A latent clinical-anatomical dimension relating metabolic syndrome to brain structure and cognition

Reviewer #1 (Public Review):

Summary:
In their study, Petersen et al. investigated the relationship between parameters of metabolic syndrome (MetS) and cortical thickness using partial least-squares correlation analysis (PLS) and performed subsequently a group comparison (sensitivity analysis). To do this, they utilized data from two large-scale population-based cohorts: the UK BioBank (UKB) and the Hamburg City Health Study (HCHS). They identified a latent variable that explained 77% of the shared variance, driven by several measures related to MetS, with obesity-related measures having the strongest contribution. Their results highlighted that higher cortical thickness in the orbitofrontal, lateral prefrontal, insular, anterior cingulate, and temporal areas is associated with lower MetS metric severity. Conversely, the opposite pattern was observed in the superior frontal, parietal, and occipital regions. A similar pattern was then observed in the sensitivity analysis when comparing two groups (MetS vs. matched controls) separately. They then mapped local cellular and network topological attributes to the observed cortical changes associated with MetS. This was achieved using cell-type-specific gene expressions from the Allen Human Brain Atlas and the group consensus functional and structural connectomes of the Human Connectome Project (HCP), respectively. This contextualization analysis allowed them to identify potential cellular contributions in these structures driven by endothelial cells, microglial cells, and excitatory neurons. It also indicated functional and structural interconnectedness of areas experiencing similar MetS effects.

Strengths:
The effects of metabolic syndrome on the brain are still incompletely understood, and such multi-scale analyses are important for the field. Despite the study's sole 'correlation-based' nature, it yields valuable results, including several scales of brain parameters (cortical thickness, cellular, and network-based). The results are robust and benefit from two 'large-scale' datasets, resulting in highly powered statistics.

Weaknesses:
However, some concerns arise regarding certain interpretations and claims made by the authors. In particular, it is not entirely convincing that the authors' results are relevant for studying insulin resistance as a clinical measure of MetS. This is due to the use of non-fasting glycemia as a metric, which the authors claim represents insulin resistance. While non-fasting blood glucose is a potential, albeit poor, indicator of insulin resistance, claiming a direct correlation between insulin resistance and cortical thickness does not seem entirely convincing. By doing so, the authors suggest that insulin resistance might have a weak contribution to cortical thickness abnormalities, with a rather low 'loading' of glycemia compared to the other MetS metrics, although this cannot be conclusively determined from these results.

Reviewer #2 (Public Review):

Summary:
In this manuscript, Petersen et al. aimed for a comprehensive assessment of the relationship between cardiometabolic risk factors and cortical thickness. They found that a latent variable reflecting higher obesity, hypertension, LDL cholesterol, triglyerides, glucose, and lower HDL cholesterol was associated with lower cortical thickness in orbitofrontal, lateral prefrontal, insular, anterior cingulate, and temporal areas. In sensitivity analyses, they showed that this pattern replicated across cohorts and was also consistent with a clinical definition of the metabolic syndrome.

Further, when including cognition in the multivariate analysis, the pattern remained unchanged and indicated that cardiometabolic risk factors were associated with worse cognitive performance across different tests. The authors investigated the cell types implicated in the regions associated with cardiometabolic risk using the Allen brain atlas and found that the density of excitatory neurons type 8, endothelial cells, and microglia reliably co-located with the pattern of cortical thickness. Furthermore, they showed that cortical regions more strongly associated with MetS were more closely structurally & functionally connected than others.

Strengths:
This study performed a comprehensive assessment of the combined association of cardiometabolic risk factors and brain structure and investigated micro- and macroscopic underpinnings. A major strength of the study is the methodological approach of Partial Least Squares which allows the authors to not single out risk factors but to take them into account simultaneously. The large sample size from two cohorts allowed for different sensitivity analyses and convincing evidence for the stability of the first latent variable. The authors demonstrated that the component was also reliably related to cognitive performance, replicating multiple previous studies that evidenced associations of different components of the MetS with worse cognitive performance.

The novel contribution of the study lies in the virtual histology and brain topology investigation of the cortical pattern related to MetS. The virtual histology provided clear evidence of the co-localization of endothelial, glial, and excitatory neuronal cells with the regions of MetS-associated cortical thinning while the brain topology analysis highlighted the disproportionate structural and functional connectivity between associated regions. This analysis provides insights into the role of inflammatory processes and the intricate link between gray matter morphology and microvasculature, both locally and in relation to long-range connectivity. This information is valuable to inform future mechanistic studies.

Weaknesses:
The study is exclusively cross-sectional which does not allow to the authors to disentangle causes from consequences. While studies indicate that most of the differences seen in middle age are probably consequences of the MetS on the vasculature, blood-brain barrier, or inflammatory processes, differences in cortical morphology might also represent a risk factor for weight gain.

Another limitation is the omission of subcortical structures and the cerebellum which might have provided additional information on the pattern of GM differences associated with MetS.

The study is exploratory in nature and for the contextualization analyses it is difficult to judge whether those were selected from a larger pool of analyses. The analysis approach taken to relate the cardiometabolic risk, brain structure, and cognition does not allow the reader to determine whether brain regions most strongly related to the MetS are the ones also most strongly associated with cognitive performance. The cortical pattern arising from the models including cognition is not thoroughly compared to the MetS-only pattern and therefore, it is difficult to estimate to which extent the MetS-related cortical patterns explain variance in cognitive performance.

Reviewer #3 (Public Review):

Summary:
This study investigates the continuous effect of MetS components - namely, obesity, arterial hypertension, dyslipidemia, and insulin resistance - on cortical thickness. It also examines the spatial correlations between MetS effects on cortical thickness with brain cellular and network topological attributes. Additionally, the authors attempt to explore the complex interplay among MetS, cognitive function, and cortical thickness.

