Chromosomal aberrations in atypical and anaplastic meningiomas: A fluorescence in situ hybridization study
Correspondence Address: Source of Support: SERB, Department of Science and Technology,, Conflict of Interest: None DOI: 10.4103/0028-3886.161994
Source of Support: SERB, Department of Science and Technology,, Conflict of Interest: None
Objective: There is significant variability in the biologic behavior of meningiomas, especially of atypical and anaplastic meningiomas, that cannot be accounted for by just histology and grade of excision. The aim of our study was to analyze deletions in regions 22q, 18p11, 1p32, and 14q32 in grade II and grade III meningiomas and their correlation with tumor grade and recurrence.
Keywords: Anaplastic meningioma; atypical meningioma; chromosomal aberration; fluorescent in situ hybridization
Meningiomas, originating from the arachnoid cap cells of brain and spinal cord, form 13% to 26% of all primary intracranial tumors.  Grades II (atypical) and III (anaplastic) variety comprise 20% of cases, are more aggressive and are associated with increased morbidity and mortality. The extent of resection and histopathologic grading determine the recurrence and outcome. , Some tumors, by virtue of their location or due to their close relation to vital structures, cannot be totally excised. Additional histopathologic, immunohistochemistry, and genetic markers may help in predicting tumor behavior and in guiding management.
Atypical meningiomas (WHO grade II) have an 8-fold increased risk of recurrence, with a statistically significant increased risk of mortality (21% at 5 years). , Anaplastic meningiomas (WHO grade III) are rare (1-2%), with recurrence rates of 50-80% after surgical resection, and an average survival rate of <2 years.  In the WHO classification, the Chordoid, Clear cell, and Brain invasive meningiomas are considered grade II while the Papillary and Rhabdoid meningiomas are classified as grade III.
The 5-year recurrence rates are 5% and 41% for grade I and grade II tumors, respectively, even following Simpson's grade I resection, and this high recurrence rate is associated with an increased risk of death.  Malignant meningiomas have high recurrence rates with an overall median survival of 2 years.  In grade III meningiomas, surgery followed by high-dose fractionated radiotherapy are associated with a recurrence risk of 80% at 5 years. While the extent of resection and histological grade are the strongest predictors of recurrence, the significant variability in biologic behavior of these tumors cannot be completely accounted for. The molecular pathogenesis has highlighted genes that serve as newer targets to improve diagnostic and therapeutic strategies, particularly for patients with aggressive meningiomas which are resistant to the conventional forms of therapy.
Multiple chromosomal aberrations have been utilized as markers for outcome and prognosis; there is a debate about the significance of each marker and their combinations. Losses of 1p, 3q, 6q, 9p, 10q, 14q, 17p, 18p, 18q, and 22q have been identified in grade II and III tumors; whether they have a prognostic significance remains to be proven. Monosomy of chromosome 22 is the most common genetic alteration in meningiomas found in almost half of the cases and is the first cytogenetic alteration described in solid tumors.  Anaplastic meningiomas share these chromosomal aberrations, but show more frequent losses on 6q, 10q, and 14q, additional losses on 9p, and gains or amplifications on 17q23. ,,, In this study, we analyze 22q, 18p11, 1p32, and 14q32 deletions in grade II and grade III meningiomas and their correlation with tumor recurrence. This is the first study from India which has studied these 4 sets of chromosomal aberrations. We hope to correlate genetic deletions in grade II and grade III meningiomas with prognostic significance.
NF2 gene (22q12.2) product belongs to the 4.1 family and has been implicated in the initiation of a meningioma. Loss of protein 4.1B (DAL-1) at 18p11.32 is found predominantly in higher grades.  Allelic losses have been reported with frequencies between 20% and 70%. , The latter represents the marker of an early progression rather than that of initiation of the tumor. Cyclin-Dependent Kinase Inhibitor 2C (CDKN2C) (encoding p18INK4c) is a candidate gene on 1p32.3 screened with inconsistent results, and in rare cases, showing deletions in atypical or anaplastic meningiomas.  Maternally-expressed-gene-3 (MEG3), an anti-proliferative factor  coded from 14q32 region, has a potential role in meningioma progression.
