Original Research—Head and Neck Surgery
Epithelial-Myoepithelial Carcinoma of the
Salivary Glands: An Analysis of 246 Cases
Alejandro Vázquez, MD1, Tapan D. Patel1,
Christine M. D’Aguillo, MD1, Rami Y. Abdou, MD1,
William Farver1, Soly Baredes, MD1,2, Jean Anderson Eloy, MD1,2,3,
and Richard Chan W. Park, MD1
No sponsorships or competing interests have been disclosed for this article.
Abstract
Objective. Epithelial-myoepithelial carcinoma (EMC) is a rare
neoplasm of the salivary glands. In this study, we aim to
examine the demographic, clinicopathologic, and survival
features of EMC using a population-based approach.
Study Design and Setting. Retrospective cohort study.
Subjects and Methods. The Surveillance, Epidemiology, and
End Result (SEER) database (1973-2010) was queried for
EMC of the major salivary glands. Data were analyzed with
respect to various demographic and clinicopathologic factors. Survival was analyzed using the Kaplan-Meier and Cox
proportional hazards models.
Results. In total, 246 cases were available for frequency analysis and 207 for survival analysis. Mean 6 SD age at diagnosis was 63.8 6 15.4 years. EMC affected females more
frequently (57.3%). Distant metastases were present at diagnosis in only 4.5% of cases. Overall disease-specific survival
(DSS) at 60, 120, and 180 months was 91.3%, 90.2%, and
80.7%, respectively. Patients with low-grade histology had
significantly better survival at 180 months relative to those
with high-grade tumors (90.6% vs 0.0%, P = .0246). When
stratified by tumor size, patients with lesions .4 cm had
the worst survival at 180 months (58.8%, P = .0003). All but
9 of the 207 cases available for survival analysis underwent
surgery. A total of 85 patients (41.1%) received radiotherapy
in addition to surgery. No survival benefit was noted for
patients who received radiotherapy compared with those
who did not (P = .4832).
Conclusion. This report represents the largest series of EMC
to date. Despite being regarded as a low-grade, indolent
tumor, a significant fraction of our cohort underwent radiotherapy in addition to surgery, with no apparent added survival benefit.
Keywords
salivary cancer; epithelial-myoepithelial carcinoma; survival;
occupational exposure; sex; epidemiology; Surveillance,
Otolaryngology–
Head and Neck Surgery
2015, Vol. 153(4) 569–574
Ó American Academy of
Otolaryngology—Head and Neck
Surgery Foundation 2015
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0194599815594788
http://otojournal.org
Epidemiology, and End Result (SEER) database; neck cancer;
salivary gland neoplasm, major salivary glands
Received October 15, 2014; revised May 18, 2015; accepted June 16,
2015.
E
pithelial-myoepithelial carcinoma (EMC) is a rare,
low-grade neoplasm of the salivary glands, first
described by Donath et al1 in 1972 but recognized in
the literature as early as 1956.2,3 EMC accounts for approximately 1% of all salivary gland tumors.4-7 The parotid gland
constitutes the majority of cases, followed by the submandibular gland and minor salivary glands. Rare cases have
also been reported in the maxillary sinus, trachea, larynx,
hypopharynx, and even the breast.2,4,8 EMC has a slight
female predominance, and the mean age of diagnosis is 60
years, although cases have been described in patients as
young as 8 years.2,6,7 Although rare, EMC carries an excellent prognosis. The local recurrence rate is estimated to be
approximately 30%, and distant metastases are rare.6,9
Histologically, EMC demonstrates biphasic morphology. A
peripheral layer of myoepithelial cells with clear cytoplasm
surrounds an inner layer of cells resembling intercalated
ducts.2-4,10 Histologic diagnosis can be challenging, however,
as the degree of differentiation between the 2 cell types may
vary considerably. The morphologic appearance of the neoplasm may mimic clear cell carcinoma, pure myoepithelial carcinoma, or even adenoid cystic carcinoma.3,5,7 Several authors
1
Department of Otolaryngology–Head & Neck Surgery, Rutgers New
Jersey Medical School, Newark, New Jersey, USA
2
Center for Skull Base and Pituitary Surgery, Neurological Institute of New
Jersey, Rutgers New Jersey Medical School, Newark, New Jersey, USA
3
Department of Neurological Surgery, Rutgers New Jersey Medical School,
Newark, New Jersey, USA
This article was presented at the 2014 AAO-HNSF Annual Meeting & OTO
EXPO; September 21-24, 2014; Orlando, Florida.
