Diverse tau pathologies in late-life mood disorders revealed by PET and autopsy assays

Abstract

Introduction: Late-life mood disorders (LLMDs) may represent prodromal manifestations of neurodegenerative dementia; however, the neuropathological basis of LLMDs, including depression and bipolar disorder, remains unclear. We aimed to investigate the involvement of Alzheimer's disease (AD) and non-AD tau pathologies in LLMD participants.

Methods: Fifty-two LLMD participants and 47 age- and sex-matched healthy controls (HCs) underwent tau and amyloid beta (Aβ) positron emission tomography (PET) imaging using ¹⁸F-florzolotau and ¹¹C-Pittsburgh compound B. Additionally, we conducted a clinicopathological correlation analysis in 208 autopsy cases, including various neurodegenerative diseases.

Results: LLMD participants were more likely to be tau PET and Aβ PET positive than HCs. The PET results were supported by the post mortem results that showed a higher likelihood of diverse tauopathies in patients with late-life mania or depression than those without.

Discussion: Our PET and autopsy assays suggest that AD and diverse non-AD tau pathologies might underlie the neuropathological basis of some LLMD cases.

Highlights: Late-life mood disorders (LLMDs) may represent prodromal states of dementia. The neuropathological basis of LLMDs remains unclear. LLMD subjects were highly likely to be tau positron emission tomography (PET) positive. Brain bank data supported our PET results. Alzheimer's disease (AD) and diverse non-AD tau pathologies can contribute to LLMDs.

Keywords: late‐life bipolar disorder; late‐life depression; late‐life mood disorder; positron emission tomography; post mortem study; prodromal states of dementia; tauopathy.

FIGURE 1

The proportion of Aβ and tau deposition assessed by PET imaging and brain bank survey in participants with LLMD and controls. A, Parametric ¹⁸F‐florzolotau‐PET SUVR images were spatially normalized to the MNI152 standard space based on the parameters for the transformation of individual MR images into templates using statistical parametric mapping. ¹⁸F‐florzolotau PET positivity was determined using a voxel‐based comparison between each participant and HCs. B, Of the 208 autopsies performed during the evaluation period, 21 patients were diagnosed with mania or depression after the age of 40 years based on evaluation of clinical records. Neuropathological results were compared between post mortem cases with (n = 21) and without late‐life mania or depression (n = 170). C, Logistic regression analyses revealed that patients with LLMD were more likely to be tau PET positive in the gray matter and Aβ PET positive compared to the HCs with sex, age, and MMSE scores as independent variables. D, The prevalence of tauopathies is significantly higher in patients with late‐life mania or depression than in those without. Aβ, DEP, depression; HC, healthy control; LLBD, late‐life bipolar disorder; LLD, late‐life depression; LLMD, late‐life mood disorder; MMSE, Mini‐Mental State Examination; MNI, Montreal Neurological Institute; MR, magnetic resonance; NCNP, National Center of Neurology and Psychiatry; PET, positron emission tomography; SUVR, standardized uptake value ratio; T1, T1‐weighted image; y.o., years old.

FIGURE 2

Representative ¹⁸F‐florzolotau PET parametric SUVR images. A, Representative ¹⁸F‐florzolotau SUVR images of ¹⁸F‐florzolotau–positive cases. White arrows, ¹⁸F‐florzolotau retention identified by voxel‐based analysis; orange arrows, off‐target binding to the choroid plexus. B, The heatmap presents ¹⁸F‐florzolotau retention patterns based on Aβ PET positivity and voxel‐based comparisons of tau PET performed separately for brain regions. Purple: Aβ PET ‐positive; pink: tau PET positive; gray: Aβ PET negative and tau PET negative. Aβ, amyloid beta; AD, Alzheimer's disease; BG, basal ganglia; HC, healthy control; LLBD, late‐life bipolar disorder; LLD, late‐life depression; PET, positron emission tomography; SUVR, standardized uptake value ratio.

FIGURE 3

Comparisons of regional ¹⁸F‐florzolotau SUVRs between LLMD participants with and without psychosis in Aβ PET–negative cases. In the Aβ PET–negative cases, ¹⁸F‐florzolotau SUVRs were significantly higher in patients with LLMD with psychosis than in those without psychosis in the frontal cortex and striatum. The red rule within the scatterplots indicates the mean. *indicates statistical significance. Aβ, amyloid beta; HC, healthy control; LLMD, late‐life mood disorder; SUVR, standardized uptake value ratio.

FIGURE 4

Symptom trajectory of autopsy cases and double staining with florzolotau and phospho‐tau antibodies. A, Clinical trajectory of autopsy cases. The clinical trajectory of 19 autopsy cases with neurodegenerative diseases who developed mania or depression preceding motor or cognitive symptoms after the age of 40 years is illustrated. Orange, purple, and blue show the onset of documented mood disorder, cognitive deterioration, and motor symptoms, respectively. We determined the onset of each symptom based on clinical records. B, Double staining with florzolotau and phospho‐tau antibodies in post mortem brain sections of diverse tauopathies. Double staining of tau aggregates in the parahippocampal gyrus of a patient with AD, in the parietal cortex of a patient with CBD, in the globus pallidus of a patient with PSP, in the amygdala of a patient with AGD, and in the frontal cortex of a patient with PiD with 25 mM non‐radiolabelled florzolotau and AT8. NFTs in AD, astrocytic plaques in CBD, globose NFTs in PSP, ballooned neurones in AGD, and Pick bodies in PiD brain sections were clearly labelled with florzolotau. AD, Alzheimer's disease; AGD, argyrophilic grain disease; CBD, corticobasal degeneration; DLB, dementia with Lewy bodies; MSA, multiple system atrophy; NFT, neurofibrillary tangle; PART, primary age‐related tauopathy; PD, Parkinson's disease; PiD, Pick's disease; PSP, progressive supranuclear palsy.