Original Research—Head and Neck Surgery
BMP-2 Expression Correlates with Local Failure in Head and Neck Squamous Cell Carcinoma
Otolaryngology– Head and Neck Surgery 2014, Vol. 150(2) 245–250 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2013 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599813513003 http://otojournal.org
Jordan P. Sand, MD1, Natalia A. Kokorina, MD1, Stanislav O. Zakharkin, PhD, James S. Lewis Jr, MD1,2, and Brian Nussenbaum, MD1
No sponsorships or competing interests have been disclosed for this article.
Keywords bone morphogenetic protein, squamous cell carcinoma
Abstract Objective. Preclinical data show that exogenous administration of recombinant human bone morphogenetic protein2 (rhBMP-2) to human oral carcinoma cell lines increases pathogenicity using a nude mouse model. The objectives of this study are to (1) describe the characteristics of baseline protein expression of BMP-2 in head and neck squamous cell carcinomas (HNSCC) and (2) determine if BMP-2 expression level correlates with worse oncologic outcomes. Study Design. Retrospective analysis of previously harvested patient samples. Setting. Academic medical center. Subjects. In total, 149 patients with oral cavity, oropharynx, larynx, and hypopharynx HNSCC treated between January 1, 1997, and December 31, 2004. Methods. A tissue microarray of HNSCC was assembled and immunohistochemistry for BMP-2 performed. Staining was quantified using a standardized scoring system. Specimens were dichotomized into high or low expression level. Statistical analyses using log-rank, Wilcoxon, and Fisher exact test were performed for associations between BMP-2 protein level and clinicopathologic features and patient survival. Results. BMP-2 expression at any level was noted in 146 of 149 (98%) of samples. Tumors with high BMP-2 expression had higher rates of local failure compared with lowexpressing tumors (17.3% vs 6.3%; P = .04). There was no significant association for BMP-2 expression level with tumor location, T stage, N stage, overall survival, regional failure, or distant failure. Conclusion. Head and neck squamous cell carcinomas with high baseline BMP-2 protein level are associated with higher rates of local recurrence. These data have important implications for using rhBMP-2 in tissue engineering reconstructive approaches in the setting of cancer-related defects.
Received September 15, 2013; revised October 14, 2013; accepted October 25, 2013.
O
steoinductive therapy using recombinant human bone morphogenetic protein-2 (rhBMP-2) was approved by the Food and Drug Administration (FDA) in 2007 for use in alveolar ridge and maxillary sinus floor augmentation. In addition, investigators have demonstrated promising clinical data using rhBMP-2 for mandibular reconstruction, dental implant stabilization, healing of LeFort osteotomies, and repair of cleft palate defects.1-9 As knowledge, technology, and experience advance in this area, there might be a paradigm shift in reconstructive techniques from current therapies using autologous bone grafts toward tissue engineering using bone morphogenetic proteins (BMPs). The anticipated advantages are avoidance of donor site morbidity, decreased operative time, decreased surgical complications, shorter hospitalization, quicker recovery from surgery, and more accurate structural restoration of surgical defects.9 Although advances using biologic therapies for reconstruction are emerging and beginning to be translated to patients, little is known about the biologic effects of BMPs on head and neck squamous cell carcinoma (SCC). This is a critically important point since most patients with major craniomaxillofacial and mandible bone defects that would
1 Department of Otolaryngology–Head and Neck Surgery, Washington University School of Medicine, St Louis, Missouri, USA 2 Department of Pathology and Immunology, Washington University School of Medicine, St Louis, Missouri, USA
This article was presented at the 2013 AAO-HNSF Annual Meeting & OTO EXPO; September 29 to October 3, 2013; Vancouver, BC, Canada. Corresponding Author: Brian Nussenbaum, MD, Christy J. and Richard S. Hawes III, Professor, Department of Otolaryngology–Head and Neck Surgery, Washington University School of Medicine, 660 South Euclid Ave, CB 8115, St Louis, MO 63110, USA. Email: [email protected]
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
246
Otolaryngology–Head and Neck Surgery 150(2)
Figure 1. (A) Typical keratinizing-type oral cavity squamous cell carcinoma (3100 magnification). (B-D) Bone morphogenetic protein-2 immunohistochemistry showing (B) no tumor staining; (C) diffuse, strong tumor staining; and (D) minimal staining in tumor-adjacent tonsil crypt epithelium. (B, C, D: 3200 magnification).
benefit from a tissue engineering approach occur in the setting of head and neck cancer. BMP-2 expression has been very prevalent in oral carcinomas.10 Our group has previously shown that oral squamous cell carcinoma cell lines treated with rhBMP-2 revealed increased invasiveness and pathogenicity.11,12 In addition, BMP-2 gene expression has been associated with regional metastasis in oral squamous cell carcinoma.13,14 The primary objectives of this study are to (1) evaluate baseline patterns of BMP-2 protein expression level in head and neck squamous cell carcinoma and (2) determine if the level of the protein correlates with clinical characteristics, pathologic features, or oncologic outcomes.
Methods Patient Cohort Approval from the Human Research Protection Office (institutional review board) at Washington University School of Medicine was obtained for conducting this study. A tissue microarray (TMA) was constructed by our study pathologist (J.S.L.) from newly diagnosed primary cases of head and neck SCC treated between January 1, 1997, and December 31, 2004. Primary tumor sites included oral cavity, oropharynx, hypopharynx, and larynx. The SCC variants—namely, verrucous, spindle cell, papillary, and basaloid—were excluded. All patients were
treated with curative intent with surgery, surgery with adjuvant therapy, radiation therapy, or chemoradiation. None of the patients had received prior treatment for a head and neck carcinoma. Formalin-fixed and paraffinembedded tissues were used for the tissue microarray. According to the amount of available biopsied or resected tumor tissue, duplicate 2-mm (or, if only limited amounts of tumor tissue present, 0.6-mm) punches were taken. Since most of the cases were treated with primary surgery, the majority of cases on the array had the larger (2-mm) punches. Clinical details and follow-up information, including survival and recurrence data, were obtained from clinician databases and electronic medical records. This TMA has been used in several previous studies, the results of which have largely matched the expected findings for a contemporary cohort of HNSCC.15-18
Immunohistochemistry Immunohistochemistry was performed for BMP-2 using the commercially available antibody from Santa Cruz Biotechnology (Dallas, TX; N-14: sc-6895; goat polyclonal) at a dilution of 1:75 with EDTA and heat antigen retrieval. Positive control staining was observed in mouse jaw tissue (positive expression in osteoblasts and odontoblasts). A negative control slide of normal human cerebellum was also used, which showed no staining. The stained tissue microarray
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
Sand et al
247
slides were then digitally scanned on an Aperio Scanscope XT digital slide scanner (Vista, California) at 3400 magnification and read visually from the digital images. The study pathologist (J.S.L.) scored the staining, blinded to all clinical and pathologic features. Cases in which there was less than 10% surface area consisting of tumor (averaged across the 2 microarray punches) were excluded from analysis. Immunohistochemical staining was graded on the basis of distribution in a quartile manner (0; 11, 1%-25% staining; 21, 26%-50% staining; 31, 51%-75% staining; and 41, 76%-100% staining). Grading was also scored on intensity (11, weak staining; 21, moderately intense staining; and 31, strong staining). A quantitative score of 0 or 2 through 7 was obtained by adding the distribution and intensity scores. Tumor samples were then divided into 2 groups: high level of protein expression (composite score of 6 or 7) and low level (composite score of 0, 2, 3, 4, or 5) for comparative analysis (Figure 1A-C).
