`o~uLioooolooo.64
`•
`¥+
`• 6
`j
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 001
`
`
`
`World Health Organization Classification of Tumours
`
`.,
`
`OMS
`
`WHO
`
`~a . ill
`'&~
`{jR
`~~
`
`~ . ,~
`
`International Agency for Research on Cancer (IARC)
`
`Pathology and Genetics of
`'
`Tumours of Endocrine Organs
`
`Edited by
`
`Ronald A. Delellis
`
`Ricardo V. Lloyd
`
`Philipp U. Heitz
`
`Charis Eng
`
`I ARC Press
`
`Lyon , 2004
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 002
`
`
`
`World Health Organization Classification of Tumours
`
`Series Editors Paul Kleihues, M.D.
`Leslie H. Sobin, M.D.
`
`/c
`:<-
`
`Pathology and Genetics of Tumours of Endocrine Organs
`
`Editors Ronald A. Delellis, M.D.
`Ricardo V. Lloyd, M.D.
`Philipp U. Heitz, M.D.
`Charis Eng, M.D., Ph.D.
`
`Coordinating Editors
`
`Wojciech Biernat, M.D.
`Janice Sych, Ph.D.
`
`LC Control Number Editorial Assistants
`
`Ill 111111\~1\1 Ill I' II" 111\1
`
`:~)1.4n17
`
`Isabelle Forcier
`Voichita Meyronein
`Agnes Meneghel
`
`Layout
`
`Vanessa Meister
`Marlen Grassinger
`Sibylle Soring
`
`Illustrations
`
`Thomas Odin
`
`Printed by
`
`Team Rush
`69603 Villeurbanne, France
`
`Publisher
`
`I ARC Press
`International Agency for
`Research on Cancer (IARC)
`69008 Lyon, France
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 003
`
`
`
`This volume was produced in collaboration with the
`
`International Academy of Pathology (lAP)
`
`The WHO Classification of Tumours of Endocrine Organs
`presented in this book reflects the views of a Working Group that
`convened for an Editorial and Consensus Conference in Lyon, France,
`April 23-26, 2003
`
`Members of the Working Group are indicated
`in the List of Contributors on page 263
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 004
`
`
`
`Published by IARC Press, International Agency for Research on Cancer,
`150 cours Albert Thomas, F-69008 Lyon, France
`
`© International Agency for Research on Cancer, 2004
`
`Publications of the World Health Organization enjoy copyright protection in
`accordance with the provisions of Protocol 2 of the Universal Copyright Convention.
`All rights reserved.
`
`The International Agency for Research on Cancer welcomes
`requests for permission to reproduce or translate its publications, in part or in fulL
`Requests for permission to reproduce figures or charts from this publication should be directed to
`the respective contributor (see section Source of Charts and Photographs).
`
`The designations used and the presentation of the material in this publication do not imply the
`expression of any opinion whatsoever on the part of the Secretariat of the
`World Health Organization concerning the legal status of any country, territory, city,
`or area or of its authorities, or concerning the delimitation of its frontiers or boundaries.
`
`The mention of specific companies or of certain manufacturers' products does not imply
`that they are endorsed or recommended by the World Health Organization in preference to others
`of a similar nature that are not mentioned. Errors and omissions excepted,
`the names of proprietary products are distinguished by initial capital letters.
`
`The authors alone are responsible for the views expressed in this publication.
`
`Enquiries should be addressed to the
`Communications Unit. International Agency for Research on Cancer. 69008 Lyon , France.
`which will provide the latest information on any changes made to the text and plans for new editions.
`
`Format for bibliographic citations:
`Delellis RA , Lloyd R.V, Heitz P.U ., Eng C. (Eds.) World Health Organization
`Classification of Tumours. Pathology and Genetics of Tumours of Endocrine Organs.
`!ARC Press: Lyon 2004
`
`IARC Library Cataloguing in Publication Data
`
`Pathology and genetics of tumours of endocrine organs I
`editors R.A. Delellis ... [et al.]
`
`(World Health Organization classification of tumours ; 8)
`
`1. Pituitary gland neoplasms - genetics 2. Pituitary gland neoplasms - pathology
`3. Thyroid neoplasms - genetics 4. Thyroid neoplasms , - pathology
`5. Endocrine pancreas neoplasms - genetics 6. Endocrine pancreas neoplasms - pathology
`7. Adrenal neoplasms- genetics 8. Adrenal neoplasms- pathology
`I. Delellis Ronald A
`II. Series
`
`ISBN 92 832 2416 7 (NLM Classification WJ 160)
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 005
`
`
`
`CHAPTER 4
`
`Tumours of the Endocrine Pancreas
`
`Tumours of the endocrine pancreas are much less frequent
`than those of the exocrine pancreas and usually have a much
`better prognosis. Hormonessecreted by endocrine neoplasms
`include insulin. glucagon.' somatostatin. gastrin, vasoactive
`intestinal polypeptide (VIP). pancreatic polypeptide (PP). sero(cid:173)
`tonin. ACTH;;: ahd calcitonin. Depending on the peptide hor(cid:173)
`mones produced. they may-e caus~distinct clinical syndromes.
