Study hypothesis and aims

We hypothesise that an interactive CBPR may increase patients' compliance with lifestyle recommendations (with respect to usual face-to-face or leaflet-based recommendations) and significantly decrease BC recurrence risk factors and BC mortality.

Prevention of BC recurrences with low-cost technologies, easily available to everybody, is a priority for public health and public finances. Informing about the major modifiable risk factors affecting the progression of BC and metastatic diffusion, and about how to reduce their impact, may also improve prognosis and prolong life in women with advanced disease.

Moreover, by improving the health literacy of patients with BC, we have the aim of reducing the risk they become prey of ‘alternative’ medicines thus declining evidence-based therapy.

The DianaWeb aims are: (1) to build a cohort of women with a previous diagnosis of BC and be willing to actively participate in the project; (2) to propose a nutritional switch towards a diet that is poor in refined carbohydrates, fats and animal proteins, and rich in unrefined cereals, pulses and other vegetable foods, according to the ECAC guidelines; (3) to promote regular physical activity of moderate intensity; (4) to monitor the change over time of dietary habits, exercise, body weight and the prevalence of metabolic syndrome parameters (including markers of chronic inflammation and of insulin resistance); (5) to monitor the occurrence of new BC-related clinical events and quality of life of the study participants and (6) to test the hypothesis that the participants who succeed in improving their pattern of risk factors can reduce the incidence of cancer recurrences (local, remote or new BC) and increase survival if already affected by metastasis.

Experimental design

DianaWeb is a CBPR for the prevention of BC recurrences and the improvement of prognosis through lifestyle modifications.

DianaWeb has a prospective, before–after study design. Since the association of metabolic syndrome or its components with cancer development and cancer-related mortality is always more frequently highlighted in the literature,34 the primary outcome of the DianaWeb in Umbria pilot study is the decrease in metabolic syndrome status, and secondary outcomes in the full-scale DianaWeb study include recurrences and mortality.

In the DianaWeb in Umbria pilot project, about 500 Italian women with a diagnosis of BC agreed to participate. The women could be with or without metastasis, local recurrence or second cancers; with in situ or invasive cancer, whatever the disease stage at diagnosis; whatever histological diagnosis; whatever the time elapsed since diagnosis. All the above traits will be taken into account as confounders in the statistical analysis.

The basic scientific strategy is to register the clinical history of the patients, provide them with evidence-based recommendations and tips for sustainable lifestyle changes, monitor participant changes, and analyse their effects on clinical outcome.

Participants will be requested to provide clinical and pathological information on their disease at the time of diagnosis and at periodic clinical follow-ups. Enrolled women will be provided with up-to-date evidence-based scientific information about the lifestyle that may prevent recurrences and cancer progression. Every few months, they will receive short questionnaires on their lifestyle, health status and treatments.

The scientific analysis will focus on the association of baseline dietary habits, physical exercise, biomarkers and anthropometric data, and of their changes over time, with clinical outcomes. Women will have access to their individual row data and to anonymised data on the whole group of participants. In the DianaWeb interactive website, individual information is protected but elaborated information (eg, descriptive statistics, case–control and cohort analyses) is open to all participants.

DianaWeb requires the following steps:

1. Setting up of a special interactive website for keeping in contact with recruited women;

2. Dissemination of the project to the public (press releases, public conferences, flyers to be distributed through pharmacies, doctors, cancer centres, charities, etc). On the basis of previous experience, word of mouth among patients who meet during clinical controls is expected to be the most effective recruitment strategy;

3. Recruitment: patients will be enrolled on a voluntary basis, will sign an informed consent form, will fill in a questionnaire on lifestyle, nutrition and medical history, and will provide anthropometric data (weight, height, waist circumference, blood pressure (BP)), results of routine biochemical analysis, and clinical information (histology report and hospital discharge letters, any other subsequent diagnosis);

4. Intervention to promote improved nutrition and physical activity through videos and lectures available on the website (theoretical lectures on preventive strategies, and practical videos on cooking techniques and specific physical exercises), and applications for smartphone; we will develop specific recommendations for women with different characteristics (lean, overweight, obese, active/sedentary lifestyle, dysmetabolic patients, patients with mutations in breast cancer genes (ie, BRCA1 and BRCA2), patients under chemotherapy, hormonal therapy or radiotherapy, patients suffering from joint pain, osteoporosis, menopausal symptoms, cardiovascular toxicity and other symptoms related to the side effects of treatments);

5. Monitoring of lifestyle changes, health changes and quality of life through periodic questionnaires; the pilot study will be completed in late 2016.

