Introduction

Meningioma accounts for at least one-third of all primary intracranial neoplasms [1]. Overall incidence, about 6–7/100,000 persons [2], increases with age and prevalence estimations reach 2% of the population, acknowledging that many meningiomas are incidentally discovered thanks to neuroimaging [3, 4] or at autopsy [5]. The great majority of meningiomas are benign (grade I), yet with the potential for recurrence, even in the event of complete resection [6].

Atypical (grade II) and anaplastic (grade III or malignant) variants account for a variable percentage of all meningiomas depending on diagnostic criteria [7, 8]. The latest 2016 update of the World Health Organization (WHO) classification of central nervous system tumors maintained this 3-grade stratification of meningiomas (Table 1) based on specific histologic features [9]. Under previous classifications, atypical and anaplastic meningiomas comprised only 5–7% and 3–5% of all meningiomas, respectively [10, 11]. However, the systematic adoption of such newer criteria has increased the proportion of referred atypical meningiomas up to 20–35% [12, 13].

Table 1 Definition of WHO grade II and III meningiomas according to the 2016 update of the WHO classification of brain tumors
Full size table

It is widely recognized that atypical and anaplastic meningiomas exhibit at least a 7- to eightfold increased risk of recurrence, and a marked increased risk of mortality, compared to benign meningiomas [13, 14]. As a general principle, maximum safe resection is the recommended primary treatment modality for all meningiomas [15]. However, at least one quarter to one-third of all meningiomas will eventually require adjuvant therapy, either after the first surgery or at the moment of recurrence [12, 13]. Therefore, it is important to confirm whether radiotherapy or other systemic therapies are clinically effective, especially among higher-grade tumors, and to identify what subset of patients benefit most.

Age, grade, and extent of resection remain the most relevant prognostic factors related to meningioma patient’s survival [16, 17]. Currently, there is relative consensus that all anaplastic and subtotally resected (STR) atypical meningiomas should undergo adjuvant radiotherapy [18,19,20]. However, controversy remains on the optimal treatment of atypical meningiomas undergoing gross total resection (GTR) [12, 13, 19, 21]. At present, GTR is defined by most authors as macroscopically complete tumor removal with or without its dural attachment or underlying bone, that is, Simpson’s grades I, II, and III [22].

Previous systematic reviews have synthesized the evidence regarding the impact of radiotherapy on atypical and malignant meningiomas, showing and discussing some conflicting results among studies. Kaur et al. [21] and Hasan et al. [23] thoroughly reviewed grade II and III meningioma case series published before 2014. More recently, Pereira et al. [24] have summarized the effectiveness of radiotherapy in atypical meningiomas according to studies published up to October 2017. Strikingly, several dozens of additional case series and a few non-randomized trials have been published since then. An updated systematic review on the impact of adjuvant radiotherapy in atypical and anaplastic meningioma patients according to published studies from 2017 to date is presented.

Methods

Search strategy

A systematic review according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analysis) guidelines was performed [25]. Studies were searched and retrieved from PubMed, SCOPUS, Cochrane Controlled of Register Trials, EMBASE, and clinicaltrials.gov databases. The search terms were as follows: meningioma, atypical, anaplastic, malignant, grade II, grade III, radiotherapy, radiation therapy, adjuvant, high grade, irradiation, radiation, resection, gross tumor resection, gross total resection, subtotal resection, postoperative, overall survival, progression-free survival, local control, progression, and recurrence. The search strategy included combinations of these key words with appropriate Boolean operators. References lists from relevant papers and related articles were also screened to make the search as comprehensive as possible. Papers published from 2017 and thereafter were gathered, and no language restriction was applied. The search was performed on April 10, 2020, by the two authors (PDL and ECG) independently, and discrepancies were solved by consensus.

