Studying Suicide: Brain Imaging, Neurochemistry, Molecular Genetics Offer Insight

By Anne Harding | Portraits by Jörg Meyer, Illustration by Scott Bakal

The questions scientists ask have shaped the unique suicide research being done at Columbia: How do the brains of suicide decedents differ from other brains? What is the link between inflammation and stress response? Is ketamine an answer to treating depression in suicidal patients? 

Understanding and preventing suicide are two paths that Columbia physicians and scientists pursue in a multipronged effort to respond to the growing problem of suicide. Other investigators devote their careers to trying to determine why suicide is considered an option for so many depressed and despairing individuals.

A recent success in research that has been translated to patient care is the finding that ketamine quickly reduces suicidal ideation in depressed patients. Because the drug, known as a club drug, can be abused and has negative side effects for long-term use, researchers are studying the drug’s mechanism of action in hopes of identifying a safer but equally fast-acting alternative. 

In one study, leading suicide researcher J. John Mann, MD, and colleagues administered the drug intravenously to depressed patients while they were undergoing brain imaging. “They basically went into the scanner depressed and most came out a lot more cheerful, which is a pretty amazing thing.”

While the mechanisms underlying ketamine’s antidepressant and antisuicidal effects are not yet known, brain imaging results indicate that glutamate, the brain’s major excitatory neurotransmitter, appears to play a role. 

“The reason this kind of research is exciting is that if the antidepressant and antisuicidal mechanism of action of ketamine can be figured out, it could open a door to a new generation of antidepressant medicines that potentially could help people who don’t benefit from current treatments, that could potentially work faster than current antidepressants which can take a month or more to take effect, whereas ketamine works within hours,” says Michael F. Grunebaum, MD, who is conducting clinical research on ketamine for depression and suicidal ideation. 

Ketamine has several side effects that make it a poor candidate for long-term treatment of depression: dissociative symptoms, interstitial cystitis, and possible brain damage. “It would be a huge benefit to have good medical options for reducing suicidal thoughts quickly and safely and that perhaps could be administered in the emergency room to acutely suicidal patients,” says Dr. Grunebaum. “If you have the potential to use a medication that reduces suicidal thoughts in a few hours that could potentially reduce length of stay in the overcrowded emergency room and avoid the need for costly inpatient hospitalizations, that would be wonderful.”

Finding a Way To Use Personalized Medicine To Prevent Suicide

Dr. Mann uses functional brain imaging, neurochemistry, and molecular genetics to investigate the causes of depression and suicide. His research, and most studies in the field, are based on the stress-diathesis model of suicidal behavior. This model proposes that genetic and nongenetic or epigenetic factors mold the brain, cognition, and stress responsivity to create a predisposition to suicidal behavior. Certain neurochemicals, such as serotonin and norepinephrine, and dysregulation within the hypothalamic pituitary adrenal axis make some individuals especially vulnerable to stress. Early-life adversity, such as physical or sexual abuse, can change the structure and chemistry of the brain through epigenetic mechanisms that are chemical modifications of the genome intended to improve adaptation to the environment. 

Understanding which depressed individuals will become suicidal is crucial. “The health care system cannot focus equally on every patient with a psychiatric disorder.”

An internal stressor such as a psychiatric illness and an external stressor such as an adverse life event can trigger suicidal behavior in such vulnerable individuals. “You rarely see a suicide in the absence of a psychiatric illness, and most of the time when people die the illness is not being treated,” says Dr. Mann. Research, prevention, and treatment should focus on why suicidal patients are not being treated so underlying problems can be addressed. 

Given the role of early life experience in molding brain development, several complementary studies are underway, including animal experiments, which replicate the effects of early-life adversity, and MRI and PET scans on the brains of people who have died by suicide. Dr. Mann’s initial studies, done with Victoria Arango, PhD, and Mark Underwood, PhD, used postmortem brain tissue from suicide decedents to evaluate the biology of suicidal behavior. Now research includes living patients who have a history of suicidal thoughts and behaviors. Data yielded by each study are used to create computational models of neural circuits, regulation of emotion, stress responses, and suicidal behavior that researchers hope will someday be able to predict risk. 

“We’ve been doing this over a number of years, and the measurements have evolved,” says Dr. Mann. “Patients go through an incredibly complex and advanced set of measures. They get multimodal brain imaging with three major modes of imaging, and then they get tested in the lab for stress responsiveness and how they handle these stressors, and then we use a method called ecological momentary assessment to see if they’re responding to social interactions and adversity in the same way that they seem to respond in the lab.”

Through this research, funded by a grant for postmortem neurochemical studies of suicide and entering its 16th year with support by a major NIMH center grant, Dr. Mann and his colleagues have identified four behavioral indicators that characterize depressed suicidal individuals and have mapped the indicators to specific areas of the brain.

Mood regulation: Depressed suicidal people tend to be pessimistic. “When they’re depressed they feel more profoundly depressed, desperate, hopeless, and pessimistic about the future than other patients,” says Dr. Mann. “They are more likely to devalue the probability of getting better.”

