ANTICANCER RESEARCH 36 : 4013-4018 (2016)

Chemokine and Receptor Profiles in Metastatic Salivary Adenoid Cystic Carcinoma ASHLEY C. MAYS, XIN FENG, JAMES D. BROWNE and CHRISTOPHER A. SULLIVAN

Department of Otolaryngology, Wake Forest School of Medicine, Winston Salem, NC, U.S.A.

Abstract. Aim: To characterize the chemokine pattern in distant (2-5). According to Ko et al. , 75% of metastatic salivary adenoid cystic carcinoma (SACC). patients with initial nodal involvement eventually developed Materials and Methods: Real-time polymerase chain distant metastasis. Patients with metastasis have a poor reaction (RT-PCR) was used to compare chemokine and prognosis (6). expression in two SACC cell lines: The development of distant metastatic disease is the chief SACC-83 and SACC-LM (lung metastasis). and cause for mortality (7, 8). Primary treatment is complete receptor were then screened and their expression surgical resection when feasible with adjuvant radiotherapy. pattern characterized in human tissue samples of non- The role of chemotherapy is debatable. Treatment of recurrent SACC and recurrent SACC with perineural metastatic ACC has been difficult to date due to lack of invasion. Results: Expression of chemokine receptors specific targets for metastatic cells (1). Though the steps that C5AR1, CCR1, CCR3, CCR6, CCR7, CCR9, CCR10, must occur in the metastatic event are well characterized, it CXCR4, CXCR6, CXCR7, CCRL1 and CCRL2 were higher remains unclear why or how ACC cells ultimately “choose” in SACC-83 compared to SACC-LM. CCRL1, CCBP2, or are ”chosen” to migrate to a specific metastatic site. A CMKLR1, XCR1 and CXCR2 and 6 chemokine genes mounting body of evidence suggests that -like (CCL13, CCL27, CXCL14, CMTM1, CMTM2, CKLF) were molecules called chemokines play a significant role in more highly expressed in tissues of patients without tumor directing the cellular traffic in metastatic melanoma, lung, recurrence/perineural invasion compared to those with breast and ACC cancers (9-15). tumor recurrence. CCRL1 (receptor), CCL27, CMTM1, In this study, we aimed to characterize the SACC CMTM2, and CKLF (chemokine) genes were more highly chemokine and receptor pattern that is seen in SACC human expressed in SACC-83 and human tissues of patients without cell lines and tissue samples with the hope that a detailed tumor recurrence/perineural invasion. Conclusion: CCRL1, characterization of the chemokine and receptor pattern in CCL27, CMTM1, CMTM2 and CKLF may play important lung metastases from SACC could lead to development of roles in the development of tumor metastases in SACC. novel therapeutic options to prevent the development of disseminated disease or to detect the presence of circulating Salivary adenoid cystic carcinoma (SACC) is the second tumor cells with metastatic potential. most common malignant tumor of the salivary glands characterized by slow, relentless growth, local recurrence, Materials and Methods and latent distant hematogenous dissemination primarily to the lung (1). Involvement of regional lymph nodes is Cell lines and human tissues. SACC-83 and -LM (lung metastasis) relatively rare but incidence of perineural invasion is high, cells were a gift from Central Laboratory, Peking University School and Hospital of Stomatology, Peking P.R. China. SACC-83 has a increasing recurrence risk. Most authors have found that low lung metastasis rate, while SACC-LM has a high lung involvement is a risk factor for subsequent metastatic rate. SACC-83 and SACC-LM cells were cultured in Dulbecco’s modified eagle medium (Gibco, Waltham, MA, USA) supplemented with 10% fetal bovine serum (Gibco), 100 U/ml penicillin and 100 U/ml streptomycin at 37˚C in a humidified Correspondence to: Christopher Sullivan, MD, Department of atmosphere containing 5% CO 2. Six SACC human tissue samples Otolaryngology, Wake Forest School of Medicine, Medical Center Blvd, (3 with and 3 without recurrence and perineural invasion) were 4th Floor Watlington Hall, Winston Salem, NC 27157, U.S.A. Tel: +1 received from Wake Forest University Comprehensive Cancer 7163854, Fax: +1 3367163857, e-mail: [email protected] Center Tissue Procurement Core Lab, Wake Forest School of Medicine Tumor Bank, Winston Salem, NC, USA. The study was Key Words: Adenoid cystic carcinoma, salivary tumors, distant approved by the Wake Forest University Health Sciences metastasis, metastatic salivary adenoid cystic carcinoma. Institutional Review Board.