Inclusion and exclusion criteria

All types of study (either observational or intervention studies) published within the last 4 years, that included a minimum of ten patients (either atypical and/or anaplastic meningiomas), were included. Studies on exclusively grade II meningiomas and studies on grade II–III and grade III meningiomas were analyzed, separately. Studies not strictly related to adjuvant radiotherapy in atypical or anaplastic meningioma patients were excluded. Review articles on radiotherapy techniques, chemotherapy or other systemic therapies, histologic features of atypical/malignant meningiomas, and studies in which the clinical course and outcome of patients [local control (LC), progression-free survival (PFS) or overall survival (OS)] were not available were also excluded (Fig. 1).

Fig. 1
figure 1

PRISMA flow chart

Full size image

Data extraction

Information extracted from studies included first author, year of publication, study design, number of participants, histologic grade, number of irradiated patients, median follow-up, extent of resection, outcome regarding LC, PFS, or OS, limitations of the study, and main conclusions. Data on adverse effects related to radiotherapy was beyond the scope of this review and was not specifically addressed. Due to the paucity of prospective trials and the great heterogeneity among studies, a quantitative analysis on the effectiveness of adjuvant radiotherapy was not attempted.

Results

Study selection and main characteristics

Out of the initially screened 3726 studies, 90 were full-text reviewed for eligibility. After applying inclusion and exclusion criteria, 43 studies were finally included in the qualitative synthesis, 22 studies on exclusively atypical meningiomas [12, 13, 20, 26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44], and 21 studies on anaplastic or atypical/anaplastic series [19, 45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64]. The main characteristics of the studies are summarized in Table 2 for atypical meningiomas, and Table 3 for anaplastic and atypical/anaplastic meningiomas.

Table 2 Summary of clinical characteristics of studies on WHO grade II meningiomas treated with adjuvant radiotherapy
Full size table
Table 3 Summary of clinical characteristics of studies on WHO grade II and III and exclusively grade III meningiomas treated with adjuvant radiotherapy
Full size table

Regarding atypical meningiomas, 20 studies were retrospective cohorts (three multi-center studies) and two were prospective trials. The median number of participants per study was 140, and the median percentage of irradiated patients per cohort was 24.8%. The median follow-up period of studies was 52 months. The RTOG 0539, a non-randomized phase II trial [34] testing the efficacy of postoperative radiotherapy (either intensity-modulated radiotherapy (IMRT) or 3D conformal, 54 Gy in 30 fractions) among the so-called Group 2 or intermediate-risk meningiomas (atypical undergoing GTR and recurrent grade I with any extent of resection), provided results in terms of LC and PFS/OS at 3 years, which were compared with historical controls. The EORTC 22,042–26,042 trial [32], a multi-cohort non-randomized phase II trial, tested the efficacy of postoperative high-dose radiotherapy on PFS at 3 years among atypical and anaplastic meningiomas separately. The majority of patients enrolled were grade II meningiomas undergoing GTR. The objective of the study was to demonstrate that 3-year PFS among completely resected and irradiated grade II patients was higher than 70%.

In the group of studies including grade II and III or exclusively grade III meningiomas, 20 studies were retrospective case series (one multi-center study and five retrospective institutional database reviews) and one non-randomized prospective trial. The median number of participants per study was 60, and the median percentage of irradiated patients per cohort was 72.4%. The reported median follow-up period of studies was 48 months. The RTOG 0539 trial [46] conductors, which recently published the initial analysis of its high-risk cohort, recruited new or recurrent grade III meningiomas of any resection extent, recurrent grade II of any resection extent, and new grade II meningiomas after subtotal resection. Patients received IMRT (60 Gy high dose or 54 Gy low dose in 30 fractions) and PFS was assessed at 3 years.