`including life-threatening hypoglycaemia. gastric and/or duo,,
`dena! ulcers. or dehydration due to diarrhoea. ,
`··
`, .
`' ,.,
`Most pancreatic neuroendocrine tumours can be surgically
`resected and this leads, to a rapid regression of clinicalsymp(cid:173)
`toms. Poorly differentiaied neoplasms may be metastatic at the
`time of clinical presentation. and this is associated with a poor
`prognosis.
`Genetic susceptibility may play an important role. Up to 20% of
`gastrinomas are associated with . the inherited MEN-1 syn(cid:173)
`drome~·
`
`~.;~~~~·t. -~-~
`"'
`
`~··~··
`
`•.
`
`I
`
`:~
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 006
`
`
`
`WHO histological classification of tumours of the
`endocrine pancreas
`
`Well-differentiated endocrine tumotJr
`Functioning
`Insulin-producing (insulinoma)
`Glucagon-producing (glucagonoma)
`Somatostatin-producing (somatostatinoma)
`Gastrin-producing (gastrinoma)
`VIP-producing (VIPoma)
`Others
`
`Non-functioning
`Microadenoma (<0.5 em)
`Others
`
`8150/1 1
`
`8151/1
`8152/1
`8156/1
`8153/1
`8155/1
`
`8150/0
`
`Well-differentiated endocrine carcinoma
`Functioning
`Insulin-producing (insulinoma)
`Glucagon-producing (glucagonoma)
`Somatostatin-producing (somatostatinoma)
`Gastrin-producing (gastrinoma)
`VIP-producing (VIPoma)
`Serotonin producing with carcinoid syndrome
`ACTH producing with Cushing syndrome
`
`Non-functioning
`
`Poorly-differentiated endocrine carcinoma -
`small cell carcinoma
`
`Mixed exocrine -endocrine carcinoma
`
`8150/3
`
`8151/3
`8152/3
`8156/3
`8153/3
`8155/3
`8241/3
`8150/3
`
`8150/3
`
`8041/3
`
`8154/3
`
`1 Morphology code of the International Classification of Diseases for Oncology (ICD-0) {664} and the Systematized Nomenclature of Medicine lhttp://snomed.org).
`Behaviour is coded /0 for benign tumours, /3 for malignant tumours, and /1 for borderline or uncertain behaviour.
`
`176
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 007
`
`
`
`'ancreatic endocrine tumours:
`ntroduction
`
`~rminology
`this chapter the term "pancreatic
`:docrine
`tumour"
`replaces earlier
`rms, e.g. pancreatic neuroendocrine
`mour, islet cell tumour, and APUOoma.
`
`epidemiology
`tumours are
`mcreatic endocrine
`1common and represent 1-2% of all
`tumours
`•.mcreatic neoplasms. The
`1ow no significant gender predilection
`r1d occur at all ages, with a peak inci(cid:173)
`ence between 30-60 years. Clinically
`nrecognised or asymptomatic, and usu(cid:173)
`:ly small tumours (diameter less than 1
`have been found in 0.4-1.5% of uns(cid:173)
`lected autopsies (779, 1086,1154,1356,
`874, 2097f
`incidence of
`he
`reported overall
`,Jmours of the endocnne pancreas has
`1creased during recent years. This is
`lrobably due to the application of more
`;ensitive diagnostic approaches such as
`•11aging techniques, reliable laboratory
`ests and careful "morphofunctional"
`malysis by immunohistochemical and
`nolecular biological techniques (2097f
`
`Endocrine function
`i::Jancreatic endocrine tumours a~e sepa(cid:173)
`ated based on their clinical manifesta(cid:173)
`'ion, into functioning and non-functioning.
`
`Functioning tumours are associated with
`clinical syndromes caused by inappro(cid:173)
`priate secretion of hormones. Within this
`group are insulinomas, glucagonomas,
`somatostatinomas, gastrinomas, VIP(cid:173)
`omas, and other less common tumours.
`The clinical syndromes are described
`under the headings of the various func(cid:173)
`tioning tumours.