Statistical Analysis Survival analysis was performed using the Kaplan-Meier method. The differences noted within survival curves were compared using log rank and Wilcoxon tests. Associations between clinical and pathologic features were analyzed by the Fisher exact test. P values at or below .05 were considered significant. All analyses were done using SAS 9.2 (SAS Institute, Cary, North Carolina).
Results A total of 149 tumors were tested for BMP-2 protein expression with immunohistochemistry. Mean follow-up time for these patients was 50 months. Twelve patients had clinical information missing with regard to their T or N stage and thus were excluded from those analyses. BMP-2 expression at any level was noted in 146 of 149 (98%) analyzed samples. Table 1 summarizes the patient characteristics with regard to clinical and pathologic variables. The TMA was specifically designed to isolate tumor-containing tissue. However, there are some areas of normal mucosa and tonsillar crypt epithelium present in some of the punches. These showed minimal staining, with rare patchy nuclear positivity in normal squamous cells (Figure 1D). Tumors with high BMP-2 protein expression had a higher rate of local failure compared with low-expressing tumors (17.3% vs 6.3%; P = .04). This is graphically shown in Figure 2. No significant correlation was observed between the level of BMP-2 protein and overall survival, regional failure, distant failure, T stage, or N stage (Table 2). These data were then reanalyzed controlling for disease site (oropharynx and nonoropharynx) and treatment type, with neither demonstrating a statistically significant difference in any of the previously mentioned variables.
Discussion The potential role of BMPs in malignant transformation and cancer progression remains poorly understood. Investigations into this area have proceeded in varying degrees in both in
Table 1. Clinical case characteristics (n = 149). Clinical Characteristics Subsites Oral cavity Oropharynx Larynx Hypopharynx Treatment type Surgery alone Surgery with adjuvant therapy Radiotherapy 6 chemotherapy Unknown T stage T1 T2 T3 T4 Unknown N stage N0 N1 N2 N3 Unknown BMP-2 expression High (score 6 or 7) Low (score 2-5) None (0)
No. (%) of Total Cases
36 (24.2) 74 (49.7) 29 (19.5) 10 (6.7) 13 (8.7) 117 (78.5) 17 (11.4) 2 (1.3) 27 (18.1) 42 (28.2) 36 (24.2) 33 (22.1) 11 (7.4) 35 (23.5) 24 (16.1) 75 (50.3) 3 (2.0) 12 (8.1) 69 (46.3) 77 (51.7) 3 (2.0)
Figure 2. Kaplan-Meier local recurrence curve comparing local failure probability of Bone morphogenetic protein-2 (BMP-2) high expression level vs BMP-2 low expression level over time. P value by log-rank test was .04.
vitro and in vivo studies, which have revealed discordant findings based on the tumor site, tumor type, and BMP studied.19-35 Most BMP-based bone tissue engineering approaches use a BMP-2 platform, either with protein or gene therapy. Previously reported data have shown that BMP-2 and the
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
248
Otolaryngology–Head and Neck Surgery 150(2)
Table 2. Bone morphogenetic protein-2 (BMP-2) expression-level associations with clinicopathologic variables. Clinical/Pathologic Variable Overall survival Local failure Regional failure Distant failure T stage T1 T2 T3 T4 N stage N0 N1 N2 N3
BMP-2 Expression Level
No. (%)
High Low High Low High Low High Low
41 41 12 5 7 6 7 11
(59.4) (51.3) (17.4) (6.3) (10.1) (7.5) (10.1) (13.8)
High Low High Low High Low High Low
9 18 22 22 17 19 19 14
(11.8) (19.8) (24.7) (23.2) (20.2) (20.7) (22.1) (16.1)
High Low High Low High Low High Low
20 16 9 15 34 41 3 0
(23.3) (18.2) (12.0) (17.2) (34.0) (36.3) (4.3)
receptor BMPR-IA were frequently expressed in human tumors as detected by immunohistochemistry.10 In addition, studies have shown that the BMP-2 gene is frequently expressed in oral SCC cell lines but not usually expressed in normal oral mucosa.12 Treatment of oral squamous cell carcinoma cell lines with rhBMP-2 resulted in increased invasiveness in vitro.11 Using a tumor xenograft in a nude mouse model, oral carcinoma cells treated with rhBMP-2 demonstrated more locally aggressive biologic behavior.12 These data suggest that further study of the possible oncologic role of BMP-2 in head and neck squamous cell carcinomas is warranted. This study demonstrates that baseline BMP-2 protein expression is found in most head and neck squamous cell carcinomas (98%). High levels of BMP-2 were considered when greater than 50% of the tumor had staining with strong intensity or greater than 75% of the tumor having staining with at least moderate staining intensity. In this study, a high level of BMP-2 protein expression was correlated with increased incidence of local failure. Notably, local failure was nearly 3 times more frequent in patients who were identified to have a higher level of BMP-2 expression in their tumors compared with patients who had a lower level. This finding correlates with the preclinical data showing that oral carcinoma cell
P Value for Association
.33 .04 .58 .62
.31
.16