0250-7005/2016 $2.00+.40 4013 ANTICANCER RESEARCH 36 : 4013-4018 (2016)

Table I. Gene table: RT² profiler polymerase chain reaction (PCR) array.

UniGene GenBank Symbol description

Hs.494997 NM_001735 C5 Complement component 5 Hs.2161 NM_001736 C5AR1 Complement component 5a receptor 1 Hs.146346 NM_001296 CCBP2 Chemokine binding 2 Hs.72918 NM_002981 CCL1 Chemokine (C-C motif) ligand 1 Hs.303649 NM_002982 CCL2 Chemokine (C-C motif) ligand 2 Hs.514107 NM_002983 CCL3 Chemokine (C-C motif) ligand 3 Hs.75703 NM_002984 CCL4 Chemokine (C-C motif) ligand 4 Hs.514821 NM_002985 CCL5 Chemokine (C-C motif) ligand 5 Hs.251526 NM_006273 CCL7 Chemokine (C-C motif) ligand 7 Hs.271387 NM_005623 CCL8 Chemokine (C-C motif) ligand 8 Hs.54460 NM_002986 CCL11 Chemokine (C-C motif) ligand 11 Hs.414629 NM_005408 CCL13 Chemokine (C-C motif) ligand 13 Hs.272493 NM_032963 CCL14 Chemokine (C-C motif) ligand 14 Hs.272493 NM_032965 CCL15 Chemokine (C-C motif) ligand 15 Hs.10458 NM_004590 CCL16 Chemokine (C-C motif) ligand 16 Hs.546294 NM_002987 CCL17 Chemokine (C-C motif) ligand 17 Hs.143961 NM_002988 CCL18 Chemokine (C-C motif) ligand 18 (pulmonary and activation-regulated) Hs.50002 NM_006274 CCL19 Chemokine (C-C motif) ligand 19 Hs.75498 NM_004591 CCL20 Chemokine (C-C motif) ligand 20 Hs.57907 NM_002989 CCL21 Chemokine (C-C motif) ligand 21 Hs.534347 NM_002990 CCL22 Chemokine (C-C motif) ligand 22 Hs.169191 NM_005064 CCL23 Chemokine (C-C motif) ligand 23 Hs.247838 NM_002991 CCL24 Chemokine (C-C motif) ligand 24 Hs.310511 NM_005624 CCL25 Chemokine (C-C motif) ligand 25 Hs.131342 NM_006072 CCL26 Chemokine (C-C motif) ligand 26 Hs.648124 NM_006664 CCL27 Chemokine (C-C motif) ligand 27 Hs.656904 NM_148672 CCL28 Chemokine (C-C motif) ligand 28 Hs.301921 NM_001295 CCR1 Chemokine (C-C motif) receptor 1 Hs.511794 NM_001123396 CCR2 Chemokine (C-C motif) receptor 2 Hs.506190 NM_001837 CCR3 Chemokine (C-C motif) receptor 3 Hs.184926 NM_005508 CCR4 Chemokine (C-C motif) receptor 4 Hs.450802 NM_000579 CCR5 Chemokine (C-C motif) receptor 5 Hs.46468 NM_004367 CCR6 Chemokine (C-C motif) receptor 6 Hs.370036 NM_001838 CCR7 Chemokine (C-C motif) receptor 7 Hs.113222 NM_005201 CCR8 Chemokine (C-C motif) receptor 8 Hs.225946 NM_006641 CCR9 Chemokine (C-C motif) receptor 9 Hs.278446 NM_016602 CCR10 Chemokine (C-C motif) receptor 10 Hs.535713 NM_003965 CCRL1 Chemokine (C-C motif) receptor-like 1 Hs.729361 NM_016557 CCRL2 Chemokine (C-C motif) receptor-like 2 Hs.15159 NM_181641 CKLF Chemokine-like factor Hs.197143 NM_004072 CMKLR1 Chemokine-receptor 1 Hs.549232 NM_181269 CMTM1 CKLF-like MARVEL transmembrane domain containing 1 Hs.195685 NM_144673 CMTM2 CKLF-like MARVEL transmembrane domain containing 2 Hs.298198 NM_144601 CMTM3 CKLF-like MARVEL transmembrane domain containing 3 Hs.643961 NM_178818 CMTM4 CKLF-like MARVEL transmembrane domain containing 4 Hs.531668 NM_002996 CX3CL1 Chemokine (C-X3-C motif) ligand 1 Hs.78913 NM_001337 CX3CR1 Chemokine (C-X3-C motif) receptor 1 Hs.789 NM_001511 CXCL1 Chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity, alpha) Hs.590921 NM_002089 CXCL2 Chemokine (C-X-C motif) ligand 2 Hs.89690 NM_002090 CXCL3 Chemokine (C-X-C motif) ligand 3 Hs.89714 NM_002994 CXCL5 Chemokine (C-X-C motif) ligand 5 Hs.164021 NM_002993 CXCL6 Chemokine (C-X-C motif) ligand 6 (granulocyte chemotactic protein 2) Hs.77367 NM_002416 CXCL9 Chemokine (C-X-C motif) ligand 9 Hs.632586 NM_001565 CXCL10 Chemokine (C-X-C motif) ligand 10 Hs.632592 NM_005409 CXCL11 Chemokine (C-X-C motif) ligand 11 Hs.522891 NM_000609 CXCL12 Chemokine (C-X-C motif) ligand 12 Hs.100431 NM_006419 CXCL13 Chemokine (C-X-C motif) ligand 13 Hs.483444 NM_004887 CXCL14 Chemokine (C-X-C motif) ligand 14

