TAR-
DNA binding protein-43 (TDP-43)
proteinopathy is seen in multiple
brain diseases. A standardized terminology was recommended recently for common age-related
TDP-43 proteinopathy:
limbic-predominant, age-related TDP-43 encephalopathy (LATE) and the underlying neuropathological changes,
LATE-NC.
LATE-NC may be co-morbid with
Alzheimer's disease neuropathological changes (ADNC). However, there currently are ill-defined diagnostic classification issues among
LATE-NC, ADNC, and
frontotemporal lobar degeneration with TDP-43 (
FTLD-TDP). A practical challenge is that different autopsy cohorts are composed of disparate groups of research volunteers: hospital- and clinic-based cohorts are enriched for
FTLD-TDP cases, whereas community-based cohorts have more
LATE-NC cases. Neuropathological methods also differ across laboratories. Here, we combined both cases and neuropathologists' diagnoses from two research centres-University of Pennsylvania and University of Kentucky. The study was designed to compare neuropathological findings between
FTLD-TDP and pathologically severe
LATE-NC. First, cases were selected from the University of Pennsylvania with pathological diagnoses of either
FTLD-TDP (n = 33) or severe
LATE-NC (mostly stage 3) with co-morbid ADNC (n = 30). Sections from these University of Pennsylvania cases were cut from amygdala, anterior cingulate, superior/mid-temporal, and middle frontal gyrus. These sections were stained for phospho-TDP-43 immunohistochemically and evaluated independently by two University of Kentucky neuropathologists blinded to case data. A simple set of criteria hypothesized to differentiate
FTLD-TDP from
LATE-NC was generated based on density of TDP-43 immunoreactive neuronal cytoplasmic inclusions in the neocortical regions. Criteria-based sensitivity and specificity of differentiating severe
LATE-NC from
FTLD-TDP cases with blind evaluation was ∼90%. Another proposed neuropathological feature related to
TDP-43 proteinopathy in aged individuals is 'Alpha' versus 'Beta' in amygdala. Alpha and Beta status was diagnosed by neuropathologists from both universities (n = 5 raters). There was poor inter-rater reliability of Alpha/Beta classification (mean κ = 0.31). We next tested a separate cohort of cases from University of Kentucky with either
FTLD-TDP (n = 8) or with relatively 'pure' severe
LATE-NC (lacking intermediate or severe ADNC; n = 14). The simple criteria were applied by neuropathologists blinded to the prior diagnoses at University of Pennsylvania. Again, the criteria for differentiating
LATE-NC from
FTLD-TDP was effective, with sensitivity and specificity ∼90%. If more representative cases from each cohort (including less severe TDP-43
proteinopathy) had been included, the overall accuracy for identifying
LATE-NC was estimated at >98% for both cohorts. Also across both cohorts, cases with
FTLD-TDP died younger than those with
LATE-NC (P < 0.0001). We conclude that in most cases, severe
LATE-NC and
FTLD-TDP can be differentiated by applying simple neuropathological criteria.