Study Design and Population
This was a longitudinal bi-center feasibility study consisting in a non-randomized experimental intervention addressing elderly patients with breast cancer. The intervention under consideration was a digital cognitive intervention using the HappyNeuron PRESCO software App on tablets. Participants with breast cancer aged 70 years and older, undergoing treatment with chemotherapy, targeted therapy or radiotherapy, regardless of cancer stage, were recruited by a neuropsychologist at the Comprehensive Cancer Centers in Caen and Rouen (France). Non-eligibility criteria included (a) neurological and/or psychological conditions, (b) the presence of brain metastases, (c) previous treatment with brain radiotherapy, (d) documented alcohol or drug abuse or medical conditions which could impact their ability to participate, (e) having severe visual and/or hearing impairment, (f) not speaking French, and (g) screening positive for overall cognitive impairment based on the Montreal Cognitive Assessment (MoCA) score [26] (cutoff based on the patient's age and education level, according to GRECOGVASC normative data [27]). All patients included in the study provided informed written consent.
Ethical approval
The Cog-Tab-Age study was approved by the local ethics committee [Ref. CNRIPH: 19.12.16.66820, Comité de protection des personnes Ile de France IV, France]. It is registered as ID-RCB 2019-A03112-55, ClinicalTrials identifier: NCT04261153, registered on 07/02/2020. All included patients gave their written informed consent to participate before any study-related assessment. They could withdraw their consent at any time. The study was conducted in accordance with Good Clinical Practice Guidelines and the ethical principles of the Declaration of Helsinki.
Objectives
The primary objective of this study was to assess the usability of the HappyNeuron PRESCO software App on tablets in elderly patients with breast cancer. Secondary objectives were to assess eligibility, acceptability, adherence, and satisfaction with regard to a digital cognitive intervention in elderly breast cancer patients.
Intervention
Tool: The HappyNeuron PRESCO software App
The HappyNeuron PRESCO software® is a digital cognitive therapy tool. For this study, it was installed in the app format on a tablet and for use offline. By creating an account, the user's progress can be recorded in the databases. Developed by a neurologist, this software is designed to improve 12 cognitive domains, including attention, processing speed, memory, and executive functions, which are common in CRCI. The software offers 41 exercises divided into nine levels of difficulty and is available in 11 different languages. It has been used in numerous studies in the field of mental health and in two studies with young cancer patients, with positive results on cognition [28–30].
Users are free to choose from the 41 exercises and levels of difficulty, or the therapist can customize the rehabilitation session by selecting the cognitive domains to be trained, the level of difficulty, and the session duration. Prior to starting the exercises, a practice example is available. Upon completion of each exercise, the software provides automatic feedback to praise or motivate the participant to persist despite any setbacks.
Procedure
After inclusion, participants were asked to use a tablet to complete three 20-minute sessions on the digital cognitive stimulation based on the software HappyNeuron. The first session was performed in the presence of a neuropsychologist while the other two sessions were performed in autonomy without the neuropsychologist’s help. Participants met the neuropsychologist for the first session in a quiet room at the cancer center. During the first session, they were given a tablet, which they would use until the end of their participation in the study. During this session, the neuropsychologist demonstrated several exercises (always at level one of difficulty) to the participants (Table 1), helped them navigate the software and answered any questions on how to use it.
For the second and third session, the participants completed the exercises in autonomy (i.e. without the neuropsychologist). They had the option to perform sessions at home or at the cancer center. They were advised to use the software for 20 minutes and to choose the exercises themselves and the level of difficulty.
Table 1. HappyNeuron® exercises used in Cog-Tab-Age study.
Memory (verbal and visual)
|
Words, Where are you?
Elephant Memory
Shapes and Colors
Heraldry
Displaced Characters
Displaced Images
N-Back
Around the World in 80 trips
I Remember You!
Restaurant
An American in Paris
Find Your Way!
Chunking
Objects, Where are You?
|
Attention / Executive functions / Information processing speed
|
Pay Attention!
Private Eye!
Sound Check!
