A randomized controlled open-label trial conducted at seven study sites in Germany (Hamburg, Berlin, Essen, Dresden, Munich, Tubingen, Kiel) with assessments at baseline (T0, before randomization), 19 weeks (T1) and 26 weeks (T2) after T0. The total study duration is 28 months. It starts with 16 months of patient recruitment. After 22 months, all enrolled patients will have completed the trial (6-months period for the intervention), and at the end of 28 months, statistical analysis and publication of results will be concluded.

Figures 1 and 2 depict the study design for patients receiving adjuvant and neoadjuvant chemotherapy, respectively. During counseling, patients randomized to the intervention group receive an app-based (Physitrack®, Bastion House, 6th Floor, 140 London Wall, London EC2Y 5DN, United Kingdom) personalized exercise and nutrition program tailored to the various phases of their treatment and recovery, as well as their individual needs. The basis for this program is the concept of exercise and nutrition applied during the pilot phase [9], which was optimized through semi-structured interviews with all stakeholders in a pre-study conducted before the main trial. The program consists of two stages. In stage I (Week 1 to 18), starting with the initiation of chemotherapy, a daily training regimen lasting approximately 15 to 30 min has been developed. This includes endurance, resistance, mobility, balance exercises with gradual intensity progression. Nerve gliding exercises are included if symptoms of polyneuropathy occur due to chemotherapy. Abdominal muscle exercises are to be introduced only after complete recovery from surgery. Every 2nd week the training program is adapted to the patient’s needs after the weekly or bi-weekly patient consultations via video call or telephone. The nutritional counseling in stage I aims to assess and potentially reduce the individual risk of malnutrition (protein-energy malnutrition) according to guidelines. Based on individual information gathered with validated instruments (Nutritional Risk Screening (NRS)-2002, 3-day dietary records, MEDAS questionnaire), nutrition plans with recipes are provided. Attention is given to chemotherapy side effects that may adversely affect patients’ eating habits. Nutritional counseling aligns with the World Cancer Research Fund (WCRF)/American Institute for Cancer Research (AICR) recommendations for cancer survivors. In stage II (Week 19 to 25) after completing chemotherapy, the intensity and duration of daily training increases. Nutritional counseling focuses more on adherence to the Mediterranean Diet, which ensures the necessary nutrient supply during cancer therapy. Both, exercise and nutrition intervention programs are designed and supervised by certified experts.

BENITA Study Design for patients receiving adjuvant chemotherapy: Patients of the intervention group receiving adjuvant chemotherapy start the exercise and nutrition program at T0. It continues for 26 weeks. Patients of the control group receive usual care

BENITA Study Design for patients receiving neoadjuvant chemotherapy: Patients of the intervention group receiving neoadjuvant chemotherapy start the exercise and nutrition intervention at T0. The intervention is paused during surgery and recovery. Patients of the control group receive usual care

Included in this study are female patients aged 18 years and older with ovarian cancer, cancer of fallopian tubes, or peritoneal cancer at FIGO stages II-IV, who are undergoing surgery and chemotherapy. Both patient subgroups receiving adjuvant and neoadjuvant chemotherapy, but who have not yet started treatment, are eligible for inclusion. Exclusion criteria are patients with an Eastern Cooperative Oncology Group (ECOG) performance status greater than 2, as well as patients with insufficient German language skills. Patients with physical or mental impairments that would interfere with the implementation of the training programs or the execution of the study procedures are also excluded.

The primary endpoint is physical performance measured by the change in the total distance (in meters) of the 6-minute walking test from T0 to T2. Table 1 shows all endpoints, data collection time points and procedures.

Existing studies in patients with advanced tumor that have primarily investigated the effectiveness of a pure exercise intervention demonstrated effect sizes ranging from Cohen’s d = 0.20 to d = 0.80 [10, 11]. Considering the combined exercise and nutrition program we assume a medium effect size (Cohen’s d = 0.50) which is considered realistic and simultaneously of sufficient clinical relevance. Using a two-sample t-test, a two-sided significance level of 5% and a power of 80%, a medium effect size of Cohen’s d = 0.50 can be demonstrated (PASS 16.0.3) when at least 128 participants (64 per group) are available for analysis. Based on the BENITA feasibility study and the number of patients treated at each center, we expect per year approximately 38 patients from Berlin, 22 each from Hamburg, Essen, and Tubingen, and 11 each from Kiel, Dresden, and Munich to meet the study inclusion criteria and be willing to participate. Over the planned 16-month recruitment period, this would result in a total of 182 patients. Assuming a conservative dropout rate of 30%, the calculated necessary number of participants would still be available.

Patients will be randomly assigned with a ratio of 1:1 to one of the two arms (intervention group and control group) using a block randomization with varying block lengths, stratified by center and type of chemotherapy (adjuvant and neoadjuvant). The randomization will be performed using the Randomizer tool (www.randomizer.at), whose compliance with Good Clinical Practice (GCP) has been confirmed by the Austrian Federal Office for Safety in Health Care. The randomization list is stored in the Randomizer, and access for authorized users is encrypted via the Randomizer website, with all transactions being logged in the Randomizer. Blinding of patients and physicians is not feasible due to the study design and intervention program.

The analysis of the primary endpoint will be conducted using a mixed linear regression model. The model includes the difference in the 6-minute walking test from baseline as endpoint and treatment group, time point (categorical), the interaction between treatment group and time point, baseline value of the 6-minute walking test, type of chemotherapy, and patient’s overall physical activity level before diagnosis (binary, measured with the GPAQ, cutoff at 150 min of moderate-intensity physical activity per week or equivalent) as fixed effects. Additionally, the study center and the patient (nested within the study center) are included as random effects. The estimation will be performed using restricted Maximum-Likelihood with an independent covariance structure for the centers and a correlation structure for the repeated measurements within a patient, reflecting the dependence of the two measurement time points (exchangeable). The primary hypothesis will be tested using the treatment contrast at time point T2 (using the Wald test). The primary analysis will be conducted according to the intention-to-treat (ITT) principle. Missing values in the endpoint will not be imputed, but the implicit missing-at-random (MAR) assumption of the mixed model will be utilized.

Evaluation of the secondary endpoint change in the total 6-minute walking test distance at T1 (as a difference from baseline), will be conducted within the primary model through the treatment contrast at T1. Analysis of secondary endpoints, including physical activity in minutes per day and MET-minutes per week, as well as various dimensions of quality of life, nutritional status, physical condition, mental health, social support, and active patient participation, will also be analyzed using mixed linear regression models analogously to the primary endpoint. Analyses will be conducted according to the ITT principle without prior multiple imputations of missing values. All analyses of secondary endpoints and subgroups (intake of probiotics and social status), along with corresponding p-values, are exploratory. No adjustment for multiple testing will be made.