Somatic copy number alterations in pleomorphic adenoma and recurrent pleomorphic adenoma

(1)

Somatic copy number alterations in pleomorphic adenoma and recurrent pleomorphic adenoma

Fernanda Viviane Mariano, DDS, PhD,^aFelipe Fidalgo, PhD,^bAndre Luıs Maion Casarim, MD,^c Ant^onio Santos Martins, MD, PhD ,^cJo~ao Figueira Scarini, DDS, PhD student,^d

Reydson Alcides de Lima Souza, DDS, MSc student,^dErika Said Egal, DVM, PhD,^a

Luiz Paulo Kowalski, MD, PhD,^eAna Cristina Victorino Krepischi, PhD,^fand Albina Altemani MD, PhD^a

Objective.As the genetic changes in recurrent pleomorphic adenoma (RPA) have not yet been investigated, the aim of this study was to assess the genomic profile of somatic copy number alteration in RPA and pleomorphic adenoma (PA) by using array comparative genomic hybridization (aCGH).

Study Design.Four cases of RPA and 13 cases of PA were evaluated by using aCGH, using a 180 K platform. Data were analyzed by using Nexus Copy Number Discovery.

Results.The RPA group rarely showed any copy number alteration, except for 1 case that exhibited losses in 5 p15.33 p15.1, 5 q13.1 q35.3 and 12 q12 q13.11. The PA group also showed few copy number alterations, and the most frequent findings involved chromosomes 8: 8p21.3-p12 (gain), 8q12.1 (loss), 8p23.3-q24.3 (gain), and 8q12.1-q21.11 (gain). Genomic amplifications were revealed in the PA group, and the relevant affected genes wereMAML2 and LIFR.

Conclusions.PA and RPA exhibit few somatic copy number alterations and show a similar genomic profile on aCGH. (Oral Surg Oral Med Oral Pathol Oral Radiol 2020;129:59 64)

Pleomorphic adenoma (PA) is the most common type of salivary gland neoplasm. Extensive cytogenetic studies revealed that these tumors have highly specific chromosome abnormalities. The genomic findings of PAs are classified into 4 major subgroups: (1) chromosomal rearrangements involving a breakpoint on chromosome 8 (q12); (2) chromosomal rearrangements involving a breakpoint on chromosome 12 (q14-15);

(3) clonal changes different from 8q12 and 12q14-15;

and (4) an apparently normal karyotype, including tumors characterized by trisomy 8 or Y as mosaics.^1,2 Molecular characterization of these loci identified some important genes in carcinogenesis (PLAG1, HMGA2, and CTNNB1).^2-7

Recurrent pleomorphic adenoma (RPA) has been associated with either increase in the number and complexity of genetic abnormalities or acquisition of promoting mutations. To date, RPA studies have focused on tumor treatment, target marker for recurrence, analysis of

proliferation indices, and comparison of the stromal patterns between nonrecurrent PAs and RPAs.^8-13 So far, there has been no investigation comparing the genomic profiles of RPA and PA. The aim of this study was to assess the genomic profile of copy number alterations associated with PA and RPA through array comparative genomic hybridization (aCGH).

MATERIAL AND METHODS

Thirteen cases of PAs and 4 of RPAs were retrieved from the archives of the Pathology Department, Faculty of Medical Sciences of the State University of Campinas (Campinas, Brazil). The cases were reviewed by 2 pathologists, and the diagnoses were reviewed and confirmed according to the criteria for the histologic typing of PA in the 4th edition of the World Health Organization Classification of Head and Neck Tumours²

Tumor DNA was extracted from a 1.5-mm-diameter punch of the paraffin-embedded tissue using the Qia- gen extraction kit (Qiagen GmbH, Hilden, Germany), according to the manufacturer’s recommendations. To improve DNA quality, the protocol included

Statement of Clinical Relevance

This study showed that pleomorphic adenoma and recurrent pleomorphic adenoma exhibit few somatic copy number alterations and that they show similar patterns of genomic profiles by array comparative genomic hybridization. These results suggest that besides the common origin of these tumors, their tumorigenesis can be similar as well.

