Hormone therapies in meningioma-where are we?

In Harvey Cushing’s 1938 landmark, “Meningiomas”, he wrote, “we have noted for many years a definite predominance in the incidence of meningiomas in females over males, the proportion being 60–40…still more striking is the fact that in our series tumors of certain loci are largely restricted to women… Meningiomas in not a few instances appear to have been incited by unusual activity, or at least to have shown their first recognizable symptoms soon after a recent pregnancy” [ 10 ]. Both Cushing’s study and contemporary quantifications show a difference between female and male incidence in meningioma, which changes over the lifespan and reaches the greatest difference of approximately 3:1 among those aged 35–55 ( Fig. 1 ) [ 10 , 11 ]. Notably, this age range coincides with perimenopause in females, a time known to be metabolically important for many organ systems, including the brain, and a process resulting in dramatic hormonal changes [ 12 ]. More recently, advances in the understanding of the genetic drivers of meningiomas has helped elucidate heterogeneity related to patient and tumor features, even within the same grade [ 13 , 14 ]. Genomic subgroups associate with tumor location, histology, and outcomes [ 13 , 14 ]. Additionally, female sex is overrepresented in KLF4/TRAF7 and POLR2A genomic subgroups, which are also known to be associated with skull base location [ 13 , 15 , 16 ]. In addition to gonadal steroid hormones, the X-chromosome has been hypothesized as a contributing factor to the observed sex predominance [ 17 ]. A population-based study in Great Britain of people with Turner Syndrome (45,X0) revealed that among all tumors, Turner Syndrome participants were much more likely to develop meningiomas, SIR = 12.0 CI 95% [4.8–24.8], compared to the general population without Turner Syndrome. Females with Turner Syndrome suffer from premature ovarian failure and lack the ability to synthesize gonadal steroid hormones estradiol and progesterone at typical post-pubertal levels and are often supplemented with hormone replacement therapy (HRT). The authors of the study postulate this relationship but report that they did not have access to information regarding patient exposure to HRT [ 18 ].

There is significant heterogeneity between treatment groups included between trials which results in overall poor quality of evidence. Because of this heterogeneity, along with the slow-growing nature of meningiomas, a recent effort has been initiated to strive for an agreed upon set of standard outcomes for studies assessing progression-free survival, referred to as the ‘Core Outcome Sets’ for Meningioma in Clinical Studies (The COSMIC Project). The authors of this project also note the need for global and multi-institutional collaborations for clinical trials in order to increase the participation of patients per study [ 65 ]. A major challenge in meningioma research is the relative lack of adequate meningioma models, with multiple reports of tumors losing features important to tumor identity upon cell culture and even in vivo modeling [ 66 ]. Considering this limitation, the use of radioligands have been beneficial in the non-invasive identification of tumors that express SSTRs, but that same technique has not been applied to other receptors in meningioma on a large scale. In 1997, a small study including 6 patients with meningioma successfully used PET imaging with Fluoroestradiol (F18-FES), a radioligand of estradiol. They demonstrated acceptable radioligand uptake and a correlation between the ligand uptake and immunohistochemical expression of an unspecified estrogen receptor [ 67 ]. As of 2020, F18-FES is FDA approved for metastatic breast cancer. There also exists radioligand progestin 21-[18F]fluorofuranyl-norprogesterone and testosterone derivatives 16β-[18F] fluoro-5α-dihydrotestosterone, both of which are in earlier phases of study in metastatic breast cancer [ 68 , 69 ]. The use of these agents in studying meningioma could help clarify the role of gonadal steroid hormones in meningiomas and be used for selecting patients for treatment; however, confirmation of ligand uptake is not equivalent to receptor isoform elucidation important for predicting biological effects, which needs further study. Historic work on the matter of gonadal steroid hormones in meningioma should be evaluated in this context of poor receptor detection and with the understanding that the gonadal-steroid hormone and IGF-1/GH signaling axes are not independent features, but rather, parts of a shared regulatory axis. Similarly, the meningioma microenvironment investigating immune cell infiltrate has gained attention and recent years with successful clinical trials targeting PD-L1 [ 70 ]. Understanding more about how these pathways intersect and influence the tumor microenvironment has promise for precision medicine and uncovering biological mechanisms of meningioma tumorigenesis. The synthesis of clinical, pharmacological, genomic, and molecular data in meningiomas may help select the patients who will benefit most from various targeted therapies.

