Oza 2015.

Methods	Accrual: August 2008 to August 2012 Multicentre phase II trial RCT, conducted across 36 sites worldwide (including USA) 130 women randomised
Participants	Median age: 66.0 (range 37 to 81 years) 80% of participants had either endometrioid or papillary serous histology 75% had stage IIIc and IVb disease, 54% had grade 3 tumours 95% had at least 1 prior treatment regimen for endometrial cancer
Interventions	Intervention: oral ridaforolimus 40 mg/day for 5 consecutive days by 2‐day dosing holiday Comparator: progestin (i.e. oral medroxyprogesterone 200 mg/day or megestrol 160 mg/day) or chemotherapy. Chemotherapy options that investigators could chose from included carboplatin, paclitaxel, topotecan, doxorubicin or liposomal doxorubicin Treatment cycle consisted of a 4‐week period; participants expected to receive 2 or more cycles of treatment Additional cycles permitted if participants continued to have at least stable disease and tolerated therapy
Outcomes	Primary outcome: ‐ PFS, defined as the time from random assignment to documented disease progression or death, whichever occurred first; PFS rates at 16 and 26 weeks also calculated Secondary outcomes: ‐ OS ‐ Response rate, according to RECIST (Response Evaluation Criteria in Solid Tumors) ‐ Adverse events, graded according to the US National Cancer Institute's Common Terminology Criteria for Adverse Events (NCI‐CTC) version 3.0
Notes	ClinicalTrials.gov record: NCT00739830 Final PFS and response rate analyses were based on full analysis set population (protocol prespecified interim analysis) OS analysis was based on intention‐to‐treat population (n = 130) Safety analyses based on all participants being treated as the population (n = 128) Funding considerations: Merck Sharp & Dohme Corp; Merck also funded writing assistance
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Stratified randomisation in a multicentre trial
Allocation concealment (selection bias)	Low risk	Randomisation was implemented with an interactive voice response system
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label study
Blinding of outcome assessment (detection bias) Overall survival	Low risk	Assessment of overall survival is unlikely to be influenced by no or incomplete blinding
Blinding of outcome assessment (detection bias) Progression‐free survival & tumour response rate	Low risk	Disease progression and objective tumour responses were evaluated every 8 weeks by site investigators and an independent review committee who used computed tomography scans, even though independent radiology review and investigator‐assessed results showed the same trend in terms of differences between the treatment arms, absolute response rate (ARR) in each group varied between the 2 assessments. PFS was assessed in full analysis set, not in ITT prespecified analysis
Blinding of outcome assessment (detection bias) Toxicity	Unclear risk	Information on assessments of toxicity not provided
Incomplete outcome data (attrition bias) All outcomes	High risk	The final PFS and RR analyses were conducted on the full analysis set (FAS) population at the time of the interim analysis in September 2010, which was scheduled after 58 events had been recorded. This resulted in 75% of participants and 71% of participants in the ridaforolimus and progestin/chemotherapy group, respectively, being included in the PFS and ORR results. The ITT population was used to determine OS and included participants enrolled up to the time of the database lock in August 2012
Selective reporting (reporting bias)	Low risk	Prespecified outcomes in ClinicalTrials.gov record (NCT0073983) and the Methods section of the trial publication are the same. All outcomes were reported in the Results section
Other bias	Low risk	None identified

AUC: area under the curve
 FAS: full analysis set
 ITT: intention‐to‐treat
 iv: intravenous
 OS: overall survival
 PFS: progression‐free survival
 RR: response rate