Meningiomas are also diagnosed in canine and feline species displaying a strong relationship with age and breeds and accounting for from 22.3% up to 50% of all brain tumors in dogs
[6][7][8] and 58% in cats
[9]. In a comparative study between canine and feline meningiomas, Wada et al. (2020) highlighted the development of tumors at median ages of 11.7 years and 14.1 years, respectively. According to the same study, the canine breeds most affected were Miniature Dachshund, Toy Poodle, Beagle, Shetland Sheepdog, Labrador Retriever, Flat-coated Retriever, Shiba Inu, Jack Russel Terrier, Welsh Corgi, and mixed breed
[10]. A Japanese study that analyzed data from 186 canine intracranial tumors showed a breed predisposition for meningiomas in Rough Collie, Golden Retriever, Miniature Schnauzer, and Scottish Terrier
[11]. Other studies associated a frequent meningioma diagnosis with canine dolichocephalic breeds
[12][13][14][15], while a female sex predisposition was not confirmed in more recent studies
[7][16]. In cats, Domestic Shorthair seems to be the most predisposed breed to meningioma and no significant difference between sexes exists
[16]. Adamo et al. (2003) reported a higher onset of meningiomas in Persian, Domestic Shorthair, and Domestic Longhair cats
[17]. In dogs, meningiomas originate in the calvarium-adjacent region, involving the olfactory and frontal regions, cranial cavity, optic chiasm, and suprasellar and parasellar regions, although it has been rarely diagnosed in other regions
[18][19]. The sites in which feline meningiomas arise are mainly the tela choroidea of the third ventricle, the supratentorial meninges, and less frequently the cerebellar meninges
[9][20][21]. Conversely to what happens in dogs, multiple meningiomas are frequently found in cats (about 17% of all meningioma cases)
[9][16][22][23]. Three theories have been proposed to explain the occurrence of multiple meningiomas: multicentric dural foci, metastasis by blood-borne spread, and metastasis via the cerebrospinal fluid. The first hypothesis seems most plausible considering the tumor’s histologically benign nature and the reports showing histological variants in the same patient
[24]. However, more research is needed to confirm them.
2. Pathogenesis
The factors contributing to the development of meningiomas in humans and domestic animals need to be further studied. Currently, several hypotheses and mechanisms have been proposed.
Ionizing radiation stands out as the primary environmental risk factor consistently linked to the development of meningiomas. Human exposure to ionizing radiation leads to a 6- to 10-fold incidence increase in this condition
[26]. Furthermore, this heightened risk is notably evident among survivors of the atomic bombings in Hiroshima and Nagasaki, where a substantial increase in meningioma cases has been documented
[27][28]. In addition to radiation exposure, occupational contact with herbicides and pesticides also appears to elevate the likelihood of developing meningiomas
[29]. Moreover, obesity has been identified as a significant positive risk factor for tumor development, likely due to its association with chronic inflammation and the signaling of insulin or insulin-like growth factors
[30][31][32][33][34].
Several receptors are over-expressed in meningiomas, particularly somatostatin receptors (SSTRs) and intracellular receptors for sexual steroid hormones, such as androgens, progesterone, and estrogen, suggesting a role for these systems in tumor pathogenesis
[27][35][36][37][38]. Among SSTRs, although all the subtypes are expressed in meningiomas
[39], SSTR2 expression was associated with a poor prognosis, while SSTR1 expression was associated with reduced incidence of relapses, with less strong evidence. Nevertheless, the in vitro activation of SSTRs demonstrates antiproliferative effects. This evidence led to the development of clinical trials exploring the use of selective agonists, such as octreotide and pasireotide, although, to date, no conclusive results have been reported
[40]. Receptors for the steroidal hormones estrogen (ERα) and progesterone (PR) have been reported to be expressed in most meningiomas: ERα presence has been associated with increased proliferation and the development of high-grade tumors, while the high expression of PRs has been correlated with Grade 1 tumors according to the WHO grading system
[41]. It is now assumed that PR expression alone represents a favorable prognostic factor in meningiomas, while its loss or the association with ER expression correlates with a worse clinical outcome. However, pharmacovigilance data indicated that the prolonged use of androgen receptor antagonists and/or progesterone receptor agonists (cyproterone acetate, nomegestrol acetate) results in increased meningioma incidence
[42][43], making rather complex the prognostic or therapeutic evaluation of these receptors.
