Modern Pathology (2012) 25, 919–929

& 2012 USCAP, Inc. All rights reserved 0893-3952/12 $32.00 919

DOG1: a novel marker of salivary acinar and

intercalated duct differentiation
Jacinthe Chênevert1, Umamaheswar Duvvuri2, Simion Chiosea1, Sanja Dacic1,
Kathleen Cieply1, Jean Kim2, Daniel Shiwarski2 and Raja R Seethala1,2
1
Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA, USA and
2
Department of Otolaryngology, University of Pittsburgh Medical Center, Pittsburgh, PA, USA

Anoctamin-1 (ANO1) (DOG1, TMEM16a) is a calcium-activated chloride channel initially described in gastrointestinal
stromal tumors, but now known to be expressed in a variety of normal and tumor tissues including salivary tissue in
murine models. We herein perform a comprehensive survey of DOG1 expression in 156 cases containing non-
neoplastic human salivary tissues and tumors. ANO1 mRNA levels were significantly higher (8-fold increase,
Po0.0001) in normal parotid tissue (n ¼ 6) as compared with squamous mucosa (n ¼ 15). By immunohistochemistry,
DOG1 showed a diffuse moderate (2 þ ) apical membranous staining pattern in normal serous acini, 1 þ apical
membranous pattern in mucous acini, and variable 1–2 þ apical staining of distal intercalated ducts. Myoepithelial
cells, striated and excretory ducts were invariably negative. All acinic cell carcinomas (n ¼ 28) were DOG1 positive
demonstrating a complex mixture of intense (3 þ ) apical membranous, cytoplasmic and complete membranous
staining. Most ductal tumor types were negative or only showed a subset of positive cases. Within the biphasic
tumor category, adenoid cystic carcinomas (18/24 cases) and epithelial–myoepithelial carcinomas (8/15 cases) were
frequently positive, often showing a distinctive combined apical ductal and membranous/cytoplasmic myoepithelial
staining profile. Thus, DOG1 staining is a marker of salivary acinar and to a lesser extent intercalated duct
differentiation. Strong staining can be used to support the diagnosis of acinic cell carcinoma. DOG1 may also be a
marker of a ‘transformed’ myoepithelial phenotype in a subset of biphasic salivary gland malignancies.
Modern Pathology (2012) 25, 919–929; doi:10.1038/modpathol.2012.57; published online 30 March 2012

Keywords: acinic cell carcinoma; DOG1; salivary gland

The gene now known as anoctamin-1 (ANO1, also surgical and even cytologic specimens.2,3,9 However,
known as FLJ10261, TMEM16A, discovered on GIST1, the discovery of its calcium-activated chloride
or DOG1) was initially noted by gene expression channel properties has suggested a potential role
profiling to be differentially expressed in gastrointest- in secretory cell types such as those of the salivary
inal tumors (GIST)1 when compared with a variety of gland, and thus perhaps tumors derived from these
other mesenchymal tumors. Since then, the expres- cell types. Interestingly, studies in murine models
sion of the ANO1, more commonly known as DOG1, have demonstrated that DOG1 is not only present
by immunohistochemistry has been validated on but also required for normal salivary gland secretory
large series of GIST, but has also been noted in other activity.10–12 However, DOG1 expression has not
tumor types such as esophageal and head and neck been well studied in human salivary tissues. Only
squamous cell carcinoma.2–4 Its main function as a a few normal salivary glands and salivary tumors
calcium-activated chloride channel was uncovered in have been stained, mainly in the context of broad
2008.5–7 Structurally, DOG1 protein is predicted to immunohistochemical surveys in reference to its
contain eight transmembrane segments, hence the performance as a marker of GIST.2
official designation ANO1.8 The aim of this study is to characterize DOG1
To date, the main clinical application of DOG1 expression patterns in normal human salivary
immunostaining has been as a marker of GIST, in tissues and a broad variety of salivary gland tumors.

Correspondence: Dr RR Seethala, MD, Department of Pathology,

University of Pittsburgh Medical Center, Scaife Hall A616.3, 200
Lothrop Street, Pittsburgh, PA 15213, USA. Materials and methods
E-mail: [email protected]
Received 24 August 2011; revised 15 February 2012; accepted 16 This study was approved by our institutional review
February 2012; published online 30 March 2012 board (IRB#PRO07050360).

