Digital Phenotyping in Women Over 70 Years of Age Treated for Breast Cancer With Any Type of Treatment

The activity tracker will register step counts for each day. the investigators will plot the average daily step counts and the 95% confidence interval across the entire study period. Then will will study the change in step count trajectory during the study. Linear mixed model will be used for describing change over time

The activity tracker will register sleep duration for each day. The investigators will plot the average sleep duration and the 95% confidence interval across the entire study period. Then will tudy the change in sleep duration trajectory during the study. Linear mixed model will be used for describing change over time

To identify digital profiles, the investigators will combine step counts profiles and sleep profiles using mixed models with latent classes. The use of a mixed model will make it possible to analyze repeat data for the population, and to determine an average profile or trajectory for the whole population. The optimal number of classes will be determined a posteriori, based on a set of statistical and clinical criteria. The most widely used statistical criterion is the "Bayesian information criterion" (BIC), which penalizes the model's likelihood according to its complexity. The BIC, which is stricter than many other criteria, has been shown to have a better performance in simulations. The number of trajectories will also be based on clinical interpretation (whether it is worthwhile retaining classes containing very small numbers of subjects, etc.).

Quality of life will be assessed with the European Organisation for Research and Treatment of Cancer (EORTC QLQ-C30) Quality of Life Questionnaire (EORTC QLQ-C30) version 3 validated in 2000, a multidimensional questionnaire validated for use with cancer patients. The QLQC30 questionnaire contains 30 items assessing five functional domains (physical, role, emotional, cognitive, and social), one overall quality-of-life domain, three symptom domains (pain, fatigue and nausea), and six individual items (dyspnea, insomnia, anorexia, diarrhea, constipation, and financial impact). Participants will respond on a Likert scale ranging from "not at all" to "very much" and from "very poor" to "excellent" for the overall quality-of-life questions only. Scores will be standardized on a scale of 0 to 100, according to the EORTC scoring manual. Higher scores correspond to better functioning, a better overall quality of life and more symptoms.

The multidimensional aspects of fatigue will be evaluated with the EORTC QLQ-FA12 version 1 module, which was validated for cancer-related fatigue in 2017. EORTC QLQ-FA12 contains 12 items assessing the physical, cognitive, and emotional domains of cancer-related fatigue. Participants will complete a four-point Likert-scale questionnaire, with responses ranging from "not at all" to "very much". All scores will be transformed to a scale of 0 to 100, with higher scores indicating a greater degree of fatigue. The estimated completion time for this questionnaire is five minutes.

Tests will be performed to compare means (Student's t test), or categorical variables (chi2test). The investigators will also perform multinomial logistic regression analyses with univariate and multivariate models, to determine the probability of belonging to a class relative to the corresponding reference class

Comorbidities will be assessed at diagnosis using the Charlson Comorbidity Index (CCI). This questionnaire takes into account the presence and severity of multiple medical conditions that may affect cancer treatment. It includes 19 comorbid conditions: myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic lung disease, connective tissue disease, ulcers, mild liver disease, diabetes, hemiplegia, moderate or severe kidney disease, organ-damaging diabetes, tumor of any kind, leukemia, lymphoma, moderate or severe liver disease, metastatic solid tumor, AIDS. Each disease has a different weighting according to the strength of its association with one-year mortality. The total CCI score is calculated by adding the weights associated with each comorbid condition presented by the patient. Higher scores indicate more severe disorders and, therefore, a poorer prognosis.

Significant life events will concern any event that occurred within a period of 6 months prior to the day of the inclusion visit, follow-up at M6 and M12, related to possible falls (the number will be specified if applicable), unplanned hospitalization for fracture or infection (or other reason), death of a close relative, change of living place or any other event considered significant by the patient. This questionnaire was developed by the project team and has not been validated by scientific publication.

Physical activity level will be measured by the Godin Leisure-Time Exercise Questionnaire (GSLTPAQ) version 1 validated in women with breast cancer in 2015. The GSLTPAQ is a validated self-administered questionnaire that includes three main questions about the frequency (lasting at least 15 minutes) in a normal week of low-intensity (e.g., easy walking), moderate-intensity (e.g., brisk walking), and high-intensity (e.g., jogging) physical activity. The total score is obtained by multiplying the frequencies of light, moderate, and high intensity physical activities by three, five, and nine metabolic equivalents, respectively, and adding them together. Finally, this score is divided into three categories (≥ 24 units equivalent to light active; between 14 and 23 units equivalent to moderately active; and <14 units equivalent to insufficiently active). The time to complete the questionnaire is estimated at 3 minutes.

The prediction of fatigue will be assessed by change from baseline in fatigue scores on FA12 questionnaire and at 6 months and at 12 months..

The prediction of a deterioration of the quality of life will be assessed by change in the global score in EORTC QLQC30 questionnaire from baseline and at 6 months and at 12 months..