Ancient Writing
Towers of Hanoi
Basketball in New-York
Hurry for Change!
Two-Timing
Under Pressure
Gulf Stream
Catch the Ladybug!
|
Language
|
Split words
Embroidery
Secret files
Speak Your Mind!
Decipher
Writing in the Stars
This Story is Full of Blanks!
Which One is Alike?
|
Logic
|
The Right Count
Countdown
Ready, Steady, Count!
|
Visuospatial abilities
|
Sleight of Hands
Entangled Figures
Point of View
Turn Around and Around
|
Assessments
Baseline assessment
Demographic variables included age, family status, living status, help at home, level of education and previous experience with digital tools. Clinical characteristics included cancer stage, previous cancer history, comorbidities, functional status with the Eastern Cooperative Oncology Group (ECOG) scale, prior and current anti-cancer therapies, and use of psychotropic medication.
Subjective cognitive complaints were assessed at baseline using the French version of the self-report Functional Assessment of Cancer Therapy-Cognitive Function (FACT-Cog) [31, 32], composed of four subscales: Perceived Cognitive Impairments (PCI), Impact on Quality of Life (QoL), Comments from Others (Oth), and Perceived Cognitive Abilities (PCA).
Assessments during intervention
Information collected in the second and third sessions performed in autonomy included the participants’ preference for onsite or at-home session, time spent navigating the interface (excluding time taken to complete exercises), number of exercises completed for each session, cognitive domains most trained, and interval between sessions.
Post-intervention assessment
Satisfaction with the software was assessed by a 9-item self-report questionnaire designed for this study. It was rated on a 5-point Likert scale: “strongly disagree”, “disagree”, “neutral”, “agree”, and “strongly agree”. It included questions concerning patients’ appreciation of the software: overall satisfaction, easiness and content of exercises, easiness of use of the software, need for more practice to feel comfortable with the software, interest to further use the software, and recommendations or not to other persons of the same age. It also contained an open-ended question concerning their greatest challenges using the software.
Study endpoints
Primary endpoint: usability
The 20 minutes spent using the tablet in the independent sessions were divided into “training time”, which was spent completing the exercises and was collected by the software, and “navigation time”, which was spent navigating the software and reading the exercise instructions, calculated by subtracting the training time from the 20-minute session. Based on our experience with the software, we hypothesized that an elderly patient would spend half of the 20 minutes with the tablet, navigating the software, choosing the exercises etc. Thus, usability was defined as completing at least three exercises during the training time (10 minutes) during one of the two independent training sessions.
If more than three exercises were completed, the software was considered to have a very high level of usability. Completion of three exercises corresponded to a high level of usability, two to a medium level, and one to a low level. Uncompleted exercises were considered as a very low level of usability of the software.
Secondary endpoints: adherence and satisfaction
The level of patients adherence was determined by their completion of three sessions of the complete intervention (high adherence), two sessions out of three (medium adherence), or one session out of three (low adherence). Patients were considered as satisfied if they answered “agree” or “strongly agree” to questions in the satisfaction questionnaire.
Statistical considerations
Assuming an acceptance rate of 70% with a 95% confidence interval of width (0.25), 49 assessable patients were required to meet the main criterion of evaluating the usability of the intervention. Anticipating a maximum of 10% of non-evaluable patients, we decided to enroll 55 patients.
Descriptive statistics were performed for the sociodemographic and clinical variables, mean and standard deviation for continuous variables and frequency with corresponding percentage for categorical variables. Reasons for non-eligibility and their corresponding percentage were reported, along with rate of acceptability and reasons for refusing to participate in the study. Frequency and percentage for each level of usability were reported. Comparisons were made with Fisher’s exact test for categorical variables and the Kruskal-Wallis rank sum test for qualitative variables.
If participants completed more than the two required sessions, only the required sessions were included in the analysis. For patients who had performed more than 20 minutes in one session, only the first 20 minutes were considered. Significant cognitive complaints were defined using the normative data of the PCI (perceived cognitive impairment) subscale of the FACT-Cog self-report questionnaire (≤ 10th percentile of normative data depending on age [32]).