Clinical recognition of frontotemporal dementia with right anterior temporal predominance: A multicenter retrospective cohort study

Abstract INTRODUCTION Although frontotemporal dementia (FTD) with right anterior temporal lobe (RATL) predominance has been recognized, a uniform description of the syndrome is still missing. This multicenter study aims to establish a cohesive clinical phenotype. METHODS Retrospective clinical data from 18 centers across 12 countries yielded 360 FTD patients with predominant RATL atrophy through initial neuroimaging assessments. RESULTS Common symptoms included mental rigidity/preoccupations (78%), disinhibition/socially inappropriate behavior (74%), naming/word‐finding difficulties (70%), memory deficits (67%), apathy (65%), loss of empathy (65%), and face‐recognition deficits (60%). Real‐life examples unveiled impairments regarding landmarks, smells, sounds, tastes, and bodily sensations (74%). Cognitive test scores indicated deficits in emotion, people, social interactions, and visual semantics however, lacked objective assessments for mental rigidity and preoccupations. DISCUSSION This study cumulates the largest RATL cohort unveiling unique RATL symptoms subdued in prior diagnostic guidelines. Our novel approach, combining real‐life examples with cognitive tests, offers clinicians a comprehensive toolkit for managing these patients. Highlights This project is the first international collaboration and largest reported cohort. Further efforts are warranted for precise nomenclature reflecting neural mechanisms. Our results will serve as a clinical guideline for early and accurate diagnoses.


Comparison of the centers (GALLAGHER M, HARES T, SPENCER J, BRADSHAW C, WEBB I. The Nominal Group
Technique: A Research Tool for General Practice?Fam Pract.1993 Mar 1;10(1): [76][77][78][79][80][81].All meetings were recorded and shared on safe platforms (SURFDrive).Generation of the symptom checklist, symptom definitions, interpretations, terminology, and how to abstract data from case notes were elucidated and a guidebook, open discussion chat (SURFDrive), and a dataset template were prepared based on the discussions.Following data collection, preliminary results such as sample characteristics, frequencies of the symptoms and cognitive test results were documented, and surveys were prepared to systematically collect the opinions of the collaborators.All results were shared and discussed in the round table meetings to (1) interpret and categorize the listed symptoms and (2) establish consensus on most frequent and distinctive symptoms and recommended terminology.
Patient selection: Among subjects registered with a clinical diagnosis of FTD and/or PPA those who demonstrated predominant RATL atrophy on the initial neuroimaging were selected.For identifying the RATL predominant atrophy, all FTD patients whose initial neuroimaging available were included.A guidebook was prepared for visual rating to score the atrophy.Visual atrophy rating was conducted by either an experienced radiologist or a dementia expert in each center by using our guidelines (see below).If the atrophy on the RATL was higher than the non-RATL areas (left temporal or frontal), this patient was considered as an rtvFTD case.Patients with frontal/left temporal atrophy  3 (max score=4) (even if they had predominant right temporal atrophy) were excluded to minimize gross effects of general neurodegeneration.
Data collection, harmonization of the data, and approach on missing data: The data were collected from the case notes retrospectively (End date: July 2022).A dataset template was prepared based on the published RATL degeneration/temporal variant FTD literature and round table discussions to fill in the pseudonymised data which comprises (1) demographic, (2) clinical data (including symptoms, family history, dementia severity, and cognitive tests and neuropsychiatric inventory (NPI) scores), (3) atrophy rating scores for each anatomical area, (4) biomarkers (CSF amyloid, tau, amyloid PET, neurofilament light chain (NfL) or other potential biomarkers for FTD-if available), (5) genetic mutations, (6) pathological information.A broad range of clinical symptoms (cognitive, behavioral, language, psychiatric, other) was coded as present, absent or not available at the initial and follow-up visits.If it is present, the collectors were asked to describe the symptom by copying the real-life examples from the case notes.
Frequencies were calculated based on the available data, excluding any missing data.No imputation technique was used to deal with the missing data since one of the goals was identifying the gaps as well.The number of available data for each symptom is displayed alongside the corresponding percentage in the results.Additionally, there was an "other" column for each section where the collectors were able to add more data/information that was not on the predefined list.All data in this section were checked by HU and YP to avoid duplications.If the details were already mentioned in other columns, they were kept as real-life examples.The rest was analyzed and similar symptom descriptions (ie: for landmarks: "topographagnosia", "problems with naming landmarks", "cannot recognize buildings", for taste: "gustatory agnosia", "cannot differentiate the taste", "flavor recognition problem" etc., ) were clustered to be discussed in the round table meetings.To provide safe, open access and transparent data, SURFdrive and CASTOR softwares were used and separate data sharing agreements were signed with each center to protect the rights of the research participants and data suppliers.and a z score lower than -1.65 was considered as impaired for each individual.Z scores couldn't be calculated for some tests that had small sample sizes and had no healthy control data.For interpretation of those tests, cut off scores were used.In the table, percentages of the patients whose z scores are lower than -1.65 and/or cut off value are displayed.

