In the group of standard of care, patients will have interviews with a clinical research associate only dedicated to the record of data for outcomes assessments.

The questionnaire is a descriptive system comprises five dimensions. Each dimension has 3 levels : no problems, some problems, and extreme problems. EQ-5D health states, may be converted into a single summary index by applying a formula that essentially attaches values (also called weights) to each of the levels in each dimension. The averages of the EQ-5D-3L scores, summarized and for each sub-dimension, will be calculated in each group at baseline, 3 months, 6 months, and 12 months post-inclusion.

Costs of the introduction of ONCORAL program on the Health Insurance budget.

For each patient, the number of oral anti-cancer units really taken by the patient will be calculated from the data obtained by the delivery of treatments (and reported units) in city pharmacy or retrocession. This number will be compared to the total number of units that should have been taken if he had received the treatment in accordance with the marketing authorization of the oral anticancer drug. In each group and for each time, the averages, standard deviations, and quartiles observed in each arm will be calculated.

For each patient, the number of oral anti-cancer units actually taken by the patient will be calculated from the data from the delivery of treatments in town pharmacy or retrocession and reported units. This number will be related to the total number of units that should have been taken if he had received the treatment in accordance with the AMM of the oral anticancer drug. For each reason type of changes in relative dose-intensity, and for the causes of the changes, the proportions observed in each of the groups will be calculated

Adherence will be measured with a 6-item medication adherence scale (adherence will be defined as a score ≤ 2) and the rate of prescription renewal by the ambulatory pharmacy (adherence will be defined as a rate ≥80%).

For each type of adverse event related to SOC oral anticancer, and for each CTCAE grade, the proportions observed in each group will be calculated between baseline and 12 months. The clinical consequences according to the NCCMERP scale will be described; their proportions will be calculated for each group.

For each type of drug interaction (contraindication, not recommended association, precaution of use) related to the oral anticancer drug between the inclusion and 12 months, the proportions observed in each of the groups will be calculated. The clinical consequences according to the DDI Predictor software will be described.

For each type of drug error confirmed and related to the oral anticancer drug between baseline and 12 months, the proportions observed in each group will be calculated. The clinical consequences according to the NCCMERP scale will be described; their proportions will be calculated for each group.

The overall survival rate of the disease will be estimated up to 12 months post-inclusion using a Kaplan-Meier model. A Log-rank test will be implemented to compare overall survival rates between the two groups.

The progression-free survival rate of the disease will be estimated up to 12 months post-inclusion using a Kaplan-Meier model. A Log-rank test will be implemented to compare progression free survival rates between the two groups.

The mean scores in SAT-MED Q (Treatment Satisfaction with Medicines Questionnaire) scale will be calculated for each of the two groups. The Sat-Med Q is a self-questionnaire of seventeen questions grouped into six elements; effectiveness, side effects, comfort of use, the general opinion of the patient, evaluation of the effect of the treatment on the patient's daily life, as well as the quality of the follow-up and the information provided by the patients. The answers are expressed on a scale of 5 points. from "No, not at all = 0 points" to "Yes, many = 4 points". Totaling the direct scores of the items yields a total composite score ranging between 0 and 68. The resultant total composite score can be transformed to a more intuitive and easier to understand metric with a minimum of 0 and a maximum of 100, using a mathematic expression.

The mean total scores and in each dimension of the B-IPQ (Brief Illness Perception Questionnaire) scale will be calculated for each of the two groups. The Brief IPQ has nine items. All of the items except the causal question are rated using a 0-to- 10 response scale. Five of the items assess cognitive illness representations, two of the items assess emotional representations and one item assesses illness comprehensibility. Assessment of the causal representation is by an openended response item. A mixed repeated measures model will be realized to estimate its evolution. The link between variation in disease representations between inclusion and final assessment and adherence, dose-relative intensity, and quality of life will be tested using a multivariate linear regression model.

The total average scores of the BMQ scale will be calculated in each dimension for each of the two groups. A mixed repeated measures model will be realized to estimate its evolution. The relationship between variation in treatment representations and adherence, dose-relative intensity, and quality of life between inclusion and final assessment will be tested using a multivariate linear regression model.

This criterion is in the form of 4-level Likert scales, concerning 3 items (support of the person with whom the patient lives in a couple, support of family members, support of friends and relatives), and for each two-dimensional item (psychological support and material support). Each dimension of each item will be analyzed independently. At each measurement, the response will be compared to that given during the previous visit. A break in perceived support will be considered if the response goes from 3 "a lot" to 0 "not at all", 1 "a little" or 2 "moderately". At each time, the percentage of patients with perceived support disruption will be calculated for each item and dimension, in each of the two groups, and these percentages will be compared. An analysis of the factors influencing the occurrence of perceived support disruption will be performed using a multivariate linear regression model.

The average scores on the Doran TSCT scale will be calculated in each group. A mixed repeated measures model will be realized to estimate its evolution. The link between self-management variation (TSCT score) and adherence, dose-relative intensity, and quality of life between inclusion and final assessment will be tested using a linear regression multivariate model.

The averages of the QLQ-C30 scores, summarized and for each sub-dimension, will be calculated in each group. A mixed repeated-measures model will be constructed to estimate the evolution of quality of life over time and compare it between the two groups.

A description of the number, the percentages according to the Standard Of Care, and a list of the modalities of management of the adverse events grades 2, 3 and 4 reported by the patient in his city-hospital liaison booklet related to the oral anticancer drug will be performed.

The averages, standard deviations, and quartiles observed will be calculated for each category of interviewee on his satisfaction with the ONCORAL program (with the use of a visual analogic scale EVA, from 0 to 10).

The averages, standard deviations, and quartiles observed will be calculated for the patient on his satisfaction with the ONCORAL program (with the use of a visual analogic scale EVA, from 0 to 10).

Number and types of pharmaceutical interventions as part of the city-hospital link according to the ACT-IP score adaptation during the 12-month post-inclusion period