Context: In people concerned by serious illness, how to anticipate the aggravation of the disease according to the patient's preferences is a challenging clinical question and an ethical key-issue to improve end-of-life care and quality of dying in France. When end of life decision occurs, many patients can no longer express themselves and advance directives do not seem to be appropriate for many patients despite the current strong incentives to write them, reinforced by the 2016 Claeys Leonetti.

The "End-of-Life Discussions" and "Advance care planning" programs developed in the United States have shown a positive impact on the aggressiveness of end-of-life care. The implementation in France of these programs has not yet been consolidated despite a first recommendation for "Planification des soins futurs", published by the French Health Authority(HAS). Inspired by the definition given in the later document, investigators propose the acronym DDA, for the Discussions Dedicated to Anticipate wishes and preferences in the event of Aggravation, defined as the dynamic and evolving process of reflection and communication between the patient, his relatives and healthcare professionals, allowing him to address his preferences and wishes regarding his care and treatment

The objective of this observational study is to characterize, in a population of patients with advanced cancer, the profile of those who take up a proposal for Dedicated Discussions on Anticipating preferences of care in the event of Aggravation (DDA) and who engage in the discussion process. Secondary objectives are to

1. evaluate the aggressiveness of end-of-life care in the group of patients who died 1 year after their inclusion, depending on their engagement in DDA occurs and whether or not their preferences are formalized; and
2. evaluate the subjective effects of the DDA with the patient and the investigating professionals, through qualitative approach by a clinical psychologist.

Method: investigators designed a mixed, quantitative and qualitative prospective, monocenter methodology to evaluate how patients take up a proposal of DDA. This proposal consists in 2 interviews: the first one is dedicated to the assessment of the patient's wishes in terms of information and participation in decision-making (API questionnaire) and to the assessment of the degree of anxiety and depression (HADS questionnaire) (E1). The second one is conducted 1 to 4 weeks later and consists in offering and initiating DDA (E2).

Quantitative outcome evaluated will be:

1. the documentation by a physician in the medical record, of patient's care preferences/values
2. the documentation by the patient of his care preference/values, either by designation of surrogate or by writing advanced directives
3. the usefulness and necessity of this approach, and the anxiety it generated, as perceived by the patient The qualitative assessment will be based on data collected during E1 and E2, and for 20 patients, during a clinical interview with a psychologist dedicated to collecting the patient's impressions of previous interviews.

Thanks to this study investigators expect to gather some data on the desire and feelings of cancer patients to engage in a DDA process and possibly formalize their end-of-life preferences, the impact of DDAs on care pathway indicators and the psychological effect for the patient with severe disease to project himself in advance into aggravation.

These expected results will provide a better understanding of the process of anticipating end-of-life situations, which is needed to improve quality of care and end-of-life conditions.

In the United States, the respect of patient's autonomy was early considered in 1990 by the Patient Self Determination Act, which stated that all persons entering a publicly funded health care facility had to be informed of their rights to formulate living wills. Meanwhile, tools have been developed to evaluate patients' expectations regarding information and participation in medical decision-making, e.g. the Autonomy Preference Index (API), initially conceived for general practice [Ende J 1989]. Since then, "End-of-Life Discussions" and "Advance care planning" programs have shown a positive impact on the aggressiveness of end-of-life care [Wright 2008, Mack 2012, Dow 2009]. However, some other studies have also suggested that patients in situations of real worsening would be less likely to seek prognostic information and could be unsettled by these programs [Michael 2013; Maciejewski 2013].

