Distribution of S100 immunopositivity:
After the immunohistochemical assessment by two researchers, a mismatch result was documented in 43 cases for which a consensus was found. The interrater consensus was 97.4% (1626/1669). Cohen’s Kappa was 0.88 (almost perfect agreement).
Among 1669 meningioma tissue samples, strong S100 immunopositivity was observed in 218 cases (13.1%). Staining was mainly cytoplasmic but additional nuclear S-100 positivity was frequently observed in tumor cells with cytoplasmic staining (Fig. 1D). Lymphocytes and tumor vessels were negative for S100. There was a significant difference with a higher rate of S100 positive meningiomas in female patients (15.0% compared to 8.3%, p = 0.0003). There was no staining difference regarding the CART-specific cut off at 41.5 years. Regarding tumor location, spinal meningiomas showed the highest rate of S100 positivity (23.8%), followed by convexity/falx (18.2%) and the lowest rate for skull base meningiomas (7.1%, p < 0.0001). The frequency of S100 immunopositivitiy was approximately twice as high in primary meningiomas compared to recurrent meningiomas (13.8% vs. 7.6%, p = 0.0147). The rate of S100 positive cases was almost 3-times as high in meningiomas from patients suffering from NF2 (11.0% vs. 4.1%, p < 0.0001). With higher CNS WHO grading the frequency of S100 immunopositivity decreased (CNS WHO grade 1: 14.2%, grade 2: 9.3%, grade 3: 0%, p = 0.109). Among the fifteen different histological subtypes marked differences were observed (p < 0.0001). Especially high S-100 positive rates were observed for transitional (22.5%), fibroblastic (24.8%) and psammomatous meningiomas (36.2%). There was no significant difference in S100 status regarding MIB1 immunopositivity. S100 expression was less frequent in cases that experienced tumor recurrence after meningiomas resection (7.3% vs. 14.3%, p = 0.0011). The distribution of S100 immunopositivity is displayed in Table 1.
Table 1
Distribution of S100 expression according to clinical and histopathological characteristics (Pearson’s chi-squared test).
| | S100 expression (N, %) | |
| N (%) | Positive | Negative | p-value |
Gender Female Male Age >/=41.5 < 41.5 Location Skull base Convexity/falx Spinal Prim/Rec Primary Recurrent NF2 Yes No Simpson grade I/II/III IV/V CNS WHO grading 1 2 3 Histology 1 Angiomatous Fibroblastic Lymphocyte rich Meningothelial Metaplastic Microcystic Psammomatous Secretory Transitional NOS 2 Atypical Chordoid Clear Cell 3 Anaplastic Papillary Rhabdoid MIB1 <=4.6% or missing >4.6% Tumor recurrence Yes No Adjuvant RT Yes No | 1197 (71.7) 472 (28.3) 1456 (87.2) 213 (12.8) 850 (50.9) 659 (39.5) 160 (9.6) 1471 (88.1) 198 (11.9) 218 (13.1) 1451 (86.9) 1159 (70.9) 475 (29.1) 1323 (79.3) 323 (19.4) 23 (1.4) 34 (2.0) 125 (7.5) 1 (0.1) 823 (49.3) 20 (1.2) 32 (1.9) 58 (3.5) 46 (2.8) 169 (10.1) 16 (1.0) 290 (17.4) 32 (1.9) 0 (-) 17 (1.0) 0 (-) 6 (0.4) 1362 (85.5) 231 (14.5) 314 (21.4) 1155 (78.6) 71 (4.8) 1401 (95.2) | 179 (15.0) 39 (8.3) 184 (12.6) 34 (16.0) 60 (7.1) 120 (18.2) 38 (23.8) 203 (13.8) 15 (7.6) 24 (11.0) 59 (4.1) 173 (14.9) 41 (8.6) 188 (14.2) 30 (9.3) 0 (-) 0 (-) 31 (24.8) 0 (-) 92 (11.2) 4 (20.0) 2 (6.3) 21 (36.2) 0 (-) 38 (22.5) 0 (-) 30 (10.3) 0 (-) 0 (-) 0 (-) 0 (-) 0 (-) 183 (13.4) 25 (10.8) 23 (7.3) 165 (14.3) 1 (1.4) 188 (13.4) | 1018 (85.1) 433 (91.7) 1272 (87.4) 179 (84.0) 790 (92.9) 539 (81.8) 122 (76.3) 1268 (86.2) 183 (92.4) 194 (89.0) 1392 (95.9) 986 (85.1) 434 (91.4) 1135 (85.8) 293 (90.7) 23 (100.0) 34 (100.0) 94 (75.2) 1 (100.0) 731 (88.8) 16 (80.0) 30 (93.8) 37 (63.8) 46 (100.0) 131 (77.5) 16 (100.0) 260 (89.7) 32 (100.0) 0 (-) 17 (100.0) 0 (-) 6 (100.0) 1179 (86.6) 206 (89.2) 291 (92.7) 990 (85.7) 70 (98.6) 1213 (86.6) | 0.0003* 0.1786 < 0.0001* 0.0147* < 0.0001* 0.0006* 0.0109* < 0.0001* 0.2756 0.0011* 0.0032* |
Nominal logistic regression of factors associated with S100 expression:
The multivariate assessment of clinical variables that potentially influence the S100 expression was done with a nominal logistic regression. Female gender, convexity/falx and spinal meningioma location as well as NF2 and WHO grade 1 were associated with higher immunopositivity rates for S100 expression. Patient age at diagnosis MIB1 immunopositivity and recurrent tumor status did not show an independent association with S100 expression. Details of the nominal logistic regression are displayed in Table 2.
