Abstract
The DNA repair gene PARP1 and NF-κB signalling pathway affect the metastasis of breast cancer by influencing the drug resistance of cancer cells. Therefore, this study focused on the value of the DNA repair gene PARP1 and NF-κB pathway proteins in predicting the postoperative metastasis of breast cancer. A nested case‒control study was performed. Immunohistochemical methods were used to detect the expression of these genes in patients. ROC curves were used to analyse the predictive effect of these factors on distant metastasis. The COX model was used to evaluate the effects of PARP1 and TNF-α on distant metastasis. The results showed that the expression levels of PARP1, IKKβ, p50, p65 and TNF-α were significantly increased in the metastasis group (P < 0.001). PARP1 was correlated with IKKβ, p50, p65 and TNF-α proteins (P < 0.001). There was a correlation between IKKβ, p50, p65 and TNF-α proteins (P < 0.001). ROC curve analysis showed that immunohistochemical scores for PARP1 of > 6, IKKβ of > 4, p65 of > 4, p50 of > 2, and TNF-α of > 4 had value in predicting distant metastasis (Se(PARP1) = 78.35%, Sp(PARP1) = 79.38%, AUC(PARP1) = 0.843; Se(p50) = 64.95%, Sp(p50) = 70.10%, AUC(p50) = 0.709; Se(TNF-α) = 60.82%, Sp(TNF-α) = 69.07%, AUC(TNF-α) = 0.6884). Cox regression analysis showed that high expression levels of PARP1 and TNF-α were a risk factor for distant metastasis after breast cancer surgery (RR(PARP1) = 4.092, 95% CI 2.475-6.766, P < 0.001; RR(TNF-α) = 1.825, 95% CI 1.189-2.799, P = 0.006). Taken together, PARP1 > 6, p50 > 2, and TNF-α > 4 have a certain value in predicting breast cancer metastasis, and the predictive value is better when they are combined for diagnosis (Se(combine) = 97.94%, Sp(combine) = 71.13%).