Search results and included studies

The initial search identified 1970 studies; 1878 were excluded based on titles and abstracts. Of the remaining 92, 72 lacked sufficient PFS data or had fewer than three patients, leaving 20 studies for meta-analysis (Fig. 1). Published between 2011 and 2024, these included six from China, four from the USA, two each from India and Switzerland, and one each from Tunisia, Mexico, Iran, France, Brazil, and Australia. Unlike adults, PMs show a nearly equal male-to-female distribution (52.1% males, 47.9% females). The NF status and WHO grade distribution were analyzed in 252 patients. Among WHO grade 1 meningiomas, 86.2% were sporadic, 1.4% had NF type 1, and 12.4% had NF type 2, representing 57.5% of the total cohort. WHO grade 2 meningiomas included 88.5% without NF and 11.5% with NF type 2, comprising 31.0% of the cohort, while grade 3 meningiomas were sporadic in 89.7% and 10.3% with NF type 2, making up 11.5%. Overall, most patients across all WHO grades had sporadic tumors (87.3%), with NF2 cases relatively consistent across WHO grades. No detailed IPD on adjuvant chemotherapy was available. Nineteen studies were retrospective single-center analyses [14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33], and one was a meta-analysis [6] (Table 1).

Fig. 1

PRISMA flowchart for study selection

Progression-free survival and overall survival in pediatric and adolescent meningioma

Mean PFS for children diagnosed with meningioma at ages 0–3, 4–12, and 13–21 were 51.3 months (95%CI: 26.4–76.2), 115.1 months (95%CI: 99.2–130.9), and 158.9 months (95%CI: 121.7–196.1), respectively. PFS was significantly shorter in those diagnosed at 0–3 years compared with 4–12 years (p = 0.04) and 13–21 years (p = 0.03, Fig. 2a). OS stratified by age groups showed no significant differences: 261.7 months (95%CI: 217.8–305.6) for 0–3 years, 212.3 months (95%CI: 191.6–233.0) for 4–12 years, and 594.5 months (95%CI: 516.1–672.8) for 13–21 years (p = 0.29, Fig. 2b).

Fig. 2

Kaplan–Meier survival analysis for progression-free survival (PFS) and overall survival (OS) based on age, WHO tumor grade, and neurofibromatosis type (NF1/NF2) status. A, B: PFS and OS stratified by age groups: 0–3 years (blue), 4–12 years (orange), and 13–21 years (red). Statistical significance between age groups is shown in each panel. C, D: PFS (C) and OS (D) based on WHO tumor grade: Grade 1 (blue), Grade 2 (orange), and Grade 3 (red). Significant differences between the WHO grades are highlighted. E, F: PFS (E) and OS (F) stratified by neurofibromatosis status: No NF (blue), NF1 (orange), and NF2 (red). p-values between groups are displayed, indicating statistical comparisons. The number of patients at risk for each time point is provided below each panel. Shaded areas represent 95% confidence intervals

Progression-free survival and overall survival among WHO grades in pediatric and adolescent meningioma

WHO grading strongly differentiated PMs regarding PFS. Mean PFS for WHO grades 1, 2, and 3 were 209.8 months (95%CI: 197.0–222.7), 137.5 months (95%CI: 104.3–170.8), and 72.1 months (95%CI: 53.7–90.5), respectively (p < 0.001, Fig. 2c). OS showed no significant difference between grades 1 and 2 (p = 0.98). Five-, 10-, and 15-year OS probabilities for grade 1 were 94.3%, 87.3%, and 76.0%; for grade 2, 93.4%, 86.4%, and 81.9%; and for grade 3, 69.7% and 62.2% at five and 10 years. OS for grade 3 was significantly shorter than grades 1 (p < 0.001) and 2 (p < 0.001, Fig. 2d).

Progression-free survival and overall survival among sporadic or NF-associated meningiomas in children and adolescents

Among 767 patients, mean PFS for sporadic, NF1-, and NF2-associated PMs were 180.6 months (95%CI: 170.3–191.0), 59.7 months (95%CI: 43.6–75.8), and 138.4 months (95%CI: 97.5–179.5), respectively (p = 0.02, Fig. 2e). NF1- (p = 0.04) and NF2-associated PMs (p = 0.045) had significantly shorter PFS than sporadic PMs. OS did not differ significantly, with a median follow-up of 51.7 months (IQR: 23.1–100.2, Fig. 2f).

