The role of microRNA expression in malignant transformation of pleomorphic adenoma: comparing carcinoma ex pleomorphic adenoma vs pleomorphic adenoma

doi:10.21203/rs.3.rs-1504827/v1

Purpose

Carcinoma ex pleomorphic adenoma (CXPA) is a rare malignant tumor which transformed from benign PA. Here we identify the characteristic miRNAs from both the PA (CPA) and carcinomatous (CA) portions of CXPA in an effort to understand their malignant transformation and then compare their expression in CPA and benign PA (BPA) samples to identify candidate miRNAs for their differential diagnosis.

Methods

We selected 13 CXPA FFPE and 16 BPA samples and then separated the CPA and CA portions of the CXPA FFPE tissues before completing total RNA extraction and miRNA profiling for each group.

Results

Downstream evaluation of these miRNAs identified 13 transcripts that were differentially expressed between the CPA and CA portions of CXPA. A total of eight of these miRNAs were up-regulated and five down-regulated in CA, where the up-regulated miRNAs were related to cancer progression and the down-regulated miRNAs were associated with tumor suppression. In addition, seven miRNAs were significantly up-regulated in CPA when compared to BPA, even though they present with the same histopathology. These transcripts also all demonstrated a clear relationship with cancer when evaluated using gene ontology and KEGG pathway tools, with almost all of these transcripts interacting with TP53.

Conclusion

We identified several differentially expressed miRNAs in the CPA and CA portions of CXPA, and confirmed that they were closely associated with TP53 and various cancer-related pathways using target gene and pathway analyses. In addition, we also identified several differentially expressed miRNAs in CPA and BPA, which are also associated with carcinogenesis and may serve as potential biomarkers for differential diagnosis.

Carcinoma ex pleomorphic adenoma

pleomorphic adenoma

microRNA

malignant transformation

TP53

Carcinoma ex pleomorphic adenoma (CXPA) is a rare malignant neoplasm of the major salivary glands, accounting for about 3–15% of all salivary gland malignancies (Lüers et al. 2009; Chen et al. 2014). In addition, CXPA has been shown to exhibit a rising incidence rate making it a growing concern for public health policy (Gupta et al, 2019). CXPA is the result of malignant transformation of the most common benign salivary gland tumor—pleomorphic adenoma (PA)—into various malignant histologies including salivary duct carcinoma (SDC), adenoid cystic carcinoma (AdCC), mucoepidermoid carcinoma (MEC), and myoepithelial carcinoma (MEC), amongst others (Lewis et al. 2001). This transformation generally occurs after the PA has been present for many years, usually decades. The prognosis of salivary malignancies evolving from PA tends to be worse than that of de novo salivary gland malignancies, with a reported 5-year overall survival-rate of as little as 25% (Antony et al. 2012). Therefore, the diagnosis of the carcinomatous portion of PA is critical in determining the appropriate treatment modality and timing. However, accurate diagnosis of CXPA via biopsy is not easy because it is often difficult to accurately target the carcinomatous portion of these tumors and this diagnosis is also only usually possible when the specimen contains a clear adenoma and carcinomatous portion, which tends to be towards the later stages of progression.

Recent molecular analysis of CXPA has shed new light on the mechanisms of carcinogenesis identifying CXPA specific mutations in TP53, PLAG1, and HMGA2 (Antony et al. 2012). However, for the most part, the landscape of mutational and copy number alterations in CXPA seems to be similar to non-malignant PAs, leaving the molecular processes involved in the transition from PA to CXPA largely unknown (El-Naggar et al. 2000).

MicroRNAs (miRNAs) are short non-coding RNAs, involved in the post-transcriptional regulation of gene expression (Bartel. 2004). Dysregulation of miRNA expression has been extensively reported in various types of tumors, including lung, breast, colon, stomach, and thyroid cancers (Hayes et al. 2014), where it is believed that they play a role in the oncogenic progression or suppression of these cancers. Several studies of salivary gland tumors (SGT) have tried, with varying degrees of success, to identify miRNA signatures for the differential diagnosis of benign and malignant SGTs in an effort to improve preoperative diagnosis using tissue or saliva samples (Denaro et al. 2019).

Here we hypothesized that the presence of the miRNAs involved in malignant transformation of CXPA might help predict the risk of carcinogenesis regardless of histology. Thus, we investigated candidate miRNAs by comparing PA (CPA) and CA samples from CXPA biopsies in an effort to elucidate the pathophysiology of malignant transformation in these cancers. We then compared the miRNA profiles of CPA and benign PA (BPA) and identified several candidate miRNAs which might be helpful for differential diagnosis between CPA and BPA (Fig. 1).

Patients and materials

This study was approved by the Seoul National University Institutional Review Board for application at both the Bundang and Dongtan Sacred Heart Hospitals, Hallym University. This committee also waived the need for written informed consent and was completed as described in Fig. 1. A total of 13 CXPA and 16 BPA patients who underwent surgical resection at one of these two hospitals between 2010 and 2019 were enrolled in this study based on their pathological presentation. Their original hematoxylin-eosin (H & E)–stained slides were reviewed by specialist head and neck pathologist (H.K), and the diagnoses in each case were confirmed following the World Health Organization's Classification of Salivary Gland Tumors (Seethala et al. 2017). All CXPA were classified according to invasiveness (intracapsular, minimally, and widely invasive) and histological subtype.