aPathology Department, Faculty of Medical Sciences/UNICAMP, Campinas-SP, Brazil.

bInternational Research Center, AC Camargo Cancer Center, S~ao Paulo-SP, Brazil.

cHead and Neck Surgery Department, Faculty of Medical Sciences/

UNICAMP, Campinas-SP, Brazil.

dOral Pathology Department, Piracicaba Dental School/UNICAMP, Piracicaba-SP, Brazil.

eHead and Neck Surgery Department, AC Camargo Cancer Center, S~ao Paulo-SP, Brazil.

fDepartment of Genetics and Evolutionary Biology, Institute of Bio- sciences, USP, S~ao Paulo-SP, Brazil.

Received for publication Apr 10, 2019; returned for revision Aug 7, 2019; accepted for publication Aug 25, 2019.

Ó 2019 Published by Elsevier Inc.

2212-4403/$-see front matter

https://doi.org/10.1016/j.oooo.2019.08.016

59

(2)

deparaffinization with xylene, followed by methanol washings, and a 24-hour incubation in 1 mol/L sodium thiocyanate. Subsequently, the tissue pellet was dried and digested for 1.5 days in a lysis buffer with high proteinase K dose. The material was column purified and eluted in buffer.

Samples of tumor DNA and reference DNA (pooled from the blood of different healthy donors) were labeled by using the Enzo Genomic DNA Labeling kit (Promega, Madison, WI), according to the manufacturer’s instructions. Five hundred nanograms of test DNA and 500 ng of reference DNA were cohybri- dized to a 180-k oligonucleotide array (SurePrint G3 Human CGH Microarray Kit 4£ 180 K design 22060;

Agilent Technologies, Santa Clara, CA), according to the manufacturer’s instructions. This design contains 24,011 exomic probes. Microarray images were obtained by Agilent Microarray Scanner Bundle (Agi- lent Technologies, Santa Clara, CA), and data were extracted by using Feature Extraction (v 9.1) (Agilent Technologies, Santa Clara, CA). aCGH data were analyzed by using the software Nexus Copy Number Dis- covery edition v 7.0. Genomic copy number alteration was called based on the FASST2 segmentation algo- rithm (significance threshold set at 5£ 10 ⁸) with threshold log₂ratios of 0.2 or 0.8 for gains and high- copy gains, respectively, and 0.2 or 1.0 for losses and homozygous losses, respectively.

The present study was carried out in accordance with the ethical guidelines of our institution (CEP/FCM- 1155/2011:22, November, 2011).

RESULTS

Clinicopathologic and histopathologic data

The PA group consisted of 7 females and 6 males (mean age 41 years). The parotid gland was the most com- monly affected site (77%), followed by the minor salivary glands (15.4%) and the submandibular gland (7.6%) (Table I). To date, none of the cases has recurred.

The mean follow-up time was 46 months (range 2 316 months). The RPA group consisted of 4 females (mean age 61 years). Mean time to recurrence was 78 months (range 12 216 months) and the mean follow-up time was 113 months (range 96 137 months). In all cases, the recurrence was local and occurred in the gland affected by the primary PA (Table II).

Microscopically, PAs were observed to be encapsu- lated, and the neoplastic parenchyma exhibited a biphasic population composed of a mixture of luminal and myoepithelial cells. The tumor exhibited variable architecture, with formation of duct-lining structures and cell proliferation permeated by myxochondroid stroma. In contrast, RPAs showed multinodular growth, and the nodules were composed of ductal elements, epithelial and myoepithelial proliferation, and

myxoid stroma. The representative morphologic features of PA and RPA are illustrated inFigure 1.

aCGH data

The PA group showed few copy number alterations.

The most frequent findings involved chromosome 8 and included 8p21.3-p12 (gain), 8-q12.1 (loss), 8- p23.3-q24.3 (gain), and 8-q12.1-q21.11 (gain), followed by loss at 19q13.32. The amplified genes found in the PA group were MAML2- 5p13.1 (1 case) and LIFR -11q21 (1 case) (Table III,Figure 2).

