Impact of Underlying Liver Disease on the Risk and Prognostic Factors of Breast Cancer Liver Metastases: A Retrospective Multicenter Cohort Study.
To examine factors influencing breast cancer liver metastases (BCLM) and assess the impact of underlying liver diseases (nonalcoholic fatty liver and HBsAg infection) on BCLM development.
Patients diagnosed with breast cancer at four affiliated hospitals in China between 2014 and 2024 were included. Logistic regression was used to identify factors associated with BCLM. Propensity score matching (PSM) and Kaplan-Meier analyses were performed to evaluate the prognostic impact of underlying liver diseases.
A total of 3653 breast cancer patients were included, among whom 387 (11%) were identified with liver metastasis (LM). Factors including nonalcoholic fatty liver (NAFL) and HBsAg (hepatitis B surface antigen) infection were independently associated with a lower risk of BCLM (NAFL: p < 0.001; HBsAg: p = 0.011). Subsequent analysis stratified by the severity of NAFL indicated that mild NAFL was associated with a lower risk of BCLM, whereas moderate-to-severe NAFL was associated with a higher risk of BCLM (p = 0.01 and p = 0.02, respectively). Survival analysis showed that HBsAg infection was associated with significantly longer liver metastasis-free survival (LMS) and overall survival (OS) (both p < 0.01). Further survival analysis, stratified by the presence of NAFL, revealed that mild NAFL could prolong both LMS and OS, while moderate-to-severe NAFL not only shortened LMS, but also shortened OS after LM (OSLM), so that OS was significantly shortened (p < 0.01 for mild NAFL; p < 0.05 for LMS and p < 0.01 for OSLM and OS in moderate-to-severe NAFL). Furthermore, consistent results on OS and OSLM were obtained even after employing 1:1 PSM to account for other covariate interferences (both p < 0.01).
Mild NAFL may be associated with reduced LM and improved prognosis, while moderate-to-severe NAFL appears to correlate with increased LM risk and worse clinical outcomes. Furthermore, HBsAg infection may be linked to suppressed LM and extended OS for patients with BCLM.
Patients diagnosed with breast cancer at four affiliated hospitals in China between 2014 and 2024 were included. Logistic regression was used to identify factors associated with BCLM. Propensity score matching (PSM) and Kaplan-Meier analyses were performed to evaluate the prognostic impact of underlying liver diseases.
A total of 3653 breast cancer patients were included, among whom 387 (11%) were identified with liver metastasis (LM). Factors including nonalcoholic fatty liver (NAFL) and HBsAg (hepatitis B surface antigen) infection were independently associated with a lower risk of BCLM (NAFL: p < 0.001; HBsAg: p = 0.011). Subsequent analysis stratified by the severity of NAFL indicated that mild NAFL was associated with a lower risk of BCLM, whereas moderate-to-severe NAFL was associated with a higher risk of BCLM (p = 0.01 and p = 0.02, respectively). Survival analysis showed that HBsAg infection was associated with significantly longer liver metastasis-free survival (LMS) and overall survival (OS) (both p < 0.01). Further survival analysis, stratified by the presence of NAFL, revealed that mild NAFL could prolong both LMS and OS, while moderate-to-severe NAFL not only shortened LMS, but also shortened OS after LM (OSLM), so that OS was significantly shortened (p < 0.01 for mild NAFL; p < 0.05 for LMS and p < 0.01 for OSLM and OS in moderate-to-severe NAFL). Furthermore, consistent results on OS and OSLM were obtained even after employing 1:1 PSM to account for other covariate interferences (both p < 0.01).
Mild NAFL may be associated with reduced LM and improved prognosis, while moderate-to-severe NAFL appears to correlate with increased LM risk and worse clinical outcomes. Furthermore, HBsAg infection may be linked to suppressed LM and extended OS for patients with BCLM.
Authors
Wang Wang, Zhang Zhang, Yang Yang, Yang Yang, Wang Wang, Zeng Zeng, Zou Zou, Guo Guo, Zhou Zhou, Jiang Jiang, Wu Wu, Zhang Zhang, Zhou Zhou, Zheng Zheng, Huang Huang, Wang Wang, Hu Hu
View on Pubmed