Role of CD34-immunopositive Microvascular Density in Juvenile Nasopharyngeal Angiofibroma

Abstract

Objective

To assess the microvascular density (MVD) in juvenile nasopharyngeal angiofibroma (JNA) with CD34 immunostaining and evaluate its relationship with clinico-demographic features.

Methods

This prospective study included patients with JNA undergoing endoscopic excision. The histopathological specimen was stained using CD-34 antibodies to calculate MVD. MVD and clinico-demographic features were correlated.

Results

The study included 12 patients with a median age of 15.5 years. The mean MVD was 39 vessels/high power field (range 5 to 151 vessels). MVD was significantly associated only with the volume of tumour (r = 0.65, p = 0.02). The recurrence occurred in one patient with an MVD of 107. The median follow-up was 38 months.

Conclusion

MVD is significantly associated with tumour volume in JNA, which implies a robust role of angiogenesis in the pathology of the tumour. Also, higher MVD may be a risk factor for recurrence.

Key message

•Juvenile nasopharyngeal angiofibroma (JNA) is currently thought to originate from remnant of first branchial arch.

•Microvascular density (MVD) is a reliable indicator of angiogenesis which can be assessed using CD34 antibodies.

•JNA has high microvascular density for a benign tumour with a mean MVD of 39 micro-vessels/high power field.

•MVD is positively correlated with tumour volume which indicates a significant role of angiogenesis in tumour growth.

•High MVD may be a risk factor for tumour recurrence.

Introduction

Juvenile nasopharyngeal angiofibroma (JNA) is a highly vascular and benign but locally aggressive tumour originating in the nasopharyngeal area. Surgery is the primary treatment for this vascular tumour with the associated difficulty of high blood loss in a young patient. Various modalities are in current practice to decrease intraoperative blood loss in an attempt to overcome the morbidity of the procedure, such as preoperative embolization, hypotensive anesthesia, and the assistance of plasma ablation [1].

Even though the etiology and pathogenesis of JNA are yet to be known, current evidence considers it to be a remnant of the first branchial arch artery. The tumour microenvironment is interesting as it influences the stages of tumour development and progression. The interaction between tumour and stromal components may dictate the growth and invasive potential of tumour. Angiogenesis is necessary for the growth of solid tumours. Microvascular density (MVD) is a known reliable indicator of neo-angiogenesis [2]. Various immunohistochemical markers are available to stain new blood vessels to assess microvascular density. These include CD34, CD31, and CD105. CD34 is a monomeric transmembrane glycoprotein, and antibodies to CD34 antigen are proven to stain endothelial cells in JNA while sparing pericytes, smooth muscle cells, and stromal fibroblasts [3]. In a retrospective study using CD105 as an immunohistochemical marker for endothelial cells, Wang et al. have shown that JNA has a higher MVD than normal middle turbinates and that higher MVD is a predictor of disease recurrence in patients with JNA after curative resection [2]. We aimed to assess the microvascular density in JNA by highlighting neo-vasculature with CD34 immunostaining and evaluating its relationship with clinico-demographic features to gain insight into its role in the etiopathogenesis of the tumour and explore therapeutic possibilities.

Materials and Methods

A prospective cohort study in the Department of Otorhinolaryngology and Head-Neck Surgery of our Institute was conducted from September 2018 to March 2020. The study received ethical clearance from the Institute Ethics Committee for Postgraduate Research. Twelve patients, clinico-radiologically diagnosed as cases of JNA, who underwent surgical treatment, were included in the study. The diagnosis was confirmed by histopathological examination of the operative specimen. Patients who had received prior surgical treatment, preoperative flutamide therapy, or any other non-surgical therapy such as radiotherapy were excluded from the study. Informed written consent was obtained from the patients or their guardians, as appropriate.

Data were collected from all the patients relating to demographic details, clinical details, preoperative radiology, intraoperative details, and follow-up radiology. Patients were radiologically staged using the system proposed by Radkowski et al. [4]. The volume of the tumour was determined on preoperative Contrast-enhanced Magnetic Resonance Imaging (CEMRI) using the manual trace method.

