Abstract
OBJECTIVE To investigate the expressions of cyclooxygenase 2 (COX-2) and human epidermal growth factor receptor-2 (HER-2) in non-small cell lung cancer (NSCLC) and their clinical significance in identifying the progression and prognosis of the NSCLC patients.
METHODS Immunohistochemical indirect method was used to detect the expressions of the COX-2 and HER-2 protein in 54 NSCLC specimens, 16 paraneoplastic specimens, and 10 normal tissue specimens.
RESULTS The positive rates of COX-2 and HER-2 protein expressions were respectively 75.9% and 40.7% in the NSCLC specimens, 25% and 12.5% in the paraneoplastic specimens, and 0 in the normal tissue. The COX-2 protein expression in lung cancer (LC) was not only related to the smoking habit of the patients and histological grades of LC, but also to the TNM stages, and lymphatic metastasis (P < 0.05). HER-2 protein expression closely correlated to the pathologic types, histological grades, TNM stages, and lymphatic metastasis (P < 0.05). The result of univariate analysis showed that all the histological grades, TNM stages, lymphatic metastasis, and expressions of COX-2/HER-2 correlated to the prognosis of NSCLC patients (mean of P value < 0.01). The multivariate survival analysis indicated that there were significant differences in comparison of the survival time between the COX-2 (++/+++) /HER-2 (++/+++) and the COX-2 (-/+)/HER-2 (-/+) groups (P < 0.001), suggesting the COX-2/HER-2 was a negative prognostic factor.
CONCLUSION COX-2 and HER-2 are valuable in identifying the progression of NSCLC and predicting the prognosis of NSCLC patients. COX-2 and HER-2 are useful for judging the NSCLC patient’s condition, and are of great value to the decision of NSCLC prognosis.
keywords
Introduction
Cyclooxygenase 2 (COX-2) is the key rate-limiting enzyme in the synthesis of prostaglandins (PGs). The roles of the mediators of inflammation and the oncogene are to induce the up-expression of the COX-2 and to mediate the synthesis of PGs. Human epidermal growth factor receptor-2 (HER-2) is an oncogene-coded transmembrane protein. The correlation between the HER-2 and COX-2 expressions has been reported to be involved in several kinds of tumors, such as breast cancer, ovarian cancer, and prostatic carcinoma etc. however, reports related to the relationship with lung cancer (LC) is rare and remains controversial. For this reason, in our study, the expressions of the enzyme and protein in 54 LC cases were determined using the immunohistochemical assay, and the clinical significance of the COX-2 and HER-2 and the mutual relationship between the 2 are discussed.
Materials and Methods
Specimens
Surgical specimens were taken from 54 non-small cell lung cancer (NSCLC) patients who were admitted to and treated in our hospital during the period from January 2004 to January 2005. Of these patients, 38 were male and 16 female. Their age ranged from 35 to 72 years, with a mean of 51. It was found that among the patients, 32 had a smoking history and the other 22 did not. Based on the standard of 2003 AJCC PTNM staging (6th Ed), 14 of the 54 cases were classified in stage I, 25 in stage II, and 15 in stage IIIa. In the 54 NSCLC cases, 38 had lymphatic metastasis and the other 16 did not. Based on the histological grading, 13 were diagnosed as grade I, 30 as grade II, and 11 as grade III. Specimen fixation was conducted in 4% formalin, a neutral buffering, after the pathological sampling, and after that the specimens were cut into 4-μm pieces and then beded in paraffin. H&E staining procedure was carried out followed by observation under a microscope. Each pathologic diagnosis was confirmed after review by 2 pathologists.
Reagents and methods
The rat-antihuman HER-2 monoclonal antibody, ratantihuman COX-2 monoclonal antibody (ZM20035), kit (PV26002) for immunohistochemical indirect method and DAB substrate kit were all purchased from Beijing Zhongshan Golden Bridge Biotechnology Co. The immunohistochemical indirect method (non-biotin) was used, and the technical manipulation was performed strictly based on the instructions of the manufacturer. DAB chromogenesis, hematoxylin staining and observation under microscopy were carried out to analyze the results of the staining. In the control, the first antibody was replaced by PBS, while the other procedures remained unchanged.
