Prognostic Factors in Patients with Breast Cancer Liver Metastases Undergoing Liver Resection: Systematic Review and Meta-Analysis.

Abstract

Simple SummaryRobust predictive and prognostic tools are needed in the management of breast cancer liver metastases (BCLMs). Until now, surgery has not been the gold standard of treatment of patients with BCLMs. The present manuscript highlights several predictive factors related to the primary tumor and the BCLM that may help to identify candidates for surgery with favorable outcomes in a large cohort of patients.Background: The role of surgical resection of liver metastases in patients with breast cancer liver metastasis (BCLM) remains controversial. A systematic review and meta-analysis of prognostic factors related to survival after BCLM resection was performed. Methods: An electronic search of relevant publications was performed. Pooled outcome measures were expressed as hazard ratios (HRs), including 95% confidence interval values (95% CIs), and calculated through a random-effects model. Heterogeneity was tested through the I2 index. Results: Thirty-five publications reported analyses on prognostic factors and survival. A total of 2782 patients who underwent liver resection for BCLM were included. Positive axillary lymph nodes at breast cancer diagnosis were an unfavorable survival factor (HR 1.74, 95% CI 1.25 to 2.41, I2 = 0%). Cumulative predictive factor HRs (multiple liver metastases, size of the metastases, short interval between primary tumor and onset of liver disease) related to the BCLM pattern were 1.32 (95% CI 1.17 to 1.48, I2 = 71%) and 1.51 (95% CI 1.15 to 1.98, I2 = 76%) for surgical and pathological features (resection margin and presence of extrahepatic disease), respectively. Conclusion: Resection of BCLM may provide a survival benefit for selected patients. For better long-term results, surgical selection should consider both primary tumor and BCLM features such as negative axillary lymph nodes at breast resection, a single hepatic lesion, a time longer than 24 months between breast and hepatic diagnosis, and a realizable R0 liver resection. However, the high heterogeneity among studies suggests the need for an RCT to validate the present findings.