Introduction

Meningiomas account for 30 % of primary brain tumors and are the most common type of brain tumors over the age of 35 [17]. Histologically, most meningiomas are World Health Organization (WHO) grade I, regarded as benign tumors [15].

The relationship between meningiomas and sex hormones is well known, and arguments for such an association are numerous, as follows: a female predominance, tumor enlargement during pregnancy, tumor shrinkage after delivery, and a positive association with uterine fibroids, endometriosis and breast cancer [4]. Additionally, meningiomas express more hormonal receptors than the meninges themselves, and two thirds of meningiomas express progesterone receptors, and in greater amounts than estrogen receptors [20]. Based on the histological analysis of progesterone receptors in meningiomas, antiprogesterone therapy has been proposed for unresectable meningiomas, with moderate success [11]. Nevertheless, the underlying mechanism of action of sex hormones on meningiomas remains unclear, and the association between meningioma and exogenous sex hormones is still debated.

Cyproterone acetate (CA) is a synthetic steroidal antiandrogen drug, which is indicated for severe signs of androgenization in women, the treatment of inoperable prostate carcinoma and the reduction of sexual drive in men. It is also widely used to treat seborrhea, acne, hirsutism and moderate androgen-related alopecia. Because it reduces acne, seborrhea and hirsutism, it is found in smaller quantities in some combined oral contraceptives. In addition to its antiandrogen properties, CA has weak progestogen activity, so it is also used to treat hot flashes.

The possible role of CA in the development of meningiomas was first described by Froelich et al. in 2008. They reported multiple meningiomas in nine women treated with CA for several years, and treatment withdrawal was associated with tumor stabilization. To date, some authors have also reported the association between meningiomas and the long-term use of CA [6].

We are presenting a series of 12 patients with meningiomas that decreased in size for 11 of them after discontinuation of CA.

Patients

From 2010, each patient, seen in the clinics in our department for a radiological diagnosis of meningioma (on MRI), was questioned about hormonal intakes. Radiological and clinical data for all patients with meningiomas and previous treatment of CA were then retrospectively reviewed.

Twelve patients harboring one or multiple meningiomas and treated with CA were identified. Another hormonal medication was associated with CA (estradiol) for one of them and discontinued at the same time. None of them presented with neurofibromatosis type 2. None had previously received whole-brain radiation. According to our previous reported experience with CA and meningiomas (Froelich, 2008) [2], patients were informed about the likely link between CA and meningiomas, and treatment was systematically discontinued.

Patients were followed up with serial MRIs after medication withdrawal. The mean duration between treatment withdrawal and the first MRI follow-up was 3.5 months.

Ethics statement

This study was approved by the local ethics committee of Pôle Neurosciences of Lariboisière Hospital. Written consent for inclusion in this study was obtained and recorded in each patient’s medical file.

Methods

OsiriX® software was used to measure and compare tumor volume. Two observers performed tumor volumetry using OsiriX®. DICOM data from each patient were stored in the OsiriX®. The outline of the meningioma was traced manually on each slice with the “closed polygon selection” and "repulsor" tools. After selecting all of the regions of interest within one series, the volume was automatically calculated using the region of interest (ROI) segmentation function.

The total volume for patients with multiple lesions was calculated by summing up the volumes of each meningioma.

Results

Patients were all females with a mean age of 49 years (Table 1). All patients had been taking CA for a long period varying from 8 to 30 years (mean: 18.6 years). The dose of CA ranged from 15 to 50 mg daily with an average dose of 40 mg. In 10 of 12 cases, MRI showed multiple meningiomas (2 to 10 lesions) at the time of diagnosis. Patients with multiple tumors had used CA for a longer period of time (mean 20.4 years; range: 8–30) than the two patients with only one tumor (mean 10 years). CA was discontinued in all cases considering the possible association between the drug and development of multiple meningiomas. Ten out of 12 patients were managed conservatively without surgical treatment and followed up with serial MRIs. In two patients, surgery was indicated because of rapid visual deterioration (Fig. 1a). The primary goal of surgery was optic nerve decompression with only partial tumor resection in both cases. In these two patients, histological analysis showed grade I meningothelial meningiomas with a strong progesterone receptor positivity. A total of 53 MRIs were analyzed. The mean interval time between two successive MRIs after medication withdrawal was 5 months (range: 1–14).

