The world of drug discovery runs at a blazing pace inside of biotechnology and pharmaceutical companies. Scientists constantly strive to make better versions of therapeutics for old diseases and novel treatments for old and new health problems. Nonetheless, even spending huge amounts of money and working for many years do not guarantee success. In the March 19, 2004, issue of Science—dedicated to drug discovery—Phil Szuromi, Valda Vinson, and Eliot Marshall wrote: “In recent years, the number of new drugs approved annually by the U.S. Food and Drug Administration has at best stayed steady, despite large increases in R&D investment.” This pattern continues. As a result, companies focus on discovering new drugs, and this requires a growing pipeline of people.


A career in drug discovery can take many forms. For example, Bruce Devens, director of research at Bayer Healthcare, describes his career as one that covers academia, as well as small and large pharmaceutical and biotechnology companies. Looking back, he says, “In my personal path, having drugs that I’ve worked on getting to patients and being useful is the greatest source of satisfaction.” Anyone interested in this field needs that desire—turning science into medicine.

Today's Paths to Discovery


In general, drug discovery starts with a particular disease. More specifically, companies often start with a mechanism of action that drives the disease. Then, scientists find a target. Michael Kurilla, director of the office of biodefense research affairs at the National Institute of Allergy and Infectious Diseases, describes a target as “some molecular entity where—if its activity is impaired or enhanced—you can intervene in some productive way in a part of a pathological process.” Once the target is validated, or shown to be fundamental to the disease, researchers look for a compound that hits the target. First, investigators need assays that show when the target is affected. “This usually involves protein biochemistry and purification,” says Brian P. Zambrowicz, executive vice president of research at Lexicon Genetics.

Looking for compounds usually involves high throughput screening of many potential compounds. “Companies have amassed large—and in some cases very high quality—compound libraries,” says Lee Flippin, senior director of medicinal chemistry at FibroGen. Such libraries can consist of millions of compounds. Companies also search for therapeutics in natural compounds. Iwao Ojima, distinguished professor in the Department of Chemistry and director of the Institute of Chemical Biology and Drug Discovery at State University of New York, Stony Brook, says, “Natural products—such as compounds from marine animals, bacteria, or fungi from soils—can also be screened against very defined targets.” He adds that this approach is not as common as screening against synthetic compounds.

During the screening, companies look for compounds, or more likely classes of compounds, with the desired activity, which are called hits. The most promising hits become leads, or compounds that a company pursues as potential drugs. To make a compound work better against a specific disease target, medicinal chemists optimize the drug, tweaking its structure or physical properties to make it more effective and safer.


Market forces also push companies to a new scope in drug discovery. “Searches for new drugs are much more focused compared to years gone by,” says Devens. “We are able to more narrowly define a distinct indication and patient population based on fairly detailed biochemical and molecular knowledge of the disease pathway.” As an example, Bayer tackled hemophilia A, which is caused by a lack of the blood clotting factor VIII. Glenn Pierce, vice president of preclinical development and U.S. medical affairs at Bayer Healthcare, says, “We developed a recombinant DNA construct for this protein, and that revolutionized therapy.” He adds that current hemophilia patients take a replacement of factor VIII every other day, but that could be much improved. Pierce says, “We will eventually create improved therapies that may enable much longer dosing intervals.”

Positions with Potential


The best job opportunities might not be at the start of the process. Zambrowicz of Lexicon Genetics says, “Much of the biotech industry is not focused on working in target discovery and validation mainly because it has been downsized, even at biotech.” He quickly adds, though, that there is always a need for scientists trained in animal pharmacology. “Not many people are trained in this area,” he says, “but it is critical to drug discovery.”

Jack Schmidt, associate director of discovery research at sanofi-aventis, agrees. When asked what backgrounds are in the highest demand, he says, “pharmacology.” Throughout sanofi-aventis, scientists see a growing need for classically trained pharmacologists. Schmidt says, “These people can test compounds or antibodies in well-defined animal models that are representative of a human disease.” Finding such employees, however, never comes easily. “The people are very difficult to find,” says Schmidt. “It has been difficult for years, and it’s not getting any easier.” Pierce of Bayer Healthcare agrees with the need for in vivo experts. “The abilities to work with animal models to develop therapies and with disease models to analyze the therapeutic targets,” he says, “are all in high demand. Quantitative and predictive analyses are skills acquired on the job, but are essential to increase the chances of success in the clinic.” As an example, Pierce points out that many drugs failed because they could not be effectively delivered to their targets. Solving these problems in preclinical models is much more cost effective, but requires a multidisciplinary team that includes formulation experts, polymer scientists, and biologists who can interface with diverse scientists who don’t normally talk to one another.

Many other experts also point out an ongoing need for medicinal chemists. According to Kurilla of the National Institute of Allergy and Infectious Diseases, “Medicinal chemistry is the most rate-limiting step in drug discovery, and chemists tend to be the most rate-limiting resource.” In drug discovery, these chemists must understand biology and chemistry. Ojima of Stony Brook says, “Medicinal chemistry is not necessarily just making molecules. It also requires a wide range of biomedical knowledge and an understanding of chemical biology.” He adds, “These chemists also need great synthetic skills.”