Progression-free survival and overall survival by extent of resection in pediatric meningioma

EoR significantly impacted PFS and OS across all grades. Mean PFS was 113.8 months (95%CI: 101.5–126.2) for GTR and 40.1 months (95%CI: 30.7–49.4) for STR (p < 0.001, Fig. 3a). Mean OS was 602.9 months (95%CI: 561.4–644.5) for GTR and 173.8 months (95%CI: 152.2–195.5) for STR (p < 0.001, Fig. 3b).

Fig. 3

Kaplan–Meier survival analysis for progression-free survival (PFS) and overall survival (OS) based on the extent of resection (GTR: Gross Total Resection, STR: Subtotal Resection) and WHO tumor grades. A, B: PFS (A) and OS (B) for WHO Grades 1–3, comparing outcomes between GTR (blue) and STR (orange). Significant differences in survival outcomes are indicated (p < 0.001). C, D: PFS (C) and OS (D) for WHO Grade 1, showing a clear survival benefit for GTR over STR (p < 0.001). E, F: Progression-free survival (E) and overall survival (F) for WHO Grade 2. GTR significantly improves PFS (p < 0.001), but the difference in OS is not statistically significant (p = 0.067). G, H: PFS (G) and OS (H) for WHO Grade 3. There is no statistically significant difference between GTR and STR for both PFS (p = 0.26) and OS (p = 0.27). The number of patients at risk at each time point is shown below each panel, and shaded areas represent 95% confidence intervals

PFS and OS by WHO grades and extent of resection in pediatric meningioma

Stratification by WHO grades revealed that GTR improved PFS and OS across all grades. For WHO grade 1, mean PFS was 243.6 months for GTR vs. 158.7 months for STR (p < 0.001, Fig. 3c), with GTR patients showing PFS probabilities of 89.6% at 60 months, 84.1% at 120 and 180 months, compared to 54.6%, 35.6%, and 28.5% for STR. Mean OS was 277.5 months for GTR vs. 177.9 months for STR (p < 0.001, Fig. 3d), with OS probabilities of 93.2% at 60 months, 84.9% at 120, and 83.5% at 180 months, vs. 79.8%, 62.4%, and 54.6% for STR.

For WHO grade 2, mean PFS was 128.2 months for GTR vs. 56.1 months for STR (p < 0.001, Fig. 3e), with GTR showing PFS probabilities of 100% at 60 months, 67.7% at 120, and 58.0% at 180 months, compared to 40.6% and 25.4% at 60 and 120 months for STR. OS probabilities for GTR were 94.4% at 60 months and 90.7% at 120 and 180 months, compared to 87.5%, 70.1%, and 52.6% for STR (p = 0.067, Fig. 3f).

For WHO grade 3, mean PFS was 75.3 months for GTR vs. 56.1 months for STR (p = 0.26, Fig. 3g), with GTR showing PFS probabilities of 41.4% at 60 months, 31.1% at 120, and 20.7% at 180 months, compared to 37.1% and 24.7% at 60 and 120 months for STR. Mean OS was 134.8 months for GTR vs. 116.1 months for STR (p = 0.27, Fig. 3h), with GTR showing OS probabilities of 93.9% at 12 months, 81.0% at 24, and 72.1% at 60 months, compared to 78.8%, 73.0%, and 66.7% for STR.

PFS and OS by adjuvant radiotherapy in pediatric WHO grade 2 and 3 meningiomas

PFS in WHO grade 2 meningiomas was higher without radiotherapy, with 60-month PFS of 72.7% vs. 20.8% with radiotherapy (p = 0.004). Mean PFS was 115.2 months (95%CI: 85.5–144.8) without radiotherapy vs. 47.9 months (95%CI: 23.4–72.4) with it (Fig. 4a). OS analysis (56 cases) showed a mean OS of 634.4 months (95%CI: 578.7–690.1) without radiotherapy vs. 194.7 months (95%CI: 132.9–256.5) with it (p = 0.43, Fig. 4b).