Analysis of the results
COX-2 scoring: based on the Robert’s classification[1], a brownish-yellow staining in the cytoplasm or on the nuclear envelope showed a positive result; based on the scoring of the stain of COX-2 protein in the cells, a score of 0 indicating negative, 1 indicating a range from 1% to 10% positive, 2 indicating from 11% to 50% positive, 3 indicating from 51% to 80% positive, and 4 indicating from 81% to 100% positive. In regards to the scoring of staining intensity, 0 meant negative, 1 weak, 2 moderate, and 3 strong. The cross product of the above-mentioned scores was the final score, i.e. score of 0 was defined as (-), a score from 1 to 4 defined as (+), score from 5 to 8 as (++), score from 9 to 12 as (+++). Those with a score of < 5 were classified into the low expression group, and those with a score of ≥ 5 were classified into the over expression group. The scoring of HER-2 was as follows: based on the scoring guidelines of HercepTest and the current criteria of clinical laboratories in the European and American countries[2], the positive results presented a brownish-yellow staining on the cancer cell membrane, while the stain of the cytoplasm in either the normal pulmonary epithelial cells or the tumor tissue was not specific. “+” indicated that the positive expression region was > 10%, and the chromogenesis of cell membrane was discontinuous. “++” meant positive expression region was > 10%, with a continuous and moderately strong positive chromogenesis on the cell membrane. “+++” indicated that positive expression region was > l0%, and the chromogenesis of cell membrane was continuously and strongly positive. If the condition of “+”, “++” and “+++” was found occurring in the same section at the same time, the scoring was conducted based on a higher grade. Negative stain showed a result without positive expression or a non-specific staining. If the difference between the value of the HER-2 staining in LC and that in paraneoplastic non-cancerous epithelium exceeded or equaled to“++”, it could be considered as an over expression of HER-2.
Statistical analysis
Stata 7.0 was used for statistical treatment. The χ2 test and Fisher test were employed to compare the relationship between the expression of COX-2 and HER-2 and the clinicopathological features of the NSCLC patients. Kaplan-Meier curve was used for the survival analysis in these cases, and the log-rank univariate analysis and multivariate analysis of the Cox regression model were carried out for investigating the correlation between the expression of COX-2 and HER-2 and prognosis in LC patients.
Results
Expression of COX-2 and HER-2 protein in pulmonary carcinoma
The positive result of COX-2 presented a brownish-yellow stain in the cytoplasm or on the nuclear membrane, and that of HER-2 showed a brownish-yellow stain situated at the cancer cell membrane (Fig. 1). Groups with a COX-2 (++/+++) and a HER-2 (++/+++) were regarded as the positive group, i.e. the over expression group, and those with a COX-2 (-/+) / HER-2 (-/+) were regarded as the negative group, i.e. the low expression group.
Positive expressions of the COX-2 and HER-2 protein in adenocarcinoma (ACA) of the lung and squamous carcinoma (SqCa) of the lung. A, COX-2 protein (ACA); B, HER-2 protein (ACA); C, COX-2 protein (SqCa); D, HER-2 protein (SqCa).
The results are shown in Table 1. The positive expression rates of COX-2 and HER-2 protein were 75.9% and 40.7% in NSCLC, respectively.
Expression of COX-2 and HER-2 protein in various pulmonary tissues.
Relationship between expression of COX-2 & HER-2 and clinicopathologic features of LC
The expression of COX-2 protein closely correlated with the factors, such as smoking habit of the patients, histological grades, TNM stages, and lymphatic metastasis (P < 0.05), while the expression of HER-2 protein was closely related to pathological types, histological grades, TNM stages, and lymphatic metastasis in the LC patients (P < 0.05), but had no relationship with smoking (P > 0.05), see Table 2.
Relationship between the expression of COX-2 & HER-2 protein and clinicopathologic features of LC.
Relationship between expressions of COX-2 and HER-2 protein in LC
Of the 54 NSCLC patients, 22 were found COX-2 (++/+++)/HER-2 (++/+++), 19 found COX-2 (++/+++)/HER-2 (-/+), and 13 found COX-2 (-/+)/HER-2 (-/+). COX-2 (-/+)/HER-2 (++/+++) was not found in any patients. The χ2 test was used to determine the correlation between the values in the 2 types of proteins. The result of the test showed that there was a significantly positive correlation between the 2 protenis (P = 0.001, r = 0.433).
Relationship between clinical parameters and prognosis of LC
Results of univariate analysis
The factors including histological grades, TNM stages, lymphatic metastasis, and COX-2/HER-2 were all related to the prognosis of NSCLC patients (mean value of P < 0.01). The NSCLC patients with the histological grade II-III, TNM stage IIIa, lymphatic metastasis, and COX-2 (++/+++) /HER-2 (++/+++) had a short survival time. No correlations were found between each of the factors, such as the age, pathological types, smoking, and the prognosis of NSCLC patients (mean value of P > 0.05) (Table 3).
Univariate analysis of the relationship between the clinical parameters and prognosis of NSCLC patients.
Results of multivariate analysis
The significant variables screened by the univariate analysis were brought into the Cox regression model for the multivariate survival analysis. After precluding the impact factors, such as histological grades, TNM stages and lymphatic metastasis upon the prognosis of NSCLC patients, the results of the analysis showed that there were significant differences in comparison of survival time between the COX-2 (++/+++)/HER-2 (++/+++) and the COX-2 (-/+)/HER-2 (-/+) groups (P < 0.001). The survival time was short in the NSCLC patients with COX-2 (++/+++)/HER-2 (++/+++).