Table 2
Nominal logistic regression of factors associated with S100 expression.
| Odds Ratio (95%CI) | p-value (Prob > Chisq) |
Female Gender Age Location Convexity/falx vs. skull base Spinal vs. skull base Spinal vs. convexity/falx Primary vs. recurrent meningioma NF2 vs. sporadic CNS WHO grading 1 vs. 2 1 vs. 3 2 vs. 3 MIB1 | 2.08 (1.39–3.09) - 3.40 (2.39–4.82) 3.42 (2.12–5.49) 1.01 (0.64–1.58) 1.42 (0.78–2.60) 4.62 (2.44–8.74) 1.82 (1.13–2.94) - - - | 0.0003* 0.1715 < 0.0001* < 0.0001* 0.9804 0.2562 < 0.0001* 0.0133* 0.9973 0.9974 0.5954 |
S100 and progression-free survival:
Information on radiographic tumor recurrence or progression was available for 1469 cases (88.0%) with a mean follow-up of 38.3 months ranging from 1.1 to 195.6 months. Tumor recurrence/progression was observed in 314 cases (21.4%).
Male patients had a higher rate of tumor recurrence compared to female patients (32.4% vs. 17.2%, p < 0.0001). At the CART-specific age cut off based on the maximum difference regarding tumor recurrence, patients younger than 41.5 years at diagnosis had a significantly higher rate of tumor recurrence (36.4% compared to 19.2%, p < 0.0001). Spinal meningiomas showed a significantly lower rate of tumor recurrence (6.3%) than meningiomas with convexity/falx or skull base location (23.8% and 22.1%, respectively, p < 0.0001). Recurrent meningiomas had a 4-fold risk of another recurrence when compared with primary tumors (63.3% vs. 15.6%, p < 0.0001). A small subgroup of patients suffering from NF2 were included in this cohort (n = 72). Such meningiomas had a higher rate of recurrence (38.9% compared to 20.5%, p = 0.0002). A higher grade of tumor resection without residual meningioma tissue (Simpson grade 1–3) as well as a lower WHO grade were associated with a markedly lower tumor recurrence rate (each p < 0.0001). An expression of the proliferative marker MIB1 reaching 4.6% or beyond was associated with a higher rate of tumor recurrences (17.3% compared to 43.8%, p < 0.0001). Tumors with immunopositivity for S100 were found to have a lower risk of recurrence (12.2% compared to 22.7%, p = 0.0011). The detailed results of the univariate analysis is shown in Table 3. In the Kaplan-Meier analysis the univariate prognostic impact was confirmed for all examined factors except adjuvant radiotherapy, as seen in the respective Kaplan-Meier curves in Figs. 2 and 3.
Table 3
Univariate analysis of factors associated with tumor recurrence (Pearson’s chi-squared test).
