Overexpression of CCR7 mRNA in nonsmall cell lung cancer: Correlation with lymph node metastasis



Tumor cell migration into the lymph nodes is an important aspect of cancer, yet little is known about the processes involved. CCR7 has been shown to play an important role in tumor cell migration and lymph node metastasis. The aim of our study was to evaluate the expression of CCR7 in nonsmall cell lung cancers (NSCLCs). We investigated CCR7 expression in 71 patients with NSCLC who underwent curative tumor resection by means of reverse transcription-polymerase chain reaction (RT-PCR) and immunohistochemistry. CCR7 mRNA was expressed in 45 cases (63.3%). The CCR7 mRNA expression was significantly associated with lymph node metastasis, stage, lymphatic invasion, CCR7 protein and CXCR4 protein (p = 0.0001 for lymph node metastasis; p < 0.0001 for stage; p = 0.0454 for lymphatic invasion; p < 0.0001 for CCR7 protein; p = 0.0013 for CXCR4 protein). Twenty-six (57.8%) of 45 cases with CCR7 mRNA expression in their cancer tissues were node-positive, whereas only 3 (11.5%) of 26 cases without CCR7 mRNA expression were node-positive. Furthermore, expression of CCR7 mRNA was shown to be an independent predictor of lymph node metastasis by multivariate analysis (p = 0.0117). Our study demonstrates that CCR7 might be related to the development of lymph node metastasis in NSCLC. The expression of CCR7 mRNA could open up a new window for the diagnostic staging and treatment of NSCLC. © 2003 Wiley-Liss, Inc.

It has been reported that tumor cell migration and metastasis share many similarities with leukocyte trafficking, which is critically regulated by chemokines and their receptors.1 Of 17 different chemokine receptors, the chemokine receptors CXCR4 and CCR7 are reported to be highly expressed in malignant breast tumors.2 Our knowledge of the molecular determinants of lymph node metastasis is clearly limited, although a small number of gene products have been shown to increase lymph node metastasis.3 Expression of the chemokine receptor gene, CCR7, has been reported to enhance murine malignant tumors to metastasize to regional lymph nodes,4 and RT-PCR analysis indicates elevated levels of CCR7 expression and a correlation with lymph node metastasis in gastric cancer.5 However, there are no previous reports studying the clinical significance of CCR7 in nonsmall cell lung cancer (NSCLC). Thus, we investigated CCR7 expression in a series of 71 cases of curatively resected NSCLC by means of RT-PCR and immunohistochemical assays to evaluate its clinical significance. We also determined whether expression of CCR7 mRNA correlates with immunohistochemical assay of CXCR4.


Tissue specimens

Tumor tissue was collected from 71 patients with NSCLC who underwent curative surgery from 1992 to 1996 at the Department of Surgery, Teikyo University School of Medicine. Patients who died within 1 month after surgery and patients with a past history of another cancer were excluded from the study. Patients were also excluded if they had received neoadjuvant chemotherapy or radiotherapy. Freshly removed pulmonary cancer tissues for RNA extraction were immediately frozen in liquid nitrogen and stored at −80°; until further use. In 5 of the cases, adjacent normal pulmonary material from the same patient was also used. Tissue samples to be used for immunohistochemistry and H&E staining were fixed in formalin and paraffin embedded.

RT-PCR analysis

Total RNA was purified from fresh soft tissues by the acid guanidinium-thiocyanate procedure.6 Total RNA (5 μg) was used for cDNA synthesis, and the first-standard cDNA solution was then used for PCR, with primers designed to amplify a 530 bp sequence as described previously7 (sense primer sequence: 5′-TCCTTCTCATCAGCAAGCTGT-3′; antisense primer sequence: 5′-GAGGCAGCCCAGGTCCTTGAAG-3′). Thirty-two 1 min cycles of denaturation at 96°C, 1 min of annealing at 58°C and 1 min extension at 72°C were then performed. S14 cDNA amplification using the same temperature profile for 30 cycles served as the internal control.8 The sense and antisense primers for S14 cDNA amplification were 5′-GGCAGACCGAGATGAATCCTC-3′ and 5′-CAGGTCCAGGGGTCTTGGTCC-3′. Control amplification was then performed adding either not-retrotranscribed RNA or no-DNA, no-RNA to the PCR tube, but no specific bands were observed. All specimens were analyzed at least twice. The amplified DNA samples were electrophoresed on 1% agarose gels and photographed with a Polaroid camera (Fig. 1).