The scientific method of the DianaWeb study includes three phases. The first phase started in January 2015 and closed in June 2015. During the first phase, we created an interactive website. The home page contains a summary of the project in plain language and instructions on how to participate in phase II of the project. During phase II, we will ask the patients interested in joining the project to fill in a personal data sheet with valid email address and a mobile phone number. In this phase, we will send by email to all participating patients an information consent (IC) form and detailed explanation about the project. Once the signed IC (by mail or certified email) with a copy of the identity card, hospital discharge letter reporting the diagnosis of BC and the histology report are received, we will send it back to the patient, in order to allow her to access the project, credentials for login (ie, identification (ID) and password (PW)). Once the patient accesses with her ID and PW, a personal DianaWeb reference code is assigned and automatically reported in the patient’s set of forms. The PDF file of the signed IC will be stored in a protected folder, whereas the paper sheets are stored in a locked cabinet. Clinical data (TNM, grading, receptor (ER, PR, ErbB2), p53, Ki67, date of the first surgery, type of treatment, date and type of recurrences) will be entered into the database. In phase III, once the patient has completed the previous steps, she will be notified that she is enrolled in the DianaWeb study and invited to open her personal information section of the website. After the patient has joined the study can access the web pages containing the forms she is requested to fill in: (1) anthropometry form (with instructions on how to make the measurements at home of height, weight, waist circumference, BP), (2) medical history questionnaire, (3) 24 hours recall on food intake and physical activity, (4) short questionnaire on Mediterranean diet and (5) results of the last routine blood tests (glycaemia, cholesterol tot, HDL-C, low-density lipoprotein cholesterol (LDL-C), triglycerides, CRP, liver tests, plasma albumin, vitamin D). The forms have to be completed within the first month since recruitment. Insulin and testosterone are not routinely evaluated on blood samples of patients with BC, therefore these analyses will be specifically required for patients recruited in the DianaWeb study.

The 24 hours recall sheet contains a list of 65 food items. The participants are asked to report whether the listed foods were consumed the previous day (serving size and weight, and ricipes of food preparations are not specified). The recall sheet also contains five questions about the time spent on physical exercise on the day before the interview and enquiring on work and recreational physical activity duration (hours/minutes) and intensity (moderate, vigorous). The 24 hours recall sheet was developed to assess dietary habits in the Diana-5 trial and was used to calculate a score of adherence to the World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) recommendations. Analysis of data obtained with 24 hours recall in the Diana-5 study has suggested that compliance with the WCRF/AICR recommendations may reduce the risk of cancer and other chronic diseases through the reduction of metabolic syndrome.35 ,36

A 24-point food-items questionnaire, an adaptation of the previously validated MEDAS (a 14-points Mediterranean Diet Adherence Screener),37 is aimed at assessing adherence to the traditional Mediterranean diet. The 24-items screener includes 12 additional items that are critical to evaluate the intake of whole grains and/or legumes, non-starchy vegetables, processed meat, milk and dairy products.

From the second month onwards, participants will be asked to post regularly, every 3 months, a 24 hours recall on food frequency and physical exercise. Periodically, every 12 months, to all women enrolled in the study, we will send a query about health status (any changes about your state of health? yes/no). If the answer is yes, a message will appear asking the patient to send by mail or certified email the documentation on the diagnosis. Each year, the patient will receive an email asking her to update anthropometry (with attached instructions to make the measurements at home of height, weight, waist circumference, BP) and new chemical chemistry data (if any). An alert will remind the patient to complete the forms in case of missing information and a smiling face will appear each time the patient will succeed. Every month, the system will email the patient with kitchen recipes, prevention tips or reminders on physical activity.