The results reveal a latent relationship between MetS and cortical thickness based on a clinical-anatomical dimension. Furthermore, the effect of MetS on cortical thickness is linked to local cell types and network topological attributes. These findings suggest that the authors achieved most, though not all, of their research objectives.

The conclusions are mostly well supported by data and results. However, the use of "was governed by" in the conclusion section suggests a causal relationship. This phrasing is inappropriate given that the study primarily employs correlational analyses.

Strengths
The study presents several strengths:

This study undertakes a comprehensive assessment encompassing the full range of MetS components, such as obesity or arterial hypertension, rather than adopting a case-control study approach (categorizing participants into MetS or non-MetS groups) as seen in some previous research. Utilizing Partial Least Squares (PLS) for correlational analysis effectively addresses issues of multicollinearity (or high covariance among MetS components) and explores the relationship between MetS and brain morphology.

The study leverages two datasets, examining a large sample size of 40,087 individuals. This substantial sample potentially aids in identifying nuanced and underexplored brain anomalies. By incorporating high-quality MRI images, standardized data, and statistical analysis procedures, as well as sensitivity analyses, the results gain robustness, which addresses the limitations of small samples and low reproducibility.

In the context of MetS, this research uniquely employs the concept of imaging transcriptomics, i.e. virtual histology analysis. This approach allows the study to explore intricate relationships between cellular types and cortical thickness anomalies.

Weaknesses
While this work has foundational strengths, the analyses and data seem inadequate to fully support the key claim and analysis. In particular:

After a thorough review of the methods and results sections, I found no direct or strong evidence supporting the authors' claim that the identified latent variables were related to more severe MetS to worse cognitive performance. While a sub-group comparison was conducted, it did not adequately account for confounding factors such as educational level. Additionally, the strength of evidence from such a sub-group comparison is substantially weaker than that from randomized controlled trials or longitudinal cohort studies. Therefore, it is inaccurate for the authors to assert a direct relationship between MetS and cognitive function based on the presented data. A more appropriate research design or data analysis approach, such as mediation analysis, can be employed to address this issue.

The use of the imaging transcriptomics pipeline (virtual histology analysis) to explore the microscale associations with MetS effects on the brain is commendable and has shown promising results. Nevertheless, variations in gene sets may introduce a degree of heterogeneity in the results (Seidlitz, et al., 2020; Martins et al., 2021). Consequently, further validation or exploratory analyses utilizing different gene sets can yield more compelling results and conclusions.

Author Response

We highly appreciate the constructive feedback provided by the reviewers, which we believe will greatly improve the quality of our work. We were encouraged to see that our manuscript was considered to be “important”, of “great interest” as well as to “yield valuable results”.

We also highly appreciate the overall positive eLife assessment. However, we were surprised to read that our “results range from solid from inadequate”. This especially applies given the positive and engaging nature of the reviews which seem to mainly concern the results interpretation being “inadequate” rather than the results themselves. Hence, we kindly request a reconsideration of this aspect of the assessment.

Moreover, there is one Reviewer comment we would like to address directly. Reviewer #3 pointed out that “this study did not conduct a direct association analysis between MetS and cognitive levels without considering subgroup comparisons.” and that “After a thor-ough review of the methods and results sections” she/he “found no direct or strong evidence supporting the authors' claim that the identified latent variables were related to more severe MetS to worse cognitive performance. While a sub-group comparison was conducted, it did not adequately account for confounding factors such as educational level.”.

We appreciate the observations of Reviewer #3 regarding the absence of a direct association analysis between Metabolic Syndrome (MetS) and cognitive levels without subgroup comparisons, and the lack of evidence linking latent variables to MetS severity and cognitive performance. Our apologies for any confusion caused by unclear presentation. Our study incorporated association analyses between MetS, brain structure, and cognition using MetS components, regional cortical thickness, and cognitive performance data in a PLS. These analyses were separately performed on the UK Biobank and HCHS datasets, due to their distinct cognitive assessments. We adjusted for age, sex, and education in the subgroup analyses by removing their effects from the input variables. The primary latent variables demonstrated significant associations with MetS components, cortical thickness, and cognitive scores, indicating that higher obesity, blood pressure, lipidemia, and glycemia levels correlate with lower cognitive performance. These relationships are detailed in supplementary figures S15b and S16b, with negligible loadings for age, sex, and education, confirming effective deconfounding. We acknowledge the reviewer's constructive feedback and will enhance the clarity of the Methods and Results sections, including conducting a mediation analysis.

Furthermore, we strive to incorporate the Reviewers’ other suggestions into the analysis. The revision will include major changes to the manuscript.

In response to Reviewer #1:

• We will revise considering non-fasting plasma glucose as a surrogate marker of insuline resistance.

• We will report Field IDs of the used UK Biobank variables.

• We aim to moderate causal interpretations and reword the indicated passages for clarity.

In response to Reviewer #2:

• We will reconsider claims of binarizing vascular dementia and Alzheimer’s dementia pathophysiology.

• We will further explore the cell type associations of the other latent variables.

• We will expand the discussion regarding conclusions from our results and the future outlook.

In response to Reviewer #3.

• We will add an additional flowchart to detail the virtual histology analysis.

• We will add a discussion of the second latent variable.

• We will conduct a mediation analysis to statistically assess the mediation effect of brain structure on the relationship between MetS and cognitive performance.

We are convinced that with these revisions, our manuscript will align even more closely with the high standards of eLife and make a strong contribution to its distinguished portfolio. We thank you for your consideration.