Molecular Aspects of Mucoepidermoid Carcinoma and Adenoid Cystic Carcinoma of the Salivary Gland

Abstract

Background

Salivary gland tumors (SGTs) are rare and highly heterogeneous lesions, making diagnosis a challenging activity. In addition, the small number of studies and samples evaluated difficults the determination of prognosis and diagnosis. Despite the solid advances achieved by research, there is still an intense need to investigate biomarkers for diagnosis, prognosis and that explain the evolution and progression of SGTs.

Methods

We performed a comprehensive literature review of the molecular alterations focusing on the most frequent malignant SGTs: mucoepidermoid carcinoma and adenoid cystic carcinoma.

Results

Due to the importance of biomarkers in the tumorigenenic process, this review aimed to address the mechanisms involved and to describe molecular and biomarker pathways to better understand some aspects of the pathophysiology of salivary gland tumorigenesis.

Conclusions

Molecular analysis is essential not only to improve the diagnosis and prognosis of the tumors but also to identify novel driver pathways in the precision medicine scenario.

Introduction

The salivary glands are exocrine glands specialized in the production and secretion of saliva [1]. The glands are formed by a complex system of ducts and acini and are surrounded by a mesenchymal stroma (Fig. 1) [2, 3].

Fig. 1.

Schematic representation of salivary gland structure. The acinar (excretory, striated, and intercalated ducts) and ductal structures are depicted. Myoepithelial cells surround the acinar portion and the intercalated duct

Salivary gland tumors are infrequent but encompass a relevant group of head and neck neoplasms [4, 5]. Their behavior and histology are peculiar, with complex clinicopathological features and diverse biological aspects; these characteristics represent some of the challenges in establishing clinical models [4].

In this review, we will address the most frequent malignant tumors - mucoepidermoid carcinoma (MEC) and adenoid cystic carcinoma (ACC) - the possible mechanisms involved in their development and progression, and potential biomarkers for diagnosis, prognosis, and prediction of response to specific targeted therapies.

Epidemiology

Salivary gland tumors comprise a heterogeneous group of benign and malignant lesions, that account for approximately 3% of head and neck neoplasms [4, 6]. The estimated global incidence ranges from 0.4 to 13.5 cases per 100,000 population per year in both genders and all age groups [4, 6, 7]. Benign salivary gland tumors account for 54–79% of cases, with pleomorphic adenoma and Warthin’s tumor being the most commonly observed histological types in clinical practice. Regarding malignant tumors, which account for 21–46% of the cases, the most commonly reported subtypes are mucoepidermoid carcinoma, adenoid cystic carcinoma, polymorphous adenocarcinoma, acinic cell carcinoma, and salivary duct carcinoma. Approximately 64–80% of salivary neoplasms occur in the parotid glands, 7–11% in the submandibular glands, and less than 1% in the sublingual glands. Minor intraoral salivary glands are also affected, but the incidence in these glands is variable [4, 8].

Salivary gland tumors usually occur in adults, with a mean age of 46.5 years and a similar gender distribution that varies with geographic incidence and histological type, although some authors report that the incidence is slightly higher in women than in men [8–10].

Tumor Behavior and Management

Several characteristics such as histological grade, tumor location, histological type, and invasion affect the overall prognosis of patients, tumor recurrence, and survival [11, 12]. Both mucoepidermoid carcinoma and adenoid cystic carcinoma tend to metastasize to nodal and distant organs presenting some symptoms such as facial paresis and pain, even when it is initial and small [13–15]. High-grade transformation (HGT) has been reported in salivary gland tumors. In adenoid cystic carcinoma, HGT is associated with abundant cytoplasm, necrosis, high mitotic rate, and loss of the biphasic ductal-myoepithelial differentiation. MEC with HGT presented solid features, necrosis, and a high mitotic rate, but no glandular or cystic components or mucous or squamous cells were present [16, 17].

Currently, the main therapeutic approach for SGTs is surgical resection, usually combined with adjuvant radiotherapy [18, 19]. Chemotherapy is indicated for metastatic disease, although the efficacy and response rates remain controversial [20]. In addition, due to the low incidence of these tumors, the studies are based on small case series, making it difficult to evaluate new therapeutic approaches. Cavalieri et al. 2020 [21] reported the development of a nomogram as a prognostic model for adenoid cystic carcinoma, that may aid clinicians in decision making and determined that gender, disease-free survival, and site of metastasis are independent prognostic factors. The use of molecular and immunohistochemical markers can complement information on tumor behavior and aggressiveness, which may benefit the indication of targeted therapeutic approaches [12, 22, 23].

