2022 Articles of Import and Impact

Glucocorticoids, such as cortisol, are essential for guiding normal brain development, but when dysregulated during critical developmental periods, can also have deleterious effects. Additionally, glucocorticoid release from the adrenal glands constitutes a primary component of the physiological response to stress. The paper by Cruceanu and coauthors (1) in the May 2022 issue of AJP uses human cerebral organoids to begin to understand the influences of glucocorticoids on human brain development. It is an exciting report and an important addition to this year’s compendium of AJP papers, foreshadowing the use of cerebral organoids to understand—at the most basic level—the molecular and cellular alterations that underpin psychiatric illnesses. This in vitro model uses pluripotent stem cells as a substrate to form a three-dimensional matrix of brain cells that allows for an understanding of interactions among different developing brain cells. In addition to highlighting the use of human brain organoids as a viable method for exploring molecular and cellular mechanisms underlying the development of psychiatric illnesses, this paper provides insights into the influences of glucocorticoids and glucocorticoid receptor activation on fetal brain development. This is very relevant to psychiatry as significant perinatal stress is well known to be a general risk factor for the development of psychiatric illnesses. This paper’s findings identify the developing brain cells that are most affected by glucocorticoid exposure and begin to provide an understanding of how glucocorticoid exposure negatively impacts normal differentiation and maturation processes. Additionally, the researchers found that the set of neuronal genes that was most affected by glucocorticoids was enriched for genes associated with psychiatric disorders. This paper provides a glimpse into the future in which the use of stem cells and cerebral organoids will not only provide insights into molecular and cellular mechanisms underlying psychiatric illnesses, but because they can be derived from individual patients, hold the promise of guiding personalized treatment approaches.

My favorite AJP article of 2022 is by Jaffe et al. (2) and examines shared versus distinct neurobiological mechanisms of posttraumatic stress disorder (PTSD) and major depressive disorder (MDD) using RNA expression data from postmortem brain samples. PTSD is highly comorbid with MDD with a large fraction of overlapping symptoms as well as shared genetic and environmental risk factors. However, how and whether this is embedded in shared neurobiological changes at the molecular and cellular level is not understood and the focus of this manuscript.

The authors use RNA sequencing data from over 300 postmortem brain samples from PTSD and MDD cases and neurotypical controls from two brain regions strongly implicated in the pathophysiology of both disorders—the prefrontal cortex and the amygdala—for their relevance in emotion and fear regulation. The authors generated data from two subregions of the prefrontal cortex, the dorsolateral prefrontal cortex and dorsal anterior cingulate cortex, as well as from two subregions of the amygdala, the basolateral amygdala and medial amygdala, for a total of 1,285 samples from 325 unique donors. This represents the largest postmortem brain RNA sequencing study in PTSD so far.

Overall, more genes differentially expressed between cases and controls were found in the cortical areas than the amygdala, and more comparing MDD than PTSD versus neurotypical controls. There was a high concordance between transcriptional effects of PTSD and MDD, and relatively few differentially expressed genes were identified in a direct comparisons between the two disorders. These analyses confirmed a shared involvement of immune processes and microglia function across both disorders. The limited number of PTSD-specific changes pointed to subpopulations of GABAergic inhibitory neurons as relevant for this disorder.

This study highlights shared neurobiological changes in MDD and PTSD, mainly related to neuroinflammation and microglia function but also identifies inhibitory neurons as potentially more specifically affected in PTSD. This large postmortem brain study greatly contributes to our understanding of the pathomechanisms involved in these disorders but also highlights the necessity of future studies increasing the number of investigated brain regions and the cell type resolution.

With the increasing legalization of medicinal and recreational marijuana use in the United States, there have been substantial increases in cannabis use. While the legalization of any substance often leads to the perception that the use of the substance is relatively safe, there is a considerable lack of scientific data about both short and long-term consequences of cannabis use. The article “Long-Term Cannabis Use and Cognitive Reserves and Hippocampal Volume in Midlife” by Meier and colleagues (3) reports on longitudinal data from over one thousand individuals in New Zealand who were assessed for socioeconomic, cognitive, health, and psychosocial and psychiatric variables (including substance use) 13 times between the ages of 3 and 45. Of importance, at the age 45 follow-up, 94% of the sample completed the assessments and 93% participated in a structural MRI. After controlling for a number of relevant factors, the authors found cognitive deficits in long-term cannabis users at age 45, including declines in IQ, which were significant compared to noncannabis users and chronic alcohol or tobacco users. They also found that the long-term cannabis users had a smaller hippocampal volume, but this did not statistically mediate the cannabis-related cognitive deficits. These findings are important in informing risk assessments about cannabis use and in the clinical evaluation of cognitive decline in mid and older age individuals. Perhaps of even greater value is the fact that there is a great deal of valuable additional information to be derived from this careful and comprehensive longitudinal study and the data that it will continue to generate. I look forward to more contributions from this research group.

Reducing the duration of untreated psychosis (DUP) in persons with first-episode psychosis to limit morbidity and mortality in affected individuals is a major goal of early intervention programs. Although multiple studies have reported positive effects of reducing DUP on short-term outcomes, it remains unclear whether these effects are enduring. In addition, the results of some analyses have suggested that the apparent predictive impact of DUP on clinical and life quality outcomes might reflect confounding by other factors such as premorbid features or lead-time bias. The study by O’Keeffe and colleagues (4) addresses these issues by applying linear mixed-model analyses to assess the relation between DUP and prospectively assessed symptoms, functioning and quality of life over a 20-year follow-up period in a large cohort of persons with first-episode psychosis. They found that shorter DUP was associated with better outcomes that persisted over two decades. These associations were robust to effects of premorbid features and were not confounded by lead-time bias. The study did reveal substantial variability across individuals in the time course of positive outcomes and in the degree to which they were influenced by baseline factors. In aggregate, the results provide additional support for the clinical importance of early intervention programs designed both to identify individuals at risk for psychosis and to apply appropriate treatment for psychosis as quickly as possible. The findings also suggest that such programs, in addition to having long-term positive benefits for many affected individuals, are likely to provide a positive return on the societal investment in such programs.