Corresponding Author:
Richard Chan W. Park, MD, Department of Otolaryngology–Head and
Neck Surgery, Rutgers New Jersey Medical School, 90 Bergen St, Suite
8100, Newark, NJ 07103, USA.
Email: [email protected]
Downloaded from oto.sagepub.com at SOCIEDADE BRASILEIRA DE CIRUR on October 2, 2015
570
Otolaryngology–Head and Neck Surgery 153(4)
have also noted the development of sebaceous differentiation
among EMC histology samples.3,6
EMC is an extremely rare tumor, with only 320 cases
having been reported in the literature since it was first
described in 1972.3,4,6 To date, the largest series of EMC
cases reported was a single-institution review of 61 cases
over a span of 30 years.3 To our knowledge, no large-scale,
multi-institution review has been attempted to characterize
this neoplasm. Using the National Cancer Institute’s
Surveillance, Epidemiology, and End Results (SEER) database, we aim to characterize the demographic, clinicopathologic, and survival features of EMC across a North
American population. This study represents not only the
largest cohort EMC studied to date but also the only study
to use a population-based approach.
Materials and Methods
Survival data were obtained from the SEER program database using SEER*Stat 8.1.2 software (National Cancer
Institute, Bethesda, Maryland); specifically, the SEER18
data set (consisting of 18 registries, covering the years
1973-2010) was queried. As SEER does not uncover sensitive patient information, institutional review board (IRB)
approval was not necessary per the standing policy of
Rutgers New Jersey Medical School (NJMS). Using the
International Classification of Diseases for Oncology,
Third Edition (ICD-O-3) codes, we extracted data for the
histologic subcategory of ‘‘epithelial-myoepithelial carcinoma’’ (histologic code 8562/3). We selected only lesions
arising from the major salivary glands, including the parotid gland (anatomic site code C07.9), submandibular gland
(C08.0), and ‘‘other salivary tissue’’ (including miscellaneous or unspecified major salivary glands under site
codes C08.1, C08.8, and C08.9). Minor salivary glands
were not included since the ICD-O-3 provides no precise,
well-defined site code for them. Data were analyzed on the
basis of age, sex, race, histologic grade, tumor size,
American Joint Committee on Cancer (AJCC) stage (for
cases diagnosed in 2004 and later), and treatment (including surgery, radiotherapy, or both). Race was divided into
white, black, and ‘‘other’’ (including American Indian,
Alaskan Native, Asian/Pacific Islander, unspecified, or
unknown). Net survival of the cohort was estimated by
Kaplan-Meier analysis (yielding disease-specific survival,
DSS). Survival and hazard analysis were carried out with
JMP Statistical Discovery 10 (SAS Institute, Cary, North
Carolina). Kaplan-Meier survival curves were compared
using the log-rank test. Hazard analysis was conducted
using the Cox proportional hazards model. Cox proportional hazards analysis did not include site-based analysis
due to the small number of cases in the submandibular
gland, which made the model too unstable. Microsoft
Office Excel 2007 (Microsoft Corporation, Redmond,
Washington) was used for additional data processing. A
probability value (P value) of \.05 was considered statistically significant for all tests.
Results
Analysis of Demographic and Clinicopathologic Factors
A total of 246 cases of salivary gland (SG)–EMC were
reported in the SEER database between 1973 and 2010
(Table 1). Males accounted for 42.68% of the SG-EMC
cases while females accounted for 57.32% of the cases.
Mean 6 SD age at diagnosis was 63.8 6 15.4 years. Whites
made up the majority of the cohort (82.11%), followed by
‘‘other’’ racial groups (9.76%) and blacks (8.13%). Overall,
32.93% of SG-EMC cases were of low histologic grade
(grades I and II), while 6.50% of the cases were of high histologic grade (grades III and IV). Histologic grade was unavailable for the remaining 60.57% of cases. Most SG-EMC
cases were localized to the primary site (68.92%), whereas
21.95% of the cases had spread to regional sites and 4.47%
of the cases had distant metastasis. AJCC staging information was available for 133 cases (ie, those diagnosed after
2004 only): stage I (35.34%), stage II (30.08%), stage III
(16.54%), and stage IV (5.26%). Most cases involved the
parotid gland (83.74%), followed by the submandibular
gland (13.01%). Most patients (59.35%) did not receive any
form of radiotherapy; nearly all patients (97.1%) received
some form of surgery.