RNA isolation

Total RNA was extracted from paraffin-embedded tissues using Trizol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’ instructions. CXPA tissues were first deparaffinization with xylene and the tissue sections stained with H & E before the CPA and CA portions were selected and carefully dissected to minimize cross contamination. RNA quality was assessed using an Agilent 2100 bioanalyzer and an RNA 6000 Pico Chip (Agilent Technologies, Amstelveen, The Netherlands), and RNA quantification was performed using a NanoDrop 2000 Spectrophotometer system (Thermo Fisher Scientific, Waltham, MA, USA).

Library preparation and sequencing

We then went on to create sequencing libraries for both the control and test RNAs using the NEBNext Multiplex Small RNA Library Prep kit (New England BioLabs, Inc., USA) according to the manufacturer’ instructions. Briefly, 1 µg of total RNA from each sample was ligated to the adaptors before being transcribed to cDNA using reverse-transcriptase and adaptor-specific primers. Libraries were then amplified by PCR and prepared for sequencing using a QIAquick PCR Purification Kit (Qiagen, Inc, German) and AMPure XP beads (Beckman Coulter, Inc., USA). The yield and size distribution of the small RNA libraries were assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies, Inc., USA) and high-throughput sequences were produced using a NextSeq500 system and single-end 75 sequencing (Illumina, San Diego, CA., USA).

Data analysis

Sequence reads were mapped using bowtie2 software and used to create the final bam file for evaluation (alignment file). Mature miRNA sequences were then used as a reference for mapping and the read counts for each mature miRNA sequence were extracted from the alignment file using bedtools (v2.25.0) and Bioconductor programmed to use R (version 3.2.2; R development Core Team, 2011) as its base. Read counts were then used to determine the expression level of each miRNA and the quantile normalization method was used for between sample comparisons. We then went on to identify the miRNA targets using the miRWalk 2.0 database (http://zmf.umm.uni-heiderlberg.de/apps/zmf/mirwalk2/) (Dweep et al. 2011; Dweep et al. 2015). Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were then applied using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) online tool (http://david.abcc.ncifcrf.gov/) (Dennis et al. 2003; Huang et al. 2009) and specific GO and KEGG pathway terms were identified using a p-value threshold of < 0.01. The protein-protein interaction data were obtained from the STRING database (http://string-db.org/) and visualized using Cytoscape_v3.4.0 software. These analyses were then combined to create the miRNA-target gene networks for these samples using miRNet 2.0 software (https://www.mirnet.ca).

Statistical analysis

All statistical analyses were completed in R (version 3.5.1) using R studio, version 1.1.383. T-test was used to evaluate differential expression between different sample types and a p-value of < 0.05 was considered statistically significant. The R “stats” package was used for all Benjamini-Hochberg corrections.

Clinicopathological data

The mean age of the CXPA group was 53.0 years (ranged 26 to 71 years) while it was only 44.7 years (ranged 28 to 75 years) in the PA group. The mean tumor size was 2.6 cm (range 1.4 to 4.0 cm) in the PA, and 3.5 cm (range 1.2 to 6.3 cm) in the CXPA groups. A total of 4 SDC, 2 epithelial-myoepithelial carcinoma, 2 MEC, 3 adenocarcinoma, 1 clear cell carcinoma, and 1 oncocytic carcinoma presented with a clear CA portion. In addition, grading revealed 5 patients with high-grade, 5 with intermediate grade, and 3 with low-grade carcinomas. In addition, when we evaluated the extent of the invasion in each sample we were able to identify 5 patients with invasive, 3 with minimally invasive, and 5 with non-invasive tumors. All of these details are summarized in Table 1.

Identification of Differently Expressed miRNAs

Pleomorphic adenoma (CPA) versus carcinomatous portion (CA) in carcinoma ex pleomorphic adenoma (CXPA)

We identified 2588 miRNAs and 13 differentially expressed miRNAs in the CA and CPA groups when applying a fold change of > 2 and an adjusted p-value of < 0.05. Table 2 describes the five upregulated (miR-455-3p, miR-140-5p miR-483-5p, miR-125b-5p, and miR-125a-5p) and eight downregulated miRNAs (miR-21-3p, miR-183-5p, miR-182-5p, miR-425-5p, miR-96-5p, miR-200a-3p, miR-181a-3p, and miR-505-3p) in these samples.

Pleomorphic adenoma (BPA) versus carcinoma ex pleomorphic adenoma (CPA/CA)

We identified a total of 17 differentially expressed miRNAs when comparing BPA and CPA/CA samples (fold change > 2, adjusted p value < 0.05). Table 3 describes the ten up-regulated (let-7a-3p, miR-27a-3p, miR-9-5p, miR-135a-5p, miR-135b-5p, miR-455-5p, miR-218-5p, miR-181d-5p, miR-369-3p, and miR-132-5p) and seven down-regulated miRNAs (miR-196a-5p, miR-193a-3p, miR-193b-3p, miR-29c-3p, miR-331-3p, miR-361-3p, and miR-423-5p) identified in these evaluations.

