December 17, 2007
Event video

Event transcript

Summary:
Genetic research may hold great promise for better treating mental illness, but psychiatrists should hold off on replacing their couches with gene-sequencing machines for now, according to panel members at the Center’s Genetics Perspectives on Policy Seminar (GenePOPS) on December 17. Topics covered at the seminar, titled “DNA and Depression: Tests, Trust, and Treatment,” ranged from specific genes and medications associated with thoughts of suicide, to the gap between rich and poor in mental health services.

Center Director Kathy Hudson welcomed audience members to the National Press Club and thanked the event’s cosponsors, the Bazelon Center for Mental Health Law and the Center for Disease Control’s (CDC’s) Office of Genomics and Disease Prevention. She then set the stage for the panel discussion. “Like with most common, complex diseases, there aren’t individual genes that can tell you that you will or will not definitely get a particular mental illness,” she said. “Instead, it’s a number of genes interacting with one another and interacting with complex environmental factors and lifestyle factors.”

Scientists are now identifying many of these factors, Hudson said, creating the potential for better, more targeted treatment. “Today, we’re going to focus on the pipeline from basic science discovery through the translation to public health benefit, and where there may be cracks in that pipeline that we can solve with public policy changes,” she said. “In addition, we’re going to talk about the social context for these new developments and the need for safe and beneficial uses of genetic tests in mental health.”

Hudson then introduced the panelists: Francis McMahon, chief of the genetic basis of mood and anxiety disorders unit at the National Institute of Mental Health at the National Institutes of Health; Kim Bechthold, founder and CEO of genetic testing company NeuroMark; Joan Scott, deputy director of the Genetics and Public Policy Center; and Robert Bernstein, executive director of the Bazelon Center for Mental Health Law.

The Scientist

McMahon explained that he would focus on pharmacogenomics, which he described as the use of genetic tests “to tailor treatment to an individual’s genetic makeup in order to improve both outcomes and side effects that might come from medications.” Only about half of patients respond to anti-depressant medications, he said, and the likelihood of response hinges on genetics as well as factors such as such as marital status, race and ethnicity, type of depression, and socioeconomic status. “One of the big questions has been to what degree genetic variation contributes to that individual variation and whether, if that could be identified, we could harness it to design better treatments and identify who is likely to respond to particular antidepressants,” he said.

McMahon’s own research has centered on the STAR*D study, he said, which enrolled 4,000 people diagnosed with clinical depression in order to study responses to antidepressants. His team correlated markers in three different genes with response to citalopram treatment. “This is a result that, from a statistical standpoint, is of interest, but it’s too small to really affect clinical decision-making and treatment,” he said.

“To get more into the clinical realm, we decided to look at a particular adverse event that’s been linked to antidepressant treatment in recent years, known as treatment-emergent suicidal ideation,” McMahon said. About 2 percent of patients who take selective serotonin reuptake inhibitors (SSRIs) for depression begin having thoughts of suicide while on the medication, he said, and though these thought probably do not lead to suicide, they are a concern. Using questionnaires and genetic analysis, his team found four genetic markers associated with treatment-emergent suicidal ideation – a finding that will need to be replicated, he said.

“What we really need in the realm of pharmacogenetics,” McMahon said, “are markers that will predict which patients will respond to a given treatment versus alternative treatments that may be available. We need to be able to identify patients who will suffer intolerable or dangerous side effects, or those who would do best with specialized care… We also need ways to provide this information at the point of service, when patients come in for treatment, so that patients and the doctors can consider the information in real time, at the time that treatment decisions are made.”

He ended by naming several “open questions” that will need to be addressed if personalized medicine is to become a reality, including “How can genetic information be combined with non-genetic information?” to inform mental health care.

The CEO

Bechthold set the scene with some numbers: The World Health Organization estimates that incidents of depression will eclipse heart disease by 2020, and 20 million people in the United States have the disease now. Fortunately, “The promise of personalized medicine in our field, in neuropsychiatry, is really quite extraordinary,” she said. This promise includes not only better ways of treating mental illness, she said, but also of preventing it.

NeuroMark works to educate clinicians, patients, and their families about its tests, Bechthold explained. The four areas her company is working in, she said, are developing a test to predict treatment-emergent suicidal ideation arising from citalopram use, developing a test to differentiate major depressive disorder from bipolar disorder, personalizing drug treatment, and research on interactions between genes and environmental factors.

“The black-box warning apparently created a rise in suicide rates in the year 2004,” Bechthold said, referring to the Food and Drug Administration’s (FDA’s) requirement that all SSRIs carry a label stating that they may cause suicidal thoughts in children. Suicide rates among teenagers rose in 2004, she said, and “the reason this is so phenomenal is that in the past 15 years, suicide rates in the United States have declined 28 percent.” Mass killings have also been committed by people taking SSRIs, she said.

