TCPR: Dr. Cuthbert, how did the Research Domain Criteria (RDoC) initiative begin?
Dr. Cuthbert: Around 2008 there was a feeling that, compared to other areas of medicine, we were not moving ahead in the prevention and treatment of psychiatric disorders. If you look at areas like leukemia, HIV, stroke, and heart disease, we have seen vast decreases in morbidity and mortality over the last 20 to 30 years, but the same is not true for mental disorders. Our vision statement is that we envision a world in which mental illnesses are prevented and cured. So we are trying to work toward transforming understanding and treatment of mental illness through research, and paving the way for prevention, recovery, and cure—not simply managing disorders.
TCPR: There have been concerns that the DSM system is no longer useful for psychiatric research. Can you explain?
Dr. Cuthbert: The DSM criteria have not kept up with developments in genetics, neuroscience, and behavioral science. The system for reviewing research grant applications has been based on DSM-defined disorders, but researchers increasingly want to study certain mechanisms, such as working memory, that cut across disorders. Thomas Insel, the director of the NIMH, proposed that we establish a research classification system looking at dimensional aspects of behavior and of corresponding neural or biological systems like fear circuits or working memory.
TCPR: From a researcher’s point of view, how is RDoC an improvement over DSM?
Dr. Cuthbert: RDoC is a framework for researchers to study behaviors, neural systems, and genetics, either to look at mechanisms that cut across disorders or to look at subtypes or dimensions within a given disorder. For instance, until recently you have had to submit a research grant organized around a DSM category, for example, schizophrenia vs. controls. But if you say you want to study schizophrenia as compared to bipolar disorder, that would be much harder to get approved in review. RDoC has freed up researchers to directly study those concepts of interest to them without being constrained by the DSM system.
TCPR: How is an office-based clinician going to benefit from RDoC?
Dr. Cuthbert: The RDoC project was never intended to be a new distinct diagnostic nosology to compete with the DSM or the ICD. Rather, RDoC is a research project about psychopathology intended to build a research literature that will tell us how we can build a DSM or an ICD that is based upon genetics, neuroscience, and behavioral science. If I am a clinician in my office, RDoC can help me think about a disorder somewhat differently. Instead of saying, “Okay, this patient has social phobia, that patient has schizophrenia.” I am actually looking at specific aspects of that patient’s dysfunction and disability and treating that, rather than treating a diagnosis.
TCPR: This is what many psychiatrists do intuitively, is it not?
Dr. Cuthbert: Yes, in many ways we know that this is what clinicians do now. Often with a complex syndrome like schizophrenia, you are treating specific aspects, such as paranoid features, cognitive problems, or social problems. RDoC would simply provide a framework to study these aspects and to think about them systematically.
TCPR: Down the road, how do you anticipate this will change psychiatric diagnosis?
Dr. Cuthbert: It is a little early to say. I think this is the type of question that the current research is designed to answer. For instance, I could see us still diagnosing schizophrenia and bipolar disorder and looking for subtypes. But I could also see us diagnosing a broader syndromal psychotic spectrum disorder, and then specifying features such as severe cognitive dysfunction or mood disruption. Overall, we consider this as psychiatry’s approach to “precision medicine.” As in other areas of medicine like cancer, cystic fibrosis, or diabetes, we are increasingly realizing that broad clinical phenotypes are not valid at a more fine-grained molecular or genetic level. And that is very important for prevention, as well. For instance, if a person showed difficulties on a very sensitive cognitive task before ever developing a symptom of a disorder, we might have an intervention that is suitable—not necessarily a psychotropic drug, but possibly a behavioral or neuroplasticity intervention to prevent more severe disease.
TCPR: There is a common perception that there is antagonism between the DSM and RDoC. Why is this?
Dr. Cuthbert: I know there is a perceived antagonism, but in fact, NIMH and the framers of the DSM-5 and WHO officials have very similar goals. DSM-5 is committed to having what they call a “living document,” by which they mean they will continually update DSM categories or subcategories as the data emerge from new research. We believe that we have complementary interests.
TCPR: DSM diagnostic criteria are the centerpiece of most drug development and pharmaceutical research today. How might the RDoC affect that area?
Dr. Cuthbert: Over the last several years, we have seen a large-scale withdrawal of the pharmaceutical industry from CNS drug development, especially for mental disorders. The usual reason given is that our disorders are so heterogeneous and vaguely defined that they lack good targets, unlike their productive targets in cancer, diabetes, and other areas. So we are trying to provide a system based on actual genetics and neuroscience that would offer them more precise targets.
TCPR: And this would be similar to how they develop drugs for other conditions?