Table I. Continued

4014 Mays et al : Chemokine Profiles in Metastatic Salivary Adenoid Cystic Carcinoma

Table I. Continued

UniGene GenBank Symbol description

Hs.708201 NM_022059 CXCL16 Chemokine (C-X-C motif) ligand 16 Hs.194778 NM_000634 CXCR1 Chemokine (C-X-C motif) receptor 1 Hs.846 NM_001557 CXCR2 Chemokine (C-X-C motif) receptor 2 Hs.198252 NM_001504 CXCR3 Chemokine (C-X-C motif) receptor 3 Hs.593413 NM_003467 CXCR4 Chemokine (C-X-C motif) receptor 4 Hs.113916 NM_001716 CXCR5 Chemokine (C-X-C motif) receptor 5 Hs.34526 NM_006564 CXCR6 Chemokine (C-X-C motif) receptor 6 Hs.471751 NM_020311 CXCR7 Chemokine (C-X-C motif) receptor 7 Hs.592212 NM_001953 TYMP Thymidine phosphorylase Hs.546295 NM_002995 XCL1 Chemokine (C motif) ligand 1 Hs.458346 NM_003175 XCL2 Chemokine (C motif) ligand 2 Hs.248116 NM_005283 XCR1 Chemokine (C motif) receptor 1

