Clear cell meningioma (CCM) is a rare subtype of meningioma. It represents only 0.2–0.8 % of all meningiomas [1, 2]. Histologically, it is characterized by sheets of polygonal cells with clear cytoplasm, which is the expression of a high glycogen concentration [3]. CCM was firstly described by Manivel and Sung in 1990 [4]. It was classified as a grade I lesion in 1993 because of its bland histological appearance [2]. However, CCM was found to behave aggressively with a high rate of local recurrence and cerebrospinal fluid metastasis, and it was changed to grade II in 2000 [5]. To date, more than 100 CCMs have been reported in the English literature [6]. Most of them were isolated case reports except for several series [2, 5–9]. The clinical features of CCM are still not clear. Here, we present the largest series of 36 CCMs and evaluate several prognostic factors of patient’s clinical outcome.

Patients and methods

Search methods and case definition

Review of the records of Beijing Tian Tan Hospital for the period from 2008 to 2015 identified 36 patients with pathologically confirmed CCM among a total of 10,529 meningioma patients. Data regarding the clinical presentation, radiological imaging, pathologic results, and treatment outcome of CCM patients were retrospectively reviewed. The microscopic pathologies of 36 CCMs were reviewed by two independent neuropathologists (L.L. and J.L.) according to the 2007 WHO grading system. The age and gender information of 10,529 meningioma patients were also collected for the purpose of comparison. We searched PubMed for relevant articles with the term “clear cell meningioma” in the title. Studies with more than seven cases were carefully reviewed.

Ethical committee approval and follow-up

This retrospective study was approved by the Beijing Tian Tan Hospital institutional review board. CCM patients were followed-up through the telephone interview or an out-patient department. Generally, patients were told to get cranial enhanced magnetic resonance imaging (MRI) examination at 3, 6, and 12 months after the operation; then they received cranial radiological examination every 1 or 2 years thereafter. Patients might also get additional radiological examination if they got suspicious symptoms that might be associated with tumor recurrence. Progression of CCM was defined as the radiological regrowth of the tumor. Progression-free survival (PFS) was defined as the time between initial surgery and tumor progression on radiology.

Statistical analysis

To select prognostic factors associated with longer PFS, the Kaplan-Meier survival analysis and log-rank test were used. The chi-squared test was used for the comparison of gender distribution in CCM and meningioma patients. Student’s independent t-test was used for the comparison of mean age at diagnosis in CCM and meningioma patients. A p value <0.05 was considered statistically significant. The analyses were performed using the SPSS statistical software, version 19 (SPSS, Chicago, IL, USA).

Results

Demographics

A total of 10,529 meningioma patents were treated in our institution from 2008 to 2015, CCM constituted 0.3 % of all meningiomas. Of the 36 CCM patients, 30 cases were intracranial and 6 cases were intraspinal. Of the 10,529 meningioma patients, 10,102 cases were intracranial and 427 cases were intraspinal. Thus, CCM constituted 0.3 % of the intracranial meningiomas and 1.4 % of the intraspinal meningiomas (chi-squared test, p = 0.003). The clinical characteristics of 36 CCM patients are summarized in Tables 1 and 2. The age at diagnosis of CCM patients ranged from 4 to 77 years; the mean age was 29.3 ± 18.4 years, which was significantly younger than that of total meningioma patients (49.8 ± 11.9 years) (t-test, p = 0, Fig. 1a–c). Of the 36 CCM patients, 13 were male and 23 were female; the male-to-female ratio (36 vs 64 %) was similar to that of total meningiomas (28 vs 72 %) patients (chi-squared test, p = 0.3, Fig. 1d).

Table 1 Clinical features of 36 patients with clear cell meningioma
Full size table
Table 2 Clinical, demographic, and pathological characteristics of 36 patients with clear cell meningioma
Full size table
Fig. 1
figure 1

Graph depicting gender and age distribution for CCM and total meningioma patients. a Distribution of age at diagnosis in total male (dotted line) and female (solid line) meningioma patients. The trendlines are normal curves. b Distribution of age at diagnosis in male and female CCM patients. c The mean age at diagnosis of CCM patients is significantly younger than that of total meningioma patients (t-test, p = 0). d The ratio of male to female CCM patients is similar to that in total meningioma patients (chi-squared test, p = 0.3)

Full size image

Symptoms and location

Of 36 CCM patients, most patients were sporadic and two patients (patients 10 and 19) had neurofibromatosis 2 (NF2). Preoperative symptoms of CCM patients were usually caused by tumor compression. Preoperative duration ranged from 1 month to 3 years, with a median duration of 3 months. Most spinal CCMs were located in the lumbosacral area. Of the 30 intracranial CCMs, the most common location was supratentorial (14 cases), followed by CPA or petrous apex region (11 cases), and infratentorial (5 cases).