GO functional enrichment and KEGG pathway analyses
Evaluation of these miRNAs in the biological process category identified several target genes with differential expression in CA and CPA, which pulled in “apoptotic process”, “cell cycle arrest”, “negative regulation of transforming growth factor beta receptor signaling pathway”, “regulation of transcription from RNA polymerase II promoter”, and “positive regulation of transcription from RNA polymerase II promoter” as their top five GO terms (Table 4). Similarly, the GO terms for the biological process category for the target genes included “peptidyl-serine phosphorylation”, “positive regulation of transcription, DNA-templated”, “protein phosphorylation”, and “negative regulation of transforming growth factor beta receptor signaling pathway” when comparing the differentially expressed miRNAs from CPA and BPA (Table 4).
Table 5 summarizes the most enriched KEGG pathways for each of the differently expressed miRNAs in both the CA vs CPA, and CPA/CA vs BPA evaluations. Many cancer-related pathways such as “viral carcinogenesis”, “Hippo signaling pathway”, “p53 signaling pathway”, and “proteoglycans in cancer” were amongst the most highly ranked processes in CA vs CPA and interestingly, there was a similar pattern in these processes, including “proteoglycans in cancer”, “ECM-receptor interaction”, Hippo signaling pathway”, “AMPK signaling pathway”, “viral carcinogenesis”, and “TGF-beta signaling pathway”, in the enriched pathways for CPA vs. BPA.

Target gene prediction, Protein-protein interaction (PPI) networks, and miRNA-Hub Gene network analyses

A total of 3082 genes were commonly identified by miRWalk, miRanda, RNA22, and Targetscan evaluations, which was then reduced to 383 genes with multiple identifications. We then used the STRING database to identify the PPIs among these 383 targets and identified those proteins that were most likely to interact with > 10 other proteins and designated these as the hub nodes. Fig. 2A shows the PPIs derived from the 15 hub nodes (CASP9, TP53, SMAD2, MAPK10, VEGFA, CXCL12, GNG12, PRKAR1A, IGF1R, SDC1, GNA01, ADCY1, ADCY6, ADCY9, and CAMK2A) for the differentially expressed miRNAs from CPA (BPA) and CA, with each network finally being constructed using Cytoscape. The most significant revelation from these networks was the identification of TP53 as a central node with more than 33 interactions in these networks.
Fig. 2B shows the PPIs for the 25 hub nodes (PML, TP53, MAPK1, MAPK14, AKT2, AKT3, GSK3B, TNS1, PDPK1, BRAF, YWHAB, ADCY1, ADCY2, CAMK2A, CALM3, PRKCA, PRKCE, RPS6KB1, TNS1, IGF1, EGFR, PTPRJ, DNAJC5, KCNAB2, and SNAP25) identified for BPA and CPA(CA). Interestingly, TP53 was once again identified as the central node with more interactions than any other target. We then used these results to create a final miRNAs-target gene network focused on their connections to the “pathways in cancer”, “p53 signaling pathway”, and “ErbB signaling pathway” terms from the KEGG database (Figs. 3A and B).

CXPA is an uncommon SGT with relatively aggressive features which requires complete excision to prevent progression making its preoperative diagnosis a critical factor in improving clinical outcome. However, its mixed cellularity can make detecting its CA component challenging, reducing the cytodiagnostic accuracy of CA in CXPA to around 50%, and increasing false-negative diagnosis of PA to 38.5% (Klijanienko et al. 1999).

Thus, several studies have tried to identify a variety of novel biomarkers to facilitate better diagnostic accuracy in SGT. Many of these have evaluated the use of miRNAs to distinguish between patients with malignant and benign SGT using both tissue and circulating miRNA targets (saliva or blood). This is because miRNA profiling is expected to create significant diagnostic, prognostic, and/or predictive value, and to help uncover the etiological role of specific miRNAs in cancer initiation, progression, and/or metastasis in various cancers (Anfossi et al. 2018; Graveel et al. 2015; Peng et al. 2016; Rupaimoole et al 2017; Sempere. 2014; Sempere et al. 2021). Numerous studies have identified a wide range of differentially expressed miRNAs in both benign and malignant SGTs, and a recent study attempted to create miRNA signatures for the diagnosis of 24 SGT patients (10 benign, and 14 malignancies) using a panel of 798 miRNAs (Denaro et al. 2019). They suggested that 46 of these miRNAs are likely to be differently expressed between benign and malignant tumors and that these may act as potential biomarkers for their differential diagnosis (Denaro et al. 2019). Another study evaluating 38 malignant and 29 benign parotid gland tumors identified four miRNA combinations (miR-132, miR-15b, miR-140, and miR-223) that could discriminate between malignant and benign parotid tumors using saliva samples (Matse et al. 2013). Most of these studies include various kinds of malignant SGTs, while our study tried to identify differentially expressed miRNAs specifically associated with the CPA and CA portions of CXPA tumors. Our study also compared the expression of these miRNAs between CPA and BPA in an attempt to help understand the pathogenesis of malignant transformation in CXPA, and identify those precancerous miRNA changes that may facilitate the move from BPA to CPA.

