Shan Ou, Qian Li, Peng Shi
LM status in colon cancer was associated with adverse pathological features, nodal burden, higher CEA levels, and higher NLR. These routinely available variables may serve as supportive clinical indicators for assessing hepatic involvement, but should not be interpreted as causal factors or as a validated prediction model.
BACKGROUND: Liver metastases (LM) are clinically important in colon cancer, but routinely available clinicopathological and laboratory profiles associated with LM status remain incompletely characterized. This study aimed to identify clinical, pathological, and laboratory variables associated with LM status in patients with colon cancer.
METHODS: This retrospective observational study included patients with pathologically confirmed colon cancer diagnosed between January 2020 and December 2023. LM was identified by imaging and/or histopathological confirmation. A pragmatic 1:2 non-propensity-score matching strategy was used to select non-LM controls. Univariable and multivariable logistic regression analyses were performed to evaluate variables associated with LM status. Sensitivity analyses and exploratory subgroup analyses of synchronous and metachronous LM were conducted.
RESULTS: Among 756 eligible patients, 101 developed LM, yielding an incidence of 13.4%. The matched cohort included 101 LM patients and 202 controls. After adjustment, lymphovascular invasion (adjusted odds ratio [aOR] 2.57, 95% confidence interval [CI] 1.38-4.76; P = 0.003), perineural invasion (aOR 1.94, 95% CI 1.05-3.59; P = 0.035), N2 stage (aOR 2.48, 95% CI 1.17-5.25; P = 0.018), carcinoembryonic antigen (CEA) per doubling (aOR 1.72, 95% CI 1.42-2.10; P<0.001), and neutrophil-to-lymphocyte ratio (NLR) per 1-unit increase (aOR 1.35, 95% CI 1.15-1.59; P<0.001) were independently associated with LM status. Sensitivity analyses showed generally consistent findings. Synchronous LM was characterized by higher carcinoembryonic antigen and NLR levels than metachronous LM.
CONCLUSIONS: LM status in colon cancer was associated with adverse pathological features, nodal burden, higher CEA levels, and higher NLR. These routinely available variables may serve as supportive clinical indicators for assessing hepatic involvement, but should not be interpreted as causal factors or as a validated prediction model.