The administration of repetitive transcranial magnetic stimulation (rTMS) can be burdensome as it requires 37.5 second duration treatments, 5 days a week for 4–6 weeks. The antidepressant response can take weeks and there are questions regarding the efficacy and durability of rTMS in treatment resistant depression (TRD). The multicenter trial by Blumberger et al. (5) demonstrated that the administration time of TMS could be shortened to 3 seconds using intermittent theta burst stimulation (iTBS) and that iTBS was noninferior to rTMS. Cole et al. (6) took advantage of the shortened stimulation time and accelerated the treatment course from 4 to 6 weeks to 1 week, administering iTBS treatments once an hour, 10 treatments a day over 1 week. In a protocol termed Stanford Neuromodulation Therapy (SNT—formerly the Stanford Accelerated Intelligent Neuromodulation Therapy, or SAINT), they also used resting-state functional connectivity to target the region in the dorsolateral prefrontal cortex functionally anticorrelated with the subgenual anterior cingulate cortex. The results were impressive in demonstrating significant decreases in depression in a sham-controlled trial of 29 subjects who were moderately treatment resistant. Practitioners and patients are clearly looking for alternatives to somatic treatments such as electroconvulsive therapy for TRD. While encouraging, the results by Cole et al. are limited by the small sample size and generalizability of this sample, the durability of the treatments including in more treatment resistant patients, and protocol refinements (i.e., do you really need neuronavigation, the optimal time between treatment sessions?). Nevertheless, this paper is an important and encouraging advance in TMS therapy.

Prior studies suggest that relations between symptoms and brain structure might differ across clinical contexts. Longitudinal studies are needed that map relations among symptoms and neural measures in different developmental contexts. Shen and colleagues (7) compare brain development among infants with fragile X syndrome, infants who later developed an autism spectrum disorder (ASD), infants at risk for ASD who did not develop an ASD, and a comparison group. The authors combined longitudinal MRI-based assessments with clinical assessments to compare trajectories of brain structure, clinical symptoms, and their interrelations.

One major finding demonstrated different patterns of brain development across groups. Infants who later developed ASD manifested a unique form of rapid amygdala growth that differentiated this group from other groups. A different pattern manifested among infants with fragile X syndrome, where abnormal growth patterns occurred in the caudate nucleus. A second major finding showed that these two patterns of unique brain development also related selectively to symptoms in each group. Symptoms related to amygdala growth only in the group with ASD; symptoms related to caudate growth only in the group with fragile X syndrome. This clinical-anatomical correlates varied with pathophysiology. This shows that relations between symptoms and brain structure vary across distinct types of neurodevelopmental disorders.

In order to have a fully functional working memory, neural circuit function must be intact. In patients with schizophrenia, working memory is impaired and associated with lower power of synchronized neural gamma oscillation in the prefrontal cortex. The cutting-edge study by Chung et al. (8) drew my attention for three reasons. First, it combines two rigorous approaches (e.g., quantification of proteins in postmortem human cortex and computational modeling). Second, the computer modeling showed that even small alterations in synaptic parameters can have large effects on gamma power. Third, the study uncovers synaptic variability as a key contributor to prefrontal cortex dysfunction in schizophrenia.

Gamma oscillations in the prefrontal cortex are shaped by activity of excitatory pyramidal neurons and GABAergic parvalbumin-expressing interneurons (PIVs). These PIVs, in turn, get excitatory input from neighboring pyramidal neurons that synchronize the firing of the pyramidal neurons at gamma frequency. Of interest to Dr. Chung and colleagues, the strength of these excitatory inputs varies and could impact gamma oscillations across the network. To test this idea, they examined pre- and postsynaptic markers of synaptic strength in PVIs in postmortem human prefrontal cortex from 20 matched pairs of schizophrenia and comparison individuals. Specifically, they found variability of these marks at excitatory inputs across PVIs was larger in schizophrenia. They additionally modeled the network, finding greater variability in excitatory synaptic strength regulated and robustly lowered gamma power. Taken together, this elegant study synergistically employs computation and neuroscience-informed approaches to reveal the mechanisms of synaptic variability and gamma band power in schizophrenia.

The American Journal of Psychiatry Residents’ Journal (AJP-RJ) started our 18th volume with a focus on minority mental health and diversity. An underlying theme of each piece was the importance of respecting and including the individual’s identity throughout medical education and patient care. I commend all the authors and guest editor, Fíona Fonseca, M.D., for bringing these pieces to readers. I want to acknowledge two pieces which highlight what the stresses and stigma of transgender identity can have on individuals seeking medical and psychiatric care. Emma Banasiak presented a case report and discussion about gender minorities and eating disorders (9). Her discussion identifies how gender identity affects biological, psychological, and social factors of patient care and the limited data we have about the interactions with eating disorders. Allegra Condiotte, M.D., M.H.A., described the stigma and discrimination transgender individuals are subjected to in health care, and the limited exposure medical trainees receive in transgender care (10). I hope these pieces inspire readers at all stages of training and practice to learn more as we strive to ultimately dismantle stigma and provide quality health care in an inclusive environment.