Yeh 2021.

Study characteristics
Methods	Design: parallel, single‐blind, 3‐arm, randomised controlled trial named SPIRIT Setting: single‐centre study in USA in Baltimore metropolitan area between August 2015 and December 2016 through mass mailing, placement of brochures in doctors' clinics, distribution of flyers in various community settings (e.g. health fairs), direct referral from study physicians, word of mouth, advertisement in local newspapers and online advertising. Follow‐up: 12 months. Weights measured at baseline and 3, 6 and 12 months
Participants	Number of participants enroled In total 121 participants randomised; 40 into control arm, 39 into weight loss arm and 42 into metformin arm Of these, 7 had endometrial cancer; 2 in control arm, 3 in weight loss arm and 2 into metformin arm Inclusion criteria Women and men aged ≥ 18 years Prior diagnosis of a solid malignant (including endometrial cancer) Completed surgical, chemotherapy or radiation therapy ≥ 3 months prior to enrolment and anticipated treatment‐free lifespan of ≥ 12 months BMI ≥ 25 kg/m2 and < 400 lbs (181 kg) Internet and telephone access Willingness to change diet, physical activity and weight Exclusion criteria Breastfeeding, pregnant or planning pregnancy within next year Medication‐treated diabetes Non‐fasting blood glucose ≥ 200 mg/dL or glycated haemoglobin ≥ 7% Current or prior regular use of metformin within past 3 months Significant renal disease or dysfunction defined as estimated glomerular filtration rate < 45 mL/minute/1.73 m2 Uncontrolled concurrent medical condition likely to limit compliance with the study interventions History of lactic acidosis by self‐report Prior or planned bariatric surgery Significant hepatic dysfunction (aspartate aminotransferase/alanine transaminase ≥ 2 × upper limit of normal or reported liver disease) Self‐reported mean consumption > 14 alcoholic drink per week Currently enroled or planned to enrol in weight loss programme Haemoglobin < 9 g/dL, platelet count < 100/μL, white blood cell count < 2.5 × 109/L Plans to relocate from the area within 1 year Use of prescription weight loss medication(s) (e.g. lorcaserin, topiramate/phentermine, phentermine, liraglutide and bupropion/naltrexone), including off‐label use of drugs for weight loss or non‐prescription weight loss medications such as orlistat within past 6 months. Baseline participant characteristics For the 7 participants with endometrial cancer, mean age 54 (SD 7.5) years, mean BMI 35.5 (SD 5.0) kg/m2. Ethnicity data specifically for participants with a history of endometrial cancer were not available. 1 participant was prediabetic and 1 reported a history of angina, arrhythmia and hypertension. Histological type of endometrial cancer not collected; 3 had stage I disease, 1 had stage III disease and 1 had stage IV disease. In 2 participants, the cancer stage was not known. Primary treatment and ECOG PS not reported. Baseline characteristics of participants according to group allocation not reported.
Interventions	Control arm Self‐directed weight loss. Meeting with trial team at beginning of study and provision of written information/websites about weight management and, if desired, after final data collection visit at 12 months. Coach‐directed behavioural weight loss arm Remote lifestyle coaching intervention‐behaviour‐based telephonic coaching with web‐based support to promote healthy lifestyle and weight loss. Goal was to achieve ≥ 5% weight loss in first 6 months and to maintain these improvements through month 12 by meeting dietary and exercise goals. Specific strategies included increased physical activity, caloric restriction, self‐monitoring (diet, exercise and weight), goal setting and problem‐solving. Metformin arm Metformin, up to 2000 mg/day. Dosing could be flexible, depending on tolerance, and given 2 or 3 times per day orally with meals for 12 months. Participants received medication‐related education and counseling from a study staff member immediately following randomisation.