Effectiveness of radiotherapy in grade II and grade III meningiomas

Among studies including exclusively grade II meningiomas, adjuvant postoperative radiotherapy was generally found beneficial for subtotally resected meningiomas in terms of LC, PFS, and OS [12, 13, 28, 30, 43]. In the case of GTR, the effect of radiotherapy was controversial, with few studies suggesting a clear benefit [29, 37, 39], the majority not confirming a clear influence on PFS or OS [13, 26, 27, 30, 33, 35, 36, 41, 43, 44], and even others suggesting a negative impact [40]. According to several studies, close observation of grade II meningiomas was considered a feasible option if they were completely resected [13, 27, 43, 44]. In general, a greater than 70% PFS rate at 3 years was expected following radiotherapy of grade II and completely resected meningiomas [32, 37]. The impact of radiotherapy on OS was not that clear, although some retrospective studies suggested a moderate or marginal beneficial effect [29, 30, 36, 39, 42, 43].

In general, authors do not recommend adjuvant radiotherapy as mandatory for grade II and GTR patients. In fact, close observation is a feasible and reasonable option, although it has been reported that salvage radiotherapy after re-intervention may not be as effective as early adjuvant radiation [34]. Patients considered of higher risk (advanced age, severe co-morbidities, tumors near critical structures, and lesions requiring large irradiation fields) are good candidates for observation. Likewise, lower-risk patients concerned about radiation-related toxicity can also be observed. Nevertheless, there seems to be relative consensus that local control is enhanced by postoperative adjuvant radiation in grade II meningiomas irrespective of resection extent [12, 34].

Regarding studies including grade II and III (15 studies) or exclusively grade III (6 studies) meningiomas, as expected, adjuvant radiotherapy was given to a larger proportion (near threefold) of patients compared to cohorts of solely grade II meningiomas. Among grade III meningioma patients, there was consensus on the beneficial effect of radiation on LC and PFS, and maybe on OS, irrespective of extent of resection [19, 45,46,47,48,49, 52, 54, 58, 59, 62,63,64]. Yet, adjuvant radiotherapy was found to have unclear or no impact on OS in some studies [50, 51, 56, 61]. Rhabdoid histology was reported to achieve better outcome after radiation in a multivariate analysis [58]. Among studies including both grade II and III meningiomas, the majority confirmed the efficacy of adjuvant radiation on LC, PFS, and likely OS, especially among subtotally resected tumors [19, 45,46,47,48, 52,53,54]. Less efficacy was reported regarding recurrent tumors, especially those located in the falx compared to convexity or skull base meningiomas [63]. Special radiation modalities, like proton therapy, brachytherapy, and boron neutron capture therapy, showed promising results, usually performed as re-irradiation or at the time of recurrence. However, these techniques reported non-negligible rates of radionecrosis and wound morbidity [55, 57, 60].

In general, it is widely accepted that grade III meningiomas need upfront postoperative radiotherapy regardless of the extent of resection. The majority of cohorts including both grades also showed benefit from adjuvant radiotherapy in terms of LC and PFS (although not so clear over OS), especially if performed immediately after the first operation. The impact of radiation at the time of recurrence seems to be less favorable and likely conditioned by the possibility of further resection. Figure 2 depicts a scheme of the current recommendations regarding grade II and III meningioma management according to the extent of resection.

Fig. 2
figure 2

Recommended indications of radiotherapy in atypical and anaplastic meningiomas

Full size image

Discussion

Since 2017, numerous grade II and III meningioma case series have been published, adding evidence toward the effectiveness of adjuvant radiation therapy in non-benign meningioma. Unfortunately, the great majority of studies were single-center retrospective cohort reviews. The synthesis of this systematic review confirms the usage and efficacy of adjuvant radiotherapy and underscores the need for prospective comparative trials. Atypical meningiomas can be cured by surgery only in 16–18% of the cases [24], and up to 70% recur within the next few months after surgery [65, 66]. Compared to benign meningiomas, atypical lesions carry a poorer overall prognosis, with 5- and 10-year PFS of 38–59% and 19–22%, respectively [12, 67], and a significantly reduced OS [65].

