To better evaluate the prognostic value of EGFR in NSCLC, the detection of activated EGFR (e.g. EGFR phosphorylation) or combined detection with other molecular markers BAY 80-6946 should be used [33]. In our study the positive rate of COX-2 protein expression was 90% for NSCLC tumors and

was significantly higher than that for normal lung (0%) and paracancerous tissue (14.3%). Therefore, it suggested that COX-2 might participate in oncogenesis of NSCLC. Similar COX-2 positivity rates ranging from 54 to 100% have been reported for NSCLC tumors as measured by immunohistochemistry [34]. In our study it was found that COX-2 protein expression in adenocarcinoma was significantly higher than that in squamous carcinoma (p = 0.022), which was consistent to previous findings of other study [21]. This might provide basis for applying COX-2 inhibitor in adenocarinoma BAY 11-7082 concentration patients receiving tyrosine kinase inhibitor (TKIs), as COX-2 inhibitor offered synergistic antitumor effects

with TKI [21]. Although COX-2 expression was also found higher in female patients, patients with ages≤60 years, non-smokers, moderate and well differentiated tumors, nodal metastasis, and in stages III-IV, the difference had no statistical significance. Studies examining the relationship between COX-2 tumor expression and survival among lung cancer patients were inconsistent, with reports of an inverse relationship with survival [35], no association [36], or a direct association with survival [37]. In our study, there was no correlation between COX-2 expression and patient’s overall survival. However, unlike check details some previously reported studies which showed that COX-2 expression was most consistently associated with poorer survival among stage I and II NSCLC patients [38, 39], our study neither showed the correlation of COX-2 expression with patient’s survival nor prognostic value in early stage adenocarcinma [21]. This might

be due to the small sample size in our study. No correlation was found between EGFR expression and COX-2 in our study, though both EGFR and COX-2 involve in the carcinogenesis and progression of NSCLC both individually and, as recently suggested, synergistically [40]. A number of in vitro studies postulated a link between EGFR activation and Farnesyltransferase subsequent COX-2 up-regulation. EGFR activation could induce COX-2 expression via the ras/raf MAPK pathway [3]. On the other hand, COX-2 could induce the activation and expression of EGFR. The lack of correlation of EGFR and COX-2 expression in our study implied that the expression of these 2 proteins might be controlled by independent mechanisms. As suggested by a recent study that examined the expression of p-EGFR, EGFR, and COX-2 by immunohistochemistry in surgically-resected stage I/II NSCLC, pathways other than EGFR activation may influence COX-2 overexpression[38].