Histopathological Evaluation and Assessment of Microvascular Density in the Specimen

The specimens were routinely processed to prepare formalin-fixed paraffin-embedded tissue blocks. Sections cut from the blocks were stained with hematoxylin and eosin (HE) and examined by a pathologist (AK) for confirmation of the diagnosis. For assessment of MVD, one of the tumour tissue blocks was selected for immunohistochemistry with a primary antibody against CD34 (Diagnostics Biosystems, 1:100 dilution). Antigen retrieval was performed in citrate buffer at pH 6. UltraVision™ Quanto Detection System (Thermo Fisher Scientific, Fremont, CA, USA) was used as a signal detection system. The vascular hotspot technique was used to calculate micro-vessel density [2, 5]. CD34-stained slides were examined under low magnification (40x) to identify “hotspots” containing micro-vessels. Four such non-overlapping hotspots were then viewed under 200x magnification, and images were captured using a digital camera attached to the microscope (Fig. 1). The number of CD-34-stained micro-vessels was counted manually in each hotspot to estimate the micro-vessel density. The mean microvessel density was calculated for each case from the four values obtained.

Fig. 1.

(a) Photomicrograph of tumour tissue stained with CD34 at 40x magnification to identify “hotspots”

(b) Photomicrograph showing tumour tissue containing normal sized as well as micro-vessels highlighted by CD34 staining

Statistical Analysis

Descriptive statistics were used to describe the clinico-demographic data in the form of mean, median, and range. Micro-vessel density was correlated with the age of the patient, duration of symptoms, intraoperative blood loss, duration of surgery, volume of the tumour, embolization, and stage of the tumour using the Pearson & Spearman correlation coefficient as appropriate. Statistical analysis was done using Stata/MP 16 (StataCorp LLC, Texas, USA). A p-value of less than 0.05 was considered to be statistically significant.

Results

Twelve patients were included in the study. All patients underwent endoscopic excision of the tumour. The clinico-demographical details are summarized in Table 1.

Table 1.

Clinico-demographical details of the study group

Age (in years)
Median Range	15.5 11 to 30
Stage of tumour#
Stage I B Stage II A Stage II B Stage II C	2 2 2 6
Preoperative embolization
Yes No	10 2
Duration of symptoms (in months)
Mean Range	15 6 to 36
Volume of tumour (in cm3)
Mean Range	94.4 18.1 to 344.3
Intraoperative Blood Loss (in mL)
Mean Range	507 70 to 1248
Duration of Surgery (in minutes)
Mean Range	108 57 to 175
Microvascular density (mean number of microvessels/ high power field)
Mean Range	39 vessels 5 to 151 vessels
Residual tumour on postoperative CEMRI*
Present Absent	None 12

^#Stage of tumour as per Radkowski et al.⁴; *CEMRI: Contrast-enhanced Magnetic Resonance Imaging at 3 months

The correlation of microvascular density with clinico-demographic features is shown in Table 2. Microvascular density was found to have a significant correlation with an increasing volume of the tumour (r = 0.65, p = 0.02). A linear regression analysis shows that the volume of the tumour could statistically significantly predict mean microvascular density (p = 0.02) and the volume of the tumour accounts for 37.2% of the explained variability in mean microvascular density (Fig. 2). One patient, with a high MVD of 107 micro-vessels/high power field, developed a recurrent tumour in the pterygopalatine fossa, which was diagnosed on follow-up CEMRI at 30 months. The patient underwent a revision endoscopic excision, and the tumour was confirmed on histopathological examination. The median follow-up duration was 38 months (range 26 to 45 months).

Table 2.

Correlation of microvascular density with clinico-demographical features

Parameter	Correlation coefficient	p-value
Age of patient	-0.21	0.5
Duration of symptoms	-0.35	0.25
Stage of tumour	0.21	0.5
Preoperative embolization	0.38	0.21
Volume of tumour	0.65	0.02
Intraoperative blood loss	-0.16	0.6
Duration of surgery	-0.26	0.39

Fig. 2.

Scatter diagram demonstrating the correlation between volume of the tumour and microvascular density

Discussion

JNA is a challenging tumour for rhinologists and skull base surgeons owing to its highly vascular nature, high recurrence rates, and occurrence in the young adolescent population. The tumour is diagnosed primarily on a clinico-radiological basis with a typical history of recurrent nasal bleeding with nasal obstruction in a young adolescent male and the presence of a globular smooth mass in the nasal cavity and nasopharynx on nasal endoscopy. The radiology shows a tumour localized around the sphenopalatine foramen, with predictable pathways of spread. Though a benign tumour, with surgery being the prime modality of treatment, its propensity to grow along the narrow crevices of the skull base and the local aggressiveness make complete excision difficult [1]. The non-surgical modalities, which include radiotherapy and chemotherapy, at best play an adjuvant role in the management of this highly vascular benign tumour [6].