In France, the right to designate a "durable power of attorney for health care" (fr. "Personne de Confiance" (PC)) dates back to 2002, and the right to formulate "living wills" (fr. "Directives Anticipées" (DA)) concerning end-of-life care was first introduced by law in 2005 and their importance was reinforced in 2016. Recently, investigators established the validity and reliability of the French version of the API among patients with incurable cancer and in primary care setting. They found high patient expectations for information on their health status but a more variable desire to be involved in medical decisions. Three supplementary items were added specifically to evaluate incurable cancer patients' attitude towards anticipating their preference of care in the event of worsening health status. These suggested that anticipating may be for patients a specific dimension, not correlated with those of information and participation in decision making. Other French surveys show the low appropriation of PC and DA and suggest that DAs are not necessarily appropriate for all patients, despite the strong current incentives to write them.

The implementation in France of "Advance care planning" programs has not yet been consolidated despite a first recommendation for "Planification des soins futurs", published by the HAS (French Health Authority). Inspired by the definition given in the later document, we propose the acronym DDA, for the Discussions Dedicated to Anticipate wishes and preferences in the event of Aggravation, defined as the dynamic and evolving process of reflection and communication between the patient, his relatives and healthcare professionals, allowing him to address his preferences and wishes regarding his care and treatment.

Considering that the strictly numerical objective of writing living wills or designating a PC is not a relevant indicator of quality of care, investigators rather aim to get a better understanding of the process of DDA, assuming that this process would contribute to improve quality of care at the end-of-life.

The objective of this observational study is to characterize, in a population of patients with advanced cancer, the profile of those who take up a proposal for Dedicated Discussions on Anticipating preferences of care in the event of Aggravation (DDA) and who engage in the discussion process. Secondary objectives are to 1) evaluate the aggressiveness of end-of-life care in the group of patients who died 1 year after their inclusion, depending on their engagement in DDA occurs and whether or not their preferences are formalized; and 2) evaluate the subjective effects of the DDA with the patient and the investigating professionals, through qualitative approach by a clinical psychologist.

For each of 240 patients with advanced metastatic cancer recruited in the oncology department of a university hospital, the mixed method of evaluation consists in :

• collecting their preferences thanks to the API, along with their degree of anxiety/depression (HADS scale) (interview E1)
• initiating DDA (interview E2)
• collecting their evaluation of E2 by quantitative criteria (self-reported anxiety, perceived usefulness,…) and their qualitative subjective impressions on E1 and E2 (by clinical psychologist in a subgroup of 20 patients)
• following up their engagement in the DDA process and its impact on healthcare resource use in the last month of life (quantitative outcomes).

Quantitative outcomes evaluated were inspired by the results of the consensus work published by Sudore and al, on outcomes design to evaluate success of Advance Care Planning. There will be:

1. the documentation by a physician in the medical record, of patient's care preferences/values
2. the documentation by the patient of his care preference/values, either by designation of surrogate or by writing advanced directives
3. the usefulness and necessity of this approach, and the anxiety it generated, as perceived by the patient The qualitative assessment will be based on data collected during E1 and E2, and for 20 patients, during a clinical interview with a psychologist dedicated to collecting the patient's impressions of previous interviews.

Inclusion Criteria:

• Patients 18 years of age or older
• Followed by oncologists from the medical oncology and pulmonology departments of Cochin Hospital,
• with advanced stage cancer (solid neoplasm, any location) (locoregional or metastatic),
• whose objective of care is not curative (in the sense of healing)
• whose life expectancy is less than 1 year, estimated by the investigating physician, by the standard question, commonly used in the literature, "would you be surprised if this patient died within the next year?"

Exclusion Criteria:

• Severe psychopathology (generalized anxiety disorder, panic disorder, schizophrenia spectrum disorders, bipolar and related disorders, and major depressive disorders), mentioned in the patient medical record
• Persistent cognitive disorders, with type of dementia or confusion, mentioned in the patient medical record
• Patient not fluent in French
• Eastern Cooperative Oncology Group (ECOG) Performance status = 4
• A refusal of the patient after proposal and information about the study
• Patient not affiliated to a social security system

Patients population in medical care for advance cancer in oncology or pneumology, for which the aim is not the cure of the disease, but rather longer survival or quality of life maintenance, whose lifespan expectation is approximately higher than a year.