Medications for major depression and psychotic disorders work well—for some people, that is. A certain medication might seem like a cure for one patient, yet do nothing for the next. Clinicians have had no choice but to use trial and error in selecting medication therapies. And psychiatric drugs can have adverse side effects that make the trial-and-error method not only costly but discouraging for the patient and family members.
However, with the advent of personalized medicine—or precision medicine as the National Institute of Mental Health (NIMH) calls it—a patient may be able to receive treatment that is more efficacious for him or her as an individual, often based on genetics or other personal characteristics. But the science is still emerging. Research won’t be at the point of determining what medication will work for which patient in psychiatric care for at least a decade. But clinicians and administrators in the industry should prepare now for the future.
Think about function instead of diagnosis
Spearheaded by the NIMH Research Domain Criteria project (RDoC), the classification of psychopathology is changing. You’ll still need the DSM for insurance claims, however, according to Bruce Cuthbert, PhD, RDoC director.
Researchers are beginning to think of schizophrenia, depression, bipolar disorder, autism and other mental disorders as a group. It turns out the conditions have a lot in common in terms of genetics, neural circuitry and the functions that interventions need to target.
For example, anhedonia is a symptom of depression, but also is present in certain anxiety disorders, personality disorders and possibly in schizophrenia (in which it is referred to as “flat affect”). And the underpinnings at the molecular level of the anhedonia are what a medication would need to target rather than the diagnosis itself.
Cuthbert compares what’s happening in psychiatric research to what’s currently emerging in cancer treatment. The diagnosis for certain types of cancer is now centered on the specific genetic signature of a particular tumor, and the treatment is developed around that genetic signature, instead of focusing on the organ system involved. For example, breast cancer patients who tests positive for certain hormone receptors will respond to hormone-therapy oncology medications, and those testing negative will not.
Mental disorders are more complicated, though, because they have more genes involved, Cuthbert says.
“You can’t put brain cells in a Petri dish,” says Pamela Sklar, MD, PhD, chief of the Division of Psychiatric Genomics at Mount Sinai Health System. But now that genetic risk factors are being revealed for schizophrenia, autism and bipolar disorder, treatments can be designed to address those genetic issues, Sklar says.
Prepare for genetic testing
This is a relatively new area, and because of its cost, most insurance companies aren’t reimbursing for it. ValueOptions, a managed behavioral health organization, in a pilot project with Genomind, a personalized-medicine organization that offers an assay tool, is looking at the variability of medications for in treating adults with new-onset psychosis.
“The time it takes for a medication to work, the multiple trials and the lack of specificity” are all concerns that ValueOptions hopes to address, says Christopher Dennis, MD, chief medical officer of the commercial division, senior vice president, and regional medical director of ValueOptions in New York City. “So many people have multiple failed trials, some are treatment resistant, some medications may not have worked for them, and genetic testing makes sense clinically,” he says, adding that “when you do the right thing clinically, the dollar usually follows.”
For example, if better selection of medications can prevent hospital readmissions, that would be a win-win for patient and payer.
Consider other predictors
It’s possible that patients don’t need to wait the customary eight weeks to find out if an antidepressant is working. Predictors can help clinicians and patients make educated choices to speed up the trial-and-error process. One predictor is an emotional battery test that involves asking patients to look at an image of a face, which gradually changes from being angry or sad, to neutral. Those with depression tend to take longer to recognize the face as neutral. What researchers found is that after only a week or two on an antidepressant, patients who were responding to the medication started seeing the neutral face more quickly. The test, called P1Vital, was developed in England.
NIMH is looking at other ways to predict medication response. For example, Helen Mayberg, MD, a neuroscientist at Emory University, has suggested that a certain neuroimaging pattern predicts a better response. NIMH is funding her study. The scans to capture the pattern–considered the gold standard–are costly. “But if you have someone with severe chronic nonresponsive depression, it may well be worth the scan to find out what would work better,” says NIHM’s Cuthbert. “If we could get this person functioning again, that would pay off in terms of years of disability.”
In addition, research might lead to an inexpensive proxy for the scan, such as an EEG or a psychological test. “It may be worth it to use the gold standard, but there’s also an interest in developing something cheaper,” he says.