The relationship between molecular subtypes and magnetic resonance perfusion in patients with brain meningioma

Abstract

Background:

Meningiomas are the most common primary brain tumors, accounting for 15–20% of all primary brain tumors. Differentiating between malignant and benign meningiomas before surgery is essential for treatment planning and prognostic assessment. One proposed method for differentiating malignant from benign meningiomas is magnetic resonance (MR) perfusion (MR perfusion), but limited studies have been conducted in this respect. Therefore, this study aimed to determine the relationship between molecular subtypes and MR perfusion in patients with meningiomas.

Methods:

In this cross-sectional study, all patients with brain meningiomas referred to Sina Hospital in Tehran between 2021 and 2024 were evaluated prospectively. MR perfusion was done for patients. After brain surgery, pathological samples were obtained from the patients, and in addition to determining the conventional tumor grade, molecular studies including Ki-67, epidermal growth factor receptor amplification, and S100 were performed by the pathologist, and the results were recorded. The relationship between MR perfusion parameters (relative cerebral blood volume [rCBV], relative cerebral blood flow [rCBF], and mean transit time) with tumor-related data was assessed. The significance level was considered <0.05.

Results:

Twenty-five patients were studied. The mean age was 55.08 ± 12.07 years, and 17 patients (68%) were female. The most common presentation (76%) was headache. The findings showed that rCBV and rCBF were statistically significantly different between low-grade and high-grade tumors (P < 0.05). Although the calculated cut-offs for rCBV and rCBF (7.55 and 6.94, respectively) were useful for differentiating the different grades of disease, neither was statistically significant (P: 0.05 and P: 0.1, respectively). Furthermore, patients with higher Ki67 levels had higher rCBV and rCBF (P < 0.05 for both).

Conclusion:

MR perfusion values (rCBV and rCBF) have statistically significant differences between different grades of meningioma.

INTRODUCTION

Meningiomas represent the most prevalent type of intracranial tumors, accounting for 30% of all primary brain neoplasms. These tumors predominantly affect middle-aged individuals and are believed to arise from arachnoid cap cells.[1,4]

While the majority of meningiomas are classified as benign, it is important to note that up to 20% of these tumors are atypical or malignant. In such instances, these variants exhibit a more aggressive nature accompanied by a higher likelihood of recurrence. The degree of initial tumor resection, in conjunction with the histological grade, serves as a significant predictor of tumor recurrence. Consequently, an understanding of the histological grade before the surgery is essential for effectively planning surgical interventions and radiotherapy.[8,14] Meningioma recurrence risk has primarily been associated with the World Health Organization histopathological grade and the extent of surgical resection. However, the importance of molecular characteristics in evaluating meningioma aggressiveness and recurrence risk was described recently.[27] Investigation of the relationship between these subtypes and imaging findings before the surgical resection can be utilized for better management of patients.[25]

The routine diagnostic approach for meningioma involves conducting a brain magnetic resonance imaging (MRI), with or without the use of intravenous (IV) contrast. This method is typically effective in detecting meningiomas; however, it may not facilitate the differentiation between various subtypes of typical meningiomas, nor between typical and atypical or malignant meningiomas, utilizing conventional MRI sequences.[9,16,28] Magnetic resonance (MR) perfusion (MR perfusion) has the potential to aid in distinguishing a primary neoplastic process from a mass. Nevertheless, there is a paucity of research examining the correlation between the molecular subtypes of brain meningiomas and MR perfusion findings.[6,23,24] Consequently, this study’s objective was to explore the relationship between molecular subtypes of meningioma and MR perfusion parameters.

MATERIALS AND METHODS

It was a cross-sectional prospective study that was conducted on all patients with brain meningiomas referred to Sina Hospital in Tehran between 2021 and 2024.

All patients presenting with newly diagnosed brain meningioma who were referred to the Sina Hospital and were candidates for surgical intervention were included in this study. The exclusion criteria were age <18 years or more than 75 years; not previously received care at this center, incomplete medical information; not having consent to participate in the study; psychological disorders; concurrent malignant diseases; pregnancy; concurrent brain lesions other than meningioma; history of stroke; previously history of brain surgery; recurrent meningioma; multiple meningiomas; history of neurofibromatosis; meningiomas secondary to radiotherapy; prior adjuvant therapy for meningioma before surgical intervention; requiring emergency surgery for meningioma; Karnofsky performance status <70; and modified Rankin scale >2.

Brain MR perfusion was conducted at the hospital utilizing a General Electric Discovery 750 GEM 3-tesla device. To enhance coordination and analysis, a member of the radiology department served as a collaborative partner on the project.