Radiological findings

The MRI characteristics of CCM were similar to those of other types of meningiomas. Most CCMs appeared as isointense or slightly hyperintense on T1-weighted images and isointense or slightly hyperintense on T2-weighted images. The tumors were usually heterogeneously enhanced after the contrast administration. Two tumors (patients 17 and 22) extended from the cerebellopontine angle (CPA) to the region of Meckel’s cave and were misdiagnosed as trigeminal schwannoma preoperatively. One tumor showed hypointense on T1-weighted image and hyperintense on T2-weighted image without evident enhancement, mimicking cholesteatoma (patient 5, Fig. 2).

Fig. 2
figure 2

Patient 5. CCM at right CPA mimicking cholesteatoma. a T2-weighted image. b T1-weighted image. c Enhanced T1-weighted image

Full size image

Treatment

Twelve patients (33 %) underwent subtotal resection (Simpson grades III and IV) and 24 patients (67 %) underwent total resection (Simpson grades I and II). During the operation, CCM tumors were usually found to be firm and hypervascular like other types of meningiomas. Seven tumors (19 %) were found to erode the skull or sacrum bone. Most tumors were attached to the dura mater. One intracranial (patient 9, Fig. 3) and one spinal (patient 34) CCM were absent of dural attachment. Most spinal CCMs were intradural extramedullary; one spinal CCM was extradural and extended outward through the intervertebral foramen (patient 33, Fig. 4). Nine patients received postoperative radiotherapy, including gamma knife surgery (GKS) in four cases.

Fig. 3
figure 3

Patient 9. One nondura-based CCM at left frontal parietal lobe. The tumor shows hypointense on T1-weighted image (a) and hyperintense on T2-weighted image (b). The tumor is heterogeneously enhanced after the contrast administration (c–e). Note that there is thin brain parenchyma between the tumor and the dura mater. Pathological examination (f) reveals the diagnosis of clear cell meningioma (HE staining, ×100)

Full size image
Fig. 4
figure 4

Patient 33. One extradural spinal CCM extends outward through the intervertebral foramen. The tumor shows hypointense on T2-weighted image (a) and isointense on T1-weighted image (b). The tumor is heterogeneously enhanced after the contrast administration (c)

Full size image

Histological findings

Most CCMs showed benign histological features, and only one had significant mitotic figures (patient 15). The histological diagnosis was aided by immunostaining. Most tumors were positive for vimentin and EMA, and negative for S-100, GFAP and CK. Progesterone receptor (PR) staining was available in 16 cases: eight were positive and eight were negative. The Ki-67 indexes were available in 19 cases, and they varied from negative to 30 %. Nine cases were below 5 %, and ten cases were above 5 %.

Recurrence and survival analysis

Follow-up time of CCM patients ranged from 2 months to 15 years, with a median time of 3 years. During the follow-up, 15 patients (42 %) suffered from tumor recurrence. Most of them were local recurrence, multiple recurrences were found in only one patient (patient 11). Imaging study of the entire neuroaxis was not performed to evaluate the possible spinal metastasis. But patients did not have relevant symptoms that might be associated with tumor occurrence in other parts of the body. The recurrence time after the first operation ranged from 10 months to 12 years, with a median time of 29 months. Kaplan-Meier survival analysis showed that patients after total resection had significantly longer PFS time than those after subtotal resection (log-rank test, p = 0.006, Fig. 5). Of 12 patients receiving subtotal resection, 9 (75 %) patients got tumor recurrence; of 24 patients receiving total resection, 6 (25 %) patients got tumor recurrence. However, age (≤20 years or >20 years, p = 0.9), gender (p = 0.3), postoperative radiotherapy (p = 0.4), PR staining (positivity or negativity, p = 0.2), and Ki-67 index (≤5 % or >5 %, p = 0.4) did not have a significant effect on patient’s PFS time. After recurrence, patients usually received repeated operation and/or radiotherapy. Most patients stayed alive to the last date of follow-up, and two patients died of the disease.

