Study design

The PASAPAS study is an interventional, controlled, randomised, single-centre, open-label, parallel-group study with two arms: an intervention (or ‘APA’) group where patients are invited to a 6-month exercise programme in supervised groups in addition to the usual and personalised dietetic care; a control group receiving the usual dietetic care and PA national guidelines. The 6-month intervention is followed by a 6-month monitoring, that is a 12-month study duration for each participant. The enrolment is planned over a period of 24 months, which corresponds to a total study duration of 36 months.

The trial was registered on the website http://www.clinicaltrials.gov (registration number: NCT01331772).

Inclusion criteria

The study sample is composed of adult women, aged 18 years or older and under 75 years, being diagnosed with a first invasive non-metastatic breast carcinoma that has been histologically confirmed, and requiring the prescription for a first line of adjuvant chemotherapy. Other inclusion criteria are: being treated in the clinical site (Léon Bérard Cancer Centre, Lyon, France); being able to participate in the APA intervention and providing a medical certificate of no contraindications to exercise issued by the general practitioner, the referring physician or the main investigator physician; living within a 60 km perimeter and agreeing to support travel expenses to attend group exercise sessions; available and willing to participate in the PASAPAS study for the duration of the programme (6 months) and during the postprogramme monitoring (6 months); able to understand, read and write French; affiliated with a social security system; having dated and signed informed consent.

Non-inclusion criteria

Non-inclusion criteria are woman with metastatic or inflammatory breast cancer; personal history or coexistence of another primary carcinoma (except another in situ carcinoma regardless of the site and/or basal cell skin carcinoma and/or non-mammary cancer in complete remission for more than 5 years); contraindications to exercise according to the investigator such as serious or unstable cardiac or respiratory pathology, uncontrolled diabetes, bone metastases and severe osteoporosis; state of severe malnutrition under the criteria of the French National Health Authority23; personal history of eating disorders; unable to be followed for medical, social, family, geographical or psychological reasons, for the study duration; deprived of their liberty by court or administrative decision; pregnant or nursing or of childbearing age without effective contraception during the study.

Dietetic intervention

All participants (intervention as well as control arms) benefit from dietetic and PA support, according to the recommendations of the second French National Health and Nutrition Program (Plan National Nutrition Santé 2, PNNS2).24 At the first visit (beginning of adjuvant chemotherapy), the participants meet the dietician and receive individualised dietary counselling according to dietary analysis, as well as print material (entitled “Tips for a healthy diet and practicing physical activity” and labelled in accordance with the PNNS2). Moreover, dietary management is offered to each participant; in case of acceptance, dietetic consultations are offered on the patient's request for the duration of the intervention (6 months).

APA intervention

Women randomised to the intervention arm are offered, in addition to dietary management (see Dietetic intervention section), an opportunity to participate in group sessions of APA over a period of 6 months, including: (1) 2 sessions/week throughout the duration of chemotherapy (four to six 3-week courses depending on the protocol, ie, for 12–18 weeks), except during the first week of each course as a recovery from exercise and to take into account fatigue and other side effects related to chemotherapy; (2) 3 sessions/week after cessation of chemotherapy, until the end of the 6-month intervention (ie, for 8–14 weeks). For the duration of chemotherapy, the APA programme includes one session of Nordic walking outdoors and one session of aerobic fitness indoors (combining cardiovascular workouts such as step, muscle strengthening and stretching or relaxation) per week. After chemotherapy, the APA programme includes two sessions of Nordic walking and one session of aerobic fitness per week.

The group sessions of a maximum of eight patients are supervised by an APA professional. Exercises consist of aerobics of moderate intensity (4–6 Metabolic Equivalents of Task, METs) adapted according to the initial PA profile of each participant. The APA programme is an individualised exercise according to comorbidities and defined by the PA level (jointly defined by the PA profile or time spent in PAs of different intensities and by aerobic capacity) and the body composition status. Each session comprises a warm-up of 10 min, a core period (20–30 min for aerobic fitness exercise or 30–40 min for Nordic walking) and a cool-down period of 10 min. The progressive increase in workload is conducted by an increased frequency (from two sessions early in the programme to three sessions at the end of the programme), then the actual duration and lastly (after the end of chemotherapy) the intensity depending on the patient's desire and capacity. In all cases, the intensity of activity remains less than 80% of the age-adjusted maximal heart rate (equal to 220—age expressed in years).