Feline and canine meningiomas also express PR
[17][44]. Adamo et al. (2003) reported a high proportion of PRs and the absence of ERs in feline meningiomas. Furthermore, a high number of cells with PRs and a significantly lower number of cells with ERs in canine meningiomas were observed. According to this study, the proportion of PR-positive cells in canine benign meningiomas was >80%, while in malignant meningioma only 32% of cells were PR-positive; in cats, the percentages were >80% and 38%, respectively. In dogs, the number of PRs correlated to more aggressive progression (with nuclear pleomorphism, severe necrosis, and histological subtypes), while in cats, such a correlation was not observed
[17].
In humans, meningiomas occur in several forms, sometimes associated with other syndromes. Table 1 shows the syndromes associated with high-frequency meningioma and the gene assumed to be involved.
Table 1. Syndromes associated with a considerable frequency of meningiomas.
Some of the syndromes listed in
Table 1 have also been modeled in animals or have been observed in other species. A murine model was developed to accurately replicate the human
NF2-related schwannoma phenotype, including the deficit in hearing and balance
[58]. Mice with mutations in
PTC, an orthologue of human
PTCH1, develop many of the characteristics of Gorlin syndrome and exhibit a high incidence of rhabdomyosarcomas
[59].
PtenM3M4 missense knock-in mutant mice present megalencephalic brains and elevated nuclear proteasome activity, also observed in patients with Cowden syndrome-related mutations in PTEN
[60]. Multiple endocrine neoplasia type-I-like syndrome was reported in two male Domestic Shorthair cats that developed symmetric alopecia, insulin-resistant diabetes mellitus, and pituitary-dependent hyperadrenocorticism at 12 and 13 years of age
[61] and in a crossbred 12-year-old male dog with abdominal enlargement, seborrhoea, and polypnea
[62]. Nevertheless, the listed studies did not highlight associations with meningioma.
Few studies reported contemporaneous and unrelated neoplasms in 3–23% of dogs with IPTs, mainly in the thoracic or abdominal cavity
[6][7][63].
As reported in a review by Motta et al. (2012), intracranial meningiomas in dogs as well as in cats have been diagnosed with concurrent neural (oligodendroglioma and meningioangiomatosis)
[7][9][64][65] or extra-neural disorders (mucopolysaccharidosis type 1 and thymic lymphoma)
[66][67]. Moreover, it has been reported that 13.9% of cats and 19% of dogs develop a meningioma in addition to another intracranial neoplasm
[7][9][64][68].
In humans, the malignant forms of meningiomas increase the tumor cell invasion processes and the risk of metastasis is higher compared to non-malignant meningiomas
[69]. Metastases developments are reported mainly in the lung, pleura, bone, and liver. In domestic animals, meningioma metastases have been described almost uniquely in dogs, and mainly pulmonary metastases were observed
[70][71]. A study performed in cats highlighted skull osteolysis. The authors hypothesized that metastases could be responsible for osteolysis
[72].
3. Histopathological Classification
The WHO classifies meningiomas in humans and domestic animals with similar criteria.
In humans, the WHO classifies meningiomas into 15 subtypes, reflecting a broad heterogeneity
[73]. These are clustered into three groups, differentiated by their histological components. The first group is composed of benign forms classified in different variants: meningothelial, fibrous, and transitional, which are the most common forms. Psammomatous, angiomatous, microcystic, secretory, lymphoplasmacyte-rich, and metaplastic variants are also included in group 1, but their incidence is significantly lower. In group 2, three additional classes are clustered: atypical, choroid, and clear cell meningiomas. The third group encompasses anaplastic, papillary, and rhabdoid meningiomas, often diagnosed as a single histological type, but including different biological and oncological aspects associated with different documented genetic mutations
[74][75].