Decision-making: Depressed suicidal individuals have a different approach to making decisions. Suicide, which will immediately relieve their psychic pain, looks like a better choice than trying an antidepressant, which can take several weeks to work and might not even help. “Everybody’s got a sort of sweet spot where they switch between the delayed and the immediate reward. These people tend to switch to the immediate reward earlier,” Dr. Mann says. “This decision-making style also has its own neural circuitry.” 

Social perceptions: Depressed suicidal patients often seem reluctant to ask for help. “The reason they’re reluctant is important and discoverable,” says Dr. Mann. “They tend to be hypersensitive to critical social cues and less sensitive to positive social cues, so they perceive the world to be more critical and less helpful than other people do. We can also see this in their brain scans. We know which bits of the brain are involved in social cognition, and the response to positive signals is weaker and the response to negative signals is stronger. So it’s not just what they tell us. We can see this in the brain.”

Learning and problem-solving: When faced with challenges, depressed suicidal individuals are not good at coming up with solutions. They are less able to formulate new approaches to problems and tend to respond more rigidly. They feel like they’ve run out of options and feel stuck. 

While brain modifications in these four areas likely represent a combination of genetic and epigenetic changes with the latter seeking to improve an individual’s resilience and survival, Dr. Mann notes, “sometimes it backfires, and when you grow up some of these modifications are really not helpful, they are potentially harmful.” By the time a person has their first episode of depression, he adds, “they probably already have the predisposition wired in.”

The good news, Dr. Mann notes, is that each of these risk factors is potentially modifiable. “The mood part is treatable, the decision-making is also educable. You can change the options by getting faster and better antidepressants, for example, or you can change the way the person approaches decision-making. With social perceptions, biofeedback can change people’s response to social signaling, and problem-solving is also something that one can address. People can be trained to tackle problems in a more systematic and constructive fashion, which may give them more options and choices.”

He adds that identifying the different components may help clinicians design a treatment strategy that is more individually relevant, “a kind of a personalized medicine approach to suicide prevention.”

Understanding which depressed individuals will become suicidal is crucial, Dr. Mann adds. “The health care system cannot focus equally on every patient with a psychiatric disorder. The clinician needs some help in figuring out ‘This is a high-risk patient, this is a low-risk patient.’ We need these kinds of multipronged tools for better classifying patients in terms of risk. It’s fundamental to prevention.”

Zero Suicide Initiative

Losing suicidal patients to follow-up is a significant problem for unintegrated health care systems, says Barbara Stanley, PhD, who has developed clinical tools to help. She collaborated with Gregory K. Brown, PhD, of the University of Pennsylvania to create the Safety Planning Intervention, an evidence-based tool that helps clinicians and suicidal patients develop a six-step plan for staying safe after hospitalization or ED visits. 

It is part of the Zero Suicide framework, an aspirational model aimed at reducing suicide in health care systems, that Dr. Stanley and Columbia colleagues are implementing as a quality improvement initiative using evidence-based tools to identify and keep at-risk patients from falling through gaps in the health care system. Of individuals who died by suicide, 40% saw a doctor in the preceding month. Many of the doctors consulted were primary care physicians, sought out because of medical issues: poor appetite, lack of energy, inability to concentrate. Because depression often masquerades as another ailment, doctors who administer tests for cancer or anemia often miss the depression. 

The Zero Suicide approach suggests suicide can be reduced by focusing attention on individuals who are patients in health care systems and by implementing simple strategies to identify, intervene, and refer patients at risk. In collaboration with the New York State Office of Mental Health, Dr. Stanley is leading an NIMH study to test the efficacy of the Zero Suicide model training in 165 behavioral health clinics across New York state. “We have provided clinicians in the community a series of strategies, kind of a template or schematic, for suicidal patients who come to their clinics,” Dr. Stanley explains. “Integrating services is a tough challenge, but we need to work toward reducing suicide among our patients to zero. These patients are right in front of us and we have the opportunity to intervene. We have to attack the problem from all these different perspectives at this point, because whatever we’re doing is clearly not enough.”

Studying Responses to Stress

Dr. Stanley also studies suicidal individuals in the lab, investigating the relationship between stress responsiveness and suicide risk, and is collaborating with M. Elizabeth Sublette, MD, PhD, to identify associations among stress, suicidal behavior, and inflammation. “While patients are doing a stress test, we measure markers of inflammation to see if there is a correlation between inflammatory markers and stress response and if that pattern is different in suicidal people versus nonsuicidal people,” Dr. Stanley explains. 

With a grant from the American Foundation for Suicide Prevention, Dr. Sublette also has investigated the association between three potentially modifiable factors linked to suicide risk in depressed patients: plasma proinflammatory cytokines, plasma omega-3 polyunsaturated fatty acids, and aggression. She previously found associations between low levels of omega-3 fatty acids and the risk of a later suicide attempt and has found correlations between levels of omega-3s in the blood and patterns of glucose uptake in the brain. 

If inflammation does turn out to play a role in suicide vulnerability in depressed patients, Columbia researchers will have discovered yet another potentially modifiable suicide risk factor.