Fig. 5
figure 5

Kaplan-Meier plots of PFS for CCM patients after total (Simpson grades I and II) or subtotal (Simpson grades III and IV) resection (log-rank test, p = 0.006)

Full size image

Discussion

CCM is a rare variant of grade II meningioma with distinctive histological features [6]. Previous studies indicated that it constituted 0.2–0.8 % of all meningiomas, which was similar to the result of our study (0.3 %). Recent studies found that CCM occurred equally in the intracranial and spinal region [6, 10]. However, the total spinal meningiomas are much less common than their intracranial counterparts. Our study showed that the proportion of CCMs among spinal meningiomas was almost 5 times that of the proportion of CCMs among intracranial meningiomas (1.4 vs 0.3 %), which could explain this interesting phenomenon. It should be noted that our institution performed far more cranial operations than spinal operations. Thus, the data in our study could not reflect the real ratio of total intracranial meningiomas to the spinal ones.

Like the other types of meningioma, CCM has a female predilection. The ratio of females to males having CCM was not significantly different from that for total meningiomas (1.8 vs 2.5, p = 0.3, Fig. 1d). It has been considered to be attributed to the sex hormones. Previous reports indicated that the female bias did not exist in CCM childhood cases [7]. CCM has also been reported to progress rapidly during pregnancy [11, 12]. All these indicate an important role of female sex hormones in the occurrence of CCM. Different from other types of meningiomas, CCMs tend to have a predilection toward younger patients [2, 6, 13]. Childhood cases of CCM are common [3, 7, 14]. Our study also found that the mean age at diagnosis of CCM patients was significantly younger than that of the total meningioma patients (29.3 ± 18.4 years vs 49.8 ± 11.9 years, p = 0, Fig. 1c).

CCMs usually have some preferred locations. The CPA is the most common intracranial area affected [1], which is seldom seen in other meningiomas [6]. In our study, CCM at the CPA or petrous apex region constituted 31 % of all CCMs. Most spinal CCMs were located in the lumbar region [15, 16], like the six patients in our study. Most CCMs are based in the dura, a few nondura-based CCMs have also been reported [17]. Two tumors in our study were also found to be nondura-based (patients 9 and 34). Clinical symptoms of CCM are mainly due to an increased intracranial pressure or a direct tumor compression. Inflammatory syndrome has also been rarely reported in patients with intracranial CCM [18, 19]. CCMs usually have similar radiological features to other types of meningiomas [20]. Wang et al. [6] reported CCM tended to have marked heterogeneous enhancement, apparent dural tail sign, prominent peritumoral edema, cystic components, and bone involvement. Despite its bland histological appearance, CCM is known for its aggressive behaviors with local recurrence and cerebrospinal fluid metastasis [7]. In six large series of CCM, the recurrence rates were reported to be 61, 50, 62.5, 22.2, 40, and 60 %, respectively [2, 5–8, 21]. This is similar to the result of our study (42 %). In contrast, the 5-year PFS rates for benign (WHO grade 1) and atypical (WHO grade 2) meningiomas were reported to be 97.5 and 69.7 % in one large series study [22]. Many patients in our study were followed-up for a relatively short time; the recurrence rate might be larger when longer follow-ups are performed. The reported average time of recurrence varied from 5.3 months to 2.3 years [5, 20]. The median recurrence time in our study was 29 months. Although multiple recurrences of CCM were common [2], only one patient in our study got multiple recurrences. Metastasis through the cerebrospinal fluid circulation has also been reported occasionally [2, 10, 13]. The imaging study of the entire neuroaxis was not performed in our study, so we could not evaluate the spinal metastasis rate. But our patients did not show any symptoms that might be associated with spinal tumor occurrence during the follow-up.

