Hudes 2007

Methods	Study design: 3‐arm, parallel‐group, open‐label RCT. Study dates: recruitment from July 2003 to April 2005, follow‐up until second interim analysis after 446 deaths (June 2006), range of follow‐up: 14 to 36 months. Setting: multicentre, international, phase III. Countries: Global Advanced Renal Cell Carcinoma (ARCC) Trial including countries of all continents, specifically: Argentina, Australia, Canada, Czech Republic, Germany, Greece, Hungary, Italy, Latvia, Lithuania, Netherlands, Poland, Russia, Serbia, Montenegro, Slovakia, South Africa, Spain, Sweden, Taiwan, Turkey, Ukraine, UK, US.
Participants	Inclusion criteria: histologically confirmed advanced RCC (stage IV or recurrent disease), KPS ≥ 60, no previous systemic therapy, measurable disease, adequate bone marrow, renal and hepatic functions (neutrophil count > 1500 cells/mm3, platelet count > 100,000 cells/mm3, haemoglobin count > 8 g/dL. People with a history of brain metastases if their condition was neurologically stable and they did not require corticosteroids after surgical resection or radiotherapy. Exclusion criteria: serum creatinine level ≤ 1.5 times ULN; aspartate aminotransferase level ≤ 3 times ULN (≤ 5 times if liver metastases present); total bilirubin level ≤ 1.5 times ULN; fasting level of total cholesterol ≤ 350 mg/dL, triglyceride level ≤ 400 mg/dL. Sample size:626. Age (years, median with range): group 1: 59 (32 to 82); group 1a: 60 (23 to 86); group 0: 58 (32 to 81). Sex (M/F, %): group 1: 69/31; group 1a: 71/28; group 0: 66/33. Prognostic factors: performance status (KPS ≤ 70%/>70%): 17/83; prior nephrectomy (n, %): 67; prior systemic therapies (n, %): not reported; risk prognosis: MSKCC risk classification: intermediate risk (1 or 2 of 5 factors)/poor risk (3 or 4 or 5 of 5 factors (n, %)): 28/72.
Interventions	Group 1 (n = 210): IFN‐α + temsirolimus Temsirolimus (Wyeth Research, 15 mg IV weekly, 30‐minute infusion) + IFN‐α (Roferon‐A, Roche, starting dose 3 MU 3 times/week for week 1 and 6 MU SC 3 times/week thereafter). Group 1a (n = 207): IFN‐α IFN‐α starting dose of 3 MU SC 3 times/week for the first week, dose was raised to 9 MU 3 times/week for the second week and to 18 MU 3 times/week for week 3, if tolerated. Participants who were unable to tolerate 9 MU or 18 MU received the highest tolerable dose (3 MU, 4.5 MU or 6 MU). Group 0 (n = 209): temsirolimus Temsirolimus 25 mg IV weekly 30‐minute infusion. Cointerventions for participants treated with temsirolimus (standard‐of‐care according to local practice): premedication with diphenhydramine 25 mg to 50 mg IV or a similar histamine H1 blocker given approximately 30 minutes before each weekly temsirolimus infusion as prophylaxis against an allergic reaction.
Outcomes	OS (primary outcome) How measured: investigator assessed, time between date of randomization and date of death. Time points measured: not reported. Time points reported: Kaplan‐Meier survival curves over up to 30 months, median with 95% CI, HR with 95% CI. Subgroups: prior nephrectomy, KPS (≤ 70, > 70). AEs, grade ≥3 How measured: AEs (NCI‐CTC 3.0) occurring in ≥ 20% of participants in any group (all grades and grade 3 or 4), number of AEs, grade 3 or 4 per treatment group, any visit at which the participant reported a symptomatic NCI‐CTC (v.3) grade 3 or 4. Time points measured: weekly or biweekly. Time points reported: treatment period. Subgroups: not evaluated. QOL How measured: EQ‐5D and EQ‐VAS questionnaire (self‐report). Time points measured: at screening, week 12, week 32. Time points reported: week 12, withdrawal or last recorded visit (only IFN‐α vs temsirolimus) (Yang 2010). Subgroups: prior nephrectomy. PFS (secondary outcome) How measured: determined by the site investigators' assessment and a blinded assessment of imaging studies (performed by Bio‐Imaging Technologies, not shown), time between date of randomization and date of disease progression or death, whichever occurred first. Time points measured: not reported. Time points reported: Kaplan‐Meier survival curves over up to 30 months, median with 95% CI, HR with P value. Subgroups: not reported. Tumour remission (secondary outcome) How measured: CT scans of the chest, abdomen and pelvis; radionuclide bone scan MRI or CT scan of the brain; classification into participants with stable disease or objective response (RECIST); % participants who had a confirmed objective response (complete or partial) as their best response to treatment. Time points measured: before treatment, repeated at 8‐week intervals. Time points reported: 24 weeks. Subgroups: not reported.
Funding sources	Wyeth Research, Cambridge, MA, US.
Declarations of interest	GH: financial support from Pfizer and Wyeth; MC, RF, IGHS‐W, RJM: financial support from Wyeth; JD: financial support from Novartis, Chiron, Bayer, Onyx, Pfizer and Wyeth; AK: financial support from Bayer and Wyeth; DMcD: financial support from Bayer, Onyx, Genentech and Novartis. TO'T, SL and LM: full‐time employee of Wyeth Research.
Notes	Trial registration: NCT00065468, study was stopped as a result of the second predefined interim analysis.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Stratified block randomization.
Allocation concealment (selection bias)	Unclear risk	No information reported.
Blinding of participants and personnel (performance bias) Subjective outcomes	High risk	Different delivery of the interventions, no placebo‐controlled trial, participants and physicians not blinded.
Blinding of participants and personnel (performance bias) Objective outcomes	Low risk	No performance bias on OS assumed.
Blinding of outcome assessment (detection bias) Subjective outcomes	High risk	Kaplan‐Meier estimates of blinded assessment for PFS not shown, outcome assessors not blinded.
Blinding of outcome assessment (detection bias) Objective outcomes	Low risk	No detection bias on OS assumed.
Incomplete outcome data (attrition bias) (OS and PFS)	Low risk	No attrition bias on PFS or OS assumed due to small censoring rates.
Incomplete outcome data (attrition bias) (safety)	Low risk	No attrition bias on safety outcomes assumed, Inclusion of all participants as treated.
Incomplete outcome data (attrition bias) (tumour remission)	High risk	Differences in postbaseline tumour assessment (group 1: 74% vs group 1a: 80% vs group 0: 92%).
Incomplete outcome data (attrition bias) (QoL)	High risk	High risk of attrition on QoL due to high differences in completion rates between treatment groups.
Selective reporting (reporting bias)	Low risk	No reporting bias assumed, nearly all outcomes reported (besides quality‐adjusted time without symptoms or toxicity).
Other bias	Low risk	Early stop for benefit, no other risk of bias assumed.