In domestic animals, meningiomas are divided into subtypes, according to the morphological characteristics of the cells
[76]. Initially, WHO classified domestic animal meningiomas into two categories: benign and malignant. The benign meningiomas included eight subtypes: meningotheliomatous, fibrous, transitional, psammomatous, angiomatous, papillary, granular cell, and myxoid. Malignant meningioma was classified as anaplastic. However, this classification presents some limitations, and considering the similarities between human and domestic animal meningiomas in pathological, immunological, molecular, and MRI aspects, an improved classification was defined. The benign meningioma classification must be used with caution as histological aspects sometimes lead to considering benign neoplasia that does not match with biological/oncological characteristics
[73][77][78][79]. In a study on feline meningiomas, concurrent benign and malignant forms were diagnosed
[68].
Dogs show a rare form of meningioma, known as cystic meningioma, characterized by cysts originating through tumoral processes, such as necrosis or release of fluids. The size of the cyst depends on the fluid volume and causes increased intracranial pressure
[80].
The observed similarities in pathological, immunological, and MRI aspects between human and canine meningiomas allowed for the classification of canine meningiomas according to three grades of the 2016 WHO human histological grading system
[77]. In order to evaluate the possibility of translating the human grading system to canine tumors in terms of accuracy and reproducibility, Belluco et al. (2022) evaluated veterinary neuropathologists’ inter-observer agreement with the application of a human grading system to canine meningioma
[77]. The reproducibility of each histologic criterion was evaluated to identify a possible disagreement. The authors proposed amendments to increase reproducibility in canine meningioma
[77]. In their study, Belluco et al. (2022) proposed a criterion for canine meningioma classification (
Table 2) based on mitotic grade in a specific area (2.37 mm
2) of tumor tissue
[77].
Table 2. Canine meningioma classification criteria proposed by Belluco et al. (2022)
[77].
Commonly, canine meningiomas present characteristics similar to the group 1 WHO classification of human meningiomas. The group 1 meningioma subtypes include meningotheliomatous, fibrous (fibroblastic), transitional (mixed), psammomatous, angiomatous (angioblastic), papillary, granular cell, myxoid, and anaplastic (malignant)
[81]. Canine meningiomas usually display transitional, meningothelial, microcystic, and psammonas histological aspects
[82]. In some cases, canine meningiomas present chondroid, osseous, myxoid, and xanthomatous-like areas in meningotheliomatous and transitional subtypes. Another aspect highlighted in meningotheliomatous and transitional subtypes was polymorphic infiltration with or without tumor cells necrosis area. Feline meningiomas are commonly classified into the transitional and fibroblastic subtypes. However, feline meningioma histology is cytologically bland and uniform; for this reason, it is very difficult to adapt to the human WHO guidelines
[82]. The Comparative Brain Tumors Consortium (CBTC) tried to establish the translational aspects of canine brain tumors as a model for their human counterparts
[83]. The work of CBTC provided the foundation for a histologic atlas of canine glioma that included astrocytoma, oligodendroglioma, and undefined glioma
[84]. The samples were collected from several institutions and analyzed with immunohistochemistry to evaluate the expression of specific markers
[84]. Other criteria studied were infiltrations, necrosis, mitosis, and vascularization. A grading classification was built reporting these criteria
[84]. The grading was compared with human tumor counterparts. Belluco et al. (2022) evaluated the reproducibility of criteria used in the human meningioma grading when applied to canine meningioma
[77].
4. Diagnosis
Meningiomas are diagnosed in both humans and domestic animals through a combination of clinical evaluations, imaging techniques, and histopathological analysis. The primary diagnostic tools employed to assess the presence of meningiomas are magnetic resonance imaging (MRI), computed tomography (CT), and positron emission tomography (PET)
[85][86].
Specific conditions such as edema, cyst formation, change in vascularity, and necrosis are well detected by MRI
[87], and novel imaging techniques are improving the diagnosis of brain lesions. The use of MRI contrast agents permits the distinction of the tumor from the normal tissues. Furthermore, neurosurgeons can use MRI to minimize the size of craniotomy, maximize tumor removal, and minimize damage to the surrounding brain. Moreover, diffusion-weighted imaging, magnetic resonance spectroscopy, and dynamic contrast-enhanced MRI are entering clinical practice
[88][89][90][91][92], whereas cerebrospinal fluid (CSF) analysis can provide clinical parameters, such as altered protein content and leukocyte count, useful for diagnosing pathology
[93]. Motta et al. (2012) reported the main MRI characteristics of dog and cat meningiomas
[16].
Table 3 summarizes meningioma MRI features observed in canine, feline, and human meningiomas
[91][94][95][96].
