Embed Size (px)
Citation preview
Clinicopathological Correlates of Pituitary Adenoma
by
Eric Monsalves
A thesis submitted in conformity with the requirements for the degree of Master
of Science
Institute of Medical Science
University of Toronto
© Copyright by Eric Monsalves 2015
Clincopathological correlates of pituitary adenoma
Eric Monsalves
Master of Science
Institute of Medical Science
University of Toronto
2015
Abstract Our current understanding of pituitary adenoma (PA) pathobiology is limited. Thus, the
purpose of this project is to gain a better understanding of PA biology through the
completion of two aims: 1) conducting a retrospective review of PA patients and 2)
establishing and characterizing an intracranial xenograft mouse model of PA and using
the model to identify novel targeted therapies.
With respect to our first aim, several tumor- and patient specific factors were correlated
to PA growth and recurrence. In our second aim, our tumor models were found to express
growth hormone (GH) and prolactin as well as proteins in the mTOR pathway. These PA
models were responsive to mTOR inhibition using RAD001 and resuted in a reduction in
tumor growth rate and the expression of mTOR and GH. A preclinical and clinical study
of PA provides a robust understanding of the clincopathologic factors contributing to PA
and may ultimately help improve treatment and outcome.
ii
Acknowledgements Thank you to my supervisor, Dr Gelareh Zadeh, for her patience and support during my
time as a masters student. In addition to Dr Zadeh, I would like to thank my committee
members: Drs Shereen Ezzat, Walter Kucharczyk, and Ozgur Mete whose mentorship
and guidance helped propel this project into reputable endocrine journals. Special thanks
to everyone in the Zadeh lab, especially to Dr Shahrzad Jalali who has played a critical
role in helping to ensure successful completion of this thesis. Finally, I would like to
thank Dr Toru Tateno from the Ezzat lab who has been an invaluable resource for
advising me on various technical aspects of this project.
iii
Contributions
Soroush Larjani (Zadeh lab) assisted with data collection in the first aim of this project
Shahrzad Jalali (Zadeh lab) assisted with the in vivo experiments
Amir Alamsehebpour (Zadeh lab) performed western blots on tumor lysates
Toru Tateno (Ezzat lab) kindly provided the GH4 cells used for in vivo experiments
iv
Table of Contents Abstract …………………………………………………………….................................ii
Acknowledgements ……………………………………………......................................iii
Contributions………………………………………………………................................iv
Table of Contents …………………………………………………..................................v
Abbreviations …………………………………………………….................................viii
List of Tables ………………………………………………………...............................xii
List of Figures ………………………………………………………............................xiii
List of Appendices …………………………………………………..............................xiv
Chapter 1 Introduction ……………………………………………................................1
1.1 Overview of PA and its Clinicopathologic Correlates ……….................................1
1.1.1 Pituitary Gland …………………………………………...............................1
1.1.2 Pituitary Adenoma ……………………………………….............................3
1.1.3 Growth Pattern of PA …………………………………….............................6
1.1.4 Cavernous Sinus Invasion ………………………………..............................7
1.1.5 PA Residual and Recurrence ……………………………...........................10
1.1.6 PA Growth Rate and Tumor Volume Doubling Time
(TVDT)………………………………………………………….........................10
1.1.7 The Ki-67/MIB-1 Labeling Index: A Measure of Cell
Proliferation……………………………………………………...........................12
1.2 The PI3K/Akt/mTOR Pathway …………………………………...............................13
1.2.1 The PI3K/Akt/mTOR Pathway …………………………............................13
1.2.2 The PI3K/Akt/mTOR Pathway in Endocrine and Non-Endocrine
Neoplasia ………………………………………………………...........................15
1.2.3 The Importance of Establishing Novel Therapeutic Targets in PA..............17
1.2.4 Analysis of the RTK/PI3K/Akt/mTOR Pathway in PA ……......................18
1.2.5 Current mTOR Inhbitors ……………………………………......................22
1.2.6 Effects of mTOR Inhibition on PA Cell Proliferation and
Viability ………………………………………………………... …...................23
1.2.7 Effects of Dual PI3K/mTOR Inhibition on PA Cell Proliferation and
Viability ……………………………………………………………....................27
v
1.2.8 Effects of mTOR Inhibition on Apoptosis in PA cells
in vitro……………………………………………………………........................29
