Accepted Manuscript

Title: Prognostic value of matrix metalloproteinase 9expression in patients with juvenile nasopharyngealangiofibroma: Tissue microarray analysis

Author: Xicai Sun Limin Guo Jingjing Wang Huan WangZhuofu Liu Juan Liu Huapeng Yu Li Hu Han Li Dehui Wang

PII: S0165-5876(14)00195-5DOI: http://dx.doi.org/doi:10.1016/j.ijporl.2014.04.003Reference: PEDOT 7071

To appear in: International Journal of Pediatric Otorhinolaryngology

Received date: 13-2-2014Revised date: 30-3-2014Accepted date: 1-4-2014

Please cite this article as: X. Sun, L. Guo, J. Wang, H. Wang, Z. Liu, J. Liu,H. Yu, L. Hu, H. Li, D. Wang, Prognostic value of matrix metalloproteinase9 expression in patients with juvenile nasopharyngeal angiofibroma: Tissuemicroarray analysis, International Journal of Pediatric Otorhinolaryngology (2014),http://dx.doi.org/10.1016/j.ijporl.2014.04.003

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

Page 1 of 28

Accep

ted

Man

uscr

ipt

Prognostic value of matrix metalloproteinase 9 expression in patients with juvenile nasopharyngeal angiofibroma: Tissue microarray analysis

Xicai Sun, Limin Guo, Jingjing Wang, Huan Wang, Zhuofu Liu, Juan Liu,

Huapeng Yu, Li Hu, Han Li, Dehui Wang

Department of Otolaryngology,

Eye, Ear, Nose and Throat Hospital,

Shanghai Medical College, Fudan University, Shanghai, China.

Corresponding contributor:

Dehui Wang,

Department of Otolaryngology,

Eye, Ear, Nose and Throat Hospital, Shanghai Medical College, Fudan University,

83 Fenyang Road, Shanghai, 200031, China

Tel. +86 21 64 377 134-788

Fax +86 21 64 377 151

E-mail wangdehuient@sina.com

Running head: Juvenile Nasopharyngeal Angiofibroma

Drs Xicai Sun, Limin Guo, Jingjing Wang, and Huan Wang contributed equally to this

work.

Acknowledgements

This work was supported by the National Science Foundation of China grant

no. 81371077. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. 

Page 2 of 28

Accep

ted

Man

uscr

ipt

Abstract

Objective

Although JNA is a benign neoplasm histopathologically, it has a propensity for locally

destructive growth and remains a higher postoperative recurrence rate. The aim of this

study was to analyze the expression and localization of MMP-9 in JNA using tissue

microarray to elucidate its correlation with clinicopathological features and

recurrence.

Materials and methods

The expression of MMP-9 was assessed by immunohistochemistry in a tissue

microarray from 70 patients with JNA and 10 control subjects. Correlation between

the levels of MMP-9 expression and clinicopathologic variables, as well as tumor

recurrence, were analyzed

Results

MMP-9 was detected in perivascular and extravascular less differentiated cells

and stromal cells of patients with JNA but not in the matured vascular endothelial

cells of these patients. The presence of MMP-9 expression in JNA was correlated with

patient’s age (p=0.001). Spearman correlation analysis suggested that high expression

of MMP-9 in JNA had negative correlation with patient’s age (r=-0.412, p＜0.001).

The recurrence rate in JNA patients with high MMP-9 expression was significantly

higher than those with low MMP-9 expression (p = 0.002). In multivariate and ROC

curve analysis, MMP-9 was a good prognostic factor for tumor recurrence of JNA.

Conclusion

Higher MMP-9 expression is a poor prognostic factor for patients with JNA who have

been surgically treated.

Key words

Juvenile nasopharyngeal angiofibroma (JNA); Matrix metalloproteinase

9(MMP-9); Tissue microarray (TMA); Prognosis

Page 3 of 28

Accep

ted

Man

uscr

ipt

INTRODUCTION

Juvenile nasopharyngeal angiofibroma (JNA) is a rare, nonencapsulated, highly

vascularized benign tumor of the nasopharynx. It accounts for less than 0.05% of all

head and neck tumors, and primarily affects adolescent males [1-3]. Although

histologically benign in appearance, JNA has a propensity for locally destructive

growth with bone erosion. JNA is considered to originate from the superior margin of

the sphenopalatine foramen or pterygoid canal [4]. Skull base invasion is present in

20% and intracranial involvement in 10-20% of all patients with JNA and are more

frequent in younger adolescents [5-8]. 

