A newly published study in the journal Nature involving dozens of past and present BBRF grantees, Scientific Council members and prize winners, among hundreds of co-authors, takes an important step forward in the effort to better understand what various psychiatric illnesses have in common and what makes them distinctive—specifically, at the level of genetics, but also, perhaps, underlying biology.

Comparing illnesses that have separate diagnostic criteria (as comprehensively captured, for instance, in the DSM-V manual used by psychiatrists in the U.S.) is trickier than it may sound. First, many patients with one disorder are ultimately diagnosed with other, ostensibly distinct disorders—for example, those who suffer from depression and anxiety; or from PTSD and anxiety and depression. Some (but by no means all) people with bipolar disorder experience psychotic episodes, which are typically associated with schizophrenia. Many people diagnosed with eating disorders also suffer from anxiety disorders and/or OCD. The list goes on. In such cases, are the comorbid illnesses in any way related? Or wholly distinct in their biological causation and underlying mechanisms? 

Even from the genetic point of view, distinguishing among different diagnoses can be more complex than might be assumed. Researchers have learned over the last 2 decades that there are many overlaps in genetic vulnerabilities for various disorders, in some cases extensive. And to make this realization even more complex, it has also become apparent that many of the genetic variants that confer above-average risk for various psychiatric disorders turn out to be “pleiotropic”: many of them have a multiplicity of potential effects, some of which may be present in one diagnosis and others in other, distinct diagnoses.

Dr. Andrew Grotzinger, of the University of Colorado, one of the lead authors of the new paper trying to “map the genetic landscape across 14 psychiatric disorders,” explains: “Right now, we diagnose psychiatric disorders based on what we [doctors] see in the [examining] room, and many people will be diagnosed with multiple disorders. That can be hard to treat, and disheartening for patients.” The team’s new research, he says, “provides the best evidence yet that there may be things that we are currently giving different names to that are actually driven by the same biological processes.”

As noted by another team leader, Jordan W. Smoller, M.D., Sc.D., a 2002 BBRF Young Investigator, the new findings, among other things, provide key insights into the biological pathways and gene expression patterns of specific brain-cell types that are implicated in various illnesses. Dr. Smoller is director of the Psychiatric and Neurodevelopmental Genetics Unit at the Massachusetts General Hospital Center for Genomic Medicine and the Center for Precision Psychiatry at Mass General Brigham, and a professor of psychiatry and epidemiology at Harvard. “These findings provide valuable clues for advancing our understanding and treatment of mental illness with greater precision,” he said.

The team noted in its paper that “the full scope of shared and disorder-specific genetic influences [upon mental illnesses] remains poorly defined.” The large international team, which included members from 10 Working Groups of the influential Psychiatric Genomics Consortium (PGC), addressed the problem by “triangulating across a suite of cutting-edge statistical and functional genomic analyses” which were applied to 14 childhood- and adult-onset psychiatric disorders. Data from genome-wide association studies (GWAS) and related studies involving over 1 million patients and 5 million healthy controls were used.

The 14 illnesses were: ADHD, alcohol-use disorder, anorexia nervosa, anxiety disorders, autism spectrum disorder, bipolar disorder, cannabis-use disorder, major depression, nicotine dependence, OCD, opioid-use disorder, PTSD, schizophrenia, and Tourette syndrome. Because of a relative paucity of genetic data for people of non-European ancestry, the sample was intentionally restricted to those with European ancestors. Small sample sizes (in this case, for individuals of non-European ancestry) would limit the mathematical power of the analysis. Much more plentiful, diverse samples will be needed as this kind of research moves forward, the team noted.

Analyses of the combined dataset revealed that five underlying “genomic factors” accounted for 66% of the genetic differences between those with any of the 14 disorders and those without them. In addition, the researchers identified 238 genetic variants associated with shared risk pathways across disorders.

The five “factors” referred to by the team are actually five groupings of the 14 disorders, each of which has a “shared genetic architecture,” i.e., genetic features that are common to all disorders in the factor. In one of the five categories, dubbed “SB” because it consists solely of schizophrenia and bipolar disorder, more than 70% of the genetic “signal” for one disorder is part of the signal for the other disorder. This is not unexpected, as it was suggested in other analyses in recent years, but its confirmation in this study underlines what is remarkable about the statistic: at the level of “signs and symptoms” as observed by doctors, these two illnesses have long been viewed as quite distinct. And yet, at the level of genetics, most of the commonly occurring DNA variations associated with elevated risk in both illnesses are the same.

This suggests underlying biological pathway and gene-expression and related irregularities in cells that may be impacted by some of the genetic variants that the two illnesses share. In some cases, and in certain combinations, where present, these gene variations may generate similar symptoms in a subset of patients, such as mood swings, or psychosis, or certain cognitive impairments. Because of pleiotropy (one gene having a role in multiple symptoms), this is very hard to untangle—but the data in the new study speaks to the importance and potential treatment benefits, for patients, of trying to do so.

The four other groupings of the 14 disorders based on genetic architecture are: disorders with compulsive features (eating disorders, Tourette syndrome, OCD, for example); “internalizing conditions” in which symptoms are directed inward (major depression, anxiety, PTSD); substance-use disorders; and neurodevelopmental disorders (autism, ADHD, for example).

In each of the groups, “we saw that the included disorders are more similar than unique,” at the level of genetics, Dr. Grotzinger says. Ideally, this knowledge might lead, he suggested, to “strategies to target them in a different way, one that doesn’t require, say, four separate pills or four separate psychotherapy interventions.”

If that optimistic vision is to be realized, it may rely upon biological observations of commonalities across diagnoses. One example, in the group of internalizing disorders, is that genetic variants related to processes that regulate a cell type called oligodendrocytes tend to be affected. These are cells that help maintain the integrity of the brain’s wiring infrastructure. Another example from the new research is that genes influencing excitatory neurons tend to be overexpressed in the “SB” group composed of schizophrenia and bipolar disorder. Genetic variants found to be characteristic in neurodevelopmental disorders suggest pathways and processes that are active prior to birth which may prove critical in developing potentially relevant future treatments.

Many of these insights are not in themselves novel, but the scope of the cross-disorder genetic analysis, in the team’s view, is part of the process of figuring out how to successfully treat symptoms across individual diagnoses, to the benefit of patients whose conditions are conventionally regarded as needing unique or wholly discrete treatment approaches.

In addition to Dr. Smoller, the team that analyzed the data from the 10 PGC working groups and composed the paper included: Kenneth S. Kendler, M.D., Sc.D., BBRF Scientific Council, 2010, 2000 BBRF Distinguished Investigator, 1995 Lieber Prize winner; Ole A. Andreassen, M.D., Ph.D., 2025 BBRF Colvin Prize winner; Gerome Breen, Ph.D., 2007 BBRF Young Investigator; John M. Hettema, M.D, Ph.D., 2014 BBRF Independent Investigator, 2001 Young Investigator; Emma C. Johnson, Ph.D., 2020 BBRF Young Investigator; James A. Knowles, M.D., Ph.D., 2009 BBRF Distinguished Investigator, 2001, 1993 Young Investigator; Roseann E. Peterson, Ph.D., 2019 BBRF Young Investigator; Elise B. Robinson, Sc.D., MPH, 2014 BBRF Young Investigator; Sandra S. Sanchez-Roige, Ph.D., 2018 BBRF Young Investigator; Murray B. Stein, M.D., MPH, BBRF Scientific Council; and Brad Verhulst, Ph.D., 2023 BBRF Young Investigator.