This project aims to understand the feasibility, acceptability and real-world evidence of a novel UK-based remote brain health clinic for patients with mild cognitive impairment (MCI). A timely and accurate diagnosis of dementia is a priority in the UK and MCI is indicative of future risk of cognitive decline. An accurate etiological diagnosis of MCI (MCI-subtyping - distinguishing those who are likely to go on to develop dementia and those who are not) is vital for treatment planning. Whilst the assessment of molecular biological markers (biomarkers) for etiological diagnosis of MCI and Alzheimer's disease (AD) is increasingly recommended and employed internationally, the uptake is low in UK memory clinics. The Brain Health Clinic (BHC) has been specifically designed as a state-of-the-art diagnostic centre for those with MCI. Procedures will include a range of clinical and biomarker assessments, with molecular biomarkers based on lumbar puncture and cerebrospinal fluid (CSF) analysis. Additionally, the clinic will employ remote neuropsychiatric assessments using digital and telephonic methods. This allows for regular contact, whilst adhering to changes in clinical practice and national guidance due to the COVID-19 pandemic. Our overarching objectives are to first establish the acceptability and feasibility of the remote Brain Health Clinic and its novel clinical and biomarker assessment programme. Then secondly establish the impact of care under the Brain Health Clinic on i) care management decisions (e.g. follow-up and treatment planning); ii) time to etiological diagnosis of MCI (MCI-subtyping); and iii) time to diagnosis of dementia and severity of dementia at the time of diagnosis.

The Brain Health Clinic is designed to complement the existing memory clinic pathway and will complete the patient journey of patients with mild cognitive impairment (MCI). All patients referred to the new brain health clinic will be assessed using a standardized battery of clinical instruments by qualified clinicians, as per South London and Maudsley (SlaM) memory clinic pathway. In line with clinical practice changes and national guidance due to the COVID-19 pandemic, these assessments will be completed remotely via telephone and/or video conferencing and online.

Consenting participants will be invited to complete the Brain health Clinic's initial assessments. All assessments will be completed remotely. Assessments will include cognitive assessments such as Telephone Interview for Cognitive Status Modified version 39 (TICS-M v39), the Clinical Dementia Rating Scale Sum of Boxes (CDR-SoB) for telephone consultations, Computerised Integrated Cognitive Assessment (ICA-comp). The Hospital Anxiety and Depression (HADS) will be used to assess the symptoms of anxiety and depression. For the information from participants' carers/study partners, the following questionnaires will used: Amsterdam Instrumental Activities of Daily Living Questionnaire (A-IADL-Q), Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE); Adult Carer Quality of Life Questionnaire (AC-QoL) and Neuropsychiatric Inventory-Questionnaire (NPI-Q). Assessment battery can be completed in more than one session as required. Participants will be given the opportunity to complete self-assessments such as HADS in their own time, prior to the first assessment session.

Participants will be reviewed in the clinic at baseline, 6 and 12 months. A biomarker program consisting of lumbar puncture with cerebral spinal fluid (CSF) analysis following international guidelines (Engelborghs et al., 2017) saliva sampling, and electroencephalogram EEG will be applied to every consenting participant. This will be completed within two months of baseline assessment. A safety questionnaire for lumbar puncture will be administered to ensure eligibility for the procedure. The lumbar puncture facilities will be located in the BRC Clinical Research Facility. The CSF samples collected from the lumbar puncture will be shipped to Affinity Biomarker Labs where the analysis will be conducted using a Cobas® 6000 analyser for three assays: Elecsys® β-Amyloid(1-42) CSF, Elecsys® Phospho-Tau (181P) CSF, Elecsys® Total -Tau CSF. Consented samples for future studies will undergo centrifugation, transfer to new tubes, aliquoting into cryovials, and storage at -80 degrees Fahrenheit.

Saliva samples will be analysed by Cytox Group Limited, who will employ a polygenic risk scoring algorithm, genoSCORE™-LAB, to identify those at the highest genetic risk of cognitive decline and Alzheimer's disease using genetic data from the saliva sample (Daunt P et al. 2020). EEG will also be offered as an optional investigation to help support the diagnostic process. The EEG will be acquired by a trained researcher or neurophysiologist.

The consented participants will also be seen by a phlebotomy trained study researcher who will take blood plasma samples and store them at -80 degrees Fahrenheit.

As per clinic guidance, participants with negative CSF biomarker results (indicating that their MCI is unlikely to progress to dementia) will be discharged from the brain health clinic back to their General Practitioners (GP) and so will not complete the 6- and 12-month assessments. They will however complete all appropriate feedback questionnaires and a sub-sample will be invited to participate in a semi-structured interview to further explore their experiences of the remote brain health clinic. The interview will be conducted remotely (by telephone/video conferencing). Questionnaires will be administered to participants concerning the feasibility and acceptability of i) the brain health clinic itself and; ii) the following brain health clinic procedures: lumbar puncture, saliva sampling, EEG ICA-comp assessment and CDR-SoB assessment. Additionally, carers/study partners will be asked to complete a questionnaire concerning their views on the acceptability and feasibility of the brain health clinic. The time from the first assessment until the final diagnosis will be recorded, as well as diagnostic confidence and the plan for future monitoring, outlining the determinants for continued monitoring vs discharge to GP. A subsample of 12 participants will be invited to participate in a semi-structured interview to further explore their experiences of the remote brain health clinic. The interview will be conducted remotely (by telephone/video conferencing). The interviews will be audio recorded with participants' consent and will be transcribed. Thematic analysis will be completed, supported by Nvivo software (Qsr International, 2012). Magnetic resonance imaging (MRI)data obtained as part of the memory service or Brain Health Clinic pathway will be analysed in order to establish the predictive validity of BrainageR software and automated volumetric measures in reference to a normative population. Novel AI and machine learning tools will also be applied to MRI scans alongside existing biomarkers. No MRI scans will be carried out as part of this study. Historical MRI scans will be retrospectively analysed. Historical data on patients receiving MCI and dementia diagnoses will be obtained from the South London and Maudsley NHS Foundation Trust Biomedical Research Centre (SLaM BRC) Clinical Record Interactive Search (CRIS) resource. The CRIS system allows research access to the anonymized electronic health record of one of Europe's specialist dementia care provider and currently data are available for 20,000 patients with dementia and more than 3,000 patients with MCI.

