Checkpoint blockade in recurrent meningiomas: Lessons for future management

Meningioma is the most common primary CNS tumor in adults. Although most meningiomas can be benign and cured by surgical resection, there is a subset that can cause significant morbidity and mortality. While we normally associate the aggressive nature of these tumors with their grades, we have observed even refractory grade I meningiomas. Despite surgery and radiation being employed as first- or second-line therapies, meningiomas can still metastasize or progress locally, and treatments for recurrent tumors are limited. As a result, assessing the efficacy of therapies from our current armamentarium to treat meningiomas is an area of significant need.

Bi et al present their prospective phase II study where 25 patients with recurrent grade 2/3 meningioma were treated with nivolumab to block programmed death 1 (PD-1).¹ In addition to following the natural history, they performed correlative studies such as tumor mutational burden (TMB) via sequencing and immunophenotyping of the tumors via immunohistochemistry (IHC). Although the PFS-6 did not significantly improve among treated patients, a subset of patients characterized by high TMB benefited from treatment with nivolumab, leading to greater infiltration and proliferation of immune cells in the tumor. None of the patients in the trial experienced unexpected adverse events following treatment with nivolumab.

Despite the lack of efficacy observed with nivolumab and the small sample size, the study highlights the need to assess TMB as a genetic criterion to determine appropriate candidates for immunotherapy among meningioma patients. Although the majority of the patients (13/15) whose tissue samples were used for exome sequencing had low TMB, the two long-term survivors had elevated TMB of more than 10/Mb. While prior studies in other tumors have shown that TMB can be positively correlated with response to immune checkpoint blockade, it remains to be seen if high TMB levels can predict favorable response rate in patients with recurrent meningioma on immunotherapy.² Although radiation and alkylating chemotherapy can increase TMB in many tumors, chemotherapy has not been effective in treating recurrent meningiomas and TMB was low in all but 2 patients despite getting at least 1 (100% of patients) or 2 radiation courses (72% of patients).^1,3 Therefore, assessment of combinatorial regimens integrating immunotherapy with ongoing radiation therapy might provide the critical window to allow robust activation of naïve T cells responding to neo-antigen presentation while also reinvigorating preexisting primed anti-tumor T cells. It must be noted that high TMB does not uniformly predict a positive response to immunotherapy across all tumors, as has been observed in gliomas where high TMB did not promote response to PD-1 blockade.⁴

The greater efficacy of nivolumab in the 2 patients with high TMB levels also highlights the need to understand the role of antigenic density and immunogenicity in predicting response to immunotherapy in recurrent meningioma. High TMB can lead to greater quantity of neo-antigens, leading to a more clonal T cell receptor (TCR) repertoire which has been associated with favorable response to PD-1 blockade in other tumors.^2,5 But not all mutations are equally immunogenic, with frameshift mutations being more immunogenic than point mutations in many cancers.⁶ Past studies have revealed meningiomas to express certain immunogenic epitopes as well, known as meningioma-expressed antigens (MGEAs).⁷ Understanding how response to nivolumab is intertwined with the quantity or quality of TCR clones would not only facilitate selection criteria for patients to be on nivolumab, but also pave the path for the incorporation of novel tools such as CAR-T therapy against meningiomas.

In addition to TMB, there might be other yet unknown factors that might alter response to immunotherapy in meningioma patients. Although the absence of the blood-brain barrier in meningioma makes them more accessible to drugs compared to other intracranial tumors such as glioblastoma (GBM), astrocytomas, or oligodendrogliomas, their location in the immune-privileged CNS might still play a role in blunting immune response in an immune-suppressive milieu. Therefore, it is imperative to understand immune interactions that are unique to meningiomas as well as shared with other CNS tumors. For example, in rodents, the meninges are known to harbor immune cells such as DCs (dendritic cells) at a level higher than the brain parenchyma, which might provide avenues for lymphocyte-DC interactions at the tumor site.⁸ Furthermore, T cells are recruited from systemic circulation to cerebrospinal fluid (CSF) through interaction with adhesion molecules found abundantly in vasculature of the choroid plexus and the meninges, compared to the parenchyma.⁹ In addition to quantifying infiltration of distinct immune populations separately, examining the cross-talk of myeloid-lymphoid arms of the immune system in meningiomas could provide further insight into the likelihood of success of immunotherapy.

To determine combinatorial regimens with a greater chance of success in high-grade meningiomas, it is important to assess the upregulation of alternative checkpoints in response to anti-PD-1. Understanding the nature of adaptive resistance in non-responders can provide additional candidates for combination therapies, such as co-blockade of multiple checkpoints (such as LAG-3, TIGIT, or TIM-3 in addition to PD-1) or agonism of co-stimulatory receptors on T cell surface (such as 4-1BB). Bi et al also report a higher abundance of macrophages in the tumors of meningioma patients, compared to that of CD4 or CD8 T cells. Most macrophages that infiltrate meningiomas are known to have tumor-promoting phenotype and play an important role in increasing tumor growth and recurrence.¹⁰ Similar to other CNS tumors such as GBM, pro-tumor macrophages, and myeloid-derived suppressor cells create an immunosuppressive environment in high-grade meningiomas, limiting the effector function of tumor-infiltrating T cells. Therefore, evaluation of therapies targeting tumor-promoting myeloid populations such as CCR2- or CSF-1R inhibitors might further synergize with lymphoid-targeted therapies in high-grade meningiomas.

Although the findings by Bi et al show no efficacy of nivolumab in patients with meningiomas, it is encouraging to see that the two long-term survivors had a high TMB. Future studies are needed to assess if elevated TMB predicts enhanced response to immunotherapy. For patients with low TMB, incorporating a multi-pronged approach could synergize with checkpoint blockade, such as increasing tumor antigen-spilling, enhancing cross-priming of T cells by antigen presenters, and reducing myeloid-predominant immunosuppression that characterizes high-grade meningiomas.