All operated cases of grade II and III meningiomas from 2001 to 2007 at our institute were included. Clinical, demographic, radiological and surgical details as well as the latest status of the patients were recorded. The study was approved by the Institutional Ethics Committee and a detailed informed consent was taken from the patients.
Paraffin blocks of selected cases were retrieved from the archives. Sections were cut and stained with hematoxylin and eosin. Their reclassification into grades was done according to the 2007 Wold Health Organisation (WHO) classification of meningiomas.
Locus specific fluorescence in situ hybridization (FISH) probes were chosen for:
• Neurofibromin 2 (NF 2, 22q12.2),
• DAL-1 or 4.1B gene or Erythrocyte Membrane Protein Band 4.1-Like 3 (EPB41 L3, 18p11.3),
• MEG-3, (14q32.2) and
• CDKN2C (1p32.3).
Highly cellular representative areas were selected and a single color FISH assay was performed. Unstained sections were cut from formalin-fixed paraffin-embedded blocks, micro waved for fixing, de-paraffinized by keeping slides in xylene, then washed with absolute alcohol, placed in Lugol's iodine for 5 min and agitated in 2.5% sodium thiocyanate. Slides were placed in 10 mM citrate (pH 6) and micro waved for 5 min at 850 watts. Following this, the sections were digested by 0.2 g pepsin. They were dehydrated in different grades of ethanol and air dried. A single fluorochrome-labeled probe (EMPIRE GENOMICS, Buffalo, NY) was added and placed on a ThermoBrite™ hybridization chamber (Vysis) and the program was set for de-naturation and hybridization of both the tissue and probe (denatured at 80°C for 3 min and hybridized for 24 h at 37°C). Posthybridization slides were washed in 2XSSC nuclei and were counterstained with 4',6-diamino-2-phenylindole [DAPI] (Vysis) nuclear stain. The sections were studied with a trinocular research fluorescent microscope (Olympus Model: BX51; Olympus, Tokyo) equipped with a set of appropriate filters (DAPI, tetramethylrhodamine isothiocyanate (TRITC) and fluorescein isothiocyanate [FITC]). Signals were scored in at least 100 nonoverlapping intact nuclei by three investigators. Every case was examined and the abnormalities were divided depending upon regional distribution of chromosomal aberrations (homogenous or heterogeneous) and the type of deletion (homozygous or heterozygous) [Figure 1].
Statistical analysis was done using SPSS 20 (IBM, Chicago). Nonparametric Spearman rank correlation coefficient was used for assessing the degree of linear correlation between pairs of variables. Mann-Whitney U-test compared mitotic count markers in recurrent and non-recurrent tumors. Kaplan-Meier multivariate analysis was done for survival using Cox proportional hazard. We conducted all statistical tests employing a significance level of 0.05. Significant correlations between two parameters were taken at 95% confidence interval.
There were a total of 57 patients, of which 7 were reclassified to grade I. The remaining 50 patients (26 males and 24 females) with a mean age of 45 ± 15 years (minimum age = 13 years, maximum age = 75 years) were included. 59 samples were available for histopathological and genetic analysis. The mean follow-up period was 67.4 months (range 27-242 months). There were 34 skull vault and 16 basal meningiomas. Two patients underwent grade 0, 11 grade 1, 21 grade 2, 13 grade 3, 3 grade 4 excision.
Thirty three patients had a recurrence out of 50 patients available at the median period of 53 months (4-216 months) after the first surgery; 20 of these were reoperated. Nine cases re-recurred after the second surgery at the median period of 34 months (6-91 months) out of which 4 required surgery again.