Table 3. Common MRI features observed in canine, feline, and human meningiomas.
Immunohistochemistry (IHC) plays a significant role in the diagnosis of meningiomas. IHC can be performed in human, canine, and feline meningiomas, using similar markers (
Table 4). Saito et al. (2021) described about 39 cases of feline meningiomas of various grades in which specific markers were observed to classify the tumors
[97]. The study analyzed markers such as Cytokeratin, Vimentin, E-cadherin, β-catenin, N-cadherin, and Ki-67. In feline meningiomas, Cytokeratin was only recognized in particular histological phenotypes (fibrous and transitional types) and showed high immunoreactivity in some studies involving dogs and humans
[97]. Vimentin tested positive in some cases of feline meningiomas. According to Saito et al., E-cadherin remained stable in all subtypes of meningiomas, making it a reliable marker for feline meningiomas. In some studies, N-cadherin was detected in many human and canine meningiomas
[98][99]. β-catenin was analyzed in both canine and feline meningiomas. In feline meningiomas, while β-catenin was detected in over half of the analyzed meningiomas, its translocation to the nucleus (indicating the active form) was observed only in specific tumor types; in particular, translocation was not evaluated in atypical and anaplastic subtypes. In dogs, β-catenin was detected at the nuclear level, mainly in anaplastic meningiomas
[100]. Ki-67 is an important proliferation index associated with many tumors, as well as in meningioma
[101]. In the human literature, KI-67 expression correlates with tumor aggressiveness in meningiomas
[102][103][104][105]. Matiasek et al. (2009) evaluated the role of KI-67 in dog meningiomas
[106] considering about 70 canine meningiomas. The samples were analyzed via immunohistochemistry and 64 cases tested positive for KI-67
[106]. Some studies reported KI-67 be predictive for the survival of dogs with non-nervous tissue tumors
[107][108][109][110]. Matiasek et al. (2009) did not find the same prediction; however, this was a retrospective study performed with a small sample size
[106]. Janssen et al. (2023) studied the expression of KI-67 in 68 canine meningiomas to correlate KI-67 expression with the WHO grading of meningioma
[111]. Many samples positive for KI-67 were classified as WHO grade I and the authors hypothesized a possible role of KI-67 in meningioma development, in particular during the early stage of these tumors
[111]. Saito et al. (2021) did not correlate Ki-67 with specific subtypes of feline meningiomas
[97].
Table 4. Immunohistochemical markers evaluated in human, canine and feline meningiomas.
Therefore, the above-mentioned observations highlight common characteristics in meningioma classification and similar markers in IHC among the three species.
5. Molecular Characteristics
Meningiomas shows numerous molecular alterations in both human and domestic animal tumors, and some similarities and differences have been reported.
In humans, many genetic alterations have been associated with meningioma development. Deletions of
NF2 seem to be a common characteristic condition of meningiomas
[114]. Indeed, alterations in
NF2 have been recognized as meningioma’s driver mutation, being present in about 50% of sporadic meningiomas
[114]. Neurofibromatosis type II determines the inactivation of merlin gene
NF2, a tumor suppressor involved in cytoskeleton dynamics, tumor-associated increased motility, and the regulation of cell proliferation
[115]. Another pivotal player in meningioma’s development is
EPB41L3, also known as
DAL-1 or
4.1B, a tumor-suppressor gene that encodes erythrocyte membrane protein-band 4.1-like 3, involved in cell–cell interaction and having an important role in the control of motility
[116][117][118]. Decreased
EPB41L3 expression is observed in about 70% of meningiomas.
Alterations in chromosome 1 were frequently found in meningiomas
[74][119] and include mutations in
TP73,
CDKN2C,
RAD54,
EPB41,
GADD45A, and
ALPL [120][121][122][123].
Loss of function in chromosome 14 is commonly found in high-grade meningiomas and includes inactivation in NDRG family protein 2 and maternally expressed gene 3 (
MEG3), which were associated with a poor prognosis
[74][124][125]
Loss of function in chromosome 9 due to the deletions of the cyclin-dependent kinase inhibitors 2A (
CDKN2A) and 2B (
CDKN2B) has been associated with the progression from Grade 2 to anaplastic meningioma (Grade 3)
[74][126].
