Correlation between immunohistochemical markers affecting vascularity in vestibular schwannoma and arterial spin labelled imaging

INTRODUCTION

Vestibular schwannomas are benign tumors originating from the Schwann cells surrounding the eighth cranial nerve. The tumors eventually spread from the internal acoustic meatus into the cerebellopontine (CP) angle. The tumor’s proximity to the brainstem and cranial nerves is the main cause of its morbidity. Among all the intracranial tumors, vestibular schwannoma contributes 6–8%, among posterior cranial fossa tumors 25–33%, and among CP angle lesions 75–80%, followed by CP angle meningioma (5–15%) and CP angle epidermoid (4.6–6.3%).^[1,2] From a histopathological perspective, they are benign, slow-growing neoplasms with an annual growth rate between 0.2 mm and 2 mm.

Conventionally, vestibular schwannomas are believed to be clinically observed as hypovascular lesions. However, in many studies, a rare entity termed hypervascular, alternatively called hemorrhagic vestibular schwannoma, has been observed in <1% of cases among all the operated cases of vestibular schwannoma.^[3] Hypervascular vestibular schwannomas have been observed to show significant bleeding during surgical decompression, reducing the visibility in the surgical field during decompression and limiting the extent of resection. These hypervascular lesions have been more commonly seen in young individuals (<45 years). Furthermore, there is a reduced hearing preservation rate and an increased risk of post-operative facial weakness.^[4]

The vascular endothelial growth factor (VEGF) is associated with the growth of vestibular schwannoma; also, its increased expression has been observed with hypervascular schwannoma and is associated with recurrent cases.^[5] CD31 and factor VIII are endothelial markers that have been used to study micro-vessel density in malignant lesions.^[6]

Arterial spin labelled (ASL) imaging is a non-invasive imaging technique to measure perfusion in the region of interest (ROI) in the brain by magnetically labelling the inflowing arterial blood using water in the blood as an endogenous tracer.^[7]

In this study, we aim to assess the extent of perfusion in the vestibular schwannoma determined radiologically using ASL and correlate that with the extent of immunohistochemical (IHC) markers-VEGF and factor VIII expression in tumor biopsy specimens obtained at surgery.

MATERIALS AND METHODS

Analysis of imaging

On the MR workstation, ASL images were sublayered with T1 MPRAGE done in 0° angulation to view the lesion and site of the thalamus accurately. Using a 100 mm² ROI over the site of maximum perfusion within the lesion, i.e., vestibular schwannoma as seen on ASL post-processed color map, mean tumor blood flow (TBF) was calculated and a similar ROI was placed over the ipsilateral thalamus to obtain mean cerebral blood flow (CBF). Relative TBF (rTBF) is calculated using the ratio of mean TBF to mean CBF. Figure 1 shows the analysis of imaging.

Close

Figure 1:

Image depicting the method of arterial spin labelled (ASL) analysis. (a) Post-processed ASL color map. (b) T1 magnetization prepared rapid acquisition gradient echo (MPRAGE) axial sublayered over ASL image, and a 100 mm² region of interest (ROI) (Purple circle) drawn over the site of maximum perfusion as observed on ASL color map to obtain mean tumor blood flow. (c) T1 MPRAGE axial at the level of the thalamus with 100 mm² ROI (Green Circle) to obtain mean cerebral blood flow.

Export to PPT

IHC assessment

The formalin-fixed paraffin-embedded blocks of the cases selected for the study were retrieved from the histopathology archives. Tissue sections of 3–5 micron in thickness were obtained on positively charged slides. The sections were subjected to deparaffinization, antigen retrieval, and automated IHC staining in Ventana BenchMark immunostainer for the following antibodies: Anti-VEGF-A (affinity, clone polyclonal, 1:50 dilution), Factor VIII (affinity, clone polyclonal, 1:50 dilution), and anti-CD31 (Path in situ, clone-negative controls) were examined. The VEGF and Factor VIII interpretations were based on cytoplasmic and cytomembranous staining.