www.modernpathology.org
 DOG1 in salivary gland
920 J Chênevert et al

qRT-PCR for ANO1 mRNA glands, and one containing normal serous (Von
Ebner) and mucinous glands of the base of tongue.
In all, 6 normal parotid and 15 non-cancerous
squamous mucosal snap-frozen tissue specimens
(stored at 80 1C) were retrieved from the institu-
DOG1 Immunohistochemistry
tional tissue biorepository for comparative analysis.
ANO1 mRNA levels were evaluated by quantitative Formalin-fixed paraffin-embedded sections were
reverse transcription PCR (qRT-PCR) Taqman pri- mounted, and serially sectioned at 4 mm intervals.
mers and probes were designed with the PRIMER Sections were deparaffinized and subjected to heat-
EXPRESS V.2.0.0 program (Applied Biosystems, induced epitope retrieval in citrate buffer. Immuno-
Foster City, CA). The reverse transcriptions were histochemical staining was performed using DOG1
carried out as previously described.13 Quantitative (Clone 1.1; Zeta Co, Sierra Madre, CA; dilution 1:50).
PCR (qPCR) was performed on the cDNA using the Labeling was performed using the I-view 20 -diami-
ABI 7700 Sequence Detection Instrument (Applied nobenzamide (DAB) detection kit (Ventana systems,
Biosystems) and analyzed using the relative quanti- Tucson, AZ) as the brown chromogen substrate.
fication method. qPCR was performed for ANO1 and Staining parameters evaluated included: cell type,
GAPDH was used as an endogenous control. For the % of cells stained, intensity (scale: 0–3), and
qPCR, the final concentrations of the reaction subcellular localization: membranous—including lo-
components were as follows: 1  Taqman Universal cation and extent (ie, apical-luminal, basolateral, and
PCR Master Mix (Applied Biosystems), 500 nM complete), cytoplasmic, and other. These parameters
Forward (GAGCCAAAGACATCGGAATCTG) and were collected for the tumor and also for adjacent
Reverse (TGAAGGAGATCACGAAGGCAT) primers, normal tissue if it was present in the same section.
200 nM Florescent Probe (FAM—CTCAGAGGCA Case was considered as ‘negative’ or as having
TTGGGAAGCTTGCTGT—TAMARA), and DNAse- ‘negligible staining’ if o2% of the tumor expressed
free H2O to a final volume of 25 ml. The primer and DOG1, as ‘focal’ if between 2 and 50%, and as
probe concentration have been described pre- ‘diffuse’ if 450% had staining. As for the intensity of
viously.4 The thermocycler conditions were 95 1C staining, the apical staining of normal parotid serous
Taq activation for 12 min and 40 cycles of 95 1C acini was used as 2 þ ; a more intense staining was
denaturation for 15 s followed by 60 1C anneal/ graded as 3 þ and weaker staining as 1 þ .
extend for 60 s. The consistency of amplification
was judged by inspection of the melting curves. The
DDCt method was used to determine relative ETV6-NTRK3 Fluorescence In-Situ Hybridization
expression. A standard curve was also performed
to evaluate the efficiency of the qPCR experiment. Six cases morphologically compatible with the
The differences in relative expression were deter- diagnosis of mammary analog secretory carcinomas
mined for each sample, run in triplicate, using a were confirmed for an ETV6-NTRK3 translocation as
Student’s t-test via GraphPad Prism software. described previously14–16 using ETV6 break apart
probe set.
Briefly, formalin-fixed paraffin-embedded sections
Case Selection for DOG1 Immunohistochemical were mounted, and serially sectioned at 5 mm inter-
Validation vals and incubated with the Vysis LSI ETV6 (TEL)
A total of 156 formalin-fixed paraffin-embedded (12p13) Dual Color, Break Apart Rearrangement
cases were retrieved from the University of Pitts- Probe overnight at 37 1C in a humidified chamber
burgh Medical Center pathology archives. The and counterstained with DAPI (Vysis). Analysis was
neoplastic cases (n ¼ 150) retrieved were as follows: performed using an Applied Imaging Workstation
28 acinic cell carcinomas, 24 adenoid cystic carci- equipped with Chroma Technology filters containing
nomas, 16 polymorphous low-grade adenocarcino- band excitors for SpectrumOrange, FITC, and DAPI.
mas, 15 epithelial - myoepithelial carcinomas, 14 Only individual and well-delineated cells were
pleomorphic adenomas, 12 myoepithelial tumors scored, overlapping cells were excluded and B60
(4 myoepitheliomas/8 myoepithelial carcinomas), 10 cells were analyzed in the targeted region. A case was
oncocytic tumors (8 oncocytomas/2 oncocytic carci- considered positive if a split green and orange signal
nomas), 9 salivary duct carcinomas, 8 mucoepider- was identified in 420% of the nuclei.
moid carcinomas, 6 mammary analog secretory
carcinomas, and 8 others (2 Warthin tumors, 2
low-grade salivary duct carcinomas/low-grade cri- Results
briform cystadenocarcinomas, 1 adenosquamous ANO1 mRNA Expression in Salivary Frozen Tissues
carcinoma, and 1 sebaceous lymphadenoma). Six
additional normal or non-neoplastic cases were also ANO1 mRNA levels were significantly higher
retrieved: three submandibular glands with chronic (8-fold increase, Po0.0001) in normal parotid tissue
sialadenitis, one parotid gland with sialadenosis, as compared with normal squamous epithelium
one nasal septum biopsy with normal mucoserous (Figures 1a and b).