The RPA group showed almost no copy number alteration, except for 1 case that exhibited losses in 5p15.33-p15.1, 5q13.1-q35.3 and 12q12-q13.11 chromosomes (Table IV, Figure 3). The only alteration found in both tumors was loss at chromosome 5p15.33-p15.31, but according to the literature, none of the genes of this chromosome is known to be related to cancer.

Table I. Clinicopathologic features of 13 patients diagnosed with pleomorphic adenoma

Clinicopathologic feature N (%)

Gender

Female 7 (53.8%)

Male 6 (46.2%)

Age (years)

Mean 41 (range 24 56)

Anatomic location

Parotid gland 10 (76%)

Submandibular gland 1 (7.7%)

Minor salivary gland 2 (15.4%)

Parapharyngeal space 1 (7.7%)

Soft palate 1 (7.7%)

Follow-up (months)

Table II. Clinicopathologic features of 4 patients diagnosed with recurrent pleomorphic adenoma

Clinicopathologic feature N (%)

Gender

Female 4 (100%)

Age (years)

Anatomic location

Parotid gland 4 (100%)

Time to recurrence (months)

Case 1 216

Case 2 36

Case 3 48

Case 4 12

Follow-up (months)

Mean 113 (range 96 137)

(3)

DISCUSSION

PA accounts for about 60% of all salivary gland tumors. The reported mean age at presentation was 46 years; however, the age ranged from the 1st to the 10th

decades. A female predominance was found, and the parotid gland was the most common site.² Important cytogenetic studies of PA have shown that 70% of the tumors are karyotypically abnormal, and the alterations Fig. 1. A, B, Pleomorphic Adenoma. A well-demarcated lesion showing an admixture of epithelial, myoepithelial, and stromal components. The tumor presents a variety of growth patterns, including ductal structures and myxochondroid stroma (hematoxy- lin and eosin [H&E]; original magnification£ 10, £ 20). C, D, Recurrent pleomorphic adenoma showing multinodular growth.

The nodules are composed of the ductal elements, epithelial and myoepithelial proliferation, and myxoid stroma (H&E; original magnification £10, £20).

Table III. Somatic copy number alteration in pleomorphic adenoma

Case Loss Gain Amplification

1 0 8p23.3-8q24.3 (PCM1, WRN, WHSC1L1, FGFR1,

HOOK3, TCEA1, PLAG1, CHCHD7, NCOA2, HEY1, COX6 C, EXT1, MYC, NDRG1, RECQL4)

0

2 5p15.33-p15.31* 0 5p13.1 (LIFR)

3 0 0 0

4 0 0 0

5 0 12p13.2 (ETV6) 11q21 (MAML2)

6 6q16.1-q25.3 (PRDM1, ROS1, GOPC, STL, MYB, TNFAIP3, ECT2L, EZR)

0 0

7 0 0 0

8 8q12.1 (PLAG1, CHCHD7) 0 0

9 0 8p21.3-p12* 0

10 9p24.3-p21.1 (JAK2, CD274, NFIB, MLLT3) 19q13.32 (ERCC2)

0 0

11 0 0 0

12 19q13.32 (ERCC2) 8q12.1-q21.11 (CHCHD7, NCOA2) 0

13 0 0 0

*Absence of genes related to cancer.

(4)

are divided into 4 groups: (1) PAs with rearrangements at 8q12; (2) PAs with rearrangements at 12q13-15, (3) PAs with other changes different from 8q12 and 12q13-15, and (4) PAs with apparently normal karyotype.¹The significant genes found in these rearrangements were PLAG1 (8q12) and HMGA2 (12q13-15).^2-7 PLAG1 is expressed as a result of translocations that cause promoter swapping between PLAG1 and a ubiq- uitously expressed translocation partner gene. CTNNB1 and LIFR are the most common partners of PLAG1,

generating the t(3;8)(p21;q12) and t(5;8)(p13;q12) translocations, respectively.^2-3,7 Cryptic gene fusions involving PLAG1 have also been described as CTNNB1-PLAG1 and TCEA1-PLAG1.^6,14

The present study investigated the somatic copy number alteration in PA and RPA and high copy gains or amplification of the PLAG1 was expected; however, the results showed few gains and amplification of PLAG1. Alterations at chromosome 8 were observed;

however, only 1 case showed gain at 8p23.3-8q24.3 involving PLAG1. Another case displayed loss at 8q12.1. Several studies have demonstrated PLAG1 positivity as possibly resulting from gene overexpression and not because of its amplification or high copy gain.