Scoring for VEGF and Factor VIII each is based on the percentage of tumor volume (that includes the blood vessels [BVs], stroma, and tumor cells). Scoring is given based on the single pathologist’s observation: Score 0 – completely negative, Score 1 – Faint/weak staining of the cytoplasm/cytomembrane in 10–25%, Score 2 – Moderate staining of the cytoplasm/cytomembrane in 10–25%, Score 3 – Strong staining of the cytoplasm/cytomembrane in 10–25%, Score 4 – Strong staining of the cytoplasm/cytomembrane in 25–50%, and Score 5 – Strong staining of the cytoplasm/cytomembrane in >50%.

Membranous staining of CD31 was considered positive staining. The scoring system is as follows: Number of high power fields (hpfs) with BV density occupying 50% of a single hpf: Score 0 – completely negative, Score 1 – 1–2 hpf, Score 2 – 3–5 hpf, Score 3 – 6–10 hpf, and Score 4 – > 10 hpf. Figure 2 shows the representative hematoxylin and eosin-stained sections of the biopsy specimen, and Figure 3 shows the IHC-stained sections.

Close

Figure 2:

(a) Microphotograph showing the neoplasm (schwannoma) with increased vascularity (asterix) and hyalinized blood vessels (arrow), hematoxylin and eosin (H&E) ×40. (b) Microphotograph showing the neoplasm (schwannoma) with increased vascularity (small delicate blood vessels shown by thin arrows), H&E ×40. (c) Microphotograph showing the neoplasm (schwannoma) with Verrocay body (arrow), H & E ×100. (d) Microphotograph showing the neoplasm (schwannoma) with delicate blood vessels (increased vascularity) and hemosiderin deposits (arrow). H & E ×40

Export to PPT

Close

Figure 3:

(a) Factor VIII Ag immunohistochemical (IHC) showing dense staining of endothelial cells (blue arrow), Score 5, ×100. (b) IHC vascular endothelial growth factor showing dense staining of endothelial (blue arrow) and stromal cells (black arrow), Score 5, ×100. (c) IHC CD31 staining showing dense collection of delicate blood vessels (white arrow), Score 4, ×100.

Export to PPT

RESULTS

Correlation between rTBF and IHC markers

Figure 4: Shows a scatter plot showing an association between MRI rTBF with VEGF, Factor VIII, and CD31. A Spearman rank correlation coefficient of 0.2 (P=0.2) was observed between IHC VEGF and rTBF, 0.3 (P=0.106) between IHC Factor VIII and rTBF, and 0.3 (P=0.146) between CD31 and rTBF [Table 3].

Close

Figure 4:

(a) Scatter plot showing correlation between relative tumor blood flow (rTBF) and vascular endothelial growth factor. (b) Scatter plot showing correlation between rTBF and Factor VIII. (c) Scatter plot showing correlation between rTBF and CD31.

Export to PPT

Table 3: Correlation between rTBF and IHC markers.

Correlation	Spearman correlation coefficient	P-value
IHC Score: VEGF with MRI: rTBF	0.2	0.200
IHC Score: Factor VIII versus MRI: rTBF	0.3	0.106
IHC Score: CD31 versus MRI: rTBF	0.3	0.146

IHC: Immunohistochemical, MRI: Magnetic resonance imaging, VEGF: Vascular endothelial growth factor, rTBF: Relative tumor blood flow. Significance of P<0.1

DISCUSSION

Hypervascular vestibular schwannomas represent a distinct entity characterized by pronounced vascularity, which poses significant challenges during surgical intervention. The primary objective of this study was to establish a reliable radiological surrogate capable of identifying hypervascular vestibular schwannomas. We attempted to correlate the increased perfusion observed on ASL imaging with specific angiogenic markers assessed through IHC analysis of tumor tissue.