Currently there are no definite predictors of clinical outcome for CCM. Several studies showed that CCM patients after gross total resection had lower recurrence rate and longer PFS time than those after subtotal resection [6, 7]. This is consistent with the results of our study (Fig. 5). To avoid distant tumor recurrence, Kuster et al. [23] suggested minimizing tumor cell spread during the CCM operation; Kobayashi et al. [24] suggested resecting tumors en bloc, without rupturing the tumor capsule. Some studies found that childhood patients seemed to have higher recurrence rates and shorter PFS time [25, 26]. However, in our study, age (≤20 years or >20 years) did not have a significant effect on the PFS time (p = 0.9). The role of adjuvant radiotherapy in CCM remains controversial [7, 25]. Due to CCM’s aggressive behavior, some authors suggested postoperative radiotherapy (gamma-knife surgery or common radiotherapy) despite total resection of the tumor [27, 28]. Postoperative radiotherapy could get satisfactory treatment effects in some CCMs; some tumors could even decrease in the volume [29]. However, some tumors still recurred despite radiotherapy [1, 25]. In our study, 9 patients received postoperative radiotherapy, and they did not have significantly longer PFS time than other patients. Considering the potential harm of radiotherapy, especially in childhood patients, some authors suggested radiotherapy reserved for cases of local recurrence or subtotal resection [20, 26, 30]. Ide et al. [31] and Hori et al. [27] reported that radiosurgery with a marginal dose of 16 Gy might not be sufficient to control CCM. On the other hand, chemotherapy does not appear to have a significant role in the management of CCM, although it is also a suggested adjuvant therapy [10]. In our study, only one patient received chemotherapy after the recurrence; he did not recover well and died of the tumor (patient 9). Considering CCM’s propensity for distant relapse, postoperative regular MRI scans of the entire neuraxis should be performed every 3–6 months for the first several years to monitor for recurrence [7, 28].

Zorludemir et al. [2] reported that no histological predictors were associated with recurrence or clinical outcome in CCM. CCMs usually show benign histological characteristics, and CCM with anaplastic features have been reported occasionally [28, 32, 33]. Some CCMs could progress to anaplastic meningiomas after recurrence [27]. The relationship between prognosis and anaplastic histological features in CCMs still remains to be clarified [28]. Some studies found that CCMs with higher Ki-67 indexes were more likely to recur [7]. However, some studies did not find this association [1, 9]. In our study, the Ki-67 indexes did not significantly affect the patients’ PFS time (≤5 % or >5 %, p = 0.4). Kwon et al. [34] reported that high expression of MMP-9 tended to decrease the PFS time for CCM, although not significantly. PR expression was documented in 77 % of CCM [11]; Kobayashi et al. [24] thought PR positivity might predict a favorable prognosis for CCM. In our study, PR staining was available in 16 patients (positivity rate 50 %), and this did not significantly affect the patients’ PFS time. Besides, some authors thought that brain invasion of the CCM was accountable for high recurrence rate [7].

Histologically, CCM appears as a solid cellular neoplasm mostly composed of clear glycogen rich cells [8]. CCM should be differentiated from other lesions with clear cell appearance, such as oligodendrogliomas, hemangioblastomas, seminomas, pleomorphic xanthoastrocytomas, lipid-rich glioblastomas, renal cell carcinoma metastases, and ependymomas [35]. The mechanisms underlying the aggressive clinical behavior of CCM still need to be elucidated. Prayson et al. [21] reported that even a minor clear cell component (10 %) in meningiomas might indicate a higher risk of recurrence. The etiology of CCM is still not clear. Some genetic mutations may be implicated in the CCM tumor genesis. SMARCE 1 mutation has been found in some CCMs [14, 36]. CCMs have also been reported in some NF2 patients [2, 8], and Hartmann et al. [37] reported 22 % NF2 mutations in CCMs by single strand conformational polymorphism. Some CCMs were found in patients with multiple meningiomas [38, 39]. Familial occurrences of CCM have also been reported occasionally [13, 40].

Conclusions

The proportion of CCMs in spinal meningiomas is likely to be much larger than that in intracranial meningiomas. CCM has a female predilection, and the ratio of female to male in CCM is similar to that in total meningiomas. The mean age at diagnosis of CCM patients was significantly younger than that of the total meningioma patients. CCM patients have a high rate of recurrence. CCMs should be resected totally when possible to decrease the risk of recurrence or prolong the patient’s PFS time.