Available online at www.sciencedirect.com

www.elsevier.com/locate/semdp

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8

http://dx.doi.org/10.0740-2570/& 2016 El

nCorresponding auE-mail address:

Acinar neoplasms of the pancreas—A summary of25 years of research

David S. Klimstra, MDa,n, Volkan Adsay, MDb

aDepartment of Pathology, Memorial Sloan Kettering Cancer Center, 1275 York Ave, New York, NY 10065bDepartment of Anatomic Pathology, Emory University, Atlanta, GA

a r t i c l e i n f o

Keywords:

Acinar

Pancreas

Carcinoma

Pancreatoblastoma

1053/j.semdp.2016.05.009sevier Inc. All rights rese

thor.klimstrd@mskcc.org (D.S

a b s t r a c t

Our understanding about the family of acinar neoplasms of the pancreas has grown

substantially over the past 25 years. The prototype is acinar cell carcinoma, an uncommon

variant of pancreatic carcinoma that demonstrates production of pancreatic exocrine

enzymes, verifiable using immunohistochemistry, and exhibits characteristic histologic

features. Related neoplasms include mixed acinar carcinomas such as mixed acinar

neuroendocrine carcinoma and mixed acinar ductal carcinoma. In the pediatric age group,

pancreatoblastoma is also closely related. Cystic and extrapancreatic forms have been

described. These neoplasms share molecular alterations that are distinct from the more

common ductal and neuroendocrine neoplasms of the pancreas. Although there is a broad

range of genetic findings, a number of potential therapeutic targets have emerged. This

review explores the clinical and pathologic features of pancreatic acinar neoplasms along

with their more common molecular phenotypes. The differential diagnosis with other

pancreatic neoplasms is explored as well.

& 2016 Elsevier Inc. All rights reserved.

Pancreatic neoplasms usually exhibit predominantly ductaldifferentiation, which are associated with mucin productionor expression of glycoproteins and display a characteristicspectrum of molecular alterations.1 Neoplasms of neuroen-docrine and acinar lineage are much less common. Acinardifferentiation is defined as the production of pancreaticexocrine enzymes by the neoplastic cells, but acinar cellneoplasms also have highly characteristic histologic features,and their emerging molecular characteristics are also largelydistinct from those of other pancreatic neoplasms.2 Thisarticle traces the evolution of our understanding of pancre-atic acinar neoplasms—including acinar cell carcinoma,mixed acinar carcinomas (ACCs), acinar cystic lesions, andpancreatoblastoma—from pathological and clinical charac-terization to molecular genetic analysis over the past

rved.

. Klimstra).

25 years. This journey was initially inspired and heavilyinfluenced by the mentorship of Dr. Juan Rosai, who not onlyrecognized a gap in our understanding of pancreatic neo-plasia but also provided the tools to stimulate pathologicalinvestigation and thematic research on this related family ofentities.

Historical perspective

Early recognition of ACC as a distinct variant of pancreaticcarcinoma was based on clinical findings rather than onpathological insight. Rare patients with pancreatic cancerwere reported to develop disseminated fat necrosis,particularly in the subcutaneous tissues, along with

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8308

polyarthralgia.3–8 These cases of the classic lipase hyper-secretion syndrome, shown to result from secretion by thetumor of massive amounts of lipase into the blood, wereultimately associated with carcinomas with a distinctivehistologic appearance. The pattern often resembled that ofthe normal acinar elements of the pancreas; thus, the term“acinar cell carcinoma” was proposed. Until the early 1980s,ACCs were reported mostly as individual cases, many ofwhich displayed the lipase hypersecretion syndrome. Onlyin 1977, a report of 11 cases of ACC appeared that emphasizedthe distinctive histologic features that could allow the diag-nosis of ACC in the absence of a paraneoplastic syndrome.9

At this time, ancillary studies to prove the presence of acinardifferentiation had not been developed, and only limitedexamples had been evaluated with electron microscopy. Infact, the high frequency of acinar cell carcinomas in thisstudy (nearly 10%) among pancreatic tumors raised someconcerns whether all of these cases would be classified asACC based on contemporary criteria.10

One this backdrop, one of the authors (D.S.K.) enteredpathology residency at Yale-New Haven Hospital in 1988,under the directorship of Dr. Rosai. In late 1989, a caseappeared in the department that sparked a series of studiesthat have redefined this entity, leading to its in depthcharacterization and ultimately to the identification of molec-ular targets for therapy. This case is worth a brief description,as it raised a number of issues that continue to make thediagnosis of ACC a challenge for pathologists. The patientwas a 62-year-old male with a prior history of organ-confinedprostatic adenocarcinoma, Gleason score 3 þ 4 ¼ 7, treatedwith radiation therapy 5 years prior to presentation. Whenseen in 1989, he had presented acutely with upper gastro-intestinal hemorrhage, and an abdominal CT scan revealed amass in the head of the pancreas or duodenum and multipleliver metastases. A core needle biopsy of a liver lesion wasinterpreted as metastatic prostatic adenocarcinoma,

Fig. 1 – Histology of pancreatic neoplasm projecting into duodenminimal fibrovascular stroma between nests and cords of cells. Aof uniform cells with basally oriented oval nuclei. The chromatievident.

although no immunohistochemical stains were performedand the prior prostate primary was not available for compar-ison. Upper endoscopy revealed a large polypoid tumorprojecting into the duodenum, and given the ongoing bleed-ing, a pancreatoduodenectomy was performed emergently.The histology of this tumor revealed a densely cellular neo-plasm that was sharply circumscribed from the underlyingpancreas (Fig. 1). The architecture was solid and trabecular,with areas showing a gyriform pattern. Although thesefeatures raised the possibility of a neuroendocrine neoplasm,the cells were well-polarized at the periphery of the nests,focal rudimentary acinar structures were present, and thenuclei had distinct nucleoli. Immunohistochemical stains forchromogranin and synaptophysin were completely negative,and the tumor had easily identifiable mitotic figures, consid-ering that the initial diagnostic impression was that of a well-differentiated neuroendocrine neoplasm. Dr. Rosai reviewedthe case and suggested the correct diagnosis of acinar cellcarcinoma, which was later supported by the ultrastructuralfinding of large (250–500 nm) homogeneous dense core gran-ules, apically located in the cells, along with a secondunfamiliar granule type that was large (3500 nm), angulated,irregular in shape, and had a fibrillary internal matrix (Fig. 2).Interestingly, despite presenting with significant hepatic dis-ease, the patient had no other distant metastases andsurvived for 42 months following a number of variablyeffective chemotherapeutic regimens.In the aftermath of this case, Dr. Rosai mentioned to D.S.K.

that ACC, although recognized for many years as a distinctentity, had never been the subject of a definitive pathologicalevaluation. Furthermore, there were no confirmatory immu-nohistochemical stains, despite the high likelihood that theneoplastic cells were producing pancreatic exocrine enzymes.A commercial source for antibodies against trypsin, chymo-trypsin, and lipase was found, and Dr. Rosai proposed thatthese antibodies be characterized for diagnostic use.

um. At low power (A) the tumor is highly cellular, witht high power (B), there is a trabecular architecture composedn pattern is coarsely granular. Mitotic figures are readily

Fig. 2 – Ultrastructural appearance of the tumor in Fig. 1.Numerous secretory granules are present, including round,dense granules and larger elongated and irregular granuleswith a fibrillary internal matrix.