lines treated with rhBMP-2 had more aggressive local biologic behavior, both in vitro and in vivo.11,12 Analysis with quantitative polymerase chain reaction (PCR) suggested that the molecular pathways of increased invasiveness in these cell lines was through upregulation of interleukin (IL)–1b, a contributor to invasion, and the downregulation of metastasis suppressor-1, a repressor of invasion and metastasis.11,36-39 Additional studies using a human phosphokinase array suggested that increased local invasiveness might be related to epithelial-mesenchymal transition, the JAK/STAT pathway, and the Wnt-b-catenin pathway.12 Recent data showed that increased baseline protein expression of BMP-6 in oral carcinomas might be related to local aggressiveness as measured by bone invasion.40 The current study was designed to determine if the results from prior studies investigating the biologic effects of BMP-2 in oral carcinoma cell lines consistently translated to patient outcome data. Future studies will focus on the mechanisms with consideration of bone invasion and other histopathologic features known to predict local aggressiveness.41 These variables were not included in the data collected on the TMA used for this study. The patients included in this study were representative of a typical head and neck cancer population in terms of tumor
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
Sand et al
249
site, T stage, and N stage. Weaknesses of this study include its retrospective nature and the possible lack of sufficient statistical power to discern a clinically significant relationship between BMP-2 expression and survival. Most patients were treated with a primary surgical-based approach. This can be considered a weakness of this study, since the findings might not be generalizable to patients primarily treated with nonsurgical therapies. The predominance of surgically treated patients is likely due to the easier accessibility of appropriate pretreatment tumor specimens for constructing the tissue microarray at our institution. There was a relatively even distribution of tumors with high and low scores for BMP-2. The scoring system that was used to categorize the high- and low-expressing tumors was not designed to obtain an even distribution into each category but rather was thought to be a logical method for categorizing these tumors. The BMP-2 immunohistochemistry staining demonstrated a nuclear rather than cytoplasmic pattern. Our positive and negative controls were appropriate, the staining was highly specific without background noise, and nuclear localization of BMP-2 has been described, particularly with the antibody clone used in this study.42,43 For head and neck squamous cell carcinomas, a high level of baseline protein expression of BMP-2 correlated with an increased risk for local recurrence. These data are consistent with available preclinical data showing that exogenous administration of rhBMP-2 to oral carcinoma cell lines affects the tumor cells by behaving more locally aggressive.12 Although overall survival, regional failure, and distant failure were not affected by level of BMP-2 expression, the available data confirm that caution needs to be applied when considering the use of BMP-2–based tissue engineering approaches for reconstructing cancer-related defects. Author Contributions Jordan P. Sand, analysis, drafting article, final approval; Natalia A. Kokorina, analysis, revising of article, final approval; Stanislav O. Zakharkin, study design, acquisition of data, interpretation of data, revising article, final approval; James S. Lewis Jr, analysis, interpretation of data, drafting, final approval; Brian Nussenbaum, study conception and design, analysis, interpretation of data, final approval.
Disclosures Competing interests: None. Sponsorships: None. Funding source: None.
References 1. Warnke PH, Coren AJ. First experiences with recombinant human bone morphogenetic protein-7 (osteogenic protein-1) in a human case in maxillofacial surgery. Plast Reconstr Surg. 2003;111:2471-2472. 2. Moghadam HG, Urist MR, Sandor GKB, Clokie CML. Successful mandibular reconstruction using a BMP bioimplant. J Craniofac Surg. 2001;12:119-127.
3. Ferretti C, Ripamonti U. Human segmental mandibular defects treated with naturally derived bone morphogenetic proteins. J Craniofac Surg. 2002;13:434-444. 4. Howell TH, Fiorellini J, Jones A, et al. A feasibility study evaluating rhBMP-2/absorbable collagen sponge device for local alveolar ridge preservation or augmentation. Int J Periodontics Restor Dent. 1997;17:124-139. 5. Boyne PJ, Marx RE, Nevins M, et al. A feasibility study evaluating rhBMP-2/absorbable collagen sponge device for maxillary sinus floor augmentation. Int J Periodontics Restor Dent. 1997;17:11-25. 6. Chin M, Ng T, Tom WK, et al. Repair of alveolar clefts with recombinant human bone morphogenetic protein (rhBMP-2) in patients with clefts. J Craniofac Surg. 2005;16:778-789. 7. Herford AS, Boyne PJ. Reconstruction of mandibular continuity defects with bone morphogenetic protein-2 (BMP-2). J Oral Maxillofac Surg. 2008;66:616-624. 8. Clokie CML, Sandor GKB. Reconstruction of 10 major mandibular defects using bioimplants containing BMP-7. J Can Dent Assoc. 2008;74:67-72. 9. Nussenbaum B, Krebsbach PH. The role of gene therapy for craniofacial and dental tissue engineering. Adv Drug Deliv Rev. 2006;58:577-591. 10. Jin Y, Tipoe GL, Liong EC, Lau TY, Fung PC, Leung KM. Overexpression of BMP-2/4, -5 and BMPR-IA associated with malignancy of oral epithelium. Oral Oncol. 2001;37:225-233. 11. Kokorina NA, Zakharkin SO, Krebsbach PH, Nussenbaum B. Treatment effects of rhBMP-2 on invasiveness of oral carcinoma cell lines. Laryngoscope. 2011;121:1876-1880. 12. Kokorina NA, Lewis JS Jr, Zakharkin SO, Krebsbach PH, Nussenbaum B. rhBMP-2 has adverse effects on human oral carcinoma cell lines in vivo. Laryngoscope. 2012;122:95-102. 13. Soares AF, Xavier RL, da Costa Miguel MC, de Souza LB, Pinto LP. Bone morphogenetic protein-2/4 and bone morphogenetic protein receptor type IA expression in metastatic and nonmetastatic oral squamous cell carcinoma. Am J Otolaryngol. 2010;31:266-271. 14. Zhou X, Temam S, Oh M, et al. Global expression-based classification of lymph node metastasis and extracapsular spread of oral tongue squamous cell carcinoma. Neoplasia. 