Application of adjuvant radiotherapy exhibits wide variability partly attributable to a lack of comparative prospective trials. Surveys among neurosurgeons indicate that a majority do not recommend routine adjuvant radiotherapy after GTR [18, 68]. The controversy about adjuvant radiotherapy in operated non-benign meningioma emanates from the conflicting evidence provided by many heterogeneous studies. Such studies commonly included a limited number of patients, the definition of atypical/anaplastic tumors evolved over time, many studies analyzed grade II and III lesions together, the definition of GTR also changed over time, and a certain selection bias (tendency to preferentially irradiate patients with poorer expected prognosis) was present in the majority of retrospective series [13, 69]. Likewise, radiation-related issues were not systematically reported in studies, like technique, dosage, scheme, field, target volume, and margins. In the recent review by Bagshaw et al. [12], adjuvant radiotherapy was offered to all atypical tumors irrespective of extent of resection upon recognizing that salvage treatments, either re-intervention or re-irradiation, do not seem to impact on survival significantly. They recommend relatively high doses of fractionated radiotherapy, at 59.4 Gy and 54 Gy, for STR and GTR, respectively, and single-dose 18–20 Gy if stereotactic radiosurgery is used.

In 2016, the European Association of Neuro-Oncology (EANO) issued guidelines for the diagnosis and treatment of meningiomas [70]. These were in concordance with previous recommendations, in which grade II tumors undergoing GTR could be observed, whereas grade II with STR and grade III should receive fractionated radiotherapy (54–60 Gy). For both grade II and III meningiomas, fractionated radiotherapy is generally preferred over stereotactic radiosurgery, which in turn might be better suited for primary treatment of small benign meningiomas or for small and deep-seated postoperative remnants. Surveillance of non-benign meningiomas require shorter control intervals compared to regular grade I lesions, therefore, repeating MRI every 3–6 months instead of annually. EANO guidelines adopted the definition of GTR as Simpson’s grades I, II, and III. Given the quality of studies supporting the use of adjuvant radiotherapy, all recommendations provided in the EANO guidelines statements were based on evidence level B, C, and good practice points.

In this view, the main issue to elucidate when confronting a high-grade meningioma is whether adjuvant therapy (primarily radiotherapy) should be proposed upfront or reserved for recurrence. Given that the extent of resection is a key prognostic factor, efforts should be made to achieve resections as complete as possible in the first surgery. Yet, it is common practice to observe grade I subtotally resected meningiomas, given that many residues remain unchanged over long periods of time, if not for life. In contrast, subtotally resected high-grade variants, generally due to involvement of vital structures or deeply located, tend to grow locally over time at a much higher speed, making further resections more complicated. These cases are commonly offered adjuvant therapy in the belief that local control is improved, and subsequently PFS and OS.

However, studies have yet failed to demonstrate a sound impact of adjuvant radiotherapy on OS likely attributable to the lack of randomized controlled trials on the matter [12, 13, 70]. It is striking that no such trials had been conducted so far given the enormous amount of eligible patients available worldwide and the conflicting evidence regarding its effectiveness among subgroups of meningiomas, stratified by prognostic factors, like age, grade, extent of resection, location, and histopathological subtypes. The latest WHO classification of meningiomas has transformed many previously considered grade I tumors into higher grades, merely upon histopathologic criteria [9]. In addition, historical surgical aggressiveness directed to deep-seated meningiomas, like cavernous sinus or petroclival region, has been revised given the intolerable surgical morbidity (especially cranial neuropathies) reported in some modern series [71,72,73]. These two factors have significantly increased the number of patients eligible for adjuvant radiation.