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8 309

Collaboration with Drs. Clara Heffess and James Oertel at theArmed Forces Institute of Pathology was further suggested byDr. Rosai, and their numerous cases were added to thosefrom the archives at Yale. After eliminating mixed acinarcarcinomas, pancreatoblastomas, and other entities, theremaining 28 cases of ACC formed the basis of the firstcontemporary study of this entity, which also establishedimmunohistochemical staining for enzymes as an extremelyhelpful tool to support the diagnosis.10 This article has beenfollowed by many others providing additional insights intoACC11 and describing mixed acinar neuroendocrine carci-noma,12 mixed acinar ductal carcinoma,13 pancreatoblas-toma,14 and so-called acinar cell cystadenoma.15 Numerousother clinical and molecular characterizations of these enti-ties have since been undertaken.

Fig. 3 – Gross appearance of acinar cell carcinoma. Thebulky, lobulated tumor involved the pancreatic tail andinvades the spleen, growing as multiple soft, fleshy noduleswith areas of hemorrhage.

Clinical and pathological features of acinar cellcarcinoma

Approximately 1–2% of pancreatic neoplasms are ACCs ormixed acinar carcinomas.16 Most patients are adults, with amean age of 60 years, but approximately 6% ACCs occur inchildhood, and 15% of pediatric pancreatic neoplasms areACCs.17,10,18–20 Males are affected more commonly thanfemales. Because ACCs have a relatively circumscribed,expansile growth pattern, invasion of the common bile ductis much less frequent than in ductal adenocarcinomas.Therefore, the presenting symptoms uncommonly includejaundice and are usually non-specific (abdominal pain,weight loss, nausea, etc.). The classic lipase hypersecretionsyndrome is actually uncommon (o10% of patients) and wasprobably over-reported in the earlier literature because of itsdramatic presentation.10,20 Patients affected by the lipasehypersecretion syndrome generally have hepatic metastasesor extremely bulky primary disease. Serum lipase levels maybe in excess of 10,000. In addition to subcutaneous fatnecrosis and arthralgia, eosinophilia may be found. Uncom-mon clinical associations of ACC include thrombotic endo-carditis9 and myeloma-like cast nephropathy.21 Alpha-

fetoprotein production can occur.22 Hereditary cases are veryunusual, although a number of cases have occurred inpatients with familial adenomatous polyposis,23,24 Lynchsyndrome,25,26 or Carney syndrome,27 providing some hintsabout the molecular underpinnings of acinar neoplasms (seebelow).ACCs are somewhat more common in the head of the

pancreas. They are usually large (mean size: 8 cm) and solid,although cystic degeneration can occur in larger exam-ples.10,20,11 Grossly, ACCs have a well-circumscribed or multi-nodular appearance and may seem to be encapsulated(Fig. 3). The tumors are red to brown, homogeneous, andhave a very soft consistency. Hemorrhage or necrosis mayoccur. Occasionally there is polypoid extension into thepancreatic ducts.28,29

ACCs are also circumscribed microscopically (Fig. 4), butinvasion through the peripheral fibrous pseudocapsule orinto vessels or nerves is common. Within the tumor lobules,ACCs are characteristically highly cellular, lacking significantfibrous stroma and devoid entirely of desmoplasia. The neo-plastic cells are uniform and arranged in various architecturalpatterns, most typically acinar or solid (Fig. 5), but alsotrabecular or even gyriform can occur (Fig. 1). Rare casesexhibit a papillary architecture, which can occur in thecomponent of intraductal extension of the tumor.28 The cellshave minimal to moderate cytoplasm that varies fromamphophilic to lightly eosinophilic (Fig. 6A). A helpful findingis eosinophilic cytoplasmic granularity, which is often morepronounced in regions with an acinar or glandular architec-ture. The granules, which are PAS-positive resistant to dia-stase, represent zymogen granules and are often orientedtoward the apical pole of the cells. The nuclei range fromround to oval, and although they are enlarged compared tothose of non-neoplastic acinar cells, they are typically uni-form and lack marked pleomorphism. Central prominentnucleoli are a helpful diagnostic feature (Fig. 6B). The mitotic

Fig. 4 – Low power appearance of acinar cell carcinoma. Theneoplasm is hypercellular and circumscribed. There isminimal stroma.

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8310

rate varies from nearly undetectable to moderate but is oftenin excess of 20 mitoses per 10 high power microscopic fields.In the presence of classic cytoarchitectural features, the

diagnosis of ACC can be strongly suspected based on routinehistology. However, there are a number of significant mimics(see “Differential Diagnosis,” below), so ancillary diagnosticstudies are generally employed to support the diagnosis.Demonstration of dPAS-positive granules can be helpful, butthis is neither sensitive nor specific.10 Currently, immunolab-eling for pancreatic exocrine enzymes is the primary meansto demonstrate acinar differentiation. Antibodies againsttrypsin and chymotrypsin are the most widely used andsensitive (Fig. 7); lipase can also be demonstrated in about65% of cases.10,20,11 Interestingly, amylase is not usuallydetectable and antibodies against this enzyme are not useddiagnostically. Another more recent acinar marker isbcl10,30,31 although in routine practice the use of trypsinand chymotrypsin is generally adequate. The ultrastructuralfeatures of ACC describing the case vignette above are alsohighly characteristic, and the so-called irregular fibrillary

Fig. 5 – Architectural pattern of acinar cell carcinoma. The acinaneoplastic pancreas, with small central lumina and well-polarizecells without clear evidence of polarization. A few lumina are id

granules appear to be specific for neoplasms with acinardifferentiation, as they recapitulate the morphology of theearliest granules observed in the acinar cells of the develop-ing embryonic pancreas.10,12,14 However, the general declinein the use of electron microscopy for tumor diagnosis hasrendered this diagnostic tool nearly obsolete.ACCs are aggressive carcinomas. Approximately, half of

patients have metastases at presentation, and another 25%develop them subsequently.10 Metastatic disease usuallyaffects the lymph nodes and liver; even late in the course ofdisease extrahepatic metastases are uncommon. However,rare cases present with ovarian metastases, where the differ-ential diagnosis can include a wide array of primary andmetastatic neoplasms.32 The outcome of ACC is better thanthat of conventional ductal adenocarcinoma. Data reportedmore than 15 years ago suggested a median survival for allstages of disease of 18 months, with patients presentingwithout metastases surviving more than 3 years on aver-age.10,18 Anecdotes also existed of patients with stage IVdisease at presentation surviving several years.33 Recent datahave shown an even more favorable prognosis, presumablydue to earlier detection and some responses to chemother-apy.34,35,19 An overall 5-year survival rate of 43% (72% forpatients undergoing resection; 22% for those with unresect-able disease) and a median survival of 57 months forresectable disease and 20 months for those with metastasesare now reported34,35,19; however, most patients ultimatelysuccumb to their disease. Prognostic factors include onlystaging features (primary tumor size, lymph node status,and presence of metastases).11 There is no predictive gradingscheme for ACC.

