Why Subtypes of Depression Disagree
Poster No:
419
Submission Type:
Abstract Submission
Authors:
Kayla Hannon1, Setthanan Jarukasemkit2, Luca Balogh2, Fyzeen Ahmad2, Petra Lenzini2, Aristeidis Sotiras3, Janine Bijsterbosch2
Institutions:
1Washington University in St Louis, St. Louis, MO, 2Washington University in St Louis, St Louis, MO, 3Washington University in St. Louis, Saint Louis, MO
First Author:
Co-Author(s):
Introduction:
Despite substantial efforts to identify depression subtypes, work has failed to converge on a consensus. However, it is unclear whether this lack of consensus is driven by differences in study cohorts, subtype approaches, or input features. Utilizing the large clinical and imaging data of the UK Biobank (UKB), we compared six data-driven depression subtyping approaches of previous work, informed by either symptom or neuroimaging data, within the same subject space. This allowed us to directly compare the findings of each approach to determine what factors drive inconsistency in results, providing guidelines for future work towards convergent subtype solutions.
Methods:
For this registered report study, depressed subjects (N=2,276) were identified using probable MDD moderate or severe status (Smith et al 2013). The six approaches implemented are described in Fig. 1. Two approaches leveraged symptom data (Maglanoc et al, 2019; Lamers et al, 2010), two approaches leveraged structural MRI data (Yang et al, 2021; Wen et al, 2022), and the final two approaches leveraged resting state functional MRI data (Price et al, 2017; Drysdale et al, 2017). We evaluated the approaches on four criteria: 1) the subject agreement between approaches measured using Adjusted Rand Index (ARI); 2) the subject agreement (ARI) between approaches when using matched input features; 3) the sensitivity of each approach's subtype solutions across phenotypes related to mental health, body health, resting state networks' amplitude and connectivity, cortical thickness and fractional anisotropy; 4) the stability of each approach across 100 bootstraps of 80% data, quantified by calculating the ARI between the subtype solution of each bootstrap to the true subtype solution of that approach.
·Figure 1 (Methods)
Results:
We found virtually no subject agreement between approaches (ARI mean= 0.006, median = 0.03, mode = 0.00), even between approaches based on the same domain (clinical/ structural MRI/ resting state fMRI). The similarity between approaches increased when using matched input features (ARI mean=0.23), suggesting that the choice of input features is a critical driver of resulting depression subtypes, even within general feature domains. Furthermore, each approach driven by its original input features showed sensitivity to multiple phenotypes (Fig. 2A) and was internally relatively stable across bootstraps (Fig. 2B). The approaches were more sensitive to phenotypes in their own domain, indicating each approach is subtyping on different sources of heterogeneity.
·Figure 2 (Results). Sorry for the poor resolution, I couldn't figure out the magic recipe to make the 1000 pixels readable
Conclusions:
This study identified important differences between subtyping approaches in depression and determined key drivers of subtypes. Despite disagreement across approaches, each approach was internally stable and sensitive. This explains why previous subtyping studies performed with high degrees of rigor (showing stability and sensitivity) have not resulted in convergent subtypes. We identified that even within the same domain, differences in the specific input features led to near complete disagreement in subtype solutions. Input features were a stronger driver of subtype results than the computational approach, because substantially greater agreement was achieved with matched input features. Based on these results, we strongly recommend future work incorporates assessments of subtype agreement across input feature choices and includes rigorous comparisons to other published subtypes.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Modeling and Analysis Methods:
Classification and Predictive Modeling 2
Novel Imaging Acquisition Methods:
Anatomical MRI
BOLD fMRI
Keywords:
FUNCTIONAL MRI
Machine Learning
Psychiatric Disorders
STRUCTURAL MRI
WHITE MATTER IMAGING - DTI, HARDI, DSI, ETC
Other - Depression
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Lamers, et al. (2010). “Identifying Depressive Subtypes in a Large Cohort Study: Results From the Netherlands Study of Depression and Anxiety (NESDA).” The Journal of Clinical Psychiatry 71 (12): 1582–89. https://doi.org/10.4088/JCP.09m05398blu.
Maglanoc, Luigi A., et al. (2019). “Data-Driven Clustering Reveals a Link Between Symptoms and Functional Brain Connectivity in Depression.” Biological Psychiatry: Cognitive Neuroscience and Neuroimaging 4 (1): 16–26. https://doi.org/10.1016/j.bpsc.2018.05.005.
Price, Rebecca B, et al. (2017). “Data-Driven Subgroups in Depression Derived from Directed Functional Connectivity Paths at Rest.” Neuropsychopharmacology 42 (13): 2623–32. https://doi.org/10.1038/npp.2017.97.
Smith, Daniel J., et al. (2013). “Prevalence and Characteristics of Probable Major Depression and Bipolar Disorder within UK Biobank: Cross-Sectional Study of 172,751 Participants.” PloS One 8 (11): e75362. https://doi.org/10.1371/journal.pone.0075362.
Wen, J., et al. (2022). Characterizing Heterogeneity in Neuroimaging, Cognition, Clinical Symptoms, and Genetics Among Patients With Late-Life Depression. JAMA Psychiatry, 79(5), 464–474. https://doi.org/10.1001/jamapsychiatry.2022.0020
Yang, X., et al. (2021). Identifying Subgroups of Major Depressive Disorder Using Brain Structural Covariance Networks and Mapping of Associated Clinical and Cognitive Variables. Biological psychiatry Global Open Science, 1(2), 135–145. https://doi.org/10.1016/j.bps