Interestingly, both miR-21 and miR-96-5p are up-regulated in CA compared to CPA and both are known to act as onco-miRNAs in head and neck squamous cell carcinoma (HNSCC) (Vahabi et al. 2021). This is further supported by the fact that miR-21 is commonly up-regulated in various malignant SGTs when compared to the benign tumor counterparts (Denaro et al. 2019; Cinpolat et al. 2017). In addition, miR-96-5p forms part of the miR-96/182/183 cluster and its overexpression is also known to be increased in patients with p53 mutations, where it acts to support cell migration and chemoradiation resistance via its targeting of PTEN and activation of the PI3K-AKT signaling pathway in HNSCC (Vahabi et al. 2021). miR-425-5p is another well-known onco-miRNA which is upregulated in breast cancers where it targets PTEN (Xiao et al. 2019), and in MEC tissues, where the function remains unconfirmed (Binmadi et al. 2018).

miR-125a-5p, miR-125b-5p, and miR-140-5p, which are all down-regulated in CA tissues when compared to CPA samples, are also known for their tumor suppressor activities in HNSCC (Vahabi et al. 2021). Of these, both miR-125b-5p and miR-140-5p are also known to be significantly down-regulated in malignant vs. benign SGT samples (Denaro et al. 2021). mir-455-3p has also been reported to act as a tumor suppressor gene downstream of TP53 inducing pro-apoptotic activity in pancreatic (Zhan et al. 2020), osteosarcoma (Yi et al. 2020), and breast cancer (Li et al. 2017).

Unlike previous studies, we identified seven miRNAs (miR-196a-5p, miR-193a-3p, miR-193b-3p, miR-29c-3p, miR-331-3p, miR-361-3p, and miR-423-5p), which demonstrated specific differential expression in CA and CPA tissues when compared to BPA samples. Interestingly, some differentially expressed mRNAs (miR-196a-5p, miR-193b-3p, miR-423-5p, and miR-361-3p) were reported to have some association with oral cavity squamous cell carcinoma (Maruyama et al. 2018; Li et al. 2020; Romani et al. 2021). In addition,miR-193a-3p was reported to be associated with MEC (Binmadi et al. 2018) and miR-331-3p has been linked to AdCC (Zhu et al. 2022). This is of value because although CPA and BPA present with nearly indistinguishable histological profiles, their onco-miRNA profiles are significantly different, with CPA samples expressing several unique transcripts when compared to BPA. These unique results suggest that these miRNAs may be associated with some of the pre-malignant changes in these samples even when the pathology remains was benign. Thus, we suggest that these pre-malignant miRNAs may facilitate better diagnostic accuracy for fine needle aspiration cytology in clinical settings.

In addition, the down-regulated miRNAs in CPA/CA were also all shown to be well-known tumor suppressor miRNAs, including the let-7 family transcripts, miR-9-5p, miR-135a-5p and miR-135b-5p, amongst others. miR-9-5p, miR-135a-5p, and miR-135b-5p, were all shown to be down-regulated in malignant vs. benign SGT samples (Denaro et al. 2019), and the differential expression of both miR-132-5p and miR-140 was consistent with the results of other studies (Matse et al. 2013).

GO enrichment and KEGG pathway analyses revealed that these differentially expressed transcripts were all linked to the regulation of several critical oncogenic signaling pathways, such as the p53 signaling pathway. Moreover, GO enrichment and KEGG pathway evaluations of the differentially expressed miRNAs from CPA vs. PA also revealed their involvement in signal regulation with these transcripts being linked to various critical signaling pathways including the AMPK and TGF-beta signaling pathways. PLAG1 or HMGA2 fusions are also useful biomarkers for distinguishing between PA and various other pathologies. Thus, we expect that CXPA tissues are also likely to encode these PA-specific gene fusions (Toper et al. 2021). In addition, the amplification of MDM2, mutations in TP53, gains and amplifications of MYC, Epidermal growth factor receptor (EGFR), HGF-A (scatter factor), c-Met (a proto-oncogene), Transforming growth factor alpha (TGF α), Fibroblast growth, factors (FGF)-2, and ErbB2 (HER2), amongst others have all been hypothesized to play a role in the progression and invasion of CXPA (Röijer et al. 2002; Martins et al. 2005; Furuse et al. 2010; Hashimoto et al. 2012). Thus it was not surprising that our PPI evaluations included several of these genes, including ErbB2, TP53, HMGA2, HIF1A, EGFR, VEGFA, and MDM2, into the miRNA-Hub gene networks created for both conditions. The most significant hub gene in both networks was shown to be TP53, which is known to be implicated in the malignant transformation of CXPA. In addition, another recent study reported the molecular events underlying PA’s malignant transformation in one patient who experienced three bouts of recurrent disease before finally receiving a CXPA diagnosis, where they identified the sequential mutation of TP53 in these recurrent tumors, all of which was consistent with our results (Valstar et al. 2021).

Despite the clear contribution of our study, it is important to acknowledge some of its potential limitations, as these should be considered when evaluating our data. Due to the rarity of this disease, we were forced to adopt a retrospective study design and were limited to a very small cohort. This limited our ability to perform any validations in a second independent cohort. However, to the best of our knowledge, there have been no other studies evaluating the independent profiles of the CPA and CA portions of CXPA nor any comparing the profiles of CPA and BPA in these patients. Given this we believe our results offer significant insight into the roles of miRNA during CXPA carcinogenesis.