Figure 1.

Agarose gel electrophoresis of RT-PCR-amplified 530 bp CCR7 cDNA and 143 bp S14 DNA as internal PCR control. Lane 1, size marker; lane 2, normal pulmonary tissue with negative CCR7 expression; lane 3, adenocarcinoma with positive CCR7 expression; lane 4, adenocarcinoma with negative CCR7 expression; lane 5, squamous cell carcinoma with positive CCR7 expression; lane 6, squamous carcinoma with negative CCR7 expression; lane 7, large cell carcinoma with high CCR7 expression; lane 8, adenocarcinoma with high CCR7 expression.

Immunohistochemical assays

To compare with the results on CCR7 mRNA expression obtained by RT-PCR, immunohistochemical studies of CCR7 and CXCR4 were performed. CCR7 and CXCR4 specimens were immunostained using formalin-fixed, paraffin-embedded tissues. The sections were immersed for 30 min in 0.3% H2O2 in absolute methanol and then treated with 10% normal rabbit serum. Overnight incubation with the anti-human CCR7 monoclonal antibody at 1:100 dilution (PharMingen, San Diego, CA) and the anti-human CXCR-4 monoclonal antibody at 1:100 dilution (R & D Systems, Minneapolis, MN) was followed by incubation with biotinylated rabbit anti-mouse IgG and an avidin-biotin-peroxidase reaction that used 3, 3′-diaminobenzidine tetrahydrochloride in the presence of 0.05% H2O2. Negative control sections were treated using nonimmunized mouse IgG as the primary antibody. The degree of monoclonal antibody reactivity within individual tissue sections was considered positive if unequivocal staining of carcinoma cells was seen in >10% of tumor cells. When <10% were stained, the samples were classified as negative. Two independent pathologists observed all the specimens.

Clinicopathologic data and statistical analysis

The clinical variables considered in our study, including the stage of disease and various pathologic factors, are shown in Table I. The relation between CCR7 mRNA expression and the clinicopathologic factors were analyzed using the χ2 test and Student's t-test. A stepwise logistic regression model was used for the multivariate analysis using the StatView software package (Abaracus Concepts, Berkeley, CA). All tests were considered significant at p < 0.05.

Table I. Correlation between CCR7 mRNA Expression and Various Clinicopathologic Factors in Patients with Lung Cancer
VariableCCR7 mRNAP-value
Negative (n = 26)Positive (n = 45)
Age69.2 ± 10.064.8 ± 9.40.1637
Tumor status   
Nodal status   
Lymphatic invasion   
Vascular invasion   
 Squamous cell carcinoma10180.9816
 Large cell carcinoma12 
CCR7 protein   
CXCR4 protein   


Expression of CCR7 mRNA and clinicopathologic data

A total of 71 NSCLC specimens were examined for expression of CCR7 mRNA, and 45 carcinomas(63%) were classified as CCR7 mRNA positive, and 26 (37%) carcinomas were classified as CCR7 mRNA negative. Five adjacent normal pulmonary materials were classified as CCR7 mRNA negative. As shown in Table I, there were no significant differences between CCR7 mRNA expression status and each clinical factor of age, sex, vascular invasion, T factor and histology. However, the expression of CCR7 mRNA was significantly associated with lymph node metastasis, stage, and lymphatic invasion (p = 0.0001 for lymph node metastasis; p < 0.0001 for stage; p = 0.0454 for lymphatic invasion). Twenty-six (57.8%) of 45 cases with CCR7 mRNA expression in cancer tissues were node-positive, whereas only 3 (11.5%) of 26 cases without CCR7 mRNA expression were node-positive.