Mixed acinar carcinomas

ACCs resemble pancreatic neuroendocrine neoplasms, andmany cases (40%) also exhibit a minor population of cells that

r pattern (A) recapitulates the architecture of the non-d nuclei. The solid pattern (B) consists of nests and sheets ofentifiable.

Fig. 6 – Cytologic features of acinar cell carcinoma. Moderate amounts of granular eosinophilic cytoplasm are characteristic (A).The granularity is accentuated in the apical cytoplasm surrounding the lumina. A helpful finding is the presence of singleprominent nucleoli (B), but many cases lack this feature.

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8 311

express neuroendocrine lineage markers (chromogranin andsynaptophysin) by immunohistochemistry.10 Less commonly,acinar neoplasms have a component of neuroendocrinedifferentiation that constitutes 425% of the neoplasm, basedon the proportion of cells labeling immunohistochemi-cally.36,12,37,38 These tumors are designated mixed acinarneuroendocrine carcinomas (Fig. 8). In almost all cases, theacinar component predominates.39 A few exceptional exam-ples have shown morphologically obvious and separatecomponents reflecting the two different cell lineages, butmost mixed acinar neuroendocrine carcinomas have a morehomogeneous morphology, with only subtle regional varia-tions raising the potential of both acinar and neuroendocrineelements being present.12 In these cases, immunohistochem-istry is required to recognize the mixed nature of the neo-plasm. In contrast to other mixed neuroendocrinecarcinomas that may have separate, definable componentsof small cell carcinoma or large cell neuroendocrine carci-noma, mixed acinar neuroendocrine carcinomas appear to befundamentally acinar neoplasms with divergent differentia-tion toward the neuroendocrine lineage that varies in extentthroughout the tumor. Individual cells appear to have

Fig. 7 – Immunolabeling for chymotrypsin of the tumorillustrated in Fig. 1 shows strong granular positivity in theapical cytoplasm.

so-called “amphicrine” differentiation, expressing both tryp-sin (or chymotrypsin) and chromogranin (or synaptophysin).There are neither prognostic nor molecular differencesbetween mixed acinar neuroendocrine carcinomas and pureACCs,12,37,11 so the main importance of this entity is to avoidits confusion with pure neuroendocrine neoplasms, whichmay be suggested based on the finding of immunolabeling forchromogranin or synaptophysin in a significant proportionof cells.Another more recently described mixed acinar carcinoma is

mixed acinar ductal carcinoma.13 In these cases, the ductalelements can resemble conventional ductal adenocarcino-mas, with individual infiltrating glands associated with stro-mal desmoplasia, or there can be frank mucin production bya subset of neoplastic cells (Fig. 9). The mucin may beintracellular or there may be pools of extracellular mucinwith suspended tumor cells, similar to colloid carcinomas. Asin the mixed acinar neuroendocrine carcinomas, there isusually a predominance of acinar differentiation, which canbe present throughout all patterns of the tumors. In fact,some cells appear to produce both mucin and exocrineenzymes based on double staining preparations. Rarely, amixed acinar ductal carcinoma will also exhibit neuroendo-crine marker expression in 425% of the cells, qualifying for adiagnosis of mixed acinar ductal neuroendocrine carcinoma.A few of these mixed acinar carcinomas with ductal differ-entiation have had mutations more characteristic of ductalthan acinar neoplasms (e.g., KRAS mutations) but in generalthese tumors resemble pure acinar cell carcinoma in theirbiology.13

Pancreatoblastoma

Pancreatoblastoma is a pediatric neoplasm of the pancreasthat has predominantly acinar differentiation.14 In fact,pancreatoblastoma can be considered the pediatric

Fig. 8 – Mixed acinar neuroendocrine carcinoma. The tumor shows variable cytologic features (A), with some cells having theeosinophilic granular cytoplasm of an acinar neoplasm whereas others have less cytoplasm lacking granularity. A doubleimmunohistochemical stain (B) reveals labeling for both trypsin (blue reaction product) and chromogranin (brown reactionproduct), largely in separate cell populations that are intimately intermixed throughout the tumor.

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8312

counterpart of ACC, in the same manner than hepatoblas-toma is the pediatric counterpart of hepatocellular carci-noma. Pancreatoblastomas usually arise in the first decade,with a mean age of 4 years, and cases in children older than10 years are rare.40 Cases in adults have also beendescribed.41,42 They are usually sporadic, but some arise inpatients with the Beckwith–Weidemann syndrome, and pan-creatoblastomas in these patients can be congenital and areoften cystic.14,43–46 A case of FAP-associated pancreatoblas-toma has been reported as well.23 Rarely, acinar neoplasmswith the histologic features of pancreatoblastomas arise inadults. Alpha-fetoprotein production can occur, with serum

Fig. 9 – Mixed acinar ductal carcinoma. The tumor grows in nescytoplasmic mucin (A). By immunohistochemistry (B), there is lablue positivity in the mucin-containing cells (blue stain). (For intereader is referred to the web version of this article.)

elevations that can be used as a biomarker, both in pancrea-toblastomas and in ACCs occurring in childhood.22,47,48

The morphology shares much with ACC (Fig. 10), includingsolid and acinar architecture and dense cellularity.14 There isusually pronounced lobulation, and the stroma between thelobules is hypercellular, sometimes with heterologous mes-enchymal differentiation. The epithelial components areformed of small, uniform cells arranged in acini, usually withless evident eosinophilic granularity than in ACC. The nucleiin these regions are small, round, and hyperchromatic. Acharacteristic feature, considered the sine qua non of thediagnosis, is the squamoid nest (or corpuscle). Squamoid

ts with marked nuclear pleomorphism and obviousbeling for trypsin (brown reaction product) as well as alcianrpretation of the references to color in this figure legend, the

Fig. 10 – Pancreatoblastoma. At lower power (A) the tumor is lobulated and the stroma between the epithelial regions ishypercellular. The neoplastic cells are arranged in nests and acini. A higher power view (B) shows a squamoid nest adjacentto an acinar region. The squamoid cells are larger, somewhat spindled and have oval nuclei and eosinophilic cytoplasm.