In conclusion, we identified several differentially expressed miRNAs in both the CA and CPA components of CXPA, and confirmed that they are related to TP53 and cancer-related pathways using target gene and gene pathway analyses. In addition, our evaluations identified a handful of unique miRNAs that may help to differentiate between BPA and CPA, which may be valuable in the development of novel liquid biopsy tools. Taken together these results suggest that miRNAs affect tumor suppressor genes such as p53 during the malignant transformation of CXPA, and that there is a possibility that CPA may have the potential to progress to CXPA via the differential regulation of several key miRNAs.

Author contributions
Conceptualization, HK and HK; methodology, HK, SE, and YJB; data analysis and interpretation, HK, JSL, and HK; investigation, WJ, and SA; writing-original draft & review, HK, JSL, and HK; supervision, SE, WJ, and SA; funding acquisition, HK. All authors discussed the results and contributed to the final manuscript.

Funding

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2019009856)

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Conflict of interest: The authors declare no conflict of interest.

Ethics approval

The study was approved by the Committee on Seoul National University Bundang Hospital (No.B-1905-540-304) and Dongtan Sacred Heart Hospital (No.2019-04-298), Republic of Korea.

undefined

Lüers JC, Wittekindt C, Streppel M, Guntinas-Lichius O (2009) Carcinoma ex pleomorphic adenoma of the parotid gland. Study and implications for diagnostics and therapy. Acta Oncol 48(1):132–136.. https://doi.org/10.1080/02841860802183604.
Chen MM, Roman SA, Sosa JA, Judson BL (2014) Predictors of survival in carcinoma ex pleomorphic adenoma. Head Neck 36(9):1324–1328. https://doi.org/10.1002/hed.23453.
Gupta A, Koochakzadeh S, Neskey DM, Nguyen SA, Lentsch EJ (2019) Carcinoma ex pleomorphic adenoma: A review of incidence, demographics, risk factors, and survival. Am J Otolaryngol 40(6):102279. https://doi.org/10.1016/j.amjoto.2019.102279.
Lewis JE, Olsen KD, Sebo TJ (2001) Carcinoma ex pleomorphic adenoma: pathologic analysis of 73 cases. Hum Pathol 32(6):596–604. https://doi.org/10.1053/hupa.2001.25000.
Antony J, Gopalan V, Smith RA, Lam AK (2012) Carcinoma ex pleomorphic adenoma: a comprehensive review of clinical, pathological and molecular data. Head Neck Pathol 6(1):1–9. https://doi.org/10.1007/s12105-011-0281-z.
El-Naggar AK, Callender D, Coombes MM, Hurr K, Luna MA, Batsakis JG (2000) Molecular genetic alterations in carcinoma ex-pleomorphic adenoma: a putative progression model? Genes Chromosomes Cancer. 27(2):162–168.
Bartel DP (2004) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 23;116(2):281–297. https://doi.org/10.1016/s0092-8674(04)00045-5.
Hayes J, Peruzzi PP, Lawler S (2014) MicroRNAs in cancer: biomarkers, functions and therapy. Trends Mol Med 20(8):460–469. https://doi.org/10.1016/j.molmed.2014.06.005.
Denaro M, Navari E, Ugolini C, Seccia V, Donati V, Casani AP, Basolo F (2019) A microRNA signature for the differential diagnosis of salivary gland tumors. PLoS One 25;14(1):e0210968. https://doi.org/10.1371/journal.pone.0210968.
Seethala RR, Stenman G (2017) Update from the 4th Edition of the World Health Organization Classification of Head and Neck Tumours: Tumors of the Salivary Gland. Head Neck Pathol 11(1):55–67. https://doi.org/10.1007/s12105-017-0795-0.
Dweep H, Sticht C, Pandey P, Gretz N (2011) miRWalk–database: prediction of possible miRNA binding sites by "walking" the genes of three genomes. J Biomed Inform 44(5):839–847. https://doi.org/10.1016/j.jbi.2011.05.002.
Dweep H, Gretz N (2015) miRWalk2.0: a comprehensive atlas of microRNA-target interactions. Nat Methods 12(8):697. https://doi.org/10.1038/nmeth.3485.
Dennis G Jr, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA (2003) DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 4(5):P3.
Huang DW, Sherman BT, Lempicki RA (2009) Systematic and integrative analysis of large gene lists using DAVID Bioinformatics Resources. Nature Protoc 4(1):44–57. https://doi.org/10.1038/nprot.2008.211.
Klijanienko J, El-Naggar AK, Vielh P (1999) Fine-needle sampling findings in 26 carcinoma ex pleomorphic adenomas: diagnostic pitfalls and clinical considerations. Diagn Cytopathol 21(3):163–166. https://doi.org/10.1002/(sici)1097-0339(199909)21:3<163::aid-dc3>3.0.co;2-2.
Anfossi S, Fu X, Nagvekar R, Calin GA (2018) MicroRNAs, Regulatory Messengers Inside and Outside Cancer Cells. Adv Exp Med Biol 1056:87–108. https://doi.org/10.1007/978-3-319-74470-4_6.
Graveel CR, Calderone HM, Westerhuis JJ, Winn ME, Sempere LF (2015) Critical analysis of the potential for microRNA biomarkers in breast cancer management. Breast Cancer (Dove Med Press) 7:59–79. https://doi.org/10.2147/BCTT.S43799.
Peng Y, Croce CM (2016) The role of MicroRNAs in human cancer. Signal Transduct Target Ther 1:15004. https://doi.org/10.1038/sigtrans.2015.4.
Rupaimoole R, Slack FJ (2017) MicroRNA therapeutics: towards a new era for the management of cancer and other diseases. Nat Rev Drug Discov 16(3):203–222. https://doi.org/ 10.1038/nrd.2016.246.
Sempere LF (2014) Fully automated fluorescence-based four-color multiplex assay for co-detection of microRNA and protein biomarkers in clinical tissue specimens. Methods Mol Biol 1211:151–170. https://doi.org/10.1007/978-1-4939-1459-3_13.