Besides, it has been suggested that PLAG1 expression is controlled by complex molecular mechanisms, such as translocations or by micro-RNA mediated transla- tional control.^6,^15,16

HMGA2, the target gene located at 12q14-15 also showed fusions as HMGA2-NFIB [ins(9;12)] and HMGA2-FHIT [t(3;12)], but the most important event was the separation of the DNA-binding domains from potential mRNA.^4,5 No high copy gain or gain of Fig. 2. Global array comparative genomic hybridization (aCGH) genomic profile exhibiting the identified copy number alterations in pleomorphic adenoma (PA) cases. The x-axis represents probes ordered according to their genomic position from chromosomes 1p to Xq. The y-axis denotes the log2 test/reference values (genomic gains in blue and losses in red are plotted above or below the 0 baseline, respectively; images adapted from the software Nexus Copy Number 7.0, Biodiscovery).

Table IV. Somatic copy number alteration in recurrent pleomorphic adenoma

Case Loss Gain Amplification

1 0 0 0

2 0 0 0

3 5p15.33-p15.1*

5q13.1-q35.3 (APC, PDGFRB, CD74, ITK, EBF1, RANBP17, TLX3, NPM1, NSD1) 12q12-q13.11(ARID2)

0 0

4 0 0 0

*Absence of genes related to cancer.

Fig. 3. Global array comparative genomic hybridization (aCGH) genomic profile exhibiting the identified copy number alterations in recurrent pleomorphic adenoma (RPA) cases. The x-axis represents probes ordered according to their genomic position from chromosomes 1p to Xq. The y-axis denotes the log2 test/reference values (genomic gains in blue and losses in red are plotted above or below the 0 baseline, respectively; images adapted from the software Nexus Copy Number 7.0, Biodiscovery).

(5)

HMGA2 was observed in our cases. Overexpression of HMGA2 from gene amplification seems to be present in malignant transformation of PA.¹⁷ The only alteration found at chromosome 12 was gain at 12p13.2 encompassing the ETV6 gene from 1 case of PA. Rear- rangements of ETV6 as ETV6-NTRK3 or ETV6-X fusions were found in secretory carcinoma of the salivary glands, secretory carcinoma of the breast, and other tumors¹⁸; however, gains in ETV6 have not been described in salivary gland tumors until now.

The most interesting finding of the present study was the somatic high copy number of MAML2 and LIFR;

amplification of these genes has not been described in salivary gland tumors until now. However, more cases presenting amplifications are necessary to draw further conclusions. MAML2 is fused with CRTC1 and CRTC3 in mucoepidermoid carcinoma originating from t(11;19)(q21;p13) and t(11;15)(q21;q26), respectively.¹⁹ The final product (fusion protein) influences the expression of other genes involved in the transformation of neoplastic cells.^20,21 In PAs, it was found that LIFR is a fusion partner of PLAG1 resulting from t (5;8)(p13;q12), and this fusion leads to upregulation of PLAG1 gene expression under the control of LIFR.⁷ Therefore, MAML2 and LIFR amplifications may be considered occasional findings, or they can be included in PA groups without known rearrangements.

In our series, no recurrence and malignant transformation were observed in the PA and RPA cases, and all patients are alive without disease. Carcinoma from RPA is reported in 1.5% to 23% of the cases, and when it occurs, it involves the acquisition of mutations over a period.²²The risk of PA recurrence ranges from 4%

to 45%, depending on the surgical technique.^23,24 Potential causative factors for recurrence are capsule rupture or incomplete resection of microscopic exten- sions beyond the pseudocapsule. Patient age ranges from 26 to 69 years (mean age 59.6 years), with no sta- tistical difference in occurrence between males and females.¹⁰ The time from the primary tumor to its recurrence ranged from 48 to 144 months, with a mean interval of 64.8 months.¹⁰

The recurrence of a tumor can be caused either by an increase in the number and complexity of gene abnormalities or by the acquisition of new mutations.¹³Clin- ical factors, surgical approach, and some markers are involved in RPA as upregulation of the expression of cell proliferation regulatory proteins.^12,13Recently, de Brito et al.¹³ found PLAG1 positivity in RPA by per- forming immunohistochemical analysis and concluded that expression of the protein is maintained from PA to RPA, suggesting that it may play a role in disease recurrence.