Surgery is the mainstay of the treatment，but the best approach to JNA has been a

dilemma to the otorhinolaryngologist. The recurrence rate after surgical excision of

JNA is high and fairly heterogeneous, varying from 20 to over 50% [5, 7, 9-13].

The factors of recurrence of JNA include age at diagnosis, extension to neighboring

structures, tumour size, previous treatment approaches, and Radkowski classification

[7, 10]. 

Although its exact nature and pathogenesis is remaining unknown, JNA is thought to

have a vascular origin [1, 7]. Angiogenesis is an essential component of solid

tumor growth and metastasis [14]. Recent studies showed that MMPs had a vital role

in tumor invasion and angiogenesis. Among all of the MMPs, MMP-9 is considered

to play a critical role in tumor aggressiveness and metastasis. Increased MMP-9

levels were associated with poor prognosis in nasopharyngeal carcinoma, non-small

cell lung cancer, breast cancer and gastrointestinal cancer [15-18]. So, the aim of this

study was to analyze the expression and localization of MMP-9 in JNA using tissue

microarray to elucidate its correlation with clinicopathological features and disease

outcome. 

Page 4 of 28

Accep

ted

Man

uscr

ipt

MATERIALS AND METHODS

Patients and tissue samples

This study protocol was approved by the institutional review board of Fudan

University. The study included 70 cases with JNA. All clinical data were obtained

from patients’ medical records. Patients with histologically proven JNA underwent

surgery between 2003 and 2007 at the Eye, Ear, Nose and Throat Hospital of Fudan

University. All patients with JNA were male and ranged in age from 9 to 41 years old

(median age, 17 years) at the time of surgery. The follow-up period ranged from 69 to

127 months, with a mean of 95 months. The clinicopathological characteristics of all

patients are shown in Table 1. Preoperative evaluation of paranasal computed

tomography (CT) and/or MRI was used to determine the extent of

tumor growth. Tumor size was determined by measuring the major diameter of the

lesion observed on a CT scan or MRI. Tumor staging was performed according to the

classification of Radkowski and colleagues [19]. Completeness of the resection was

defined when no macroscopic residual tumor remained and was further confirmed

with surgical bed histology, as well as postoperative CT scan or MRI scan. A

diagnosis of relapse was based on clinical or imaging study demonstrating tumor

presence in the nasopharynx or neighboring structures, with confirmed major

symptoms including epistaxis or nasal obstruction after the first surgical treatment.

Different hospitals have different therapeutic principles and surgical techniques for

patients with JNA. Therefore, the history of previous operations at other hospitals was

not considered for the relapse analysis in this study. 

Tissue microarray

The preparation of tissue microarray for JNA was reported in our previous report [7].

Briefly, a total of 70 formalin-fixed (buffered neutral aqueous 4% formalin solution),

paraffin-embedded tumors were obtained from the Department of Otolaryngology–

Head and Neck Surgery, Eye, Ear, Nose, and Throat Hospital, Fudan

University, China. A representative paraffin block was selected from each case after

Page 5 of 28

Accep

ted

Man

uscr

ipt

histologic review of slides stained with hematoxylin – eosin

(H&E). Two representative tumor areas were selected, and the corresponding areas

were marked on the surface of the standard paraffin block. Tissue cores were punched

from the designated area using a biopsy needle. The normal middle turbinates of 10

male patients with a median age of 21.4 years (range, 13–26 years) were used as

normal controls. Two cores were taken from each representative tissue, and all tissue

cores (1.5 mm in diameter) were put into a tissue microarray (TMA) block that

covered up to 160 cores. Samples were processed in collaboration with Shanghai

Biochip Company, Shanghai, China. Serial 4 μ m sections were placed on

3-aminopropyltriethoxysilane-coated slides. 

Immunohistochemical study and microscopic analysis

Immunodetections of MMP-9 was performed using monoclonal

antibodies (1:100; Abcam, UK) with a microwave antigen retrieval method. 0.01

M phosphate-buffered saline (PBS) (pH 7.4) was used as the dilutions of all the

antibodies. Tissue sections were deparaffinized and rehydrated. Endogenous

peroxidase was blocked by placing slides in 3% hydrogen peroxide in methanol for

15 min. Slides were then washed with deionized water and placed in PBS for 5 min.

After preincubation with 10% normal serum in 1% bovine albumin

phosphate-buffered saline solution for 30 min to avoid non-specific binding, the

primary antibodies were stored overnight at 4℃. Slides were washed three times with

PBS and consecutively incubated with biotinylated secondary antibody for 30

min; again washed three times with PBS and incubated with streptavidin horseradish

peroxidase conjugate for 20 min; washed again three times with PBS.