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8 313

nests vary in number and distribution throughout the tumorand are composed of larger cells with more cytoplasm thanthe surrounding acinar components. The cells are vaguelyspindled and arranged in whorls, sometime with focal kera-tinization. The nuclei are also larger, more oval, and have anopen chromatin pattern with clearing due to nuclear biotinaccumulation.49,50 Rare pancreatoblastomas include ductalstructures with intracellular mucin.Pancreatoblastomas exhibit acinar differentiation consis-

tently, with expression of trypsin, chymotrypsin, and otheracinar markers.51,14 There is also commonly labeling forneuroendocrine and occasionally for ductal lineage markers.AFP is expressed in cases with serum elevation. The squa-moid nests do not seem to reflect a particular cell lineage,although they preferentially express EMA and CEA.

Fig. 11 – Acinar cell cystadenocarcinoma. The multicystic tumorare cytologically acinar (B), with lumina showing apical eosinopnucleoli.

Interestingly, they also show nuclear labeling for β-catenin,52

which is implicated in the molecular histogenesis of acinarneoplasms, but the acinar components of the tumors retainnormal membranous labeling for β-catenin.Pancreatoblastomas can behave aggressively, but like

pediatric ACCs, they are less aggressive in children than inadults.53,14,54,40,47 Patients without metastases at presentationmay be cured, and favorable responses to chemotherapy havebeen documented.

Cystic acinar lesions

Although some large ACCs have degenerative cystic change,due to necrosis, truly cystic ACCs in which the neoplastic

is lined by flattened to complex epithelium (A). The noduleshilic granularity and basal atypical nuclei with prominent

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8314

cells line cystic locules are very rare (Fig. 11). Other than a fewold case reports, acinar cell cystadenocarcinomas have beenoccasionally included among large studies of conventionalACCs.20,55–57 Acinar cell cystadenocarcinomas are fully malig-nant and share all other features of ACCs, other than theirmulticystic nature, which presumably results from the mas-sive dilatation of the acinar lumina that constitute themajority of the tumor architecture.Another pancreatic cystic lesion with acinar differentiation

has been more recently described as acinar cell cystade-noma.58,15,59,60 These lesions can represent incidental micro-scopic cystic structures lined by benign-appearing acinarcells, some of which are likely pancreatic ducts with acinarmetaplasia. Macroscopically identified cases have a micro-cystic gross appearance and may diffusely involve largeregions of the pancreas. Microscopically, there is cysticdilatation of the entire acinar and ductal units, resulting incomplex, branching cysts intermingled with more normal-appearing parenchyma (Fig. 12). The cysts are lined variablyby flat, non-mucinous ductal cells and by acinar cells, whichsometimes form grape-like clusters budding from the cystwalls. These lesions are cytologically bland and uniformlybenign.15,61 In fact, the initial name of acinar cell cystade-noma, which connote a benign neoplasm, has been ques-tioned recently when molecular studies failed to prove thelesions to be neoplastic.62 An alternate designation of acinarcystic transformation of the pancreas has been proposed.63

Extrapancreatic acinar neoplasms

A few neoplasms with the morphologic and immunohisto-chemical features of pancreatic acinar neoplasms have beenreported outside of the pancreatic itself. Some appear to ariseon the basis of neoplastic transformation of heterotopicpancreatic tissue; at least, they arise in locations wherepancreatic heterotopia is well recognized to occur. But evi-dence of residual non-neoplastic heterotopic tissue is rarelydocumented. Most examples have involved the liver64–66 andstomach (Fig. 13). In the stomach, mixed acinar neoplasmswith neuroendocrine differentiation are described.67–70 Giventhe propensity for the gastric mucosa to undergo pancreaticacinar metaplasia,71 acinar neoplasms in this location may

Fig. 12 – So-called acinar cell cystadenoma (acinar cystictransformation). The cysts are partially lined by benignacinar cells and partially by ductal-type epithelium. Nodulesof acini with an interanastomosing pattern are found in thecyst septa.

arise on the basis of aberrant differentiation of a gastricmucosal neoplasm, rather than from heterotopia. The gastricexamples have also been smaller, less aggressive neoplasmsthan the few cases reported in the liver, which are morefaithful phenocopies of the primary pancreatic neoplasm. Anadditional unusual scenario for the origin of extrapancreaticacinar neoplasms is somatic transformation of a germ celltumor. Non-neoplastic pancreatic tissue occurs in teratomasof the mediastinum and gonads,72 and presumably primaryacinar neoplasms arising in these locations represent secon-dary malignancy of these teratomatous elements.22

Molecular features of acinar neoplasms

Molecular data related to pancreatic acinar neoplasms havebeen accumulating steadily over the past 20 years as thetechnology to study them has improved.16 But the rarity ofthese neoplasms has hindered their comprehensive molec-ular analysis until only recently. Initially, the genes found tobe involved in ductal adenocarcinoma of the pancreas werestudied in acinar neoplasms, and it was shown that the mostfrequent genetic abnormalities (mutations in KRAS, TP53, andSMAD4) are uncommon.73,20,74 A more recent, comprehensivestudy of TP53 mutations, deletions, and promoter methyla-tion has shown much more frequent TP53 abnormalities,however.75 The hereditary associations of ACC and pancrea-toblastoma provided hints as to the genetic alterations insporadic cases. The rare association of both neoplasms withFAP led to the study of the APC/β-catenin pathway, and inabout 20–50% of cases, abnormalities in either APC or β-cateninwere found, including activating mutations, losses, andpromoter methylation of APC, and inactivating mutations inCTNNB1.23,24,76 Copy number alterations and methylation ofthe promoter of APC have also been described. Curiously,immunohistochemical staining of pancreatoblastoma for β-catenin shows aberrant nuclear labeling largely restricted tothe squamoid nests.52 Beckwith–Wiedemann syndrome withwhich pancreatoblastomas may be associated is caused byvarious abnormalities involving chromosome 11p15, includ-ing methylation defects, uniparental disomy, or mutations inthe CDKN1C, H19, IGF2, or KCNQ1OT1 genes.77 Losses at 11pwere also found in both pancreatoblastoma (80%) and ACC(50%).23,24,78,79 Due to the rare cases of ACC in Lynch syn-drome, mismatch repair abnormalities were studied in ACCand were found both in Lynch-associated and in 14% ofsporadic cases.26,25 Finally, Carney syndrome has rarely beenassociated with ACC,27 and the PRKAR1A gene responsible forCarney syndrome is also mutated in some sporadic ACCs.80

More recent genomic studies have taken advantage of next-generation sequencing to perform more comprehensivegenomic analysis. Whole exome sequencing as well as moretargeted broad-spectrum sequencing studies81,80 haverevealed a high degree of genomic instability on both thechromosome and base pair levels in acinar neoplasms. Manydifferent genes were mutated across the tumors studied, withno single gene being mutated in more than 30% of cases. Thelack of common alterations in ductal adenocarcinoma (KRAS,SMAD4, TP53, and CDKN2A), cystic neoplasm (GNAS andRNF43), and neuroendocrine tumor (MEN1, DAXX, and ATRX)

Fig. 13 – Extrapancreatic acinar neoplasms. An acinar cell carcinoma of the liver (A) is histologically similar to its pancreaticcounterpart. The tumor was resected and there was no evidence of a primary in the pancreas. Another case arising in thestomach (B) shows residual overlying gastric glands. This tumor expressed both acinar and neuroendocrine differentiation byimmunohistochemistry.