Modern Pathology (2012) 25, 919–929

 DOG1 in salivary gland
J Chênevert et al 921

DOG1 Immunohistochemical Staining Profile in was weaker than the apical pattern, with an intensity
Normal, Non-Neoplastic Salivary Gland of 1 þ or 2 þ . Two acinic cell carcinomas (7%; 2/28)
showed only focal DOG1 staining. However, these
Normal salivary gland tissue was present in 109 cases had a prominence of vacuolated and non-
cases (103 peritumoral normal cases and 6 non- specific ductal cells with a paucity of acinar cells
neoplastic cases). Staining was greatest in salivary (Figures 3c and d).
acini. Regardless of salivary subsite, normal serous Six cases of mammary analog secretory carcinoma
acini uniformly demonstrated a sharp, crisp, mod- (Figure 4a), a new entity historically categorized as
erately (2 þ ) intense membranous staining in an acinic cell carcinomas, were confirmed by fluores-
apical-luminal pattern with occasional basolateral cence in-situ hybridization (FISH) (Figure 4b) and
staining (Figures 2a and c). Mucous acini showed a subsequently tested for DOG1. Staining was ob-
similarly well-delineated apical pattern of membra- served in 2/6 (33%) of cases. One case showed fairly
nous staining, but with less intensity (1 þ , Figures diffuse (80%) staining reminiscent of acinic cell
2b and c). Intercalated ducts were variably positive carcinomas with 2 þ apical membranous and 1 þ
in an apical-luminal membranous pattern. Staining cytoplasmic staining pattern. A second case was
of intercalated ducts was more frequent in the more focally positive (40%) with faint (1 þ ) apical and
distal portions toward the acini and ranged from 1 þ complete membranous staining. The remainder of
to 2 þ (Figure 2a, single arrow). In all cases, striated mammary analog secretory carcinomas were nega-
ducts (Figure 2a, letter ‘x’) and more proximal larger tive (Figure 4c).
excretory ducts were negative for DOG1. No normal Of the other phenotypically ductal tumors, poly-
constituent showed cytoplasmic staining, or com- morphous low-grade adenocarcinoma most fre-
plete membranous staining. Three cases of sialade- quently demonstrated DOG1 immunopositivity,
nitis and one case of sialadenosis (Figure 2d) present in 31% (5/16) cases; however, staining was
showed a similar staining pattern in each of the very focal in all but one case. Staining was not nearly
ductoacinar components. as strong as in acinic cell carcinomas and was more
reminiscent of normal intercalated ducts with a 1–2 þ
apical-luminal pattern of staining. The single diffu-
DOG1 Expression in Acinar and Ductal Tumors sely positive case also showed areas of more complete
membranous staining (Figures 5a and b). There were
Immunostaining of all salivary tumors is summar-
no striking differences in histologic growth pattern
ized in Table 1. All 28 acinic cell carcinomas showed
(ie, proportion of papillary components) between the
DOG1 staining, and 93% (26/28) tumors showed
positive and negative polymorphous low-grade ade-
diffuse (450%) staining. Overall, staining was more
nocarcinoma. All nine salivary duct carcinomas
intense and more complex than in normal acini with
(Figures 5c and d), eight oncocytomas, two oncocytic
25/28 (89%) of acinic cell carcinomas showing 3 þ
carcinomas, and two low-grade salivary duct carci-
apical/luminal membranous staining as well as
nomas (a.k.a. low-grade cribriform cystadenocarcino-
scattered foci of complete membranous and even
mas) tested were negative for DOG1.
cytoplasmic staining (Figures 3a and b). Acinar cell
types stained most intensely, while intercalated
ductal, non-specific ductal, and vacuolated cells DOG1 Expression in Biphasic Salivary Gland Tumors
were more variable and somewhat weaker in DOG1
staining intensity, and in some areas negative. When Twenty-four adenoid cystic carcinomas were eval-
present, the membranous and cytoplasmic staining uated: 4 tubular pattern predominant, 14 cribriform

Figure 1 qRT-PCR amplification of ANO1 in normal parotid tissue and normal squamous mucosa. (a) Parotid tissue samples (curve
aggregate 1) show a Ct that is earlier that of the normal squamous mucosa (curve aggregate 2). (b) Relative expression levels of ANO1
mRNA showing an 8-fold higher level of expression in parotid tissue as compared with normal squamous mucosa (***Po0.0001). Box
edge—mean value, whiskers standard error of mean.

Modern Pathology (2012) 25, 919–929

 DOG1 in salivary gland
922 J Chênevert et al

Figure 2 DOG1 staining pattern in normal salivary gland tissues (all  400). (a) Serous acini of parotid with moderate (2 þ ) apical-
luminal membranous staining and weaker (1 þ ) intercalated duct (arrow) staining. Striated ducts are negative (x mark). (b) Mucous acini
in a minor salivary gland with 1 þ apical staining. (c) Sinonasal mucoserous glands with stronger expression in serous acini compared
with mucous acini. (d) Sialadenosis of parotid showing almost equivalent DOG1 intensity as compared with normal parotid.