2006;8: 925-932. 15. Scantlebury JB, Luo J, Thorstad WL, et al. Cyclin D1—a prognostic marker in oropharyngeal squamous cell carcinoma that is tightly associated with high risk human papillomavirus status. Hum Pathol. 2013;44:1672-1680. 16. Ukpo OC, Thorstad WL, Lewis JS Jr. B7-H1 expression model for immune evasion in human papillomavirus-related oropharyngeal squamous cell carcinoma. Head Neck Pathol. 2013;7:113-121. 17. Ukpo OC, Flanagan JJ, Ma X, et al. High-risk human papillomavirus E6/E7 mRNA detection by a novel in-situ hybridization assay strongly correlates with p16 expression and patient outcomes in oropharyngeal squamous cell carcinoma. Am J Surg Pathol. 2011;35:1343-1350. 18. Lewis JS Jr, Ukpo OC, Flanagan JJ, et al. Transcriptionallyactive high risk human papillomavirus is rare in oral cavity and laryngeal/hypopharyngeal squamous cell carcinomas—a
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
250
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
Otolaryngology–Head and Neck Surgery 150(2) tissue microarray study utilizing E6/E7 mRNA in-situ hybridization. Histopathology. 2012;60:982-991. Raida M, Clement JH, Leek RD, et al. Bone morphogenetic protein 2 (BMP-2) and induction of tumor angiogenesis. J Cancer Res Clin Oncol. 2005;1:1-10. Clement JH, Raida M, Sanger J, et al. Bone morphogenetic protein 2 (BMP-2) induces in vitro invasion and in vivo hormone dependent growth of breast carcinoma cells. Int J Oncol. 2005;27:401-407. Langenfeld EM, Calvano SE, Abou-Nukta F, et al. The mature bone morphogenetic protein-2 is aberrantly expressed in non– small cell lung carcinomas and stimulates tumor growth of A549 cells. Carcinogenesis. 2003;24:1445-1454. Langenfeld EM, Langenfeld J. Bone morphogenetic protein-2 stimulates angiogenesis in developing tumors. Mol Cancer Res. 2004;2:141-149. Kleeff J, Maruyama H, Ishiwata T, et al. Bone morphogenetic protein 2 exerts diverse effects on cell growth in vitro and is expressed in human pancreatic cancer in vivo. Gastroenterology. 1999;166:1202-1216. Horvath LG, Henshall SM, Kench JG, et al. Loss of BMP2, Smad8, and Smad4 expression in prostate cancer progression. Prostate. 2004;59:234-242. Sulzbacher I, Birner P, Trieb K, et al. The expression of bone morphogenetic proteins in osteosarcoma and its relevance as a prognostic parameter. J Clin Pathol. 2002;55:381-385. Orui H, Imaizumi S, Ogino T, et al. Effects of bone morphogenetic protein-2 on human tumor cell growth and differentiation: a preliminary report. J Orthoped Sci. 2000;5:600-604. Asano N, Yamakazi T, Seto M, et al. The expression and prognostic significance of bone morphogenetic protein-2 in patients with malignant fibrous histiocytoma. J Bone Joint Surg Br. 2004;86:607-612. Buckley S, Shi W, Driscoll B, et al. BMP4 signaling induces senescence and modulates the oncogenic phenotype of A549 lung adenocarcinoma cells. Am J Physiol Lung Cell Mol Physiol. 2004;286:L81-L86. Ro TB, Holt RU, Brenne A, et al. Bone morphogenetic protein5, -6, and -7 inhibit growth and induce apoptosis in human myeloma cells. Oncogene. 2003;23:3024-3032. Shepherd TG, Nachtigal MW. Identification of a putative autocrine bone morphogenetic protein signaling pathway in human ovarian surface epithelium and ovarian cancer cells. Endocrinology. 2003; 144:3306-3314.
31. Raida M, Sarbia M, Clement JH, et al. Expression, regulation and clinical significance of bone morphogenetic protein 6 in esophageal squamous cell carcinoma. Int J Cancer. 1999;83:38-44. 32. Ong DB, Colley SM, Norman MR, et al. Transcriptional regulation of a BMP-6 promoter by estrogen receptor alpha. J Bone Miner Res. 2004;19:447-454. 33. Haudenschild DR, Palmer SM, Moseley TA, et al. Bone morphogenetic protein (BMP)–6 signaling and BMP antagonist noggin in prostate cancer. Cancer Res. 2004;64:8276-8284. 34. Masuda H, Fukabori Y, Nakano K, et al. Increased expression of bone morphogenetic protein-7 in bone metastatic prostate cancer. Prostate. 2003;54:268-274. 35. Franzen A, Heldin N. BMP-7 induced cell cycle arrest of anaplastic thyroid carcinoma cells via p21 and p27. Biochem Biophys Res Comm. 2001;285:773-781. 36. Parr C, Jiang WG. Metastasis suppressor 1 (MTSS1) demonstrates prognostic value and anti-metastatic properties in breast cancer. Eur J Cancer. 2009;45:1673-1683. 37. Liu K, Wang G, Ding H, et al. Downregulation of metastasis suppressor 1 (MTSS1) is associated with nodal metastasis and poor outcome in Chinese patients with gastric cancer. BMC Cancer. 2010;10:428. 38. Vidal-Vanaclocha F, Alvarez A, Asumendi A, et al. Interleukin 1 (IL-1)–dependent melanoma hepatic metastasis in vivo: increased endothelial adherence by IL-1–induced mannose receptors and growth factor production in vitro. J Natl Cancer Inst. 1996;88:198-205. 39. Voronov E, Shouval DS, Krelin Y, et al. IL-1 is required for tumor invasiveness and angiogenesis. Proc Natl Acad Sci U S A. 2003;100:2645-2650. 40. Kejner AE, Burch MB, Sweeny L, et al. Bone morphogenetic protein 6 expression in oral cavity squamous cell cancer is associated with bone invasion [published online June 17, 2013]. Laryngoscope. 41. Brandwein-Gensler M, Teixeira MS, Lewis CM, et al. Oral squamous cell carcinoma: histologic risk assessment, but not margin status, is strongly predictive of local disease-free and overall survival. Am J Surg Pathol. 2005;29:167-178. 42. Karaoz E, Aksoy A, Ayhan S, et al. Characterization of mesenchymal stem cells from rat bone marrow: ultrastructural properties, differentiation potential and immunophenotypic markers. Histochem Cell Biol. 2009;132:533-546. 43. Felin JE, Mayo JL, Loos TJ, et al. Nuclear variants of bone morphogenetic proteins. BMC Cell Biol. 2010;11:20.