Clinical and Histological Features

Mucoepidermoid carcinoma accounts for approximately 10–15% of salivary gland tumors and 30% of all salivary gland malignancies. It is composed of mucous, intermediate, and squamous (epidermoid) cells. In low-grade tumors, mucous cells, and cystic structures predominate, whereas in high-grade tumors, epidermoid cells are the major tumor component, and lesions tend to be solid (Fig. 2). Other high-grade features include perineural infiltration and necrotic areas in mucoepidermoid carcinoma [14, 24, 25].

Fig. 2.

Histological aspects of mucoepidermoid carcinoma and adenoid cystic carcinoma. A - C: mucoepidermoid carcinoma: in A, panoramic view of low-grade mucoepidermoid carcinoma composed of mucous spaces surrounded by mucous, intermediate, and squamous cells. In B and C: details of mucous and intermediate cells. Original magnification x40, x400 and x400, respectively. D - F: adenoid cystic carcinoma: in D and E: adenoid cystic carcinoma cribriform islands composed of basaloid cells and pseudocystic spaces. In F, note tumoral infiltration in nerve bundles. Original magnification x40, x250 and x250, respectively

Adenoid cystic carcinoma accounts for 1% of all malignant head and neck tumors and 10% of salivary gland tumors, with a reported annual incidence of 3-4.5 cases per million population [26] It is composed of epithelial (ductal) and myoepithelial cells, forming tubular, cribriform, and solid patterns (Fig. 2). It shows a propensity for early invasion of peripheral nerves or blood vessels, a high incidence of local recurrence, and distant metastases (mainly lungs and bone). Recurrent tumors are often considered [26, 27].

Biomarkers

Biomarkers are cell or tissue representative components (proteins, genes, non-coding RNAs) that can identify or indicate a specific characteristic of the process under investigation (histogenesis, molecular characteristics, and mutations, among others) [28]. In oncology, they are considered in many categories, as they are employed as indicators of mutational susceptibility, prognostic factors, and disease aggressiveness, as well as to determine neoplastic differentiation and immunologic sensitivity or resistance to therapies such as chemotherapy, radiotherapy, or immunotherapy [29]. These biomarkers can be identified through several techniques such as immunohistochemistry, PCR, in situ hybridization, and DNA or RNA sequencing [23].

One of the most routinely used methodologies is immunohistochemistry (IHC). It is a surgical pathology laboratory technique that allows the characterization of histological subtypes by observing specific antigens present in the tissue. This technique has been widely used due to its reliable results in terms of sensitivity, specificity, reproducibility, and overall cost-effectiveness, despite its limitations. It is used in pathological diagnosis as an aid for morphological evaluation [30–34]. There are some immunohistochemical markers used for histological characterization of salivary gland tumors [30, 33, 34]. Cytokeratins for identification of basal-luminal cells and differentiation of metastatic cases [13, 22, 35], p63 as a marker of myoepithelial differentiation [36], Ki-67 as a marker of cell proliferation and often associated with tumor aggressiveness [37, 38], vimentin for identification of epithelial-mesenchymal transition [39, 40].

In addition to pathological review and immunohistochemical expression mapping, genomic profiling using more complex technologies, such as Next Generation Sequencing (NGS), are essential to improve diagnostic differentiation of the tumors and identify novel driver pathways to determine the most appropriate therapeutic approach for salivary gland tumors [41]. This approach is particularly beneficial for recurrent and metastatic tumors [19, 30, 42]. Gene rearrangements might be used as diagnostic, prognostic, and predictive molecular markers helping to stratify the tumors. Recently, additional methods such as sequencing, microarray, and FISH analysis have been applied to identify these rearrangements [31, 43–45].

Epigenetic alterations might also be responsible for the development and progression of salivary gland tumors. Epigenetic mechanisms are defined as a set of alterations in DNA that do not alter the sequence of nucleotide bases but alter the expression of proteins through DNA methylation (associated with gene repression and activation by DNA methyltransferases), alterations in the expression of coding RNAs, including microRNAs (miRNAs), and histone structural modifications [46].

Many studies have reported some consistent genetic alterations in salivary gland tumors [18, 29, 31, 47]. In Tables 1 and 2 we summarize the alterations reported in adenoid cystic carcinoma and mucoepidermoid carcinoma and discuss them in the following topics.

