

MI: And after England you pursued an academic career. The professorial life certainly must have appealed to you—you remained at the University of Colorado Health Sciences Center for ten years before taking the Chair at Penn. What finally brought you to industry?

PM: The thing that brought me to industry was really the belief that advances, largely fueled by molecular biology, had reached the point where it would be feasible to approach neurological and psychiatric diseases through programs of rational drug discovery. We can now think in terms of identifying a specific molecular target and then finding a specific compound to interact with that target to treat disease. This is equivalent to what was being done in cardiovascular research about twenty years ago—where you knew many of the targets and you could devise therapies for them. But in the treatment for neurological and psychiatric disease at that time, what was happening really amounted to throwing compounds at crude animal models with limited predictive validity.

I went to Bristol-Myers Squibb thinking that it was possible to do something different. And, clearly, I was correct. Some of the targets that we worked on at Bristol-Myers Squibb, and that other companies are working on, are very carefully and precisely defined and are specifically relevant to neurological and psychiatric diseases. Just as one example, I think most people in the field believe that lowering levels of β-amyloid may lead to the development of a neuroprotective approach to the treatment of Alzheimer's disease. We now have knowledge of the synthetic pathway that leads to the production of β-amyloid. We are starting to get an understanding of the pathways involved with the degradation of β-amyloid. And so you can now think about developing specific therapies that act on these pathways. That couldn't be done before.

And of course there is always both a push and a pull when one makes a career change. So the pull to industry was clearly the opportunity for a new challenge. But the push in that case was that I had just been given my third six-year term as Chair at Penn. And I was starting to get a little bit of the feeling that I had been there, done that, and even had the T-shirt. And so I wanted to do something different.

MI: Did the burden of administrative responsibilities also become part of the push from Penn and into industry?

PM: No. My experience at Penn was actually a very good one. Yes, I had to be involved in administrative issues and at the same time try to run a lab. The thing that made it possible is that I also had an excellent support system. And in fact my laboratory was larger and more productive at Penn than it had been in Colorado. When I went to Penn and they asked me what kind of staffing I needed, I based my request on what had been available to the Chair at Colorado. And again, my motivation in moving is always that I never want to do things that somebody else can do equally well. I want to do things where I have something unique to contribute. And I was fortunate enough to have some superb colleagues—administrative as well as scientific—at Penn that made my life easier. At Penn there was strength and depth throughout the whole scientific enterprise. And so you had collaborations and interactions that made a lot of things possible. It is very important to be able to leverage positive collaborations, both in industry and in academia.

MI: And what kinds of collaborations exist in industry?

PM: One of the most exciting things about industry is the level of collaboration. Everyone is in the same boat and everyone recognizes the need to collaborate. The most obvious collaboration is between the biologists and the chemists. Each discovery project is co-chaired by a biologist and a chemist, and involves people from a variety of disciplines, including experts in pharmacokinetics, drug safety, process chemistry, and drug formulation. In addition, there is close collaboration among discovery scientists, clinical scientists, and people from the business world with expertise in sales and marketing. Another type of collaboration exists with scientists at universities and other companies with whom a specific alliance is established. I was actually involved in four significant external alliances for products, two of which are potentially breakthrough products. The most interesting and exciting of these was with Otsuka Pharmaceuticals, a company based in Japan that discovered a new antipsychotic called aripiprazole. Bristol-Myers Squibb and Otsuka are co-developing this product and I was a member of the product development committee that oversaw the whole process. This is a very important project to both companies and a lot of effort and resources are going into the program.

MI: And scientifically, what would you say are some of the most exciting areas in industry that you've been involved with?

PM: One of the projects that continues to receive the most media attention is the clinical development program for Alzheimer's disease, which I led. Alzheimer's is a very complicated system to study. We don't really understand what the etiology is. And the medical need is so enormous. Up to fifty percent of people over the age of eighty-five have pathologic signs of Alzheimer's. It's a disease with tremendous cost both to the individual and to society, and with the aging population, the problem is getting bigger. There is really nothing out there that is therapeutically effective. Cholinesterase inhibitors provide mild symptomatic treatment for a short time, but there are no data suggesting that they have a major effect on disease progression. Alzheimer's is probably the brass ring right now on the pharmaceutical industry merry-go-round, at least as far as neurophsychiatric disease is concerned.

MI: You mentioned the role of β-amyloid in Alzheimer's. Has it been established whether β-amyloid plaques are necessary for disease symptoms, or is that an open issue?

PM: It's still an open issue, to some degree. What we do know is that genetic abnormalities that lead to an increase in β-amyloid—soluble β-amyloid, that is—are sufficient to lead to Alzheimer's at an early age. And there are a number of different mutations, on different chromosomes, that can have this effect. Mutations of the genes that encode enzymes called presenilins are associated with early onset: if gene constructs containing these mutations are put into cells, you get an increase in the release of β-amyloid. So there's an awful lot of data to suggest that the release and accumulation of β-amyloid is integral to the process. Can you prove yet that it's directly responsible for the disease? No. Can you say that it's the only thing? No. Can you say it's first? No. But it is integral.