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
BMP-2 Expression Correlates with Local Failure in Head and Neck Squamous Cell Carcinoma
Otolaryngology– Head and Neck Surgery 2014, Vol. 150(2) 245–250 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2013 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599813513003 http://otojournal.org
Jordan P. Sand, MD1, Natalia A. Kokorina, MD1, Stanislav O. Zakharkin, PhD, James S. Lewis Jr, MD1,2, and Brian Nussenbaum, MD1
No sponsorships or competing interests have been disclosed for this article.
Keywords bone morphogenetic protein, squamous cell carcinoma
Abstract Objective. Preclinical data show that exogenous administration of recombinant human bone morphogenetic protein2 (rhBMP-2) to human oral carcinoma cell lines increases pathogenicity using a nude mouse model. The objectives of this study are to (1) describe the characteristics of baseline protein expression of BMP-2 in head and neck squamous cell carcinomas (HNSCC) and (2) determine if BMP-2 expression level correlates with worse oncologic outcomes. Study Design. Retrospective analysis of previously harvested patient samples. Setting. Academic medical center. Subjects. In total, 149 patients with oral cavity, oropharynx, larynx, and hypopharynx HNSCC treated between January 1, 1997, and December 31, 2004. Methods. A tissue microarray of HNSCC was assembled and immunohistochemistry for BMP-2 performed. Staining was quantified using a standardized scoring system. Specimens were dichotomized into high or low expression level. Statistical analyses using log-rank, Wilcoxon, and Fisher exact test were performed for associations between BMP-2 protein level and clinicopathologic features and patient survival. Results. BMP-2 expression at any level was noted in 146 of 149 (98%) of samples. Tumors with high BMP-2 expression had higher rates of local failure compared with lowexpressing tumors (17.3% vs 6.3%; P = .04). There was no significant association for BMP-2 expression level with tumor location, T stage, N stage, overall survival, regional failure, or distant failure. Conclusion. Head and neck squamous cell carcinomas with high baseline BMP-2 protein level are associated with higher rates of local recurrence. These data have important implications for using rhBMP-2 in tissue engineering reconstructive approaches in the setting of cancer-related defects.
Received September 15, 2013; revised October 14, 2013; accepted October 25, 2013.
O
steoinductive therapy using recombinant human bone morphogenetic protein-2 (rhBMP-2) was approved by the Food and Drug Administration (FDA) in 2007 for use in alveolar ridge and maxillary sinus floor augmentation. In addition, investigators have demonstrated promising clinical data using rhBMP-2 for mandibular reconstruction, dental implant stabilization, healing of LeFort osteotomies, and repair of cleft palate defects.1-9 As knowledge, technology, and experience advance in this area, there might be a paradigm shift in reconstructive techniques from current therapies using autologous bone grafts toward tissue engineering using bone morphogenetic proteins (BMPs). The anticipated advantages are avoidance of donor site morbidity, decreased operative time, decreased surgical complications, shorter hospitalization, quicker recovery from surgery, and more accurate structural restoration of surgical defects.9 Although advances using biologic therapies for reconstruction are emerging and beginning to be translated to patients, little is known about the biologic effects of BMPs on head and neck squamous cell carcinoma (SCC). This is a critically important point since most patients with major craniomaxillofacial and mandible bone defects that would
1 Department of Otolaryngology–Head and Neck Surgery, Washington University School of Medicine, St Louis, Missouri, USA 2 Department of Pathology and Immunology, Washington University School of Medicine, St Louis, Missouri, USA
This article was presented at the 2013 AAO-HNSF Annual Meeting & OTO EXPO; September 29 to October 3, 2013; Vancouver, BC, Canada. Corresponding Author: Brian Nussenbaum, MD, Christy J. and Richard S. Hawes III, Professor, Department of Otolaryngology–Head and Neck Surgery, Washington University School of Medicine, 660 South Euclid Ave, CB 8115, St Louis, MO 63110, USA. Email: [email protected]
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
246
Otolaryngology–Head and Neck Surgery 150(2)
Figure 1. (A) Typical keratinizing-type oral cavity squamous cell carcinoma (3100 magnification). (B-D) Bone morphogenetic protein-2 immunohistochemistry showing (B) no tumor staining; (C) diffuse, strong tumor staining; and (D) minimal staining in tumor-adjacent tonsil crypt epithelium. (B, C, D: 3200 magnification).
benefit from a tissue engineering approach occur in the setting of head and neck cancer. BMP-2 expression has been very prevalent in oral carcinomas.10 Our group has previously shown that oral squamous cell carcinoma cell lines treated with rhBMP-2 revealed increased invasiveness and pathogenicity.11,12 In addition, BMP-2 gene expression has been associated with regional metastasis in oral squamous cell carcinoma.13,14 The primary objectives of this study are to (1) evaluate baseline patterns of BMP-2 protein expression level in head and neck squamous cell carcinoma and (2) determine if the level of the protein correlates with clinical characteristics, pathologic features, or oncologic outcomes.
Methods Patient Cohort Approval from the Human Research Protection Office (institutional review board) at Washington University School of Medicine was obtained for conducting this study. A tissue microarray (TMA) was constructed by our study pathologist (J.S.L.) from newly diagnosed primary cases of head and neck SCC treated between January 1, 1997, and December 31, 2004. Primary tumor sites included oral cavity, oropharynx, hypopharynx, and larynx. The SCC variants—namely, verrucous, spindle cell, papillary, and basaloid—were excluded. All patients were
treated with curative intent with surgery, surgery with adjuvant therapy, radiation therapy, or chemoradiation. None of the patients had received prior treatment for a head and neck carcinoma. Formalin-fixed and paraffinembedded tissues were used for the tissue microarray. According to the amount of available biopsied or resected tumor tissue, duplicate 2-mm (or, if only limited amounts of tumor tissue present, 0.6-mm) punches were taken. Since most of the cases were treated with primary surgery, the majority of cases on the array had the larger (2-mm) punches. Clinical details and follow-up information, including survival and recurrence data, were obtained from clinician databases and electronic medical records. This TMA has been used in several previous studies, the results of which have largely matched the expected findings for a contemporary cohort of HNSCC.15-18
Immunohistochemistry Immunohistochemistry was performed for BMP-2 using the commercially available antibody from Santa Cruz Biotechnology (Dallas, TX; N-14: sc-6895; goat polyclonal) at a dilution of 1:75 with EDTA and heat antigen retrieval. Positive control staining was observed in mouse jaw tissue (positive expression in osteoblasts and odontoblasts). A negative control slide of normal human cerebellum was also used, which showed no staining. The stained tissue microarray
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
Sand et al
247
slides were then digitally scanned on an Aperio Scanscope XT digital slide scanner (Vista, California) at 3400 magnification and read visually from the digital images. The study pathologist (J.S.L.) scored the staining, blinded to all clinical and pathologic features. Cases in which there was less than 10% surface area consisting of tumor (averaged across the 2 microarray punches) were excluded from analysis. Immunohistochemical staining was graded on the basis of distribution in a quartile manner (0; 11, 1%-25% staining; 21, 26%-50% staining; 31, 51%-75% staining; and 41, 76%-100% staining). Grading was also scored on intensity (11, weak staining; 21, moderately intense staining; and 31, strong staining). A quantitative score of 0 or 2 through 7 was obtained by adding the distribution and intensity scores. Tumor samples were then divided into 2 groups: high level of protein expression (composite score of 6 or 7) and low level (composite score of 0, 2, 3, 4, or 5) for comparative analysis (Figure 1A-C).