Sempere LF, Azmi AS, Moore A (2021) microRNA-based diagnostic and therapeutic applications in cancer medicine. Wiley Interdiscip Rev RNA 12(6):e1662. https://doi.org/10.1002/wrna.1662.
Matse JH, Yoshizawa J, Wang X, Elashoff D, Bolscher JGM, Veerman EC, Bloemena E, Wong DT (2013) Discovery and prevalidation of salivary extracellular microRNA biomarkers panel for the noninvasive detection of benign and malignant parotid gland tumors. Clin Cancer Res 19(11):3032–3038. https://doi.org/10.1158/1078-0432.CCR-12-3505.
Vahabi M, Blandino G, Di Agostino S (2021) MicroRNAs in head and neck squamous cell carcinoma: a possible challenge as biomarkers, determinants for the choice of therapy and targets for personalized molecular therapies. Transl Cancer Res 10(6):3090–3110. https://doi.org/10.21037/tcr-20-2530.
Cinpolat O, Unal ZN, Ismi O, Gorur A, Unal M (2017) Comparison of microRNA profiles between benign and malignant salivary gland tumors in tissue, blood and saliva samples: a prospective, case-control study. Braz J Otorhinolaryngol 83(3):276–284. https://doi.org/10.1016/j.bjorl.2016.03.013.
Xiao S, Zhu H, Luo J, Wu Z, Xie M (2019). miR–425–5p is associated with poor prognosis in patients with breast cancer and promotes cancer cell progression by targeting PTEN. Oncol Rep 42(6):2550–2560. https://doi.org/10.3892/or.2019.7371.
Binmadi NO, Basile JR, Perez P, Gallo A, Tandon M, Elias W, Jang SI, Alevizos I (2018) miRNA expression profile of mucoepidermoid carcinoma. Oral Dis 24(4):537–543. https://doi.org/10.1111/odi.12800.
Zhan T, Zhu Q, Han Z, Tan J, Liu M, Liu W, Chen W, Chen X, Chen X, Deng J, Tian X, Huang X (2020) miR-455-3p Functions as a Tumor Suppressor by Restraining Wnt/β-Catenin Signaling via TAZ in Pancreatic Cancer. Cancer Manag Res 12:1483–1492. https://doi.org/10.2147/CMAR.S235794.
Yi X, Wang Y, Xu S (2020) MiR-455-3p downregulation facilitates cell proliferation and invasion and predicts poor prognosis of osteosarcoma. J Orthop Surg Res 15(1):454. https://doi.org/10.1186/s13018-020-01967-1.
Li Z, Meng Q, Pan A, Wu X, Cui J, Wang Y, Li L (2017) MicroRNA-455-3p promotes invasion and migration in triple negative breast cancer by targeting tumor suppressor EI24. Oncotarget 8(12):19455–19466. https://doi.org/10.18632/oncotarget.14307.
Maruyama T, Nishihara K, Umikawa M, Arasaki A, Nakasone T, Nimura F, Matayoshi A, Takei K, Nakachi S, Kariya KI, Yoshimi N (2018) MicroRNA-196a-5p is a potential prognostic marker of delayed lymph node metastasis in early-stage tongue squamous cell carcinoma. Oncol Lett 15(2):2349–2363. https://doi.org/10.3892/ol.2017.7562.
Li X, Rui B, Cao Y, Gong X, Li H (2020) Long non–coding RNA LINC00152 acts as a sponge of miRNA–193b–3p to promote tongue squamous cell carcinoma progression. Oncol Lett 19(3):2035–2042. https://doi.org/10.3892/ol.2020.11293.
Romani C, Salviato E, Paderno A, Zanotti L, Ravaggi A, Deganello A, Berretti G, Gualtieri T, Marchini S, D'Incalci M, Mattavelli D, Piazza C, Bossi P, Romualdi C, Nicolai P, Bignotti E (2021) Genome-wide study of salivary miRNAs identifies miR-423-5p as promising diagnostic and prognostic biomarker in oral squamous cell carcinoma. Theranostics 11(6):2987–2999. https://doi.org/10.7150/thno.45157.
Zhu QH, Meng Y, Tang YT, Hou CX, Sun NN, Han W, Wang CX, Ye J (2022) Identification of pivotal microRNAs involved in the development and progression of salivary adenoid cystic carcinoma. J Oral Pathol Med 51(2):160–171. https://doi.org/ 10.1111/jop.13261.
Toper MH, Sarioglu, S (2021) Molecular Pathology of Salivary Gland Neoplasms: Diagnostic, Prognostic, and Predictive Perspective. Adv Anat Pathol 28(2):81–93. https://doi.org/10.1097/PAP.0000000000000291.
Röijer E, Nordkvist A, Ström AK, Ryd W, Behrendt M, Bullerdiek J, Mark J, Stenman G (2002) Translocation, deletion/amplification, and expression of HMGIC and MDM2 in a carcinoma ex pleomorphic adenoma. Am J Pathol 160(2):433–440. https://doi.org/10.1016/S0002-9440(10)64862-6.
Martins C, Fonseca I, Roque L, Pereira T, Ribeiro C, Bullerdiek J, Soares J (2005) PLAG1 gene alterations in salivary gland pleomorphic adenoma and carcinoma ex-pleomorphic adenoma: a combined study using chromosome banding, in situ hybridization and immunocytochemistry. Mod Pathol 18(8):1048–1055. https://doi.org/ 10.1038/modpathol.3800386.
Martins C, Fonseca I, Roque L, Pereira T, Ribeiro C, Bullerdiek J, Soares J (2005) PLAG1 gene alterations in salivary gland pleomorphic adenoma and carcinoma ex-pleomorphic adenoma: a combined study using chromosome banding, in situ hybridization and immunocytochemistry. Mod Pathol 18(8):1048–1055. https://doi.org/10.1038/modpathol.3800386.
Furuse C, Miguita L, Rosa ACG, Soares AB, Martinez EF, Altemani A, De Araújo VC (2010) Study of growth factors and receptors in carcinoma ex pleomorphic adenoma. J Oral Pathol Med 39(7):540–547. https://doi.org/10.1111/j.1600-0714.2009.00858.x.
Hashimoto K, Yamamoto H, Shiratsuchi H, Nakashima T, Tamiya S, Nishiyama K, Higaki Y, Komune S, Tsuneyoshi M, Oda Y (2012) HER-2/neu gene amplification in carcinoma ex pleomorphic adenoma in relation to progression and prognosis: a chromogenic in-situ hybridization study. Histopathology 60(6B):E131-142. https://doi.org/10.1111/j.1365-2559.2012.04201.x.
Valstar MH, Mast H, Ten Hove I, Moonen LR, Balm AJ, Smeele LE, Koljenović S, Dinjens WN, van Velthuysen MF (2021) Malignant transformation of salivary gland pleomorphic adenoma: proof of principle. J Pathol Clin Res 7(5):432–437. https://doi.org/10.1002/cjp2.216.