“So we believe that the work that Dr. McMahon and his colleagues have done is the basis for a history-making test,” Bechthold said, referring to the Mark-C test that NeuroMark is developing to assess the chances that a patient will develop suicidal ideation while taking citalopram. The company soon will launch a study to confirm McMahon’s findings, she said, and expects the study to finish in 2009, at which time NeuroMark will seek FDA approval for the Mark-C test kit. She showed how the Mark-C results could be fed into a risk prediction chart to make them easier for physicians to understand and use.

Bechthold concluded that preventative medicine would require genetic information nondiscrimination legislation and better reimbursement policies for genetic tests.

The Panelist

Scott began by explaining that she would be “playing the role of Jeff Botkin,” a University of Utah researcher who was unable to attend due to weather. Both Botkin and Scott are members of the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) working group; Scott’s talk would focus on EGAPP’s just-released recommendations on the use of CYP450 testing in patients beginning SSRI treatment.

SSRIs “are the first drugs of choice because of their tolerability and safety as compared to other categories of anti-depressants,” Scott said. “The problem, as you’ve just heard, is it can be very challenging for clinicians, and individuals with depression, to find the right balance of the right drug, and the right dose, and it can take a while to go through that entire process.” Some patients don’t respond to the drugs at all, she said, or experience bad reactions. “So it would be nice to have a method, or a test to do on individuals who have depression, and who are starting treatment, to determine which is the best drug, what is the right dose, to narrow that therapeutic window and start getting responses fairly quickly and avoid adverse reactions.”

CYP450 refers to a category of genes that, among other things, are involved in the metabolism of most drugs, Scott explained. “There are two specifically that are involved in the metabolism of the SSRIs, and they come in different variations.” These variations cause people to metabolize drugs at different rates, she said, affecting how much of the substance actually circulates in the blood, and thus, in theory, individual response to treatment.

Tests of the SSRI-related CYP450 genes already are commercially available. The question EGAPP took up, Scott said, was “what do the data show to suggest that doing CYP450 testing ahead of time will affect clinical outcomes?” EGAPP is sponsored by the CDC, she explained, but is an independent non-federal working group set up to develop and test methods to rigorously evaluate new genetic tests. Specifically, the group looked at whether the CYP450 tests accurately predicted circulating levels of SSRIs, whether they predicted how well patients would respond to SSRI treatment, and whether treatment outcomes improved as a result of testing.

The group found “no consistent association” between patients’ genotype and their responses to treatment, Scott said. However, she noted, there were few studies on the subject, and those that have been done involved small patient populations. “And then there were no studies at all that actually tried to address the overarching question, which is, does testing really affect patient outcomes?” she said. The group found insufficient evidence to recommend for or against routine use of CYP450 testing before SSRI treatment, but taking into account “clinical contextual issues” such as cost and the possibility of treatment being delayed or misapplied, it discouraged testing pending additional clinical trials.

“I think what we have done is outlined what is the evidence that we have to date, and where are the gaps and the information that needs to be filled in,” Scott said. “It very well may be that there are subsets of patients that would benefit from testing.”

The Advocate

Bernstein titled his talk “DNA and Depression: Test, Trust, Treatment… and a Touch of Trepidation.” He explained, “I didn’t say ‘terror,’ but I think there is a good deal of unease in the mental health community about what all of this means.” There are, he said, “two very different worlds of mental illness”: Those with means have a much different experience than those who must use public health systems.

“The idea that genetic testing might validate more precise tailoring is a very positive thing,” Bernstein said. He questioned, though, what genetic medicine will mean in a public health system where “numerous studies have documented the discrepancy between care that is known to be effective and care that is actually delivered. And often the gap is: How cheaply can we get away with serving people who are regarded as a social burden?”

Bernstein related statistics on the high readmission rates for patients discharged from mental health programs in state hospitals. With America’s mental health system “in shambles,” he said, “we’re talking about some very good science, some very promising science that is going to be delivered to a system that really is ill-equipped to even deal with the most basic of care for people with mental health needs.”

Science focuses on symptoms and response to medications, Bernstein said, and asked, “will [genetic testing] risk further blurring the attention to the individual who has the symptoms?” He concluded, “I don’t know where this is all going to go. I do know that our history doesn’t bode well for where this might end up.”

Discussion

Hudson asked Scott whether the current turnaround time for genetic tests would allow physicians to make use of them, or whether tests need to be designed to be performed in doctors’ offices. Scott said she did not know how long CYP450 testing takes, but that ideally tests would be performed at the point of care. Bechthold said the Mark-C test would have a turnaround time of 48 hours, and that “in the future, that test can be done in the physician office.” McMahon predicted that technology would soon solve the turnaround problem. “Many people have predicted that the complete genetic sequence will be part of everyone’s medical record within the next five years or so,” he noted.

