When words first fail: Predicting the emergence of primary progressive aphasia variants from unclassifiable anomic...

The majority of patients with primary progressive aphasia (PPA) can be distinguished into one of three variants: semantic, non-fluent/agrammatic, or logopenic. However, many do not meet criteria for any one variant.

To identify aspects of cognitive-linguistic performance that yield an early unclassifiable PPA designation that predicted the later emergence of a given variant.

Of 256 individuals with PPA evaluated, 19 initially were unclassifiable and later met criteria for a variant. Receiver operating characteristic curves were used to evaluate the binary ability of a given task to predict eventual classification as a given variant. Tasks with a high area under the curve were examined using regression analyses to determine their ability to predict variant.

High mean predictive value was observed for multiple naming assessments targeting nouns and verbs. The Boston Naming Test (BNT) was the only test that, in isolation, resulted in a significant model and high classification accuracy.

Although naming impairment is common across PPA variants, very low initial BNT scores emerged as a uniquely accurate basis for predicting eventual semantic variant, and normal BNT scores predicted eventual nonfluent/agrammatic variant. High performance on picture-verb verification was useful in identifying future lvPPA.