BBR3464, a novel tri-nuclear platinum complex, forms long-range DNA adducts and is highly potent when compared with cisplatin in vitro. Preclinical studies demonstrated activity in cisplatin-resistant tumours and tumours with mutated p53 status. Phase I & II clinical studies gave preliminary indications of activity in melanoma, pancreatic, lung and ovarian cancers. The aim of this study was to determine the efficacy and confirm the toxicity of BBR3464 when given either as first- or second-line treatment for advanced disease in patients with gastric and gastro-oesphageal adenocarcinoma. Two multicentre, open label, Gehan design studies were conducted; one study used BBR3464 as first-line and the other as second-line treatment for metastatic or locally advanced disease. Nineteen first-line and 26 second-line patients were enrolled receiving a total of 74 and 53 infusions, respectively. Initially, seven patients in the second-line study received BBR3464 using the planned schedule of 1.1 mg/m2 every 4 weeks; however, 5 of these patients experienced dose-limiting grade 3 or 4 febrile neutropenia; subsequent patients in both studies were treated using the modified schedule of 0.9 mg/m2, every 21 days. In 1 of 17 evaluable, previously untreated patients, regression of multiple skin lesions was noted with stabilisation of lung metastases and maxillary sinus mass, lasting 155 days. In the first-line study, the median time to progression was 85 days [95% Confidence Interval (CI): 42, 127] (2.8 months) and in the second-line study, the median time to progression was 71 days [95% CI: 42, 109] and 38 days [95% CI: 32, 73] in the 1.1 and 0.9 mg/m2 dose level groups, respectively. Toxicity data were available for 45 patients. Neutropenia was the main toxicity seen (G3: 40%, G4: 40%). Febrile neutropenia was observed in six patients (15%) treated with 0.9 mg/m2 compared with five patients (71%) treated with 1.1 mg/m2 BBR3464. Other drug-related toxicities (G3/4) included: anaemia, thrombocytopenia, nausea, vomiting, diarrhoea, mucositis and fatigue. Diarrhoea and nausea/ vomiting were adequately controlled by the use of loperamide and antiemetics, respectively. Recruitment to the second-line study was closed early due to the poor response rate (1/17 evaluable, 6%; 95% CI: 1%, 27%) and short time to progression noted in the first-line study. Further studies with BBR3464 in this tumour type are not recommended.