The patients underwent examination by a neurosurgeon, during which demographic information was systematically recorded using a checklist. Subsequently, surgical procedures were conducted on the patients. Following the surgical intervention, conventional tumor grading was performed, alongside molecular assessments, including Ki-67, epidermal growth factor receptor (EGFR) amplification, and S100. These analyses were conducted by a pathologist at the hospital, and the findings were duly documented.

Statistical methods

The collected data were analyzed using the Statistical Package for the Social Sciences version 26 software. The statistical indicators of mean and standard deviation for continuous variables and frequency (number and percentage) for categorical variables were used to describe the data. The analysis of variance (pairwise Bonferroni) test was used to examine the differences between groups in continuous variables. The statistical significance level in this study was considered <0.05.

Ethical consideration

This study was approved by the ethical committee of Tehran University of Medical Sciences (IR.TUMS.SINAHOSPITAL. REC.1401.054).

RESULTS

Twenty-five patients were evaluated. The mean age of the participants was 55.08 ± 12.07 years, and 32% were male (n = 8) and 68% were female (n = 17). The data on presentation, location, and type of tumor are presented in Table 1.

Table 1:

Data of presentation, location, and type of meningioma in patients.

Ki67 was >10% in 4 patients (16%), 5–10% in 1 patient (4%), and <5% in 20 patients (80%). In patients with Grade 1, 2, and 3, 9 patients (75%) had Ki67 <5%, three patients had Ki67 >10%, and Ki67 was >10% in one case, respectively. EGFR was negative in all patients, and S100 was strongly positive in only one with Grade 2 (4%).

The mean value of MR perfusion parameters, including relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), and mean transit time (MTT) of all patients, were 9.74 ± 7.64, 19.08 ± 13.61, and 18.15 ± 9.44, respectively. The comparison of findings of MR perfusion parameters based on the grade of the tumor is presented in Table 2. Based on this table’s data, rCBV and rCBF values were significantly associated with the grade of meningioma, and both were significantly higher in grade 3 meningioma (P < 0.001 and P = 0.001, respectively). MTT had no association with the grade of tumors (P: 0.573).

Table 2:

Comparison of MRI perfusion parameters between different tumor grades.

We assessed the association between rCBV, rCBF, and MTT with different types of tumors. We observed that rCBV and rCBF had significantly higher values in patients with higher grade (atypical and anaplastic) meningiomas than patients with benign types of meningiomas (meningothelioma and transitional) (P < 0.001 and P: 0.005, respectively). MTT had no significant difference between the different types of meningiomas (P: 0.091). The data are presented in Table 3.

Table 3:

Comparison of cerebral perfusion parameters in different histopathological types of meningiomas.

To establish the diagnostic cut-off of MR perfusion parameters for differentiating meningioma grades, we conducted a receiver operating characteristic (ROC) analysis. We specifically focused on the indices that demonstrated statistically significant differences across grades (rCBV and rCBF). The findings are detailed in Tables 3 and 4. Due to the limited number of grade 3 patients in this study, with only three cases identified, we amalgamated patients with grades 2 and 3 for statistical analysis, categorizing them as grade 2/3.

Table 4:

Determining the cut-off of rCBV for tumor grade detection.

The area under the curve (AUC) for rCBV was 0.731, suggesting a moderate-to-good level of diagnostic accuracy for grading differentiation based on rCBV measurements. The standard error associated with this AUC value was 0.106. The P-value was at 0.050, on the statistical significance threshold (P ≤ 0.05). This may be attributed to the limited sample size used in the analysis. The optimal cut-off for detecting various grades of meningioma for rCBV was 7.55. At this threshold, the sensitivity was 83.3% and the specificity was 30.8% [Table 4].

The AUC for rCBF was 0.692, indicating a moderate to acceptable diagnostic ability, but the significance level (P-value) was 0.103. Based on Youden’s index, the best cut-off for differentiating the two groups was 6.94. At this threshold value, the specificity was 100%, while the sensitivity was 50%. The related results of rCBF are seen in Table 5. Furthermore, ROC diagrams for rCBF and rCBV are presented in Diagram 1.

Table 5:

Results of determining the rCBF cut-off for tumor grade differentiation.

Diagram 1:

Receiver operating characteristic (ROC) diagram for relative cerebral blood flow (rCBV) and relative cerebral blood volume.

