Oncogeriatric: a collaboration between oncologists and geriatricians which aims to ensure that all elderly cancer patients receive treatment adapted to their condition, thanks to a multidisciplinary and multi-professional approach.

This project aims to gain a better understanding of the motivational determinants of PA and nutrition in elderly cancer patients.

It has a dual objective:

1. to identify clusters/groups in patients on the basis of daily motivational factors focusing on PA and nutrition
2. on the basis of the results obtained in (1), to propose an interventional study based on the previously established clusters, in order to examine the effects of a behavioral intervention on patients' adherence to PA and nutrition, both agreed according to an individualized goal and defined in agreement with the patient and the multidisciplinary team, taking into account the recommendations.

While cancer is the leading cause of death in people aged 75 to 85, and the second leading cause of death in people over 85 , there is little evidence in the elderly due to their under-representation in clinical trials. This under-representation increases the risk of under- or over-treatment in this population, making the elderly even more vulnerable to chemotherapy-related toxicities [3]. There is considerable heterogeneity in the population of patients over 70. Some advance in age with few comorbidities and maintained independence, while others combine several chronic pathologies and deficits . Treatments to combat these comorbidities are sometimes subject to drug interactions with anticancer therapies, which makes prescribing treatments all the more difficult in oncogeriatrics. The balance between quality of life and quantity of life is fundamental in drawing up a care plan [5]. Support for the elderly cancer patient must be comprehensive and individualized, incorporating the patient's opinion and a multidisciplinary approach to meet all identified needs .

Physical inactivity and undernutrition: risky behaviours leading to over-toxicity of treatments, worsening of co-morbidities and increased risk of mortality.

Ageing leads to physiological changes (hormonal, metabolic, etc.) and an increase in risk behaviours, including physical inactivity and undernutrition [6,7]. These are responsible for a loss of strength and muscle mass [8], exacerbated by cancer and its treatments [9], a phenomenon all the more marked in the elderly.

This progressive and generalized loss of muscle mass and strength is associated with a deterioration in physical capacity and high rates of hospitalization and mortality.Reduced muscle mass is associated with over-toxicity to chemotherapies, with a direct impact on survival.Reduced muscle strength is an important predictor of adverse events such as falls .Beyond muscle-related issues, these risk behaviors associated with aging also lead to the onset of other comorbidities and a higher risk of mortality .

In this context, nutritional monitoring and physical activity (PA) are two complementary and effective interventions for maintaining muscle status and preventing undernutrition during oncology treatment. More specifically, prevention and management of this loss of muscle mass and strength should be based on adequate energy and protein intake and multimodal PA (including muscle strengthening coupled with exercise conditioning) at moderate intensity.

These recommendations apply to both primary, age-related mass loss, and secondary, disease-related mass loss.

Although an intervention combining nutritional monitoring and regular PA practice is recognized as effective in maintaining muscle status and preventing undernutrition, it is difficult to achieve high adherence and lasting behavioural changes to move the elderly towards a more active lifestyle and to change their eating habits .

The theory of planned behavior: a theoretical approach that facilitates behavior change towards an active lifestyle associated with adapted nutritional behaviors.

Identifying the motivational factors associated with PA and nutrition in this population could help facilitate the adoption and sustainability of these behaviours.

The literature indicates that using a theoretical approach to identify motivational factors associated with PA and nutrition is important to facilitate behavior change and hope for higher adherence . A meta-analysis has showń that no theory is superior to another in terms of effectivenesś for modifying PA and that interventions are more effective when based on a single theory rather than a combination of theories . In this respect, the theory of planned behavior (TCP) is commonly used in the study of behavior change, particularly for an aging population (Figure 1). This theory assumes that the adoption of a behavior stems from the formation of intentions in the individual. These intentions are facilitated by attitudes, subjective norms (particularly in the caregiver) and self-perceived behavioral control:

• Attitudes represent the benefits perceived by the individual in performing the behavior, i.e. benefits related to the usefulness of the behavior and those related to the pleasure of performing it.
• Subjective norms correspond to the individual's social influence, represented by what those around him think of the behavior he wants to undertake.
• Perceived behavioral control refers to the perceived ease or difficulty of performing the behavior, and thus to the individual's belief that he or she possesses the necessary resources to perform the behavior. Moreover, this variable is assumed to be directly related to behavior.

However, TCP has been criticized for its variable prediction of behavioral intentions. Indeed, the designs used are based on a limited number of measures, failing to take into account the temporal fluctuation of perceptions and intra-individual variability. In response to these methodological limitations, Maher and colleagues investigated the use of repeated-measures methods to assess the relationship between intentions and behavior. In particular, the study revealed that intentions can vary according to the time of day, reflecting the dynamics of motivational variables. As a result, repeated-measures methods are effective for measuring TCP variables, as they identify the contexts most conducive to putting intentions into action.

In this respect, the Ecological Momentary Assessment (EMA) method enables repeated sampling of behaviors and psychological variables, in real time , which could take into account the temporal fluctuation of perceptions.

This data collection method is suitable for measuring TCP variables and could improve their predictive level. It is feasible with high compliance, above 80%, in the elderly.

Behavior change techniques: an effective, individualized intervention to optimize the adoption and sustainability of health behaviors

In oncology and geriatrics, intervention research indicates that an increase in TCP variables is accompanied by a significant increase in PA levels. Interventions based on behavior change techniques (BCT) , alone or in combination, have been shown to be effective in promoting PA and nutritionhave validated a taxonomy that lists 16 groups of CBTs: 1. goal and planning, 2. behavioral monitoring and feedback, 3. social support, 4. knowledge modification, 5. behavioral consequences, 6. behavioral comparison, 7. associations, 8. repetition and substitution, 9. comparison of outcomes, 10. behavior-related rewards and threats, 11. regulation, 12. antecedents, 13. identitý, 14. anticipating consequences, 15. self-conviction, 16. imagery/hidden learning.

In a context where the scientific literature does not allow for the reproducibility of protocols, which are often insufficiently described, this taxonomy standardizes the language used in intervention studies, and contributes to their readability and dissemination within the scientific community.

By improving the observability and feasibility of the components of an intervention, it becomes possible to develop effective, individualized interventions that take into account the singularity of the individual and his or her situation. With the aim of developing an optimal behavior-change intervention for elderly cancer patients, further research is needed into understanding the mechanisms of action and the effect of CBTs on PA and nutrition behaviors. To this end, N-of-1 designs facilitate the development of individualized interventions, taking into account intra-individual variability and the fluctuation of measured variables [27]. In particular, these N-of-1 designs have been recommended by the authors in recent research perspectives on CBT and PA behaviors and nutrition.

Inclusion Criteria:

1. Patient aged 70 or over
2. With cancer (solid tumors, all sites)
3. G8 ≤ 14
4. Inactive (patient not meeting recommended physical activity levels of 150 minutes per week at moderate intensity)
5. Patient has given informed, written and express consent
6. Patient affiliated to a French social security scheme.

Exclusion Criteria:

1. Known presence of brain metastases
2. Inability to participate in digital platform assessments or physical tests
3. Inability to eat orally
4. Contraindication or inability to engage in physical activity
5. Patient unable to follow up regularly for psychological, family, numerical, social or geographical reasons
6. Person deprived of liberty or under protective custody or guardianship.