Hormones mediate control of gene regulation by binding to nuclear and trans-membrane receptors resulting in cell signaling cascades and epigenetic modifications. Nuclear hormone receptors relevant to meningioma include PGR , AR , estrogen receptor alpha ESR1 (ERα) and estrogen receptor beta ESR2 (ERβ) [ 24 ]. These bind to the DNA and act to influence chromatin configuration by recruitment of different coregulatory complexes responsible for different biological activity [ 25 ]. In the case of estrogen receptors, these represent distinct gene products on different chromosomes that lack cross-affinity when detected by antibodies and can have opposing gene regulation [ 26 ]. Another class of hormone receptors includes the trans-membrane receptors, many of which are G-coupled membrane receptors, such as somatostatin receptors 1–5 (SSTRs), gonadotropin-releasing hormone receptor (GnRHR or LHRHR), and the G-coupled Protein Estrogen Receptor 1 (GPER1) [ 27 ]. The clinical importance of distinguishing between receptors binding to the same ligand can be exemplified by the selective estrogen receptor modulator (SERM) tamoxifen. A SERM is a mixed agonist/antagonist with activity that differs between estrogen receptors; specifically, it is known to be an agonist to ERα but an antagonist to ERβ and GPER1 [ 25 , 28 ]. The clinical consequence of this mixed activity is demonstrated in tamoxifen’s use in treating breast cancer but its risk for developing endometrial cancer [ 29 ]. Our unpublished meta-analysis and comprehensive review of the literature identifies that ERα was the only estrogen receptor assessed in almost all studies of meningioma, with few notable exceptions [ 24 ]. Some authors who have detected ERα have reported its association with increased proliferation and high grade meningiomas [ 30 , 31 ]. Most meningiomas are PGR positive and its expression has been associated with WHO grade I and inversely correlated with chromosome 22 loss, where NF2 is located [ 31 - 33 ]. GNRHR expression is important because luteinizing hormone agonists are sometimes used to treat prostate cancer and meningioma growth is a potential complication of this treatment [ 34 , 35 ]. Multiple publications have identified SSTR1-5 expression in meningioma at high frequencies [ 3 ]. A large study of meningiomas with over 700 tumors identified that SSTRs are associated with different tumor characteristics, such as SSTR2A overrepresentation in skull-base tumors and a negative association with high grade and NF2 mutant meningiomas [ 3 ] and reported to be a prognostic marker [ 2 ]. While meningioma literature has often evaluated steroid-hormone receptors and SSTRs separately, ERα in particular is known to play a fundamental role in normal physiological regulation of GH, and the relationship between SSTRs, ERα and PGR expression in cancer is reported in prostate and breast cancer cells [ 36 - 38 ]. Additionally, high BMI is associated with meningioma development and it has been found to correlate with leptin receptor (LEPR) expression in meningiomas, a receptor known to communicate with the IGF-1/GH and gonadal steroid hormone pathways [ 39 - 41 ].

We reviewed clinical trials related to hormonal agents in meningiomas ( Table 2 ) using MEDLINE PubMed, Scopus, and the NIH clinical trials database ( clinicaltrials.gov ). Among these studies, only one was a randomized controlled trial ( NCT03015701 , Table 4 ), which assessed mifepristone in meningiomas, summarized below [ 42 ]. Several other studies were conducted but all had thirty-four patients or less with heterogenous patient and tumor inclusion characteristics. Studies defined complete response as no tumor after treatment, partial response (PR) as > 50% tumor reduction, and progressive disease > 25% tumor increase, unless otherwise specified. Early in vitro studies of tamoxifen in meningioma revealed a complex interaction with other hormones that could result in the stimulation or inhibition of growth [ 43 , 44 ]. Some population-based studies have suggested that patients who received tamoxifen for breast cancer treatment may have a decreased risk of meningioma relative to other breast cancer patients, but these results are limited due to the retrospective observational nature of the study design and relatively weak associations [ 45 , 46 ]. In 1985, Markwalder et al. conducted the first pilot study of tamoxifen for the treatment of meningioma patients and included 6 patients treated for a median duration of 31 months ( Table 2 ) [ 47 ]. Four were considered inoperable and method of meningioma confirmation was not specified. Two other tumors were considered non-operative recurrences. Four patients were evaluated at the end of the study, with one patient showing what the author describes as a nonspecific response to tamoxifen, two were stable, and one progressed ( Table 2 ). In 1993, Goodwin et al. conducted a single-arm Phase II clinical trial to assess the efficacy of tamoxifen on progressive meningiomas secondary to resection or radiation treatment failure [ 48 ]. There were twenty-one patients enrolled in the study but