Utility of the Addenbrooke’s Cognitive Examination III online calculator to differentiate the primary progressive aphasia variants

Abstract The Addenbrooke’s Cognitive Examination III is a brief cognitive screening tool that is widely used for the detection and monitoring of dementia. Recent findings suggest that the three variants of primary progressive aphasia can be distinguished based on their distinct profiles on the five subdomain scores of this test. Here, we investigated the utility of the Addenbrooke’s Cognitive Examination III to differentiate the primary progressive aphasia variants based on their item-by-item performance profiles on this test. From these results, we created an interactive primary progressive aphasia Addenbrooke’s Cognitive Examination III calculator which predicts the variant based on a patient’s unique item-by-item profile. Twenty-eight logopenic variant, 25 non-fluent variant and 37 semantic variant primary progressive aphasia patients and 104 healthy controls completed the Addenbrooke’s Cognitive Examination III at first clinical presentation. Multinomial regression analyses were conducted to establish performance profiles among groups, and R Shiny from RStudio was used to create the interactive Addenbrooke’s Cognitive Examination III diagnostic calculator. To verify its accuracy, probability values of the regression model were derived based on a 5-fold cross-validation of cases. The calculator’s accuracy was then verified in an independent sample of 17 logopenic, 19 non-fluent and 13 semantic variant primary progressive aphasia patients and 68 Alzheimer’s disease patients who had completed the Addenbrooke’s Cognitive Examination III (or an older version of this test: Revised) and had in vivo amyloid-PET imaging and/or brain autopsy pathological confirmation. Cross-validation of cases in the calculator model revealed different rates of sensitivity in classifying variants: semantic = 100%, non-fluent = 80.6% and logopenic = 79.9%; healthy controls were distinguished from primary progressive aphasia patients with 100% sensitivity. Verification of in vivo amyloid and/or autopsy-confirmed patients showed that the calculator correctly classified 10/13 (77%) semantic variant, 3/19 (16%) non-fluent variant and 4/17 (24%) logopenic variant patients. Importantly, for patients who were not classified, diagnostic probability values mostly pointed toward the correct clinical diagnosis. Furthermore, misclassified diagnoses of the primary progressive aphasia cohort were rare (1/49; 2%). Although 22 of the 68 Alzheimer’s disease patients (32%) were misclassified with primary progressive aphasia, 19/22 were misclassified with the logopenic variant (i.e. falling within the same neuropathological entity). The Addenbrooke’s Cognitive Examination III primary progressive aphasia diagnostic calculator demonstrates sound accuracy in differentiating the variants based on an item-by-item Addenbrooke’s Cognitive Examination III profile. This calculator represents a new frontier in using data-driven approaches to differentiate the primary progressive aphasia variants.

The PPA groups showed distinct ACE-III profiles at the item-by-item level. Relative to controls, the PPA groups were impaired on most ACE-III items. Notable exceptions across PPA groups were, however, present. The sv-PPA group did not differ significantly from controls on all visuospatial items (i.e., loops, cube, clock, dot counting, fragmented letters), most repetition items (i.e., single words, 'glitters' sentence), three step commands, and on several attention items (orientation to time, threeword registration/learning, serial 7s). Relative to controls, the nfv-PPA group's performance was spared on aspects of visuospatial ability (i.e., clock drawing, fragmented letters), semantic association, and on free recall of a name and address. The lv-PPA group did not differ significantly from controls on basic visuoperceptual skills (i.e., dot counting, fragmented letters) and the semantic association item.
Direct comparisons of the PPA groups revealed cognitive profiles characteristic of each variant. Comparison of the two non-fluent variants (i.e., lv-PPA, nfv-PPA) revealed the lv-PPA group were disproportionately impaired on the name and address learning item and clock drawing, whereas the nfv-PPA group were disproportionately impaired on single word repetition and dot counting.
Relative to the non-fluent groups, the sv-PPA were more impaired on retrograde memory, confrontation naming, semantic knowledge (i.e., word-picture matching) and reading, and were additionally more impaired than the nfv-PPA group on orientation to location (i.e., geography), and aspects of memory (i.e., three-word recall, free memory recall of name and address). By contrast, the sv-PPA group performed better than the non-fluent groups on aspects of attention (i.e., orientation to time, serial 7s), repetition (i.e., single word, sentences) and visuospatial abilities (i.e., loops, cube); the sv-PPA group were also superior to the nfv-PPA group on letter fluency, writing and dot counting, and superior to lv-PPA on clock drawing. Language production is telegraphic, with evident oversimplification of sentence structure.

Naming No anomia
Nearly normal with just one or two naming errors or omissions, usually for uncommon items (e.g., tongs, rhinoceros, etc.). Frequent and marked selfcorrections, subtle delays in retrieving names, or naming doubtfully should be included in this level.
Although may correctly name some well-known items, there is consistent and unmistakable naming errors.
Just able to correctly name a few well-known items (less than a third of items), with marked delays in retrieving names.

Single word repetition Flawless repetition
Occasional (1)(2) and subtle errors mainly due to mispronunciation or word fragmentation, involving only longer, or more complex words.
Evident and consistent mistakes due to mispronunciation, omissions, substitutions, phoneme reversal or word fragmentation.
Frequent and evident errors. Only able to repeat the simpler-more-common words (e.g., banana, potato) or one to two syllables of the longer-morecomplex words.

Single word comprehension
Normal comprehension Low frequency words (e.g., ferocious, indigenous) are occasionally unknown. In general, most words are somehow correctly defined, pointed to, or related to their concept.
Marked impairment in defining or object-word matching, able to provide broad superordinate category (e.g., an animal), but not able to elaborate on details. Preservation of more familiar items.
Only a few highly familiar words might be correctly recognised, with consistent failures in pointing to, or matching well-known items, such as tiger or crocodile.

Sentence repetition Flawless repetition
Subtle mistakes due to word errors, self-corrections, or long latency. Although minimal omissions may be present, there is frequent repetition of t the main sentence's components.
Consistent errors (namely disorganisation, omissions, or mispronunciation) when the patient tries to repeat longer sequences (e.g., The Chinese fan…).
Unable to correctly repeat phrases with more than two short words.

Normal comprehension
Occasional or minimal mistakes. Longer latency in replying, asking for instructions to be repeated, and self-corrections should be considered.

Consistent mistakes involving complex instructions.
Marked and consistent mistakes in direct instructions (e.g., touch the pen then the toothbrush).  (16) 15.6 (0.6) 15.6 (0.7)   Table 7. Patients shaded in grey were deemed atypical due to either the neurologist's clinical opinion or atypical findings on PiB-PET or brain autopsy. ^ Patient 12 was included in the initial ACE-III evaluation and modelled data (i.e., Phase 1) but was also included in Phase 2

Language_3_Commands
Performance scores on the ACE-R three-stage command were used for the ACE-III three single-step item.

Language_Writing
The ACE-R writing item is out of /1 but the ACE-III writing item is out of /2. ACE-R performance scores were converted to equivalent ACE-III scores: If the patient scored 1/1 on the ACE-R, then their performance was converted to 2/2. If the patient scored 0/1 on the ACE-R, then their score on the ACE-III writing item remained as 0/2. Notes: * The first two naming items on the ACE-R ("pencil" and "watch") were replaced/superseded in the ACE-III with other highly familiar objects ("spoon" and "book"). All other naming items in the ACE-R and ACE-III are the same.