Staining of the CXCR4 and CCR7 protein was identified in the cytoplasm and/or cell membrane of cancer cells and was not detected in the normal lung cells taken from noncancerous regions in lung cancer tissue. Weak staining for the CXCR4 and CCR7 proteins was observed in a majority of the infiltrating inflammatory cells in the specimens. The CXCR4 and CCR7 proteins were detected in the endothelial cells of relatively large vessels around tumor lesions (Fig. 2). In most cases of CCR7 mRNA expression, immunostaining was intense and seen uniformly in >10% of the tumor cells (Fig. 3). As shown in Table I, immunohistochemical results for CCR7 protein agreed well with those for the RT-PCR of CCR7, and immunohistochemical results for CXCR4 also agreed well with those for the RT-PCR of CCR7 (p < 0.0001 for CCR7 protein; p = 0.0173 for CXCR4 protein). In cases of discrepancies in the CCR7 assays, the results of RT-PCR were used for specimen classification.

Figure 2.

Immunohistochemical expression of CCR7 protein in a relatively large vessel (original magnification 240×) Scale bar = 10 μm.

Figure 3.

Immunohistochemical expression of CCR7 protein (a,a′) and CXCR4 protein (b) in pulmonary adenocarcinoma. CCR7 staining and CXCR4 were detected in the membrane and cytoplasm of cancer cells. (a,b, original magnification 40×; a′, original magnification 160×).Scale bars = 100 μm (a,b) and 10 μm (a′).

The independent predictors of lymph node metastasis were determined by stepwise logistic regression analysis. As shown in Table II, T factor, lymphocytic invasion, vascular invasion, CCR7 mRNA and CXCR4 protein were correlated with lymph node metastasis by univariate analysis. According to the multivariate analysis of these variables, lymphatic invasion tended to correlate with lymph node metastasis without significance (p = 0.0507), and CCR7 mRNA was proven to be the only independent predictor of lymph node metastasis (p = 0.0177).

Table II. Statistical Associations between Lymph Node Metastasis and Examined Variables
FactorsLymph node metastasisUnivariate p-valueMultivariate p-value
Negative (n = 42)Positive (n = 29)
T factor    
Lymphatic invasion    
Vascular invasion    
 Squamous cell carcinoma1711  
 Large cell carcinoma120.64800.6850
CCR7 mRNA    
CXCR4 protein    


NSCLC is a leading cause of morbidity and mortality among both men and women in the U.S.9 Despite major advances in cancer treatment in the past 2 decades, the prognosis of patients with lung cancer has improved only minimally.9 Lymph node metastasis is one of the most powerful prognostic markers for resected NSCLC.10 With a view to clinical usefulness, early detection techniques that clearly predict lymph node metastasis before an operation may improve disease treatment planning. Information on important predictive factors for lymph node metastasis is presently obtained by studying resected specimens after operation. In our study also, only resected samples were available. But if it proves to be a useful marker in the resected samples, the methodology can be applied to preoperative bronchofiberscopic biopsy samples. We would then be able to get relevant prognostic information on lymph node metastasis before surgery. The advent and development of PCR, a highly sensitive and efficient method of amplifying specific DNA segments present at low concentrations, provides an alternative approach for estimating relative gene expression in a small amount of tissue.11 We thus focused on RT-PCR expression of CCR7 mRNA in resected specimens as a possible useful molecular predictor of lymph node metastasis, and this is the first study to our knowledge on CCR7 mRNA expression determined by RT-PCR in NSCLC.

