DOI: 10.1002/alz.083090 ISSN: 1552-5260

Amygdala volume as neuroimaging marker of limbic‐predominant age‐related TDP‐43 encephalopathy (LATE) in neurodegenerative disease: a post‐mortem MRI and pathology study

Alex Wesseling, Ismael Luis Calandri, Rik Ossenkoppele, Wilma D.J. van de Berg, Annemieke J.M. Rozemuller, Yolande A.L. Pijnenburg, Jeroen Hoozemans, Laura E. Jonkman
  • Psychiatry and Mental health
  • Cellular and Molecular Neuroscience
  • Geriatrics and Gerontology
  • Neurology (clinical)
  • Developmental Neuroscience
  • Health Policy
  • Epidemiology



Limbic‐predominant age‐related TDP‐43 encephalopathy (LATE) is characterised by hyper‐phosphorylated TAR DNA‐binding protein 43 (TDP‐43) aggregates within the amygdala and (para)hippocampal regions. LATE neuropathological change can occur in patients with neurodegenerative disorders including Alzheimer’s (AD) and Parkinson’s disease (PD), as well as in healthy controls. LATE is associated with lower cognitive performance in AD patients. No neuroimaging marker for LATE is currently available, but identification would be of importance for accurate disease prognosis and as potential exclusion criteria for e.g. AD clinical trials. The current study aims to assess whether MRI‐measured amygdala and hippocampal volume can differentiate LATE status in neuropathological‐confirmed AD and PD brain donors.


Post‐mortem in‐situ 3DT1 3T‐MRI data were collected for 15 cases (9 AD and 6 PD) with post‐mortem confirmed LATE neuropathological change, as well as 30 non‐LATE (18 AD and 12 PD) donors. Each donor with LATE was matched with two non‐LATE donors on age, sex and neuropathological (i.e. Thal and Braak) staging. Neuropathological assessment was done on tissue of the right hemisphere. Amygdala volume, including amygdala subfield volumes, as well as hippocampal volume, were calculated using Freesurfer (version 7.1.1) and corrected for total intracranial volume. Group differences were assessed with univariate analyses using post‐mortem delay as covariate, subfield analysis were Bonferroni corrected.


Brain donors with LATE neuropathological change showed significantly lower right amygdala (p = .002) and right hippocampal (p = .007) volumes (figure 1). Amygdala subfield volumes were available for 13 LATE and 27 non‐LATE cases. When investigating the amygdala subfields, we found significant LATE compared to non‐LATE differences within the anterior amygdaloid area (p = 0.005) and the lateral nucleus (p = 0.002) (figure 2). In AD (n = 27), LATE cases showed significantly lower right amygdala (p = .023) and right hippocampal (p = .003) volumes compared to non‐LATE cases. In PD (n = 18), only the volume of the right amygdala was significantly lower in LATE cases compared to non‐LATE cases (p = .043) (figure 3).


MRI‐measured amygdala volume, especially of the anterior and lateral nucleus, can differentiate LATE neuropathological change, and may therefore serve as a promising neuroimaging biomarker in neurodegenerative disease.

More from our Archive