Table 1. Clinicopathologic characteristics

	Pleomorphic adenoma	Carcinoma ex pleomorphic adenoma	P value
Mean age (range)	44.7 (26-71)	53.0 (28-75)	0.120
Sex (male:female)	7:9	6:7	0.897
Size (cm, range)	2.6 (1.4-4.0)	3.5 (1.2-6.3)	0.093
Subtype of carcinomatous portion	NA
Salivary duct carcinoma		4 (30.7%)
Epithelial-myoepithelial carcinoma		2 (15.4%)
Mucoepidermoid carcinoma		2 (15.4%)
adenocarcinoma		3 (23.1%)
Clear cell carcinoma		1 (7.7%)
Oncocytic carcinoma		1 (7.7%)
Grade of carcinomatous portion	NA
Low grade		3 (23.1%)
Intermediate grade		5 (38.45%)
High grade		5 (38.45%)
Invasiveness	NA
Non-invasive		5 (38.45%)
Minimally invasive		3 (23.1%)
Invasive		5 (38.45%)
Total	16	13
NA, not applicable

Table 2. Significantly differentially expressed miRNAs between pleomorphic adenoma portion (CPA) and carcinomatous portion (CA) in carcinoma ex pleomorphic adenoma.

miRNAs, CPA vs CA		Fold change	p-value	Adjusted p-value
Up-regulated
miR-455-3p	3.409		<0.001	<0.001
miR-140-5p	2.91		<0.001	0.002
miR-483-5p	2.504		0.016	0.025
miR-125b-5p	2.173		0.002	0.003
miR-125a-5p	2.05		0.004	0.007
Down-regulated
miR-21-3p	0.124		0.025	0.037
miR-183-5p	0.148		0.015	0.023
miR-182-5p	0.173		0.012	0.018
miR-425-5p	0.225		0.016	0.049
miR-96-5p	0.246		<0.001	0.001
miR-200a-3p	0.443		0.006	0.006
miR-181a-3p	0.446		0.03	0.045
miR-505-3p	0.46		0.016	0.024

CPA, pleomorphic adenoma portion of carcinoma ex pleomorphic adenoma; CA, carcinomatous portion of carcinoma ex pleomorphic adenoma

Table 3. Significantly differentially expressed miRNAs between pleomorphic adenoma portion (CPA) in carcinoma ex pleomorphic adenoma (CXPA) and pleomorphic adenoma (BPA).