The data source has been enhanced through more than 100 natural language processing algorithms, facilitating the extraction of in-depth data from free text, including the degree of cognitive impairment at MCI and dementia diagnosis (Couch et al., 2021). This source will be utilized to identify patients referred to SLAM memory clinics who have been diagnosed with MCI (approximately 10%). The objectives are to determine 1) the time from discharge from the memory clinic with MCI diagnosis to a new referral and dementia diagnosis, and 2) the severity of dementia at the time of diagnosis.

A sub-study involving a programme of cognitive training games will be offered to a maximum subset of 30 participants. The games target episodic memory, semantic and spatial abilities, and working memory. Participants will train 5 days per week for 10 weeks. Each training session lasts 24 min, i.e., participants will play 3 games per session each for 8 min.

The Quality assurance The monitoring of this trial to ensure compliance with Good Clinical Practice and scientific integrity will be managed by the study team. The research team will have quarterly meetings to monitor the conduct of the trail. A proportion of research projects undertaken within the Institute of Psychiatry, Psychology and Neuroscience, King's College London are selected at random and subjected to internal audit each year. It is anticipated that no further monitoring or auditing will be necessary; however, the Chief Investigator will be available to provide support with any situations that arise which have not been accounted for.

Data Management and Monitoring

An electronic Case Report Form (CRF) will be used which will be stored in a password protected database which is stored on a password protected encrypted secure network. Access to this document will be restricted to the Chief Investigator and clinical researchers. Study progress and all other aspects of data monitoring will be completed by the Chief Investigator and research team. Quarterly meetings will be held to facilitate this.

Statistical analysis:

The statistical analysis will mainly consist of descriptive analyses such as mean or median of scale scores with diversity measures (Standard Deviation or Interquartile Range) as appropriate) or proportions with Confidence Interval. Given the cross-sectional nature of the study, significant drop-out is not anticipated; however, a 10% drop-out rate will be factored into the analysis. An interim analysis is scheduled for completion at 12 months.

Feedback questionnaires: Acceptability of the brain health clinic and its procedures will be assessed from the perspective of participants via a series of short questionnaires that include both 5-point Likert scales and open-ended questions. The likert scales will contain ordinal variables therefore nonparametric methods will be used to analyse the data collected; the central tendency will be summarised by median or the mode calculations and dispersion summarised by the range across quartiles. The qualitative information obtained by the open-ended questions will be manually coded and analysed by thematic analysis; the software package NVivo will be utilised to support with this. All coding and themes will be checked by a second researcher to ensure that no aspect of the data has been overrepresented. Brain health clinic procedures will be considered acceptable if 70% of participants rate each aspect of acceptability explored as satisfactory or above. Study partners will also complete a questionnaire concerning acceptability, feasibility and clinical value of the brain health clinic. The questionnaire will be analysed utilising the same methods as the participant's questionnaires. Participant semi-structured interviews: Interviews will be audio recorded with participants consent and the results transcribed. An inductive thematic analysis will be completed, supported by Nvivo software (Qsr International, 2012). Braun and Clarke's six phase method for conducting a thematic analysis will be followed (Braun & Clarke, 2006). Themes across the dataset will be checked by a researcher with qualitative experience who is independent of the initial analysis to ensure that no aspect of the data has been under or overrepresented. Themes will be reported to inform any amendments or improvements to the brain health clinic.

Lab analysis: CSF - The following Assay system will be used to analyse CSF: ElectroChemiLuminescence Immunoassay Instrument: Cobas® 6000 analyzer series. The Assays are: Elecsys® β-Amyloid(1-42) CSF, Elecsys® Phospho-Tau (181P) CSF & Elecsys® Total -Tau CSF. Saliva - Samples will be analysed by Cytox Group Limited, who will employ a polygenic risk score algorithm, genoSCORE™-LAB, to identify those at highest genetic risk of cognitive decline and Alzheimer's disease using genetic data from the saliva sample. genoSCORE-LAB is a genetic test, that analyses patient genotypes generated from an array of 114,000 single nucleotide polymorphisms - common genetic variations - that are associated with, or protective against, the risk of developing Alzheimer's disease. Blood plasma - Blood samples will be immediately centrifuged. Plasma will be extracted from this sample and stored for future diagnostic dementia research studies (subject to ethical review). Sub-study data analysis: Data will be analysed using frequentist and Bayesian analysis methods. Variables of interest are the maximum level reached and learning slope in each game, reaction times, number of correct answers, and omission/commission errors.

Inclusion Criteria:

• Suspected Mild Cognitive Impairment requiring further diagnostic work-up to secure etiologic diagnosis, subjective cognitive impairment and mild dementia.
• Aged 18 years or over
• Fluent in the English language

Exclusion Criteria:

• A diagnosis of moderate-severe dementia
• Unwilling or unable to provide informed written consent to participate

Participants will be recruited via South London and Maudsley (SLaM) NHS Foundation trust memory services and referred by the memory clinic Psychiatrists.