Letter to the Editor: New neural perspective on suicidal behavior in adolescent depression: Specific amygdala subregional dysconnectivity with visual and socio-cognitive networks

Abstract

This letter discusses a recent case-control study by Cheng et al, which investigated alterations in resting-state functional connectivity (rsFC) in adolescents with major depressive disorder (MDD) and a history of suicide attempt (sMDD), with a specific focus on the subregions of the amygdala and their connectivity with whole-brain networks. The study included 32 sMDD patients, 33 MDD patients without suicide attempt (nsMDD), and 34 healthy controls, using resting-state functional magnetic resonance imaging. The results revealed that, compared with the nsMDD group, the sMDD group exhibited significantly reduced rsFC between the right lateral amygdala and both the right inferior occipital gyrus and the left middle occipital gyrus. Furthermore, within the sMDD group, rsFC between the right medial amygdala and the right superior temporal gyrus was negatively correlated with self-esteem scores, while rsFC between the left lateral amygdala and the right parahippocampal gyrus was positively correlated with interpersonal stress scores. These findings underscore abnormal functional connectivity between amygdala subregions and brain areas involved in visual processing, emotion, and social cognition, which may represent a potential neural mechanism underlying suicidal behavior in depressed adolescents. Future studies should integrate multimodal neuroimaging and longitudinal designs to validate these findings and provide a neurobiological basis for early identification and intervention of suicide risk in adolescent depression.

Key Words: Major depressive disorder; Adolescent; Suicide attempt; Amygdala; Functional connectivity

Core Tip: Adolescents with major depressive disorder and a history of suicide attempts exhibit reduced resting-state functional connectivity between the right lateral amygdala and visual association areas (bilateral occipital lobes), along with abnormal connections involving the parahippocampal gyrus and superior temporal gyrus. These neural patterns are further correlated with lower self-esteem and higher interpersonal stress, offering an integrated neurobiological model linking emotional regulation, self-evaluation, and environmental stress in adolescent suicidality. The findings highlight the potential of amygdala subregion-based analysis in identifying specific neural circuits associated with suicide risk, paving the way for future biomarker development and targeted interventions.

TO THE EDITOR

We read with great interest the recent study by Cheng et al[1], which investigated alterations in resting-state functional connectivity (rsFC) of amygdala subregions in adolescents with major depressive disorder (MDD) and a history of suicide attempts. Globally, suicide has become the second leading cause of death among adolescents, and MDD is a primary risk factor for adolescent suicide[2,3]. In adolescents diagnosed with MDD, the lifetime prevalence of suicide attempts is approximately 25%[4]. Moreover, adolescents with MDD show a higher prevalence of suicidal ideation compared to adults with the condition[5]. Therefore, it is particularly important to investigate the underlying mechanisms of suicide in adolescents with MDD. The authors compared rsFC patterns among three groups: Patients with MDD who attempted suicide (sMDD), MDD patients without suicide attempts (nsMDD), and healthy controls. The study revealed that, compared to the nsMDD group, the sMDD group exhibited decreased rsFC between the right lateral amygdala (LA) and bilateral occipital regions, including the right inferior occipital gyrus and the left middle occipital gyrus. Additionally, both patient groups showed aberrant connectivity in other brain regions such as the parahippocampal gyrus (PHG) and fusiform gyrus. These findings provide valuable evidence for understanding the neural mechanisms underlying suicide risk in adolescent depression.

While this study offers novel perspectives, several aspects warrant further discussion. First, the observed weakened connectivity between the LA and visual association cortex may reflect impairments in the integration of emotional and sensory information in adolescents with suicidal tendencies, and such impairments could further compromise their threat perception or emotion regulation abilities[6,7]. However, it would be more illuminating if future analyses could explore whether these connectivity patterns are linked to specific cognitive-affective processes, such as attentional bias toward negative stimuli or impaired reward processing. Furthermore, the correlation between rsFC of the right medial amygdala (MeA)-superior temporal gyrus (STG) and self-esteem scores underscores the potential role of social interaction and self-referential processing in suicide risk[8,9]. Future studies that incorporate task-based functional magnetic resonance imaging (MRI) to examine the dynamic interactions within these networks during specific social or self-evaluation tasks could help clarify the underlying mechanisms more precisely. We commend the authors for their valuable contributions to elucidating the neurobiological underpinnings of suicidal behavior in adolescents. We look forward to seeing these findings stimulate further longitudinal and multimodal research to clarify causal relationships and identify potential therapeutic targets.