`
`Non-functioning tumours (or inactive,
`clinically silent, nonsyndromic) are not
`associated with a distinct hormonal syn(cid:173)
`drome but may still show elevated hor(cid:173)
`mone levels in the blood or immunoreac(cid:173)
`tivity in tissue sections. For this reason,
`term "nonsyndromic" pancreatic
`the
`endocrine tumour may more accurately
`describe this group, but is not widely
`used (2094l Therefore, tumours with the
`majority of cells expressing (and often
`secreting) pancreatic polypeptide (PP),
`or neurotensin are included in the group
`of non-functioning tumours (as are many
`"0-cell tumours" or "somatostatin pro(cid:173)
`ducing tumours"), because they do not
`cause a distinct hormonal syndrome.
`Non-functioning tumours only become
`clinically apparent due to their large size,
`to Invasion of adjacent organs, or the
`occurrence of metastases. Rarely they
`may present as acute pancreatitis.
`
`lnsulinomas
`
`Glucagonomas Somatostatlnomas Gastrlnomas
`
`Tumours
`producing ectopic
`hormones
`Fig. 4.01 Frequency of various types of pancreatic endocrine tumours, based on a series of 638 cases.
`
`Vipomas
`
`Nonfunctioning
`tumours
`
`Ph.U. He1tz
`P Komminoth
`A. Perren
`O.S. Klimstra
`Y Dayal
`
`C. Bordi
`J. Lechago
`BA Centeno
`G. Kl6ppel
`
`lncreas1ngly, they are incidentally detect(cid:173)
`ed on imaging tests.
`Tumours with a diameter of less than 0.5
`em, the minimum size required for gross
`detection, are defined as microadeno(cid:173)
`mas and are, as a rule, non-functioning.
`
`Macroscopy
`The majority of the tumours are well
`demarcated and solitary, showing a
`white-yellow or pink-brown colour. The
`consistency is variable. Rarely, they are
`cystic (1301l
`Their diameter ranges usually around 1-5
`em. Among
`the functioning tumours,
`insulinomas are usually smaller (less
`than 2 em in diameter) than glucagono(cid:173)
`mas, somatostatinomas, gastrinomas or
`VIPomas, but the size of the tumours is
`not related to the severity of the hormon(cid:173)
`ally induced symptoms.
`Non-functioning tumours are generally
`larger than 2 em in diameter (often 5 em
`or more). They are probably detected rel(cid:173)
`atively late because they do not induce a
`clinical syndrome due to inappropriate
`hormone secretion.
`Tumours with a diameter of more than 2
`em have an increased risk of malignant
`behaviour and those over 3 em are usu(cid:173)
`ally malignant.
`
`Cytopathology
`Fine needle aspiration biopsy (FNAB) is
`a useful method for investigating pancre(cid:173)
`atic endocrine tumours and their metas(cid:173)
`techniques
`include
`tases. Guidance
`computed tomography (CT),
`transab(cid:173)
`dominal ultrasonography (TUS) and,
`more recently, endoscopic ultrasonogra(cid:173)
`phy (EUS) The cytomorphological fea(cid:173)
`tures of pancreatic endocrine tumours
`are the same for functioning and non(cid:173)
`functioning tumours (25, 161 ,401 ,2072).
`Smears are usually uniformly cellular and
`composed of a relatively monotonous
`cells
`predominantly
`population of
`arranged singly, but also in loose clus(cid:173)
`ters or pseudorosettes. The round
`to
`contoured
`nuclei
`ovoid,
`smoothly
`demonstrate a salt-and-pepper chro(cid:173)
`matin pattern. The
`cytoplasm
`is
`
`177
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 008
`
`
`
`and an elevated mitotrc rndex (> 10 p
`10
`
`Markers of neuroendocrine
`differentiation
`Pancreatic endocnne tumours can cleE
`ly be identified by usrng antibodies •
`markers common to all or most neuroe
`docrine cells, 1.e. synaptophysin, an int•
`gral membrane glycoprotein of synap
`vesicles, or protein gene product
`9.5, a cytoplasmic protein. Neuron sp·
`cifrc enolase
`is widely reported
`stain
`these tumours. but the resu
`should be
`with caution rn lip
`of the low specificity of this marker. T,
`presence of
`immunoreactive chron
`granins, which are glycoprotein comr
`nents of the matrix of neuroendocri:
`secretory granules. indicates the pre
`ence of secretory
`i.e. sor:
`degree of differentiation of the tum1
`cells 1187 4,20971. In less well granulat•
`examples, chromogranin staining is g1
`erally less intense and extensive,
`diffuse strong
`tor synaptophy':
`Pancreatic endocrine tumours also cc.
`tain cytokeratins 8, 18 and 19 and n
`often contain neurofilaments as well.
`
`Hormonal markers
`The use of these markers
`
`tumours
`functioning
`However,
`defrned on the basis of clinrca! symptc"
`due to inappropriate hormone secret
`rather than immunohistochemrcal fr·
`
`In the maJority of functronrng tumours.