Table 6 shows the rCBV and rCBF values based on the Ki67 level. There were statistically significant differences between the rCBV and rCBF values at different Ki67 levels (P < 0.05), but this significant difference was not found in MTT (P: 0.09).

Table 6:

MR perfusion parameters based on Ki67 levels.

DISCUSSION

This study evaluated 25 patients, with a mean age of 55.08 ± 12.07 years, of which 17 patients (68%) were female. The mean Ki67 was 0.06 ± 0.04, and headaches were the most common symptom, reported by 76% of participants. Grade 1 tumors were most frequently operated on, comprising 52% of the cases. rCBV and rCBF significantly increased with higher tumor grades, with anaplastic meningiomas exhibiting the highest values. The study identified cut-off values for differentiating grade 1 from grade 2/3 meningiomas, with rCBV at 7.55 (sensitivity: 83.3%, specificity: 30.8%, P: 0.05) and rCBF at 6.94 (sensitivity: 50%, specificity: 100%, P: 0.1). However, neither cut-off achieved statistical significance, likely due to the small sample size. A significant association was found between rCBV and rCBF with Ki67 levels, indicating that higher Ki67 levels were associated with higher rCBV and rCBF.

MRI is an important imaging modality for evaluating pathophysiological processes. Specifically, the characterization of brain tumors has significantly benefited from advancements in techniques such as diffusion-weighted imaging and perfusion imaging. Nevertheless, despite the numerous computational methods that have been proposed, a comprehensive analysis of meningiomas remains insufficient. Furthermore, it has been suggested that MR perfusion parameters may serve as promising biomarkers for tumor staging.[11,12]

In a study examining the role of MR perfusion utilizing arterial spin labeling (ASL) and dynamic susceptibility contrast in the grading of skull base meningiomas, Eissa et al. concluded that MR perfusion is effective in distinguishing between low-grade and high-grade meningiomas.[5] The findings of the present study corroborate the effectiveness of MR perfusion in diagnosing meningiomas. Our investigation revealed a significant association between meningioma grading and MR perfusion parameters, with the rCBV and rCBF exhibiting higher values in higher tumor grades compared to lower grades. Furthermore, the current study demonstrated that different grades of meningiomas could be differentiated based on established cutoff values. However, it is important to note that the cutoffs for rCBV and rCBF did not achieve statistical significance, which may be attributed to the limited sample size. This concern necessitates further studies with larger populations to validate the proposed cutoffs.

Qiao et al.[21] investigated the effectiveness of MR perfusion with ASL in differentiating benign from malignant meningiomas. They identified three perfusion patterns: Pattern 1 showed homogeneous hyperperfusion, associated with grade I meningiomas; Pattern 2 exhibited heterogeneous hyperperfusion; and Pattern 3 revealed no significant hyperperfusion, correlating with grade II and III meningiomas. The study concluded that qualitative assessment of cerebral blood flow (CBF) maps could aid in distinguishing benign from higher-grade tumors, potentially influencing treatment strategies.[20] The current study found that rCBV and rCBF were positively and significantly associated with increasing tumor grade. Although the MTT was different between the groups, it was not statistically significant, possibly due to the small sample size.

The study conducted by Rohilla et al.[21] assessed the grading of meningiomas by analyzing rCBV. The findings indicated that the maximum rCBV within the tumor did not significantly aid in distinguishing benign meningiomas from malignant ones. Notably, an rCBV threshold of 2.5 mL/100 g in peritumoral edema demonstrated a sensitivity of 75%, specificity of 84.6%, and an accuracy of 83.3% for differentiating between benign and malignant meningiomas. Consequently, it was concluded that benign and malignant meningiomas can be effectively differentiated based on the maximum rCBV in peritumoral edema. Nevertheless, intratumoral rCBV values may provide valuable insights into the subtyping of meningiomas, particularly in the assessment of transitional and meningothelial types.[21] In the current study, rCBV values were found to exhibit a statistically significant association with the grading of meningiomas, with elevated rCBV values associated with higher tumor grades. Both studies corroborated the notion that rCBV may serve as a potential indicator of meningioma grading, although employing differing methodologies. Despite these methodological differences, existing research appears to substantiate the diagnostic utility of MR perfusion in assessing meningioma status. In our study, the diagnostic cutoff established for differentiating meningioma grades based on rCBV was identified as 7.55, yielding a sensitivity of 83.3% and a specificity of 30.8%. The discrepancies observed between our findings and those reported by Rohilla et al.[21] may be attributable to differences in the target sites for rCBV measurement; our study focused on the tumor itself, whereas Rohilla et al.[21] concentrated on the peritumoral edema. Previous research has indicated that the evaluation site, the type of imaging device, and the software utilized can significantly influence the values of MR perfusion parameters, which may explain the difference in results observed across studies.[10,13]

