De novo and secondary anaplastic meningiomas: natural history, prognosis, and the TERT promoter

See the article by Peyre et al., pp. 1113–1121.

Anaplastic meningiomas (AMs) are World Health Organization (WHO) grade III primary dural cancers.¹ AMs make up only 1% to 3% of all meningiomas. Three hundred and forty-nine malignant meningiomas, including WHO grade II (atypical) meningiomas and AMs, were diagnosed yearly in the United States in 2010–2014.² AMs arise de novo or lower-grade meningiomas de-differentiate and recur as secondary AMs. Determining the prognosis for individual patients with an AM can be difficult. Standard treatment following surgical resection includes adjuvant radiotherapy, although utilization of radiotherapy in community practice may occur in only 40% to 60% of patients.^3,4 Single institution series using varying radiation techniques in adjuvant and recurrent settings report progression-free survival (PFS) at 2 or 3 years between 0 and 80%.⁵ The 5-year overall survival (OS) rate between 2004 and 2012 in the National Cancer Center Database was 41.4%.³ In a single institution series of 34 AMs treated with surgical resection followed by adjuvant radiotherapy, the 2-year recurrence-free survival rate was 80% and the 5- and 10-year OS rates were 61% and 40%, respectively.⁶ The variability of the retrospective data and lack of prospective survival outcomes make it difficult to set benchmarks when comparing studies or setting expectations for our patients.

In this issue, Peyre et al report on 57 cases of AMs, excluding papillary and rhabdoid types, from the French Brain Tumor and a departmental neuropathology database.⁷ The cohort includes 28 de novo and 29 secondary tumors. De novo AMs were classified as anaplastic based upon mitotic index with or without anaplasia or based on anaplasia alone. Seventy-five percent had a gross total resection and 88% received adjuvant radiotherapy. As reported in 2 smaller studies,^8,9 improved survival outcomes were seen in de novo (OS 3.1 y) compared with secondary AMs (OS 2.1 y). Seven de novo AMs with anaplastic features alone had a median OS of 6.2 years compared with an OS of 2.6 years in 21 de novo AMs with a high mitotic count. Interesting clinical observations were made among secondary tumors, including wide variability of time to anaplastic transformation, but OS for secondary tumors did not differ after anaplastic transformation.

Telomerase reverse transcriptase promoter (TERTp) mutations are common across different cancer types, including pediatric and adult primary CNS neoplasms.¹⁰ TERTp mutations were initially described in a small number of higher-grade and recurrent meningiomas.^10,11 In a study by Sahm et al, TERTp mutations were found in 16 of 252 (6.4%) meningioma cases and in 9 of 45 (20%) AMs. In a separate study by Juratli et al, TERTp mutations were identified in 6 of 26 patients (23%); 5 of these tumors were AMs. In both studies TERTp mutations were associated with a reduced PFS.^12,13 TERTp mutation appears to be acquired in many cases at recurrence, and expression of TERTp mutations is spatially heterogeneous throughout individual tumor specimens.¹³ The study by Peyre et al adds to the TERT story in AMs. Five of 28 (18%) de novo AMs and 3 of 29 (10%) secondary AMs had TERTp mutations. There was no difference in OS in TERTp mutated versus wild type cases, although PFS in secondary AMs with TERTp mutations was significantly shorter.⁷

Methylation of TERTp at cytosine-phosphate-guanine sites upstream of the transcription start site results in an overexpression of TERT in some pediatric primary CNS neoplasms.¹⁴ The resulting upregulation of TERT expression observed with TERTp methylation is opposite to the epigenetic silencing of gene expression observed with promoter methylation in genes such as O⁶-methylguanine methyltransferase in glioblastoma. The mechanism by which TERTp methylation promotes TERT expression is unknown, but several mechanisms have been suggested.¹⁵ Furtjes et al first reported on TERTp methylation in meningiomas and found that 3 of 16 benign, 9 of 21 atypical, and 15 of 25 AMs had TERTp methylation. TERTp methylation was associated with TERT expression in 80% of cases. TERTp methylation was associated with increasing meningioma grade and decreased PFS and OS.¹⁶ The current study assessed TERTp hypermethylation in 14 de novo and 19 secondary AMs. The results are striking: 11 of the 14 de novo and 19 of 19 secondary AMs tested had TERTp methylation.⁷ Taking into account cases with TERTp mutation, TERTp methylation, or both, 12 of 28 de novo and 19 to 22 of 29 secondary AMs would be expected to have increased TERT expression. Acknowledging TERT expression was not assessed, 54% or more of AM cases in this study cohort would be expected to have upregulation of TERT expression.

Increased TERT expression appears to be an important mediator of meningioma progression from a benign to an atypical to an anaplastic tumor.^12,13,16 Accumulating the results of studies to date, the majority of AMs have either TERTp methylation or TERTp mutation (some with both) and a small proportion may have TERT gene fusions¹³ or other undiscovered mechanisms of upregulation of TERT expression. Assuming upregulation of TERT expression leads to a more aggressive clinical course, determining the prognostic significance of TERTp mutations and TERTp methylation in AMs will require concurrent determination of TERT expression and comparing OS outcomes in cases with and without upregulation of TERT expression.

The study by Peyre et al gives us additional prognostic data to consider when guiding the multidisciplinary management of our patients. In line with recognition of a benign subtype of rhabdoid meningiomas without other malignant features as discussed in the 2016 WHO classification,¹⁷ there was a subset of AMs without evidence of cellular proliferation with an improved prognosis in the current study. Further correlation of WHO grade, mutational profile, and DNA methylation profile will help reconcile the improved prognosis observed in de novo AMs and the overall variability in survival outcomes. There is already evidence supporting this assertion within recent publications. In a study of DNA methylation-based classification of meningiomas, 6 rhabdoid and papillary meningiomas clustered with either benign or atypical meningiomas.¹⁸ The study by Peyre et al strongly suggests that the majority of AMs have acquired upregulation of TERT expression by TERTp methylation or mutation, and confirms the poor prognosis for most secondary and de novo AMs. Identifying patients with higher-risk DNA methylation profiles and TERTp mutation or methylation may provide additional support for administering adjuvant therapies, particularly radiotherapy, when the tumor is WHO grade I or II, as the retrospective data to date suggest that maximal resection followed by adjuvant radiotherapy has limited efficacy once a meningioma meets full anaplastic criteria. WHO Grade I and II meningiomas with TERT upregulation may be a population in which to test TERT-targeted therapeutic strategies.

We need to improve outcomes for AM patients. Results from systemic therapeutic trials in AMs are difficult to interpret and compare as expertly outlined by the Response Assessment in Neuro-Oncology group.¹⁹ The results of 2 phase II clinical trials, RTOG 0539 and EORTC 22042–22062, will provide much needed prospective survival outcome data in uniformly treated AM patients.

Disclaimer

The views expressed are those of the author and do not reflect the official policy of Walter Reed National Military Medical Center, the Department of the Army, the Department of Defense, or the US government.