The good, the bad and the ugly: surgical resection for non-colorectal, non-neuroendocrine liver metastases.

Background/aims: The aim of this retrospective study was to analyze outcomes after partial hepatectomy for non-colorectal non-neuroendocrine metastases (mNCRNNE) and to identify prognostic factors associated with survival.

Methodology: Patient demographics, tumor characteristics, treatment and outcome of 30 consecutive patients operated between January 1995 and June 2012 were analyzed. The Kaplan-Meier method was used to analyze survival and Cox regression models were applied to identify independent prognostic variables.

Results: NCRNNE primaries included stomach (n=13), breast (n=5), pancreas (n=3), choroid melanoma (n=2), unknown primary (n=1) and others (n=6). Thirty-day mortality. rate was 3.3% and morbidity rate was 23.3%. Following hepatectomy, median overall (OS) and disease- free survival (DFS) were 24 and 12 months, respectively. In univariate analysis, longer disease-free interval (p=0.04) was associated with better OS. Histology of the primary tumor (p=0.003) and the presence of extrahepatic disease (p=0.04) were associated with a lower DFS. Cox regression analysis identified histology as an independent prognostic factor for DFS (p=0.013).

Conclusions: In selected patients, resection of mNCRNNE appears associated with prolonged survival and may be performed in a specialized environment with acceptable morbidity and mortality. Patients with a longer disease-free interval, specific primary location and histology and absence of extra-hepatic disease seem to benefit the most from this approach.