Table 1.

Summary of molecular alterations and potential immunohistochemical biomarkers detected in adenoid cystic carcinoma

Biomarkers	Description	Alteration	Potential role	Reference
AXL	Gene encoding a receptor tyrosine kinase that transduces signals from the extracellular matrix to the cytoplasm	Overexpression	Prognostic/ Diagnostic	FERRAROTTO et al. [43] HUMTSOE et al. [62]
BCL2	Gene encoding the apoptosis regulatory protein BCL 2	Overexpression	Prognostic/ Treatment	JIANG et al. [68]; NOR et al. [69]
CD117/c-Kit	Gene encoding receptor tyrosine kinase and considered a proto-oncogene	Overexpression	Prognostic	TANG et al. [51]; PHUCHAREON et al. [52]; HOU et al. [53]
Cytokeratins	Epithelial cell identification protein, present in the cytoskeleton	Overexpression and low expression.	Differentiation	NAGANO et al. [13]; NAMBOODIRIPAD [22]; GAO et al. [35]
EGFR	Transmembrane glycoprotein encoder members of the kin superfamily that is an epidermal growth receptor	Overexpression	Prognostic/ Diagnostic	FERRAROTTO et al. [43]; GUAZZO et al. [74];WANG et al. [75]; WANG et al. [76]
HCN2	Hyperpolarization-activated cyclic nucleotide-activated	Hypomethylation	Diagnostic and Prognostic	LING et al. [79]
Histone	Histone modifications (Acetylations or methylations)	Overexpression	Prognostic	XIA et al. [77]
Ki-67	Proliferating cell identification protein	Overexpression	Prognostic	BAGULKAR et al. [38]; SKALOVA et al. [16]; HELLQUIST et al. [17]; FREIBERGER et al. [54]
mTOR	Encodes protein kinases that mediate cellular stress response	Overexpression	Treatment	MOORE et al. [41]; YU et al. [66]; CHAE et al. [67]
MYB/NFIB	Transcription factor considered a proto-oncogene, being important for the regulation of hematopoiesis	Translocation	Prognostic	YAN et al. [63]; PERSSON et al. [58]; MITANI et al. [82]
MYC	Protein-coding proto-oncogene associated with cell cycle progression	Overexpression	Prognostic	FERRAROTTO et al. [43]; FUJII et al. [64]; ANDERSSON et al. [109]
NOTCH1	NOTCH pathway-related transmembrane receptor encoder	Overexpression	Prognostic	FERRAROTTO et al. [43]; LORINI et al. [56]; FERRAROTTO et al. [59]; EVEN et al. [44]; FERRAROTTO et al. [60]; SU et al. [110]; SAJED et al. [111]
p27	Inhibitor of cyclin-dependent kinases	Hypermethylation	Diagnostic and Prognostic	DAA et al. [78]
p53	Protein associated with cell cycle regulation and had tumor suppression action	Overexpression and low expression.	Prognostic	LI et al. [70]; NOR et al. [69]
PTEN	Tumor suppressor gene and is part of the signaling pathway PI3K/AKT/mTOR	Low expression	Prognostic	ETTL et al. [72]; CAO et al. [78]
RET	Receptor of tyrosine kinase	Overexpression	Prognostic	MOORE et al. [41]
Vimentin	Mesenchymal cell identification protein	Overexpression and low expression.	Differentiation	WARNER et al. [39]; BELULESCU et al. [40]; FUKAI et al. [55]
miR-150	MicroRNAs involved in post-transcriptional regulation of gene expression	Low expression	Diagnostic and Prognostic	BROWN et al. [80]
miR-155	MicroRNAs involved in post-transcriptional regulation of gene expression	Overexpression and low expression.	Prognostic	KERCHE et al. [83]; LIU et al. [84]; KUMAR et al. [85]
miR-375	MicroRNAs involved in post-transcriptional regulation of gene expression	Low expression	Diagnostic and Prognostic	BROWN et al. [80]
miR-455	MicroRNAs involved in post-transcriptional regulation of gene expression	Overexpression	Diagnostic	BROWN et al. [80]
miR-1180	MicroRNAs involved in post-transcriptional regulation of gene expression	Overexpression	Prognostic	ANDREASEN et al. [81]
miR-4676	MicroRNAs involved in post-transcriptional regulation of gene expression	Overexpression	Prognostic	ANDREASEN et al. [81]
miR-6835	MicroRNAs involved in post-transcriptional regulation of gene expression	Overexpression	Prognostic	ANDREASEN et al. [80]

Table 2.

