The clinical value of CEA and CA 15-3 in predicting the curative effect of fulvestrant in breast cancer treatment

ZHU Wen-zhen; ZHANG Guo-chun; WANG Yu-lei; REN Chong-yang; LIAO Ning

doi:10.3969/j.issn.1001-9448.2018.06.010

Abstract：Objective To investigate the predictive significance of CEA and CA 15-3 in the fulvestrant treatment for breast cancer. Methods Thirty - two patients with breast cancer were recruited. The patients were grouped into clinical benefit group (CB, n=21) and clinical non - benefit group (non - CB, n = 11). The levels of CEA and CA 15 -3 before treatment, at the third month, at the sixth months and after the whole treatment were analyzed. The patients were then grouped into disease progression group, disease response group and disease stable group based upon the judgement criteria of CEA and CA 15 - 3 with a threshold of 25％. The sensitivities and specificities of CEA and CA 15-3 were calculated with the threshold of 25％ as well as adjusted thresholds. The ORR and CBR in each group were compared. The continuous variation of CEA and CA 15 -3 in the period of the fulvestrant treatment were analyzed. Results CEA had the sensitivities of 36％ and 48％, and specificities of 76％ and 73％ in predicting non - CB and CB, respectively. The sensitivity and specificity of CA 15-3 were 18％ and 86％ in predicting non - CB, respectively; and 67％ and 73％ in predicting CB, respectively. CEA had its best sensitivity of 46％ in predicting non - CB with the threshold of 10％. However, CA 15-3 had its best specificity of 82％ in predicting CB with the threshold of 30％. The CBR of the disease stable group was significantly different from that of the disease response group (P = 0.001). The CEA and CA 15 -3 of patients in the disease response group continuously declined during treatment. Conclusion The changes of CEA and CA 15 -3 are useful in predicting patients with breast cancer to obtain clinical benefit from the fulvestrant treatment.