In this study, only 1 RPA case (25%) displayed changes in the number of copies (loss in 5p15.33-

p15.1), whereas more than half the PAs (61.5%) pre- sented some type of alteration, such as gene gain or amplification. According to our interpretation, this result may be a reflection of the small number of RPA samples studied (4 cases). This result, however, sug- gests that the change in the number of copies may not be responsible for recurrence and that other complex molecular mechanisms, such as translocations, gene fusions, or micro-RNA mediated translation control, may play a more important role. Further studies are necessary for better understanding of the pathogenesis of these tumors.

CONCLUSIONS

Our data show that PA and RPA exhibit few somatic copy number alterations and that they show similar genomic profiles on aCGH, suggesting similar mechanisms of tumorigenesis. The copy number alteration at chromosome 8 and the amplified genes MAML2 and LIFR may be a contributory factor in PA tumorigenesis. However, these genetic alterations might not play a role in tumor recurrence.

FUNDING

This study was supported by S~ao Paulo Research Foun- dation (FAPESP) (Process: 11/23204-5; 11/23366-5;

15/07304-0; 17/00831-0).

REFERENCES

1. Bullerdiek J, Wobst G, Meyer-Bolte K, et al. Cytogenetic sub- typing of 220 salivary gland pleomorphic adenomas: correlation to occurrence, histological subtype, and in vitro cellular behav- ior. Cancer Genet Cytogenet. 1993;65:27-31.

2. Bell D, Bullerdiek J, Gnepp DR, Schwartz MR, Stenman G, Tri- antafyliou A. Pleomorphic adenoma. In: Barnes L, ed. World Health Organization Classification of of Head and Neck Tumours—Pathology and Genetics, ed 4, Lyon, France: IARC Press; 2017:185-186.

3. Kas K, Voz ML, Roijer E, et al. Promoter swapping between the genes for a novel zinc finger protein and beta-catenin in pleomorphic adenomas with t(3;8)(p21;q12) translocations. Nat Genet. 1997;15:170-174.

4. Geurts JM, Schoenmakers EF, Roijer E, Astrom AK, Stenman G, Van de Ven WJ. Identification of NFIB as recurrent translocation partner gene of HMGIC in pleomorphic adenomas. Onco- gene. 1998;16:865-872.

5. Geurts JM, Schoenmakers EF, Roijer E, Stenman G, Van de Ven WJ. Expression of reciprocal hybrid transcripts of HMGIC and FHIT in a pleomorphic adenoma of the parotid gland. Cancer Res. 1997;57:13-17.

6. Astrom AK, Voz ML, Kas K, et al. Conserved mechanism of PLAG1 activation in salivary gland tumors with and without chromosome 8 q12 abnormalities: identification of SII as a new fusion partner gene. Cancer Res. 1999;59:918-923.

7. Voz ML, Astrom AK, Kas K, Mark J, Stenman G, Van de Ven WJ. The recurrent translocation t(5;8)(p13;q12) in pleomorphic adenomas results in upregulation of PLAG1 gene expression under control of the LIFR promoter. Oncogene. 1998;16:1409- 1416.

(6)

8.Carew JF, Spiro RH, Singh B, Shah JP. Treatment of recurrent pleomorphic adenomas of the parotid gland. Otolaryngol Head Neck Surg. 1999;121:539-542.

9.Stennert E, Wittekindt C, Klussmann JP, Arnold G, Guntinas- Lichius O. Recurrent pleomorphic adenoma of the parotid gland:

a prospective histopathological and immunohistochemical study.

Laryngoscope. 2004;114:158-163.

