News FocusMysteries of the Brain

Why Is Mental Illness So Hard to Treat?

Science  05 Oct 2012:
Vol. 338, Issue 6103, pp. 32-33
DOI: 10.1126/science.338.6103.32

The human brain is complex and difficult to study, which has impeded development of drug treatments for mental illnesses. But new tools and new ways of thinking could help the field gain new traction.

“The last quarter century has seen many forward strides in the management of patients with mental disease. During this period modes of therapy have been available that are far superior to the old-time method of whirling patients on wheels or ducking them into cold water.” So begins a 1954 article in the Journal of the American Medical Association on the calming effects of reserpine, a compound derived from an Indian plant, on psychotic patients in a California mental hospital. Its effects were so remarkable, the authors wrote, that if their findings held up, “reserpine will be the most important therapeutic development in the history of psychiatry.”

Serious side effects ultimately prevented reserpine from being the game changer its early champions had envisioned, but other drugs discovered around the same time—including lithium for bipolar disorder, monoamine oxidase inhibitors for depression, and chlorpromazine for schizophrenia—gave clinicians their first real weapons against mental illness.

Flash-forward to the 21st century. Current psychiatric drugs are not much more effective than those initial medicines. Pharmaceutical companies have few promising candidates in the pipeline and show signs of giving up (Science, 30 July 2010, p. 502). Meanwhile, mental illness remains a major cause of disability throughout the world. Why has it been so hard to develop new treatments?

Don't despair.

Progress on new treatments for disorders of the mind has been frustratingly slow, but some researchers see new hope on the horizon.


“We just don't know enough,” says Thomas Insel, director of the U.S. National Institute of Mental Health (NIMH) in Bethesda, Maryland. “Research and development in this area has been almost entirely dependent on the serendipitous discoveries of medications. From the get-go, none of it was ever based on an understanding of the pathophysiology of any of the illnesses involved.”

It's little wonder that uncovering the roots of psychiatric illness has not been easy. Not only is the brain the most complex organ in the body, but it's also harder to study. Doctors and researchers can't biopsy a patient's brain as they would a diseased kidney or swollen prostate. Genetic studies of psychiatric disorders may yet uncover solid leads for drug developers, but so far they've mostly uncovered bewildering complexity.

There's also the question of how well disorders of the human mind can be studied in other species. Many of the animal models of mental illnesses in use today were developed decades ago to screen for compounds with effects similar to those of the early antidepressant and antipsychotic drugs, says Steven Hyman, who directs the Stanley Center for Psychiatric Research at the Broad Institute in Cambridge, Massachusetts, and preceded Insel as NIMH director. In the “forced swim test,” for example, researchers put a rat or mouse in a tub of water and clock how long it takes for the animal to stop struggling and just float, as if it's given up. “This is called behavioral despair,” Hyman says. Of course, nobody knows if the rat experiences anything like what a person in the grips of depression experiences, but rats given imipramine, one of the early antidepressants, struggle longer. The forced swim test has in fact identified other drugs that turn out to have antidepressant effects in humans, Hyman says, but by relying on a test designed to find imipramine-like effects, researchers may have missed drugs that work by means of other, potentially more effective mechanisms.

Hyman, Insel, and others argue that too much effort has been spent searching in the narrow beam of light cast by the early psychiatric drugs, looking for similar compounds or tweaking their chemistry to eke out improvements in efficacy, reduce their side effects, and—not insignificantly—preserve the revenue stream generated by patent-protected drugs. The field of psychiatric medicine, it seems, got lucky early on, and then it got in a rut.

However, several new tools and new ways of thinking could help the field gain new traction. One example, Insel says, comes from drugs that target receptors for the neurotransmitter glutamate, which recent evidence suggests can reduce hopelessness and suicidal ideation in people with depression far faster than current drugs do. “That story, while it's still developing, is extraordinary because it tells us we need to rethink our expectations,” Insel says. “It may be possible to treat this in hours instead of weeks.”

Another encouraging inroad into depression comes from deep brain stimulation (DBS), in which surgeons implant electrodes in brain regions thought to be involved in regulating emotion and cognition. The approach is still experimental, and only severely depressed patients who've failed to respond to less invasive treatments are eligible, but DBS seems to help about three-quarters of them, says Helen Mayberg, a neurologist at Emory University in Atlanta and a pioneer of this therapy. The success of DBS, Mayberg and others suggest, undermines the decades-old concept of mental illness as primarily a chemical imbalance—too much or too little serotonin floating around the brain, for example—and points instead to faulty neural circuits as the core problem. In their quest for the next generation of treatments, researchers should focus less on single molecules and individual brain regions, Mayberg says, and think of the brain as “a dynamic system that has to be properly choreographed.”

One powerful new tool for examining neural circuits at the cellular level is optogenetics, which combines laser optics and genetic engineering to stimulate or inhibit specific classes of neurons in rodents and monkeys. Researchers have employed optogenetic methods in mice to investigate the mechanisms of DBS therapy for Parkinson's disease (Science, 20 March 2009, p. 1554) and to study the neural circuits involved in addiction, anxiety, and other conditions.

On a larger scale, human brain imaging research is illustrating that psychiatric conditions are multifaceted, with different symptoms that can be traced to different networks of brain regions, says Cameron Carter, a cognitive neuroscientist at the University of California (UC), Davis. The delusions and hallucinations of schizophrenia, for example, may involve malfunctions in one network, while disordered thinking and other cognitive problems involve another. This brain-based view is at odds with the traditional approach of diagnosing disorders according to the behavioral problems and inner anguish they cause—the approach taken by psychiatry's go-to diagnostic guide, the Diagnostic and Statistical Manual of Mental Disorders (DSM), published by the American Psychiatric Association. “There's no doubt that these categories [in the DSM] don't map very well onto nature,” Carter says. An effort to create a new diagnostic scheme rooted in biology, spearheaded by NIMH, is already under way (Science, 19 March 2010, p. 1437).

And studies of the human genome may yet lead to innovative drugs for mental illness. Researchers have identified hundreds of genetic variants that increase the risk of autism, for example. Making sense of this deluge of new data is a challenge, but there are some hints of convergence: Many of the autism risk genes appear to be involved in common biological functions, such as synaptic signaling and brain development. Researchers now have novel tools at their disposal to follow up on these leads, including the ability to create neurons from reprogrammed stem cells from patients (Science, 26 November 2010, p. 1172). Gene expression profiling—investigating the activity of thousands of genes—in these human neurons could help researchers identify the biological pathways disrupted by autism risk genes and screen drugs to correct them, says Daniel Geschwind, a neurogeneticist at UC Los Angeles. “I'm extremely optimistic that [by] using a combination of these methods we're going to be developing new classes of drugs,” he says. “I think we're on the threshold of something really exciting.”


Navigate This Article