The following subgroups were analyzed:
Paired sample analysis was done comparing deletion states of tumors at primary surgery with tumors at recurrence. There was a statistically significant increase in the frequency of deletions in all the sites at first recurrence as compared to the primary cases. There was a progressive increase in the number of deletions in all the sites, from primary surgery through recurrence and re-recurrence [Table 1].
Paired sample analysis was done comparing the two cohorts following primary excision of the tumor. Deletion states of tumors which recurred (20/37) were compared to tumors which did not recur (17/37). [Table 2] shows that tumors which recurred had a clear increase in the frequency of deletion in all the studied sites compared to tumors which did not recur.
Cases with 18p deletion had a trend towards occurring in a younger population [Table 3]. There were no statistically significant differences in the deletion status in samples of either sex or of those occurring in different locations. Though the frequency of deletions was more in anaplastic meningiomas compared to the atypical ones, it was statistically not significant due to the small group of anaplastic meningiomas (n = 4). 22q deletions had a trend towards occuring in hypercellular tumors. 18p11 deletions had a significant association with sheeting. Samples with mitosis ≥7 had a statistically significant association with 1p, 14q, and 1p14q codeletion. The rest of the histopathologic features did not have a statistically significant relation with the deletion states. The mean mitotic index in samples having the deletion was significantly higher than in cases without the deletion [Table 4].
The progression-free survival was calculated taking 37 cases where samples of primary surgery were available. Twenty (54.1%) of these cases had recurred (median recurrence free survival [RFS] was 50.5 months; range 9-216 months). Of the 37 cases where survival analysis done, the mean follow-up was 67.4 months (range 27-232 months). Seventeen cases had no recurrence at a mean follow-up period of 63.8 months (range 27-137 months).
The median overall survival was 60 months (range 27-132 months). Better grades of excision had higher estimated overall survival time. The total estimated survival period increased from grade 0 to grade IV [Figure 2]a and b. The overall survival and time for recurrence was not significantly correlating with the deletion status [Figure 2]c and d, [Table 5] and [Table 6]. All cases that had shown recurrence for the 2 nd time (n = 4) had deletions in all 4 sites.
Three patients had a chordoid meningioma out of which samples of two primary tumors and only one recurrent tumor were available. All three samples had all 4 sites deleted. Three patients had an anaplastic meningioma and only one had both primary and recurrent samples while the others had only recurrent samples available. All 4 samples had deletions in all 4 sites.
There is a significant variability in the biologic behavior of meningiomas that cannot be accounted for by just histology and grade of excision. There remains considerable variability in clinical outcomes within each histologic grade, especially among atypical meningiomas. Existing criteria do not adequately predict the rates of tumor growth or the likelihood of tumor recurrence.
Frequency of loss of heterozygosity (LOH) in 1p, 6p, 9q, 10q, and 14q increases with tumor progression.  1p and 14q deletions are associated with meningioma progression and increasing histologic grade.  Pascal et al. studied LOH on chromosome 22 and 1 and found LOH in 50% and 28% samples, respectively. Weber et al.  studied for chromosomal imbalances by comparative genomic hybridization. In grade II tumors, losses in 1p (76%), 22q (71%), 14q (43%), 18q (43%) were observed, and in grade III tumors, most of these alterations were found at similar frequencies, but an increase in losses on 14q (63%) was observed.
Pfisterer et al.  in their study of genetic abnormalities in chromosomes 1, 14, and 22 in 77 meningioma samples detected by the FISH technique, found aberrations to be significantly associated with the radiographic evidence of recurrence. Maillo et al.  studied 70 meningiomas of all grades by the FISH technique and observed that tumor grade, chromosome 14 status and age were the best combination of independent variables for predicting recurrence free survival (RFS). Asirvatham et al.  studied 27 meningiomas for LOH in 22q, 10q, 14q, and 17p. They found LOH in 33% atypical and 50% anaplastic meningiomas but without statistical significance. Kumar et al.  in their FISH study on 64 convexity and 20 petroclival meningiomas found a significant progressive increase in deletions with the increasing grade of meningiomas. Two of the 10 (20%) petroclival tumors with STR showed a 1p/14q codeletion. However, no case with gross total resection showed the codeletion. They suggested that petroclival tumors harboring 1p/14q codeletion, a genetic marker for aggressive behaviour, should be treated with caution. Jansen et al.  studied 86 completely resected atypical meningiomas using the dual-color interphase FISH and clarified the association of 1q gain with reduced progression-free survival.
