von Gruenigen 2009.

Methods	Design Comment: Parallel design, two‐arm, randomised controlled trial Quote: "...prospective, two‐group randomized controlled trial..." Setting Comment: Single‐centre study in Ohio, USA Quote: "...women included in the cancer registry at the Ireland Cancer Center diagnosed from 2001–2004..." Follow‐up Comment: 24 months
Participants	Number of participants enrolled 45 participants enrolled; 23 into the intervention arm and 22 into the control arm Inclusion criteria Histologically‐confirmed endometrial cancer Stage I or II Undergone surgical treatment of endometrial cancer in the form of total abdominal hysterectomy and bilateral salpingo‐oophorectomy +/‐ lymphadenectomy No evidence of disease at time of enrolment Performance status 0‐2 BMI > 25 kg/m2 Exclusion criteria Clear cell or papillary serous histology Baseline participant characteristics There were no significant differences in the baseline characteristics of participants between the two groups. The mean age of participants in the intervention arm was 54 years (SEM 2.0 years) compared with 55.5 years (SEM 1.6 years) in the control arm. Overall, the mean BMI was 42.3 kg/m2; 43.5 kg/m2 (SEM 2.1) in the intervention arm and 41.1 kg/m2 (SEM 2.2) in the control arm. The co‐morbidities hypertension, diabetes and metabolic syndrome were present in 65.2%, 17.4% and 26.1% of participants in the intervention arm compared with 36.4%, 27.3% and 27.3% of participants in the control arm, respectively. All participants had a performance status of 0‐2. All participants underwent surgical treatment of their endometrial cancer, on average, 2 years earlier. Details of adjuvant treatment and grade, stage and histological type of endometrial cancer were not provided.
Interventions	Intervention arm Comment: Consisted of group sessions based on other nutrition and exercise goals and delivered by a registered dietitian, principle investigator and psychologist for 6 months. Participants were encouraged to gradually increase walking or other aerobic activity to 5 days per week for 45 minutes or more per session. Reinforcement of the content of group sessions was provided on an individual basis by the principle investigator at 3, 6 and 12 months. Quote: "...Group session topics included: weight loss readiness and goal‐setting, physical activity, portion sizes and food intake per mypyramid.gov, emotional eating/negative thinking, behavior modification, grocery shopping and reading food labels, relapse prevention, eating out and in social situations, and stress management...(The groups) met weekly for 6 weeks, bi‐weekly for 1 month, and monthly for 3 months. Participants were contacted by the RD (MBK) by phone or newsletter every week that the group did not meet. Phone calls were structured in content and included reinforcement and discussion regarding the previous week's topic. Participants were also given feedback on individual progress towards physical activity and nutrition goals...Pedometers were provided to and used by the LI group for patient feedback...Study participants in both groups saw the PI at 3, 6 and 12 months. Both groups received counselling regarding overall health concerns and LI participants received specific reinforcement of group session topics. " Control arm Comment: Received usual care and were provided with a generic booklet on improving health. Individual meetings were held with the principle investigator at 3, 6 and 12 months, however, these consisted of counselling regarding overall health concerns rather than a discussion about weight loss and physical activity. Quote: "the (control arm) received only an informational brochure after randomization ("Better Health and You," Weight Control Information Network, June 2004)...(control arm) participants did not receive any advice related to weight loss, physical activity or nutrition at these visits..."
Outcomes	Primary outcomes Overall survival: Deaths reported for the duration of the study period (24 months) but insufficient data available to determine hazard ratio. Quote: "Within 24‐months, 2 patients deceased: n=1 to brain aneurysm and n=1 to kidney cancer." Adverse events: No reported adverse events in either intervention and control arms Quote:"(Adverse events reported) as required by the IRB. No adverse events due to study procedures occurred Second outcomes Recurence‐free survival: Not reported Cancer‐specific survival: Deaths reported for the duration of the study period (24 months) but insufficient data available to determine hazard ratio Quote: "During the study period...2 patients deceased: n=1 to brain aneurysm and n=1 to kidney cancer. Both deaths were in the control arm..." Weight loss: Weight change from baseline at 3, 6, 9, 12 and 24 months reported Cardiovascular and metabolic event frequency: No events reported up to 24 months follow‐up QoL: Change in QoL from baseline at 3, 6, 9 and months reported using FACT‐G questionnaire Quote: "QoL and self‐efficacy were assessed at baseline and at 3, 6, and 12 months... QoL was measured by the Functional Assessment of Cancer Therapy‐General (FACT‐G), a valid and reliable questionnaire evaluating physical, functional, family‐social, and emotional well‐being domains. A fatigue subscale (‐F) and an endometrial symptom subscale (‐En) were also used..." Power Comment: A power calculation was performed and sufficient detail was provided to allow it to be replicated Quote: "Approximately 25 patients per group were needed to provide 80% power to detect a difference between groups in mean weight change from baseline to 12 months of 5 kg (11 lb) or greater, representing approximately 5% for an obese female (alpha = 0.05, two‐sided, SD = 5.0). Five percent weight change is considered clinically relevant and a recommended goal for weight loss over 6 months."