Although the mechanisms of CCR7 governing entry into the lymph nodes are not yet as well understood, the migration of cells to regional lymph nodes via lymphatic vessels resembles the normal migration of antigen-presenting cells such as dendritic cells during the course of inflammation.1 Tumor cells use chemokine-mediated mechanisms such as those regulating leukocyte trafficking during the process of metastasis.1 Cancerous epithelial cells undergo clonal proliferation, invade local tissue, induce angiogenesis and express CCR7 on their surface. Two constitutively expressed chemokines, CCL21 and CCL19, share a common receptor, CCR7. CCL21 is expressed in the high endothelial venules of lymph nodes and in the T-cell zone of lymph nodes. CCL19 is expressed predominantly by stromal cells within the T-cell zones of lymph nodes.2 Both CCL19 and CCL21 had shown the most abundant expression in lymph nodes. Thus, the abundant expression of CCL19 and CCL21 in lymph nodes makes such expression a likely candidate for attracting CCR7-positive tumor cells and the subsequent migration of these cancer cells to the regional lymph nodes. Previous studies12, 13 showed that CCL21 is produced by lymphatic endothelial cells as well as lymph nodes. The migration of malignant cells from peripheral tissue into lymphatics may be an active process mediated by CCL21. CCL21 and CCR7 participate in the emigration of cancer cells from peripheral tissue to lymph nodes via lymphatics. CCR7 staining detected endothelial cells in the relative large vessels around the tumorous lesions observed in our study. This finding may suggest that CCR7 is essential for organ vascularization.

CCR7 mRNA was expressed at higher levels in cancer tissue, and it was not expressed in the adjacent normal lung tissues. CCR7 mRNA was detected in 63% of tumor samples in our patient population with NSCLC. We demonstrated that the expression of CCR7 mRNA correlated with histologic lymph node metastasis, stage and lymphatic invasion. This suggested that the tumors with an overexpression of CCR7 mRNA might have more potential to recur after surgery. There were 13 CCR7-positive patients in stage IA/IB, which is a state of the absence of nodal metastasis. These patients may have been on the cusp of lymph node metastasis, or they may have had micro-lymph node metastasis that was not found by our routine histologic examination.

We also investigated the correlation between CCR7 mRNA and CCR7 protein and CXCR4 protein expression using the immunohistochemical staining of the monoclonal antibodies; CCR7 mRNA was well associated with CCR7 protein, and 85.9% of cases had no discrepancy. A significant correlation was also found between CCR7 mRNA and CXCR4 protein in NSCLC. Multivariate analysis demonstrated that CCR7 was the only independent factor for lymph node metastasis. Although we did not investigate CXCR4 mRNA expression in our study, CXCR4 protein was not picked up as an independent factor for lymph node metastasis by multivariate analysis. Muller et al.2 reported that 2 receptors, CCR7 and CXCR4, may be involved in breast cancer metastasis and that neutralizing the interaction of CXCR4 may be useful in interfering with tumor progression and metastasis in tumor patients. But in our study, no significant difference in the rate of patient survival was detected between those patients whose tumors were CXCR4 protein negative or those whose tumors were CXCR4 positive (data not shown). CCR7 may have a more critical role in determining the metastatic destination of tumor cells than CXCR4. Our results indicate that the CCR7 expressed in cancer cells may be relevant to the process by which cancer cells preferentially migrate to the lymph nodes and subsequently cause lymph node metastasis. Lymphatic invasion tended to be an independent predictor for lymph node metastasis without significance in multivariate analysis. The status of lymphatic invasion is diagnosed by pathologic examination of the resected primary tumors, and the status of CCR7 mRNA is more useful than the status of lymphatic invasion as a predictor for lymph node metastasis in NSCLC. The expression status of CCR7 mRNA can be performed by RT-PCR using preoperative biopsy samples, so the study of CCR7 mRNA could become a useful tool for preoperative evaluation of lymph node metastasis.

If adequate information about lymph node metastasis was obtained before surgery, one could add neoadjuvant chemotherapy before surgery or avoid lymph node dissection. Most NSCLC patients without an overexpression of CCR7 could be treated by lobectomy only and without lymph node dissection or, at most, a sentinel node biopsy only, in spite of undergoing a complete mediastinal dissection. According to our findings, many early-staged NSCLC patients may be curable with much less invasive surgery. Treatment with anti-CXCR4 antibodies resulted in a considerable reduction of tumor spread in an animal model for breast cancer metastasis.12 In the future, small molecule antagonists of chemokine receptors such as CCR7 may be used to interfere with lymph node metastasis in tumor patients.

In conclusion, our study suggests that CCR7 may be related to the development of lymph node metastasis. Thus, the utility of expression of CCR7 mRNA could open up a new window for the diagnosis of lymph node metastasis and the treatment of NSCLC.