Summary of molecular alterations and potential immunohistochemical biomarkers detected in mucoepidermoid carcinoma

Biomarkers	Description	Alteration	Potential role	Reference
BCL2	Gene encoding the apoptosis regulatory protein BCL 2	Overexpression	Prognostic/ Treatment	JIANG et al. [68]; XU et al. [102]; NOR et al. [69]; FLORES et al. [101]; CHOI et al. [103]
CRTC1/ MAML2	Regulatory transcription coactivation gene in the SIK/TORC signaling pathway	Translocation	Prognostic	YAN et al. [63]; MORITA et al. [91]; CHEN et al. [92]; JEE et al. [112]
Cytokeratins	Epithelial cell identification protein, present in the cytoskeleton	Overexpression and low expression	Differentiation	NAMBOODIRIPAD [22]; MEYER et al. [23]
EGFR	Epidermal growth factor receptor	Overexpression	Prognostic/ Treatment	KANG et al. [98]; LUJAN et al. [99]
HER2	Oncogene associated with tumor progression when dysregulated	Overexpression	Prognostic	CORRÊA et al. [96]; YOSHIMURA et al. [95]; FIRWANA et al. [114]
Ki-67	Proliferating cell identification protein	Overexpression	Prognostic	BAGULKAR et al. [38] Da Silva et al. [105]
LINC00473	Specific long non-coding RNA (lncRNA) mediator of the oncoprotein CRTC1- MAML2	Overexpression	Diagnostic and Prognostic	CHEN et al. [93]; NAAKKA et al. [94]
miR-155	MicroRNA involved in post-transcriptional regulation of gene expression	Overexpression	Prognostic	KERCHE et al. [83]
miR-20	MicroRNA involved in post-transcriptional regulation of gene expression	Low expression	Prognostic	FLORES et al. [101]
miR-205-5p	MicroRNA involved in post-transcriptional regulation of gene expression	Overexpression	Prognostic	BINMADI et al. [100]
miR-22	MicroRNA involved in post-transcriptional regulation of gene expression	Overexpression	Prognostic	NAAKKA et al. [94]
miR-302a	MicroRNA involved in post-transcriptional regulation of gene expression	Overexpression	Prognostic	BINMADI et al. [100]
miR-34a	MicroRNA involved in post-transcriptional regulation of gene expression	Overexpression	Prognostic	KERCHE et al. [83];FLORES et al. [101]
p14	Inactivate the mechanisms of DNA repair in the cell cycle	Hypermethylation	Diagnostic and Prognostic	NIKOLIC et al. [97]
p53/TP53	Gene and protein that acts as a tumor suppressor	Overexpression and Low expression	Prognostic	LI et al. [70]; NOR et al. [69]; MORITA et al. [91]; KANG et al. [98]; FLORES et al. [101]; ANDREWS et al. [113]
PDL-1	Programmed cell death ligand 1	Overexpression	Prognostic/ Treatment	HARADA et al. [87]
PTEN	Tumor suppressor gene and is part of the signaling pathway PI3K/AKT/mTOR	Low expression	Prognostic	ETTL et al. [72]; CAO et al. [73]; BOU et al. [104]

Molecular Alterations in Adenoid Cystic Carcinoma

Different molecular pathways, such as PI3K/AKT/mTOR, NOTCH, TP53, and HER2 have been associated with adenoid cystic carcinoma by Moore et al. (2020) [41]. CD117/c-kit is overexpressed in adenoid cystic carcinoma [48–50] and is considered a prognostic factor for salivary and non-salivary sites as it is associated with aggressive cases, perineural invasion, distant metastasis, and worse survival rates [51–53]. Ki-67 expression was positive in high-grade adenoid cystic carcinoma as it is associated with cell proliferation [16, 17, 38]. Freiberger et al. (2021) [54] described an adenoid cystic carcinoma with an unusual morphology, a spindle cell, and pseudo angiomatoid pattern, which presented diffuse p40 expression, absence of CD117, and focal expression of Ki-67.