MI: And what's been done to target that system?

PM: The targets that have seemed to be most attractive are two proteases, called β- and γ- secretase, that cleave the amyloid precursor protein to produce β-amyloid. There is a lot of speculation around the question of which secretase would be the best drug target. The question became sort of academic in our research in that our screen would have detected inhibitors of either the β- or γ-secretase. Despite a lot of work, we found only inhibitors of the γ-secretase.

MI: And so what can you say about the efforts at Bristol-Myers Squibb, or about efforts industry-wide, and what's in store for the future?

PM: I think that what I can say as far as Bristol-Myers Squibb is concerned is what has been publicly announced—that we have a clinical candidate that is now in clinical development. We know that a number of other companies, including Merck, Lilly, and DuPont, are close. They may or may not have initiated human studies. Another exciting approach is the vaccine approach developed by Elan Pharmaceuticals, where they are immunizing patients now with β-amyloid. This approach showed surprising results in studies with animal models: The pathological progression of the disease was slowed or, in some cases, reversed. They have also reported behavioral data indicating an improvement in cognitive function following immunization of experimental animals.

MI: Besides Alzheimer's disease, what else is hot in the pharmaceutical industry?

PM: The second disease on my list would be depression, where we have large numbers of effective drugs, but they have problems. They have signigicant side effects, they frequently require four to six weeks to take effect, and about forty percent of the patient population is resistant to treatment. And so there's an enormous need. The treatment of depression represents an incredibly large and well-developed market, and so most drug companies are putting considerable effort into identifying novel antidepressants. But just another serotonin uptake inhibitor is not so interesting at this point. On the other hand, finding a novel approach to treatment is very exciting. And there are a couple of targets that are getting a lot of attention. An example comes from studies of antagonists of NK1 (neurokinin 1; substance P), which Merck has reported. These compounds may have a role in the treatment of depression and/or anxiety. Another new approach that may be important involves finding antagonists of the CRF1 receptor (corticotropin releasing factor receptor 1). This was an approach first investigated by Neurocrine Bioisciences, Inc., a relatively small company. Some very interesting results of preliminary studies have been reported.

MI: You've mentioned the market-driven aspect of drug development for Alzheimer's and depression. To what extent can “you” in the pharmaceutical industry pursue targets based strictly on scientific interest, as opposed to the financial “bottom line?”

PM: The pharmaceutical industry is profit based and is interested in targets that have the potential for a big commercial return. The cost of developing a drug today can easily exceed 500 million dollars. You're not going to invest that kind of money unless you can perceive a return that will make the investment worthwhile. But the issue about choosing targets out of market considerations versus scientific interest is really not a major consideration when it comes to finding novel treatments for neurological and psychiatric disease: both the market and the science are wide open. I think it is fair to say that there is no neurological or psychiatric disease that is adequately treated with available medications. Take Alzheimer's disease, Parkinson's disease, and stroke as examples. We've spoken about depression, but consider also schizophrenia—the antipsychotics are all plagued with serious side effects.

MI: But there must certainly be ways that market pressures still differentiate industrial approaches from academic approaches to research.

PM: There are major differences. The real business of the academic scientist is to make discoveries and write papers that make information available for the general community, for the benefit of mankind. But the function of industry is entrepreneurial: the function of industry is to make money. If you have intellectual property and you've made it public, its value to industry is dramatically reduced. Some people have the misconception that if you submit a new patent, for example on a new gene, that your discovery then “belongs” to you. But the truth is that, once you've published your paper, any company can, in a matter of a few weeks, repeat your work and you have no protection until your patent is approved, which can be two years downstream. Any company that's worth anything should be able to clone the published target, screen a library, and prepare a drug to go into the clinic before your patent is solidified. Many people don't understand that, and I don't know how to deal with it, because if you discourage people from publishing until a patent is issued—that would be a terrible restriction on the process of discovery. On the other hand, if you give somebody patent protection just on the basis of publication, then you may actually stifle drug development in the long run. I also have real reservations about the growing popularity of so-called use patents. People are patenting ideas and then waiting for someone to do the work of developing a drug. I think that you should have to play a role in reducing an idea to practice if you want to benefit from the idea.

MI: Yet any measure that seeks to protect the market conjures up the politically charged image of the big, evil pharmaceutical industry…

PM: Well, “big and evil” is what we hear. The pharmaceutical industry has been painted as an incredible villain over the AIDS epidemic, particularly in sub-Saharan Africa. It's absolutely true that the population there is ill and needs treatment and cannot afford the prices that are being charged. But they can't really afford even the basic cost of manufacturing the drugs. For example, the 600 dollars per year that the Indian company Cipla is planning to charge, to make these drugs available to underdeveloped communities, will be too high. That amount is already ten to fifteen times higher than the per capita investment of those regions in health care itself. So even if the drugs are available at cost, you won't solve the problem of poverty. And an obvious question arises in that, if the people in southern Africa who can't afford these drugs are entitled to them, what about the poor in the US—aren't they entitled to them, too?

