30 Dec 2016
by Hakon Heimer
What course will the US National Institute of Mental Health take with new director Joshua Gordon at the helm? At a press conference at the 2016 Society for Neuroscience (SfN) meeting in San Diego, and in a follow-up interview at the 2016 meeting of the American College of Neuropsychopharmacology in Hollywood, Florida, Gordon offered a few details about his views, but reserved judgment on the overall NIMH research direction.
The appointment earlier this year of Gordon, both an animal researcher and practicing psychiatrist, has led to speculation and editorializing about what this will mean for the allocation of research funds. In particular, questions abound about whether there will be a sea change from the priorities of his predecessor Tom Insel, who now works for Verily, the life sciences division of Alphabet, Google's parent company.
In early interviews and in his inaugural Director's Message blog entry, Josh Gordon indicated that he was going to spend a year learning about the NIMH research portfolio and listening to all the constituents in the community. But he also signaled that a significant departure from the recent past was unlikely (see interviews with US News and World Report and Science).
As Gordon told Nature News, Insel and his predecessor Steven Hyman, now director of the Stanley Center for Psychiatric Research at the Broad Institute, "embedded into the NIMH is the idea that psychiatric disorders are disorders of the brain, and to make progress in treating them we really have to understand the brain." (See also interviews with US News and World Report and Science.)
The move to genetics and neuroscience
During his tenure, Insel was criticized for some of his choices, especially for increasing the funding of genetics while reducing funding for clinical research, and for not supporting enough research on psychosocial treatments and barriers to access to current treatments. The position of some critics was that Insel was sacrificing the improvement of people living with mental illness today in favor of benefiting future generations.
Soon after Insel retired, a group of current and former NIMH advisory council members wrote an editorial in the British Journal of Psychiatry calling for a realignment of the institute's research portfolio (Lewis-Fernández et al., 2016), likening it to a personal investment account. "A disproportionate investment in neuroscience is as imprudent as investing only in growth stocks and neglecting less risky investments that yield immediate albeit potentially more modest benefits," write Roberto Lewis-Fernández of Columbia University and colleagues. [Disclosure: The author of this news story, SRF executive editor Hakon Heimer, is a member of the NIMH Council but not an author of the editorial.]
Insel's position was that researchers had exhausted most of the research leads of past decades, and the best chance to find new clues was with the tools of the genetics revolution combined with new neuroscience methods (see 2007 SRF interview with Insel, as well as his many blogs).
The view from SfN
On November 13, at the Society for Neuroscience meeting, Josh Gordon joined directors or representatives of the other neuroscience-focused NIH institutes for a press conference. The panelists offered their thoughts on major developments in their fields, especially highlights from the meeting.
Gordon said that he favored "a diversity of research," and organized his highlights on a time scale. In the short term, he is intrigued by recent research led by Todd Gould of the University of Maryland suggesting that depression might be treated with metabolites of ketamine that do not have abuse potential (Zanos et al., 2016).
For results in the medium term, Gordon has high hopes for techniques to modulate neural circuits—optogenetics and chemogenetics (aka DREADDs)—to help map circuits underlying mental disorders. In the longer term, he thinks that using cerebral organoids can help us understand the events that create vulnerability to mental illness during early neurodevelopment.
These examples, Gordon told SRF in a later interview, were neuroscience specific because of the audience at the SfN conference.
However, he was pressed on the question of whether the institute was too devoted to neuroscience. During the Q&A session at the press conference, an audience member pointed to two op-ed pieces published soon after Gordon took up his post, which assailed the current program. Lewis-Fernández, who conducts research on how Latinos experience mental illness and the mental health system, had reiterated the criticisms of the BJP article in a Washington Post editorial.
Another former Columbia colleague of Gordon's, John Markowitz, reinforced this in a letter to The New York Times, titled "There's such a thing as too much neuroscience." Markowitz's research focus is clinical research on psychosocial treatments of PTSD and depression.
"My colleagues at Columbia have important things to say," Gordon replied in San Diego. "We need to invest in services for people who have illnesses today. But the issue is less about whether one part of the portfolio is funded over another. It's more important that we fund good science."
Too much neuroscience?