Only two prospective non-randomized trials have studied the effect of radiotherapy on non-benign meningiomas. The RTOG 0539 phase II trial [37] analyzed cohorts of intermediate risk (new grade II undergoing GTR or recurrent grade I) and high risk (grade III or new grade II undergoing STR or recurrent grade II). All patients received 54 Gy as adjuvant therapy. They reported a remarkable 3-year PFS of 93.8%. Interestingly, both subgroups within the intermediate risk achieved statistically similar PFS rates. Among high-risk patients, 3-year PFS, LC, and OS were 59.2%, 68.9%, and 78.6%, respectively. The EORTC 22,042–26,042 phase II trial included grade II and III meningiomas. Atypical lesions undergoing GTR and adjuvant radiotherapy (60 Gy) achieved a 3-year PFS and OS of 88.7% and 98.2%, respectively. However, none of these studies compared adjuvant radiotherapy against observation nor randomized patients to treatment and control arms.

The currently ongoing ROAM/EORTC 1308 phase III multi-center trial [69] was designed to randomize GTR atypical meningioma patients to either radiation (60 Gy in 30 fractions) or observation. This trial was intended to determine whether early adjuvant radiotherapy reduced the risk of tumor recurrence following GTR of atypical meningiomas. Embedded within this trial and pre-randomization, a qualitative study was performed [74], in which specific researchers examined how information about the trial is exchanged between physicians and patients during recruitment interviews and its influence on consent obtention. In a sub-sample of patients, they compared discussions during recruitment consultations with clinicians’ and patients’ interpretations of such consultations. This has been shown to be effective to improve patient and doctor acceptability and maximize recruitment [75]. In fact, this analysis identified several challenges that practitioners face in conveying equipoise, addressing difficulties with communication, and exploring patient treatment preferences, with potential impact on improving informed consent and recruitment. A second trial is also underway (phase III NRG-BN-003 trial or NCT 03,180,268), comparing radiation (59.4 Gy) versus observation in GTR patients [13]. Table 4 shows the main characteristics of both trials and the estimated completion date. These trials address a simple but important question about meningioma management, and their preliminary results, likely available in a few years from now, have the potential for changing the current neuro-oncologic practice worldwide. Other clinical trials have been recently activated or are currently recruiting patients in order to test the efficacy of carbon ion radiotherapy in atypical meningioma (NCT01166321), proton dose escalation in atypical and anaplastic meningiomas (NCT02978677), and the efficacy of postoperative radiotherapy for atypical meningiomas without venous sinus invasion after GTR (NCT04127760).

Table 4 Main characteristics of the ongoing ROAM/EORTC 1308 and NRG-BN-003 trials on the effect of adjuvant radiotherapy following total surgical removal of atypical meningiomas
Full size table

Radiation therapy modalities have not been compared between them in well-designed studies as to provide evidence of one treatment modality over the others. In general, higher doses (at least 54 Gy) seem to provide better overall outcomes [70]. According to the review by Hwang et al. [76], PFS was significantly higher when 60 Gy were given compared to lower doses, and when combined proton and photon irradiation was used [77]. Conventional fractionated external beam radiotherapy (typically using 1.8–2.0 Gy/fraction over 5- to 7-week course) benefits from the conformality of radiation dose delivery provided by intensity-modulated RT (IMRT) and volumetric arch therapy [78]. For grade II and III meningiomas, doses ranging between 60 and 66 Gy are typically used [21, 79]. Applied to skull base meningiomas, stereotactic radiotherapy with IMRT provides 3- and 5-year local control rates over 93%, although this technique is generally viewed as a second-tier option once surgery and/or conventional RT have failed [80]. Single fraction stereotactic radiosurgery (SRS) is an effective option for lesions < 10 cc, a maximum diameter less than 3–4 cm, and sufficient distance from critical structures [76]. According to Kano et al. [81], a marginal dose of 16–20 Gy associates a 5-year local control rate of 60–75%. Interestingly, the inclusion of the dural tail in the treatment volume has proven beneficial in benign meningioma but has not been yet studied in high-grade meningioma [76]. Likewise, hypofractionated stereotactic radiotherapy (five fractions or less), indicated for larger lesions, have yielded good results in benign meningioma but has not been studied in aggressive meningiomas [82, 83]. Interstitial brachytherapy has been reported to be an effective adjunct to surgery and external beam radiotherapy, especially for malignant, recurrent, and large meningiomas, yet associating over 25% rate of wound complications and radiation necrosis, many of which needed further surgery [84]. Salvage brachytherapy with I-125 or Cs-131 seeds, occasionally used in recurrent grade II–III meningiomas already re-operated and irradiated, has yielded a 40% need for reoperation due to wound complications [57]. Finally, heavy particle therapy with carbon ion radiotherapy offers the advantages of highly localized deposition of energy as in proton therapy, reduced cell cycle-dependent radiosensitivity, and increased efficacy for cancer stem-like cells [76]. The NCT01166321 phase II open-label trial (expected to be completed by December 2020) is currently recruiting patients with atypical meningiomas undergoing partial resection (Simpson 4 and 5) treated with carbon ion boost in combination photon radiotherapy.