Cytokeratin (CK) expressions such as AE1/3, 34betaE12, CK5/6, CK7, CK14, and CK18 are consistently positive in adenoid cystic carcinoma [13, 22]. Gao et al. (2017) [35] demonstrated that CK14 overexpression was associated with high-grade, advanced stage, and invasion in adenoid cystic carcinoma. Nagano et al. (2016) [13] reported loss of CK expression when comparing metastatic lesions and the primary tumor from ACC patients.

Vimentin expression is associated with the acquisition of mesenchymal phenotype, suggesting the involvement of the epithelial-mesenchymal transition (EMT) in the tumorigenesis and progression of adenoid cystic carcinoma, as it confers the ability of migration, invasion, and metastatic properties [39, 40, 55].

NOTCH1 is expressed in 25% of metastatic tissues and has been evaluated as a therapeutic target since mutations have been frequently observed in solid tumors and more advanced, recurrent, or metastatic ACC [43, 56]. Its overexpression is associated with the promotion of cell growth, migration, and invasion, with NOTCH inhibitors being able to block these pathways in vitro and tumorigenicity in vivo by inducing apoptosis.

Ho et al. (2019) [57] performed an integrated genomic analysis comparing primary, recurrent, and metastatic tumors from different anatomic sites (head and neck/salivary, breast, and lung) to identify alterations and classify tumor subtypes. Recurrent and metastatic adenoid cystic carcinomas were found to be enriched for NOTCH alterations, ARID1B, TERT mutations, and MYB-MYBL1 fusions. MYB-MBL1 fusions are the major oncogenic driver in the tumorigenesis process of ACC and are considered as prognostic biomarker for short overall survival (OS) rates. The presence of this fusion leads to deregulation of transcriptional activation domains [58].

It has been shown that NOTCH signaling and chromatin remodeling genes, such as the ARID gene family, act synergistically during adenoid cystic carcinoma tumorigenesis, whereas TERT mutations represent an alternative pathogenic mechanism independent of the MYB/MYBL1 translocation. Key alterations involving MYB, NOTCH, and TERT defined distinct molecular subgroups with different overall survival rates, in particular, MYB^wt/NOTCH1⁺ and MYB⁺/NOTCH1⁺ status conferred the worst prognosis. From this analysis, we can understand the genomic progression through the presence of pathogenic mutations that facilitate recurrence and metastasis in these tumors and help to guide therapeutic strategies. Currently, the identification of activating NOTCH1 mutations is used for prognostic purposes, as they are associated with high rates of distant metastasis in salivary and non-salivary adenoid cystic carcinoma [59]. Clinical trials targeting the NOTCH pathway are ongoing. Even et al. (2020) [44] investigating a small molecule inhibitor of NOTCH observed no objective response, in contrast to the phase I trial of the monoclonal antibody brontictuzumab, which demonstrated a 17% clinical benefit and efficacy in the ACC subset with Notch1 pathway activation [60].

Karpinets et al. (2021) [61] identified that, in adenoid cystic carcinoma, loss of chromosome 12q was dominant in tumors positive for MYB or MYBL1 fusion, and tumors negative for these fusions harbor NOTCH mutations. Humtsoe et al. (2022) [62] used the ACC-01 cell line, carrying the MYB/NFIB translocation, to verify which pathways are altered and to evaluate effective drugs for targeted therapy. The presence of this fusion was also discussed in the review of Yan et al. (2018) [63]. For example, MYC is one of the genes that can be affected by the translocation present in adenoid cystic carcinoma and that is observed in a subset of cases [56]. In the study carried out by Fujiiet al. (2017) [64] MYC overexpression was associated with shorter disease-free survival of patients. This overexpression might occur by direct amplification or be influenced by NOTCH mutations. Shibata et al. (2021) [65] showed that fusions of alternative genes to NFIB, such as EPB41L2, MAP7, MCMDC2, C8orf34, and CASC20, could promote MYB overexpression, suggesting the possibility of future therapeutic targets.

The mTOR pathway has been implicated in the evasion of apoptosis and tumor growth in adenoid cystic carcinoma. It has been investigated as a potential therapeutic target as observed in the studies of Yu et al. (2014) and Chae et al. (2015) [66, 67]. Alterations in BCL2 expression have been described in adenoid cystic carcinoma, leading to tumor progression, with this protein being considered a possible therapeutic target [68, 69]. Inhibition of the MDM2-p53 interaction prevents recurrence of these tumors. Murine double minute 2 (MDM2) interacts with p53 leading to the degradation of 26S proteasomal pathways and consequently maintaining low cellular levels of p53, which is involved in cell cycle control and DNA repair. MI773 is a small molecule inhibitor of MDM2 and in combination with cisplatin leads to reactivation of p53 and induction of tumor regression [69].