RNA extraction and microarrays . Total RNA was isolated from Results SACC-LM, SACC-83 cells and human SACC tissues using Trizol reagent (Life Technologies, Carlsbad, CA, USA). Generally, cell Expression of chemokine receptors C5AR1, CCR1, CCR3, lines and tissue samples were homogenized in Trizol reagent and incubated for 5 min at room temperature. Cell debris was removed CCR6, CCR7, CCR9, CCR10, CXCR4, CXCR6, CXCR7, by centrifuge and supernatant solution was transferred to new tube. CCRL1 and CCRL2 was higher in SACC-83 compared to Chloroform was added into the Trizol reagent with a ratio (1:5), SACC-LM cell lines, p< 0.05 (Table II). Table III vortexed for 15 sec, incubated at room temperature for 3 min and demonstrates the statistically significant chemokine gene centrifuged at 12,000 × g for 15 min at 4˚C. The upper aqueous expression between SACC-83 and SACC-LM ( p< 0.05). Six phase was carefully removed into new tube. Then the same volume human SACC were analyzed for expression of chemokines of isopropyl alcohol was added, mixed up and down for 10 sec, and receptors. Three samples were recurrent tumors and all incubated at room temperature for 10 min and centrifuged at 12,000 × g for 15 min at 4˚C. RNA pellets were visible on the side and had perineural invasion (See Table IV for detailed bottom of the tube, then washed once with 75% ethanol and information of patients). CCRL1, CCBP2, CMKLR1, XCR1 centrifuged at 7,500 × g for 5 min at 4˚C. RNA pellets were air- and CXCR2 receptors were more highly expressed in tissue dried for 5-10 min, dissolve with nuclease-free water and quantified samples of patients without tumor recurrence and perineural with NanoDrop 1000 Spectrophotometer (Thermo Scientific, invasion compared to those with (Table V). CCL13, CCL27, Wilmington, DE, USA). CXCL14, CMTM1, CMTM2 and CKLF chemokine genes were more highly expressed in tissue samples of patients Reverse transcription for cDNA synthesis and real-time quantitative polymerase chain reaction (qPCR) for chemokine and without tumor recurrence and perineural invasion compared receptor detection and quantification. RT2 First Strand Kit to those with tumor recurrence (Table VI). (QIAGEN, Valencia, CA, USA) was used for cDNA synthesis. RT2 Profiler PCR Array (QIAGEN) was used for real-time qPCR Discussion quantification of known human chemokines and receptors (genes tested in this study are shown in Table I). Amplification reactions The complexity of the events that incite and control the were performed in triplicate in the Agilent MX 3000P qPCR migration of cells to metastatic sites remains a considerable System (Agilent Technologies, Inc, Folsom, CA, USA). All the procedures were performed strictly following the manufacturer’s challenge in oncology research. Recent studies have instructions. Each RNA expression was normalized to the demonstrated that tumor cells express functionally active housekeeping gene as internal control provided by the chemokine receptors and that expression of these receptors manufacturer. appears to regulate cellular functions associated with the process of metastasis (9, 14, 15). Chemokines are a Data analysis. The empirical Bayes statistics implemented in superfamily of small cytokine-like that selectively Bioconductor package Limma was used for the data analysis and attract and activate different cell types (12, 13). The CXC detailed method as previously described (16). The modified Student’s t- test was used to determine differentially expressed chemokine receptor CXCR4 is expressed in several cancer chemokine and receptors between SACC-83 and SACC-LM cell types, including breast cancer (11), melanoma (13) and lung lines as well as the six human samples coded with or without SACC cancer (13). Studies have suggested that, in these cancer recurrence and perineural invasion. types, CXCR4 may play a role in metastases. Furthermore,