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8 315

genes was confirmed, although 24% of acinar neoplasms hadpoint mutations or truncations in TP53, 18% had SMAD4mutations, and CDKN2A/CDKN2B was also deleted in aminority of cases. Also confirmed were the alterations inAPC and CTNNB1 described previously. Rare mutations werefound in DNA repair genes associated with familial pancre-atic cancer, such as MLH1 and MSH2 as well as ATM, PALB2,BRCA2, the last three of which were mutually exclusive.Additional recurrently altered genes included JAK1, BRAF,RB1, PTEN, ARID1A, NF1, SKT11, MLL3, PRKAR1A, and BAP1.Thus, the spectrum of mutations is broad but does include avariety of potential therapeutic targets, such as JAK1, BRAF,and genes of the mTOR and DNA repair pathways. An addi-tional molecular alteration of potential therapeutic signifi-cance is the finding of BRAF fusions in 23% of acinarneoplasms.80 Most such fusions (which are rare in other solidtumor types) involved SND1, although several other fusionspartners were identified. The fusions are functional, leadingto activation of the MAPK pathway, sensitive in vitro to MEKinhibitors. A rapid FISH assay to identify BRAF fusions inpancreatic acinar neoplasms has also been developed.82

Differential diagnosis

The differential diagnosis for pancreatic acinar neoplasmslargely includes other tumors with a solid, hypercellular lowpower appearance—which typically excludes ductal adeno-carcinoma and cystic neoplasms, unless a mixed acinarductal carcinoma or an acinar cystic lesion is under consid-eration.2,1 Conventional ductal adenocarcinomas are uncom-monly confused for ACC, although many cases originallyclassified as “microadenocarcinoma,” a microglandular var-iant of ductal adenocarcinoma, proved to be ACC or mixedacinar neuroendocrine carcinoma after additional studyusing IHC for pancreatic enzymes.83 The most common solid,

cellular pancreatic neoplasms that are confused for acinarneoplasms are pancreatic neuroendocrine neoplasms [bothwell differentiation pancreatic neuroendocrine tumor (Pan-NET) and poorly differentiation neuroendocrine carcinoma(PanNEC)] and solid pseudopapillary neoplasm.PanNETs share many architectural features with ACC,

especially when the latter has solid or trabecular patterns.The nuclei in both are uniform, the cytoplasm can beamphophilic, and ACCs can show focal labeling for neuro-endocrine markers. And mixed acinar neuroendocrine carci-nomas show extensive labeling for chromogranin orsynaptophysin. Markedly prominent nucleoli favor an acinarneoplasm, although PanNETs can also have prominent nucle-oli. Apical granular eosinophilic cytoplasm and exocrine cellpolarization also favor ACC. When an organoid architecturesuggests a well-differentiated PanNET, recognition of fre-quent mitoses (more than 20 per 10 high power fields) canhelp raise the possibility of ACC. But the resemblance of ACCto poorly differentiated PanNEC can prove even more vexing:15.9% of cases referred from specialty institutions for a studyon poorly differentiated PanNECs proved to be ACC or mixedacinar neuroendocrine carcinoma on further review.84 Forthese reasons, immunolabeling for trypsin, chymotrypsin, orbcl10 is suggested to exclude an acinar neoplasm whenever aputative neuroendocrine neoplasm of the pancreas does nothave perfectly classic histologic features. On cytologic prep-arations, this difficulty is also apparent, and immunohisto-chemistry can also be helpful in this situation.85,86

Solid pseudopapillary neoplasms have a number of highlycharacteristic histologic features that strongly suggest thediagnosis, including loosely cohesive epithelial cells thatform degenerative pseudopapillae around the microvascula-ture, large cytoplasmic hyaline globules (much larger thanzymogen granules), and nuclear grooves.87,88 Peripheral bloodlakes, stromal hyalinization, and cytoplasmic clear vacuolesare also typical of solid pseudopapillary neoplasm. In

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8316

contrast, the formation of true luminal spaces is not found insolid pseudopapillary neoplasm and strongly suggests analternative diagnosis. If the distinction from ACC is challeng-ing, immunohistochemistry will resolve the issue. Solidpseudopapillary neoplasms never express exocrine enzymes,instead labeling for vimentin, CD10, CD56, alpha-1-antitryp-sin, synaptophysin, and other assorted markers.88,89 Nuclearlabeling for β-catenin is also consistent found,90 although thepresence of APC—β-catenin pathway abnormalities in acinarneoplasms suggests a degree of caution in the use of thismarker as the sole diagnostic support for solid pseudopapil-lary neoplasm.Finally, the distinction of ACC from pancreatoblastoma

may be challenging. In the pediatric age group, pancreato-blastomas usually have well-developed squamoid nests andcellular stroma, and the former features are regarded asnecessary to confirm the diagnosis. In adults, these twofeatures may be more subtle. In the end, the distinctionmay be less critical than it seems, since the treatment andprognosis of these two tumors is essentially the same inchildren and in adults. For cases with unclear histologicfindings (e.g., for biopsies with limited tumor for review), itis suggested to favor pancreatoblastoma in children and ACCin adults.

Conclusions

Much new information about the family of pancreatic acinarneoplasms has appeared in the past 25 years. The distin-guishing histologic features are better defined, and immuno-histochemical markers to support the presence of acinardifferentiation are more widely available. An awareness ofmixed acinar carcinomas has helped properly classify theseunusual neoplasms among the other members of the acinarfamily. A wealth of molecular information has providedhistogenetic insights and suggested therapeutic targets. Inmany ways, this evolution has paralleled the increase inpathologic knowledge about many tumor entities over thistime period. Yet it is tempting to believe that a single seminalcase of ACC, in the hands of a talented morphologist andpotent mentor, contributed to this advance in knowledge, aspart of the broad academic legacy of Dr. Juan Rosai.

r e f e r e n c e s

1. Klimstra DS, Pitman MB, Hruban RH. An algorithmic approachto the diagnosis of pancreatic neoplasms. Arch Pathol Lab Med.2009;133(3):454–464.