Detection system with 0.05% diaminobenzidine (DAB) as chromagen (DAB; DAKO,

Denmark) was used. Then, the samples were counterstained with hematoxylin for 2

seconds, washed, dehydrated, and mounted. Negative controls, which were included

in all assays, were treated identically to the samples without the addition of primary

Page 6 of 28

Accep

ted

Man

uscr

ipt

antibodies.

The intensity of the immunoreactivity and the percentage of positive cells were

initially estimated semi-quantitatively by light microscopy in stromal cells of

JNA. The overall immunoreactivity in stromal cells was scored in a 4-tier scale

system based on the percentage of positive cells and the intensity of immunoreactivity

in a semi-quantitative method as previously described[1, 20]. Immunoreactivity was

scored as follows: - = negative result; + = weak immunoreactivity regardless of the

percentage of the cells being positive; ++ = moderate immunoreactivity in less

than 75% or strong immunoreactivity in less than 25% of the cells; and +++ =

moderate immunoreactivity in 75% or more or strong immunoreactivity in 25% or

more of the cells. Only cytoplasmic reactivity was considered specific for

MMP-9. For further analysis, the patients were divided into two groups, low MMP-9

expression (0—+) versus high MMP-9 expression (++—+++). 

 

Microvessel density（MVD） count was carried out using immunostaining with the

anti-CD105 antibody, as described in our previous report[7]. The vascular hot

spot technique was used according to Weidner et al [21]. For further survival

analysis, MVD was classified into two groups, high(≥10 vessels per ×200 visual field)

versus low(＜10 vessels per ×200 visual field). 

Statistics

The association between clinicopathologic variables and expression of MMP-9 in

JNA were analyzed by Pearson’s chi-square test. Overall survival was calculated by

the Kaplan–Meier survival curves. The impact of the expression of MMP-9 on the

time to recurrence of JNA patients was investigated by Kaplan–Meier survival curves,

and the difference was determined by the log-rank test. The Cox regression

model served for multivariate survival analysis. Receiver operating characteristic

Page 7 of 28

Accep

ted

Man

uscr

ipt

(ROC) curve analysis was used to determine the predictive value of the parameters. A

p values < .05 were considered statistically significant. All statistical analyses were

performed with SPSS 19.0 statistical software (SPSS, Chicago, IL).

Results

MMP-9 expression detected by JNA tissue microarray

JNA was composed of vascular and stromal components. Vessels in JNA have a

bicellular characteristics consisted of endothelial and perivascular cells. There are a

lot of perivascular and extravascular less differentiated cells between pericytes and

myofibroblasts. From the immunohistochemistry of TMA experiments, MMP-9 was

detected in perivascular and extravascular less differentiated cells and stromal cells of

patients with JNA but not in the matured vascular endothelial cells of these patients

(Figure 1). Immunohistochemical analysis of MMP-9 in patients with JNA and

normal middle turbinate are summarized in Table 2. There was a significant

difference in expression of MMP-9(p < 0.05, Pearson’s chi-square test) between JNA

and normal middle turbinate.

Association between clinicopathologic variables and levels of MMP-9 expressions

To elucidate the biologic significance of MMP-9 expression in patients with JNA, the

correlation between clinicopathologic variables and levels of MMP-9 expressions was

analyzed (Table 3). History of JNA operation (p=  0.931), operation approach (p= 

0.934), tumor stage (p= 0.563), and volume of intraoperative hemorrhage (p= 0.732)

didn’t associate with the expression of MMP-9 in patients with JNA. However, the

presence of MMP-9 expression in JNA was correlated with patient’s age (p=0.001).

Furthermore, Spearman correlation analysis suggested that high expression of MMP-9

in JNA had negative correlation with patient’s age (r=-0.412, p＜0.001, Figure 2).

The correlation between tumor recurrence of JNA and levels of MMP-9 expressions

was also analyzed. Until August 31, 2013, complete follow-up data were available

for 55 patients, all of whom were eligible for further analyses (Table 4 and Table 5).

Page 8 of 28

Accep

ted

Man

uscr

ipt

Statistical analysis revealed that patients with higher MMP-9 expression had

significantly higher recurrence rates when compared to those with low MMP-9

expression (p=0.018). Moreover, patients with advanced tumor staging (>stage IIa)

showed significantly higher recurrence rates than those with lower tumor staging

(p=0.016). However, patient's age (p=0.054), operation history (p=0.171), operation

approach (p=0.529), and volume of intraoperative hemorrhage (p=0.054) didn’t

associated with the tumor recurrence.

