Impossible Medicine

This Glorious Quest

William Chin, Harvard Medical School’s executive dean for research, offers a vision for partnerships that move discoveries from bench to bedside

Q: You’ve likened the researcher’s path of discovery to Don Quixote’s quest. Can you elaborate?

A: Don Quixote relied on imagination, determination, and luck, and researchers need all three to prevail—along with an unwavering sense of optimism in the face of ongoing defeat. That’s the nature of scientific inquiry. Nowhere is that sense of optimism more needed than in drug discovery, which seeks to identify and develop potential medicines that provide a favorable and delicate balance of effectiveness against disease while ensuring safety.

Q: How has drug discovery evolved?

A: Thirty or more years ago, the discovery of new medicines was based on the observation that a chemical or an herbal extract had a biological effect. Scientists would then tweak the chemical or molecule to make it safe and effective in humans. That’s the classic pharmacological approach.

Another, newer approach is based on knowledge we’ve gained during the genomic revolution. We can now begin with a gene that we believe relates to a particular disease, then either increase or suppress the activity of that gene in the hope of discovering its precise role in that particular condition. We take multiple shots in hopes of scoring a goal.

Q: And then you screen a million compounds?

A: Yes, but that adds a new level of complexity. Even if we find a molecule that hits a target, we still don’t know exactly how that target links to a disease. A major reason the pharmaceutical industry has failed to be as productive as we’d like is that we have yet to pinpoint the root causes of most diseases. Nor do we understand the natural variation in disease that occurs across patients. Not everyone with type 2 diabetes, for example, has exactly the same disease. Not all breast cancers or lung cancers stem from the same cause, even if they behave similarly in a biological sense. If you don’t know the cause of a disease and you try to cure that disease with a guess, chances are you’ll fail.

So if you think about the notion of “impossible medicine”—treatment goals that might seem unreachable—yes, medicine becomes far less possible when you don’t know what causes disease. To make good medicine possible, we need to focus on the basic biology of disease.

Q: Can you give some examples of biomedical discoveries that proved attainable against great odds?

A: Vaccines have been tremendous successes. When I was a kid, everyone was deathly afraid of polio. Polio has since been largely cured, as have many childhood diseases. Antibiotics have proved to be triumphs as well—against pneumonia, urinary tract infections, even tuberculosis. And multiple drug therapies can now control HIV disease—provided people with the virus can access those medicines.

Going back a bit further, the discovery of insulin was remarkable, and that finding isn’t yet a century old. Before insulin, doctors essentially lowered patients’ blood sugar through starvation!

QDid the discovery of insulin occur in academia or in industry?

A: The answer to that is fascinating. In the early 1920s, two university researchers, Frederick Banting and Charles Best, discovered how to make an extract from the pancreases of cows and pigs that could treat diabetes in humans. This finding, incidentally, earned Banting and a collaborator a Nobel Prize. But when these researchers tried to make batches of insulin, they lacked a method for making a pharmacologically uniform drug. Standardization is critical in drug production. So the head of research at Eli Lilly—where I worked for eleven years—went to Toronto and said, “Look, we can help you make uniform batches.” Lilly became the first company to produce a standardized insulin.

More recently, statins—medicines aimed at lowering blood cholesterol—have helped many people, as have antihypertensives. And in the realm of cancer, we’ve had Gleevec, Tarceva, Iressa, and Herceptin. These are just a handful of the contributions pharmaceutical companies have made during the past few decades, most in collaboration with academia.

QHow many of these drugs were developed under the more classic, randomized approach?

A: Most of them, although the development of statins came from a combination of new and old approaches. In truth, the original statin was discovered almost by accident: the original researcher was actually looking for an antifungal, and then others realized the statin’s possible role in lowering bad cholesterol. In another example of serendipity, the breast cancer drug tamoxifen also was found to be a treatment for osteoporosis. And Viagra, remember, was supposed to be a heart medication.

QWhat are the biggest challenges in drug discovery?

A: We still don’t have many medications to offer for schizophrenia, obesity, and a number of cancers, to name a few unmet medical needs. We know how to control type 1 diabetes mellitus, but we can’t stop its development. The list goes on. Millions of people worldwide suffer from major illnesses for which we have no effective therapies.

QCan you give examples of how our understanding of a disease’s cause led to a drug?

A: One great example is Gaucher’s disease, a condition that affects the liver, bone marrow, and other tissues. People with that disease lack the enzyme glucocerebrosidase, which regulates the production of blood fats, or lipids, that are important for these organs. But Genzyme has been able to reproduce that enzyme and thus save lives.

Another example is sickle cell anemia, in which a gene defect causes hemoglobin to malfunction. Ordinarily, such a mutation would have been selected out through the process of evolution, except that it proved to confer a genetic advantage to people living in parts of West Africa; it reduced their chances of getting malaria. Our understanding of the cause of sickle cell anemia has led to medications that relieve its symptoms, but we still haven’t cured it.