Guidebook for data collection
The excel file contains 4 pages.Page 1 is designed for initial features.Page 2 and 3 are for follow-up visits.Page 4 is for abbreviations and coding system.If your patient does not have a follow-up visit, please type not available (n.a.) into the follow-up pages.If your patient has more than 2 follow up visits please copy the template and create a new follow-up page.We prepared 2 sample cases as an example for you.

DEMOGRAPHICS
This heading comprises 5 subheadings 1. CASE: Data must be anonymized.Please do not share personal ID data.The case ID will be the combination of your city and the case number (ie: Montreal12)

CLINICAL SYMPTOMS
This heading comprises 5 subheadings.If the mentioned symptom is present in the case notes, please type 1, if it is absent type 0. If it is not mentioned in the case notes, please type "n.a.".This part is the limitation of our retrospective study design.However, we expect that characteristic dementia symptoms are systematically asked in the tertiary care dementia centers.
Additionally, if your patient has additional symptoms please write those symptoms into the "other" column.If the family history is specifically asked for and if it is negative, please type "0".If the patient has a positive family history for psychiatric disorders, amyotrophic lateral sclerosis (ALS) or other neurodegenerative disorders please write in detail (see sample cases).

DEMENTIA SEVERITY AND COGNITIVE ASSESSMENT
In this section, we used the routine tests and scales for assessing the dementia severity and the cognitive impairment in the Alzheimer Centrum Amsterdam (for the details of the tests and daily patient assessment please see; Van der Flier M. and Scheltens P., 2018.Amsterdam Dementia Cohort: Performing Research to Optimize Care).If your center uses the same tests please fill in the excel file, otherwise please prepare a separate cognitive assessment table.Of note, for statistical analysis, after collecting the data, we'll need either z scores or comparison with healthy controls.

NEUROPSYCHOLOGICAL ASSESSMENT
Similar to item 4.However, for neuropsychiatric inventory (NPI), please use multiplication of the frequency and severity scores.For instance; depression frequency score: 2, depression severity score: 1, frequency X severity= 2.

NEUROIMAGING
Right anterior temporal lobe (RATL) predominant impairment is the inclusion criterion of our study.Therefore, all cases have to be examined by at least one structural neuroimaging technic.
Since the specificity of PET imaging is relatively low, we considered to include only MRI or CT.
We will use the visual assessment method to identify the RATL atrophy.For the mesial temporal atrophy (MTA) scoring please see below (Scheltens et al., 1992 Atrophy of medial temporal lobes on MRI in "probable" Alzheimer's disease and normal ageing: diagnostic value and neuropsychological correlates).For the frontotemporal atrophy scoring please see below (Kipps

GENETIC AND PATHOLOGICAL CONFIRMATION
In this part, we will use the same coding system.Positive: 1, negative: 2 and Not available: n.a.
If the patient has a genetic mutation please write the locus in the file.(ie: MAPT; Arg406Trp) for C9orf72, please write the number of repeat.
For follow visits, follow-up duration time and clinical data will be sufficient.

Data Sharing Procedure
In this retrospective multicenter project, Alzheimer Center Amsterdam (ACA), will be responsible to control the entire data management process along the research lifecycle; from study preparation, data acquisition, data processing and statistical analysis, writing and publishing, to archiving and open data, in a transparent and traceable way.Our data management plan is prepared under the supervision of our professional research support team (Amsterdam UMC, Locatie VUmc -Research Support).
Privacy and security safeguards: Data set is pseudonymised.Each center will use the name of their city and case number.In case of any correction, update or more information, the responsible PI will be able to reach the patient and de-identify it to correct/ update the data.However, other centers will be blind, including us.Subjects will be registered in a key locked subject identification log, therefore no personal data will be shared.For the collaboration with the research partners, written agreements on data management, privacy and intellectual properties will be made.Data will not be transferred to any external parties.Data sharing agreement protect the rights of the research participants and data suppliers.
Data acquisition: Each center will collect the already recorded demographic, clinical, neuroimaging, biomarker, genetic and pathological data from their cohorts, retrospectively by using our guidebook.We will re-analyze the harmonized data.The data will not be reused.