It is unfortunate that industry has chosen patent protection as the issue on which to make a stand. It might have been better to have given low-cost licenses to manufacture drugs to treat HIV to governments in underdeveloped countries. This might have prevented the pharmaceutical industry from becoming the scapegoat. Innovative drug development needs to remain a priority. We have a system set up that is designed to reward the innovation that we need and value. We must permit a high return on the investment that takes place, recognizing that drug discovery is a high-risk undertaking. On the other hand, we want to make drugs generally available at a price that people who will benefit can afford to pay. But it is crucial to continue asking how a policy change may affect the next generation of innovation. And it then becomes the same question: Who's going to pay the bills? How much are they willing to pay and what is the impact going to be on profitability and the incentive to do drug discovery? The people who probe into questions of drug pricing must also ask the critical question, “What value do you set on innovation?”

MI: Besides the issue of drug pricing, what other pressures are posing potential threats to innovation in drug research and development?

PM: There are several that are really major. One issue is animal research. The pressure is probably greater in Europe than in the US today, but there is enormous pressure against doing animal research, which is truly shortsighted. None of the major advances in treatments could have occurred, or will occur in the future, without the use of animals in research. That is certainly not to say that animals should be used casually or that they should be used without concern for their welfare and their comfort, but you have to do animal research if you want to discover drugs.

But you've got lots of other big pressures that are facing the industry as well. You've got the whole area of prescription benefits for the elderly or the poor, again with the obvious question of who's going to pay the bill and how big of a bill are they willing to pay. And what's the impact of reduced income of the pharmaceutical industry going to be on investment and thus on innovation.

I think another important area relates to the question of how we can increase the translation of intellectual property generated by academic science into new therapies to treat disease. And I don't think this is an area where we do terribly well today.

MI: Do issues of translation and patents tend to divide academia and industry?

PM: I see the issues of translation as an opportunity to bring academic and industrial research together. There needs to be increased efficiency in things like material transfers and communication of willingness to work together on both sides. And academic institutions are in fact beginning to talk about facilitating the translation of academic discoveries, for example, in the form of so-called “incubators,” which refers to space adjacent to or even part of a university where a nascent discovery can be developed into a drug candidate. And that sort of process probably does require independent space, because it is very important to protect the integrity and mission of the academic investigator.

MI: So what kinds of philosophical approaches need to be taken, and who needs to take them?

PM: I'm not sure who needs to make the bigger step—I'm not sure how that will play out. If the universities take the position that they're willing only if industry pays for it, progress may be slow. If, on the other hand, the universities take the position that they're going to benefit from commercial development and they're willing to finance some of those costs, I think the interplay could be quick and efficient.

MI: Are there other pressures, either negative or positive, with regard to innovation in research?

PM: A new element that is very exciting is the genomics revolution and the sequencing of the human genome. We are seeing a dramatic increase in the number of targets, which is causing a basic paradigm change. Whereas we used to be guided by the relatively defined range of limited targets, now we are faced with a more basic question, namely, “Which target shall we choose?” Somewhere along the lines of an order of magnitude increase in the number of available targets is clearly going to put a stress on the system. I don't think that pharmaceutical companies have come to grips yet with the question of how you select among the myriad targets that are available. There is some technology that is being developed to move quickly from the gene to some kind of a high-throughput screen, but this is still pretty early in that game. And the drug discovery organizations do not have the capacity to increase by an order of magnitude the number of targets that they're working on. Companies have to confront the likelihood that they will have to invest in a large number of additional, new, unvalidated targets, most of which will fail to lead to new treatments.

But changes are going to continue to confront not only the industrial scientist, but also university scientists. Academicians are becoming much more focused and interested in translation and commercialization. I personally feel that they are being required to face a steep learning curve. It used to be that somebody at the university would discover something and they'd give a few seminars and then they'd write up the results and publish. But now they must consider patent issues before they discuss some aspects of their findings. So some of the new realities of what it means to be doing biomedical research are affecting both industry and academia, sometimes in similar ways.

MI: Do you think genomics will change pharmaceutical science, as has been predicted, such that every patient will have an individual strategy tailored in terms of his or her gene expression patterns?

PM: Yes, I do. The industry is looking at this issue with a little bit of a jaundiced eye. By abandoning the concept that one drug fits all, the market will clearly undergo fragmentation. However, there is already pretty convincing evidence that you can segment the population according to probabilities of seeing efficacy and side effects. This will benefit not only patients in terms of being able to get drugs with increased efficacy and reduced adverse reactions, but it also has implications for clinical trials. For example, if you can predict patients that will respond, you can dramatically decrease the size of your clinical trial, recognizing that you're then committing yourself to using that same technology to identify the population of patients to be treated.

MI: But how can the pharmaceutical industry undergo such market fragmentation, especially given the blockbuster mentality that has characterized the industry? Can pharmaceutical research be practiced in such a fragmented scheme?

PM: I don't think it's anything that the pharmaceutical industry, or anyone else, has a vote in. It's happening! There is already technology, with at least some classes of drugs, that allows you to pick the agent that is more likely, from a set of similar drugs, to be efficacious. New technology is already generating these sorts of data. It's not stoppable.

• © American Society for Pharmacology and Experimental Theraputics 2001