BACKGROUND MABp1, an antibody that targets interleukin 1α, has been associated with antitumour activity and relief of debilitating symptoms in patients with advanced colorectal cancer. We sought to establish the effect of MABp1 with a new primary endpoint in patients with advanced colorectal cancer. METHODS Eligible patients for the double-blind phase of this ongoing, placebo-controlled, randomised, phase 3 trial, had metastatic or unresectable disease, Eastern Cooperative Oncology Group performance status score 1 or 2, systemic inflammation, weight loss, and other disease-related morbidities associated with poor prognosis, and were refractory to oxaliplatin and irinotecan. Patients were randomly assigned 2:1 to receive either MABp1 or placebo. Randomisation codes were obtained from a centrally held list via an interactive web response system. Patients received an intravenous infusion of 7·5 mg/kg MABp1 or placebo given every 2 weeks for 8 weeks. The primary endpoint was assessed in patients who received at least one dose of MABp1 or placebo (modified intention-to-treat population), and was a composite of stable or increased lean body mass and stability or improvement in two of three symptoms (pain, fatigue, or anorexia) at week 8 compared with baseline measurements. This study is registered with ClinicalTrials.gov, number NCT02138422. RESULTS Patients were enrolled between May 20, 2014, and Sept 2, 2015. The double-blind phase of the study was completed on Nov 3, 2015. Of 333 patients randomly assigned treatment, 207 received at least one dose of MABp1 and 102 at least one dose of placebo. 68 (33%) and 19 (19%) patients, respectively, achieved the primary endpoint (relative risk 1·76, 95% CI 1·12-2·77, p=0·0045). The most common grade 3-4 adverse events in the MABp1 group compared with in the placebo group were anaemia (eight [4%] of 207 vs five [5%] of 102 patients), increased concentration of alkaline phosphatase (nine [4%] vs two [2%]), fatigue (six [3%] vs seven [7%]), and increased concentration of aspartate aminotransferase (six [3%] vs two [2%]). After 8 weeks, 17 (8%) patients in the MABp1 group and 11 (11%) in the placebo group had died, but no death was judged to be related to treatment. The incidence of serious adverse events was not significantly different in the MABp1 group and placebo groups (47 [23%] vs 33 [32%], p=0·07). CONCLUSIONS The primary endpoint was a useful means of measuring clinical performance in patients. MABp1 might represent a new standard in the management of advanced colorectal cancer. BACKGROUND XBiotech.