Survival Analysis
The SEER18 data set was queried for survival data (Table
2). Net survival was estimated by calculating DSS, which
considers only deaths attributable to the cause of interest (in
this case, SG-EMC). Overall survival was 91.3% at 60
months, 90.2% at 120 months, and 80.7% at 180 months.
When stratified by histologic grade, low-grade SG-EMC
had a better survival at 120 months compared with highgrade SG-EMC (96.7% vs 79.6%; P = .0772). When stratified by tumor size, patients with tumors larger than 4 cm
had a DSS of 58.8% at 180 months, while patients with
tumors smaller than 2 cm had a DSS of 96.8%; patients
with tumors 2 to 4 cm had survival of 85.9% at 180 months
(P = .0003). When stratified by tumor location, patients
with parotid gland EMC had a DSS of 77.8% at 180
months. Patients with submandibular gland EMC had a DSS
of 100.00% at 180 months. However, this difference in DSS
between the 2 tumor locations was not statistically significant (P = .0632). All but 9 of the 207 cases available for
survival analysis underwent surgery. A total of 85 patients
(41.1%) received radiotherapy in addition to surgery.
However, no survival benefit was noted for patients who
received radiotherapy compared with those who did not
(P = .4832).
Cox proportional hazards analysis for SG-EMC was carried out at 60 months (Table 3). Women exhibited a higher
hazard of death than did men; however, this difference was
not statistically significant (hazard ratio [HR], 2.76; 95%
confidence interval [CI], 0.76-13.27; P = .1279). Although
not significant, higher hazard of death was possibly related
to old age (specifically, 75 years or older) compared with
age between 15 and 44 years (HR, 3.08; 95% CI, 0.50-
Downloaded from oto.sagepub.com at SOCIEDADE BRASILEIRA DE CIRUR on October 2, 2015
Vázquez et al
571
Table 1. Demographic and Clinicopathologic Characteristics of
SG-EMC (1973-2010).
Characteristic
Total
Age at diagnosis, mean 6 SD, y
Sex
Male
Female
Race
White
Black
Other
Age, y
0-14
15-44
45-54
55-64
65-74
751
Histologic grade
Low grade (grades I and II)
High grade (grades III and IV)
Grade unknown
SEER LRD stage
Localized
Regional
Distant
Unstaged
AJCC stage (20041 only)b
Stage I
Stage II
Stage III
Stage IV
Unknown
T classification (20041 only)b
T1
T2
T3
T4a
Tx
N classification (20041 only)b
N0
N1
NX
M classification (20041 only)b
M0
MX
Anatomic site
Parotid gland
Submandibular gland
Sublingual gland
Overlapping lesion of major SGs
Major salivary gland NOS
Radiation therapy
Yes
No
Unknown
Valuea
%
246
63.8 6 15.4
100.00
105
141
42.68
57.32
202
20
24
82.11
8.13
9.76
1
25
37
55
65
63
0.41
10.16
15.04
22.36
26.42
25.61
81
16
149
32.93
6.50
60.57
172
54
11
9
69.92
21.95
4.47
3.66
47
40
22
7
17
35.34
30.08
16.54
5.26
12.78
51
43
22
7
10
38.35
32.33
16.54
5.26
7.52
120
5
8
90.23
3.76
6.02
125
8
93.98
6.02
206
32
1
1
6
83.74
13.01
0.41
0.41
2.44
96
146
4
39.02
59.35
1.63
Abbreviations: AJCC, American Joint Committee on Cancer; EMC, epithelialmyoepithelial carcinoma; LRD, localized, regional, distant; NOS, not otherwise
specified; SEER, Surveillance, Epidemiology, and End Result; SG, salivary gland.
a
Values are presented as numbers unless otherwise indicated.
b
Percentages are out of n = 133 (cases diagnosed 2004 and onward).
59.01; P = .2528). Tumors larger than 4 cm carried a higher
hazard of death compared with tumors smaller than 2 cm
(HR, 32.99; 95% CI, 4.46-726.93; P = .0003). In addition,
high-grade histology was associated with an HR of 4.67
(95% CI, 0.40-61.05; P = .2100). However, despite P values
\.05, the broad CIs suggest that these results are not significant. HR for those treated with surgery with adjuvant radiotherapy was 0.75 (95% CI, 0.21-7.81; P = .6562) relative to
surgery alone.