Attending the group sessions is required to determine the compliance of each participant with the APA programme, but participants are free to practice more exercise individually or attend additional group sessions. In case a participant cannot attend a group session (for medical or personal reasons or due to exceptional circumstances such as holidays), she must replace it during the same week with an individual session of another aerobic exercise (eg, Nordic walking, free swimming, hiking, biking), which is planned in advance with the APA professional; this intends to avoid a decrease in physical fitness. To verify attendance in the programme, exercise is recorded in a notebook and the number of steps is recorded by a pedometer given to participants.

If side effects occur during 3-week cycles of chemotherapy, patients may be prescribed a weekly chemotherapy (mainly paclitaxel). The APA programme remains the same for these patients, meaning they have no session during each weekly chemotherapy cycle and resume APA at the end of chemotherapy, but they can attend group sessions if they wish.

Physical activity

The level and profile of PA are assessed at each evaluation using the Physical Activity Questionnaire Activité Physique (PAQAP) questionnaire25 administered by interview; the International Physical Activity Questionnaire (IPAQ) is filled at the same time for international comparison. PA before the time of diagnosis is also assessed retrospectively at T1, to distinguish between activities that are no longer practiced since diagnosis and new activities practiced since surgery. The PAQAP questionnaire provides the estimated aerobic capacity (VO_2max, mL/min/kg), usual average daily energy expenditure (kJ/day), PA profile (ie, time spent in activities of the following intensities: below 2 METs or sedentary, 2–2.9 METs, 3–3.9 METs, 4–5.9 METs, 6–9.9 METs and >10 METs) and time spent sitting and lying.

Anthropometrics and body composition

The following anthropometric data are measured at each evaluation: weight (kg) using the same scale throughout the study, height (cm) using a fathom, waist circumference (cm) and hip circumference (cm) using a measuring tape. For standardising measurements, waist circumference is measured midway between the last floating rib and the iliac crest and the hip circumference is measured at the tip of the pubis. Body mass index (BMI) is calculated as weight (kg) divided by the square of height (m). Weight 1 month ago (close to surgery), 6 months ago (close to diagnosis) and 1 year ago is also declared retrospectively at baseline.

Risk of metabolic diseases is identified for this population as follows: metabolic risk26 if waist circumference to height ratio >0.5; insulin-resistance risk27 if waist circumference >80 cm; and additionally cardiovascular risk28 if waist circumference >88 cm.

Body composition is evaluated using a multifrequency bioelectrical impedance analysis (QuadScan4000, Bodystat; measure duration of about 30 s, after lying for 10 min) that provides body fat (%, kg), lean mass (%, kg), dry lean mass, basal metabolic rate, basal metabolic rate to weight ratio and body water (%, l; total, intracellular, extracellular). Normal-weight obese women are identified using the waist circumference and more particularly the percentage of body fat in comparison to BMI status.29 ,30 Body composition is also assessed at the time of diagnosis by analysing a CT scan (systematically performed for clinical management) at the L3 lumbar position and measuring the area occupied by adipose (subcutaneous and visceral) and lean tissue (Slice-O-Matic Software, V.4.3, Tomovision).

Nutritional intake

Dietary intake is assessed by a 3-day food diary (2-weekdays and 1-weekend day) completed by the patient in the week preceding the evaluation and reviewed by the dietician using standardised portions.31 A dietary analysis using the GENI software (Micro 6, Villers-les-Nancy, France, V.7.4) provides the caloric, macronutrient and fluid intake. The dietician also assesses the usual consumption of alcoholic beverages over the preceding 6 months, the use of dietary supplements and the compliance with the PNNS2 guidelines.32

Biological samples

A fasting blood sample is collected at each evaluation as part of the biological study (in particular, blood will be drawn before infusion of chemotherapy at T1 and T2). Two tubes are drawn each time (one dry tube of 10 mL and one EDTA tube of 10 mL). Blood parts (ie, serum, plasma and buffy coat) are separated and stored in aliquots of 300 µL in liquid nitrogen. After completion of the study, the following analyses will be performed longitudinally:

• Levels of plasma biomarkers of dietary intake of antioxidants (total carotenoids, lycopene, β-carotene, α-tocopherol);

• Lipidomics analysis: complete profile of about 50 fatty acids, including the circulating levels of saturated, monounsaturated cis, n-6 polyunsaturated, n-3 polyunsaturated and trans fatty acids, as well as the ω3 to ω6 ratio;

• Levels of cytokinic factors (leptin, adiponectin, tumour necrosis factor-α, transforming growth factor-β) and endocrine factors (insulin, insulin-like growth factor 1, sex hormone-binding globulin) in plasma;

• Metabolomics analysis: measurement of serum metabolites identified by the profile obtained by nuclear MR such as amino acids, sugars, organic acids and all identifiable metabolites of molecular weight below 1 kDa; identification of multivariate metabolic profile characteristics for each study arm.