MAIN FINDINGS AND IMPLICATIONS OF THIS STUDY

The main findings of this study reveal that adolescents with MDD and a history of suicide attempts exhibit specific alterations in functional connectivity between the amygdala, a key region for emotional processing, and brain areas involved in visual processing, memory, and social cognition, such as the occipital lobe, PHG, and STG[10,11]. More importantly, these neural connectivity patterns are closely correlated with clinical psychological features: Stronger functional connectivity between the right MeA and the STG was associated with lower self-esteem, whereas stronger connectivity between the left LA and the PHG was positively correlated with the level of interpersonal stress experienced by the individuals.

Theoretically, these findings establish a neurobiological model that integrates abnormalities in brain circuits responsible for emotional regulation with external environmental factors such as interpersonal stress and internal psychological characteristics, particularly low self-esteem, thus providing a more comprehensive framework for understanding the mechanisms that underlie suicidal behavior. Clinically, these specific connectivity patterns offer promise for the future development of objective biomarkers to identify adolescents at high risk of suicide, and present new targets for developing precise interventions aimed at specific neural circuits, such as targeted psychotherapy or neuromodulation techniques. Ultimately, from a neuroscience perspective, this study underscores the importance of focusing on interpersonal relationships and self-esteem in adolescents, offering key scientific evidence to guide early prevention and mental health education efforts in schools, families, and society.

The functional connectivity alterations observed in our study, specifically, the diminished connectivity between the right amygdala and visual cortex, along with aberrant connections involving the PHG and STG, are supported by recent multimodal and neurodevelopmental evidence. Firstly, these functional disruptions may reflect underlying structural abnormalities. A 2025 postmortem MRI study demonstrated significant reductions in gray matter volume in the amygdala, PHG, and broader anterior limbic regions, including the STG, among individuals who died from suicide[12]. This suggests that the weakened functional connectivity identified in our study may stem from compromised structural integrity in these key nodes. Secondly, these alterations should be interpreted within the developmental context of adolescence. Pubertal surges in gonadal and adrenal hormones exert organizational effects on the structure and function of brain regions such as the amygdala and hippocampus, shaping their response to social stress[13]. During this sensitive period, pre-existing functional weaknesses in circuits supporting amygdala-visual integration and parahippocampal self-referential processing may impair an individual’s ability to effectively regulate interpersonal stress (reflecting high interpersonal pressure) and maintain stable self-evaluation (reflecting low self-esteem), thereby jointly increasing suicide risk. Thus, the specific functional connectivity patterns revealed in our study align with emerging structural and neurodevelopmental frameworks, and together delineate a multimodal, developmentally sensitive neurocircuitry model of suicide vulnerability.

Strengths of the study

The strengths of this study are primarily reflected in its precise design and multidimensional correlation strategy. First, the study performs a fine-grained analysis at the amygdala subregion level, distinguishing between the functionally distinct lateral and MeA. The discovery that these subregions correlate with different brain networks and psychological dimensions overcomes the ambiguity inherent in prior studies treating the amygdala as a unified structure, thereby providing more precise insights into neural mechanisms[14]. Secondly, the study successfully establishes correlations between brain functional connectivity data and clinically meaningful psychological scales such as self-esteem and life events. This approach not only builds a bridge between brain activity and behavior but also constructs a multidimensional explanatory model integrating neural, psychological, and environmental factors, substantially enhancing the depth and practical value of the research findings.

Limitations of the study

This study has several methodological limitations that merit careful consideration, as they may significantly affect the interpretation and generalizability of the results. In terms of sample representativeness, all participants were recruited from the inpatient population of a single specialty hospital. While this approach ensured diagnostic consistency, it may have introduced considerable selection bias. Inpatients generally represent a clinical subgroup with more severe symptoms requiring urgent intervention, which calls for great caution when extrapolating the findings to outpatient, community-based populations, or adolescent depression groups in regions with varying healthcare resources. Moreover, after applying stringent head motion criteria, the final sample size for each of the three subgroups was only about 30. This limited sample size not only reduces statistical power and may obscure true group differences but also undermines the reproducibility and stability of the results. This issue is especially relevant in analyses involving multiple group comparisons and covariate adjustments, where the statistical models may demonstrate limited robustness.