`hormone
`the syndrome can
`detected by
`immunohistochemis
`However. staining rntensity or the nurp·
`of positive cells is not related to
`severity of symptoms. This is in part c
`to the impairment of the
`
`of horm ·
`and secretion. There is a i
`of heterogeneity among the rr
`vidual tumour cells rn
`the conten·
`immunoreactive
`hormone::
`rnRI-.JA.
`correspondrng
`In addi:
`rmrnunmeactive horrncmes with redu·
`biological activity or with a greatly s!
`ened or
`half-life :n the
`may be produced.
`functror
`tumours m;w paradoxically lack rrnm:'
`hrstochemrcaily detecta!Jie horrno:
`seen:
`1-rrR!\jA, oetect;or·1 r
`
`E
`F
`Fig. 4.02 Growth patterns of pancreatic endocrine tumours. A Solid growth pattern. B Trabecular growth
`pattern. C Gland formation. The cells lining the lumina are cytologically identical to those of the remainder
`of the tumour. D Gyriform growth pattern: Nested architecture and peripheral palisading of nuclei. E Clear
`lipid-rich cells.H&E. F Solid growth pattern and focal oncocytic metaplasia.
`
`amphophilic and varies in quantity and
`density Some cells may be stripped of
`their cytoplasm whereas others may
`have abundant cytoplasm and a plasma(cid:173)
`cytoid appearance.
`
`Histopathology
`Most pancreatic endocrine tumours are
`well differentiated showing various histo(cid:173)
`logical patterns. characterised by a solid,
`trabecular. glandular. gyriform, tubuloaci(cid:173)
`nar or pseudorosette arrangement of their
`cells. The cells are relatively uniform.
`show
`cyto(cid:173)
`plasm and a centrally located round to
`oval nucleus that may display a distinct
`nucleolus. Occasionally, clear cells, vac(cid:173)
`uolated
`lrpid-rich
`cells.
`oncocytes
`or"rhabdoid'' features 117171 may be
`observed. The amount of stroma anci
`
`tumour. However, in most instances these
`features are sufficiently drstinctive to per(cid:173)
`mit recognition of the endocrine nature of
`a
`tumour.
`In general. the histological pattern of a
`tumour does not allow a conclusion as to
`its functional state or type of the hormone
`to
`produced. There are two
`this rule amyloid deposits are indicative
`of insulinomas. and
`structures
`contarning psammoma bodies are com(cid:173)
`monly observed in somatostatin produc(cid:173)
`ing tumours of the periampullary duode(cid:173)
`num.
`Poorly differentiated endocnne carcino(cid:173)
`mas are uncommon. These highly
`neoplasms are hardly recog(cid:173)
`nizable as endocrine tumours at
`first
`sight and
`Immunohistochemical
`their neuroen(cid:173)
`to reveal
`show rather
`
`178
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 009
`
`
`
`niques may be useful. On the other hand .
`an immunoreactive hormone may not be
`secreted due to an impaired secretory
`pathway. Thus,
`immunoreactive hor(cid:173)
`mones very often can be localized to
`ce lls of non-functioning tumours.
`Upon careful investigation it has become
`obvious that many tumou rs are composed
`of more than one phaenotype (multihor(cid:173)
`monal tumou rs). As a rule however, only
`one cell type correlates with an associat(cid:173)
`ed syndrome of endocrine hyperfunction.
`The classification of the tumours must
`therefore be "morphofunctionat", i.e. not
`only based on cell typing. It must primari(cid:173)
`'y take into cons ideration the clinical signs
`and symptoms , and determination of cir(cid:173)
`culating hormone concentrations. Meta(cid:173)
`stases may produce hormones other than
`those found in the primary 120971
`
`Staging and prognosis
`No stag ing system, such as the UICC
`TNM system, has been applied to pan(cid:173)
`c reatic endocrine tumours .
`The most reliable evidence of malignant
`1n
`pancreatic endocrine
`behaviour
`tumours is metastasis to the regional
`lymph nodes or the liver or gross infiltration
`of adjacent organs. Many of the smatter
`examples. includ ing most insulinomas.
`probably have malignant potential , but
`interruption of their natural history by sur(cid:173)
`gical resection prevents the expression of
`such potential. A proposal has been made
`to separate pancreatic endocrine tumours
`1nto prognostic groups based on mitotic
`rate and necrosis 18991.
`Among the functioning tumours, most
`1nsutinomas show benign behaviour. In
`contrast. the oth er types of functioni ng
`tumours fall either into the categories of
`Nell-d ifferentiated tumours wi th uncertain
`behaviour (approx. 10-15%) or. more fre-
`
`Table 4.01
`lmmunophaenotyping of pancreati c endocrine
`tumours.
`
`Table 4.02
`Criteria tor the clinic opathological classification of
`pancreatic endocrine tumours.