QThat’s true of Huntington’s disease, amyotrophic lateral sclerosis, and cystic fibrosis—we know the causes but don’t have cures.

A: Yes. I should qualify what I said earlier: Knowing the cause of a disease is just the first step. In the case of cystic fibrosis, we know all the mutations involved, yet we still don’t have a cure. Imagine just how much harder it is, then, if you don’t even know the cause.

The problem is that the drug pipeline functions like a lottery—one in which you invest a half-billion dollars per ticket. And you have to keep on investing in that ticket until it either wins or loses.


QWe often hear that the pharma pipeline is drying up.

A: The problem is that the drug pipeline functions like a lottery—one in which you invest a half-billion dollars per ticket. And you have to keep on investing in that ticket until it either wins or loses. Industry needs to recoup those investments. You can hope that every medicine will be like Lipitor, an effective cholesterol-lowering drug that can be used safely and indefinitely. But if you keep aiming for that jackpot, what will you miss? Rare diseases, ones that affect only a few thousand people. It’s amazing, too, how few companies invest in preventive medicine, like vaccines. That’s one of my pet peeves.

Another problem lies in the culture of Big Pharma. Pharmaceutical companies used to think they could do everything themselves. They’re slowly realizing that they need collaborative partners in academia, in government, and in nongovernmental agencies.

QWhat role can an institution like HMS play?

A: First of all, our faculty members—some of the best scientists in the world—are studying biological processes in all kinds of cells, tissues, and model organisms to understand the fundamentals of how the body works. If we at HMS could improve the way we collaborate with our colleagues in our affiliated hospitals, we could make a big dent in this question of what causes disease, because patients hold the key to that question. A vital role for the School, then, is to solve the problems medicine faces by combining this intellectual capital, this passion for learning, and this power to implement experiments.

HMS can also provide leadership both in creating new ways of learning about the causes of diseases—by discovering new biomarkers and relevant animal models, for example—and in developing new and growing disciplines, such as chemical biology and systems pharmacology.

QWhat is systems pharmacology?

A: Think of a spider web. If you touch even a fragment of the web, the spider knows, because it immediately senses the vibrations along the web’s fine filaments. When a bug lands on a web, the spider doesn’t go on a long trek to find its prey. It goes directly to the spot. Why? Because it understands its network.

Systems pharmacology strives to understand how medicines work—not just how they affect one biological pathway or one molecular or genetic target, but how they affect multiple tissues, organs, and cellular pathways in an entire physiological system. HMS has just entered this arena. We have experts in systems biology, chemical biology, human genetics, biochemistry, classical pharmacology, animal models, and informatics, but we’ve never brought all that expertise together.

Systems pharmacology is, in fact, one of our new initiatives. When Jeff Flier became dean of HMS, he made it a central part of his mission to have the School participate more fully in the development of new therapies. So we’ll soon launch an HMS program in translational science and therapeutics, which will aim to harness all the strengths of HMS both to gain a deeper understanding of what causes disease and to create molecules that can combat those causes.

QHow do we bring people from the basic and clinical sciences together?

A: First, through education. We need to provide training to scientists who are interested in embarking on translational research. We plan to sponsor fellowships, courses, and seminars. We’ll also create a culture that better appreciates the value of translational work.

The School has and always will place tremendous value on basic science. In many ways, that’s central to its identity. But a critical goal is to encourage bench scientists to shift their awareness, to consider how they might apply their research to the solution of clinical problems.

What’s more, we’d like to create a Harvard therapeutics commons, a meeting place—both virtual and real-life—that we could use to unite people with diverse experience and scientific backgrounds who like exploring issues surrounding translational research, not just at HMS and its affiliates, but at other institutions as well. A scientific ecosystem that is populated by experts from academia, industry, government, nongovernmental agencies, and foundations could help patients, who must be the focus of these collaborations.

Now, any collaboration between academia and industry brings up the specter of conflicts of interest. Clearly, researchers must strive always do the right thing. But that doesn’t prohibit them from working with industry or government. We live in a context of conflict—whether of interest, time, or resources. We must be honest, open, and transparent in all situations, at all times.

QCould you elaborate on the role of the patient in translational medicine?

A: Several months back, Paul Farmer [’90], chair of our Department of Global Health and Social Medicine, reminded me that we can have the perfect medicine—one that’s safe and effective—but if our patients can’t take that medicine because they can’t afford it or don’t have a physician or don’t have access to medications, then we haven’t succeeded. Translational medicine means more than just developing therapies. It means getting those therapies to patients.

All HMS departments—whether basic science, clinical science, health care policy, or global health—are critical to achieving the objectives of translational medicine. That’s why HMS is such a great place.

David Cameron is director of external relations at Harvard Medical School.

Images: Mark Ostow