Statistical Analysis Survival analysis was performed using the Kaplan-Meier method. The differences noted within survival curves were compared using log rank and Wilcoxon tests. Associations between clinical and pathologic features were analyzed by the Fisher exact test. P values at or below .05 were considered significant. All analyses were done using SAS 9.2 (SAS Institute, Cary, North Carolina).
Results A total of 149 tumors were tested for BMP-2 protein expression with immunohistochemistry. Mean follow-up time for these patients was 50 months. Twelve patients had clinical information missing with regard to their T or N stage and thus were excluded from those analyses. BMP-2 expression at any level was noted in 146 of 149 (98%) analyzed samples. Table 1 summarizes the patient characteristics with regard to clinical and pathologic variables. The TMA was specifically designed to isolate tumor-containing tissue. However, there are some areas of normal mucosa and tonsillar crypt epithelium present in some of the punches. These showed minimal staining, with rare patchy nuclear positivity in normal squamous cells (Figure 1D). Tumors with high BMP-2 protein expression had a higher rate of local failure compared with low-expressing tumors (17.3% vs 6.3%; P = .04). This is graphically shown in Figure 2. No significant correlation was observed between the level of BMP-2 protein and overall survival, regional failure, distant failure, T stage, or N stage (Table 2). These data were then reanalyzed controlling for disease site (oropharynx and nonoropharynx) and treatment type, with neither demonstrating a statistically significant difference in any of the previously mentioned variables.
Discussion The potential role of BMPs in malignant transformation and cancer progression remains poorly understood. Investigations into this area have proceeded in varying degrees in both in
Table 1. Clinical case characteristics (n = 149). Clinical Characteristics Subsites Oral cavity Oropharynx Larynx Hypopharynx Treatment type Surgery alone Surgery with adjuvant therapy Radiotherapy 6 chemotherapy Unknown T stage T1 T2 T3 T4 Unknown N stage N0 N1 N2 N3 Unknown BMP-2 expression High (score 6 or 7) Low (score 2-5) None (0)
No. (%) of Total Cases
36 (24.2) 74 (49.7) 29 (19.5) 10 (6.7) 13 (8.7) 117 (78.5) 17 (11.4) 2 (1.3) 27 (18.1) 42 (28.2) 36 (24.2) 33 (22.1) 11 (7.4) 35 (23.5) 24 (16.1) 75 (50.3) 3 (2.0) 12 (8.1) 69 (46.3) 77 (51.7) 3 (2.0)
Figure 2. Kaplan-Meier local recurrence curve comparing local failure probability of Bone morphogenetic protein-2 (BMP-2) high expression level vs BMP-2 low expression level over time. P value by log-rank test was .04.
vitro and in vivo studies, which have revealed discordant findings based on the tumor site, tumor type, and BMP studied.19-35 Most BMP-based bone tissue engineering approaches use a BMP-2 platform, either with protein or gene therapy. Previously reported data have shown that BMP-2 and the
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
248
Otolaryngology–Head and Neck Surgery 150(2)
Table 2. Bone morphogenetic protein-2 (BMP-2) expression-level associations with clinicopathologic variables. Clinical/Pathologic Variable Overall survival Local failure Regional failure Distant failure T stage T1 T2 T3 T4 N stage N0 N1 N2 N3
BMP-2 Expression Level
No. (%)
High Low High Low High Low High Low
41 41 12 5 7 6 7 11
(59.4) (51.3) (17.4) (6.3) (10.1) (7.5) (10.1) (13.8)
High Low High Low High Low High Low
9 18 22 22 17 19 19 14
(11.8) (19.8) (24.7) (23.2) (20.2) (20.7) (22.1) (16.1)
High Low High Low High Low High Low
20 16 9 15 34 41 3 0
(23.3) (18.2) (12.0) (17.2) (34.0) (36.3) (4.3)
receptor BMPR-IA were frequently expressed in human tumors as detected by immunohistochemistry.10 In addition, studies have shown that the BMP-2 gene is frequently expressed in oral SCC cell lines but not usually expressed in normal oral mucosa.12 Treatment of oral squamous cell carcinoma cell lines with rhBMP-2 resulted in increased invasiveness in vitro.11 Using a tumor xenograft in a nude mouse model, oral carcinoma cells treated with rhBMP-2 demonstrated more locally aggressive biologic behavior.12 These data suggest that further study of the possible oncologic role of BMP-2 in head and neck squamous cell carcinomas is warranted. This study demonstrates that baseline BMP-2 protein expression is found in most head and neck squamous cell carcinomas (98%). High levels of BMP-2 were considered when greater than 50% of the tumor had staining with strong intensity or greater than 75% of the tumor having staining with at least moderate staining intensity. In this study, a high level of BMP-2 protein expression was correlated with increased incidence of local failure. Notably, local failure was nearly 3 times more frequent in patients who were identified to have a higher level of BMP-2 expression in their tumors compared with patients who had a lower level. This finding correlates with the preclinical data showing that oral carcinoma cell
P Value for Association
.33 .04 .58 .62
.31
.16
