Background: Diagnosis for colorectal cancer is infrequently diagnosed and not accurate staging. Trace elements may be involved in the whole process of intestinal tumors. Can trace elements be used to predict condition and staging scores for colorectal cancer?
Methods: Patients(n=74) and the control group (n=32) were enrolled. Data from the first acceptance were recorded including age, gender, clinical staging scores of tumors, and tumor biomarkers. The serological specimens were tested for magnesium, iron, copper, zinc, calcium, and selenium with ICP-MS. T-test was used to analyze serum trace elements in patients and the control group, and LASSO regression was executed to screen which factors contributed most to clinical staging scores.
Results: The level of copper in the control group is lower than in the patients (p=0.0004, ∆mean±SEM=210.7±57.79, and 95%CI=96.14to325.4). For the other five trace elements in the control group is higher than in the patients, respectively manganese (p=0.0003, ∆ mean±SEM=-1.658±0.4470, 95%CI=-2.544to-0.7713), iron (p=0.0343, ∆ mean=-0.3672±0.1712, 95%CI=-0.7068to-0.02761), zinc (p=0.0059, ∆ mean=-0.1348±0.04794, 95%CI=-0.2298to-0.03968), calcium (p<0.0001, ∆ mean=-15.10±3.482, 95%CI=-22.00to-8.195), selenium (p=0.0216, ∆mean=-12.23±5.242, 95%CI=-22.62to-1.832). The results of LASSO regression show variables selected are Ca125 ( coefficient=5.8e-04 ) , Ca199 (coefficient=7.4e-05), zinc(coefficient=-0.2047) , and intercept(3.1951), which are regarded as the most contributing to the clinical staging scores, the coefficients for other factors are zero.
Conclusions: Ca125 and Ca199 are positively contributing to the clinical staging scores, and zinc is negatively correlated with them. Therefore, we suggest that serum zinc might be used to predict the clinical staging scores of patients with colorectal cancer.