2. Klimstra DS. Nonductal neoplasms of the pancreas. ModPathol. 2007;20(suppl 1):S94–S112.

3. Alcantara EN Jr. Functioning acinar cell carcinoma of thepancreas. Can Med Assoc J. 1962;87(11):970–973.

4. Auger C. Acinous cell carcinoma of the pancreas with exten-sive fat necrosis. Arch Pathol. 1947;43(4):400–405.

5. Belsky H, Cornell NW. Disseminated focal fat necrosis follow-ing radical pancreatico-duodenectomy for acinous carcinomaof head of pancreas. Ann Surg. 1955;141(4):556–562.

6. Burns WA, Matthews MJ, Hamosh M, Weide GV, Blum R,Johnson FB. Lipase-secreting acinar cell carcinoma of thepancreas with polyarthropathy. A light and electron

microscopic, histochemical, and biochemical study. Cancer.1974;33(4):1002–1009.

7. MacMahon HE, Brown PA, Shen EM. Acinar cell carcinoma ofthe pancreas with subcutaneous fat necrosis. Gastroenterology.1965;49(5):555–559.

8. Robertson JC, Eeles GH. Syndrome associated with pancreaticacinar cell carcinoma. Br Med J. 1970;2(5711):708–709.

9. Webb JN. Acinar cell neoplasms of the exocrine pancreas. JClin Pathol. 1977;30(2):103–112.

10. Klimstra DS, Heffess CS, Oertel JE, Rosai J. Acinar cell carci-noma of the pancreas. A clinicopathologic study of 28 cases.Am J Surg Pathol. 1992;16(9):815–837.

11. La Rosa S, Adsay V, Albarello L, et al. Clinicopathologic studyof 62 acinar cell carcinomas of the pancreas: insights into themorphology and immunophenotype and search for prognos-tic markers. Am J Surg Pathol. 2012;36(12):1782–1795.

12. Klimstra DS, Rosai J, Heffess CS. Mixed acinar-endocrine carci-nomas of the pancreas. Am J Surg Pathol. 1994;18(8):765–778.

13. Stelow EB, Shaco-Levy R, Bao F, Garcia J, Klimstra DS.Pancreatic acinar cell carcinomas with prominent ductaldifferentiation: mixed acinar ductal carcinoma and mixedacinar endocrine ductal carcinoma. Am J Surg Pathol. 2010;34(4):510–518.

14. Klimstra DS, Wenig BM, Adair CF, Heffess CS. Pancreatoblas-toma. A clinicopathologic study and review of the literature.Am J Surg Pathol. 1995;19(12):1371–1389.

15. Zamboni G, Terris B, Scarpa A, et al. Acinar cell cystadenomaof the pancreas: a new entity? Am J Surg Pathol. 2002;26(6):698–704.

16. Wood LD, Klimstra DS. Pathology and genetics of pancreaticneoplasms with acinar differentiation. Semin Diagn Pathol.2014;31(6):491–497.

17. Shorter NA, Glick RD, Klimstra DS, Brennan MF, Laquaglia MP.Malignant pancreatic tumors in childhood and adolescence:The Memorial Sloan-Kettering experience, 1967 to present. JPediatr Surg. 2002;37(6):887–892.

18. Holen KD, Klimstra DS, Hummer A, et al. Clinical character-istics and outcomes from an institutional series of acinar cellcarcinoma of the pancreas and related tumors. J Clin Oncol.2002;20(24):4673–4678.

19. Lowery MA, Klimstra DS, Shia J, et al. Acinar cell carcinoma ofthe pancreas: new genetic and treatment insights into a raremalignancy. Oncologist. 2011;16(12):1714–1720.

20. Hoorens A, Lemoine NR, McLellan E, et al. Pancreatic acinarcell carcinoma. An analysis of cell lineage markers, p53expression, and Ki-ras mutation. Am J Pathol. 1993;143(3):685–698.

21. Reducka K, Gardiner GW, Sweet J, Vandenbroucke A, Bear R.Myeloma-like cast nephropathy associated with acinar cellcarcinoma of the pancreas. Am J Nephrol. 1988;8(5):421–424.

22. Cingolani N, Shaco-Levy R, Farruggio A, Klimstra DS, Rosai J.Alpha-fetoprotein production by pancreatic tumors exhibit-ing acinar cell differentiation: study of five cases, one arisingin a mediastinal teratoma. Hum Pathol. 2000;31(8):938–944.

23. Abraham SC, Wu TT, Klimstra DS, et al. Distinctive moleculargenetic alterations in sporadic and familial adenomatouspolyposis-associated pancreatoblastomas: frequent altera-tions in the APC/beta-catenin pathway and chromosome11p. Am J Pathol. 2001;159(5):1619–1627.

24. Abraham SC, Wu TT, Hruban RH, et al. Genetic and immu-nohistochemical analysis of pancreatic acinar cell carcinoma:frequent allelic loss on chromosome 11p and alterations inthe APC/beta-catenin pathway. Am J Pathol. 2002;160(3):953–962.

25. Liu W, Shia J, Gonen M, Lowery MA, O'Reilly EM, Klimstra DS.DNA mismatch repair abnormalities in acinar cell carcinoma

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8 317

of the pancreas: frequency and clinical significance. Pancreas.2014;43(8):1264–1270.

26. Karamurzin Y, Zeng Z, Stadler ZK, et al. Unusual DNAmismatch repair-deficient tumors in Lynch syndrome: areport of new cases and review of the literature. Hum Pathol.2012;43(10):1677–1687.

27. Gaujoux S, Tissier F, Ragazzon B, et al. Pancreatic ductal andacinar cell neoplasms in Carney complex: a possible newassociation. J Clin Endocrinol Metab. 2011;96(11):E1888–E1895.

28. Basturk O, Zamboni G, Klimstra DS, et al. Intraductal andpapillary variants of acinar cell carcinomas: a new addition tothe challenging differential diagnosis of intraductal neo-plasms. Am J Surg Pathol. 2007;31(3):363–370.

29. Kobayashi S, Asakura T, Ohike N, et al. Mixed acinar-endocrine carcinoma of the pancreas with intraductal growthinto the main pancreatic duct: report of a case. Surg Today.2010;40(4):380–384.

30. La Rosa S, Franzi F, Marchet S, et al. The monoclonal anti-BCL10 antibody (clone 331.1) is a sensitive and specific markerof pancreatic acinar cell carcinoma and pancreatic metapla-sia. Virchows Arch. 2009;454(2):133–142.

31. Hosoda W, Sasaki E, Murakami Y, Yamao K, Shimizu Y,Yatabe Y. BCL10 as a useful marker for pancreatic acinar cellcarcinoma, especially using endoscopic ultrasound cytologyspecimens. Pathol Int. 2013;63(3):176–182.

32. Vakiani E, Young RH, Carcangiu ML, Klimstra DS. Acinar cellcarcinoma of the pancreas metastatic to the ovary: a report of4 cases. Am J Surg Pathol. 2008;32(10):1540–1545.