Prognostic roles of MMP-9 and tumor stage in patients with JNA

Until August 31, 2013, twenty-one patients (38.18%) exhibited recurrence, with the

mean relapse time between surgical intervention and tumor recurrence being 21.38±

23.52 months. In univariate analysis, the 1-year, 3-year, and 5-year recurrence rates

were 20.00%, 32.70%, and 36.40%, respectively. The recurrence rate for patients of

JNA with high MMP-9 expression was 57.14%, with 1-year, 3-year, and 5-year

recurrence rates of 32.10%, 57.10%, and 57.10%, respectively. However, the

recurrence rate was only 18.52% in patients of JNA with low MMP-9 expression, and

these patients exhibited 1-year, 3-year, and 5-year recurrence rates of 7.40%, 7.40%,

and 14.80%, respectively. The recurrence rate in JNA patients with high MMP-9

expression was significantly higher than those with low MMP-9 expression (p =

0.002).

Univariate and multivariate analyses were performed using Cox proportional hazard

model to evaluate the impact of MMP-9 and clinicopathological factors on the time to

recurrence (TTR) of JNA patients (Table 6). In univariate analysis, factors that

influenced the time to recurrence were MMP-9(p=0.004), MVD (p=0.015), and age

(p=0.039). In multivariate analyses, higher MMP-9 level, higher MVD

and advanced-stage tumors(＞stage IIa) were associated with the time to recurrence.

Of all clinicopathological features, MMP-9 level and MVD were

independent prognostic factors for TTR. Moreover, Kaplan–Meier analysis also

Page 9 of 28

Accep

ted

Man

uscr

ipt

indicated that patients with higher MMP-9 level, higher MVD, advanced-stage tumors,

and younger age had significantly shorter median TTRs than the patients with lower

MMP-9 level and MVD, early-stage tumors, and older age (Figure 3). In

addition, receiver operating characteristic (ROC) analysis also showed that

MMP-9 was the best predictor for tumor recurrence (Figure 4).

Disccusion

Despite tremendous progress in treatment of JNA, the recurrence rate after surgical

treatment remains high. Previous reports on JNA showed fairly

heterogeneous recurrence rates (20–50%)[5, 7, 10-13, 22, 23]. In the present study,

the overall recurrence rate was 38.18% (21 of 55) in the 55 patients with JNA, and the

mean recurrence time was 21.38 months after the operation, which was consistent

with our previous reports [5, 7, 10]. So, finding the prognostic factors for recurrence

of JNA is useful, which can help doctors select the appropriate treatment according to

the status of individual patient. 

Currently, several clinicopathological factors have been evaluated with the recurrence

of JNA, including age at diagnosis, extension sites, tumour size, preoperative

embolization, tumor stage, surgical approaches, previous treatments, and

intraoperative blood loss [24-26]. But these prognostic factors for JNA

recurrence remain controversial and have not been well established. Accumulating

evidence showed that some biological factors involved in oncogenesis may play an

essential role in the prognosis of JNA and seemed more specific than markers

currently used in clinical such as age at diagnosis, tumor stage, and surgical

approaches [7, 17]. Our previous report suggested that microvessel density MVD is a

useful predictor for poor prognosis of patients with JNA after curative resection [7]. 

The extracellular matrix (ECM) plays a key role in both normal and disease process as

angiogenesis, inflammation, and tumor invasion [27]. Matrix metalloproteinase 9

Page 10 of 28

Accep

ted

Man

uscr

ipt

(MMP-9), as a member of matrix metalloproteinases (MMPs), involved in

extracellular matrix cleavage, was shown to be associated with tumor progression in

several solid tumors [15-17, 28, 29]. To date, the cell types involved in the

production of MMP-9 in the tumor progression have not been identified

definitely. There have been rare reports found regarding researches associated with

the MMP-9 in JNA. Duerr et al [30] detected the expression of metalloproteinases in

9 juvenile angiofibroma and 2 inferior nasal turbinate specimens, and

found significantly elevated expression levels for MMP-9 in juvenile

angiofibromas compared with inferior nasal turbinates at the transcriptional and

translational levels. Moreover, immunofluorescence localization of MMP-9 was

found in the stromal compartment and in the vessels of juvenile angiofibromas. In our

study, MMP-9 was detected in perivascular and extravascular less differentiated cells

and stromal cells of patients with JNA but not in the matured vascular endothelial

cells of these patients. These findings suggested that the production of MMP-9 in the

JNA might play an active role in oncogenesis and tumor invasion. 