lines treated with rhBMP-2 had more aggressive local biologic behavior, both in vitro and in vivo.11,12 Analysis with quantitative polymerase chain reaction (PCR) suggested that the molecular pathways of increased invasiveness in these cell lines was through upregulation of interleukin (IL)–1b, a contributor to invasion, and the downregulation of metastasis suppressor-1, a repressor of invasion and metastasis.11,36-39 Additional studies using a human phosphokinase array suggested that increased local invasiveness might be related to epithelial-mesenchymal transition, the JAK/STAT pathway, and the Wnt-b-catenin pathway.12 Recent data showed that increased baseline protein expression of BMP-6 in oral carcinomas might be related to local aggressiveness as measured by bone invasion.40 The current study was designed to determine if the results from prior studies investigating the biologic effects of BMP-2 in oral carcinoma cell lines consistently translated to patient outcome data. Future studies will focus on the mechanisms with consideration of bone invasion and other histopathologic features known to predict local aggressiveness.41 These variables were not included in the data collected on the TMA used for this study. The patients included in this study were representative of a typical head and neck cancer population in terms of tumor
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
Sand et al
249
site, T stage, and N stage. Weaknesses of this study include its retrospective nature and the possible lack of sufficient statistical power to discern a clinically significant relationship between BMP-2 expression and survival. Most patients were treated with a primary surgical-based approach. This can be considered a weakness of this study, since the findings might not be generalizable to patients primarily treated with nonsurgical therapies. The predominance of surgically treated patients is likely due to the easier accessibility of appropriate pretreatment tumor specimens for constructing the tissue microarray at our institution. There was a relatively even distribution of tumors with high and low scores for BMP-2. The scoring system that was used to categorize the high- and low-expressing tumors was not designed to obtain an even distribution into each category but rather was thought to be a logical method for categorizing these tumors. The BMP-2 immunohistochemistry staining demonstrated a nuclear rather than cytoplasmic pattern. Our positive and negative controls were appropriate, the staining was highly specific without background noise, and nuclear localization of BMP-2 has been described, particularly with the antibody clone used in this study.42,43 For head and neck squamous cell carcinomas, a high level of baseline protein expression of BMP-2 correlated with an increased risk for local recurrence. These data are consistent with available preclinical data showing that exogenous administration of rhBMP-2 to oral carcinoma cell lines affects the tumor cells by behaving more locally aggressive.12 Although overall survival, regional failure, and distant failure were not affected by level of BMP-2 expression, the available data confirm that caution needs to be applied when considering the use of BMP-2–based tissue engineering approaches for reconstructing cancer-related defects. Author Contributions Jordan P. Sand, analysis, drafting article, final approval; Natalia A. Kokorina, analysis, revising of article, final approval; Stanislav O. Zakharkin, study design, acquisition of data, interpretation of data, revising article, final approval; James S. Lewis Jr, analysis, interpretation of data, drafting, final approval; Brian Nussenbaum, study conception and design, analysis, interpretation of data, final approval.
Disclosures Competing interests: None. Sponsorships: None. Funding source: None.
References 1. Warnke PH, Coren AJ. First experiences with recombinant human bone morphogenetic protein-7 (osteogenic protein-1) in a human case in maxillofacial surgery. Plast Reconstr Surg. 2003;111:2471-2472. 2. Moghadam HG, Urist MR, Sandor GKB, Clokie CML. Successful mandibular reconstruction using a BMP bioimplant. J Craniofac Surg. 2001;12:119-127.
3. Ferretti C, Ripamonti U. Human segmental mandibular defects treated with naturally derived bone morphogenetic proteins. J Craniofac Surg. 2002;13:434-444. 4. Howell TH, Fiorellini J, Jones A, et al. A feasibility study evaluating rhBMP-2/absorbable collagen sponge device for local alveolar ridge preservation or augmentation. Int J Periodontics Restor Dent. 1997;17:124-139. 5. Boyne PJ, Marx RE, Nevins M, et al. A feasibility study evaluating rhBMP-2/absorbable collagen sponge device for maxillary sinus floor augmentation. Int J Periodontics Restor Dent. 1997;17:11-25. 6. Chin M, Ng T, Tom WK, et al. Repair of alveolar clefts with recombinant human bone morphogenetic protein (rhBMP-2) in patients with clefts. J Craniofac Surg. 2005;16:778-789. 7. Herford AS, Boyne PJ. Reconstruction of mandibular continuity defects with bone morphogenetic protein-2 (BMP-2). J Oral Maxillofac Surg. 2008;66:616-624. 8. Clokie CML, Sandor GKB. Reconstruction of 10 major mandibular defects using bioimplants containing BMP-7. J Can Dent Assoc. 2008;74:67-72. 9. Nussenbaum B, Krebsbach PH. The role of gene therapy for craniofacial and dental tissue engineering. Adv Drug Deliv Rev. 2006;58:577-591. 10. Jin Y, Tipoe GL, Liong EC, Lau TY, Fung PC, Leung KM. Overexpression of BMP-2/4, -5 and BMPR-IA associated with malignancy of oral epithelium. Oral Oncol. 2001;37:225-233. 11. Kokorina NA, Zakharkin SO, Krebsbach PH, Nussenbaum B. Treatment effects of rhBMP-2 on invasiveness of oral carcinoma cell lines. Laryngoscope. 2011;121:1876-1880. 12. Kokorina NA, Lewis JS Jr, Zakharkin SO, Krebsbach PH, Nussenbaum B. rhBMP-2 has adverse effects on human oral carcinoma cell lines in vivo. Laryngoscope. 2012;122:95-102. 13. Soares AF, Xavier RL, da Costa Miguel MC, de Souza LB, Pinto LP. Bone morphogenetic protein-2/4 and bone morphogenetic protein receptor type IA expression in metastatic and nonmetastatic oral squamous cell carcinoma. Am J Otolaryngol. 2010;31:266-271. 14. Zhou X, Temam S, Oh M, et al. Global expression-based classification of lymph node metastasis and extracapsular spread of oral tongue squamous cell carcinoma. Neoplasia. 2006;8: 925-932. 15. Scantlebury JB, Luo J, Thorstad WL, et al. Cyclin D1—a prognostic marker in oropharyngeal squamous cell carcinoma that is tightly associated with high risk human papillomavirus status. Hum Pathol. 2013;44:1672-1680. 16. Ukpo OC, Thorstad WL, Lewis JS Jr. B7-H1 expression model for immune evasion in human papillomavirus-related oropharyngeal squamous cell carcinoma. Head Neck Pathol. 2013;7:113-121. 17. Ukpo OC, Flanagan JJ, Ma X, et al. High-risk human papillomavirus E6/E7 mRNA detection by a novel in-situ hybridization assay strongly correlates with p16 expression and patient outcomes in oropharyngeal squamous cell carcinoma. Am J Surg Pathol. 2011;35:1343-1350. 18. Lewis JS Jr, Ukpo OC, Flanagan JJ, et al. Transcriptionallyactive high risk human papillomavirus is rare in oral cavity and laryngeal/hypopharyngeal squamous cell carcinomas—a