A meta-analysis demonstrated the prognostic significance of p53 immunohistochemical expression in adenoid cystic carcinoma, with higher expression observed in the solid type of adenoid cystic carcinoma [70].

Overexpression of AXL may be associated with resistance to chemotherapy and immunotherapy, although it has been described in less aggressive adenoid cystic carcinoma [43, 54, 71]. Low PTEN expression is observed in adenoid cystic carcinoma tumors with more aggressive and high-grade behavior and worse disease-free survival, as observed by Ettl et al. (2012) and Cao et al. (2019) [72, 73]. PTEN loss was associated with increased EGFR and HER2 expression.

Overexpression of tyrosine kinase receptors such as EGFR is commonly observed in tumors. In adenoid cystic carcinoma, its expression can be observed in more aggressive tumors. Regarding the use of EGFR inhibitory therapies, the efficacy is still under debate, as discussed by Ferrarotto et al. (2021), Guazzo et al. (2021), Wang et al. (2015), and Wang et al. (2018) [43, 74–76].

Moore et al. (2020) [41] presented the first report on the treatment of advanced adenoid cystic carcinoma with anti-RET therapy (TRIM33-RET fusion), achieving disease stability (no progression) for 14 months. The authors suggest that tumor genome sequencing should be used to guide clinical decisions, as it could benefit patients with recurrent and metastatic salivary gland tumors and increase patient survival.

In adenoid cystic carcinoma, histone H3 lysine 9 trimethylation (H3K9me3) was associated with worse OS outcomes whereas histone lysine 9 acetylation (H3K9Ac) was associated with significantly better OS. H3K9me3 and H3K9Ac expressions were correlated with histologic patterns and the presence of distant metastasis [77].

Hypermethylation of p27, a specific inhibitor of cyclin-dependent kinases (CDKs), may lead to its downregulation and contribute to cell cycle alteration in adenoid cystic carcinoma tumorigenesis [78]. HCN2 (Hyperpolarization-activated cyclic nucleotide-activated) hypomethylation has also been reported in adenoid cystic carcinoma xenografts [79].

Some studies have described alterations in microRNA expression in salivary gland tumors. In adenoid cystic carcinoma, the expression of miRNA-455-3p, miR-375, and miR-150 was significantly decreased compared to normal tissues. Furthermore, miRNA-455-3p expression was increased in adenoid cystic carcinoma when compared to polymorphous adenocarcinoma, suggesting that it could be used as a complementary tool in the diagnosis of challenging cases [80]. Increased expression of miR-6835-3p, miR-4676, and miR-1180 was associated with decreased overall survival (OS) and recurrence-free survival (RFS) rates [81], and increased expression of miR-20a and miR-17 was associated with poor outcomes [82] miR-155 downregulation and overexpression are observed in adenoid cystic carcinoma [83, 84]. Other authors have reported that miR-155 may play a role in oncogenesis and as a tumor suppressor in different tumors as described in the review of Kumar et al. (2023) [85].

The expression of specific surface proteins (PD-1/PD-L1 receptor) is often used by neoplastic cells to evade the immune response and might be used as a therapeutic route to reactivate the immune cell response [86–88]. Mosconi et al. (2019) and Nakano et al. (2019) [89, 90] demonstrated a low expression of programmed cell death-ligand 1 (PD-L1) but an increase of PD-L2 in adenoid cystic carcinoma. Adenoid cystic carcinoma microenvironment has low immunogenicity, which might be associated with poor prognosis of these tumors.

Molecular Alterations in Mucoepidermoid Carcinoma

CRTC1-MAML2 or CRTC3-MAML2 fusions are associated with better prognosis and survival in mucoepidermoid carcinoma and are observed in more than 50% of the cases [91, 92].