Table 1 Summary of meningioma cases associated with long-term use of CA
Full size table
Fig. 1
figure 1

Two illustrating cases from our series. a Case (1) of a 51-year-old female who was referred with a clinoid meningioma for a right-sided decrease in visual acuity for 3 months. She had been taking 15 mg of CA daily for 30 years. MRI (June 2013) showed three meningiomas. CA was stopped, and the right optic nerve was decompressed 1 month after discontinuation because of non-improvement of visual acuity with a good final recovery. The pathological examination revealed a grade I meningothelial meningioma with strong immunohistochemical positivities for the progesterone receptor. We observed the shrinkage of the two other meningiomas. b Case (4) of a 39-year-old female presenting with hearing loss and nonspecific headaches. An MRI showed a right cerebellopontine angle meningioma (13.5 cm3) with slight tumor enlargement at the beginning of the follow-up. She had been taking 50 mg of CA daily for acne and dysmenorrhea for more than 10 years. CA was discontinued in February 2012 followed by a significant decrease in the size of the cerebellopontine meningioma over a 28-month period

Full size image

Tumor shrinkage was observed in 11 out of 12 cases within a mean period of 5 months (ranging from 2 to 11 months), and there was no regrowth of the tumor in any cases during a mean 12- month follow-up period (Figs.1b and 2). In several patients with multiple tumors, we observed different length curves of shrinkage for each tumor. We observed a mean tumor volume reduction of around 10 cm3/year (range: 0.0; 76). In one patient, the tumor remained stable.

Fig. 2
figure 2

Cases in our series: Evolution of tumor volume before and after CA discontinuation

Full size image

The mean follow-up (from medication withdrawal time) was 12 months (range: 5–35).

Discussion

The observed increased incidence of meningioma during the last decades has been most pronounced among women [4, 20, 12]. One explanation for this phenomenon is the increased use of postmenopausal hormone replacement therapy (HRT) [13]. However, epidemiological studies of the association between HRT and risk of meningioma have yielded conflicting results [21, 5, 14, 4].

Several authors have reported the link between meningioma growth and treatment by progesterone agonists [9, 18]. Shimizu in 2004 [24] reported a case of meningioma regression after the discontinuation of chlormadinone acetate (progesterone agonist).

Adams in 1990 [1], describing in vitro effects of of cerebral meningiomas’ hormonal dependency, reported that CA inhibited growth of meningioma cells. However in 2008, Froelich et al. (Does cypropeterone actate promote multiple meningiomas? In: Proceedings of the 10th European Congress of Endocrinology, Berlin) first described the link between multiple meningiomas and prolonged treatment with CA in a series of nine women with tumor stabilization after treatment withdrawal. To date, there have been several reports of meningiomas associated with the long-term use of CA, and tumor shrinkage was also observed in some cases after discontinuation of the drug [2, 8, 21]. In addition, similar cases of meningiomas related to the long-term use of other progesterone agonists have been also reported [25, 19, 18, 24, 10] (Table 2). Interestingly, among all previously reported cases, most (14/15 cases; 93.3 %) of the patient harboring multiple meningiomas were treated with either CA or megesterol acetate, suggesting that different mechanisms may exist among progesterone agonists to explain the development of meningiomas.

Table 2 Case reports of meningiomas associated with the use of progesterone agonists
Full size table

Because all of our patients except one had been taking CA for more than 10 years (mean 18 years), the duration of exposure to CA seems to be a factor for developing meningiomas. In addition, patients with multiple tumors had used CA for a longer period of time (mean of 20.4 years) than the patients with only one tumor (10 years), so longer exposure to the CA may induce the development of multiple tumors.

