Surgical treatment for non-colorectal non-neuroendocrine liver metastases: a systematic review and meta-analysis

Department of Surgery, Prince of Wales Hospital, Hong Kong SAR Aim: Liver-only metastasis from primary tumor elsewhere is common. Hepatectomy for colorectal or neuroendocrine liver metastases have favourable survival outcomes and acceptable surgical morbidities and mortalities. The role of hepatectomy for non-colorectal nonneuroendocrine liver metastases (NCNNLM) is still controversial. This study aims to perform a systematic review and meta-analysis of important clinical outcomes after hepatectomy for NCNNLM. Method: Electronic search of Pubmed and Embase databases were performed to identify relevant published studies. Patient demographics, surgical procedures, peri-operative and survival outcomes were documented. Primary end-points were 3-and 5-year overall survival (OS) and diseasefree survival (DFS). Secondary end-points were post-operative morbidity, 30-day mortality and prognostic factors associated with survival. Results: A total of 161 studies were identified involving patients with NCNNLM undergone hepatectomy. Thirty-six studies including 3827 patients met the inclusion criteria. Fourteen studies analysed NCNNLM as a whole group, while 32 studies described tumors with a specified primary, with gastric cancer being the most common. Pooled data showed that the 3-and 5-year OS were 45% (95% CI 0.40–0.50) and 35% (95% CI 0.31–0.19), respectively, whereas the 3-and 5-year DFS were 33% (95% CI 0.29–20.37) and 27% (95% CI 0.23–0.32), respectively. The postoperative morbidity rate was 24% (95% CI 0.20–0.29), whereas the 30-day mortality was 2% (95% CI 0.01– 0.03). Gastric cancer with liver metastasis had 3-year and 5-year OS of 39% (95% CI 0.32–0.46) and 25% (95% CI 0.20–0.31). Conclusion: Hepatectomy for NCNNLM may achieve favourable survival outcome, with low surgical morbidities and mortalities in selected patients. PR2