Surgery has become well established for patients with colorectal and neuroendocrine liver metastases. However, the value of this procedure in non-colorectal and non-neuroendocrine metastases (NCRNNELMs) remains unclear. We analyzed the outcomes of patients that underwent liver surgery for NCRNNELMs and for colorectal liver metastases (CRLMs) between 2012 and 2017 at our institution. Prognostic factors of overall and recurrence-free survival were analyzed, and a comparison of survival between two groups was performed. Seventy-three patients (30 NCRNNELM and 43 CRLM) were included in this study. Although the mean age, extrahepatic metastases, and rate of reoperation were significantly different between the groups, recurrence-free survival was comparable. The 5-year overall survival rates were 38% for NCRNNELM and 55% for CRLM. In univariate analysis, a patient age of ≥60 years, endodermal origin of the primary tumor, and major complications were negative prognostic factors. Resection for NCRNNELM showed comparable results to resection for CRLM. Age, the embryological origin of the primary tumor, and the number of metastases might be the criteria for patient selection.

Primary hepatic angiosarcoma (PHA) is a rare malignant tumor of the liver, and data on patient outcome after surgical treatment are scarce. The aim of this study was to evaluate postoperative morbidity and overall survival (OS) of patients who underwent hepatectomy for PHA. This is a bicentric retrospective analysis of all consecutive patients who underwent liver resection in curative intent for PHA between 2012 and 2019 at the University Hospital of Muenster and the University Hospital of Bern. Nine patients (five female, four male) were included from both centers. Median age was 72 years (44–82). Most lesions (77.8%) were larger than 5 cm, and mean size of the biggest lesion was 9.4 ± 4.5 cm. Major hepatectomy was performed in four (44.4%), and radical resection (R0) was achieved in six (66.7%) patients. Postoperative complication rate was 88.8%, including 44.4% higher than 3a in the Clavien–Dindo classification. OS survival rates at 1, 2, and 3 years were 44.4%, 22.2%, and 12.5%, respectively, and median OS was 5 months. OS was significantly better after radical resection (R0: 15 months vs. R1: 0 months, p = 0.04), whereas presentation with tumor rupture at diagnosis was associated with the worst OS (0 months vs. 15 months, p = 0.005). Disease recurrence occurred in three patients (33.3%) between three and seven months after surgery. Radical resection remains the only potentially curative treatment option for PHA. However, postoperative morbidity is high, and the overall prognosis remains poor. Multimodal therapy options and management strategies are urgently needed and could improve the prognosis of patients suffering from PHA in the future.