Meningioma classification by immunohistochemistry: A replicability study

Abstract

Introduction: Meningiomas account for nearly 40% of intracranial tumors. Recently, the immunohistochemistry (IHC) markers S100B, SCGN, ACADL and MCM2 have been shown to be associated with underlying biological subtypes of meningioma (MG1-MG4). We aimed to evaluate these IHC markers in a clinical setting.

Research question: Are the new proposed IHC markers clinically useful?

Methods: In total, 244 patients with meningiomas with tissue in TMAs were included and the IHC markers S100B, SCGN, ACADL and MCM2 were analyzed. Two sets of analyses were performed; the first included all samples with any staining considered positive, the second only samples with >10% immunopositivity. PFS and OS were analyzed in correlation to immunopositivity in the second analysis set.

Results: In the first set of analyses only 26.2% of samples could be to allocate to one group. No further analyses were performed with this selection. In the second set of analyses 52.0% could be allocated to a group. There was an enrichment of WHO grade 2 and 3 tumors in MG3 and MG4 as compared to MG1 (24.1% and 25.7% vs. 12.1%). Both the molecular group (p ​= ​0.032) and WHO grade (p ​= ​0.005) had significant impact on PFS, but only WHO grade predicted OS (p ​= ​0.033).

Conclusion: We studied the proposed new method of classifying meningiomas into groups MG1, MG2, MG3 and MG4 using IHC markers, but found difficulties applying the classification system in our material mainly due to lack of exclusivity of markers. Thus, in its present form the classification method lacks clinical applicability.

Keywords: Immunohistochemistry; Meningioma; Molecular marker; Recurrence.

Fig. 1

Immunohistochemical staining of S100B, SCGN, ACADL and MCM2 in six different tumor samples. Sample A: All negative; Sample B: S100B negative, the rest positive; Sample C: S100B and SCGN negative, ACADL and MCM2 positive; Sample D: S100B positive, the rest negative; Sample E: ACADL positive, the rest negative; Sample F: MCM2 positive, the rest negative.

Fig. 2

a) PFS amongst molecular group 1, 3 and 4∗, n ​= ​126 (p ​= ​0.032). b) PFS amongst WHO grade 1–3∗∗, n ​= ​127 (p ​= ​0.005). c) OS amongst molecular group 1, 3 and 4∗, n ​= ​126 (p ​= ​0.302). d) OS amongst WHO grade 1–3∗∗, n ​= ​127 (p ​= ​0.033). ∗MG2 excluded from analysis due to only one patient assigned to this group. ∗∗ Classified according to WHO classification used at time of surgery (2000, 2007 or 2016).

Fig. 3

a) PFS amongst molecular group 1, 3 and 4∗ when stratified for tumors with WHO grade 1, n ​= ​99 (p ​= ​0.137). b) PFS amongst molecular group 1,3 and 4∗ when stratified for tumors with WHO grade 2, n ​= ​25 (p ​= ​0.269). c) OS amongst molecular group 1, 3 and 4∗ when stratified for tumors with WHO grade 1, n ​= ​99 (p ​= ​0.587). d) OS amongst molecular group 1,3 and 4∗ when stratified for tumors with WHO grade 2, n ​= ​25 (p ​= ​0.177). ∗MG2 excluded from analysis due to only one patient assigned to this group.

Fig. 4

a) PFS among molecular group 1, 3 and 4∗ when stratified for tumors with Simpson grade 1 ​at resection (p ​= ​0.34), b) PFS among molecular group 1, 3 and 4∗ when stratified for tumors with Simpson grade 2 ​at resection (p ​= ​0.49) c) OS among molecular group 1, 3 and 4∗ when stratified for tumors with Simpson grade 1 ​at resection (p ​= ​0.6), d) OS among molecular group 1, 3 and 4∗ when stratified for tumors with Simpson grade 2 ​at resection (p ​= ​0.26). ∗MG2 excluded from analysis due to only one patient assigned to this group.