Lum et al.[15] study aimed to evaluate the efficacy of intra-arterial (IA) MR perfusion in characterizing the vascular supply of meningiomas. The findings indicated that full width at half maximum, rCBV, and rCBF showed statistically significant differences between regions of interest for IA MR perfusion. Moreover, rCBV and MTT were significantly lower for IA perfusion in the dural external carotid artery compared to IV perfusion, while rCBF was notably higher in the internal carotid artery region for IA MR.[15] The current study demonstrates that MR perfusion can effectively differentiate different grades of meningioma, highlighting its diagnostic utility for preoperative assessments. Although MTT did not show significant differences among grades, the parameters of rCBV and rCBF proved valuable due to their sensitivity and specificity. This comparison suggests that MR perfusion is a promising imaging technique for grade differentiation in meningiomas, but further studies are necessary to confirm its diagnostic value.

Regarding the utility of MR perfusion parameters, some other studies were performed, but they did not evaluate our parameters. Saito et al.,[22] evaluated the relationship between cerebral blood volume (CBV) and CBF with pathologic features of meningioma and they found that patients with larger peritumoral edema had higher values of CBF and CBV. Furthermore, skull base meningioma showed higher levels of these values.[22] Naghath et al., concluded that MR perfusion had a significant relationship with the primary type of tumors that are diagnosed as metastatic meningioma.[17] Kang et al.,[11] determined the predictive power of MR perfusion for the prediction of intraoperative blood loss of meningioma. They mentioned that rCBV could predict intraoperative blood loss in meningiomas who underwent surgical management.[11] In this respect, our study was similar to the previous study regarding the importance of MR perfusion relationships with meningioma’s pathologic features.

In our study, the expression of Ki67 was observed to increase with escalating tumor grades from 1 to 2 and 3, particularly noted in grades 2 and 3, where a higher proportion of patients exhibited elevated Ki67 levels. However, it is important to note that the small sample size precluded the establishment of statistical significance. Numerous studies have indicated that Ki67 serves as an independent prognostic indicator for meningioma. Nevertheless, the prognostic implications of Ki67 in meningioma remain inconclusive. Some research has identified a negative correlation between Ki67 expression levels and meningioma prognosis, while other investigations have reported findings that lack significance. These discrepancies may be attributed to the substantial differences in cut-off values used for Ki67 analysis, the definition of these cut-off values, and the tumor grades considered across studies.[2,3,12,18,19,26]

Our findings indicate a statistically significant association between rCBF and Ki-67. This observation aligns with the study conducted by Ginat et al., which reported a positive and statistically significant relationship between maximum rCBV and Ki-67.[7] Notably, our investigation reveals that no existing studies have assessed the association between Ki-67 and rCBF, underscoring the necessity for further research in this area and highlighting the novelty of this evaluation in the current study.

A small sample size was one of the important limitations of the study, and this limitation occurred due to the low prevalence of meningioma. On the other hand, our center is a referral center, and most referred patients are complicated. Most of our tumor were falx and parasagittal meningioma, and the high percentage of grades 1 and 2 could be due to this issue. Other molecular and genetic tests could be associated with MR perfusion parameters. However, we could not investigate other molecular and genetic factors because they were inaccessible and costly.

CONCLUSION

The values of MR perfusion parameters, including rCBV and rCBF, demonstrate statistically significant differences between different grades of meningioma. These parameters may be potential as preoperative diagnostic tools that may enhance the management of patients presenting with these tumors. According to our findings, the established differential cutoffs for rCBV and rCBF were 7.55 and 6.94, respectively. While these cutoffs exhibited commendable sensitivity and specificity for the diagnosis of meningioma grades, they were not statistically significant, possibly attributable to the small sample size utilized in this study.

Ethical approval:

The research/study was approved by the Institutional Review Board at the ethical committee of Tehran University of Medical Science, number IR.TUMS.SINAHOSPITAL. REC.1401.054, dated 2023.

Declaration of patient consent:

The authors certify that they have obtained all appropriate patient consent.

Financial support and sponsorship:

Nil.

Conflicts of interest:

There are no conflicts of interest.

Use of artificial intelligence (AI)-assisted technology for manuscript preparation:

The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.

Disclaimer

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Journal or its management. The information contained in this article should not be considered to be medical advice; patients should consult their own physicians for advice as to their specific medical needs.