Outcomes	Primary outcomes Overall survival: no deaths reported in any arm during study. Adverse events: 1 hospitalisation for heart failure (metformin arm). No other adverse events reported in any arm during study. Secondary outcomes Recurrence‐free survival: no known recurrences in any arm during study. Cancer‐specific survival: no cancer‐specific deaths reported in any arm during study. Weight loss: change in weight loss from baseline to 12 months reported. Cardiovascular and metabolic event frequency: 1 patient had congestive heart failure and was hospitalised (metformin arm). QoL: change in QoL from baseline to 12 months measured, but data were not analysed at time of undertaking this review. Power Power calculation performed, and sufficient detail provided to allow it to be replicated. Quote: "The required sample size for the study was powered for comparing change in IGF‐1 at 6 months from baseline between: (1) coach‐directed weight loss intervention and self‐directed arm and (2) metformin and self‐directed arm, respectively, each with 80% power using a 2‐sided z‐test with α of 0.025. The comparison between the coach‐directed and metformin arms was exploratory and not included in sample size and power considerations."
Notes	Clinical Trials.gov identifier: NCT02431676
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Comment: computer‐generated algorithm, generated prior to study commencement by a blinded statistician. Quote: "Randomization assignments were computer‐generated and stratified according to baseline BMI category (BMI < 30; BMI ≥ 30 kg/m2) and race (black, non‐black)." Quote: "The statistician has never seen participants, did not see randomization arm or outcome data until data analysis stage."
Allocation concealment (selection bias)	Low risk	Comment: computer‐generated, randomly selected block sizes of 3 and 6. Quote: "Assignments were generated with equal allocation to the three study arms within randomly selected block sizes of 3 and 6 using a computerized program by the study statistician."
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comment: participants and personnel were unblinded. Quote: "Intervention assignment was not blinded to the trial participants, nor the intervention staff."
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Comment: outcome assessments performed by blinded study and laboratory staff. Quote: "study staff involved in follow‐up data collection and lab staff involved in lab measures were masked to the randomization assignments."
Incomplete outcome data (attrition bias) All outcomes	High risk	Follow‐up Entered study: 2 in control arm, 3 in weight loss arm and 2 in metformin arm Withdrew from study: 1 withdrew from study between 6 and 12 months. Completed study: 6 participants completed study. Missing data for 1 participant for both laboratory and weight measurements. Intention‐to‐treat analysis Comment: analyses conducted according to an intention‐to‐treat protocol. Missing data were imputed by multiple imputation. Quote: "Intervention effects were analyzed using an intention‐to‐treat approach." Quote: "Sensitivity analysis through multiple imputation based on sensible missing not at random scenarios was conducted to evaluate the robustness of the finding under MAR [missing at random] assumption."
Selective reporting (reporting bias)	High risk	Comment: protocol not published but trial registered prospectively on ClinicalTrials.gov. QoL data were not analysed or reported.
Other bias	Unclear risk	Source of funding: (quote) "This project was supported by the Maryland Cigarette Restitution Fund and Johns Hopkins Sidney Kimmel Comprehensive Cancer Center. Drs. Yeh and Kanarek, were also supported in part by the National Cancer Institute's Cancer Centers Support Grant (5P30CA006973)." Ethical approval: (quote) "The SPIRIT trial was reviewed and approved by an Institutional Review Board at Johns Hopkins University School of Medicine." Conflicts of interest: declared, may have impacted on trial. Quote: "Healthways, Inc. developed the website for the original POWER trial in collaboration with Johns Hopkins investigators (Appel, Dalcin, Jerome). On the basis of POWER trial results, Healthways developed and commercialized a weight‐loss intervention program called Innergy. This project used the Innergy website to deliver the weight loss intervention. Under an agreement with Healthways, Johns Hopkins faculty (Appel, Dalcin, Jerome) monitored the Innergy program's content and process (staffing, training, and counseling) and outcomes (engagement and weight loss) to ensure consistency with the original POWER Trial. Johns Hopkins received fees for these services and faculty members (Appel, Dalcin, Jerome) who participated in the consulting services receive a portion of these fees. Johns Hopkins receives royalty on sales of the Innergy program. After completion of this project, Healthways sold the Innergy platform to Sharecare, which ended the relationship with Johns Hopkins. Dr. Maruthur is co‐inventor of virtual diabetes prevention program technology; under a license agreement between Johns Hopkins HealthCare Solutions and the Johns Hopkins University, Dr. Maruthur and the University are entitled to royalty distributions related to this technology. This technology/intervention is not discussed in this publication. This arrangement has been reviewed and approved by the Johns Hopkins University in accordance with its conflict of interest policies." Other sources: none identified.