Notes	The follow‐up was described as being of 12 months duration in the publication, however, when contacted the authors were able to provide data for weight change up to 24 months This was the pilot study preceding the definitive trial, which is also included in this review (McCarroll 2014)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Comment: stratified block randomisation based on BMI employed Quote: "After enrolment, participants were randomly assigned (to intervention or control arm)... Randomization was stratified according to patient BMI (25–39.9 versus N40 kg/m2 ) using a stratified blocked randomization scheme."
Allocation concealment (selection bias)	Unclear risk	Comment: No details of allocation concealment provided by study authors
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comment: Blinding of participants and personnel not possible due to nature of intervention. Principle investigator involved in delivery of intervention so aware of randomisation. Quote: "Due to the interventions performed by the study team (dietitian, Physical therapist, psychologist, etc.), they were able to know who was in each group."
Blinding of outcome assessment (detection bias) All outcomes	High risk	Comment: Principle investigator performed outcome assessments and was unblinded to treatment group allocation. This is unlikely to affect weight measurements but may impact upon quality of life assessments
Incomplete outcome data (attrition bias) All outcomes	High risk	Follow‐up Entered into the study: 23 in intervention arm, 22 in control arm Withdrew from study: 5 in intervention arm, 2 in control arm Completed study: 17 in intervention arm, 20 in control arm (though data from assessment at 12 months missing for 2 participants in the control arm) Two withdrawals in the intervention arm were due to issues with work, the reason for the other three withdrawals in this group were not stipulated. The two withdrawals from the control arm occurred prior to the first assessment at 3 months and the reasons were not stipulated. Quote: "Attrition in the trial overall was 16% [2 patients (10%) in the UC group versus 5 (22%) in the LI group; P = 0.242], therefore 84% completed follow‐up assessments. Specifically, 78% of patients [LI: 17/23 (74%), UC: (18/22) (82%)] completed the 12‐month assessment time point and there was no difference between groups (P = 0.523)" Intention‐to‐treat analysis Comment: Analyses were conducted according to an intention‐to‐treat protocol. There was, however, significant missing data; 19% of weight values and 15% to 19% of QoL data were missing. Missing data were imputed using three different techniques; last and next average (average of last and next known values), previous row mean method and last observation carried forward. All produced similar findings and so only the results obtained using the first approach were included in the journal publication. Quote: "Imputation was done for 19% (35/ 180) of weight values, 10 patients (LI: 6 and UC: 4) had weight values imputed for the final weight. These patients opted to not complete the assessment and values were imputed based on the most recent physician visit, if they had one or were imputed...Imputation was done on between 15–19% of values for the various QoL and eating behavior measures."
Selective reporting (reporting bias)	Low risk	Comment: Protocol not published but trial registered prospectively on clinicaltrials.gov and all prespecified outcomes reported.
Other bias	High risk	Source of funding Comment: Source of funding described Quote:"This research was supported by a grant from the Lance Armstrong Foundation " Ethical approval Comment: Ethical approval was obtained Quote: "Institutional review board approval was obtained ..." Conflicts of interest Comment: No significant conflicts of interest noted Other sources Study failed to recruit sufficient numbers to meet a priori total in time frame. One patient in the intervention arm underwent gastric bypass at 9 months after the start of the intervention and another between 12 and 24 months. Both were included in the final analysis.

BMI: body mass index; ECOG: European Cooperative Oncology Group; FACT‐G: Functional Assessment of Cancer Therapy‐General; QoL: quality of life; RCT: randomised controlled trial; SD: standard deviation; SEM: standard error of the mean