The result of the present study demonstrated that the recurrence rate in JNA patients

with high MMP-9 expression was significantly higher than those with low MMP-9

expression. Tumor stage is an important prognostic indicator for JNA [5, 7,

10]. The results of our study showed that patients with advanced tumor

staging (>stage IIa) showed significantly higher recurrence rates than those with

lower tumor staging. However, tumor stage didn’t associate with the expression of

MMP-9 in patients with JNA. This result may be caused by a small number of

patients collected in our study, especially those with early stage tumors. So, we

cannot perform subgroup analysis by Radkowski staging system. Moreover, our study

found that the presence of MMP-9 expression in JNA was correlated with patient’s

age, as the expression of MMP-9 in JNA had negative correlation with patient’s

age. Our previous reports demonstrated that age was one of the factors that had an

effect on recurrence [5, 7, 10]. Whether this result is caused by its effect on the

expression of MMP-9 need to be further clarified in future studies. In this study, we

Page 11 of 28

Accep

ted

Man

uscr

ipt

further determined whether overexpression of MMP-9 is associated with prognosis in

JNA patients. In univariate analysis, patients with overexpression of MMP-9 had

significantly poorer prognosis than patients with low MMP-9. Moreover,

we demonstrated that using multivariate and ROC curve analysis, MMP-9 showed to

be an independent prognostic factor for JNA.

MMP-9 activity can be regulated at multiple levels including gene transcription,

posttranslational activation of zymogens, and the interaction with endogenous

inhibitors, especially the tissue inhibitor of metalloproteinase (TIMP)-1. It has been

shown that genetic factors contribute significantly to the variability in MMP-9 levels.

Belo et al [31] studied 175 healthy control children and 127 obese children, and found

that two functional MMP-9 genetic polymorphisms affect the circulating MMP-9

levels in obese children and adolescents. Whether MMP-9 gene polymorphisms or

haplotypes affect MMP-9 levels in JNA patients need to be further studied in the

future. MMP-9 genetic variants can also affect the responsiveness to therapy in

several diseases. In a study of hypertensive disorders of pregnancy, it has been found

that MMP-9 haplotypes affect the susceptibility to hypertensive disorders of

pregnancy and the responsiveness to antihypertensive therapy in these conditions [32].

Although the effects of MMP-9 polymorphisms on the overall therapeutic responses

including surgery and recurrence are not evaluated in our study, it is possible that the

responsiveness to surgery and the overall recurrence may be affected by MMP-9

genetic variants in JNA patients.

As an important endogenous MMP-9 inhibitor, TIMP-1 may play a role in the therapy

of JNA as an adjuvant to surgery. The equilibrium between MMPs and TIMPs

determines extracellular matrix degradation, and alterations in the MMPs/TIMPs

ratios are implicated in the pathophysiology of a variety of diseases. Therefore, the

assessment of MMP-9/TIMP-1 may be a better marker of net MMP activities. A

research carried out in polycystic ovary syndrome (PCOS) has indicated that MMP-9

and TIMP-1 levels were similar in PCOS patients and in healthy controls. However,

Page 12 of 28

Accep

ted

Man

uscr

ipt

patients with PCOS have higher MMP-9/TIMP-1 ratios when compared with healthy

controls [33]. A limitation of our study is that we have not measured the expression of

TIMP-1and the MMP-9/TIMP-1 ratio. For providing much improved information,

additional studies are needed to be carried out to examine the expression of TIMP-1

and the MMP-9/TIMP-1 ratio in JNA patients.

In conclusion, our results demonstrate that significantly higher level of MMP-9 is

a poor prognostic factor for patients with JNA who have been surgically treated. The

overexpression of MMP-9 might be useful to identify patients at higher risk for

recurrence of JNA. As a potentially useful biomarker for predicting prognosis in JNA,

MMP-9 expression can help clinicians select the appropriate treatment according to

the status of individual patient. In addition, further prospective studies with more

samples are required to investigate the prognostic effect of MMP-9 in JNA.

Page 13 of 28

Accep

ted

Man

uscr

ipt

References

[1] M. Zhang, X. Sun, H. Yu, L. Hu, D. Wang, Biological distinctions between

juvenile nasopharyngeal angiofibroma and vascular malformation: an

immunohistochemical study, Acta Histochem 113 (2011) 626-630.

[2] J.W. Bremer, H.B. 3rd Neel, L.W. DeSanto, G.C. Jones, Angiofibroma: treatment

trends in 150 patients during 40 years, Laryngoscope 96 (1986) 1321-1329. 

[3] P.J. Gullane, J. Davidson, T. O'Dwyer, V, Forte, Juvenile angiofibroma: a review

of the literature and a case series report, Laryngoscope 102 (1992) 928-933. 

