Liver transplantation (LT) for malignant liver diseases represents up to 50 per cent of all indications for this operation. Hepatocellular cancer is the most common of these indications and, when the Milan criteria are applied, 5-year disease-free survival (DFS) rates of 75-85 per cent have been obtained. Recently, these criteria have been widened successfully, based on both tumour morphology and biology combined with different locoregional treatments designed to downstage disease. The success of LT in the treatment of primary liver tumours has led to renewed interest in its role as a treatment for secondary liver tumours (SLTs) 1 -4 . Standard treatment for SLTs includes a combination of chemotherapy and liver resection. Lately, liver resectability has been increased markedly as a result of more effective chemotherapy, parenchyma-sparing operations and/or staged surgical procedures. Further extension of survival can also be achieved by incorporating ablative procedures, arterial embolization, and hormone and radiolabelled chemotherapies into treatment protocols including resection. Despite these advances, curative liver resection remains applicable in only 20 per cent of patients, owing mainly to multifocality and the extent of bilobar disease.Total hepatectomy followed by transplantation for liver-only secondaries is by definition an R0 procedure and is therefore a potentially curative therapy. For metastases from neuroendocrine tumours (NETs), 5-year overall survival and DFS rates ranging from 36 to 92 per cent and from 9 to 78 per cent respectively have been reported. For colorectal metastases, 5-year overall survival rates of 60 per cent have been reached 1,2 . These results are comparable to those of state-of-the-art oncological treatments and therefore merit further investigation based on refined selection criteria and neoadjuvant and adjuvant therapies.Gastroenteropancreatic NETs are often diagnosed at an advanced stage. Fortunately the metastatic disease often remains confined to the liver, so R0 primary tumour resection and liver resection may represent a curative option. Liver resection is rarely appropriate, however, as detailed intraoperative imaging and thin-slice examination of the resected parenchyma often reveals many more lesions than those identified by routine imaging and pathology 5,6 . Until recently LT experience for NET was based on small single-centre and heterogeneous multicentre experiences. Recent reports, however, have shed new light on the value of LT 1,7 . The European Liver Transplant Registry 1 and Milan 7 series, including 213 and 24 recipients respectively, identified a number of factors as a prerequisite for success: presence of a well differentiated, low-grade tumour with Ki-67 less than 5-10 per cent; primary tumour localization within the portal venous drainage area; tumour burden less than 50 per cent of liver volume; age below 55 years; response to pre-LT treatment with somatostatin analogues and m-Tor inhibitors; and stable or controlled disease for at least 6 months. To...