Mood Biomarkers FAQ
Can you summarize your current research?
We found patterns of change in the blood that reflect whether a person has low mood (depression) or high mood. These changes occur at the level of gene expression (how active a gene is) and involve genes previously thought to act primarily in the brain.
Unlike cases of cancer, you can’t biopsy the target organ (brain) in psychiatry. Having a readily accessible peripheral readout for blood would be great - even better (i.e. easier) than having a readout for cerebro-spinal fluid (CSF), which has been another area studied. However, such study requires a painful procedure for the patient (spinal tap).
Gene expression is more informative than genetic mutations in DNA - especially in terms of disease state - as it reflects gene activity. It is the place where rubber-meets-the-road in terms of gene-environment interactions: effects of drugs, stress, and medications. It may be more informative to know your biomarkers than to have your genome sequenced.
What are the advantages of a blood test for mood disorders?
In psychiatry, we currently rely on patients’ self-reported symptoms, as well as the clinician’s impression. Patients are not sure as to the severity of their illness, and neither is the clinician (sometimes dismissing patients' symptoms, sometimes overestimating them). I am a psychiatrist as well as a researcher, and I often face that dilemma in my own clinical practice, where I see complicated cases.
Having an objective test to measure disease state, severity, and (especially) response to treatment would be a big step forward. Such a measure would be similar to your primary care doctor or cardiologist ordering a baseline lipid panel, starting you on a medication, and conducting repeat lab testing to see how well the medication is working. It would place psychiatry on par with other medical specialties.
For example - nowadays for depression - a patient gets started on an antidepressant. Weeks or months may pass before the patient and doctor know if the particular medication is working, or if a different antidepressant should be administered. A blood test for mood state could objectively reflect the efficacy of current treatment. Perhaps changes in the blood could be caught early-on, reducing the duration of uncertainty (trial and error). Knowing early-on that a medication is moving things in the right direction is reassuring - and frankly - therapeutic to patients. Hope and optimism are powerful (yet poorly understood) neurobiological players: we know this due to the high rate of initial placebo responses with anti-depressants.
Another example: a patient is diagnosed as depressed, but is actually bipolar (manic-depressive); this is a common problem. The patient may be started on an antidepressant initially, and if bipolar, will be flipped by the antidepressant into a mixed state or frank mood elevation (hypomania or mania). Patients could be monitored with a panel of mood-state markers after they have begun taking an antidepressant. If these markers indicate a shift to elevated mood, several interventions can be employed: medications can be changed; a mood stabilizer added to the regimen. A potentially dangerous - and certainly, miserable - episode for the patient can be averted. This is especially important in children and adolescents, who are difficult to diagnose using only clinical criteria. In such younger patients, mood states may quickly change, sometimes dangerously.
Last but not least, pharmaceutical companies can use mood-state biomarkers to aid in the discovery of new medications: the process will become less hit-or-miss. Currently, developing a new drug is like developing a Hollywood blockbuster: no guarantees that it will pan-out, and sequels are a safer bet.
We emphasize that we are in the early stages of this work, and a lot of prospective work needs to be finished. We need to establish: which markers work best; potential subpopulation, age, and gender differences, and; response to different medications. Psychiatry - like the rest of medicine - will likely become tailored to the individual based on their biomarker profiles.
Aside from depression and bipolar disorder, what other diseases could these findings diagnose?
Mood state is important and sometimes overlooked in favor of complex disorders: postpartum depression (trivialized as "baby blues"), post-traumatic stress disorder, depression following myocardial infarction and cardiovascular procedures, depression associated with cancer, fibromyalgia, and mood changes associated with other serious diseases. Having an objective measure would expedite appropriate attention and treatment to patients.
We are also doing similar work in schizophrenia, looking for blood markers associated with delusions and hallucinations. Schizophrenia is a complex disorder, with subjectivity and uncertainty about illness severity contributing to it intricacy. We are excited about our unpublished findings in this area, and hope to quickly advance such work. NIH funding is on the decline these days. I think philanthropy can play a major role for funding such early-stage, unconventional research.
Last but not least: we are doing some work on biomarkers for anxiety and stress, using a new animal model we have developed, as well as studies in human blood samples. This work, in addition to my collaborators at Indiana University (notably Helen Le-Niculescu, and John Nurnberger) is conducted in collaboration with colleagues at The Scripps Research Institute in La Jolla, CA, my PhD alma mater (notably Dan Salomon, and more recently Nik Schork), and with advice from Ming Tsuang and others at UC San Diego, my clinical alma mater. We are looking at ways of collaborating with the military for expanding such studies, since they are of high-relevance to the effects of combat stress.
There is a role for industry - including biotech - in this area. We hope the FDA will take an active role in evaluating, regulating and approving any such particpation. I predict that tests like these may hit the market in about five years, plus-or-minus two years. A lot more work in this area needs to be done... and will be done.
Two drugs were found to exert similar effects on both high and low mood. Can you expound on the implications of this and how this data could aid drug development?
In brief: we are looking at patterns changed in the blood with relationship to mood state, and comparing these pattern changes with a database at the Broad Institute at MIT (containing signature patterns of a large number of known drugs). Using the database, we can ascertain which drugs exert a similar effect. The database might suggest new drugs (or at least chemical structures to be tweaked and developed by pharmaceutical companies) to be explored in the treatment of mood disorders.
Can you expound on the implications of the overlap between some genes identified in your study and those involved in cancer?
It is a surprise to us, and we do not want to over-interpret it. Other colleagues in the field (Husseini Manji at NIMH, Francine Benes at Harvard, Huda Akil, colleagues at Michigan, and others) obtained similar results, and now the full picture is beginning to make sense. Nature is a tinkerer, and has a limited number of building blocks at its disposal: nature seems to be re-using some of the same molecules used for "lower" functions (such as cell proliferation and cell death) and higher mental functions (such as mood regulation). From philosophical and evolutionary standpoints, one can speculate about the normal roles of this particular overlap-expansiveness of cells and of behavior (high mood) when the environment is favorable for expansion. Shrinkage of cells, apoptosis (cell death), retraction into self (low mood), and suicide occur when the environment is not supportive or hostile.
An interesting example: some athletes may take performance-enhancing drugs such as steroids. These athletes may have an expansive - and sometimes boorish - behavior in tandem with their expanding muscles, and are depressed when they get off these drugs, in tandem with their muscle shrinkage. We speculated about this "brain as a muscle" model in an article in the journal Genome Biology in 2005.
Pragmatically, we do not know for certain if mood drugs may help with treating cancer, or if cancer drugs may help with treating mood, but there are interesting hints in both directions, and this may turn out to be a very fruitful avenue to walk.
Also: are mood disorders associated with different types of cancer? A lot of empirical work needs to be done.