Table 4 shows information on neck dissections based on
nodal status. Of the 120 patients with N0 disease, 53.33%
of the patients received neck dissections, while 46.67% of
the patients did not receive neck dissections. Table 5 shows
radiotherapy DSS analysis for SG-EMC stratified by AJCC
stage. AJCC stages I and II were grouped into the ‘‘early
stage’’ category, while stages III and IV were grouped into
the ‘‘advanced stage’’ category. Although not statistically
significant, patients with an early stage disease who
received radiotherapy had better DSS compared with the
patients who did not receive radiotherapy (100.00% vs
93.75% at 120 months; P = .3496).
Discussion
SG-EMC is a rare malignancy, with just over 320 cases
reported in the literature since its description in 1972. In our
study, we examined a cohort of 246 patients diagnosed with
SG-EMC in the United States; this represents the largest
single study to date. Overall survival for SG-EMC is excellent at 60 months, with 91.3% of patients alive at this time
point. Survival at 120 months and 180 months was 90.2%
and 80.7%, respectively.
Similar to prior reports, we found that SG-EMC occurred
most frequently in females (57.32%), with a female-to-male
ratio of 1.34:1. This is somewhat lower but still similar to
the 1.5:1 ratio reported by Seethala et al3 and significantly
lower than the roughly 2:1 ratio reported in most other
series. As with most neoplasms that exhibit a predilection
for 1 of the 2 sexes, it is unclear why this discrepancy
exists; however, one can surmise that hormonal or other
genetic influences might play a role in the pathogenesis of
these lesions. Univariate regression analysis of our survival
data showed a trend toward worse prognosis among
females; however, this was not statistically significant.
The mean age at diagnosis was 63.8 years (similar to the
60.9 years reported previously).3 Univariate regression analysis of our survival data showed a trend toward worse prognosis among those aged 75 years and older; however, this
was not statistically significant. In our study, lesions preferentially involved the parotid gland (83.74%), at a rate much
greater than that reported by Seethala et al3 (62.1%;
although it should be noted that 4 of these cases were of
minor salivary gland origin, and 2 were extra-salivary).
Univariate regression analysis by anatomic site could not
confirm a statistically significant difference between the
submandibular and parotid glands. It is difficult to draw a
strong conclusion from these given the breadth of the CI
and the low case numbers for submandibular neoplasms (32, or
Downloaded from oto.sagepub.com at SOCIEDADE BRASILEIRA DE CIRUR on October 2, 2015
572
Otolaryngology–Head and Neck Surgery 153(4)
Table 2. Disease-Specific Survival Analysis for Epithelial-Myoepithelial Carcinoma.
Overall survival
Histologic grade
Low grade (grades I and II)
High grade (grades III and IV)
Tumor size, cm
\2
2-4
.4
Location
Parotid gland
Submandibular gland
Treatment modality
Radiation
No radiation
Surgery
Surgery 1 radiation
P Value (Log-Rank)a
n
60 mo, %
120 mo, %
180 mo, %
207
91.33
90.20
80.71
65
14
96.67
79.55
96.67
79.55
90.63
0.00
.0772
45
113
23
96.77
93.04
73.48
96.77
93.04
58.78
96.77
85.88
58.78
.0003
178
25
89.53
100.00
88.13
100.00
77.08
100.00
.0632
86
118
113
85
90.46
91.80
93.23
90.30
87.80
91.80
93.23
87.56
78.04
81.83
86.57
77.83
.8115
.4832
a
Bolding denotes significance.
Table 3. Cox Proportional Hazards Analysis for SG-EMC at 60
Months.a
Hazards Ratio
Age, y
15-44
751
Sex
Male
Female
Race
White
Black
Other
Tumor size, cm
\2
2-4
.4
Histologic grade
Low grade (I and II)
High grade (III and IV)
Treatment modality
Surgery
Surgery 1 radiation therapy
Reference
3.08
Reference
2.76
95% CI
Table 4. Neck Dissection and Nodal Status.