[4] Z.F. Liu, D.H. Wang, X.C. Sun, J.J. Wang, L. Hu, H. Li, P.D. Dai, The site of

origin and expansive routes of juvenile nasopharyngeal angiofibroma (JNA), Int J

Pediatr Otorhinolaryngol 75 (2011) 1088-1092. 

[5] Y. Huang, Z. Liu, J. Wang, X. Sun, L. Yang, D. Wang, Surgical management of

juvenile nasopharyngeal angiofibroma: Analysis of 162 cases 1995-2012,

Laryngoscope (2013) doi: 10.1002/lary.24522. [Epub ahead of print] 

[6] P. Nicolai, A. Schreiber, A. Bolzoni Villaret, Juvenile angiofibroma: evolution of

management, Int J Pediatr (2012); 2012:412545. doi: 10.1155/2012/412545. Epub

2011 Nov 17. 

[7] J.J. Wang, X.C. Sun, L. Hu, Z.F. Liu, H.P. Yu, H. Li, S.Y. Wang, D.H. Wang,

Endoglin (CD105) expression on microvessel endothelial cells in juvenile

nasopharyngeal angiofibroma: Tissue microarray analysis and association with

Page 14 of 28

Accep

ted

Man

uscr

ipt

prognostic significance, Head Neck 35 (2013) 1719-1725. 

[8] A.W. Scholtz, E. Appenroth, K. Kammen-Jolly, L.U. Scholtz, W.F. Thumfart,

Juvenile nasopharyngeal angiofibroma: management and therapy, Laryngoscope 111

(2001) 681-687. 

[9] M. Naraghi, A. Kashfi, Endoscopic resection of nasopharyngeal angiofibromas by

combined transnasal and transoral routes, Am J Otolaryngol 24 (2003) 149-154. 

[10] X.C. Sun, D.H. Wang, H.P. Yu, F. Wang, W. Wang, J.J. Jiang, Analysis of risk

factors associated with recurrence of nasopharyngeal angiofibroma, J Otolaryngol

Head Neck Surg 39 (2010) 56-61. 

[11] J.F. Carrillo, F. Maldonado, O. Albores, M.C. Ramírez-Ortega, L.F. Oñate-Ocaña,

Juvenile nasopharyngeal angiofibroma: clinical factors associated with recurrence,

and proposal of a staging system, J Surg Oncol 98 (2008) 75-80. 

[12] P.H. Roche, J. Paris, J. Régis, G. Moulin, M. Zanaret, J.M. Thomassin, W. Pellet,

Management of invasive juvenile nasopharyngeal angiofibromas: the role of a

multimodality approach, Neurosurgery 61 (2007) 768-777; discussion 777. 

[13] D.J. Howard, G. Lloyd, V. Lund, Recurrence and its avoidance in juvenile

angiofibroma, Laryngoscope 111 (2001) 1509-1511. 

[14] B.R. Zetter, Angiogenesis and tumor metastasis, Annu Rev Med 49 (1998)

407-424. 

[15] D. Cheng, H. Kong, Y. Li, Prognostic value of interleukin-8 and MMP-9 in

nasopharyngeal carcinoma, Eur Arch Otorhinolaryngol (2013) Jun 9. [Epub ahead of

print] 

[16] W.J. Peng, J.Q. Zhang, B.X. Wang, H.F. Pan, M.M. Lu, J. Wang, Prognostic value

of matrix metalloproteinase 9 expression in patients with non-small cell lung cancer,

Clin Chim Acta 413 (2012) 1121-116. 

[17] J. Song, H. Su, Y.Y. Zhou, L.L. Guo, Prognostic value of matrix

metalloproteinase 9 expression in breast cancer patients: a meta-analysis, Asian Pac J

Cancer Prev 14 (2013) 1615-1621. 

[18] C.Y. Li, P. Yuan, S.S. Lin, C.F. Song, W.Y. Guan, L. Yuan, R.B. Lai, Y. Gao, Y.

Wang, Matrix metalloproteinase 9 expression and prognosis in colorectal cancer: a

Page 15 of 28

Accep

ted

Man

uscr

ipt

meta-analysis, Tumour Biol 34(2013) 735-741. 

[19] D. Radkowski, T. McGill, G.B. Healy, L. Ohlms, D.T. Jones, Angiofibroma.

Changes in staging and treatment, Arch Otolaryngol Head Neck Surg 122 (1996)

122-129. 

[20] P.J. Zhang, R. Weber, H.H. Liang, T.L. Pasha, V.A. LiVolsi, Growth factors and

receptors in juvenile nasopharyngeal angiofibroma and nasal polyps: an

immunohistochemical study, Arch Pathol Lab Med 127 (2003) 1480-1484. 