33. Ang C, Herran LA, Lagunes DR, Klimstra DS, Kemeny NE. Acase report of a patient with advanced acinar cell carcinomaof the pancreas: long-term survival with regional, systemicand targeted therapy. Tumori. 2013;99(2):e61–e64.

34. Kitagami H, Kondo S, Hirano S, Kawakami H, Egawa S, TanakaM. Acinar cell carcinoma of the pancreas: clinical analysis of115 patients from Pancreatic Cancer Registry of Japan Pan-creas Society. Pancreas. 2007;35(1):42–46.

35. Wisnoski NC, Townsend CM Jr., Nealon WH, Freeman JL, RiallTS. 672 patients with acinar cell carcinoma of the pancreas: apopulation-based comparison to pancreatic adenocarcinoma.Surgery. 2008;144(2):141–148.

36. Jakobsen M, Kloppel G, Detlefsen S. Mixed acinar-neuroendocrine carcinoma of the pancreas: a case report anda review. Histol Histopathol. 2016:11767 [Epub ahead of print].

37. Ohike N, Kosmahl M, Kloppel G. Mixed acinar-endocrinecarcinoma of the pancreas. A clinicopathological study andcomparison with acinar-cell carcinoma. Virchows Arch. 2004;445(3):231–235.

38. Yu R, Jih L, Zhai J, et al. Mixed acinar-endocrine carcinoma ofthe pancreas: new clinical and pathological features in acontemporary series. Pancreas. 2013;42(3):429–435.

39. Ogbonna OH, Garcon MC, Syrigos KN, Saif MW. Mixed acinar-neuroendocrine carcinoma of the pancreas with neuroendo-crine predominance. Case Rep Med. 2013;2013:705092.

40. Bien E, Godzinski J, Dall'igna P, et al. Pancreatoblastoma: areport from the European cooperative study group for paedi-atric rare tumours (EXPeRT). Eur J Cancer. 2011;47(15):2347–2352.

41. Dunn JL, Longnecker DS. Pancreatoblastoma in an older adult.Arch Pathol Lab Med. 1995;119(6):547–551.

42. Hoorens A, Gebhard F, Kraft K, Lemoine NR, Kloppel G.Pancreatoblastoma in an adult: its separation from acinarcell carcinoma. Virchows Arch. 1994;424(5):485–490.

43. Chisholm KM, Hsu CH, Kim MJ, Rangaswami A, Gray HazardFK. Congenital pancreatoblastoma: report of an atypical caseand review of the literature. J Pediatr Hematol Oncol. 2012;34(4):310–315.

44. Sorrentino S, Conte M, Nozza P, et al. Simultaneous occur-rence of pancreatoblastoma and neuroblastoma in a newborn

with beckwith-wiedemann syndrome. J Pediatr Hematol Oncol.2010;32(5):e207–e209.

45. Sugai M, Kimura N, Umehara M, et al. A case of pancreato-blastoma prenatally diagnosed as intraperitoneal cyst. PediatrSurg Int. 2006;22(10):845–847.

46. Muguerza R, Rodriguez A, Formigo E, et al. Pancreatoblastomaassociated with incomplete Beckwith-Wiedemann syndrome:case report and review of the literature. J Pediatr Surg. 2005;40(8):1341–1344.

47. Defachelles AS, Martin De Lassalle E, Boutard P, Nelken B,Schneider P, Patte C. Pancreatoblastoma in childhood: clinicalcourse and therapeutic management of seven patients. MedPediatr Oncol. 2001;37(1):47–52.

48. Morohoshi T, Sagawa F, Mitsuya T. Pancreatoblastoma withmarked elevation of serum alpha-fetoprotein. An autopsycase report with immunocytochemical study. Virchows Arch APathol Anat Histopathol. 1990;416(3):265–270.

49. Tanaka Y, Ijiri R, Yamanaka S, et al. Pancreatoblastoma:optically clear nuclei in squamoid corpuscles are rich inbiotin. Mod Pathol. 1998;11(10):945–949.

50. Hasegawa Y, Ishida Y, Kato K, et al. Pancreatoblastoma. A casereport with special emphasis on squamoid corpuscles withoptically clear nuclei rich in biotin. Acta Cytol. 2003;47(4):679–684.

51. Morohoshi T, Kanda M, Horie A, et al. Immunocytochemicalmarkers of uncommon pancreatic tumors. Acinar cell carci-noma, pancreatoblastoma, and solid cystic (papillary-cystic)tumor. Cancer. 1987;59(4):739–747.

52. Tanaka Y, Kato K, Notohara K, et al. Significance of aberrant(cytoplasmic/nuclear) expression of beta-catenin in pancrea-toblastoma. J Pathol. 2003;199(2):185–190.

53. Salman B, Brat G, Yoon YS, et al. The diagnosis and surgicaltreatment of pancreatoblastoma in adults: a case series andreview of the literature. J Gastrointest Surg. 2013;17(12):2153–2161.

54. Glick RD, Pashankar FD, Pappo A, Laquaglia MP. Managementof pancreatoblastoma in children and young adults. J PediatrHematol Oncol. 2012;34(suppl 2):S47–S50.

55. Stamm B, Burger H, Hollinger A. Acinar cell cystadenocarci-noma of the pancreas. Cancer. 1987;60(10):2542–2547.

56. Cantrell BB, Cubilla AL, Erlandson RA, Fortner J, Fitzgerald PJ.Acinar cell cystadenocarcinoma of human pancreas. Cancer.1981;47(2):410–416.

57. Colombo P, Arizzi C, Roncalli M. Acinar cell cystadenocarci-noma of the pancreas: report of rare case and review of theliterature. Hum Pathol. 2004;35(12):1568–1571.

58. Chatelain D, Paye F, Mourra N, et al. Unilocular acinar cellcystadenoma of the pancreas an unusual acinar cell tumor.Am J Clin Pathol. 2002;118(2):211–214.

59. Albores-Saavedra J. Acinar cystadenoma of the pancreas: apreviously undescribed tumor. Ann Diagn Pathol. 2002;6(2):113–115.

60. Khor TS, Badizadegan K, Ferrone C, et al. Acinar cystadenomaof the pancreas: a clinicopathologic study of 10 cases includ-ing multilocular lesions with mural nodules. Am J Surg Pathol.2012;36(11):1579–1591.

61. Wang G, Ji L, Qu FZ, et al. Acinar cell cystadenoma of thepancreas: a retrospective analysis of ten-year experiencefrom a single academic institution. Pancreatology. 2016.

62. Singhi AD, Norwood S, Liu TC, et al. Acinar cell cystadenomaof the pancreas: a benign neoplasm or non-neoplastic bal-looning of acinar and ductal epithelium? Am J Surg Pathol.2013;37(9):1329–1335.