`
`General neuroendocrine markers
`Synaptophysin
`Protein Gene Product IPGP) 9.5
`CD 56
`MAP18
`
`Markers of the matrix of secretory granules
`Chromogranins
`
`Hormone (cell type)- specific markers
`Insulin
`Glucagon
`Somatostatin
`Gastrin
`Vasoactive Intestinal Polypeptide (VIP)
`Pancreatic Polypeptide (PP)
`Serotonin
`ACTH
`Neurotensin
`Calcitonin
`
`Well-differentiated endocrine tumour
`
`1.1 'Benign' behaviour
`Confined to the pancreas, non-angioinvasive,
`no pe rineu ral invasi on, <2 em in diameter,
`<2 mitoses/10 HPF and <2% Ki-67 positive cells
`
`1.2 Uncertain behaviour
`Confined to the pancreas and one or more of
`the following features: ?2 em in diameter, 2-10
`mitoses/10 HPF, >2% Ki-67 positive cells,
`angioinvasion, perineural invasion
`
`2 Well-differentiated endocrine carcinoma
`Low grade malignant
`Gross local invasion and/or metastases
`
`3 Poorly-differentiated endocrine carcinoma
`High grade malignant
`> 10 mitoses I 10 HPF
`
`quently, of well-differentiated carcinomas
`(approx. 85-90%).
`A small number of non-functioning
`tumours are well-differentiated tumours
`showing benign or uncertain behaviour;
`however. the vast majority (approx. 90-
`95%) are well-differentiated carcinomas.
`Poorly differentiated endocrine carci no(cid:173)
`mas are uncommon 120971.
`
`Clinicopathological correlations
`There is an obvious need to establish a
`close correlation between morphological
`classification and
`tumour-associated
`syndromes. This is emphasized by the
`difficulty in predicting the biological
`behavi our of well-differentiated endo(cid:173)
`crine tumours based on histological cri(cid:173)
`teria alone . In addition, available follow(cid:173)
`up studies most often refe r to tumours
`
`diagnosed according to the associated
`clinical syndrome due to inappropriate
`hormone secretion.
`To defin itely establish the benign nature
`of a tumour, a long clinical follow-up peri(cid:173)
`od is needed , because metastases may
`develop years after removal of the pri(cid:173)
`mary lesion.
`With the exception of the poorly differen(cid:173)
`tiated endocrine carcinomas, the pro(cid:173)
`gression of the disease is often remark(cid:173)
`ably slow. Survival for five to ten years
`after appearance of liver metastases is
`not uncommon. However, inappropriate
`secretion of hormones may cause life(cid:173)
`threatening hypog lycaemia , gastric
`and/or duodenal ulcers. or important toss
`of fluid by watery diarrhoea.
`The final 'morpho-functional' classifica(cid:173)
`tion of an endocrine tumour of the pan-
`
`~- .. -
`
`.·.• ;s
`?
`
`\
`4 / .
`~- .i,
`-~..;y
`
`.
`
`~" t •
`'\'
`
`';<
`
`,
`
`•
`·"'-
`-l
`Fig. 4.03 Fine needle aspiration (FNA) of a pancre-
`atic endocrine tumour. Amphophili c cyto plasm,
`uniform nuclei and finely granu la r chromatin,
`imp arting a sa lt-a nd-pepper appearance.
`
`A
`Fig. 4.04 Gastrinoma. Endocrine differentiation is more unive rsa l than cell-specific hormone production.
`A Virtually all tumour cells contain synaptophysin. B The majority, but clearly fewe r cells contain chromo(cid:173)
`granin A.
`
`8 ~-..
`
`179
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 010
`
`
`
`.. .
`
`•
`
`Fig. 4.06 A Pancreatic endocrine tumour. Vascular invasion. B Well differentiated endocrine carcinom c
`with massive vascular invasion.
`
`cytokeratin , but expresses vimentin ,
`alpha-1-antitrypsin and CD10, 8) it lacks
`immunohistochemical expression of pep(cid:173)
`tide hormones, 9) it occurs predominant(cid:173)
`ly
`in young women, and 1 0) most
`tumou rs show a benign biological behav(cid:173)
`iour 11874)
`Further tumours which may be contused
`with pancreatic endocrine tumours are
`acinar cel l carcinoma, pancreatoblas(cid:173)
`tomas 11537), poorly differentiated ductal
`adenocarcinoma. clear cell carcinoma,
`epithelioid
`gastrointestinal
`stromal
`tumours,
`primitive neuroectodermal
`tumours (PNET) and pancreatic metas(cid:173)
`tases (e.g., renal cel l carcinoma, small
`cell lung carcinoma. melanoma)
`Acinar cell carcinomas, which histologi(cid:173)
`cal ly may be very difficult to distinguish
`from endocrine tumours of the pancreas,
`usually produce trypsin (-ogen) and
`other pancreatic (pro- ) enzymes. The
`same is true for pancreatoblastomas .