pattern predominant, 3 solid pattern predominant, In all, 15 epithelial–myoepithelial carcinomas

and 3 adenoid cystic carcinoma with high-grade were evaluated: 14 cases had classic clear cell
transformation. Overall, 75% (18/24) adenoid cystic morphology, while 1 case was an oncocytic-sebac-
carcinoma stained for DOG1, and 38% (9/24) eous variant. Three of the classic epithelial–myo-
showing diffuse staining. When present, staining epithelial carcinomas arose in a pleomorphic ade-
was frequently noted in both ductal and myoepithe- noma, and one was associated with intercalated
lial cells (Figures 6a and b). However, the ductal duct hyperplasia (see also below). Epithelial–myo-
components showed an apical-luminal staining epithelial carcinoma frequently showed DOG1 po-
pattern similar to normal intercalated ducts (1–2 þ sitivity with 53% (8/15) showing positivity, and
intensity), while the myoepithelial cells showed 47% (7/15) showing diffuse positivity. Staining was
diffuse cytoplasmic staining (2 þ intensity). The similar to that seen in adenoid cystic carcinoma
tubular and cribriform predominant adenoid cystic (Figures 6c and d). When positive, the ductal
carcinoma was more frequently DOG1 positive components showed 1–2 þ apical-luminal staining,
(89%, 16/18 cases) as compared with solid adenoid while the myoepithelial components showed 12 þ
cystic carcinoma and adenoid cystic carcinoma with complete membranous or cytoplasmic staining.
high-grade transformation (33%, 2/6 cases). One There were no morphologic differences between
case of adenoid cystic carcinoma with high-grade positive and negative cases. The oncocytic-sebac-
transformation did show some DOG1 positivity, but eous epithelial–myoepithelial carcinoma showed
only in the adjacent conventional adenoid cystic staining in only the myoepithelial component with
carcinoma component. 2 þ complete membranous positivity.

Modern Pathology (2012) 25, 919–929

 DOG1 in salivary gland
J Chênevert et al 923

Table 1 DOG1 staining in salivary gland neoplasms staining and 1 þ cytoplasmic staining of myoepithe-
lial cells as well (Figures 6e and f).
Diagnosis Negative Focal Diffuse Total
or r 2% positivity positivity
(3–50%) (450%)
Discussion
Acinic cell carcinoma — 2 26 28
Mammary analog 4 1 1 6 The discovery of the function of ANO1 (DOG1) as a
secretory carcinoma calcium-dependent chloride channel suggests poten-
Polymorphous low-grade 11 4 1 16 tial roles for this protein beyond that of a marker of
adenocarcinoma
Salivary duct carcinoma 9 — — 9 GIST. Indeed, recently, DOG1 has been characterized
Adenoid cystic carcinoma 7a 8 9 24
in pancreatic centroacinar cells and a subset of islet
Epithelial-myoepithelial 7 1 7 15 cells suggesting a potential exocrine/endocrine se-
carcinoma cretory role.17,18 With respect to salivary gland, DOG1
Pleomorphic adenoma 13 1 — 14 has been shown to be expressed and perhaps even
Mucoepidermoid 5 3 — 8 required for normal salivary secretion in murine
carcinoma
Myoepithelioma/ 4/7 /1 12 models.10–12 Thus perhaps not surprisingly, we were
myoepithelial carcinoma able to confirm high levels of ANO1 mRNA and
Oncocytoma/oncocytic 8/2 — — 10 DOG1 protein by immunohistochemistry in human
carcinoma salivary tissues as well in this study, which repre-
Other 6b 1c 1d 8 sents the first comprehensive survey of non-neoplas-
tic salivary tissues and salivary gland neoplasms.
a
Includes three high-grade transformations of adenoid cystic The high level of DOG1 expression in salivary
carcinomas.
b tissues was found to be localized immunohisto-
Two Warthin tumors, two low-grade cribriform cystadenocarcinomas,
one adenosquamous carcinoma, and one sebaceous lymphadenoma. chemically mainly to salivary serous acini, which
c
One basal cell adenocarcinoma. invariably showed an apical pattern of staining of
d
One intercalated duct adenoma. moderate intensity. Mucous acini also consistently
showed a similar pattern but with lesser intensity.
This staining diminished at the level of the inter-
Of the other biphasic salivary gland neoplasms, the calated ducts and was completely absent more
one basal cell adenocarcinoma tested was focally proximally. This pattern of staining is in keeping
positive (40% of tumor) in the basal cell component in with the function of DOG1 as a transmembrane
a 2 þ cytoplasmic and membranous pattern; predo- anion channel with a secretory role. Of note, no
minant growth patterns were solid and membranous. cytoplasmic staining was noted in normal salivary
Only 1/13 (8%) pleomorphic adenoma showed even gland. This is in contrast to other cell types in which
focal (5%) DOG1 staining, with a 1 þ apical staining DOG1 is constitutively expressed such as interstitial
pattern in the ductal component only. cells of Cajal where cytosolic localization of DOG1 is
typical.19,20 Potential reasons for this may include a
Other Salivary Gland Neoplasms and Precursor differential distribution of isoforms, and differences
Lesions in post-translational modification and processing in
the endoplasmic reticulum and Golgi apparatus, and
In all, 38% (3/8) of the mucoepidermoid carcinomas potentially even the presence in some cell types of
(1 low grade and 2 intermediate grade) studied lipid microvacuoles with DOG1 that are exocytosed
showed focal (3–20%) apical 1 þ staining in the in response to increases in cytosolic calcium.11,19
mucous cell components. Only 14% (1/7) myoepithe- As expected, acinic cell carcinomas are the tumor
lial carcinomas with spindle cell predominant histol- types with the strongest expression of DOG1. In
ogy showed DOG1 staining, but only focally with 1 þ contrast to normal acini, staining was noted to be
cytoplasmic staining in 20% of the tumor. All four more intense and more complex often with cyto-
myoepitheliomas were negative. All Warthin tumors plasmic staining. This study was not intended to
and other tested tumors were negative. address the potential reasons for this apparent
Interestingly, four cases showed putative precursor increase in DOG1 staining. The high expression in
lesions (three intercalated duct hyperplasias and one acinic cell carcinomas may simply represents an
intercalated duct adenoma). The intercalated duct ‘exaggerated acinar’ phenotype rather than a specific
hyperplasias were seen adjacent to or in association event such as gene amplification as seen in some
with tumors mentioned above (one mammary analog tumors with DOG1 overexpression. For instance in
secretory carcinoma, one epithelial–myoepithelial squamous cell carcinoma of the head and neck, the
carcinoma, and one cystadenoma). The intercalated chromosome 11q13 region in which the ANO1 gene
duct adenoma on the other hand was seen in resides is frequently amplified suggesting a poten-
isolation. All precursor lesions were positive for tial mechanism for DOG1 overexpression in a subset
DOG1 and 3/4 (75%) showed diffuse positivity. All of these tumors.4,21 However, this region has not
cases showed apical staining ranging from 1 to 3 þ . been described to be amplified in salivary acinic cell
The intercalated duct adenoma showed 3 þ apical carcinomas.