A specific long non-coding RNA (lncRNA) (LINC00473) acts as an essential mediator of CRTC1-MAML2 oncoprotein leading to cell growth and survival in mucoepidermoid carcinoma. lncRNA LINC00473 is overexpressed by activation of the CRTC/CREB transcriptional program and serves as a promising biomarker and therapeutic target [93, 94]. Furthermore, large-scale molecular mapping revealed that tumors with CRTC1-MAML2 fusion have frequent BAP1 mutations, and tumors without this fusion have high LRFN1 mutation rates. Thus, these alterations are associated with tumor phenotype and progression and may serve as biomarkers [61]. Negative CRTC1-MAML2 tumors also presented TP53 mutation [91] which is a common genomic alteration in mucoepidermoid carcinoma, in addition to PI3K/AKT/mTOR, NOTCH, and HER2 pathways. HER2 overexpression is correlated with a worse prognosis, although it can be used as a therapeutic target as suggested by Yoshimura et al. (2021) and Corrêa et al. (2018) [95, 96]. p14 gene hypermethylation, which is crucial for the regulation of p53 activity, was reported as a rare epigenetic phenomenon in the development of mucoepidermoid carcinoma by Nikolic et al. (2018) [97].

Kang et al. (2017) [98] and Lujan et al. (2010) [99] identified a high frequency of epidermal growth factor receptor (EGFR) mutation in intermediate or high-grade mucoepidermoid carcinoma tumors, which showed significantly shorter disease-free intervals and overall survival. However, the cases with EGFR gene gain demonstrated better outcomes and chemotherapy response.

Expression of cytokeratins such as CK8, CK14, CK17, and CK19 is observed in different components of mucoepidermoid carcinoma, and the use of these markers helps in more accurate and effective diagnosis and classification of SGTs [22, 23].

Different microRNAs have been described as important modulators of cellular functions and have been associated with the development/progression of mucoepidermoid carcinoma. Binmadi et al. (2018) [100] reported increased expression of miR-302a, which was associated with in vitro cellular invasion. Downregulation of miR-20a was associated with apoptotic deregulation in mucoepidermoid carcinoma [101]. Naakka et al. (2022) [94] observed that miR-205-5p and miR-22 upregulation was associated with increased cellular viability, migration, and invasion and consequently with worse overall survival rates, and Kerche et al. (2022) [83] identified that upregulation of miR-34a was associated with better overall survival in mucoepidermoid carcinoma patients. miR-34a can regulate c-Kit, a tyrosine kinase involved in regulation of the developmental processes, and β-catenin which is involved in cell adhesion and gene expression. Furthemore, Kerche et al. (2022) [83] identified that miR-155 is associated with low-grade mucoepidermoid carcinoma, but in adenoid cystic carcinoma, is associated with poor prognosis and low overall survival.

Harada et al. (2018) [87] demonstrated that PD-L1 expression in mucoepidermoid carcinoma is correlated with tumor aggressiveness; higher stage and worse prognosis due to immune-suppression.

Xu et al. (2013) and Choi et al. (2021) [102, 103] investigated the use of traditional Chinese medicine substances in a mucoepidermoid cell line (MC3 demonstrating the induction of apoptosis through the regulation of BCL-2 family proteins and suggesting a potential role for these substances as cancer therapeutics. B-cell Lymphoma 2 (BCL2) is a pro-apoptotic family associated with tumorigenesis and chemotherapy resistance by protecting cells from apoptotic mechanisms. Therefore, studies with anti-apoptotic BCL2 may provide a potential therapeutic strategy [102, 103].

Phosphatase and tensin homolog (PTEN) is associated with mTOR pathways and cell proliferation regulation, although genetic alterations are observed at a low incidence in mucoepidermoid carcinoma [104].

Immunoexpression of Ki-67, a protein related to cell proliferation and aggressiveness, is associated with lymph node metastasis and poor overall survival rates in mucoepidermoid carcinoma patients, suggesting a prognostic role [105].

Potential Therapeutic Targets

The development of novel and effective therapies for salivary gland tumors is a paramount issue since the current options are still quite limited [32, 106, 107]. Different signaling pathways are involved in the tumorigenic process and understanding the molecular mechanisms is necessary to identify biomarkers and targets for tailored treatments.

Precision medicine allows for the development of personalized therapies, which individualize the best treatment for each patient. In this direction, the article published by Kurzrocket et al. (2020) [108] highlights the importance of this approach for more advanced cases, including metastatic ones for which there are currently few treatment options. In this study, 63% of patients with advanced SGC demonstrated objective response rates (ORR) using a combination of therapies based on the genomic characteristics of their tumors.

Regarding the mTOR pathway, Yu et al. (2014) [66] demonstrated that the use of rapamycin can reduce tumor growth and decrease EGFR activity, suggesting that mTOR inhibitors may be a potential candidate for the treatment of adenoid cystic carcinoma.

