Association of LATE‐NC with other neuropathologies and cognition in the oldest‐old: Evidence from The 90+ Study and NACC

Background

Limbic‐predominant age‐related TDP‐43 encephalopathy neuropathologic change (LATE‐NC) is a prevalent and important contributor to dementia. A previous study suggested that associations between LATE‐NC and other neuropathologies, namely Alzheimer’s disease neuropathologic change (ADNC) and Lewy bodies (LB), were attenuated in oldest‐old (≥90 years at death) compared to younger‐old individuals. However, this remains an open question. Our objectives were: Objective‐1) Examine the association of LATE‐NC with other neuropathologies and cognition in oldest‐old participants from The 90+ Study and National Alzheimer’s Coordinating Center (NACC). Objective‐2) Compare the associations of LATE‐NC with other pathologies and cognition in oldest‐old vs younger‐old NACC participants.

Method

We included participants from both The 90+ Study and NACC, excluding those with frontotemporal lobar degeneration and rare pathologies. LATE‐NC was defined as presence of TDP‐43 in amygdala, hippocampus, or neocortex. Neuropathologies of interest were dichotomized as follows: ADNC: intermediate/high; LB: limbic/neocortical; hippocampal sclerosis (HS): presence; arteriolosclerosis, atherosclerosis, and cerebral amyloid angiopathy (CAA): moderate/severe. We used diagnoses of dementia and clinical AD as measures of cognitive function. For Objective‐1 we performed logistic regressions of LATE‐NC with neuropathologies and cognitive measures in oldest‐old participants from The 90+ Study and NACC. For Objective‐2 we tested the interaction between LATE‐NC and age‐group (≥90 vs 65‐90) for associations with neuropathologies and cognitive measures in NACC. Regressions were adjusted for sex and education.

Result

Objective‐1): Both cohorts had similar prevalence of LATE‐NC in the oldest‐old (NACC, N = 393, 44% with LATE‐NC; 90+Study, N = 394, 37% with LATE‐NC, Table‐1). LATE‐NC was consistently associated with HS (P<0.001, 90+Study & NACC), ADNC (P = 0.008, 90+Study; P = 0.002, NACC), dementia (P<0.001, 90+Study & NACC), and clinical AD (P<0.001, 90+Study & NACC), but the association with arteriolosclerosis was specific to NACC (P<0.001; P = 0.7 for 90+Study, Figure‐1).

Objective‐2): NACC participant characteristics are shown in Table‐1. We found trends for the interaction between LATE‐NC and age‐group for ADNC (P = 0.055, stronger in younger‐old) and arteriolosclerosis (P = 0.051, stronger in oldest‐old, Figure‐2).

Conclusion

In the oldest‐old, associations of LATE‐NC with other neuropathologies and cognitive measures were consistent across two independent cohorts. The association of LATE‐NC with ADNC and arteriolosclerosis might be age dependent although further studies are needed.