Resting state functional connectivity subtypes predict discrete patterns of cognitive-affective functioning across levels of analysis among patients with treatment-resistant depression

Dimensional Neuroimaging Endophenotypes: Neurobiological Representations of Disease Heterogeneity Through Machine Learning

Machine learning has been increasingly used to obtain individualized neuroimaging signatures for disease diagnosis, prognosis, and response to treatment in neuropsychiatric and neurodegenerative disorders. Therefore, it has contributed to a better understanding of disease heterogeneity by identifying disease subtypes with different brain phenotypic measures. In this review, we first present a systematic literature overview of studies using machine learning and multimodal magnetic resonance imaging to unravel disease heterogeneity in various neuropsychiatric and neurodegenerative disorders, including Alzheimer's disease, schizophrenia, major depressive disorder, autism spectrum disorder, and multiple sclerosis, as well as their potential in a transdiagnostic framework, where neuroanatomical and neurobiological commonalities were assessed across diagnostic boundaries. Subsequently, we summarize relevant machine learning methodologies and their clinical interpretability. We discuss the potential clinical implications of the current findings and envision future research avenues. Finally, we discuss an emerging paradigm called dimensional neuroimaging endophenotypes. Dimensional neuroimaging endophenotypes dissects the neurobiological heterogeneity of neuropsychiatric and neurodegenerative disorders into low-dimensional yet informative, quantitative brain phenotypic representations, serving as robust intermediate phenotypes (i.e., endophenotypes), presumably reflecting the interplay of underlying genetic, lifestyle, and environmental processes associated with disease etiology.

Dimensional Neuroimaging Endophenotypes: Neurobiological Representations of Disease Heterogeneity Through Machine Learning

Machine learning has been increasingly used to obtain individualized neuroimaging signatures for disease diagnosis, prognosis, and response to treatment in neuropsychiatric and neurodegenerative disorders. Therefore, it has contributed to a better understanding of disease heterogeneity by identifying disease subtypes that present significant differences in various brain phenotypic measures. In this review, we first present a systematic literature overview of studies using machine learning and multimodal MRI to unravel disease heterogeneity in various neuropsychiatric and neurodegenerative disorders, including Alzheimer's disease, schizophrenia, major depressive disorder, autism spectrum disorder, multiple sclerosis, as well as their potential in transdiagnostic settings. Subsequently, we summarize relevant machine learning methodologies and discuss an emerging paradigm which we call dimensional neuroimaging endophenotype (DNE). DNE dissects the neurobiological heterogeneity of neuropsychiatric and neurodegenerative disorders into a low-dimensional yet informative, quantitative brain phenotypic representation, serving as a robust intermediate phenotype (i.e., endophenotype) largely reflecting underlying genetics and etiology. Finally, we discuss the potential clinical implications of the current findings and envision future research avenues.

Biotypes of major depressive disorder identified by a multiview clustering framework

BACKGROUND

The advances in resting-state functional magnetic resonance imaging techniques motivate parsing heterogeneity in major depressive disorder (MDD) through neurophysiological subtypes (i.e., biotypes). Based on graph theories, researchers have observed the functional organization of the human brain as a complex system with modular structures and have found wide-spread but variable MDD-related abnormality regarding the modules. The evidence implies the possibility of identifying biotypes using high-dimensional functional connectivity (FC) data in ways that suit the potentially multifaceted biotypes taxonomy.

METHODS

We proposed a multiview biotype discovery framework that involves theory-driven feature subspace partition (i.e., "view") and independent subspace clustering. Six views were defined using intra- and intermodule FC regarding three MDD focal modules (i.e., the sensory-motor system, default mode network, and subcortical network). For robust biotypes, the framework was applied to a large multisite sample (805 MDD participants and 738 healthy controls).

RESULTS

Two biotypes were stably obtained in each view, respectively characterized by significantly increased and decreased FC compared to healthy controls. These view-specific biotypes promoted the diagnosis of MDD and showed different symptom profiles. By integrating the view-specific biotypes into biotype profiles, a broad spectrum in the neural heterogeneity of MDD and its separation from symptom-based subtypes was further revealed.