two were immediately excluded after pathologist review of tissue, though they are still included in aggregate demographics ( Table 2 ). Only one tumor (5%) was tested for estrogen receptor (isoform not specified) and it was negative ( Table 3 ). No information regarding the histology, grade, or location of tumors was provided. In total, one patient achieved partial response (PR) (> 50%), two (11%) had minor response (unspecified definition) of short duration (4–20 months), six (32%) were stable, and ten (53%) demonstrated progression (unspecified definition) with an overall median time to progression of 15.1 months. Megestrol acetate is a progestin with a similar profile to CPA in terms of antiandrogenic and PGR agonist activity, but is typically used as a hunger stimulant [ 49 ]. In 1990, Grunberg et al. conducted a single-arm study on a prospective cohort consisting of nine patients with measurable or evaluable unresectable tissue-confirmed meningiomas with recurrence after prior surgery [ 50 ]. They reported that two (22%) meningiomas had PGR status assessed, which were both negative. Seven patients (78%) experienced progression, with four (44%) being removed from therapy within 2.5 months. Two (22%) patients remained stable and stopped treatment due to weight gain without tumor progression. There were no cases where megestrol acetate resulted in regression of the tumor by any observable amount, but recent reports of progestin therapy contributing to skull base tumor growth, and the lack of assessment of grade, proliferation index, and other features, should be considered as explanations for rapid progression in this trial. Mifepristone is an antagonist to glucocorticoid receptor (GR) and PGR and a weak androgen/AR agonist [ 53 ]. In vitro and in vivo studies revealed effectiveness, but the authors also noted that mifepristone was effective in PGR negative meningiomas, implying its possible action on other receptors that were not evaluated [ 54 ]. In 2006, Grunberg et al. conducted the first human prospective study using mifepristone for persistent or recurrent unresectable meningioma [ 51 ]. Twenty-four patients (86%) had undergone prior surgery and tissue confirmation of meningioma; whereas the other four patients were not operated on and diagnosed based on clinical and radiological evidence alone. No tumors were tested for PGR, GR, or AR status but radiographic evidence of tumor regression equaling about 5–10% was seen in five (18%) patients and an objective improvement in visual field examination without radiographic improvement was noted in three (11%) patients. In 2014, Touat, et al. published a case-series of three female patients with multiple-meningiomas and evaluated the long-term treatment with mifepristone [ 52 ]. One tumor was evaluated for PGR where it was found to be expressed at the protein level. All three patients saw reduction in tumor volume of 22–72% after a period of 5–10 years. These initial findings prompted the initiation of a phase III randomized clinical trial of mifepristone that included one-hundred and sixty four patients with eighty in the placebo and eighty-four in the mifepristone arms [ 42 ]. Patients with pathology confirmed meningiomas demonstrating radiotherapy treatment failure defined by radiographic evidence of recurrence or progression after surgical resection, were eligible for the trial and were randomly assigned to either oral mifepristone 200 mg daily or placebo for 2 years. Notably, this study reported that 32% of patients had non-measurable—but evaluable—meningiomas at baseline, which others have pointed out as being unlikely to see any improvement compared to baseline [ 55 ]. While 52% of tumors were tested for PGR, nearly 12% of those tumors were found to be PGR negative and the remaining 48% were not evaluated. Of the 164 patients who began the trial, 30% of the mifepristone arm and 33% of the placebo arm completed the 2 years of treatment as planned. The median failure free survival (FFS) for placebo was 11 months CI 95% [6–18] and 10 months CI 95% [7–13], (p = 0.9) adjusted for sex, prior radiotherapy and disease status [HR 1.02, CI 95% (0.72–1.48)]. One patient each in the placebo and mifepristone trials achieved partial response (PR). Histology (aside from 10% atypical), grade, and location were not specified. The study intended to recruit 200 patients but did not achieve this. In combination with the inclusion of patients unlikely to show radiographic response (non-measurable tumors) and receptor negative meningiomas, along with missing relevant information, the study is likely to be more underpowered than originally calculated with limited interpretation of results. Thus, a future clinical trial with more specific patient inclusion criteria and documented tumor characteristics may be considered. In vitro studies of somatostatin showed efficacy, but the author of at least one highly cited study noted that one of the four tumors demonstrating growth inhibition was devoid of any SSTR expression [ 59 ]. A recent systematic review with meta-analysis of individual patient data was published