The diversity of the drug discovery process demands an equally diverse collection of employees. Flippin reflects that thinking when he reels off FibroGen’s list of needed specialists: medicinal chemists, bioanalytical experts, cell biologists, enzyme biologists, in vivo pharmacologists, and molecular biologists. Then he says, “Expertise across all of these functions is essential.”

Another topic that can never be ignored is drug safety. This area constantly gains more importance, says Devens of Bayer Healthcare. He says, “Companies are really putting a huge emphasis on making sure that what we do—in as many ways as we can test—is safe for patients.” He adds, “Many scientists could make a career out of getting an ever better understanding of what makes a safe drug.”


Scientists with project management skills can also find opportunities in drug discovery. "It's a logistical nightmare to keep multiple research tracks in line, so that the output from one line is ready as input for another," says Kurilla. "The process of launching a new drug is like launching an aircraft carrier." So companies often seek scientists who can manage complex and multidisciplinary processes.

In the long run, though, it’s not about obtaining today’s most needed skill, because that could change tomorrow. Zambrowicz says, “Lots of times, people get obsessed with trying to train for some specific task or the next best thing coming along. When we look for people, we may have a skill in mind, but we mostly look for bright, well-trained people, because things change all the time.”

Outside Industry


Drug discovery also goes on in academia. For example Ojima founded the Institute of Chemical Biology and Drug Discovery at Stony Brook. Academic efforts in drug discovery also exist at other institutions, such as Baylor University’s Center for Drug Discovery, Harvard University’s Laboratory for Drug Discovery in Neurodegeneration, and many others.

Academic settings provide unique opportunities in drug discovery. “We can target diseases that impact very few people,” says Ojima. “For example, some cancers may not impact many patients and some antibacterial drugs, such as that for drug-resistant tuberculosis, may not be profitable, but it is almost a responsibility of biomedical researchers in academics to do something about it.”

Often, though, the financial challenge of drug discovery leads academic scientists to collaborate with industry. Ojima says, “Once we have drug candidates that are ready to go into animal models, preclinical studies, toxicology, and such things, it really takes companies to do that.”

Drug companies also see the value of such collaborations. Devens of Bayer Healthcare says, “One of the new models of drug discovery involves collaborative efforts between little companies, big companies, and university scientists.” He adds, “This is developing as a path where you can have very focused, specialized university researchers and/or scientists from smaller companies interacting with larger companies to facilitate getting drugs developed.” In this way, even large pharmaceutical companies do not need every possible expertise in house.

An Evolving Process


In many ways the world of drug discovery sits at the edge of enormous changes. At FibroGen, for example, Flippin expects that in silico methods will soon play a major role in drug discovery. He adds, “We are on the beginning of a serious upsurge of structure-based discovery.” Flippin believes that some of the upcoming capabilities in searching for the proper structure of a drug will even replace some of today’s high throughput screening. “With robotic methods of purification and crystallization of proteins,” says Flippin, “it is much easier to gain access to high quality, crystal protein structures.” At Bayer Healthcare, Pierce also sees growing applications of structure-function knowledge of compounds and targets. “We are able to make new protein constructs based on evolving scientific principles,” he says, “and then screen for candidates that can enter clinical development. These designer molecules, which are improvements on naturally occurring proteins, will see an increased role in the armamentarium of therapies to treat diseases poorly treated with natural molecules.”

Issues with data handling will also play a crucial role in the future of drug discovery. “We have a huge amount of genomics data,” says Kurilla of the National Institute of Allergy and Infectious Diseases, “and we need this [information] in functional databases.” He adds, “We are barely keeping up.” As a result, people skilled in bioinformatics will probably remain needed for years to come. The need for computationally savvy scientists already exists. At sanofi-aventis, Schmidt says, “A background in biostatistics is valuable, and we have put an increased emphasis on rigorous statistical analysis of even early pharmacological data.” Part of the increase in statistical testing comes from regulatory mandates, but Pierce also says, “This helps us know if a result is significant and biologically meaningful.”

From his perspective at Stony Brook, Ojima sees a world of chemical biology ahead for drug discovery. This expertise that mixes chemistry and biology is already important, says Ojima, but it will become even more so. He says, “Chemical biology will play the predominant role in the future of drug discovery.”

On the horizon as well is gene therapy, and Pierce believes that the industry will keep moving in that direction. He says, “All of the work on structure and function and the rational design of proteins can translate into improved products offering a better chance of success in gene therapy.” He adds that drug discoverers of the future might show that modifying a gene for therapy can work up to 10 times better than simply replacing the native protein. “That can make all the difference in the world,” he says. Moreover, Pierce mentions that scientists already understand the concepts underlying this technology platform. Now, it’s just a matter of putting it all together.

What lies ahead, though, depends primarily on today’s research. As Zambrowicz at Lexicon Genetics says, “New technologies could have a huge impact on the drug discovery process.” No doubt, drug discoverers of the future will find many new approaches.

Mike May ( mikemay@mindspring.com ) is a publishing consultant for science and technology based in the state of Minnesota, U.S.A.

DOI: 10.1126/science.opms.r0600014

www.sciencecareers.org

Visit and plan to attend upcoming meetings and job fairs that will help further your career.

10.1126/science.opms.r0600014