For the cross-sectional study at surgery, a tumour sample (frozen if tumour size ≥20 mm) and a circulating blood sample that were collected whenever possible during surgery will be obtained retrospectively from the clinic biobank. Then, a complete profile of fatty acids, measurement of cytokinic and endocrine factors and metabolomics analysis will also be performed on them.

Finally, levels of fasting glucose, albumin, lipids (total cholesterol and its high-density lipoprotein and low-density lipoprotein fractions, triglycerides) are measured at each evaluation.

Psychological criteria

Psychological criteria are assessed at each evaluation using self-administered questionnaires: quality of life using the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) questionnaire with its module BR-23 specific for breast cancer33 ,34 and the Medical Outcomes Study 36-Item Short-Form Health Survey (MOS SF-36) questionnaire35 ,36 (V.4.0 in the French language); body satisfaction and global self-perception using the Questionnaire de Satisfaction Corporelle et de Perception Globale de Soi (QSCPGS) questionnaire37; self-esteem using the Scale of Rosenweig38 ,39; depression using the Beck Depression Inventory (BDI) questionnaire40–42; and anxiety using the State Trait Anxiety Inventory (STAI) questionnaire.43 ,44

Patients’ satisfaction

Overall satisfaction of the patients for the intervention in both arms is assessed at the end of the programme (T3) using a graduated scale.

Other data collection

Clinical data, demographic data, medical history and reproductive history are collected at baseline. Tobacco use is collected at each evaluation.

In the intervention group, adherence to the APA sessions (primary outcome) is monitored by the APA professionals and reasons for non-adherence are recorded.

All adverse events (except haematological adverse events, gastrointestinal and hepatobiliary events that are frequently linked to adjuvant treatment of cancer) and all concomitant medications are reported by the patients continuously during the 6-month intervention period using a booklet.

Globally, the duration of each evaluation is estimated to be approximately 2 h/patient (10 min for anthropometric measurements, 10 min for the rest necessary for impedance measurements, 45–60 min for PA assessment and 35–45 min for the dietary evaluation).

Cost analysis

Costs will be assessed prospectively for each patient. The identification, measurement and valuation of costs will be performed based on the hospital's point of view. Hence, cost computations will focus on inpatient, outpatient and home care. Length of stay within a conventional medical unit, rehabilitation unit, home care, etc will be identified and multiplied by their respective unit costs. The time frame will be 12 months. Mean costs will be assessed for APA intervention as well as control groups. All costs will be related to 2013 and will be given in euros.

Budget impact analysis

The hospital perspective will be retained and a 1-year baseline period presented. Target and prevalent populations will be estimated based on this study and the Léon Bérard Cancer Centre data.

Sample size

The number of patients to include was defined empirically. A sample size of 60 patients seems sufficient to explore the feasibility of such a programme and to identify the potential corrective measures to take into account in interventions in the context of a future randomised clinical trial for efficacy. To emphasise the size of the group of patients supported by the APA intervention (main target of the study), an unbalanced randomisation (2:1 ratio) has been proposed.

Statistical analyses

The analysis will be performed in the intent-to-treat population, including all patients randomised in the arm allocated by randomisation, in order to limit the possibility of bias associated with participants not receiving the intervention they were allocated for. Baseline characteristics will be summarised by treatment arms using standard descriptive statistics. Primary outcome will be described as the proportion of patients who participate in at least two planned APA sessions per week, with its 95% CI. Participation in the APA programme will be evaluated by the rate of patients who participate in all APA sessions and reasons for non-participation will be described. Tolerance analysis will be performed by studying the rate and its 95% CI of patients who had at least one adverse event in the APA arm.

For secondary outcome variables, changes in the anthropometric measurements, body composition variables, PA profile and dietary, psychological and biological variables will be analysed by the absolute and relative variations of each of these endpoints between randomisation (initial assessment at T1) and follow-up assessments (T2–T4). These variations will be compared between both arms of randomisation using a non-parametric test. A method for imputing missing data will be considered if necessary.

Health economics analysis

Descriptive statistics will be used to present costs. Ninety-five per cent CIs will be calculated. Costs will be compared using the Mann-Whitney test. When necessary, multiple regression analyses will be performed to examine the relationship between costs and a range of potentially explanatory clinical variables.