Background Adrenocortical carcinoma is a rare, aggressive cancer for which few treatment options are available. Linsitinib (OSI-906) is a potent, oral small molecule inhibitor of both IGF-1R and the insulin receptor, which has shown acceptable tolerability and preliminary evidence of anti-tumour activity. We assessed linsitinib against placebo to investigate efficacy in patients with advanced adrenocortical carcinoma. Methods In this international, double-blind, placebo-controlled phase 3 study, adult patients with histologically confirmed locally advanced or metastatic adrenocortical carcinoma were recruited at clinical sites in nine countries. Patients were randomly assigned (2:1) twice-daily 150 mg oral linsitinib or placebo via a web-based, centralised randomisation system and stratified according to previous systemic cytotoxic chemotherapy for adrenocortical carcinoma, Eastern Cooperative Oncology Group performance status, and use of one or more oral antihyperglycaemic therapy at randomisation. Allocation was concealed by blinded block size and permuted block randomisation. The primary endpoint was overall survival, calculated from date of randomisation until death from any cause. The primary analysis was done in the intention-to-treat population. This study is registered with ClinicalTrials.gov, number NCT00924989. Findings Between Dec 2, 2009, and July 11, 2011, 139 patients were enrolled, of whom 90 were assigned to linsitinib and 49 to placebo. The trial was unblinded on March 19, 2012, based on data monitoring committee recommendation due to the failure of linsitinib to increase either progression-free survival or overall survival. At database lock and based on 92 deaths, no difference in overall survival was noted between linsitinib and placebo (median 323 days [95% CI 256-507] vs 356 days [249-556]; hazard ratio 0.94 [95% CI 0.61-1.44]; p= 0.77). The most common treatment-related adverse events of grade 3 or worse in the linsitinib group were fatigue (three [3%] patients vs no patients in the placebo group), nausea (two [2%] vs none), and hyperglycaemia (two [2%] vs none). No adverse events in the linsitinib group were deemed to be treatment related; one death (due to sepsis and megacolon) in the placebo group was deemed to be treatment related. Interpretation Linsitinib did not increase overall survival and so cannot be recommended as treatment for this general patient population. Further studies of IGF-1R and insulin receptor inhibitors, together with genetic profiling of responders, might pave the way toward individualised and improved therapeutic options in adrenocortical carcinoma.