The Relationship Between Polygenic Risk of Depression and Regional Brain Atrophy in Older Adults.

BACKGROUND: Depression has been identified as a risk factor for the development of dementia, but the relationship between the two conditions is complex. If the mechanism by which depression increased the risk of dementia was better understood then new treatment targets could be identified. Previous studies have found evidence of reduced hippocampal volume in depressed patients, but because depression may also be an early sign of dementia, these studies may have inadvertently included individuals in the early stages of dementia. We used genetic risk of depression as our exposure and examined whether older adults with the highest genetic risk of depression had reduced regional brain volumes or greater longitudinal regional atrophy. METHODS: We used clinical, genetic and neuroimaging data from the ROSMAP and ADNI studies, which are both older adult focused longitudinal cohort studies with derived regional brain volumes a subset of available from MRI scans. Polygenic scores for depression were generated in PLINK from genotyped data imputed using 1000 genomes Phase 3 v5 using the clumping and thresholding method, based on the Howard et al., 2019 GWAS of depression. In ADNI n = 473 at baseline and 342 at 12 months, in ROSMAP n = 187 at baseline and n = 113 at 12 months. In a secondary analysis we excluded those with dementia at baseline. Our primary outcome was the effect of PRS-D on hippocampal volume in both studies using linear regression. The results from the 2 studies were then meta-analysed. RESULTS: We did not find any evidence of an effect of polygenic risk for depression on regional brain volume changes at baseline or after 12 months, either in ROSMAP or ADNI individually or in the meta-analysis. This finding was not affected by excluding individuals with evidence of dementia at baseline in ADNI or in ROSMAP those with AD at post-mortem examination. CONCLUSIONS: We found no strong evidence of an effect of polygenic risk of depression on regional brain volumes in older adults either at baseline or after 12 months. Our work was limited to genetic risk of depression and further research is required to investigate the complex relationship between depression and dementia.