[21] N. Weidner, J.P. Semple, W.R. Welch, J. Folkman, Tumor angiogenesis and

metastasis--correlation in invasive breast carcinoma, N Engl J Med 324 (1991) 1-8. 

[22] M.A. Khalifa, S.M. Ragab, Endoscopic assisted antral window approach for

type III nasopharyngeal angiofibroma with infratemporal fossa extension, Int J Pediatr

Otorhinolaryngol 72 (2008) 1855-1860. 

[23] M. Zhang, W. Garvis, T. Linder, U. Fisch, Update on the infratemporal fossa

approaches to nasopharyngeal angiofibroma, Laryngoscope 108 (1998) 1717-1723. 

[24] K.H. Chan, D. Gao, P.G. Fernandez, T.T. Kingdom, D.A. Kumpe, Juvenile

nasopharyngeal angiofibroma: Vascular determinates for operative complications and

tumor recurrence, Laryngoscope (2013) Aug 8. doi: 10.1002/lary.24337. [Epub ahead

of print] 

[25] K. Petruson, M. Rodriguez-Catarino, B. Petruson, C. Finizia, Juvenile

nasopharyngeal angiofibroma: long-term results in preoperative embolized and

non-embolized patients, Acta Otolaryngol 122(2002) 96-100. 

[26] Z. Boghani, Q. Husain, V.V. Kanumuri, M.N. Khan, S. Sangvhi, J.K. Liu, J.A.

Eloy, Juvenile nasopharyngeal angiofibroma: a systematic review and comparison of

endoscopic, endoscopic-assisted, and open resection in 1047 cases, Laryngoscope 123

(2013) 859-869. 

[27] E.A. McKenzie, Heparanase: a target for drug discovery in cancer and

inflammation, Br J Pharmacol 151 (2007) 1-14. 

[28] C.S. Moore, S.J. Crocker, An alternate perspective on the roles of TIMPs and

MMPs in pathology, Am J Pathol 180 (2012) 12-16. 

[29] H. Nagase, J.F. Woessner Jr, Matrix metalloproteinases, J Biol Chem 274 (1999)

Page 16 of 28

Accep

ted

Man

uscr

ipt

21491-21494. 

[30] S. Duerr, O. Wendler, T. Aigner, S. Karosi, B. Schick, Metalloproteinases in

juvenile angiofibroma--a collagen rich tumor, Hum Pathol 39(2008) 259-268.

[31] Belo VA, Souza-Costa DC, Luizon MR, Lanna CM, Carneiro PC, Izidoro-Toledo

TC, Ferraz KC, Tanus-Santos JE, Matrix metalloproteinase-9 genetic variations affect

MMP-9 levels in obese children, Int J Obes (Lond) 36(2012)69-75.

[32]  Palei AC, Sandrim VC, Amaral LM, Machado JS, Cavalli RC, Lacchini R,

Duarte G, Tanus-Santos JE, Matrix metalloproteinase-9 polymorphisms affect plasma

MMP-9 levels and antihypertensive therapy responsiveness in hypertensive disorders

of pregnancy, Pharmacogenomics J 12(2012)489-98.

[33] Gomes VA, Vieira CS, Jacob-Ferreira AL, Belo VA, Soares GM, Fernandes JB,

Ferriani RA, Tanus-Santos JE, Imbalanced circulating matrix metalloproteinases in

polycystic ovary syndrome, Mol Cell Biochem 353(2011)251-7.

Page 17 of 28

Accep

ted

Man

uscr

ipt

Legends to the figures

Figure 1. MMP-9 immunoreactivity in JNA and normal middle turbinate (Original

magnification ×200). (A, B, and C), Representative positive stainings of MMP-9 in

JNA. (D), Negative staining of MMP-9 in normal middle turbinate.

Figure 2. Spearman correlation analysis suggested that high expression of MMP-9 in

JNA had negative correlation with patient’s age (r=-0.412, p＜0.001).

Figure 3. Kaplan–Meier analysis of time to recurrence (TTR) for MMP-9 (A),

microvessel density (MVD) (B), tumor stage (C), and age (D). The p values

were determined by the log-rank test.

Figure 4. Receiver operating characteristic (ROC) analysis showed that MMP-9 was

the best predictor for time to recurrence (TTR). The area under the curve for this

Page 18 of 28

Accep

ted

Man

uscr

ipt

factor was 0.704 (95% confidence interval [CI], 0.562–0.847; p=0.011) for

recurrence. AUC, area under the curve; SE, standard error.

