The Expression Of COX-2 And TRX-1 In Breast Cancer And The Correlation With Clinicopathologic Characteristic And Prognosis

The expression of COX-2 and TRX-1 in breast cancer and the correlation with clinicopathologic characteristic and prognosisIntroductionMany studies indicated that over-expression of cyclooxygenase-2(COX-2) was correlated with occurrence and development of many solid cancers(such as colorectal cancer, lung cancer, ect.). Over-expression of COX-2 protein in cancers was likely to occur via several different mechanisms involving complex signaling pathways. Over-expression had been associated with a poor clinical outcome. High incidence rates of over-expression COX-2 was also found in breast cancer patients. At present although we used selective inhibitor to improve prognosis, the angiomyocardiac adverse effect constrained the serviceable range. Furthermore the curative effect was not sure. So if we could inhibit expression of upstream control factors, it would be better for downregulation of COX-2. But now the mechanism of regulation was not clear. The thioedoxin family played an important role on the oxidation-reduction reaction, the metabolism of nucleic acid and the growth of cell. An abroad document indicated TRX-1 up-regulated expression COX-2 via HIF-1αin non-small cell lung cancer cells in vitro. Another document reported that over-expression TRX-1 was correlated with prognosis in colorectal cancer patients. In this study, we detected the expression of COX-2 and TRX-1 using immunohistochemical SP methods to investigate the relationship between them, as well as to correlate with clinicopathological parameters, Disease Free Survival Time in human breast cancer. Methods1. Patients122 breast cancer patients underwent completely resected who were enrolled from the first affiliated hospital, China Medical University(CMU). None of them underwent chemotheray or radiotherapy before operation. But 22 patients(18.0％) received chemotherapy of taxane after operation. All the cases were female with median age of 50.2 years. Among of them there were 32 cases(26.2％) of the tumout size≤2cm, 66 cases(54.1％)of lymph node metastasis, 88 cases(72.1％) of infitrating ductal carcinoma and 34(27.9％) advanced cases. 32 patients(26.2％)occurrenced relapse and matastasis. The Diseas Free Survival Time(DFS)ranged from 6 to 59 months. Normal breast tissue of 15 cases were selected as controls.2. MaterialsThe polyclonal antibodies against TRX-1 were purchased from Cell Signaling biotechnology, Inc; the polyclonal antibodies against COX-2 and SP immunohistochemical kits and DAB substrate solution were purchased from Fujian Maixin Biotechology, Inc.3. MethodsImmunohistochemical staining SP method was performed. All slides were treated with 0.01 mol/L citric acid buffer(PH=6.0) and recovered the antigen activity by autoclave.4. AssessmentOutcome definition of TRX-1 and COX-2 were stained palm yellow particle in cytosol, positive cells≤10％as negative,＞10％as positive at high magnification(400×).5. Statistical analysisStatistical methodsχ~2 test, Spearman rank correlation test and Kaplan-Meier survival analysis were performed by SPSS software version 13.0. Results1. Expression of COX-2 and TRX-1 in breast cancer patientsPositive expression of COX-2 and TRX-1 in 122 cases was 58.2％and 54.1％, respectively, however, was negative in 15 cases normal breast tissue. Expression of TRX-1 was positive correlated with COX-2(1=0.286, P=0.001).2. The relationship between expression of COX-2 and TRX-1 and clinicopathologic parameters in breast cancer patientsExpression of COX-2 was positive correlated with tumour size(r=0.192, P=0.034), lymph node metastasis(r=0.186, P=0.040), clinical stage(r=0.193, P=0.033) and negative correlated with expression of ER(r=-0.183, P=0.044), HER-2(r=0.194, P=0.032), pathology grading(r=-0.197, P=0.029) but not correlated whith expression of PR(r=-0.175, P=0.054). Expression of TRX-1 was positive correlated with tumour size(r=0.236, P=0.009) and clinical stage(r=0.206, P=0.023), but not correlated with lymph node metastasis(r=0.175, P=0.054), expression of ER(r=-0.033, P=0.722) and PR(r=-0.043, P=0.635) and HER-2(r=-0.041, P=0.654), pathology grading(r=-0.107, P=0.242).3. The relationship between expression of COX-2 and TRX-1 and Disease Free Survival Time in breast cancer patientsCOX-2 expression was negative correlated with DFS by Kaplan-Meier survival analysis(P=0.046). The recurrence rate of patients with positive COX-2 expression(31.0％) was significantly higher than that with negative COX-2 expression(19.6％). TRX-1 expression was not correlated with DFS by Kaplan-Meier survival analysis(P=0.056). But the value P of COX-2 combined with TRX-1 expression was more significant(P=0.042).4. Multiple analysis outcome among clinicopathologic parameters effecting brest cancer patients prognosisExpression of ER and PR and HER-2, tumour size, lymph node metastasis, pathology grading were unique factors effecting prognosis. Expression of COX-2 or combined with TRX-1 was not unique factor.5. The relationship between expression of TRX-1 and resistance to taxane22 cases in taxane group, among them the relapse rate of positive expression TRX-1 was 40.0% (4/10) and negative expression was 25.0% (3/12). The value of P was 0.652.Conclusion1. The expression rate of COX-2 and TRX-1 was higher in human breast cancer and the correlation between them was positive. Expression of COX-2 was positive correlated with tumour size, clinical stage, lymph node metastasis, and negative correlated with expression of ER and HER-2, pathology grading, but not correlated whith expression of PR. Expression of TRX-1 was positive correlated with tumour size and clinical stage, but not correlated with lymph node metastasis, expression of ER and PR and HER-2, pathology grading.2. The patients with COX-2 negative expression had longer DFS than COX-2 positive expression. Detection of COX-2 in bresat cancer tissues might be significant in prognosis prediction. The recurrence rate of TRX-1 positive expression was not significantly different compared with that of TRX-1 negative expression, but the value of P approached 0.05. So detection of TRX-1 in bresat cancer tissues might be significant in prognosis prediction if adding enrolled samples or prolonging following-up time. The multiple analysis outcome indicated that despite expression of COX-2 or COX-2 combined with TRX-1 was not unique factor for judging prognosis of brest cancer patients. Lymph node metastasis, expression of ER and PR and HER-2, pathology grading were respectively unique factors. The value P of tumour size approached 0.05 and possibly correlated with prognosis.3. Over-expression TRX-1 was not correlated with resistance to taxane. Although the value of P was not significant, it was difficult to explain using such a alone factor. So we needed adding samples and prolonging following-up time.4. Immunohistochemical staining SP using tissue array possessed merits of higher output, lower cost, smaller errors and so on.