Dr. Cuthbert: I think the industry is realizing that in an age of precision medicine they are going to have to revamp their business models. An example is the drug ivacaftor (Kalydeco) for cystic fibrosis. This is a rationally-engineered drug developed for one particular mutation of the CFTR gene that is present in only four percent of patients with cystic fibrosis. It was approved by the FDA after a very short clinical trial, which showed it was extremely effective for the patients who had that mutation, but didn’t do much of anything for the other 96 percent. This drug is also spurring feverish research to look at all the other genotypes and engineer new drugs that would target those. [Editor’s note: as of July 2014, the price of Kalydeco is over $300,000 per year.]
TCPR: Have you introduced any RDoC-based research initiatives to discover new psychiatric drugs?
Dr. Cuthbert: We have one contract mechanism in place called the Fast-Fail Trials Initiative, to screen new drugs for further testing. Rather than looking at broad symptom categories and performing a clinical trial to see if the drug affects symptoms, we require investigators to show that their drug engages a particular receptor or affects neurotransmitter signaling (eg, through PET or fMRI imaging), or that it alters brain function in some way (eg, by cognitive testing). That way, if a drug fails, we know why it failed: either it didn’t engage the target we thought, or it didn’t really change anything in the brain even though it reached the molecular target. The idea is to fail quickly and move onto the next prospective drug rather than spending many years and millions of dollars in animal models and then failing in a large clinical trial.
TCPR: Switching gears for a moment, there is a criticism that psychiatry is too biologically oriented—and that human behaviors and emotions transcend mere biology. With RDoC, might we exclude people who would benefit from psychiatric care simply because they don’t have particular biomarkers or genotypes?
Dr. Cuthbert: That is a very good question. I will give you a four-part answer. First, a good interview will always be an important part of psychiatric care, so we are not disavowing that. Second, we are trying to include neuroscience to help us to a more precise diagnosis, but that doesn’t mean we are throwing out ideas about phenomenology. We are not just proposing biological tests like MRIs or EEGs; we may also develop very precise psychometric tasks to measure things like reward-related behavior and reports of pleasurable activities. We advocate for looking at the whole range of how people actually function.
TCPR: And the third and fourth points?
Dr. Cuthbert: The third point is that RDoC includes a focus on environmental effects on neurodevelopment. Just because we include biology in a diagnosis doesn’t mean that many of the etiological factors in a person’s psychosocial milieu or environment are ignored. And finally, we are trying to move toward primary prevention: developing measures that can pick up incipient problems before they become manifest. You can’t do that unless you have these measures that are presymptomatic. We don’t know if down the line, there may be some subtypes of patients for whom no predictive biosignature exists. But our view is, “let’s do the research to find out.”
TCPR: Can you explain the RDoC matrix?
Dr. Cuthbert: It is a four-dimensional matrix—but we visually present it as a 2 x 2. The first dimension is divided into five major domains of functioning. They include:
- Aversive properties—those systems that respond to aversive/negative situations. We call it “negative valence.”
- Appetitive properties, or a “positive valence”—working toward rewards, habits, responding to reward, and so forth
- Cognitive systems.
- Systems for social processes.
- Arousal and regulatory systems, like serotonin systems and the ascending reticular activating system.
[Editor’s note: Learn more about the RDoC matrix.]
TCPR: And the second dimension?
Dr. Cuthbert: The second dimension includes what we call “units of analysis.” These are different ways that you can study the functional domains of the first dimension. They include genetics; molecular processes and cellular processes; measures of circuits; physiological measures, like heart rate, skin conductance, or serum cortisol; behavioral measures, like an assessment battery; and self-reports, which are defined to include questionnaires as well as structured diagnostic interviews. We are not trying to reduce everything down to the molecular level, but rather relate measurements at all these different units of analysis to each other to get a comprehensive understanding of how a particular concept, such as working memory works at multiple levels. This is integrative, not reductionistic.
TCPR: And the third dimension?
Dr. Cuthbert: The third dimension is neurodevelopment. This is important because we now understand most psychiatric disorders to be developmental disorders. And it is vital to understand how the functional domains and the units of analysis described above evolve with the maturing organism.
TCPR: And finally the fourth dimension?
Dr. Cuthbert: The fourth dimension is environmental influences. When something happens in the environment at a certain stage of neurodevelopment it can influence the functioning of all of the above parameters. TCPR: So clinically speaking, we have to look at a patient as a combination of all four dimensions, to guide us towards better therapeutics. Dr. Cuthbert: Yes, although clinicians shouldn’t have to master the whole matrix. It is simply a framework so that we, the researchers, can help deliver more precise diagnoses and more precise ways to deliver clinical care.
TCPR: Thank you, Dr. Cuthbert.