4015 ANTICANCER RESEARCH 36 : 4013-4018 (2016)

Table II . Analysis of chemokine receptor mRNA expression between Table III. Analysis of chemokine mRNA expression between salivary salivary adenoid cystic carcinoma (SACC)-LM and SACC-83 cells using adenoid cystic carcinoma (SACC)-LM and SACC-83 cells using real- real-time quantitative polymerase chain reaction (PCR), with associated time quantitative polymerase chain reaction (PCR). chemokines. Gene p-Value Gene p-Value Associated chemokines C5 7.39E-06 C5AR1 0.009707 C5 CCL2 0.000652 CCR1 0.004494 CCL5 CCL5 8.96E-05 CCL15 CCL15 0.010738 CCR3 0.001134 CCL15 CCL17 0.000333 CCL24 CCL26 CCL28 CCL18 0.000459 CCR6 0.017816 CCL20 CCL20 3.12E-06 CCR7 0.000144 CCL22 2.37E-07 CCR9 0.000243 CCL25 CCL24 0.011392 CCR10 0.01992 CCL27 CCL25 0.001088 CCL28 CCL26 4.03E-05 CXCR4 0.000493 CCL27 0.001522 CXCR6 0.025469 CCL28 1.67E-05 CXCR7 5.15E-05 CXCL11 CXCL2 2.35E-08 CCRL1 0.002157 CCL25 CXCL3 7.57E-05 CCRL2 0.00343 CXCL10 8.03E-06 CXCL11 5.14E-07 SACC-LM had lower expression of every chemokine receptor gene CXCL14 0.000828 compared to SACC-83 ( p<0.05). Underline represents greater XCL1 0.000794 expression of SACC-LM compared to SACC-83 in the associated XCL2 0.000118 chemokines. CMTM1 0.008204 CMTM2 0.001772 CMTM3 0.010851 CMTM4 0.003763 CKLF 0.013026 CXCR4 receptor expression has been implicated in SACC-LM had lower expression of all genes except those underlined metastatic spread in vivo , in animal models of breast cancer when compared to SACC-83 ( p<0.05). and melanoma (13, 14). A group of researchers in Germany, working with ACC and squamous cell carcinoma (SCCA) cell lines, discovered that ACC and SCCA expressed a distinct, non-random pattern of chemokine receptors that influenced their metastatic behavior. The chief finding was study, we found differential expression of a panel of that ACC cells expressed high levels of CXCR4 both in vitro chemokines and receptors between two SACC metastatic cell and in pathologic tumor specimens, whereas CXCR4 lines. These chemokines and receptors may provide direction expression was undetectable or low in SCC. Chemokine into further studies investigating chemokine-driven receptor CCR7 was abundantly expressed in primary tumors metastasis. Chemokines and receptors were also analyzed in and lymph node metastases of SCCA but restricted in ACC human tissue looking at differential expression between (17). The authors concluded that significant differences in perineural positive and negative samples and recurrent and chemokine expression correlated with known clinical non-recurrent in order to investigate chemokine-driven local behavior of these two head and neck tumor entities. tissue invasion and progression. We found differential These data taken collectively suggest that chemokine expression of a panel of chemokines and receptors between receptor status in primary ACC tumors could predict organ- these low and high locally aggressive groups. Interestingly, specific metastases in vivo . In order to gain incite into the we found a panel of chemokines and receptors that showed process of metastases in ACC, others have attempted to create the same pattern of expression between the SACC- clinically relevant animal models of metastatic disease. LM/SACC-83 group and the recurrent/non-recurrent human Recently, Liu et al. used 2 ACC cell lines, one with a low and tissues. CCRL1, CCL27, CMTM1, CMTM2 and CKLF one with a high metastasis rate, and found that autocrine demonstrated lower expression in both the SACC-LM group, expression of epiregulin-induced epidermal as well as the tumor recurrence/perineural invasion group. As receptor activation promotes lung metastasis via epithelial- previously stated, the SACC-LM is well established as having mesenchymal transition (18). We aimed in this study to use high metastatic potential. Similarly those patients with tumor the same SACC cell lines and human SACC samples to study recurrence and/or perineural invasion represent a higher risk the chemokine receptor pattern of these metastatic foci. In our group with increased propensity to poor outcomes when

4016 Mays et al : Chemokine Profiles in Metastatic Salivary Adenoid Cystic Carcinoma

Table IV. Human salivary adenoid cystic carcinoma (SACC) samples.