`Both neoplasms may contain scattered
`endocrine cells . Truly mixed acinar(cid:173)
`endocrine or ductal-endocrine carcino(cid:173)
`mas are very rare. In these tumours the
`should
`compon ent
`endocrine
`cell
`account for at least one third of the entire
`cell population. Poorly differentiated duc(cid:173)
`tal adenocarcinomas as well as clear cell
`carcinomas reveal focal expression of
`
`mucin (MUC 1) and carcinoembryonic
`antigen (CEA). Most epithelioid gastroin
`testinal stromal tumours (G IST) are ch ar
`acterized by the expression of C-Kil
`(CD117) an d absence of stain ing to:
`neuroen docrine markers. PNETs exprest
`a set of markers, incl uding CD99. Meta
`stases of clear ce ll carcinomas of the kid·
`ney lack neuroendocrine markers, but ir:
`addition
`to cytokeratins
`frequentl y
`express vimentin and CD 1 0.
`
`Cytopathology
`The key entity in the differential diagnos i ~
`is acinar cell carcinoma. It can be distin
`guished
`from pancreatic endoc rin E
`tumours by its arrangement in
`loose
`grapelike cl usters, granular cytoplasn
`and prominent cherry
`red nucleol
`11191 ). The neoplastic cell s from a solid
`pseudopapillary tumour. when detachec
`from the fibrovascular cores may be mis
`taken for those of a pancreatic endocrinf
`tumour A search in the remainder of thf
`smear for structures with the characteri s
`tic three-laye red papillary architecture (<
`central capillary, a middle layer of myx
`oid stroma and an outer layer of neoplas
`tic cells) will yield the correct interpreta
`lion 1760)
`Pancreati c endocrine tumours may bt
`mistaken for lymphomas because the•
`
`Table 4.03
`Adverse prognostic factors of. well-differentiated pancreatic endocrine tumours.
`
`Metastasis
`Gross invasion
`Tumour diameter
`Angioinvasion
`Perineural invasion
`Mitose s
`Proliferative index Ki-67 I MIB-1
`Necrosis
`Functioning tumours except insulinoma
`
`Regional lymph nodes, liver
`Adjacent organs
`2 em or more
`Veins, lymphatic vessels
`lntrapancreatic nerves
`>2 per 10 HPF
`>2%
`
`creas should take into consideration ( 1)
`the clinical syndrome induced by or asso(cid:173)
`ciated with , a tumour, (2) determination of
`the blood concentration of hormone(s) to
`identify the hormone(s) secreted by the
`tumour, (3) the size (mass) of the tumour,
`( 4) the histological differentiation and
`probable biological behaviour of the
`tumour, (5) the phaenotype(s) of the vari(cid:173)
`ous tumour cells and , if necessary and
`feasible, (6) molecular genetic analysis of
`the tumour.
`
`Differential diagnosis
`Histopathology
`Most pancreatic endocrine tumours are
`recognizable without much difficulty. The
`use of immunohi stochemical markers of
`th e neuroendocrine phenotype and of
`hormonal content most often can estab(cid:173)
`lish the diagnosis unequivocally.
`An important differential diagnostic prob(cid:173)
`lem is to distinguish solid-pseudopapil(cid:173)
`lary neoplasms from endocrine tumours
`of the pancreas. Solid-pseudopapillary
`neoplasms morpholog ically resemble
`endocrine tumours, and furthermore,
`produce CD56, NSE and sometimes
`synaptophysin, as has been demonstrat(cid:173)
`ed by immunohistochemistry. Arguments
`in favour of the diagnosis of a solid(cid:173)
`pseudopapillary neoplasm of the pan(cid:173)
`creas are the following 1) it does not pro(cid:173)
`duce a hormonal syndrome but only local
`symptoms, 2) it is usually large , with a
`diameter often over 5 em, 3) it contains
`clusters of cel ls with a clear foamy cyto(cid:173)
`plasm, 4) it often shows aggregates of
`PAS-positive hyaline globules in and
`between the tumour cells, 5) it contains
`broad, hyalinized septa incl udmg small
`blood vessels, 6) it displays haemor(cid:173)
`rhages . necrotic foci and occ asional ly
`cholesterol crystals, 7) it lacks expres(cid:173)
`sion of chromogran1 n and usually also
`
`180
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 011
`
`
`
`;able 4.04
`!enetic alterations detected by loss of heterozygosity analysis (LOH), comparative genomic hybridization (CGH) and mutation analysis.