Various theories have been proposed for the etiopathogenesis of JNA, which in itself explains the enigma of its origin. The currently accepted theory considers the tumour to originate from incomplete regression of the first branchial arch artery [7]. Beham et al. have characterized JNAs as vascular malformations, which have been supported by recent molecular genetic findings [8, 9]. The progressive growth of the tumour has been attributed to the presence of various growth factors such as VEGF, b-FGF, TGF-β, and PCNA [10]. These factors, likely originating from the stromal cells of the tumour, are also known to promote angiogenesis [11]. This tumour angiogenesis can be measured by assessing microvascular density (MVD) by performing immunohistochemistry with stains that highlight endothelial cells. MVD has been used to predict tumour aggressiveness in various solid tumours, including breast cancer, cutaneous melanoma, and prostate cancer [12]. Even though JNA is a benign tumour, it has significant local aggressiveness and recurrence rates to warrant a search for new therapeutic targets.

Beham et al. studied the histopathological samples of 24 patients with JNA for the immuno-expression of CD34 [3]. The authors found that there was exclusive staining of endothelial cells, establishing CD34 as a reliable immunohistochemical marker. Zhang et al. have calculated microvascular density in 28 patients with JNA, which had an average MVD of 49 vessels per HPF [13]. The authors have also calculated MVD in 20 patients with orbital cavernous hemangioma, which was 29 vessels per HPF. Wang et al. have used CD105 to assess microvascular density from a tissue microarray of 70 patients with JNA [2]. The average MVD was 10 vessels per high-power field (HPF). We found that the mean MVD in our case was 39 vessels per HPF, which is high for a benign tumour, in agreement with the previous studies. The increased MVD implies significant neo-angiogenesis in JNAs, which might be playing an important role in the etiopathogenesis.

Though JNA had a higher MVD, the wide range of MVD needed searching for determining factors. The association between MVD and the clinico-pathological characteristics of the patient has been done previously by Wang et al. in a retrospective study [2]. The authors have found that higher MVD correlated with recurrence in JNA (p = 0.013) and was also a significant predictor of time to recurrence (p = 0.009). In our prospective study, we had one patient with a recurrent tumour who had an MVD of 107 vessels per HPF, suggesting that high MVD may be a risk factor for recurrence in JNA. Wang et al. further found that MVD was not associated with age, history of JNA surgery, the approach of surgery, stage of the tumour, or intraoperative blood loss. Our findings were also similar to those of the previous study. In addition, we evaluated the correlation of MVD with tumour volume. A significant positive correlation (r = 0.65, p = 0.02) is present between them. This additional finding of an association between MVD and tumour volume indicates the role of angiogenesis in promoting tumour growth. Also, the knowledge that higher-volume tumours have high MVD may be used for therapeutic advantage. Brieger et al. have shown that in patients with JNA, increased MVD is associated with elevated VEGF expression [14]. This may help justify the use of anti-angiogenic drugs in tumours with high volume or the selection of patients for percutaneous embolization. To our knowledge, this is the first prospective study to correlate MVD and tumour volume in JNA. This finding may have a therapeutically promising role in the management of the tumour.

The number of patients included in our study is small, owing to the rare tumour. We found that the duration of symptoms had a moderate negative correlation with MVD (r = -0.35), which may suggest the aggressive nature of tumours with higher MVD. The finding may also point that early intervention might be beneficial to prevent a recurrence. However, the correlation was not statistically significant (p = 0.25), which may be because of the small numbers. Also, apart from angiogenesis, other factors determining the aggressiveness and growth of tumours in JNA have not been investigated. The findings of our study need to be substantiated by a larger sample size.

Conclusion

Juvenile nasopharyngeal angiofibroma is a benign tumour with high microvascular density, suggesting increased neo-angiogenesis. There is a significant positive correlation between MVD and tumour volume, which implies a robust role for angiogenesis in the pathology of the tumour. Also, higher MVD may be a risk factor for recurrence. The findings of our study may be further used to explore the role of antiangiogenic therapy in JNA.

Author Contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Avinash Shekhar Jaiswal and Rakesh Kumar. The first draft of the manuscript was written by Avinash Shekhar Jaiswal and all authors commented on subsequent versions of the manuscript. All authors read and approved the final manuscript.

Declarations

Financial support

This research has not received specific grant from any funding agency, commercial or not-for-profit sectors.

Ethical Standards

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional guidelines on human experimentation (Institute Ethics Committee, All India Institute of Medical Sciences, Ansari Nagar, New Delhi) and with the Helsinki Declaration of 1975, as revised in 2008.

Competing Interests

The authors have no competing interests to declare that are relevant to the content of this article.