Fig. 4

Kaplan–Meier survival analysis for progression-free survival (PFS) and overall survival (OS) based on the use of radiotherapy in WHO Grade 2 and Grade 3 meningiomas. A&B: PFS (A) and OS (B) for WHO Grade 2 meningiomas, comparing patients who received radiotherapy (orange) versus those who did not (blue). Radiotherapy significantly improved PFS (p = 0.004), while no significant difference in OS was observed (p = 0.43). C, D: PFS (C) and OS (D) for WHO Grade 3 meningiomas, showing a significant improvement in both PFS (p = 0.009) and OS (p = 0.002) for patients who received radiotherapy. The number of patients at risk at each time point is displayed below each panel, and shaded areas represent 95% confidence intervals

In WHO grade 3 meningiomas (22 cases), radiotherapy improved outcomes significantly. At 60 months, PFS was 18.2% without radiotherapy vs. 55.4% with it, rising to 41.6% by 100 months (p = 0.009, Fig. 4c). Mean PFS was 23.8 months (95%CI: 0.28–47.4) without radiotherapy vs. 72.5 months (95%CI: 38.1–106.9) with it. OS was higher with radiotherapy: mean OS was 63.0 months (95%CI: 39.8–86.2) without radiotherapy vs. 140.7 months (95%CI: 115.8–165.6) with it, with 60-month survival of 44.9% vs. 85.1% (p = 0.002, Fig. 4d).

Subgroup analysis of reconstructed progression-free survival (PFS) data incorporating multiple shared covariates

Subgroup analysis of studies providing IPD with multiple covariates was performed [17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34]. Previous IPD analyses showed a role of adjuvant radiation therapy in aggressive meningiomas. Hence, PFS data of subtotally resected WHO grade 2 and 3 meningiomas was stratified regarding adjuvant radiation therapy. In pediatric WHO grade 1 meningiomas following subtotal resection, the 36-month progression-free survival (PFS) rate for patients receiving adjuvant radiation therapy was 72.2%, compared to 71.3% for those without radiation therapy. At 72 months, the PFS rate for the radiation therapy group declined to 36.1%, while the group without radiation therapy maintained a rate of 63.4%. The results, shown in Part A of Supplementary Fig. 1, indicate no statistically significant difference in PFS between the two groups (p = 0.23). The PFS analysis in pediatric WHO grade 2 and 3 meningiomas following subtotal resection with and without adjuvant radiation therapy showed that adjuvant radiation therapy significantly enhanced median PFS time to 55.4 months (95% CI: 25.1–85.7) compared to 5.0 months (95% CI: 0–32.3) for those without radiation (p = 0.049). The Kaplan–Meier curve in Supplementary Fig. 1(B) illustrates this difference, highlighting the potential benefit of adjuvant radiation in improving PFS for these higher-grade PMs.

To further investigate prognostic factors regarding PFS in PM, we analyzed 184 of those 1010 patients, who share the following common available covariates: Age, sex, EoR (subtotal resection, gross total resection), and neurofibromatosis status (NF2 or sporadic). We performed uni- and multivariable Cox regression analyses of all factors potentially predicting PFS among these patients separately for each WHO grade to determine independent risk factors of patients sharing common available covariates (see Supplementary Tables 2, 3, 4). In WHO grade 1 meningioma, univariable analysis revealed that STR significantly increased the risk of progression compared to GTR (HR = 7.86, 95% CI: 3.30–18.75, p = 0.001), and the absence of adjuvant radiation was similarly associated with poorer PFS (HR = 5.50, 95% CI: 2.30–13.16, p = 0.001). Multivariable analysis confirmed only STR as an independent risk factor (HR = 4.51, 95% CI: 1.67–12.25, p = 0.003) (see Supplementary Fig. 2). These findings emphasize the importance of achieving GTR in pediatric WHO grade 1 meningiomas. The analyses of PFS in WHO grade 2 and 3 meningiomas were concluded after univariable Cox regression due to the statistical significance of only one variable in each group. For WHO grade 2 meningiomas, STR was the sole significant predictor of worse PFS (HR = 3.57, 95% CI: 1.43–8.93, p = 0.007)​. In WHO grade 3 meningiomas, the lack of adjuvant radiation was the only variable associated with significantly poorer PFS (HR = 3.98, 95% CI: 1.28–12.36, p = 0.02)​. These results underscore the importance of gross total resection in WHO grade 2 meningiomas and adjuvant radiation in WHO grade 3 meningiomas for improving PFS.

Bias and quality evaluation

The NIH Quality Assessment Tool revealed most studies had clear objectives, defined populations, adequate recruitment, and measured exposures before outcomes with sufficient follow-up. Limitations included missing sample size justification, unblinded assessors, and limited exposure measure validation, introducing potential bias. Despite this, most studies addressed confounding variables, resulting in a moderate but manageable risk of bias and reliable findings on PMs. The scores for all 14 NIH-QAT domains are summarized in Supplementary Fig. 3.