`
`Locus
`
`1p36-
`1q32-
`3p23-
`3p25-26-
`6q22-
`9p-
`9q+
`10q23-
`11p14-
`11q13
`11q22-23
`12p12+
`15q-
`17p13-
`17p+
`18q21-
`22q12.1
`Xq-
`Y-
`
`LOH
`
`10/29 (34%)
`8/29 (28%)
`23/31 (74%)
`31/73 (42%)
`43/69 (62%)
`12/37 (32%)
`
`8/16 (50%)
`
`75/111 (67%)
`20/37 (54%)
`
`15/40 (38%)
`
`23/68 (34%)
`9/12 (75%)
`11/23 (48%)
`5/14 (36%)
`
`Gene
`
`Mutation
`
`CGH {2105,2107,2154,2490}
`
`Reference
`
`VHL
`
`1/75 (1 %)
`
`CDKN2Npl6
`
`1/44 (2%)
`
`PTEN
`
`MEN I
`SDHD
`K-Ras
`SMAD3
`TP53
`
`DPC4
`
`1/31 (3%)
`
`33/155 (21%)
`0/20 (0%)
`1/39 (3%)
`0/18 (0%)
`1/40 (3%)
`
`0/41 (0%)
`
`21/102 (21 %)
`16/102 (16%)
`19/102 (19%)
`19/102 (19%)
`29/102 (28%)
`0/102 (0%)
`29/102 (28%)
`14/102 (14%)
`28/102 (27%)
`31/102 (30%)
`31/102 (30%)
`23/102 (23%)
`6/102 (6%)
`2/102 (2%)
`32/102 (31%)
`6/102 (6%)
`4/102 (4%)
`14/46 (30%)
`14/56 (25%)
`
`{541}
`{1830}
`{120}
`{382,879, 1530, 1830}
`{121, 1830}
`{1531 ,2008}
`
`{1723}
`
`{441 ,750,879,880, 1530,2018,2350,2495}
`{1721 '1830}
`{1531}
`{2025}
`{1531 '1830}
`
`{879,1531,1722}
`{2385}
`{1512}
`{1512}
`
`re dyscohesive and may lack much
`ytoplasm /161). However, an absence
`I lymphoglandular bodies in the back(cid:173)
`:round and the formation of loosely
`ohesive epithelial
`structures will
`·xclude the diagnosis of lymphoma.
`'ossibly more likely to occur is the misdi(cid:173)
`gnosis of a pancreatic endocrine
`.Jmour as a plasmacytoma, since pan(cid:173)
`reatic endocrine tumours may have a
`ery plasmacytoid appearance (505)
`eatures that will help to avoid this error
`re the presence of a salt-and-pepper
`hromatin pattern in the nuclei rather
`1an the clock-face chromatin pattern
`een in a plasmacytoma, greater vari(cid:173)
`:bility in nuclear size and shape, and an
`:bsence of a paranuclear halo in the
`)ancreatic endocrine tumour.
`
`Molecular genetic analysis
`iVhereas the molecular basis of familial
`Jancreatic endocrine tumours associat(cid:173)
`:d with multiple endocrine neoplasia
`ype 1 (MEN 1) and von Hippei-Lindau
`VHL) syndrome has recently been
`:stablished (351, 1241). little is known
`!bout the oncogenesis and the molecu ..
`ar basis of progression of sporadic
`umours.
`~ small number of published studies
`:ldicate that, in contrast to other human
`'Jmours, the activation of oncogenes is
`10t a common event
`in
`In
`•ndocrine tumours
`/903,980,1346).
`uarticular, the common genetic muta-
`
`tions identified in pancreatic ductal ade(cid:173)
`nocarcinomas
`(e.g., TP53, K-RAS,
`CDKN2A/p16, DPC4) are not found in
`pancreatic endocrine tumours (931 ).
`Molecular and cytogenetic analyses
`have identified a number of chromoso(cid:173)
`mal alterations in pancreatic endocrine
`tumours.
`
`Chromosomal imbalances
`Comparative genomic hybridization
`(CGH) studies of 102 pancreatic
`endocrine tumours revealed that chro(cid:173)
`mosomal losses occur slightly more fre(cid:173)
`quently than gains, while amplifications
`are uncommon (2105,2107,2154.2490).
`Furthermore, the total number of genom(cid:173)
`ic changes per tumour appears to be
`associated with both tumour volume and
`disease stage, indicating that genetic
`alterations accumulate during tumour
`progression (2105). Thus, large tumours
`or those with increased malignant poten(cid:173)
`tial, and especially metastases, harbour
`more genetic alterations than small and
`clinically ben1gn neoplasms (2105,2490).
`These findings point toward a tumour
`suppressor pathway and genomic insta(cid:173)
`bility as important mechanisms associat(cid:173)
`ed with tumour progression.