miRNAs, BPA vs CPA	Fold change	p-value	Adjustive p-value
Up-regulated
let-7a-3p	3.342	<0.001	<0.001
miR-27a-3p	3.136	<0.001	<0.001
miR-9-5p	2.669	0.002	0.002
miR-135a-5p	2.63	0.019	0.028
miR-135b-5p	2.201	0.001	0.002
miR-455-5p	2.474	0.033	0.045
miR-218-5p	2.123	0.001	0.004
miR-181d-5p	2.201	<0.001	<0.001
miR-369-3p	2.101	0.028	0.028
miR-132-5p	2.041	0.003	0.009
Down-regulated
miR-196a-5p	0.218	0.009	0.022
miR-193a-3p	0.464	0.001	0.004
miR-193b-3p	0.459	<0.001	<0.001
miR-29c-3p	0.458	<0.001	0.002
miR-331-3p	0.424	<0.001	<0.001
miR-361-3p	0.49	0.003	0.004
miR-423-5p	0.466	<0.001	<0.001

Table 4. Gene Ontology enrichment of miRNAs which differently expressed in two groups.

Biology process Term	Number of involved genes	P-value
Pleomorphic adenoma (CPA) versus carcinomatous portion (CA) in carcinoma ex pleomorphic adenoma
GO:0006915~apoptotic process	27	1.39E-04
GO:0007050~cell cycle arrest	11	7.19E-04
GO:0030512~negative regulation of transforming growth factor beta receptor signaling pathway	7	0.00207
GO:0006357~regulation of transcription from RNA polymerase II promoter	20	0.00222
GO:0045944~positive regulation of transcription from RNA polymerase II promoter	35	0.00232
GO:0045165~cell fate commitment	6	0.00254
GO:0071456~cellular response to hypoxia	8	0.00381
GO:0009791~post-embryonic development	7	0.00403
GO:0010468~regulation of gene expression	8	0.00477
GO:0006366~transcription from RNA polymerase II promoter	21	0.00524
Pleomorphic adenoma (BPA) versus carcinoma ex pleomorphic adenoma (CPA/CA)
GO:0018105~peptidyl-serine phosphorylation	15	2.98E-07
GO:0045893~positive regulation of transcription, DNA-templated	28	9.20E-06
GO:0006468~protein phosphorylation	26	9.32E-06
GO:0030512~negative regulation of transforming growth factor beta receptor signaling pathway	9	4.85E-05
GO:0016477~cell migration	14	5.91E-05
GO:0008284~positive regulation of cell proliferation	24	1.08E-04
GO:0045944~positive regulation of transcription from RNA polymerase II promoter	39	1.36E-04
GO:0045892~negative regulation of transcription, DNA-templated	24	2.87E-04
GO:0046777~protein autophosphorylation	12	8.95E-04
GO:0018107~peptidyl-threonine phosphorylation	6	0.00104

Table 5. KEGG pathway analysis of the differentially expressed miRNAs and targets in two groups.

KEGG pathway	P-value	Numbers of involved genes	Numbers of involved miRNAs
Pleomorphic adenoma (CPA) versus carcinomatous portion (CA) in carcinoma ex pleomorphic adenoma
Viral carcinogenesis	2.54E-12	88	7
Adherens junction	8.65E-10	40	7
Cell cycle	3.23E-09	61	7
Hippo signaling pathway	1.80E-08	60	7
Other types of O-glycan biosynthesis	2.61E-08	11	6
p53 signaling pathway	3.05E-08	40	7
Hepatitis B	5.90E-08	61	7
Proteoglycans in cancer	6.20E-08	80	7
Fatty acid biosynthesis	1.07E-07	3	5
Bacterial invasion of epithelial cells	1.22E-07	37	7
Oocyte meiosis	1.47E-07	51	7
Prostate cancer	9.44E-07	45	7
Pleomorphic adenoma (BPA) versus carcinoma ex pleomorphic adenoma (CPA/CA)
Proteoglycans in cancer	5.23E-14	89	6
ECM-receptor interaction	6.12E-11	34	5
Hippo signaling pathway	2.22E-09	69	6
Prion diseases	2.23E-09	12	5
Bacterial invasion of epithelial cells	9.41E-09	44	6
AMPK signaling pathway	5.16E-07	66	6
Chronic myeloid leukemia	3.29E-06	40	6
Viral carcinogenesis	3.32E-06	83	6
TGF-beta signaling pathway	3.51E-06	42	6
Glioma	3.75E-06	34	6
Fatty acid biosynthesis	4.45E-06	3	4
Adherens junction	1.65E-05	38	6
Thyroid hormone signaling pathway	1.65E-05	57	6

No competing interests reported.

The role of microRNA expression in malignant transformation of pleomorphic adenoma: comparing carcinoma ex pleomorphic adenoma vs pleomorphic adenoma

Status:

Version 1

Abstract

Figures

Introduction

Patients And Methods

Patients and materials

RNA isolation

Library preparation and sequencing

Data analysis

Statistical analysis

Results

Discussion

Declarations

References

Tables

Additional Declarations

Status:

Version 1