At the research design level, the cross-sectional nature of the study constitutes a fundamental methodological limitation. As all data were collected at a single time point, it remains unclear whether the observed amygdala-cortical functional connectivity alterations represent a prospective risk marker for suicidal behavior, an adaptive or injury-related change following the traumatic event of a suicide attempt, or a result of confounding by other unmeasured variables. This design is inherently inadequate for revealing the dynamic causal relationship between brain functional connectivity and suicide risk over time.

Several noteworthy limitations exist regarding the control for confounding variables in this study. First, positive psychotic symptoms and anxiety symptoms were not assessed in the sMDD and nsMDD groups. Therefore, if there were significant differences in the severity of these symptoms between the two groups, the observed differences in brain connectivity might reflect more than just the specificity of suicidal behavior[15,16]. Second, the exclusion criteria did not clearly indicate whether systematic assessments were conducted for comorbid diagnoses such as borderline personality disorder or post-traumatic stress disorder, which are strongly associated with suicidal behavior. The distinct neural correlates of these disorders may confound the interpretation of the current findings. Furthermore, although participants were required to be medication-free prior to admission, the study did not specify whether sedative medications or non-antidepressant drugs administered between admission and the MRI scan were considered. Such medications could potentially modulate functional brain connectivity in the short term.

At the methodological and technical level, the inherent limitations of functional MRI must be taken into consideration. Although a strict head motion exclusion criterion of 2 mm/2° was applied, residual micromotion in adolescent populations and potentially more agitated depressed patients may still systematically affect functional connectivity estimates. This effect is particularly sensitive for signals from deep gray matter nuclei such as the amygdala[17]. The residual head micro-motion correction methods should be detailed. Furthermore, the study did not clearly specify whether blinding procedures were implemented between clinical group assessment and imaging data analysis. If clinical assessors were aware of the neuroimaging results, or if imaging analysts knew the clinical group assignments, confirmation bias could have been introduced. It is recommended that future studies should strictly implement double-blinding for both assessors and analysts. Finally, rs-FC, being a correlational measure, cannot reveal dynamic abnormalities in these circuits during specific cognitive or emotional processing. This inherent constraint limits direct inference regarding the underlying cognitive mechanisms. Finally, while this study aimed to establish abnormal amygdala subregion connectivity as a biomarker for adolescent suicide risk, it did not assess its diagnostic efficacy (e.g., sensitivity and specificity). Future work should validate this biomarker by incorporating receiver operating characteristic curve analysis. Moreover, based on such findings, a framework for targeted interventions using existing neuromodulation techniques (e.g., transcranial magnetic stimulation) could be proposed to modulate the implicated aberrant circuits.

DISCUSSION

This study systematically investigated abnormal patterns of amygdala subregion functional connectivity in adolescent depression with suicide attempts using resting-state functional MRI. The results revealed that, compared to depressed adolescents without suicide attempts, patients with a history of suicide attempts exhibited significantly weakened functional connectivity between the right LA and the right inferior occipital gyrus, as well as the left middle occipital gyrus. This finding suggests that disrupted integration between the amygdala, which is responsible for primary emotional responses, and higher-order visual cortical areas may play a critical role in suicidal behavior among adolescent depression patients. More importantly, this study is the first to identify intrinsic relationships between specific neural circuits and psychosocial factors in an adolescent cohort: The strength of the right MeA-right temporal pole: STG (TPOsup) circuit was negatively correlated with self-esteem levels, whereas the left LA-right PHG circuit was positively correlated with interpersonal stress. These associations provide a novel perspective for understanding the neuropsychological mechanisms of suicidal behavior, indicating that the brain’s emotional processing network not only interacts with external stressors but is also closely related to an individual’s self-evaluation system.

CONCLUSION

By employing a refined analytical approach at the level of amygdala subregions, this study moves beyond the traditional paradigm of treating the amygdala as a unitary functional entity, thereby enabling more precise identification of neural circuits specifically associated with suicidal behavior. These findings not only deepen the understanding of the neural basis of suicidal behavior in adolescent depression but also provide an important foundation for the future development of targeted biomarkers and intervention strategies.