In our interview with Gordon, we delved further into this question. He agrees with Lewis-Fernández and the other authors of the BJP article on the need for better treatment today.
"Implementation and disparities research, among the other things that they discussed, are crucial to ensuring that as many people as possible get access to the treatments that we know work and, for that matter, to future treatments that may work better," he said.
"But I would challenge the assumption that a reallocation of resources is 1) a necessary component of that, and 2) warranted by comparison to research efforts in other areas of the NIH," said Gordon.
By "research efforts in other areas," Gordon clarified that he means the extent to which other NIH institutes fund implementation or disparities research in their disease fields, something he said he will investigate.
Gordon views this process of assessing NIMH's research in terms of promise and opportunities. He clearly thinks that studying the brain holds the greatest promise for new and better treatments for mental illnesses, and acknowledges a relative increase in mental illness research funds going to neuroscience.
"The question is whether that's out of proportion to the promise that the different areas of research afford in terms of making transformative treatments and improving mental healthcare," said Gordon.
Regarding opportunity, Gordon said, "What I intend to do is actively look for opportunities in the implementation sphere, in the translational sphere, in the clinical sphere, and in the basic neuroscience sphere that will fill needed gaps."
In the second blogpost of his tenure, Gordon identifies research into suicide prevention as one of those opportunities.
Addressing comments such as those by Markowitz that the NIMH should revert to spending more money on clinical trials, Gordon said that clinical trials, when done right, are enormously expensive, and a single trial could constitute a large part the NIMH budget.
Nonetheless, Gordon said that the NIMH is not a laggard relative to other institutes at the NIH: The institute falls close to the average for clinical trials as a percentage of budgets across the NIH.
RDoC and big data
Gordon seems unlikely to abolish or radically alter the Research Domain Criteria (see SRF webinar on RDoC for hallucinations), another project of the Insel era that has drawn criticism from different quarters of the research world. According to Gordon, about 50 percent of current NIMH grants are for projects working within the RDoC framework.
"I kind of think RDoC is a good idea," Gordon told SRF. He endorses the idea of breaking down the complex phenotypes of behavior into components in order to inform research on mental illnesses.
"The process by which it was initiated was similar to DSM: Put a bunch of experts in a room, and that process has faults," Gordon acknowledged.
But RDoC, he said, should be seen as a living text, which changes over time to reflect new knowledge. "It's too early to assess the success or failure of RDoC," Gordon added.
One thing that RDoC may well need is a big-data approach, Gordon said. Indeed, he has mentioned in earlier interviews that psychiatry could benefit from computational approaches (see especially Nature News interview).
Gordon's hope is that the field will be able to "use computational analyses to guide diagnosis, to help clarify and modernize RDoC, and to analyze the utility of biomarkers."
One of the challenges to adopting computational approaches is expertise, and there is some good news on that front. The BRAIN Initiative will get a small increase in its budget in 2017, and some of that funding will go beyond the BRAIN project's focus on tool creation to benefit tool utilizers as well, Gordon said.
He added that researchers should keep an eye out for a program of supplemental grant funding, to be introduced in February of 2017, that will allow researchers to hire data scientists for both new and existing projects.
What about research investment in other areas, such as genetics? Gordon thinks that the focus on genetics over recent years has proven to be a valuable investment. "We now have potentially hundreds of genes that impact different psychiatric disorders," he said.
The question he will seek to answer over the next year, Gordon said, is how the NIMH should divide its resources between finding more genes through more or larger genetics studies versus studying the biological roles of the genes that have already been fingered. An advisory group made up of current and former members of the NIMH Council will look at this issue and provide Gordon with recommendations.
Given that our interview happened at the American College of Neuropsychopharmacology meeting, we asked Gordon about the future of psychopharmacology, a field that was such a significant driver of research into mental illness for decades.
"I think the future of psychopharmacology is neuroscience. I think neuroscience discoveries will result in the reinvigoration of psychopharmacology," said Gordon.
But he cautioned that this will not happen quickly.
"The trick will be maintaining the expertise in clinical psychopharm when we don't have a lot of easy targets to go after now. And I don't have easy answers for that," Gordon said.