Patient Age Gender Stage Recurrence with perineural invasion Recurrence type

1 74 Male T1N0 No None 2 61 Male T3N1 No None 3 47 Male T2N0 No None 4 90 Female T4N0 Yes Local 5 52 Male T4N0 Yes Regional 6 55 Male T4N0 Yes Local

Table V. Analysis of chemokine receptor mRNA expression in human Conclusion salivary adenoid cystic carcinoma (SACC) tissues.

Gene p-Value CCRL1, CCL27, CMTM1, CMTM2 and CKLF may play important roles in the development of tumor metastases in CCRL1 0.002668 salivary ACC. Further investigation into the mechanisms CCBP2 0.005916 underlying this chemokine-mediated tumor cell migration and CMKLR1 0.030776 invasion could lead to development of novel prognostic and XCR1 0.031393 CXCR2 0.041186 therapeutic options to predict metastatic potential, prevent the development of metastases or treat disseminated disease. All genes demonstrated less expression in recurrence samples versus non-recurrent ( p<0.05). Acknowledgements

Jeff Chou Ph.D, is acknowledged for his assistance with statistical analysis. Table VI. Analysis of chemokine mRNA expression in human salivary adenoid cystic carcinoma (SACC) tissues. References Gene p- Value 1 Suarez C, Barnes L, Silver CE, Rodrigo JP, Shah JP, CCL13 0.00707 Triantafyllou A, Rinaldo A, Cardesa A, Pitman KT, Kowalski CCL27 0.012257 LP, Robbins KT, Hellquist H, Medina JE, de Bree R, Takes RP, CXCL14 0.026939 Coca-Pelaz A, Bradley PJ, Gnepp DR, Teymoortash A, Strojan CMTM1 0.043589 P, Mendenhall WM, Eloy JA, Bishop JA, Devaney KO, CMTM2 0.044802 Thompson LD, Hamoir M, Slootweg PJ, Vander Poorten V, CKLF 0.04394 Williams MD, Wenig BM, Skálová A and Ferlito A: Cervical All genes demonstrated less expression in recurrence samples versus lymph node metastasis in adenoid cystic carcinoma of oral non-recurrent ( p<0.05). cavity and oropharynx: A collective international review. Auris Nasus Larynx 43(5) : 477-484, 2016. 2 Douglas JG, Laramore GE, Austin-Seymour M, Koh W, Stelzer K and Griffin TW: Treatment of locally advanced adenoid cystic compared to non-recurrent. Finding a panel of chemokines carcinoma of the head and neck with neutron radiotherapy. Int J Radiat Oncol Biol Phys 46 : 551-557, 2000. and receptors that were associated with lower expression in 3 Huber PE, Debus J, Latz D, Zierhut D, Bischof M, both high-risk groups (SACC-LM and recurrent tumor/ Wannenmacher M and Engenhart-Cabillic R: Radiotherapy for perineural invasion) certainly suggests that these may play a advanced adenoid cystic carcinoma: neutrons, photons or mixed more significant role in chemokine-driven pathways of beam? Radiother Oncol 59 : 161-167, 2001. metastasis and local progression. This may also suggest an 4 Lee SY, Shin HA, Rho KJ, Chung HJ, Kim SH and Choi EC: activity of gene suppression in those patients with distant Characteristics, management of the neck, and oncological metastases and perineural invasion. Further elucidating these outcomes of malignant minor salivary gland tumours in the oral and sinonasal regions. Br J Oral Maxillofac Surg 51 : e142-147, chemokines and their expression patterns could lead to the 2013. development of therapeutic targets to limit local and distant 5 Lloyd S, Yu JB, Wilson LD and Decker RH: Determinants and spread. In addition, based on these profiles, drug development patterns of survival in adenoid cystic carcinoma of the head and could be directed toward the prevention of disseminated neck, including an analysis of adjuvant radiation therapy. Am J disease with considerable impact on mortality from ACC. Clin Oncol 34 : 76-81, 2011.