As already standardized for glioma [9] or ependymoma [84], the eventual upcoming genetic- and molecular-based classification of meningiomas will likely introduce a new way of stratifying risk in meningioma, and may provide a surrogate prognostic factor of survival and recurrence.

Advances on the immunogenetic landscape of meningioma have been recently reviewed [85]. Using unsupervised clustering of DNA methylation data, Sahm et al. [86] were able to identify two major epigenetic patterns across the three grades of meningioma: group A, including four methylation classes (three benign and one intermediate), and group B which included two additional classes (one intermediate and one malignant). Interestingly, the majority of NF2 (Neurofibromatosis-2) gene-mutated tumors gathered in group A, and most TERT mutations in group B. This 6-methylation class subdivision predicted PFS better than the current WHO 15-histology variants and 3-grade classification [86]. Additionally, a meningioma recurrence score has been proposed, based on methylation status, extent of resection, and WHO grade which can be used in the clinical setting to inform choices for follow-up and to decide about adjuvant therapy [87]. However, meningioma is characterized by a relatively low mutational burden compared to other brain tumors; therefore, identification of prognostic biomarkers has yet provided modest clinical impact for decision making. In general, TERT promoter and BAP1 mutations are linked to poorer prognosis [85]. This molecular subgrouping may eventually simplify meningioma stratification and will necessarily need to be acknowledged when designing future trials.

Among unresectable meningiomas, in which tumor growth is confirmed, especially if suspicious of being high grade, upfront radiation or systemic therapy is mandatory [70]. A detailed description on the effectiveness of chemotherapy and other systemic therapies in non-benign meningioma is beyond the scope of this review and can be found elsewhere [88,89,90]. At the time of recurrence, if feasible, re-intervention is the first-choice irrespective of grade or previous extent of resection [12, 13, 24, 70]. Irradiation is the preferred modality when re-intervention is deemed not possible and for small deep tumor remnants. Salvage re-irradiation can also be an option. Studies show that PFS and OS are improved the longer the interval between prior radiotherapy and re-irradiation [13]. Although metastasis is rare and can occur throughout the natural course of any type of meningioma, including grade I lesions [91], it is much more common among higher-grade tumors [13, 70]. Anecdotally, the recent paper by Golub et al. [92] reports on a possible abscopal effect after IMRT treatment of an intracranial meningioma, with radiographically significant response of another untreated second intracranial lesion, distant to the radiation field, also suggestive of meningioma.

Conclusions

According to the recently updated WHO classification of central nervous system tumors, grade II and III meningiomas comprise a larger proportion of all meningiomas. Maximum safe resection including the dural attachment is the recommended first treatment option for all types of meningiomas. Atypical and anaplastic meningiomas carry a significantly increased recurrence rate compared to benign meningiomas, which negatively impacts survival. Adjuvant radiation is currently advised for subtotally resected grade II and for all grade III meningiomas. In the case of grade II meningiomas undergoing gross total resection, close radiologic and clinical observation is feasible. The potential benefit of adjuvant radiation in this subset of patients is currently being tested in ongoing clinical trials.