Relationship between COX-2/HER-2 and postoperative survival time
The postoperative survival time in the groups with COX-2 (++/+++) /HER-2(++/+++) and COX-2 (-/+)/HER-2 (-/+) (Fig. 2) suggested that the COX-2 (++/+++)/HER-2 (++/+++) was a negative prognostic factor for the NSCLC patients.
Survival curves in the groups with an over and a low expressions of COX-2/HER-2.
Discussion
Several mediators of inflammation, growth factors, and oncogenes may activate the transcription of COX-2 and its protein expression[3]. There is a correlation between COX-2 and the occurence and progress of the rat squamous cell lung cancer, which might be one of the major ties linking the inflammation, apoptosis, and tumor development[4]. It was shown in our study, after the immunohistochemical assay of LC, that the expression of COX-2 was increased by turns in the normal tissue, paraneoplstic tissue, and LC. There were statistically significant differences among the expression of COX-2 in the 3 tissues, which indicated that COX-2 might be an early event in the onset of LC. This conclusion has also been accepted by most scholars[5,6]. The findings also reveal that the over expression of COX-2 correlates to smoking, histological grades, TNM stages, and lymphatic metastasis. The rate of COX-2 expression in the patients with a smoking history, high histological grades, high TNM stages, and lymphatic metastasis, is significantly higher than that in the patients with low histological grades, low TNM stages, and without a smoking history or lymphatic metastasis (mean value of P < 0.05). This indicates that COX-2 may play a role in facilitating cells transforming into malignant cells and in enhancing tumor invasion, and that it can be involved in the course of LC progression. The expression of COX-2 protein significantly correlates with the smoking habit of the NSCLC patients, suggesting that smoking closely correlates with the occurrence of NSCLC. This mechanism might be in relation to the fact that COX-2 can synergistically facilitate the cell proliferation together with the carcinogen of tobacco[7]. Laga et al.[5]confirmed in a study with 259 NSCLC cases that the increase of COX-2 expression predicted an unfavourable prognosis for NSCLC patients.
HER-2 is a protein tyrosine kinase-active transmembrane glycoprotein. It can transfer the signal of proliferation after the automatic phosphorylation, thus regulating the growth, proliferation, and differentiation of the cells[8-11]. This suggests that HER-2 expression in the cell membrane is the main modality of the HER-2 function to transfer the signal of cell proliferation, which is in accord with the report of Tan[12-15]. It was shown in our study that the HER-2 expression increases in order in the normal tissue, paraneoplastic tissue, and LC, with significant differences in the expression of COX-2 among the 3 tissues. The results in our study also revealed that the expression of HER-2 protein closely correlates with the histological grades, pathological types, TNM stages, and lymphatic metastasis of the LC patients. A significantly higher rate of HER-2 expression presented in the cases with ACA, high histological grades, high TNM stages, and with lymphatic metastasis compared with that of the cases with SqCa, low histological grades, low TNM stages, and without lymphatic metastasis (P < 0.05). This indicates that HER-2 may also correlate with the onset, progress and prognosis of LC.
In our study, the univariate analysis showed that histological grades, TNM stages, and lymphatic metastasis and COX-2/HER-2 correlate to the prognosis of NSCLC patients (P < 0.01). The survival time was short in the NSCLC patients with grade II-III, stage-IIIa, and with presence of lymphatic metastasis and COX-2 (++/+++)/HER-2 (++/+++). The results of multivariate analysis showed, after eliminating the effects of histological grades, TNM stages, and lymphatic metastasis on the prognosis of NSCLC patients, that there were significant differences in comparison of survival time between the cases with COX-2 (++/+++) /HER-2 (++/+++) and the cases with COX-2 (-/+) /HER-2 (-/+) (P < 0.001). The survival time was short in the NSCLC patients with COX-2 (++/+++)/HER-2 (++/+++).
Several oncogenes can up-regulate the expression of COX-2. The reports of several studies on breast cancer[16-18] demonstrated that HER-2 played a decisive role in the COX-2 expression, i.e. in HER-2-positive breast cancer specimens, both the rate and amount of the COX-2 expression obviously raised. In the 54 pulmonary ACA patients in our study, there was a positive correlation between the expressions of COX-2 and HER-2 protein (P = 0.001, r = 0.433), which indicated that COX-2 and HER-2 may have synergistic action in facilitating cancer development and progression, as well as invasion and metastasis of the human NSCLC. Prostaglandin E, a COX-2 product, and HER-2 can all stimulate the synthesis of the vascular endothelial cell growth factors; induce the immunological tolerance; and inhibit the apoptosis, therefore, increase the capability of tumor cell infiltration, which can be regarded as potential targets to be attacked in the prevention and systemic therapy for NSCLC[16-20]. Therefore, combining inhibitors of COX-2 and HER-2 into the treatment plan for NSCLC patients may improve the therapeutic effects and may achieve significant clinical outcomes.
Conflict of interest statement
No potential conflicts of interest were disclosed.
- Received April 4, 2010.
- Accepted June 10, 2010.
- Copyright © 2010 by Tianjin Medical University Cancer Institute & Hospital and Springer
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.