Modern Pathology (2012) 25, 919–929

 DOG1 in salivary gland
924 J Chênevert et al

Figure 3 DOG1 in acinic cell carcinoma. (a) Acinic cell carcinoma, with abundant acinar cell components (H&E,  200). (b) DOG1
staining showing intense (3 þ ) apical membranous staining around lumina as well as complete membranous and variable cytoplasmic
staining (12 þ ) (  400). Inset: adjacent non-neoplastic parotid for comparison (  400). (c) Acinic cell carcinoma with predominant
vacuolated and non-specific ductal cells and only scattered acinar cells (H&E,  200). (d) DOG1 staining showing only focal 2 þ apical
positivity (left) and faintly positive 1 þ to negative complete membranous staining in the vacuolated areas (right) (  400).

Regardless of mechanism, our findings would for acinar differentiation is very low. Thus, DOG1
indicate that this consistently strong DOG1 staining staining offers a sensitive and robust marker to
can be utilized diagnostically to support the diag- support the diagnosis of acinic cell carcinomas.
nosis of acinic cell carcinomas. Typically, acinic cell The differential diagnosis for acinic cell carcino-
carcinomas show readily recognizable serous acinar mas can potentially be quite broad depending on
differentiation on a routine hematoxylin and eosin- histologic pattern and cell type predominance;
stained slide. When this cell type is less prominent, however, in most cases, it can be resolved to tumors
traditionally, the histochemical stain using period that are phenotypically ductal. Thus when consid-
acid-Schiff reaction after diastase digestion can be ering this subgroup of tumors, DOG1 also appears to
used to highlight zymogen granules. However show a good deal of specificity, with only occasional
occasionally, even these may be scarce, and in some ductal tumors showing immunopositivity that ap-
cases, non-acinic cell carcinomas may show small proaches the extent seen in acinic cell carcinomas.
globules of intracytoplasmic mucin or even hemo- For instance, all salivary duct carcinomas, oncocy-
siderin that may be periodic acid-Schiff positive and tomas, oncocytic carcinomas, and low-grade sali-
diastase resistant. In such situations, it is appro- vary duct carcinomas tested were negative. Only a
priate to exclude these ‘zymogen granule mimics’ by subset of polymorphous low-grade adenocarcinoma
performing a mucicarmine, and in some instances, was DOG1 positive, with only one diffusely positive
an iron stain. Anti-amylase immunostains may be tumor. Interestingly, the focality and heterogeneity
useful for the diagnosis of acinic cell carcinomas if of reactivity in polymorphous low-grade adenocar-
positive;22,23 however, the sensitivity of this marker cinoma is reminiscent of the normal staining profile