Tang et al. (2014) [51] suggested that targeting c-kit may be a promising therapeutic strategy for adenoid cystic carcinoma metastasis, as treatment with Gleevec – the inhibitor of c-kit – can modulate the migration and invasion of adenoid cystic carcinoma cells.

Andersson et al. (2017) [109] demonstrated that pharmacological inhibition of IGF1R leads to downregulation of MYB-NFIB expression and MYC regulation, decreasing the proliferation of salivary and non-salivary adenoid cystic carcinoma cells.

Therapeutic inhibition of MDM2-p53 interaction by the small molecule inhibitor MI-773 can sensitize adenoid cystic carcinoma xenograft tumors to cisplatin, promoting p53 activation, induction of apoptosis, and regression of adenoid cystic carcinoma patient-derived xenografts (PDX) tumors [69]. Li et al. (2017) [70] demonstrated that adenoid cystic carcinoma cases with p53 overexpression can be treated with paclitaxel.

Recent research suggests that the use of P. kaempferi extract (EEPK) - a traditional Chinese natural product with cytotoxic effects - inhibited the growth of mucoepidermoid carcinoma cells and stimulated the induction of caspase-mediated apoptosis [103].

Some trials of targeted therapies that are already in clinical development for salivary gland tumors are summarized in Table 3 (https://clinicaltrials.gov/).

Table 3.

Current clinical trials for ACC and MEC based on targeted therapies

Drugs	Target	Tumor	Phase	References
AL101	Inhibitor of gamma secretase-mediated Notch signaling	Recurrent and metastatic ACC	II	FERRAROTO et al. [115]
Avelumab, Axitinib	Inhibitor of enzymes needed for cell growth	Progressive, recurrent, and metastatic ACC	II	FERRAROTO et al. [116]
CB-103	Inhibitor of Notch signaling	Locally advanced or metastatic ACC	I/II	MIRANDA et al. [117]
PRT543	Inhibitor of PRMT5	Advanced ACC	I	MISHRA et al. [118]
VMD-928	Inhibitor of Trk	Advanced ACC	I	CHUNG et al. [119]
A166	Anti-HER2 monotherapy	Her2-positive MEC	I/II	HU et al. [120]

PRMT5: Protein arginine methyltransferase 5; Trk: tyrosine kinases

Conclusion and Future Perspectives

The etiopathogenesis of many salivary gland tumors is still unknown and there are no validated biomarkers for specific diagnosis/classification of salivary gland tumors. Immunohistochemistry remains one of the most used approaches in daily clinical practice. However, molecular analysis is essential to improve the diagnostic differentiation of the tumors and to identify novel driver pathways in the precision medicine scenario.

Abbreviations

MEC

Mucoepidermoid Carcinoma

ACC

Adenoid cystic carcinoma

HGT

High-grade transformation

IHC

Immunohistochemistry

NGS

Next Generation Sequencing

SDC

Salivary Duct carcinoma

miRNAs

MicroRNAs

CK

Cytokeratins

CDKs

Cyclin-dependent kinases

EMT

Epithelial-Mesenchymal Transition

MDM2

Murine double minute 2

PD-L1

Programmed cell death - ligand1

EGFR

Epidermal growth factor receptor

HCN2

hyperpolarization-activated cyclic nucleotide-activated

H3K9me3

histone H3 lysine 9 trimethylation

H3K9ac

histone H3 lysine 9 acetilation

IncRNA

long non-coding RNA

SGC

Salivary gland carcinoma

CASP

Caspases

BCL2

B cell Lymphoma 2

PTEN

Phosphatase and tensin homologue

OS

overall survival

ORR

objective response rates

RFS

recurrence free survival

PDX

patient derived xenograft

Author Contributions

R.F.C. performed the literature search and wrote the main manuscript text; C.A.O. performed the literature search and wrote the main manuscript text; A.N.M.G prepared the figures and critically revised the work; S.V.L. and C.M.C.C. prepared the figures, wrote the main manuscript and critically revised the work. All authors read and approved the final manuscript.

Funding

The São Paulo Research Foundation (FAPESP), grant number 14/50943-1 and the Brazilian National Research Council (CNPq), grant number 465682/2014-6, funded this research. The authors were supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) – Finance Code 001 (RFC, CAO, ANMG).

Declarations

Ethical Approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Consent for Publication

For this type of study consent for publication is not required.

Informed Consent

For this type of study informed consent is not required.

Competing Interests

The authors declare no competing interests.