Whole-genome sequencing approaches have identified mutations occurring in
TRAF7,
AKT1,
KLF4,
PIK3CA, and
SMO, although these mutations seem to be mutually exclusive with those associated with
NRF-2 [127][128][129].
TRAF7 is mutated in about 20% of all meningiomas and is frequently associated with
KLF4,
AKT1, and
PIK3CA mutations
[74][128].
AKT1 mutations are found in about 12% of Grade 1 meningiomas and, although less frequently, in Grade 2 and 3. Interestingly, about 50% of all
AKT1-mutated meningiomas also show alterations in
TRAF7 [74][128][130].
PIK3CA mutations occur in about 7% of all meningiomas; they are mutually exclusive with
NF2 and
AKT1 and often associated with
TRAF7 mutations
[127]. Other somatic mutations associated with meningiomas include
BAP1,
SMARCB1,
SMARCE1,
BRAF-V600E,
NOTCH2,
CHEK2,
PTEN,
CDKN2A,
CDKN2B, and
DMD [52][131][132][133][134][135][136][137][138].
In humans, telomerase alterations are frequently associated with an increased risk of meningioma development
[139]. Telomerase mutations occur in all grades of meningiomas, the frequency being associated with the tumor grade
[140][141][142]. The main somatic mutations occur in the promoter region at two specific hotspots, C228T and C250T, resulting in up-regulation of the protein and the increased survival of cancer cells
[141]. Slavik et al. (2022) performed RNA-seq in 64 meningiomas to identify novel prognostic markers for these tumors. This study found the dysregulation of many transcripts involved in the WNT signaling pathway, highlighting the importance of the WNT pathway in meningioma development, as already reported for dogs
[143].
Initially, meningiomas were studied in athymic nude mouse models, after PDX meningioma cell inoculation. The study by Rath et al. (2011) demonstrated that meningioma PDX retained the characteristics of the original tumors and that a stabilized cell line of meningiomas exhibited similar features to the tumors of origin
[144]. A limitation of the xenograft models is the absence of the tumor microenvironment, which restricts the study of the aspects of meningiomas, including drug resistance. GEM overcomes this limit by having an intact tumor microenvironment. Peyre et al. (2018) utilized a model of transgenic mice in which alterations in
NF-2 and
CDKN2AB were induced
[145]. Nevertheless, as above reported, studying meningiomas in mice models does not yield the same results obtained with spontaneous animal models and, in general, the existence of spontaneous models is a fundamental tool for studying cancer and other pathologies. Indeed, they mimic key aspects in the investigation of novel therapeutical approaches that are absent in non-spontaneous models, i.e., the role of the microenvironment and of the immune system
[146].
Few available studies explored the genetic factors involved in the formation of canine and feline meningiomas. However, canine and feline models are reported to be strong translational models for CNS diseases, like stroke, epilepsy, movement disorders, lysosomal storage diseases, Alzheimer’s and cognitive disorders, and neuro-oncology
[113]. Differential expression regulation in orthologue genes of Homo sapiens in canine and feline meningiomas has been reported (
Table 5). Partridge et al. (2020) highlighted the advantages of using canine and feline models to study meningiomas since the larger size of the animals allows for surgical handling of the CNS; moreover, the presence of an intact immune system and the molecular, histological, and neuroimaging characteristics make them comparable to human neoplasia
[113].
Table 5. Reported differentially expressed genes in canine and feline meningiomas.
Besides genetics, it is important to consider the role of epigenetics in meningioma development. Many studies aimed at the identification of specific methylation profiles that could explain the mechanisms of oncogenesis, and some have demonstrated an epigenetic role in human meningioma onset
[74][154][155]. These epigenetics mechanisms include DNA methylation, defective chromatin remodeling, alterations in microRNAs, and the hypermethylation of
TIMP3. Alterations in the methylation of the
TIMP3 promoter determine inhibition of the metalloproteinases and can be associated with a poor prognosis
[74][156][157].
TP73 inactivation by hypermethylation has been investigated as a possible risk factor for malignant meningiomas
[156][158]. Interestingly, experimental evidence suggests that methylation profiles might be suitable to predict the clinical outcome of patients. Indeed, some studies showed that an altered methylation profile is associated with a worse prognosis
[142].