Nodal Status
n
%
P Value
0.50-59.01 .2508
0.76-13.27 .1279
Reference
0.74
\0.0001
0.04-4.33 .7772
0-1.71 .1238
Reference
4.68
32.99
0.62-106.60 .1489
4.46-726.93 .0003
Reference
4.67
0.40-61.05 .2100
Reference
0.75
0.21-2.81 .6562
Abbreviations: CI, confidence interval; EMC, epithelial-myoepithelial carcinoma; SG, salivary gland.
a
Bolding denotes significance.
13.01%). One possible explanation is that due to its intricate
relationship with the facial nerve, surgical resection of the parotid gland is technically more challenging than the submandibular
gland. Therefore, complete resection and obtaining negative
N0
Neck dissection
No neck dissection
N1
Neck dissection
No neck dissection
NX
Neck dissection
No neck dissection
120
64
56
5
4
1
8
3
5
53.33
46.67
80.00
20.00
37.50
62.50
margins for tumors arising from the parotid gland may be more
difficult. Nonetheless, this point is purely speculative.
Although generally regarded as a low-grade neoplasm,
high-grade or dedifferentiated tumors have been described.
In our series, there were 16 cases (6.5%) of high-grade SGEMC. There were significantly more cases of low-grade histology (81, or 32.93%); unfortunately, 60.57% of cases did
not have a reported histologic grade. Regression analysis in
our study showed a near 5-fold greater hazard of death
among high-grade tumors compared with low-grade tumors;
however, this was not statistically significant (P = .2100).
Again, low case numbers (ie, only 16 high-grade lesions)
resulted in a broad CI.
In our survival analysis, we discovered a significant
effect of tumor size on survival. Patients with tumors \2
cm showed survival of 96.77% at 180 months, whereas
those with tumors 2 to 4 cm showed survival of 85.9%
and patients with tumors .4 cm had a survival of 58.8%
Downloaded from oto.sagepub.com at SOCIEDADE BRASILEIRA DE CIRUR on October 2, 2015
Vázquez et al
573
Table 5. Radiotherapy Disease-Specific Survival Analysis for
SG-EMC Stratified by AJCC Stage.
AJCC Stage
Early stage (I and II)
RT
No RT
Advanced stage (III and IV)
RT
No RT
P Value
(Log-Rank)
n
120 mo, %
30
40
100.00
93.75
0.3496
16
7
79.55
100.0
0.4697
Abbreviations: AJCC, American Joint Committee on Cancer; EMC,
epithelial-myoepithelial carcinoma; RT, radiotherapy; SG, salivary gland.
(P = .0003). Given the low case numbers for clinical stage
group (ie, contemporary AJCC conventions), we were unable
to conduct survival analysis based on these factors. In this
case, tumor size serves as a proxy for clinical stage and very
clearly shows a strong influence on survival. It should be
noted, however, that even very large tumors showed survival
of greater than 70% at 5 years.
In our cohort, radiotherapy was administered to 39.02%
of patients (and not administered to 59.35%). There was no
survival advantage to administering radiotherapy (up to 180
months). Moreover, there was no difference in survival
between those treated with surgery only (93.2% at 10 years)
and those treated with surgery and adjuvant radiotherapy
(87.6%, P = .4832). Although it is tempting to conclude that
surgery alone is sufficient for the treatment of this lowgrade tumor, it is ultimately impossible to draw strong conclusions regarding treatment using registry-based data. Most
authors agree that surgical resection is the treatment of
choice for SG-EMC. It is unclear whether adjuvant radiotherapy is beneficial, although a prior study has argued that
it may be effective in preventing local recurrence.11 As seen
in Table 5, the survival benefit of radiotherapy took stage
into account. However, other adverse features such as grade
and positive margins were not taken into account in this
analysis. Hence, it is difficult to determine the true survival
benefit of radiotherapy. Given the nature of our data, it is
impossible to evaluate the role of combination therapy on
DSS.
To our knowledge, our study represents the first largescale, population-based analysis of SG-EMC. However, certain limitations inherent to a population-based data set such
as SEER must be taken into consideration. ICD-O-3 histological and topographical codes were used to select our
patient population from the SEER database. However, one
limitation of using such a coding system is that it introduces
the possibility of inaccuracy through miscoding of tumors.
