Multicentre analysis of long-term outcome after surgical resection for gastric cancer liver metastases

Background: The efficacy of surgical resection for gastric cancer liver metastases (GCLMs) is currently debated. Hitherto, no large-scale clinical studies have been conducted. Methods: This retrospective multicentre study analysed a database of consecutive patients with either synchronous or metachronous metastases who underwent surgical R0 resection for GCLM between 1990 and 2010. Clinical data were collected from five cancer centres in Japan. Survival curves were assessed, and clinical parameters were evaluated to identify predictors of prognosis. Results: A total of 256 patients were enrolled. The mean(s.d.) number of hepatic tumours resected was 20(24). The surgical mortality rate was 16 per cent. Median follow-up was 65 (range 1-261) months. Recurrences were detected in 192 patients (750 per cent). The median interval from hepatic resection to recurrence was 7 (range 1-72) months, and the dominant site of recurrence was the liver (724 per cent). Actuarial 1-, 3- and 5-year overall and recurrence-free survival rates were 773, 419 and 311 per cent, and 436, 324 and 301 per cent, respectively. Median overall and recurrence-free survival times were 311 and 94months respectively. Multivariable analysis identified serosal invasion of the primary gastric cancer (hazard ratio (HR) 150; P=0012), three or more liver metastases (HR 233; P<0001) and liver tumour diameter at least 5cm (HR 162; P=0005) as independent predictors of poor survival. Conclusion: Clinically resectable GCLM is rare, but strict and careful patient selection can lead to long-term survival following R0 surgical resection.