Few studies analyzed the molecular alterations driving canine and feline meningioma pathogenesis
[153]. Different genes resulted in down-regulation, for instance
MYOC,
ALP,
PRKD1,
FHL5,
TIE1,
MCAM and
PECAM1. Courtay-Cahen et al. (2008) highlighted the following alterations in canine meningiomas: histone acetyltransferase p300, PDGF-β, thioredoxin reductase 1, mutS homolog 2 and 6, Dal-1, Clusterin-like 1 (Retinal), B-cell lymphoma, T-cell differentiation protein, BCL-2-like 11, IL-1α, and IL-1β
[159]. An RNA-seq transcriptome analysis performed in canine meningioma determined a series of over-expressed genes, for example,
FOSB,
KLF5,
WNT,
GEM,
EGR1, and
DACT2 [153]. The entire altered gene list is reported in
Table 5 and includes also genes altered in both humans and dogs:
FOS,
KLF5,
BMPR1B,
NR4A1,
WNT5A,
PDPN,
MYRF,
MYC,
ALP,
COL14A1,
THSB1,
MMRN2,
FHL5,
SFRP1,
MCAM,
DAAM2,
AQP1,
KITLG,
PECAM1,
OLFML3,
ADCY2,
NOTCH3,
FBLN2,
APOD,
PTCH2,
IGFBP6,
GAS6,
COL16A1,
PIK3R1, and
FMN2 [160][161][162][163][164][165][166][167][168][169][170][171][172][173][174][175][176][177][178][179][180][181][182][183][184][185][186]. A case-report study speculated about the relationship between mucopolysaccharidosis I (caused by alpha-L-iduronidase deletions) and the onset of meningiomas in cats
[66].
The role of PR receptors in the pathogenesis of human and animal meningiomas has been already mentioned
[17][44][187][188][189][190][191][192][193]. Other studies reported the important role of matrix metalloproteinases
MMP-2 and
MMP-9 in canine and human meningiomas, where they are involved in extracellular matrix degradation, required for tumor progression and recurrence
[194][195][196][197]. Mandara et al. (2007) evaluated the expression of
MMP-2 and
MMP-9 in canine meningiomas, establishing correlations with
TIMP-1 and
TIMP-2 expression
[198]. The authors found that
TIMP-1 levels were elevated in Grade 1 and Grade 2 meningiomas, but not in Grade 3 cases and hypothesized different pathways in which
TIMP-1 can be involved in the progression of meningiomas
[138]. Cyclooxygenase-2 (
COX-2) is reported to have an impact on canine meningiomas
[199] and is over-expressed in some feline meningiomas
[200]. However, only in humans,
COX-2 seems to be correlated with meningioma grade and local invasion
[113] and is a marker used to classify the tumors
[201]. Vascular Endothelial Growth Factor (
VEGF) has been proposed as a prognostic marker in canine, where it seems to be inversely correlated with survival time
[113], while in human meningiomas, it is a tumor recurrence marker
[202]. The role of E-cadherin, N-cadherin, β-catenin and doublecortin (
DCX) have been evaluated in both canine and human meningiomas. In canine tumors, N-cadherin, β-catenin, and
DCX seem to have a positive correlation with invasion and anaplastic subtypes of meningiomas, instead in human meningiomas, it has not demonstrated a correlation and this role is still debated
[99][113][203][204]. Alterations of
NF-2 in chromosome 22 do not seem to play an important role in the onset of canine meningiomas
[147], as opposed to human meningiomas, where they have been reported to predispose to the development of the tumors
[203].
Glucose transporter (Glut-1) is expressed at high levels in malignant meningiomas both in dogs and humans
[113]. Boozer et al. (2012) evaluated the immunological infiltrate in canine meningiomas
[205], reporting a prevalence of CD18+ microglia and macrophages surrounding and infiltrating the tumors; CD11d+ cells were also present. Lymphocyte infiltrate included mainly CD3+ T-cells and a sparse number of CD79a+ B-cells. In human meningiomas, similar T-cell and B-cell infiltrate have been found, but in both cases, the biological role of this infiltrate was not determined
[206][207][208][209]. The proteins altered in human and dog meningiomas are reported in
Table 6.
Table 6. Proteins frequently altered in human and dog Meningiomas.