Commentary: The Ki-67 Proliferation Index as a Marker of Time to Recurrence in Intracranial Meningioma

In the December 2020 issue of Neurosurgery, Mirian and colleagues¹ evaluated the Ki-67 proliferation index as a marker for recurrence and time to recurrence in meningioma by evaluating a prospective cohort with long-term follow-up (10 yr). Meningiomas, the most common central nervous system (CNS) primary neoplasia (38.3% of CNS tumors),² have an estimated incidence of 8.81 cases per 100 000 people per year. The incidence of meningioma increases with aging, and it is more common in females and African Americans.²

The 2016 World Health Organization (WHO) CNS classification recognizes 13 meningioma histologic subtypes with WHO grades ranging from 1 to 3.³ Nine histologic subtypes (meningothelial, fibrous, transitional, psammomatous, angiomatous, microcystic, secretory, lymphoplasmacyte-rich, and metaplastic) are considered WHO grade 1 unless features of aggressiveness are identified. Two histologic variants (clear cell and chordoid) are considered WHO grade 2, and one variant (papillary) is considered WHO grade 3.³ The rhabdoid variant was previously considered WHO grade 3, but it is now recommended that it should be graded based on the presence or absence of aggressive features, like the other 9 variants.

While most meningiomas (∼80%) are considered benign and classified as WHO grade 1, 17.9% and 1.6% are considered WHO grades 2 and 3, respectively.² Although this classification is somewhat predictive for prognosis, it has several pitfalls, inherent from purely histologic evaluation.^4-6 The diagnosis of atypical meningioma, WHO grade 2, requires ≥4 mitotic figures in 10 microscopic high-power fields (HPFs), and/or brain invasion, and/or 3 of the following 5 features: necrosis, prominent nucleoli, small cell formation, sheeting architecture, or hypercellularity.³ However, brain invasion, as a sole criterion for atypia, has been strongly debated in multiple studies.^7-9 The diagnosis of anaplastic meningioma, WHO grade 3, is based on the presence of ≥20 mitotic figures in 10 HPFs.

The Ki-67 proliferation index is expressed by proliferating cells and it is widely utilized in the neuropathology evaluation of CNS tumors. This index has been evaluated in multiple studies and it has been correlated with recurrence with a tentatively proposed cutoff of 4%.^10-12 The study from Miriam et al¹ presents a continuous, longitudinal, and prospective cohort with long-term follow-up, avoiding most of the biases of prior studies. The authors identified that a 1 percentage point increase in the Ki-67 proliferation index yielded a 12% risk increase of recurrence in multivariable analysis.¹ Despite this finding and the fact that the cumulative incidence of recurrence with a 4% cutoff of Ki-67 was substantially different after 1 yr, the 10-yr cumulative incidence was not significantly different.¹ Demonstrating that Ki-67 was a predictor of time to recurrence, rather than recurrence itself, patients with meningiomas with a Ki-67 < 4% had a recurrence after 4.8 yr compared to 0.60 to 0.75 yr in meningioma patients with higher Ki-67. Even though a larger sample size might show a difference in recurrence cumulative incidence with a 4% cutoff, the study methodology (prospective and consecutive) provides an important addition to the literature. Moreover, there are important clinical implications, as this study suggests patients with a high Ki-67 should be followed up closely, especially in the first years after resection, due to their shorter time to recurrence.

The evaluation of genetic alterations in meningiomas has demonstrated that a high percentage (50%-60%) of meningiomas harbored loss of the NF2 suppressor gene at chromosome 22.¹³ Multiple studies have shown that meningioma location correlates with specific genetic mutations (NF2-lateral and posterior skull bases, parafalcine, and convexity meningioma; KLF4-sphenoid wing and midline skull base; AKT1 and TRAF7-anterior skull base; SMO-olfactory groove; PIK3CA and POLR2A-anterior and middle fossae).^14-17 Additionally, certain genetic mutations (TERTp mutation and CDKN2A/B loss),^4,18-21 chromosomal abnormalities (1p, 10 deletions),⁴ and epigenetic modifications (H3K27me3)²² have been independently associated with poor prognosis and higher grade. Moreover, TERTp mutations have been proposed to be routinely incorporated in the grading of meningiomas.^19,20 Furthermore, DNA methylation-based classifiers have been demonstrated to be superior to histologic grade to predict progression-free survival (PFS) in large meningioma cohorts.⁴ While the advancements in meningioma molecular classification will probably influence tumor grading in the upcoming new WHO classification,⁵ the wide implementation of this classification incorporating methylation and genetic characteristics might be slow, especially in centers with limited resources. Therefore, the utilization of widely available markers like Ki-67, as shown by Miriam and collaborators,¹ is an attractive and easily applicable tool that can raise the alert to close follow-up, making the authors' valiant efforts to evaluate patients prospectively for a decade very commendable. However, it is important to keep in mind that pathologists often evaluate Ki-67 as a visual estimate. This approach suffers from interobserver variability. In contrast, computer-assisted quantitation of the Ki-67 index could provide more consistent results and perhaps more accurate predictions of the risk of recurrence.

Funding

This study did not receive any funding or financial support.

Disclosures

Dr Ballester receives support from the National Cancer Institute of the National Institutes of Health under the Award Number K08CA241651. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.