LIMITATIONS

The power of clinical effects is limited and the cross-sectional nature cannot predict the treatment effects of the biotypes.

CONCLUSIONS

Our findings not only contribute to the understanding of heterogeneity in MDD, but also provide a novel subtyping framework that could transcend current diagnostic boundaries and data modality.

Distinct patterns of functional brain network integration between treatment-resistant depression and non treatment-resistant depression: A resting-state functional magnetic resonance imaging study

BACKGROUND

Previous neuroimaging has paid little attention to the differences in brain network integration between patients with treatment-resistant depression(TRD) and non-TRD (nTRD), and the relationship between their impaired brain network integration and clinical symptoms has not been elucidated.

METHOD

Eighty one major depressive disorder (MDD) patients (40 in TRD, 41 in nTRD) and 40 healthy controls (HCs) were enrolled for the functional magnetic resonance imaging (fMRI) scans. A seed-based functional connectivity (FC) method was used to investigate the brain network abnormalities of default mode network (DMN), affective network (AN), salience network (SN) and cognitive control network (CCN) for the MDD. Finally, the correlation was analyzed between the abnormal FCs and 17-item Hamilton Rating Scale for Depression scale (HAMD-17) scores.

RESULTS

Compared with the HC group, the FCs in DMN, AN, SN, CCN were altered in both the TRD and nTRD groups. Compared with the nTRD group, FC alterations in the AN and CCN were more abnormal in the TRD group, and the FC alterations were generally decreased at the SN in the TRD group. In addition, the FC values of right dorsolateral prefrontal cortices and left caudate nucleus in the TRD group and the FC values of right subgenual anterior cingulate cortex and left middle temporal gyrus in the nTRD group were positively correlated with HAMD-17 scale scores.

CONCLUSIONS

Abnormal FCs are present in four brain networks (DMN, AN, SN, CCN) in both the TRD and nTRD groups. Except of DMN, FCs in AN, SN and CCN maybe underlay the neurobiological mechanism in differentiating TRD from nTRD.

Altered functional connectivity in first-episode and recurrent depression: A resting-state functional magnetic resonance imaging study

Background

Functional magnetic resonance imaging (fMRI) studies examining differences in the activity of brain networks between the first depressive episode (FDE) and recurrent depressive episode (RDE) are limited. The current study observed and compared the altered functional connectivity (FC) characteristics in the default mode network (DMN), cognitive control network (CCN), and affective network (AN) between the RDE and FDE. In addition, we further investigated the correlation between abnormal FC and clinical symptoms.

Methods

We recruited 32 patients with the RDE, 31 patients with the FDE, and 30 healthy controls (HCs). All subjects underwent resting-state fMRI. The seed-based FC method was used to analyze the abnormal brain networks in the DMN, CCN, and AN among the three groups and further explore the correlation between abnormal FC and clinical symptoms.

Results

One-way analysis of variance showed significant differences the FC in the DMN, CCN, and AN among the three groups in the frontal, parietal, temporal, and precuneus lobes and cerebellum. Compared with the RDE group, the FDE group generally showed reduced FC in the DMN, CCN, and AN. Compared with the HC group, the FDE group showed reduced FC in the DMN, CCN, and AN, while the RDE group showed reduced FC only in the DMN and AN. Moreover, the FC in the left posterior cingulate cortices and the right inferior temporal gyrus in the RDE group were positively correlated with the 17-item Hamilton Rating Scale for Depression (HAMD-17), and the FC in the left dorsolateral prefrontal cortices and the right precuneus in the FDE group were negatively correlated with the HAMD-17.

Conclusions

The RDE and FDE groups showed multiple abnormal brain networks. However, the alterations of abnormal FC were more extensive and intensive in the FDE group.