Hudson asked about the relative importance of socioeconomic status and genetics in predicting the outcome of mental health treatment. McMahon said that in the STAR*D study, socioeconomic status “was a huge predictor, much better than any of the markers we found so far.” Bernstein said that “people of lower SES tend to get into treatment later on than people who have ready access through commercial insurance.” Genetic screening might allow earlier intervention by revealing who is most at risk for depression, he said; on the other hand, it could “indelibly” label people. McMahon said that even when patients have equal access to treatment, socioeconomic status remains a significant predictor of outcome. “There are other ways that struggling to make ends meet can affect depression adversely, even when you’re getting the same care as somebody who is of higher SES,” he said.

Hudson asked about the connection between suicidal thoughts and suicide. “There really is, as far as I am aware, no good data suggesting that antidepressants actually increase suicidal deaths,” McMahon said. “In fact, the overwhelming data is that death by suicide is prevented by antidepressants, and that by far the greatest risk factor for completed suicide is depression.” Bechthold said, “The important thing is, we have three times as many suicides as homicides – three to two – in terms of suicides versus homicides in the United States. No matter how you cut it, that’s too much; it’s a treatable illness.” Bernstein said that both suicide and homicide are relatively rare, and that “a percentage increase may look large because the denominator is pretty small to begin with.” Referring to the rise in suicide rates in 2004, he said, “Whether this has anything at all to do with prescribing practices, I don’t know, because we’re living in a time when there is a war going on; there are people coming home with PTSD, and they affect family functioning.”

Audience member Andrea Kalfglou of the University of Maryland Baltimore County asked whether, if there were a genetic test for predisposition to post-traumatic stress disorder, it should be used to screen enlistees in the military. Bechthold responded, “do we want to eliminate a career Army officer from serving in Iraq because he has a particular genetic predisposition, or do we want to take a smarter approach, and when he returns home and he’s suffering from post-traumatic stress disorder, do we look at his genetics and use those to help him receive treatment and help his family understand him…” Bernstein said that while PTSD is a serious problem, “it may mean that these are people who have their eyes open to what war is all about and how awful it is. And in screening those people out, are we perhaps going this science fiction direction of looking for some ruthless people who will just be machines for us?”

Khaled Bouri of George Washington University asked Scott, “Does the FDA have an EGAPP-like group that can rigorously oversee the hundreds of genetic tests that are in the pipeline? And if not, is EGAPP coordinating any results with the FDA?” Scott responded that the FDA looks at the analytic and clinical validity of test kits, “but with exception of one small category of laboratory-developed tests, they have not taken on doing reviews of every laboratory-developed test that’s out there.” Bechthold added that the agency has a legal right to review laboratory-developed tests.

Hudson asked Bechthold what led to NeuroMark’s decision to make the Mark-C a test kit, which will require FDA pre-approval, rather than a laboratory-developed test, which would not. Bechthold replied, “The decision is really not what we need to do for the FDA; the decision is what do we need to do for the clinician, what do we need to do for the scientific community, for them to embrace a test such as this?” FDA approval will take about a year, but the company considers it worth the time, she said.

“I think it’s one of the big challenges we have in personalized medicine: How do we make that translation process much quicker so it’s in the hands and in the minds of physicians when they’re seeing their patients?” Hudson said. Bechthold commented that “in years past, it was totally up to the physician for adoption. And we may find now that parents and young adults are very, very interested in what they find online, and so we have a major program that we’re going to be instituting to reach those people who have a concern, who are going to pick up their mouse and see what’s going on in the field of depression.”

The Center’s Gail Javitt asked Bechthold whether the Mark-C test would be offered directly to consumers, and if not, why not. Bechthold said that the test will be available only through physicians, and that the company had shifted its focus to developing clinical tests for various reasons, such as privacy concerns.

Joe McInerney of the National Coalition for Health Professional Education in Genetics asked, “What would you like the gatekeepers that are prescribing physicians in particular to know about these kinds of tests so they can use them effectively and explain them to their patients before they order the tests? And second, how would you like to see that educational content delivered to the providers? Who should do it?” McMahon said that because consumers are now able to order genetic tests themselves online, “we don’t have a monopoly on the information anymore.” Genetics education should be improved for doctors of all types, he said, because “physicians need to be able to intelligently consume the literature, keep up with it so they can interpret information that increasingly shows up in the popular press at about the same time it shows in the medical journals.” Bernstein said he sees promise in computer programs that have patients answer questions about their symptoms and learn about treatment options before they see their physicians, so that they can have a more informed conversation.

Ed Abrahams of the Personalized Medicine Coalition asked, “how can we generate more evidence around these diagnostic tests, particularly how drugs are metabolized?” He also asked Scott, “as a genetic counselor, would you advise a patient who presented for depression to learn how he or she metabolizes drugs?” Scott responded that funding would be needed for studies of tests’ effectiveness. She said her response to the second question would depend heavily on individual circumstances, but that “at this point in time, I don’t see that the evidence suggests that having that information will really help determine how to treat that individual.” – Shawna Williams

Video Link:
http://www.dnapolicy.org/video/genepops/121707/index.htm

<< Back to Past Events