4017 ANTICANCER RESEARCH 36 : 4013-4018 (2016)

6 Ko YH, Lee MA, Hong YS, Lee KS, Jung CK, Kim YS, Sun DI, 13 Spano JP, Andre F, Morat L, Sabatier L, Besse B, Combadiere Kim BS, Kim MS and Kang JH: Prognostic factors affecting the C, Deterre P, Martin A, Azorin J, Valeyre D, Khayat D, Le clinical outcome of adenoid cystic carcinoma of the head and Chevalier T and Soria JC: Chemokine receptor CXCR4 and neck. Jpn J Clin Oncol 37 : 805-811, 2007. early-stage non-small cell lung cancer: pattern of expression and 7 Bier H, Hoffmann T, Hauser U, Wink M, Ochler M, Kovar A, correlation with outcome. Ann Oncol 15 : 613-617, 2004, Müser M and Knecht R: Clinical trial with escalating doses of 14 Zlotnik A: Chemokines in neoplastic progression. Semin Cancer the antiepidermal growth factor receptor humanized monoclonal Biol 14 : 181-185, 2004. antibody EMD 72 000 in patients with advanced squamous cell 15 Zlotnik A and Yoshie O: Chemokines: A new classification carcinoma of the larynx and hypopharynx. Cancer Chemoth system and their role in immunity. Immunity 12 : 121-127, 2000. Pharmacol 47 : 519-524, 2001. 16 Smyth GK: Linear models and empirical bayes methods for 8 Fordice J, Kershaw C, El-Naggar A and Goepfert H: Adenoid assessing differential expression in microarray experiments. Stat cystic carcinoma of the head and neck: predictors of morbidity Appl Genet Mol Biol 3: Article 3, 2004. and mortality. Arch Otolaryngol Head Neck Surg 125 : 149-152, 17 Hashitani S, Urade M, Zushi Y, Segawa E, Okui S and Sakurai 1999. K: Establishment of nude mouse transplantable model of a 9 Campbell JJ and Butcher EC: Chemokines in tissue-specific and human adenoid cystic carcinoma of the oral floor showing microenvironment-specific homing. Curr Opin metastasis to the lymph node and lung. Oncol Rep 17 : 67-72, Immunol 12 : 336-341, 2000. 2007. 10 Liotta LA: An attractive force in metastasis. Nature 410 : 24-25, 18 Liu S, Ye D, Xu D, Liao Y, Zhang L, Liu L, Yu W, Wang Y, He 2001. Y, Hu J, Guo W, Wang T, Sun B, Song H, Yin H, Liu J, Wu Y, 11 Muller A, Sonkoly E, Eulert C, Gerber PA, Kubitza R, Schirlau Zhu H, Zhou BP, Deng J and Zhang Z: Autocrine epiregulin K, Franken-Kunkel P, Poremba C, Snyderman C, Klotz LO, activates EGFR pathway for lung metastasis via EMT in salivary Ruzicka T, Bier H, Zlotnik A, Whiteside TL, Homey B and adenoid cystic carcinoma. Oncotarget 2016. doi: 10.18632/ Hoffmann TK: Chemokine receptors in head and neck cancer: oncotarget.7940. (Epub ahead of print) association with metastatic spread and regulation during chemotherapy. Int J Cancer 118 : 2147-2157, 2006. 12 Murakami T, Maki W, Cardones AR, Fang H, Tun Kyi A, Nestle FO and Hwang ST: Expression of CXC chemokine receptor-4 Received June 16, 2016 enhances the pulmonary metastatic potential of murine B16 Revised July 4, 2016 melanoma cells. Cancer Res 62 : 7328-7334, 2002. Accepted July 6, 2016

4018