that obtained complete datasets for 7 of 8 clinical trials assessing somatostatin as a monotherapy and combined therapy in meningioma [ 60 ]. When using somatostatin as a monotherapy for disease control, the meta-analysis found an odds ratio of 1.42 CI 95% [1.11–1.81, p = 0.005] for each 100 mg increase in total somatostatin. This effect for disease control was most pronounced in WHO I tumors, with the relative odds ratio for WHO II and WHO III tumors decreasing to 0.31 CI 95% [0.09–1.05, p = 0.06] and 0.08 CI 95% [0.02–0.33, p < 0.0001]. The authors note that their meta-analysis was compromised by poor quality of evidence due to different treatment regimens and heterogeneity between patients within trials and state they could not conclusively support or disqualify somatostatin as a treatment for meningioma. Several combination therapies with octreotide have been attempted including with everolimus, an mTOR inhibitor, and use of radiolabeled somatostatin analogs 90Y-DOTATOC, 177Lu-DOTATOC, 177Lu-DOTATATE. Studies generated in vitro suggested that everolimus has a synergistic effect with octreotide [ 61 ]. This led to two clinical trials including a total of 34 patients, to study the combination of octreotide and everolimus for disease control [ 62 , 63 ]. These studies were combined in the aforementioned meta-analysis, which found an odds ratio for disease control of 1.44, CI 95% [1.00–2.08, p = 0.05] [ 60 ]. Seperately, an individual patient data meta-analysis of radiolabeled somatostatin analogs for radiopeptide therapy synthesized 6 cohort studies including 111 patients and found a PFS6 of 94% for WHO I, 48% for WHO II, and 0% for WHO III [ 64 ]. The potential benefit of radiopeptide therapy over conventional radiation therapy is the ability to target tumors more precisely; however, the side effect profile is similar to other forms of radiation treatment. Nearly all studies required SSTR positive status but showed inconsistent efficacy; however, most studies included only somatostatin-radioligand detection of SSTR activity in the tumor and not a specific assay of the SSTR present (by IHC or another receptor-specific method). Future studies could investigate whether all SSTRs are functional in meningioma, or whether the receptors result in differences in treatment efficacy.

In this review of hormone-related clinical trials in meningioma, it is apparent that a disparity in the assessment of receptors in these tumors indeed exists between trials that employed gonadal steroid hormones versus somatostatin analogs. Understandably, the need for receptor status obtained by tissue confirmation complicates clinical trials that primarily includes unresectable tumors. However, given that the mechanism of action for most hormone therapy requires binding to specific receptors, it is imperative to assess that status before treating patients. Further complicating results in all trials are the small number of patients included and the lack of stratification based on grade and mutation status, which was not known when many of these studies were conducted, but influence recurrence of the tumors even within the same grade [ 14 ]. The use of radiolabeled somatostatin greatly enhanced clinicians to assess receptor status non-invasively; however, it does not provide specific receptor status of SSTR1-5, which may have influenced treatment outcome. Almost all current recruiting clinical trials are radiopeptide trials of somatostatin analogs with the exception of an observational study of meningiomas in a group of patients being treated with CPA ( NCT04372095 , Table 4 ), but there are no ongoing trials investigating other hormone-based therapy. Combination therapy appears to be a promising avenue, but given meningioma heterogeneity, more work should be done to evaluate what particular features—such as grade, receptor status, and somatic mutations—might predict an efficacious response in monotherapy, which can be used to further employ strategic combination therapy.

Meningiomas have a female predominance with an overall ratio of 2:1, and assessments of risk factors have implicated gonadal steroid hormones and receptor expression relevant to the growth hormone (GH/IGF-1) signaling axis [ 1 - 3 ]. Most meningiomas are progesterone receptor ( PGR ) positive, but inconsistent results between studies and incomplete assessment of all relevant hormone receptors have obscured the results for androgen receptors ( AR ) and estrogen receptors [ 4 , 5 ]. Scientific interest in the effects of gonadal steroid hormones on meningioma tumorigenesis and treatment has been renewed after evidence that some patients with long-term progestin/antiandrogen [cyproterone acetate (CPA)] exposure share molecular, histological, and anatomical features in addition to an increased risk of developing a meningioma [ 6 - 9 ]. These developments, and still poorly understood contributions of hormones to meningioma growth and development, led us to review what is known about this subject, the clinical trials that have been conducted, and considerations for the future of hormone therapy in meningiomas.