Retrospective population-based studies are dependent on
accurate coding and consistent data collection among
numerous sites, which can be imperfect. In addition, SEER
does not contain complete information regarding the extent
of therapeutic intervention, such as the number of cycles of
radiotherapy applied or the specific surgical techniques
used. Likewise, data about chemotherapy were unavailable
for analysis, even though some of the patients in our study
might have received chemotherapy. Furthermore, in our
study, histologic grade was available only for 97 (39.9%) of
243 cases. Similarly, AJCC stage and TNM classification
information were available for 133 cases diagnosed from
2004 and onward. Such a small sample size is another limitation of analyzing a rare tumor such as SG-EMC.
Conclusion
This report represents the largest series of EMC to date.
SG-EMC is a low-grade tumor with good overall survival
(approximately 80% at 15 years). There is no apparent benefit to the use of adjuvant radiotherapy in terms of DSS.
Author Contributions
Alejandro Vázquez, acquisition of data, analysis, drafting, final
approval; Tapan D. Patel, acquisition of data, analysis, drafting,
final approval; Christine M. D’Aguillo, acquisition of data, analysis, revision, final approval; Rami Y. Abdou, acquisition of data,
revision, final approval; William Farver, acquisition of data, analysis, revision, final approval; Soly Baredes, conception, revision,
final approval; Jean Anderson Eloy, conception, revision, final
approval; Richard Chan W. Park, conception, design, analysis
and interpretation of data, revision, final approval.
Disclosures
Competing interests: None.
Sponsorships: None.
Funding source: None.
References
1. Donath K, Seifert G, Schmitz R. Diagnosis and ultrastructure
of the tubular carcinoma of salivary gland ducts: epithelialmyoepithelial carcinoma of the intercalated ducts [in German].
Virchows Arch A Pathol Pathol Anat. 1972;356:16-31.
2. Angiero F, Sozzi D, Seramondi R, Valente MG. Epithelialmyoepithelial carcinoma of the minor salivary glands: immunohistochemical and morphological features. Anticancer Res.
2009;29:4703-4709.
3. Seethala RR, Barnes EL, Hunt JL. Epithelial-myoepithelial
carcinoma: a review of the clinicopathologic spectrum and
immunophenotypic characteristics in 61 tumors of the salivary
glands and upper aerodigestive tract. Am J Surg Pathol. 2007;
31:44-57.
4. Esposito E, Cassiano B, Cinquegrani F. Salivary glands: report
of a rare case of myoepithelial carcinoma involving tongue
base treated by CO2 laser. Acta Otorhinolaryngol Ital. 2009;
29:156-159.
5. Miliauskas JR, Orell SR. Fine-needle aspiration cytological
findings in five cases of epithelial-myoepithelial carcinoma of
salivary glands. Diagn Cytopathol. 2003;28:163-167.
6. Shinozaki A, Nagao T, Endo H, et al. Sebaceous epithelialmyoepithelial carcinoma of the salivary gland: clinicopathologic and immunohistochemical analysis of 6 cases of a new histologic variant. Am J Surg Pathol. 2008;32:913-923.
Downloaded from oto.sagepub.com at SOCIEDADE BRASILEIRA DE CIRUR on October 2, 2015
574
Otolaryngology–Head and Neck Surgery 153(4)
7. van Tongeren J, Creytens DH, Meulemans EV, de Bondt RB, de
Jong J, Manni JJ. Synchronous bilateral epithelial-myoepithelial
carcinoma of the parotid gland: case report and review of the literature. Eur Arch Otorhinolaryngol. 2009;266:1495-1500.
8. Teppo H, Paronen I. Epithelial-myoepithelial carcinoma in
minor salivary gland of the hard palate. J Craniofac Surg.
2008;19:1689-1691.
9. Senis-Segarra L, Sahuquillo-Arce E, Davo R, Hamad-Arcis P,
Floria-Garcia LM, Baquero MC. Salivary gland epithelialmyoepithelial carcinoma: behaviour, diagnosis and treatment.
Med Oral. 2002;7:391-395.
10. Wax T, Layfield LJ. Epithelial-myoepithelial cell carcinoma
of the parotid gland: a case report and comparison of cytologic
features with other stromal, epithelial, and myoepithelial cell
containing lesions of the salivary glands. Diagn Cytopathol.
1996;14:298-304.
11. Deere H, Hore I, McDermott N, Levine T. Epithelial-myoepithelial carcinoma of the parotid gland: a case report and
review of the cytological and histological features. J Laryngol
Otol. 2001;115:434-436.
Downloaded from oto.sagepub.com at SOCIEDADE BRASILEIRA DE CIRUR on October 2, 2015