A comparative study of amplitude of low-frequence fluctuation of resting-state fMRI between the younger and older treatment-resistant depression in adults

Background

Treatment-resistant depression (TRD) may have different physiopathological neuromechanism in different age groups. This study used the amplitude of low frequency fluctuations (ALFF) to initially compare abnormalities in local functional brain activity in younger and older patients with TRD.

Materials and methods

A total of 21 older TRD patients, 19 younger TRD, 19 older healthy controls (HCs), and 19 younger HCs underwent resting-state functional MRI scans, and the images were analyzed using the ALFF and further analyzed for correlation between abnormal brain regions and clinical symptoms in TRD patients of different age groups.

Results

Compared with the older TRD, the younger TRD group had increased ALFF in the left middle frontal gyrus and decreased ALFF in the left caudate nucleus. Compared with the matched HC group, ALFF was increased in the right middle temporal gyrus and left pallidum in the older TRD group, whereas no significant differences were found in the younger TRD group. In addition, ALFF values in the left middle frontal gyrus in the younger TRD group and in the right middle temporal gyrus in the older TRD were both positively correlated with the 17-item Hamilton Rating Scale for Depression score.

Conclusion

Different neuropathological mechanisms may exist in TRD patients of different ages, especially in the left middle frontal gyrus and left caudate nucleus. This study is beneficial in providing potential key targets for the clinical management of TRD patients of different ages.

Precuneus connectivity and symptom severity in chronic depression✰

Although abnormal resting state connectivity within several brain networks has been repeatedly reported in depression, little is known about connectivity in patients with early onset chronic depression. We compared resting state connectivity in a homogenous sample of 32 unmedicated patients with early onset chronic depression and 40 healthy control participants in a seed-to-voxel-analysis. According to previous meta-analyses on resting state connectivity in depression, 12 regions implicated in default mode, limbic, frontoparietal and ventral attention networks were chosen as seeds. We also investigated associations between connectivity values and severity of depression. Patients with chronic depression exhibited stronger connectivity between precuneus and right pre-supplementary motor area than healthy control participants, possibly reflecting aberrant information processing and emotion regulation deficits in depression. Higher depression severity scores (Hamilton Rating Scale for Depression) were strongly and selectively associated with weaker connectivity between the precuneus and the subcallosal anterior cingulate. Our findings correspond to results obtained in studies including both episodic and chronic depression. This suggests that there may be no strong differences between subtypes of depression regarding the seeds analyzed here. To further clarify this issue, future studies should directly compare patients with different courses of depression.

Immediate Modulation of Transcutaneous Auricular Vagus Nerve Stimulation in Patients With Treatment-Resistant Depression: A Resting-State Functional Magnetic Resonance Imaging Study

BACKGROUND

Previous studies found that transcutaneous auricular vagus nerve stimulation (taVNS) was clinically effective in treating a case of treatment-resistant depression (TRD). However, the brain neural mechanisms underlying the immediate effects of taVNS treatment for TRD have not been elucidated.

MATERIALS AND METHODS

Differences in the amplitude of low-frequency fluctuations (ALFF) between TRD and healthy control (HC) groups were observed. The TRD group was treated with taVNS for 30 min, and changes in ALFF in the TRD group before and after immediate treatment were observed. The ALFF brain regions altered by taVNS induction were used as regions of interest to analyze whole-brain functional connectivity (FC) changes in the TRD group.

RESULTS

A total of 44 TRD patients and 44 HCs completed the study and were included in the data analysis. Compared with the HC group, the TRD group had increased ALFF in the left orbital area of the middle frontal gyrus. After taVNS treatment, ALFF in the left orbital area of the middle frontal gyrus and right middle frontal gyrus decreased in the TRD group, while ALFF in the right orbital area of the superior frontal gyrus increased. The FC in the left orbital area of the middle frontal gyrus with left middle frontal gyrus and the right inferior occipital gyrus was significantly increased.

CONCLUSION

Transcutaneous auricular vagus nerve stimulation demonstrates immediate modulation of functional activity in the emotional network, cognitive control network, and visual processing cortex, and may be a potential brain imaging biomarker for the treatment of TRD.