Assessing the role of neuroimaging connectomes and neuropsychological profiles in predicting cognitive decline in late-life depression: A prospective study

Summary

Background: Dementia is a syndrome typically caused by progressive cognitive decline in neurodegenerative diseases resulting in impairment of activities of daily living. There is a substantial economic burden at family and community level. The Longitudinal Aging Study in India (LASI) estimated dementia prevalence for adults ages 60+ in India is 7.4%. A systematic review and meta-analysis estimated the pooled prevalence of depression among the elderly population in India to be approximately 34.4%. The 2020 Lancet commission on Dementia lists 12 modifiable risk factors that together account for 40% of dementias worldwide and theoretically they could be prevented or delayed. LLD accounts for 4% of the risk. Late-life depression is associated with cognitive impairments that may persist even after successful treatment of depressive symptoms, potentially serving as an early indicator or prodrome of dementia.  LLD is considered both a risk factor and prodrome of Dementia, but the link between LLD and Dementia has not been fully elucidated. Not all patients with LLD progress into Dementia. While there is a well-established association between late-life depression and increased dementia risk, the causal pathways remain unclear. It is not clear as to which clinical and neuropsychological profile of patients develop Dementia. Dementia and late-life depression LLD both involve significant alterations in brain networks, which are linked to their symptoms and progression. Neuroimaging offers several advantages as a biomarker for dementia. Human Connectome Project (HCP), Alzheimer’s Disease Neuroimaging Initiative (ADNI) and Alzheimer’s Disease Connectome Project (ADCP) are ongoing connectome based projects looking at neuroimaging as a biomarker for AD. To the best of our knowledge, non of the studies looked at comparing connectomes of LLD which is considered a risk factor/prodrome with that of Dementia. In this study we intend to fill this gap thereby facilitating early diagnosis and treatment initiation for Dementia.

Novelty: To build neuroimaging connectome datasets for Late Life Depression (LLD), Dementia and healthy controls for our population. Many studies have looked at neuropsychological profiles and single MRI sequence to explore the correlation between LLD and Dementia. This study looks at multimodal imaging which will allow us to build a predictive model for cognitive decline in LLD with more accuracy. To the best of our knowledge this is the first study from India exploring the connection between LLD and Dementia through multimodal imaging.

Methods and procedure: A prospective study of persons with LLD(n=35) with comparison groups of Dementia(n=35) and healthy controls(n=35). Connectomes are built using multimodal neuroimaging ( T1,T2,FLAIR,DWI,DTI,ASL and rsfMRI) sequences and neuropsychological assessments are carried out at 6 month and 12 month time points. A predictive model for cognitive decline in LLD will be built using the neuroimaging, neuropsychological and clinical data.

Outcome: Datasets for LLD, Dementia and healthy population. A predictive model that’ll inform us about which LLD patients are more likely to develop cognitive decline. This helps in early diagnosis and treatment planning Dementia. Having these datasets for our population will nurture future research in the field.

Keywords: Late Life Depression, Dementia, Connectome, Datasets, Predictive model

Problem Statement

LLD is considered both a risk factor and prodrome of Dementia, but the link between LLD and Dementia has not been fully elucidated. Not all patients with LLD progress into Dementia. It is not clear as to which clinical and neuropsychological profile of patients develop Dementia. Though many prospective studies have established the temporal association of LLD and Dementia, the known biomarkers didn’t yield conclusive evidence for the same. Neuroimaging findings are one of the biomarkers that will inform us about the risk of progression to Dementia. Till now many prospective studies have looked at individual sequences and techniques of neuroimaging to establish a link between LLD and Dementia and identified patterns similar to both the conditions. But there is lot of inconsistency due to multiple factors like symptom heterogeneity, lack of experimental rigor, individual marker/ imaging technique etc. Use of multimodal neuroimaging techniques, each having its own strengths yields more comprehensive understanding of brain abnormalities, pathology, and their relationship to clinical symptoms.

Hypothesis/ Research question: This study hypothesizes that Late life depression (LLD) and Dementia share more underlying structural and functional changes in the brain than previously understood. We intend to investigate this overlap by creating connectomes using advanced neuroimaging techniques and study specific patterns of connectome that will predict the cognitive decline in LLD patients.

The research question: How do neuroimaging connectomes in LDD and Dementia patients overlap and differ and can specific connectome patterns in LLD patients predict the extent of cognitive decline at one year follow-up?