The suspicion of a link between CPA and meningiomaswas raised by our group in 2008[6]and was confirmedby numerous case seriesshowing regression or stabilization of meningioma(s) after CPA discontinuation[7, 8, 12, 13, 15, 17, 18, 27, 29–32]. Gill et al.[14]described an increased risk of meningioma(s)in patients under high dose CPA and with a longer than 1 year exposure. Recently, Weill et al.[19]proved this increased risk of meningioma(s) development with long term use of CPA and a strong dose effect[19].
CPA-related meningiomas also seem to be more often multiple and localized on the anterior or mid-skull base and the anterior convexity[21, 22].It is well known that the meninges have different embryological origins[33] depending on the anatomical site; this embryological origin might be associated with the pathophysiology of various diseases such as meningiomas[26, 34]. Recently, Okano et al[24] demonstrated that oncogenic mutations and meningioma characteristics (tumor location, histological findings) are associated with meningeal origins. Tumors from meninges originating from the neural crest more frequently harbor NF2 mutations or 22q loss[23, 24, 35], and are associated with more frequent fibrous histological type[24].Meningiomas with NF2 mutations are also associated with the posterior skull baselocation that originates from the dorsal mesoderm. The tumors arising from the meninges originating from the paraxial mesoderm harbor preferentially AKT1, SMO, KLF4 and PIK3CA mutations and are frequently associated with meningothelial meningioma type[24].
The mechanism(s) of pathogenesis of multiple sporadic meningiomas remain unclear, and two hypothesis have been raised: a monoclonal origin[36] or an independent development[36, 37] suggesting several mutational landscapes. Concerning progestin-associated meningiomas which account for almost one third of the multiple meningioma, they seem to be characterized by a specific mutational landscape. PIK3CA pathogenic mutations were identified in progestin-associated meningiomas in comparison with a control population[21, 22].
This is the first report of CPA related meningiomatosis with opposed evolution of meningiomas in the same patient after CPA discontinuation with regressing tumors while other meningiomas continued to growth. The most striking example of this differential growing pattern is case 2 in which two components exhibiting opposite behavior with volume increase of the convexity component along with a decrease in size of the skull base component. In our patients, NF2-mutations were identified in the growing tumors, although a PIK3CA mutation was only found in the tumor shrinking after CPA discontinuation.
Two mainhypotheses could be evoked to clarify this atypical phenomenon.
The first hypothesis for the opposed profile evolution of meningiomas in a same patient after cessation of CPA is the coexistence of hormone-related and not meningiomas with specific mutational landscapes.
For case 2, the hypothesis of two different meningiomas which would later unify in a single entityis plausible considering their distinct genetic profiles (Fig 3). It would imply the coexistence of two tumoral clones (hormonal or not induced) in a same patient, characterized by a different mutational landscape and probably different underlying tumorigenesismechanisms.
Furthermore, our results are in concordance with previous reports of an association between the mutational lansdscape of meningiomas and the embryological origins of the meninges from which they arise. All growing-meningiomas harboring NF2-mutations were localized in the convexity or the tentorium (who is originated from the neural crest) and were fibroblastic meningiomas. In the case 2, the spheno-orbital part of the meningioma, decreasing in volume after CPA-discontinuation, harbored a PIK3CA-mutation that is common in CPA-related meningiomas[21]. However, not only driver genetic mutations can explain the tumorigenesis of CPA-related meningiomas but also the epigenetic regulation and methylation status which is commonly implicated in tumor progression[38].
A second hypothesisthat all meningiomas were hormone-related but prolonged exposure could generate loss of progesterone receptors.
PR are found in most meningiomas[5] and are believed to play a significant role in meningioma growth[39]. Progestin exposition could induce the growth of meningeal cells through their intracellular receptors in concert with coactivators and corepressors[10]. Previous reports indicated thatloss ofPR status has been associated with a high cellular proliferation, mitotic index (Ki-67), recurrenceand high-grade tumors[40, 41]. Moreover, according to the tumor location, meningiomas express different patterns of sex hormonal receptors[42]. CPA-related meningiomas often present a higher PR expression than non-CPA meningiomas[42], and PIK3CA mutations occurred also more frequently in the presence of the hormone receptor overexpression[43].
Long-term and high-level progesterone stimulation could generate tumoral genomic alterations in CPA-meningiomas leading to progressive loss of progestin tumor receptors, thus explaining a continued growth of these tumors even after CPA withdrawal.This hypothesis refers to the mechanisms of drug resistance in patients with advanced cancers, thus selecting some clonal tumor cells. One of themincludesgenetic alterations leading to anabnormal conformation protein that impair the binding of the drug.
This hypothesis is also consistent with the pathological findings of our case 2in which we observed a PR level lower than 30% and Ki-67 at 12 % in the convexity growing part of tumor harboring a NF2-mutation. However, in the cases3 and 4, we observed in the skull base growing meningiomas a classic higher level of PR with a low Ki-67 rate as well asin the tentorial meningioma of the case 1.
If a genetic predisposition to CPA meningiomatosis exists, as suggested by familial cases[44], it must not be the only explanation considering the coexistence of two hormone response patterns. Others factors in the genesis of these tumors remain to be determinated.As the number of patients followed for progestin-related meningiomas increased significantly in the last years, it is important to understand why some meningiomas continue to grow meanwhile most of them regresses or stabilize.