| | Tumor recurrence (N, %) | |
| N (%) | Yes | No | p-value |
Gender Female Male Age >/=41.5 < 41.5 Location Skull base Convexity/falx Spinal Prim/Rec Primary Recurrent NF2 Yes No Simpson grade 1/2/3 4/5 CNS WHO grading 1 2 3 Histology 1 Angiomatous Fibroblastic Lymphocyte rich Meningothelial Metaplastic Microcystic Psammomatous Secretory Transitional NOS 2 Atypical Chordoid Clear Cell 3 Anaplastic Papillary Rhabdoid Adjuvant radiotherapy Yes No MIB1 <=4.6% or missing >4.6% S100 expression Yes No | 1064 (72.4) 405 (27.6) 1282 (87.3) 187 (12.7) 752 (51.2) 589 (40.1) 128 (8.7) 1292 (88.0) 177 (12.1) 72 (4.9) 1397 (95.1) 1018 (70.9) 418 (29.1) 1169 (79.6) 278 (18.9) 22 (1.5) 29 (2.0) 112 (7.6) 1 (0.1) 729 (49.6) 20 (1.4) 28 (1.9) 48 (3.3) 42 (2.9) 146 (9.9) 15 (1.0) 248 (16.9) 29 (2.0) 0 (-) 16 (1.1) 0 (-) 6 (0.4) 71 (4.8) 1398 (95.2) 1200 (85.1) 210 (14.9) 188 (12.8) 1281 (87.2) | 183 (17.2) 131 (32.4) 246 (19.2) 68 (36.4) 166 (22.1) 140 (23.8) 8 (6.3) 202 (15.6) 112 (63.3) 28 (38.9) 286 (20.5) 137 (13.5) 171 (40.9) 167 (14.3) 128 (46.0) 19 (86.4) 2 (6.9) 11 (9.8) 0 (-) 118 (16.2) 0 (-) 4 (14.3) 2 (4.2) 5 (11.9) 23 (15.8) 3 (20.0) 116 (46.8) 11 (37.9) 0 (-) 15 (93.8) 0 (-) 4 (66.7) 25 (35.2) 289 (20.7) 208 (17.3) 92 (43.8) 23 (12.2) 291 (22.7) | 881 (82.8) 274 (67.7) 1036 (80.8) 119 (63.6) 586 (77.9) 449 (76.2) 120 (93.8) 1090 (84.4) 65 (36.7) 44 (61.1) 1111 (79.5) 881 (86.5) 247 (59.1) 1002 (85.7) 150 (54.0) 3 (13.6) 27 (93.1) 101 (90.2) 1 (100.0) 611 (83.8) 20 (100.0) 24 (85.7) 46 (95.8) 37 (88.1) 123 (84.3) 12 (80.0) 132 (53.2) 18 (62.1) 0 (-) 1 (6.3) 0 (-) 2 (33.3) 46 (64.8) 1109 (79.3) 992 (82.7) 118 (56.2) 165 (87.8) 990 (77.3) | < 0.0001* < 0.0001* < 0.0001* < 0.0001* 0.0002* < 0.0001* < 0.0001* < 0.0001* 0.0036* < 0.0001* 0.0011* |
Cox proportional hazard of progression-free survival:
For the multivariate analysis of meningioma recurrence/progression a Cox proportional hazard model was applied. Female gender remained an independent protective factor (p = 0.015) while patient age of 41.5 years or older merely showed a prognostically positive trend without statistical significance (p = 0.0627). Meningioma location and NF2 also missed statistical significance regarding progression-free survival. Primary tumor status and lower WHO grade were independent positive prognostic factors leading to a favorable progression-free survival (each p < 0.0001). Resection of all visible tumor tissue (Simpson 1–3) and adjuvant radiotherapy were treatment modalities that also showed and independent positive prognostic effect (each p < 0.0001). Expression of the proliferation marker MIB1 exceeding 4.6% was associated with a two-fold risk of tumor recurrence (p < 0.0001) while S100 immunopositivity had no independent influence on progression-free survival (p = 0.6140). Details of the Cox proportional hazard analysis can be seen in Table 4).
Table 4
Multivariate analysis of prognostic factors of tumor recurrence (Cox proportional hazard).
| Risk Ratio (95%CI) | p-value (Prob > Chisq) |
Female Gender Age > = 41.5 Location Convexity/falx vs. skull base Convexity/falx vs. spinal Skull base vs. spinal Recurrent vs. primary meningioma NF2 vs. sporadic Simpson Grade 1–3 vs. 4/5 CNS WHO grading 1 vs. 2 1 vs. 3 2 vs. 3 Adjuvant radiotherapy MIB1 > 4.6% S100 immunopositivity | 0.73 (0.56–0.94) 0.73 (0.53–1.02) 0.80 (0.62–1.03) 1.44 (0.66–3.12) 1.80 (0.84–3.88) 3.46 (2.61–4.59) 0.84 (0.51–1.41) 0.34 (0.26–0.44) 0.34 (0.25–0.45) 0.06 (0.03–0.12) 0.19 (0.10–0.24) 0.33 (0.21–0.50) 2.15 (1.61–2.86) 0.89 (0.56–1.41) | 0.0150* 0.0627 0.0853 0.3565 0.1312 < 0.0001* 0.5182 < 0.0001* < 0.0001* < 0.0001* < 0.0001* < 0.0001* < 0.0001* 0.6140 |