Page 19 of 28

Accep

ted

Man

uscr

ipt

 

Clinicopathological characteristics of 70

patients with JNA Number of Patients with JNA Total number 70 Sex Male 70 Female 0 Median age, y 17

≤17 39

＞17 31

Tumor stage

Stage Ⅰa 4

Stage Ⅰb 7

Stage Ⅱa 6

Stage Ⅱb 7

Stage Ⅱc 34

Stage Ⅲa 5

Stage Ⅲb 7

Operation history No 40 Yes 30 Operation approach With endoscope 17 Without endoscope 53 Volume of intraoperative

hemorrhage（ml）

≤800 26

＞800 44

Abbreviations: JNA, juvenile nasopharyngeal angiofibroma

Page 20 of 28

Accep

ted

Man

uscr

ipt

 

Table 2  

Immunohistochemical staining intensity levels of MMP-9 in patients with JNA and normal middle turbinate

The intensity of

immunoreactivity − + ++ +++ JNA(n=70) 3 29 14 24 Normal middle turbinat(n=10) 8 2 0 0

 

Page 21 of 28

Accep

ted

Man

uscr

ipt

 

Pearson chi-square test analysis of association between clinicopathologic

variables and levels of MMP-9 expressions in patients with JNA Levels of MMP-9

expressions Variables No.of Patients − + ++ +++

p value

Age, y

＜18 39 0 10 10 19

≥18 31 3 19 4 5 0.001

JNA operation history No 40 2 17 7 14 Yes 30 1 12 7 10

0.931

Operation approach With endoscope 17 1 7 4 5 Without endoscope 53 2 22 10 19

0.934

Tumor stage

≤stage IIa 17 1 5 3 8

＞stage IIa 53 2 24 11 16 0.563

Volume of intraoperative

hemorrhage（ml）

≤800 26 2 11 5 8

＞800 44 1 18 9 16 0.732

 

Page 22 of 28

Accep

ted

Man

uscr

ipt

Table 4 Analysis of correlations between levels of MMP-9 expressions

and tumor recurrence in patients with JNA levels of MMP-9 expression Tumor recurrence

− + ++ +++ Yes 1 4 5 11 No 2 20 6 6

 

Page 23 of 28

Accep

ted

Man

uscr

ipt

Table 5

Pearson chi-square test analysis of association between clinicopathologic

variables and tumor recurrence in patients with JNA

RecurrenceVariables No.of Patients

Yes No Recurrence rate p value

Age, y        

＜18 33 16 17 48.48%

≥18 22 5 17 22.73% 0.054

JNA operation history          

No 30 9 21 30.00% Yes 25 12 13 48.00%

0.171

Operation approach          

With endoscope 13 4 9 30.77% Without endoscope 42 17 25 40.48%

0.529

Tumor stage          

≤stage IIa 12 1 11 8.33%

＞stage IIa 43 20 23 46.51% 0.016

Volume of intraoperative

hemorrhage（ml）          

≤800 22 5 17 22.73%

＞800 33 16 17 48.48% 0.054

 

Page 24 of 28

Accep

ted

Man

uscr

ipt

 

Univariate and multivariate Cox regression analyses of prognostic factors

associated with recurrence in patients with juvenile nasopharyngeal

angiofibroma. Univariate Multivariate

Variables HR 95 % CI P HR 95 % CI P

MMP-9(high vs. low) 4.499 1.624-12.463 0.004 4.447 1.311-15.086 0.017MVD (high vs. low) 2.989 1.235-7.237 0.015 3.532 1.420-8.787 0.007Tumor stage

(≤stage IIa vs.＞

stage IIa)

7.197 0.965-53.682 0.054 12.344 1.607-94.833 0.016

Age (＜18 vs. ≥18 y) 0.344 0.124-0.950 0.039 0.864 0.260-2.874 0.864

Hemorrhage during the

operation (≤800mL

vs. ＞800mL)

2.469 0.904-6.745 0.078

JNA operation history (yes vs. no)

1.888 0.794-4.489 0.150

Operation approach (with endoscope vs. without endoscope)

0.758 0.253-2.270 0.620

HR=hazard ratio; CI = confidence interval at 95% level; MVD, microvessel density; JNA, juvenile nasopharyngeal angiofibroma.  

Page 25 of 28

Accep

ted

Man

uscr

ipt

Figure 1

Page 26 of 28

Accep

ted

Man

uscr

ipt

Figure 2

Page 27 of 28

Accep

ted

Man

uscr

ipt

Figure 3

Page 28 of 28

Accep

ted

Man

uscr

ipt

Figure 4