Data collection:
The data collection system is hosted by the ACA, namely CASTOR EDC (Electronic Data Capture EDC eConsent eCRF ePRO clinical research | Castor (castoredc.com).
A data definition (dictionary), data sharing agreement, the recommendations of our legal office, the platform for questions and open discussion, documents for ethical approval, data collection template, tutorials for MRI rating, guidebook will be available and will be stored in another safe electronic system, namely SURFdrive (SURFdrive: store and share your files securely in the cloud | SURF.nl) to be shared with our collaborators.Accounts and passwords have been established in both systems and data will be secured by password-protected system.Both electronic systems have an automatic track changes function which allows all collaborators to control the data.
Open data: Signed data sharing agreements between centers do not permit open data sharing.

1 . 2 2. 3 . 5 4 6 .
Member list of the international working groupPage Roadmap of the international working group Page 5 Establishment of the international working group Page Data collection, harmonization of the data, and approach on missing data cognitive tests in the collected datasetPage 17

Analysis:
Variables were reported as means and standard deviations or proportions when appropriate.The frequency of individual symptoms was reported as proportions.All cases had symptom check list charts generated by the IWG and nearly half of the sample had details of the case notes (real life examples) for each symptom (n=134).Symptoms were listed based on their frequencies at the initial visit and each reported symptom was associated with real life examples and available cognitive tests.The list of the tests can be found below.Harmonization of the cognitive test scores: To harmonize the heterogeneous cognitive test scores, we used the following methodology.Based on the numbers and the relevance of the tests based on the published literature, a selection was made for each cognitive domain.Similar test scores of widely used tests were combined.If centers shared different versions of the well-known tests, the scores were adapted based on the published literature.Uncommon scoring systems were excluded even if a well-known test was used (e.g.: seconds instead of longest span in the digit span tests).Z-scores were calculated with the formula: (patient's score)-mean (norm)/SD(norm)

2 .
Age: years 3. Gender: female (code:0), male (code:1) 4. Symptom duration: from symptom onset to initial visit.Years NOTE: In the follow pages you will see follow-up duration.Please be sure that you typed the duration from initial visit to follow-up visit.Months (Not years) 5. Handedness: right, left, ambidexter.( If your patient is not right handed please mention the hemispheric dominance) Here you will see 4 subheadings.For dementia, please score the family history based on modified Goldman score (Please see below, Rohrer et al., 2009.The heritability and genetics of frontotemporal lobar degeneration) MODIFIED GOLDMAN SCALE 1 is an autosomal dominant family history of FTLD, MND, CBS, or PSP, defined as the presence of at least 3 affected people in 2 generations with 1 person being a first-degree relative of the other 2, 2 is familial aggregation of 3 of more family members with dementia but not meeting criteria for 1, 3 is 1 other affected family member with dementia (modified to give a score of 3 only if there is a history of young-onset dementia within the family, i.e., less than 65, and 3.5 if onset above 65), 4 is no or unknown family history.

1 .
et al., 2007 Clinical Significance of Lobar Atrophy in Frontotemporal Dementia: Application of an MRI Visual Rating Scale ).Additionally you will see a protocol for anterior temporal atrophy rating (Credit: Giorgio Fumagalli and also Harper et al., 2015 Using visual rating to diagnose dementia: a critical evaluation of MRI atrophy scales) The most important issue is that the frontal or the left temporal atrophy scores must be lower (at least 1 grade) than the RATL atrophy score.Patients with frontal/left temporal atrophy  3 (max score=4) (even if they had predominant RATL atrophy) will be excluded to minimize gross effects of general neurodegeneration.Anterior Temporal Rating Protocol 1.a.SLICE SELECTION Select the first slice where the connection between the frontal and temporal lobes is NOT visible: rate temporal pole a) Pre-rating slice: connection between the frontal and temporal lobes is still visible b) Correct slice: NO visible connection between the frontal and temporal lobes c) Post-rating slice 1.b.RATING GUIDE 0: Normal appearances (closed sulci) 1: Small sulcal slit of anterior temporal sulci 2: Temporal sulci definitely widened 3: Triangular shape of gyri 4: Severe atrophy of Temporal pole (convex sulci) 3. Kipps Frontotemporal Scale Image is from Harper et al., 2015 Using visual rating to diagnose dementia: a critical evaluation of MRI atrophy scales 7. BIOMARKER If the amyloid status of your patient is not available, please type "n.a."If the CSF or amyloid PET result of your patient is suggestive for Alzheimer's Disease, please type "1", if not in other words amyloid negative, please type "0".If you have the neurofilament data such as neurofilament light chain (NfL) or other potential biomarkers please write them into the respective column.

Figure 1 :Figure 2 :
Figure 1: Comparison of the centers.In order to avoid any potential selection bias, the mostly rated symptoms from 5 centers were compared.