Modern Pathology (2012) 25, 919–929

 DOG1 in salivary gland
J Chênevert et al 925

intercalated duct region based on immunohisto-

chemical and ultrastructural findings, and one of the
initial names for this entity was ‘terminal duct
carcinoma’.24,25 Other tumors that occasionally enter
into the differential diagnosis such as mucoepider-
moid carcinomas were only focally and weakly
positive, mainly in mucocytes. Only one myoepithe-
lial tumor showed any staining for DOG1, and this
tumor did not have epithelioid, clear cell, or
oncocytic morphology that would even raise the
consideration for acinic cell carcinomas. Warthin
tumors that rarely enter into the differential diag-
nosis were also uniformly negative. One caveat is
that we did not test any ‘cystadenocarcinomas, not
otherwise specified,’ which as a class of lesions can
show overlap with papillary cystic variant of acinic
cell carcinoma.
Perhaps, most interesting to the differential
diagnosis of acinic cell carcinomas is the recently
described entity, mammary analog secretory carci-
nomas. Historically, mammary analog secretory
carcinomas were most often classified as ‘zymogen
granule poor’ acinic cell carcinomas or as adeno-
carcinoma, not otherwise specified.14–16 These
carcinomas are composed of cells with moderate-
to-abundant eosinophilic, vacuolated cytoplasm
show a lobulated growth pattern with solid, papil-
lary, microcystic, or glandular spaces filled with
dense eosinophilic mucinous material. Tumors are
characteristically strongly S100, vimentin, mamma-
globin, and STAT5a positive 14–16,26 No true serous
acinar differentiation is noted in these tumors.
Mammary analog secretory carcinomas also recapi-
tulate the morphology of juvenile secretory carcino-
ma of the breast (hence the name, mammary analog
secretory carcinoma) and harbor the same ETV6-
NRTK3 translocation. The lack of true serous acinar
differentiation, the characteristic morphology, and a
reproducible translocation justify separation of
mammary analog secretory carcinomas from acinic
cell carcinomas. And though we only had a few
cases to test, the DOG1 staining profile in mammary
analog secretory carcinomas was indeed different
from acinic cell carcinomas, and more in keeping
with an acinar-intercalated duct junction pheno-
type. Only one case showed staining reminiscent of
acinic cell carcinomas. Thus, DOG1 adds yet
another marker to the armament (including muci-
carmine, S100, and mammaglobin) that is readily
available to help distinguish mammary analog
Figure 4 Mammary analog secretory carcinoma. (a) Tumor shows secretory carcinomas from acinic cell carcinomas.
cribriform nests of cells with moderate amounts of somewhat
vacuolated eosinophilic cytoplasm and luminal secretions remi- In completing our survey of salivary gland tumors,
niscent of non-specific ductal cells and vacuolated cells in acinic we encountered a high frequency of DOG1 positivity
cell carcinoma (H&E,  200). (b) ETV6 FISH break apart probe in adenoid cystic carcinoma and epithelial–myo-
showing cells with one split orange signal indicative of a epithelial carcinoma, though not as consistently as
translocation (arrows). (c) Most mammary analog secretory
carcinomas were DOG1 negative (  200).
seen in acinic cell carcinomas. Our findings with
adenoid cystic carcinoma are somewhat similar to
those of Lopes et al27 who found positivity in 3/4
of distal intercalated ducts. Polymorphous low- tumors using the K9 antibody. Additionally,
grade adenocarcinoma is one of a few tumors we found that more solid adenoid cystic carcinoma
thought to recapitulate the phenotype of distal and adenoid cystic carcinoma with high-grade

Modern Pathology (2012) 25, 919–929

 DOG1 in salivary gland
926 J Chênevert et al

Figure 5 DOG1 staining in other ductal neoplasms. (a) Polymorphous low-grade adenocarcinomas (H&E,  200). (b) DOG1 was focally
positive in a subset of tumors in an apical membranous pattern. This particular case was diffusely positive (2 þ ) with some complete
membranous staining as well (  400). (c) Salivary duct carcinoma (H&E,  200). (d) All tested cases were negative. Note the perineural
invasion in this case (bottom right) (  400).

transformation were more frequently DOG1 negative branous pattern reminiscent of normal intercalated
as compared with tubular and cribriform predomi- ducts. But surprisingly, the myoepithelial compo-
nant adenoid cystic carcinoma. Adenoid cystic nents were frequently positive and showed a
carcinoma and epithelial–myoepithelial carcinoma complete membranous and even cytoplasmic
tumors fit into the ‘biphasic’ category of salivary pattern of DOG1 reactivity in adenoid cystic
gland tumors. These are tumors that comprising a carcinoma and epithelial–myoepithelial carcino-
bilayered arrangement of inner (luminal) ductal and ma. This contrasts with normal salivary gland
outer (abluminal) basal and/or myoepithelial cells. myoepithelial and basal cells, which were uni-
This pattern is often readily visible by routine formly negative suggesting that this is a ‘trans-
light microscopy, and by immunostaining shows a formed’ phenotype in salivary tumors. It must be
sharp distinction between luminal and abluminal noted, however, that this may not hold true at
components. As such, neither adenoid cystic carci- other sites with myoepithelial cells. For instance,
noma nor epithelial–myoepithelial carcinoma is this pattern of staining has been described in
usually in the differential diagnosis for acinic cell normal basal/myoepithelial cells in breast and
carcinomas (a ‘monophasic’ acinar malignancy) and prostate.2,27 Any role for an anion channel such as
do not invalidate the use of DOG1 as discussed DOG1 in myoepithelial cells, normal or neoplastic
above. at any site is currently unknown.
However, this is still of interest, particularly given Another interesting finding was that this dual
the unique pattern of staining identified. The ductal staining pattern was mainly restricted, in our study,
constituents of these two tumor types, when to biphasic malignant tumors and a few putative
positive, showed a heterogeneity and apical mem- precursor lesions. One case of intercalated duct