We observed that most of these meningiomas were located principally on the anterior skull base.

Meningiomas of the convexity seem to regress faster than in other locations in our series independently of the initial volume (cases 1, 7 and 10) and number of tumors (Fig. 2).

Is there a particular role of the localization in the growth of these tumors? Several reports that have described a rapid growth of meningiomas during pregnancy have also reported a higher basal localization, approximately 60 %, compared with the hemispheric localization, suggesting differences in the hemodynamics and biology of these meningiomas [16, 22]. Clark [3] correlated the mutation spectrum of 300 intracranial meningiomas with anatomical distribution and histological subtype. They found that meningiomas originating from the anterior and medial skull base were non-NF2 mutant and nearly always benign with chromosomal stability, in contrast to meningiomas with cerebral and cerebellar hemispheric localization that presented mutant NF2 and/or chromosome 22 loss and were more likely atypical with genomic instability [3]. Our observation that CA promotes preferentially basal localization of meningiomas may suggest that CA is directly involved in these signaling pathways.

Gil et al. [7] conducted a population-based cohort study in Spain and observed a higher risk of meningioma among high dose users (over 50 mg/day) of CA for more than 1 year. The risk of meningioma associated with a high dose of CA was also reported by Gazerri et al. [6]. They reported the case of a male-to-female transsexual treated with a high dose of CA (100 mg/day) for several years who developed a giant olfactory groove meningioma. In our study, 8 out of 12 patients had been taking a high dose of CA (50 mg/day) before their first presentation, whereas 4 other patients had been taking a relatively low daily dose (15–25 mg/day). Not only the higher dose of CA but also the length of exposure seem to be important factors for the development of meningiomas. Considering our experience and other reports, interruption of CA is mandatory in case of a meningioma diagnosed under CA treatment. However, CA may be the only efficient option to treat hyperandrogenic symptoms, and the psychological difficulties caused by the reappearance of hyperandrogenic symptoms after CA discontinuation should not be underestimated. More data are needed on the dose effect of CA on meningiomas in order to evaluate whether low CA could be pursued in case of severe hyperandrogenic symptoms under careful radiological follow-up.

In both our operated cases, histological examination showed a strong progesterone receptor positivity, which may account for the effect of CA on meningiomas. However, the multiplicity of meningiomas under the use of CA seems to be a typical feature of CA compared to others progestative agonists.

The link between administrations of CA and the increased growth rate or development of meningiomas now seems evident, but the underlying mechanisms are still unknown. In the only report on the effect of CA on meningioma culture cells, Schrell et al. [23] noted inhibited cell growth. Thus, in vivo effects of CA are probably mediated through an indirect signaling pathway.

To clarify the role of CA in the development of meningiomas, many parameters need to be examined, and further experimental studies are necessary. Currently, even if many unanswered questions remain, practical lessons can be drawn from our cases, as well as other reported cases. Care must be taken with long-term use of high doses of CA. In such cases, the risk of meningiomas must be taken into account and a serial brain imaging study must be considered. When the first injected CT or MRI should be performed remains unclear and seems to be related to the dose of CA. When a meningioma is diagnosed, CA should be stopped, and if one considers the strong probability of reduction in size after CA discontinuation, conservative treatment with serial MRIs should be the first treatment option except if surgery is required because of the rapidly worsening neurological deficits. Given the potential psychological consequences of CA withdrawal, the toxic doses must be identified to optimize the management of some patients.

Conclusions

We reported our recent experience with a series of meningiomas associated with the long-term use of CA. Because discontinuation of CA led to shrinkage of tumors in all but one case, treatment discontinuation should be considered as the first line treatment in such cases. Past or present history of CA treatment should be documented systematically in case of meningiomas and even more so in cases of multiple meningiomas.