`In the majority of tumour types chromo(cid:173)
`somal alterations are not randomly dis(cid:173)
`tributed but are particularly common in
`certain chromosomal regions, including
`4pq (17%). 5q (25%), 7pq (41%). 9q
`(28%), 12q (23%). 14q
`17pq
`
`(31%) and 20q (27%) (gains) and 1p
`(21%), 3p (19%). 6q (28%). 10pq (14%),
`11q (30%), Y (31%) and X (31%) (loss(cid:173)
`es) Additional losses of 3p, 6pq, 10pq
`and gains of 5q, 12q, 18q and 20q are
`associated with malignant behaviour
`(2490).
`Losses of chromosome 1 and 11 q as well
`as gains of 9q appear to be early events
`in
`the development of pancreatic
`endocrine
`tumours, since
`they are
`already present in a substantial number
`of small (<2 em) tumours (2490). The
`other aforementioned alterations appear
`to occur later, accumulating during pro(cid:173)
`gression and are frequently associated
`with malignant biological behaviour.
`Prevalent chromosomal aberrations com(cid:173)
`mon in metastases include gains of both
`chromosomes 4 and 7, and losses of 21 q
`(2490), implying that these chromosomal
`
`80
`
`60
`
`60
`
`Fig. 4.07 Pancreatic endocrine tumours. Summary
`of the results obtained by CGH. Gains of chromoso(cid:173)
`mal material are prominent on chromosomes 4, 5, 7,
`9, 14, 17 and 20, while losses are concentrated on
`chromosomes 1, 3, 6 and 11.
`
`181
`
`Roxane Labs., Inc.
`Exhibit 1012
`Page 012
`
`
`
`A
`
`CGH
`
`---·· • .. . • I •
`... ~
`B •• ,..:.
`• .,
`
`4
`
`6 FISH
`
`9
`
`11
`
`9c-9q34
`
`9c-9q34
`
`6c-6q21
`
`3c-4p16
`
`Fig. 4.08 Pancreatic endocrine tumour. Analysis by comparative genomic hybridization(CGH) and fluores
`cence in-situ hybridization (FISH). Small functioning and non -functioning tumours of the pancreas: Gain 0•
`9q34 is an early event in insulinomas (green fluorescence in FISH; red: centromere of chromosome 9). The ~<'
`is a deletion of one copy of 6q21 (green fluorescence) in the presenc e of both centromeres of chromosom•
`6 (red). A duplication of 4p16 as shown by FISH (green) is prese nt as compared with the centromere 3 (red
`From: E.J. Speel et al. {2107}.
`
`• ·.
`
`··"
`t."-
`"It
`
`P149
`
`---
`
`.. - GCT·>lTT ; Ala->Ph(
`
`1 T ~ r C r 0 T '! T G 7 CA A C ~ C
`220
`
`. J·
`:'
`
`---
`
`PCR·SSCP
`
`Exon 2; ASOF
`
`Fig. 4.09 Sporadic pancreatic neuroendocrine tumour. Loss of one chromosome 11 (red), including th·
`MEN-1 locus (green) in the majority of tumour cell s. PCR-SSCP shows a band shift in exon 2 (red arroc
`heads) which is caused by a A50F missense mutation as shown by sequence analysis. From: B. Gortz et a
`{750}
`
`alterations than the other tumour types,
`(4) the frequency of MEN1 mutations in
`insulinomas is remarkably low, (5) func(cid:173)
`tioning tumours other than 1nsulinomas
`exhibit a higher frequency of MEN1
`mutations and associated LOH at 11 q 13
`than non-function ing tumours.
`These observations indicate that different
`types of pancreatic endocrine tumours
`
`evolve along different genetic pathway
`and that somatic inactivation of MEN1 i
`involved in a significant proportion (
`functioning tumours, but only exce ptior
`ally in insul inomas.
`
`imbalances may contribute to
`dissemination.
`
`tumour
`
`Loss of heterozygosity (LOH)
`When comparing th e resu lts of LOH
`studies using PCR microsatellite markers
`with those of CGH , similar chromosomal
`regions exhibit genetic losses. However.
`in general the rate of LOH is roughly
`twice that of allelic losses detected by
`CG H. At regions 3p23. 6q22. 9p , 11q1 3,
`18q2 1 and 22q12.1 the differences are
`even more pronounced , indicating that
`small deletions. not detectable by CGH,
`are involved 12385/. Only a small number
`of candidate genes located at some of
`the above mentioned chromosomal loci
`have been thoroughly investigated and
`many genes remain to be identified .
`Pooled data indicate that somatic MEN 1
`mutations are present in 21% (33/ 155) of
`spontaneous neoplasms and that 68%
`(75/ 111 ) harbo