To allow the patient to contact us, a section for questions and proposals has been created with buttons to indicate with whom the patient wants to interact (principal investigator, administrator, medical doctor, nutritionist, biologist, exercise specialist), or the other participating women. The maximum number of characters allowed is 500. The frequently asked questions (FAQs) section has been created in the home page to be progressively implemented. When the patient completes all the steps, she will receive a free phone number for emergencies, active 2 hours a day for 5 days a week.

During this phase, we will encourage DianaWeb participants to organise meetings, conferences, kitchen classes, walking groups, exchange of information on recipes and on their strategy to comply with the recommendations. The DianaWeb staff will participate in these activities.

The basic lifestyle recommendations are those in the ECAC.38–41 For overweight patients, we will stress the evidence-based recommendations for losing weight: reduce protein intake,42 avoid chips, potatoes, processed meat, red meat, sugared beverages, white bread, commercial pastries, refined flour.43 We will teach how to cook whole grain cereals, legumes, seeds, vegetables and low glycaemic index desserts, which have proven to help losing weight.44 Specific dietary and exercise counselling will be offered to patients with hypertension, hyperglycaemia, diabetes, dyslipidaemia, heart diseases, osteoporosis, liver steatosis and chronic inflammation.45–47

Feasibility

The primary purposes of the DianaWeb in Umbria pilot study are to guarantee that the study implementation is practical and to reduce threats to the validity of the study's outcomes. The aspects of feasibility that will be examined with the DianaWeb in Umbria pilot study are: (1) number of enrolled patients; (2) refusal rates for participation after recruitment, retention and follow-up rates as the participants move through the study and (3) adherence rates to study procedures, intervention attendance and engagement. We will also evaluate if eligibility criteria are clear, if the participants are able to understand the questions and the data collection methods, if they respond with missing or unusable data. Regarding data collection, we will assess if the participants have enough time and are able to complete data collection procedures and if the overall data collection plan involves a reasonable amount of time, or if it creates a burden for the participants. Furthermore, we will estimate the relevance and affordability of the intervention to obtain significant change.

Statistical power

In the ongoing Diana-5 study, a randomised controlled trial aimed at evaluating the effectiveness of diet modification in reducing BC recurrences,36 we have observed a marked improvement in metabolic parameters associated with cancer progression in the intervention group (ie, patients who participated monthly in kitchen classes and physical exercise sessions). Moreover, we have also observed a significant reduction in several metabolic syndrome parameters (ie, body weight, waist circumference, serum cholesterol and triglycerides) in patients enrolled in the control group who have received only written lifestyle recommendations (unpublished data), the same that we are proposing in the DianaWeb study. On this basis, we have estimated that in the DianaWeb in Umbria pilot study we would have over 90% power of obtaining significant differences in the considered metabolic parameters between compliant and non-compliant women, with a target sample of 500 patients.

In the full-scale study, we expect to reach a sufficient number of patients to eventually test the prognostic effect of lifestyle and modification of lifestyle both in the early and advanced stages of disease, and in relevant subgroups of 1–2000 patients (eg, triple negative BC), with statistical power of the order of 85–95% for a 25–33% risk reduction in the upper quintile of compliance.

Ethics

DianaWeb does not interfere with prescribed oncological treatments; rather, it recommends participants to follow the prescriptions that they have received. The participants themselves, however, may exchange information on the treatment that they have received, and this may stimulate patients to become more critically aware of their health decisions and to request a second opinion.

Potential biases and limits

Given that in the DianaWeb study the greater part of the contact with patients will be through the website, the study design could be prone to high dropout rates of participants.

Another limitation is that we will use a food and physical activity 24 hours recall that depends on the respondent's full cooperation. However, diet and physical activity history will be taken regularly every 3 months and, as such, should give more valid estimates of patients' habits.

Finally, since participation is limited to women who have an email address, our findings may not necessarily reflect the overall situation of BC survivors. In fact, the inability to use internet might reduce the participation of older patients. However, there is no reason to suppose that this aspect could bias the results of those women who will be able to participate.