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
250
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
Otolaryngology–Head and Neck Surgery 150(2) tissue microarray study utilizing E6/E7 mRNA in-situ hybridization. Histopathology. 2012;60:982-991. Raida M, Clement JH, Leek RD, et al. Bone morphogenetic protein 2 (BMP-2) and induction of tumor angiogenesis. J Cancer Res Clin Oncol. 2005;1:1-10. Clement JH, Raida M, Sanger J, et al. Bone morphogenetic protein 2 (BMP-2) induces in vitro invasion and in vivo hormone dependent growth of breast carcinoma cells. Int J Oncol. 2005;27:401-407. Langenfeld EM, Calvano SE, Abou-Nukta F, et al. The mature bone morphogenetic protein-2 is aberrantly expressed in non– small cell lung carcinomas and stimulates tumor growth of A549 cells. Carcinogenesis. 2003;24:1445-1454. Langenfeld EM, Langenfeld J. Bone morphogenetic protein-2 stimulates angiogenesis in developing tumors. Mol Cancer Res. 2004;2:141-149. Kleeff J, Maruyama H, Ishiwata T, et al. Bone morphogenetic protein 2 exerts diverse effects on cell growth in vitro and is expressed in human pancreatic cancer in vivo. Gastroenterology. 1999;166:1202-1216. Horvath LG, Henshall SM, Kench JG, et al. Loss of BMP2, Smad8, and Smad4 expression in prostate cancer progression. Prostate. 2004;59:234-242. Sulzbacher I, Birner P, Trieb K, et al. The expression of bone morphogenetic proteins in osteosarcoma and its relevance as a prognostic parameter. J Clin Pathol. 2002;55:381-385. Orui H, Imaizumi S, Ogino T, et al. Effects of bone morphogenetic protein-2 on human tumor cell growth and differentiation: a preliminary report. J Orthoped Sci. 2000;5:600-604. Asano N, Yamakazi T, Seto M, et al. The expression and prognostic significance of bone morphogenetic protein-2 in patients with malignant fibrous histiocytoma. J Bone Joint Surg Br. 2004;86:607-612. Buckley S, Shi W, Driscoll B, et al. BMP4 signaling induces senescence and modulates the oncogenic phenotype of A549 lung adenocarcinoma cells. Am J Physiol Lung Cell Mol Physiol. 2004;286:L81-L86. Ro TB, Holt RU, Brenne A, et al. Bone morphogenetic protein5, -6, and -7 inhibit growth and induce apoptosis in human myeloma cells. Oncogene. 2003;23:3024-3032. Shepherd TG, Nachtigal MW. Identification of a putative autocrine bone morphogenetic protein signaling pathway in human ovarian surface epithelium and ovarian cancer cells. Endocrinology. 2003; 144:3306-3314.
31. Raida M, Sarbia M, Clement JH, et al. Expression, regulation and clinical significance of bone morphogenetic protein 6 in esophageal squamous cell carcinoma. Int J Cancer. 1999;83:38-44. 32. Ong DB, Colley SM, Norman MR, et al. Transcriptional regulation of a BMP-6 promoter by estrogen receptor alpha. J Bone Miner Res. 2004;19:447-454. 33. Haudenschild DR, Palmer SM, Moseley TA, et al. Bone morphogenetic protein (BMP)–6 signaling and BMP antagonist noggin in prostate cancer. Cancer Res. 2004;64:8276-8284. 34. Masuda H, Fukabori Y, Nakano K, et al. Increased expression of bone morphogenetic protein-7 in bone metastatic prostate cancer. Prostate. 2003;54:268-274. 35. Franzen A, Heldin N. BMP-7 induced cell cycle arrest of anaplastic thyroid carcinoma cells via p21 and p27. Biochem Biophys Res Comm. 2001;285:773-781. 36. Parr C, Jiang WG. Metastasis suppressor 1 (MTSS1) demonstrates prognostic value and anti-metastatic properties in breast cancer. Eur J Cancer. 2009;45:1673-1683. 37. Liu K, Wang G, Ding H, et al. Downregulation of metastasis suppressor 1 (MTSS1) is associated with nodal metastasis and poor outcome in Chinese patients with gastric cancer. BMC Cancer. 2010;10:428. 38. Vidal-Vanaclocha F, Alvarez A, Asumendi A, et al. Interleukin 1 (IL-1)–dependent melanoma hepatic metastasis in vivo: increased endothelial adherence by IL-1–induced mannose receptors and growth factor production in vitro. J Natl Cancer Inst. 1996;88:198-205. 39. Voronov E, Shouval DS, Krelin Y, et al. IL-1 is required for tumor invasiveness and angiogenesis. Proc Natl Acad Sci U S A. 2003;100:2645-2650. 40. Kejner AE, Burch MB, Sweeny L, et al. Bone morphogenetic protein 6 expression in oral cavity squamous cell cancer is associated with bone invasion [published online June 17, 2013]. Laryngoscope. 41. Brandwein-Gensler M, Teixeira MS, Lewis CM, et al. Oral squamous cell carcinoma: histologic risk assessment, but not margin status, is strongly predictive of local disease-free and overall survival. Am J Surg Pathol. 2005;29:167-178. 42. Karaoz E, Aksoy A, Ayhan S, et al. Characterization of mesenchymal stem cells from rat bone marrow: ultrastructural properties, differentiation potential and immunophenotypic markers. Histochem Cell Biol. 2009;132:533-546. 43. Felin JE, Mayo JL, Loos TJ, et al. Nuclear variants of bone morphogenetic proteins. BMC Cell Biol. 2010;11:20.
Downloaded from oto.sagepub.com at HOWARD UNIV UNDERGRAD LIBRARY on March 9, 2015