10.Leonetti JP, Marzo SJ, Petruzzelli GJ, Herr B. Recurrent pleomorphic adenoma of the parotid gland. Otolaryngol Head Neck Surg. 2005;133:319-322.

11.Soares AB, Altemani A, de Araujo VC. Study of histopathological, morphological and immunohistochemical features of recurrent pleomorphic adenoma: an attempt to predict recurrence of pleomorphic adenoma. J Oral Pathol Med. 2011;40:352-358.

12.de Souza AA, Altemai A, Passador-Santos F, et al. Dysregula- tion of the Rb pathway in recurrent pleomorphic adenoma of the salivary glands. Virchows Arch. 2015;467:295-301.

13.de Brito BS, Gaspar NG, Egal ES, et al. PLAG1 expression is maintained in recurrent pleomorphic adenoma. Virchows Arch.

2016;469:477-481.

14.Asp J, Persson F, Kost-Alimova M, Stenman G. CHCHD7- PLAG1 and TCEA1-PLAG1 gene fusions resulting from cryptic, intrachromosomal 8 q rearrangements in pleomorphic salivary gland adenomas. Genes Chromosomes Cancer. 2006;45:

820-828.

15.Zhang X, Cairns M, Rose B, et al. Alterations in miRNA proc- essing and expression in pleomorphic adenomas of the salivary gland. Int J Cancer. 2009;124:2855-2863.

16.Rotellini M, Palomba A, Baroni G, Franchi A. Diagnostic utility of PLAG1 immunohistochemical determination in salivary gland tumors. Appl Immunohistochem Mol Morphol. 2014;22:390-394.

17.Roijer E, Nordkvist A, Strom AK, et al. Translocation, deletion/

amplification, and expression of HMGIC and MDM2 in a carcinoma ex pleomorphic adenoma. Am J Pathol. 2002;160:

433-440.

18.Skalova A, Vanecek T, Simpson RH, et al. Mammary analogue secretory carcinoma of salivary glands: molecular analysis of 25 ETV6 gene rearranged tumors with lack of detection of classical ETV6-NTRK3 fusion transcript by standard RT-PCR: report of

4 cases harboring ETV6-X gene fusion. Am J Surg Pathol.

2016;40:3-13.

19. Skalova A, Vanecek T, Simpson RH, et al. CRTC1-MAML2 and CRTC3-MAML2 fusions were not detected in metaplastic War- thin tumor and metaplastic pleomorphic adenoma of salivary glands. Am J Surg Pathol. 2013;37:1743-1750.

20. Tonon G, Modi S, Wu L, et al. t(11;19)(q21;p13.1) translocation in mucoepidermoid carcinoma creates a novel fusion product that disrupts a Notch signalling pathway. Nat Genet. 2003;33:

208-213.

21. Behboudi A, Enlund F, Winnes M, et al. Molecular classification of mucoepidermoid carcinomas-prognostic significance of the MECT1-MAML2 fusion oncogene. Genes Chromosomes Can- cer. 2006;45:470-481.

22. Mariano FV, Giovanetti K, Saccomani LF, et al. Carcinoma ex- pleomorphic adenoma derived from recurrent pleomorphic adenoma shows important difference by array CGH compared to recurrent pleomorphic adenoma without malignant transformation. Braz J Otorhinolaryngol. 2016;82:687-694.

23. Laccourreye H, Laccourreye O, Cauchois R, Jouffre V, Menard M, Brasnu D. Total conservative parotidectomy for primary benign pleomorphic adenoma of the parotid gland: a 25-year experience with 229 patients. Laryngoscope. 1994;104:1487-1494.

24. Zbaren P, Tschumi I, Nuyens M, Stauffer E. Recurrent pleomorphic adenoma of the parotid gland. Am J Surg. 2005;189:203- 207.

Reprint requests:

Fernanda Viviane Mariano

Departamento de Anatomia Patologica Faculdade de Ci^encias Medicas Universidade Estadual de Campinas Av. Tessalia Vieira de Camargo 126 Bar~ao Geraldo

Cidade Universitaria

Campinas-SP, CEP: 13083-887 Brazil

fevimariano@gmail.com