Nunes et al. studying 63 sporadic meningiomas implicated the DAL-1/4.1B locus in sporadic meningiomas less commonly than reported previously and suggested that it is a progression rather than an initiation locus. Martinez-Glez et al. showed a low mutational frequency of variations in DAL-1/4.1B gene in the neoplastic transformation of meningiomas and suggested that inactivating mechanisms like epigenetic changes may be participating in the DAL1/4.1B silencing. 
In 6 different studies done on patients with grade II and III meningiomas, the 5-year overall survival rates ranged from 28% to 91%.  The overall survival was 81.3 (SE 53.8) months in this study, which was better compared to that in the available literature. This may be explained by the higher number of Simpson's grade 1 excisions. Ko et al. and Palma et al. have reported that a greater extent of resection of an atypical or anaplastic meningioma improved survival.  In this study, there was a gradual increase in the overall survival with a better grade of excision (P = . 017), but the recurrence free survival was not statistically significant owing to the presence of only a few cases that recurred (P = . 141).
There was no single study available in literature where the FISH study was done simultaneously for these sites. There was a distinct increase in the frequency of deletions in all the sites associated with tumors that recurred. In recurrent tumors, the frequency of deletions was increased compared to that seen in primary tumors but the results were not statistically significant. This may be due to the fewer number of patients in different categories as well as due to the large number of deletions identified when compared to the fewer samples of non-deleted tumors.
Perry et al.  and Pasquier et al.  have reported a high mitotic count to be a negative prognostic marker. Vranic et al.  used mitotic counts to stratify further risk of recurrence. In this study, the mean mitotic count was 7.81 (median 6, range 4-20) which is comparable to that quoted in the available literature. The overall survival of patients with a mitotic index ≥7 was 54 (standard error [SE] 5.6) months; in comparison, the overall survival in those patients with a mitotic count <7 was 75 (SE 24.2) months (P = 0.122) [Figure 2]c and d. The mean mitotic indices with deletion in 22q, 18p, 14q, and 1p14q codeletion were significantly higher compared to samples which did not have the deletion (P = 0.017, 0.001, 0.001, 0.004, <0.001, respectively).
Cai et al.  reported an overall survival of 6.7 years in atypical and anaplastic meningiomas. The 5-year and 10-year survival of atypical meningiomas was 77.6% and 59.5%, respectively; and, in cases having an anaplastic meningioma, the survival was 37.3% and 26.6%, respectively. The overall survival for patients with 14q deletions was low both in atypical and anaplastic meningiomas. Maillo et al. showed abnormalities of chromosome 22 (61%), as well as chromosomes 1 (40%) and 14 (33%). The tumor grade, numerical abnormalities for chromosome 14 and age were the best combination of independent variables for predicting recurrence free survival in meningioma patients. Our results of survival were comparable to the results of Cai et al. but the results were not statistically significant.
There was progressive increase in the frequency of deletions with each recurrence and cases that recurred had an increased frequency of deletions compared to the cases that did not. There was an increase in the frequency of deletions with an increase in histologic grade of the tumor. The mean mitotic index of patients with deletion was significantly higher compared to that in patients without deletions. This indirectly reflects the higher grade and consequently, a poorer prognosis. This is a first study analyzing deletions in all these four sites using the FISH technique. The significance of these deletions in predicting recurrence and survival needs further studies with larger cohorts of patients along with their clinical correlation. This would help in categorizing patients with a high risk of recurrence and thus, help in guiding their clinical management.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]