Modern Pathology (2012) 25, 919–929

 DOG1 in salivary gland
J Chênevert et al 927

Figure 6 DOG1 staining in adenoid cystic carcinomas, epithelial–myoepithelial carcinomas, and intercalated duct lesions. (a) Adenoid
cystic carcinoma, cribriform pattern (H&E,  200). (b) DOG1 frequently showed an apical membranous staining pattern (2 þ in this case)
in the ductal component (arrows), and a cytoplasmic pattern in the abluminal myoepithelial cells (1 þ in this case) (  400). (c) Epithelial
- myoepithelial carcinoma with a classic clear cell myoepithelial morphology (H&E,  200). (d) Similarly to adenoid cystic carcinomas,
epithelial - myoepithelial carcinoma had moderate apical-luminal DOG1 staining (2 þ arrows) complete membranous staining of the
abluminal myoepithelial cells (2 þ ) (  400). (e) A 3-mm intercalated duct adenoma with partial encapsulation (H&E,  20). Inset shows
transition of adenoma from adjacent acini. The proliferation consists of ducts the same size and caliber of a normal intercalated duct
(arrow). (f) DOG1 shows staining of ductal and myoepithelial components reminiscent of epithelial–myoepithelial carcinoma (  400).

Modern Pathology (2012) 25, 919–929

 DOG1 in salivary gland
928 J Chênevert et al

hyperplasia in association with an epithelial–myo- 5 Caputo A, Caci E, Ferrera L, et al. TMEM16A, a

epithelial carcinoma, and one intercalated duct membrane protein associated with calcium-dependent
adenoma showed a similar staining pattern to chloride channel activity. Science 2008;322:590–594.
epithelial–myoepithelial carcinoma and adenoid 6 Schroeder BC, Cheng T, Jan YN, et al. Expression
cloning of TMEM16A as a calcium-activated chloride
cystic carcinoma. Intercalated duct lesions are not
channel subunit. Cell 2008;134:1019–1029.
well characterized but have been mainly seen in 7 Yang YD, Cho H, Koo JY, et al. TMEM16A confers
association with epithelial–myoepithelial carcinoma receptor-activated calcium-dependent chloride con-
and basal cell salivary gland neoplasms (adenoma ductance. Nature 2008;455:1210–1215.
and carcinoma). Thus, if these lesions can truly serve 8 Ferrera L, Caputo A, Galietta LJ. TMEM16A protein: a
as precursors to epithelial–myoepithelial carcinoma, new identity for Ca(2+)-dependent Cl channels. Phy-
then it is reasonable to find that some of them may siology (Bethesda) 2010;25:357–363.
show a similar ‘transformed’ DOG1 staining pattern 9 Hwang DG, Qian X, Hornick JL. DOG1 antibody is a
in the myoepithelial component.28,29 On the other highly sensitive and specific marker for gastrointest-
hand, the main biphasic benign tumor category inal stromal tumors in cytology cell blocks. Am J Clin
Pathol 2011;135:448–453.
tested was pleomorphic adenoma, which was also 10 Almaca J, Tian Y, Aldehni F, et al. TMEM16 proteins
largely negative. One limitation to this study how- produce volume-regulated chloride currents that are
ever is that we did not test any basal cell adenomas reduced in mice lacking TMEM16A. J Biol Chem
and only tested one basal cell adenocarcinoma 2009;284:28571–28578.
(focally positive), which are also often considered 11 Kunzelmann K, Kongsuphol P, Aldehni F, et al.
as biphasic tumors. Additionally, tumors composed Bestrophin and TMEM16-Ca(2+) activated Cl(-) chan-
only of myoepithelial cells were negative in all nels with different functions. Cell Calcium
but one case of myoepithelial carcinoma that showed 2009;46:233–241.
only focal staining. 12 Ousingsawat J, Martins JR, Schreiber R, et al. Loss of
Thus in summary, we confirm high levels of TMEM16A causes a defect in epithelial Ca2+-depen-
dent chloride transport. J Biol Chem 2009;284:
ANO1/DOG1 expression in human salivary tissues 28698–28703.
with predominant apical membranous serous acinar 13 Ferrera L, Caputo A, Ubby I, et al. Regulation of
staining pattern. Acinic cell carcinomas almost TMEM16A chloride channel properties by alternative
uniformly show diffuse and intense DOG1 immu- splicing. J Biol Chem 2009;284:33360–33368.
nopositivity, a finding that can be used in conjunc- 14 Skalova A, Vanecek T, Sima R, et al. Mammary
tion with the traditional periodic acid-Schiff analogue secretory carcinoma of salivary glands,
reaction after diastase treatment to distinguish containing the ETV6-NTRK3 fusion gene: a hitherto
acinic cell carcinomas from other morphologic undescribed salivary gland tumor entity. Am J Surg
mimics, particularly mammary analog secretory Pathol 2011;34:599–608.
carcinomas. A subset of biphasic malignancies, 15 Chiosea SI, Griffith C, Assaad A, et al. The profile of
acinic cell carcinoma after recognition of mammary
adenoid cystic carcinoma, and epithelial–myo- analog secretory carcinoma. Am J Surg Pathol 2012;
epithelial carcinoma show a distinctive DOG1 36:343–350.
staining profile in both ductal and myoepithelial 16 Chiosea SI, Griffith C, Assaad A, et al. Clinicopatho-
cell components. logic characterization of mammary analogue secretory
carcinoma of salivary glands. Histopathology 2012;
e-pub ahead of print.
Disclosure/conflict of interest 17 Bergmann F, Andrulis M, Hartwig W, et al. Discovered
on gastrointestinal stromal tumor 1 (DOG1) is ex-
The authors declare no conflict of interest. pressed in pancreatic centroacinar cells and in
solid-pseudopapillary neoplasms-novel evidence for
a histogenetic relationship. Hum Pathol 2011;42:
References 817–823.
18 Ardeleanu C, Arsene D, Hinescu M, et al. Pancreatic
1 West RB, Corless CL, Chen X, et al. The novel marker, expression of DOG1: a novel gastrointestinal stromal
DOG1, is expressed ubiquitously in gastrointestinal tumor (GIST) biomarker. Appl Immunohistochem Mol
stromal tumors irrespective of KIT or PDGFRA muta- Morphol 2009;17:413–418.
tion status. Am J Pathol 2004;165:107–113. 19 Davis AJ, Forrest AS, Jepps TA, et al. Expression
2 Lee CH, Liang CW, Espinosa I. The utility of discovered profile and protein translation of TMEM16A in murine
on gastrointestinal stromal tumor 1 (DOG1) antibody in smooth muscle. Am J Physiol Cell Physiol 2010;
surgical pathology-the GIST of it. Adv Anat Pathol 299:C948–C959.
2010;17:222–232. 20 Gomez-Pinilla PJ, Gibbons SJ, Bardsley MR, et al. Ano1
3 Miettinen M, Wang ZF, Lasota J. DOG1 antibody in the is a selective marker of interstitial cells of Cajal in the
differential diagnosis of gastrointestinal stromal tumors: human and mouse gastrointestinal tract. Am J Physiol
a study of 1840 cases. Am J Surg Pathol 2009;33: Gastrointest Liver Physiol 2009;296:G1370–G1381.
1401–1408. 21 Huang X, Godfrey TE, Gooding WE, et al. Comprehen-
4 Carles A, Millon R, Cromer A, et al. Head and neck sive genome and transcriptome analysis of the 11q13
squamous cell carcinoma transcriptome analysis by amplicon in human oral cancer and synteny to the 7F5
comprehensive validated differential display. Onco- amplicon in murine oral carcinoma. Genes Chromo-
gene 2006;25:1821–1831. somes Cancer 2006;45:1058–1069.