The mean rTBF observed on ASL in the current study was 1.56, ranging from 0.73 to 2.81 among 31 patients. This is consistent with the off value of rTBF values of 1.55 obtained in a study by Tanaka et al., who assessed perfusion of 103 patients diagnosed as unilateral sporadic vestibular schwannoma on ASL and correlated the same with angiographic and clinical vascularity evaluated by studying the degree of intraoperative bleeding and observed that rTBF values (rTBF = TBF/CBF; above 1.55) showed statistically significant correlation with hypervascularity on angiography and increased bleeding clinically.^[7] At this cut-off of 1.55 for rTBF, the sensitivity and specificity of correlation with angiographic vascularity were 93.9% and 72.9%, respectively, and the sensitivity and specificity of correlation with clinical hypervascularity were 79.4% and 66.7%, respectively.^[7] Noguchi et al. evaluated perfusion in 35 cases of brain tumors using perfusion imaging in a semi-quantitative manner employing ASL imaging and studying the percentage of signal intensity and investigated its correlation with increased histopathological vascular density; they observed statistically significant correlation between two parameters in glioma (P < 0.00005) and hemangioblastoma (P < 0.05), but in schwannomas and meningiomas through positive correlation was observed, it was not statistically significant given small sample size.^[9] While other studies by Yamamoto et al., Kleijwegt et al., and Qiao et al. have tried using ASL, semi-quantitatively and qualitatively to assess perfusion in intracranial lesions such as skull base meningiomas, pediatric intracranial lesions, and glioma, respectively, there are not many studies evaluating the same quantitatively for vestibular schwannomas.^[10-12]

We have observed a statistically significant positive relationship between VEGF and Factor VIII expression and microvessel density assessed using CD31 staining (rho = 0.6, P < 0.001; rho = 0.5, P = 0.005 respectively) and between VEGF and Factor VIII expression (rho = 0.8, P < 0.001), indicating a causative role of these factors in promoting neovascularization in these acoustic neuromas. Similarly, Koutsimpelas et al.^[5] demonstrated a statistically significant positive correlation between increased VEGF mRNA and protein expression with elevated microvessel density assessed using CD31 analysis, accelerated growth rates, and larger tumor volumes in vestibular schwannomas. CayeThomasen et al.,^[13] in their retrospective study of 18 cases of vestibular schwannoma specimens using polyclonal VEGF antibody, observed a positive correlation between VEGF staining intensity and the absolute growth rate (R = 0.5785; P < 0.05) as well as the relative growth rate (R = 0.7018; P < 0.01) of the tumor as seen on serial pre-operative MRI. Plotkin et al.^[14] observed that VEGF expression and VEGF receptor-2 expression were seen in 100% cases and 32% cases, respectively, of vestibular schwannoma in NF-2 and sporadic cases; furthermore, the study concluded an indirect role of VEGF in tumor angiogenesis when tumor size reduced following treatment with bevacizumab. Walker et al. observed elevated synthesis and secretion of bioactive Factor VIII in bladder cancer cells and in other malignant lesions such as thyroid carcinoma, lung adenocarcinoma, colon carcinoma, hepatocellular carcinoma, and papillary renal cell carcinoma, highlighting the extracoagulative function of Factor VIII in cancer pathophysiology.^[15]

In our study, we observed a positive correlation between rTBF and IHC factors in vestibular schwannomas, although these correlations did not reach statistical significance (MRI rTBF and VEGF: rho = 0.24, P = 0.20; MRI rTBF and Factor VIII: rho = 0.106; MRI rTBF and CD31: rho = 0.27, P = 0.146). While inconclusive, these findings contribute to the existing literature by suggesting the potential for identifying a radiological, non-invasive surrogate to predict the presence of hypervascular schwannomas pre-operatively. There are certain limitations observed in the current study. First, although our study provides preliminary evidence of a potential correlation between ASL-derived rTBF and IHC markers in vestibular schwannoma, a statistically significant definite correlation can be obtained only after extrapolating findings over a larger sample size. Second, ASL imaging reveals a heterogenous perfusion pattern and a sample taken for IHC evaluation can be from a low perfusion area, which might not show a correlation with the exact area of the ASL hotspot.^[2]

This can be overcome using the navigation to sample the exact site of hyperperfusion. Szeremeta et al. reported maximum proliferation in the tumor capsule of vestibular schwannomas, with lower proliferative activity observed in specimens from the tumor parenchyma.^[16] The third limitation observed is that only radiological perfusion was observed and correlated with the extent of expression of IHC markers; intra-operative bleeding was not studied during this study.

CONCLUSION

A positive association between VEGF and Factor VIII, along with heightened microvessel density, indicates that the elevated expression of both growth factors may contribute to hypervascularity in vestibular schwannoma. The strong correlation between elevated IHC expression and ASL imaging presents a significant opportunity for future research to establish ASL as a pre-operative radiological marker for the detection and management of hypervascular vestibular schwannoma. We also investigated Factor VIII for the first time in vestibular schwannoma and its causal relationship and as a potential target for tumor control can be validated through further studies.