63. Couvelard A, Terris B, Hammel P, et al. Acinar cystic trans-formation of the pancreas (or acinar cell cystadenoma), a rareand recently described entity. Ann Pathol. 2002;22(5):397–400.

64. Wildgruber M, Rummeny EJ, Gaa J. Primary acinar cellcarcinoma of the liver. RoFo. 2013;185(6):572–573.

65. Zundler S, Erber R, Agaimy A, et al. Pancreatic panniculitis ina patient with pancreatic-type acinar cell carcinoma of the

S E M I N A R S I N D I A G N O S T I C P A T H O L O G Y 3 3 ( 2 0 1 6 ) 3 0 7 – 3 1 8318

liver—case report and review of literature. BMC Cancer.2016;16(1):130.

66. Agaimy A, Kaiser A, Becker K, et al. Pancreatic-type acinar cellcarcinoma of the liver: a clinicopathologic study of fourpatients. Mod Pathol. 2011;24(12):1620–1626.

67. Fukunaga M. Gastric carcinoma resembling pancreatic mixedacinar-endocrine carcinoma. Hum Pathol. 2002;33(5):569–573.

68. Jain D, Eslami-Varzaneh F, Takano AM, et al. Compositeglandular and endocrine tumors of the stomach with pancreaticacinar differentiation. Am J Surg Pathol. 2005;29(11):1524–1529.

69. Kusafuka K, Bando E, Muramatsu K, et al. Pancreatic-typemixed acinar-endocrine carcinoma with alpha-fetoproteinproduction arising from the stomach: a report of anextremely rare case. Med Mol Morphol. 2009;42(3):167–174.

70. Lee H, Tang LH, Veras EF, Klimstra DS. The prevalence ofpancreatic acinar differentiation in gastric adenocarcinoma:report of a case and immunohistochemical study of 111additional cases. Am J Surg Pathol. 2012;36(3):402–408.

71. Doglioni C, Laurino L, Dei Tos AP, et al. Pancreatic (acinar)metaplasia of the gastric mucosa. Histology, ultrastructure,immunocytochemistry, and clinicopathologic correlations of101 cases. Am J Surg Pathol. 1993;17(11):1134–1143.

72. Suda K, Mizuguchi K, Hebisawa A, Wakabayashi T, Saito S.Pancreatic tissue in teratoma. Arch Pathol Lab Med. 1984;108(10):835–837.

73. Terhune PG, Heffess CS, Longnecker DS. Only wild-type c-Ki-ras codons 12, 13, and 61 in human pancreatic acinar cellcarcinomas. Mol Carcinog. 1994;10(2):110–114.

74. Moore PS, Orlandini S, Zamboni G, et al. Pancreatic tumours:molecular pathways implicated in ductal cancer are involvedin ampullary but not in exocrine nonductal or endocrinetumorigenesis. Br J Cancer. 2001;84(2):253–262.

75. La Rosa S, Bernasconi B, Frattini M, et al. TP53 alterations inpancreatic acinar cell carcinoma: new insights into themolecular pathology of this rare cancer. Virchows Arch. 2016;468(3):289–296.

76. Furlan D, Sahnane N, Bernasconi B, et al. APC alterations arefrequently involved in the pathogenesis of acinar cell carci-noma of the pancreas, mainly through gene loss and pro-moter hypermethylation. Virchows Arch. 2014;464(5):553–564.

77. Soejima H, Higashimoto K. Epigenetic and genetic alterationsof the imprinting disorder Beckwith-Wiedemann syndromeand related disorders. J Hum Genet. 2013;58(7):402–409.

78. Kerr NJ, Chun YH, Yun K, Heathcott RW, Reeve AE, SullivanMJ. Pancreatoblastoma is associated with chromosome 11ploss of heterozygosity and IGF2 overexpression. Med PediatrOncol. 2002;39(1):52–54.

79. Kerr NJ, Fukuzawa R, Reeve AE, Sullivan MJ. Beckwith-Wiedemann syndrome, pancreatoblastoma, and the wntsignaling pathway. Am J Pathol. 2002;160(4):1541–1542 [authorreply 1542].

80. Chmielecki J, Hutchinson KE, Frampton GM, et al. Compre-hensive genomic profiling of pancreatic acinar cell carcino-mas identifies recurrent RAF fusions and frequentinactivation of DNA repair genes. Cancer Discov. 2014;4(12):1398–1405.

81. Jiao Y, Yonescu R, Offerhaus GJ, et al. Whole-exome sequenc-ing of pancreatic neoplasms with acinar differentiation.J Pathol. 2014;232(4):428–435.

82. Wang LBO, Chmielecki J, Ross J, et al. Development of BRAFFISH assay for the detection of BRAF gene rearrangementsidentified in pancreatic acinar cell carcinomas [abstract]. ModPathol. 2015;28(suppl 2):451A.

83. Lonardo F, Cubilla AL, Klimstra DS. Microadenocarcinoma ofthe pancreas—morphologic pattern or pathologic entity? Areevaluation of the original series. Am J Surg Pathol. 1996;20(11):1385–1393.

84. Basturk O, Tang L, Hruban RH, et al. Poorly differentiatedneuroendocrine carcinomas of the pancreas: a clinicopatho-logic analysis of 44 cases. Am J Surg Pathol. 2014;38(4):437–447.

85. Sigel CS, Klimstra DS. Cytomorphologic and immunopheno-typical features of acinar cell neoplasms of the pancreas.Cancer Cytopathol. 2013;121(8):459–470.

86. Labate AM, Klimstra DL, Zakowski MF. Comparative cytologicfeatures of pancreatic acinar cell carcinoma and islet celltumor. Diagn Cytopathol. 1997;16(2):112–116.

87. Tang LH, Aydin H, Brennan MF, Klimstra DS. Clinicallyaggressive solid pseudopapillary tumors of the pancreas: areport of two cases with components of undifferentiatedcarcinoma and a comparative clinicopathologic analysis of34 conventional cases. Am J Surg Pathol. 2005;29(4):512–519.

88. Klimstra DS, Wenig BM, Heffess CS. Solid-pseudopapillarytumor of the pancreas: a typically cystic carcinoma of lowmalignant potential. Semin Diagn Pathol. 2000;17(1):66–80.

89. Notohara K, Hamazaki S, Tsukayama C, et al. Solid-pseudopapillary tumor of the pancreas: immunohistochem-ical localization of neuroendocrine markers and CD10. Am JSurg Pathol. 2000;24(10):1361–1371.

90. Abraham SC, Klimstra DS, Wilentz RE, et al. Solid-pseudopapillary tumors of the pancreas are genetically dis-tinct from pancreatic ductal adenocarcinomas and almostalways harbor beta-catenin mutations. Am J Pathol. 2002;160(4):1361–1369.