Modern Pathology (2012) 25, 919–929

 DOG1 in salivary gland
J Chênevert et al 929

22 Childers EL, Ellis GL, Auclair PL. An immunohisto- 26 Griffith C, Seethala R, Chiosea SI. Mammary analogue
chemical analysis of anti-amylase antibody reactivity secretory carcinoma: a new twist to the diagnostic
in acinic cell adenocarcinoma. Oral Surg Oral Med dilemma of zymogen granule poor acinic cell carcino-
Oral Pathol Oral Radiol Endod 1996;81:691–694. ma. Virchows Arch 2011;459:117–118.
23 Ihrler S, Blasenbreu-Vogt S, Sendelhofert A, et al. 27 Lopes LF, West RB, Bacchi LM, et al. DOG1 for the
Differential diagnosis of salivary acinic cell carcinoma diagnosis of gastrointestinal stromal tumor (GIST):
and adenocarcinoma (NOS). A comparison of (immu- comparison between 2 different antibodies. Appl
no-)histochemical markers. Pathol Res Pract Immunohistochem Mol Morphol 2010;18:333–337.
2002;198:777–783. 28 Chetty R. Intercalated duct hyperplasia: possible
24 Batsakis JG, Pinkston GR, Luna MA, et al. Adenocarci- relationship to epithelial-myoepithelial carcinoma
nomas of the oral cavity: a clinicopathologic study of ter- and hybrid tumours of salivary gland. Histopathology
minal duct carcinomas. J Laryngol Otol 1983;97:825–835. 2000;37:260–263.
25 Araujo V, Sousa S, Jaeger M, et al. Characterization of 29 Weinreb I, Seethala RR, Hunt JL, et al. Intercalated
the cellular component of polymorphous low-grade duct lesions of salivary gland: a morphologic spectrum
adenocarcinoma by immunohistochemistry and elec- from hyperplasia to adenoma. Am J Surg Pathol 2009;
tron microscopy. Oral Oncol 1999;35:164–172. 33:1322–1329.

Modern Pathology (2012) 25, 919–929