A dementia patient participates in an Alzheimer’s prevention trial. (Andre Chung/For The Washington Post)

Thomas C. Südhof is a professor of neurosurgery, psychiatry, neurology, and molecular and cellular physiology at Stanford University. He received the 2013 Nobel Prize in medicine.

When the news emerged in November that a major clinical trial for a drug to treat Alzheimer’s disease had failed, the questions naturally focused on the trial itself: Was the drug poorly designed? Was the science bad?


Yet these questions fall short in explaining why the $1 billion trial fizzled. The real problem is that there simply is not enough underlying science to begin with. The trial, led by Eli Lilly, was based on a vague hypothesis without a fundamental understanding of Alzheimer’s disease itself. The trial — like too many recently — was essentially a moonshot in the dark.

The episode underscores that it is time to rethink how we pursue biomedical research and drug development. This shift is especially pressing for diseases of the brain, such as autism and Alzheimer’s disease, which are becoming increasingly important for our society but lack effective treatments. One reason for this lag in treatments is that we have failed to invest enough in obtaining a fundamental understanding of diseases. Instead, we have spent too much on costly clinical trials and studies on human patients that, for ethical reasons, have to be guided by the interests of the patients rather than research goals.

Understanding the mechanisms of a disease is essential for developing an effective treatment. For example, we don’t know why people lose their memory in Alzheimer’s disease or why an autistic child has difficulty communicating. But we won’t be able to solve these mysteries without learning more about the underlying biology of memory and communication. To find out what goes wrong when a person become sick, we need to know how things work.

Inspired by the Human Genome Project, the science of genetics has made revolutionary progress in recent years. We now know of many genes implicated in specific diseases. For example, scientists have identified genes in which genetic variation strongly predisposes to Alzheimer’s disease and autism. However, this information does not tell us much on its own because we don’t know enough about he functions of most of the affected genes.

As we learn more about the brilliance of our bodies, we realize that few genes perform identifiable, single functions. Most genes are multitasking and participate in intricate networks. Their functions have to be understood before genes can be targeted for therapy. Without that knowledge, clinical trials risk more multibillion-dollar failures.

To cite one example: The most successful drugs developed over the past 50 years have been statins, which reduce blood LDL-cholesterol levels and have saved millions of lives. These drugs act by lowering the synthesis of cholesterol in our body, which in turn leads to an elevation in receptors for particles in the liver that remove “bad” cholesterol from the blood.

Statins were not discovered in a clinical trial for cholesterol-lowering compounds, and they were not discovered by serendipity. Biologists discovered them through painstaking research into how the body makes cholesterol. Luck did play a role, in that a fungus was found to produce a natural inhibitor of cholesterol synthesis. But without the critical underlying knowledge about how the body controls cholesterol, this observation would have gone unnoticed.

Can’t we get lucky if we simply try hard enough with enough compounds? Why not try any plausible target, conduct multiple clinical trials and hope one will work? That scattershot approach won’t be successful. Each clinical trial is so expensive that going down this path would devour billions of dollars without much prospect of a return. In my field, brain chemistry, no major advances have been made in the treatment of most brain disorders for the past 50 years — too long.

Some people see a focus on basic research as insistence on wastefully pursuing knowledge for its own sake. That assessment is false. Basic research provides the underpinnings for any understanding of disease, so we need to reassess how we spend our precious funds for development of therapies.

The wisest investment in many cases may be to understand disease biology first and move into clinical trials second, only after we have conceived a rational plan for how to treat a disease. Otherwise, clinical trials risk continuing to be shots in the dark, costly and frustrating not only for scientists but also for patients who badly need new treatments.