1.2.9 Effects of PI3K and mTOR inhibition on PA cell cycle ………..................30
1.2.10 mTOR Inhibition and Radiation on PA cells ………………….................31
1.2.11 mTOR Inhibition Using Xenograft in vivo Models of PA ........................32
Chapter 2 Outline and Overview of Aims …………………………………................34 2.1 Purpose …………………………………………………………………...................34
2.2 Hypthesis#1.................................................................................................................34
2.2.1 Aim#1 ……………………………………………………………………..............34
2.2.2 Rationale for Aim#1 ……………………………………………………............... 34
2.3 Hypothesis#2................................................................................................................36
2.3.1 Aim#2 ……………………………………………………………………..............36
2.3.2 Rationale for Aim#2 ……………………………………………………................36
Chapter 3 Growth Patterns of PA and Histopathological Correlates ………...........38 3.1 Introduction.............................. …………………………………………….............38
3.2 Materials and Methods……………………………………………………..............39
3.2.1 Patient Population ………………………………………………...............39
3.2.2 Magnetic Resonance Imaging (MRI) ……………………………….........39
3.2.3 Growth Pattern and MRI Characteristics…………………………….........41
3.2.4 Tumor Volumetry and Growth Rate ………………………………….......46
3.2.5 Correlating Tumor Growth Rate to Histopathology of PA ………….........46
3.2.6 Statistical analysis …………………………………………………….......47
3.3 Results…………………………………..................................................................47
3.3.1 Clinical Demographics and Radiological Characteristics ……………......47
3.3.2 Preoperative PA Volume and Growth Rate (TVDT) …………………......48
3.3.3 Association of PA Residual with Growth Patterns and Patient
Characteristics …………………………………………………………………..50
3.3.4 Postoperative Tumor Volume and Growth Rate (TVDT) ………………..51
3.3.5 Correlation of TVDT with Expression of MIB-1, FGFR4, and p27 ……..52
3.4 Discussion …………………………………………………………………………..54
Chapter 4 Preclinical Model of PA For Interrogating Therapeutic Strategies: Role
of mTOR Inhibitors …………………………………………………………………....56 4.1 Introduction ………………………………………………………...........................56
vi
4.2 Materials and Methods …………………………………………………………….58
4.2.1 Cell lines and culture ………………………………………………..........58
4.2.2 Plasmids and Transfections ……………………………………………....58
4.2.3 Animals …………………………………………………………………...58
4.2.4 In vivo Intracranial Tumor Models …………………………………….....59
4.2.5 Magnetic Resonance Imaging (MRI) ………………………………….....59
4.2.6 Treatment Schedule ……………………………………………………....60
4.2.7 Tumor Size Measurement ………………………………………………...60
4.2.8 Immunohistochemistry …………………………………………………...61
4.2.9 Western Blot Analysis …………………………………………………....61
4.2.10 ELISA …………………………………………………………………...62
4.2.11 Statistical Analysis ………………………………………………………62
4.3 Results ………………………………………………………………………………62
4.3.1 Intracranial Tumor Formation Assay ……………………………………..62
4.3.2 MRI analysis ……………………………………………………………...63
4.3.3 Tumor Volume and Growth Rate ……………………………………….. 64
4.3.4 mTOR Pathway expression ……………………………………………....65
4.3.5 mTOR inhibition treatment using RAD001 ……………………………....66
4.3.6 The Growth Rate of GH4 tumors ………………………………………...66
4.3.7 mTOR Expression Patterns Following RAD001 Treatment ……………..68
4.3.8 Hormone Expression Patterns Following RAD001 Treatment …………..70
4.3.9 Mouse Serum IGF-1 Levels Following RAD001 Treatment …………….71
4.4 Discussion …………………………………………………………………………..73
Chapter 5: Discussion ………………………………………………………………….75 5.1 Future Directions: Clinical Applications for mTOR inhibitors of PA................ 84
5.2 Conclusions …………………………………………………………………………85
5.3 Limitations ………………………………………………………………………….85
5.4 Future directions …………………………………………………………………...87
References…………………………………………………………………………….....88
Appendix 1……………………………………………………………………………..108
vii
Abbreviations
4EBP1 Eukaryotic translation initiation factor 4E-binding protein 1
ABC Avidin Biotin Complex
ACTH Adrenocorticotropic Hormone
ADH Antidiuretic Hormone
CS Cavernous Sinus
CSI Cavernous Sinus Invasion
CT Computed tomography
DAB 3,3’ - Diaminobenzidine
DMSO Dimethyl Sulfoxide
FSH Follicule-Stimulating Hormone
EGFR Epidermal Growth Factor Receptor
eIF4E Eukaryotic Translation Initiation Factor 4E
ELISA Enzyme-Linked Immunosorbent Assay
ER Estrogen Receptor
viii
FBS Fetal Bovine Serum
FGFR4 Fibroblast Growth Factor Receptor 4
GH Growth Hormone
HS Horse Serum
IHC Immunohistochemistry
ICA Intracavernous carotid artery
IGF-1 Insulin Growth Factor-1
IRS-1 or 2 Insulin Receptor Substrate 1, -2
IP Intraperitoneal
ITK-SNAP Insight Toolkit Snake Automatic Partioning
mTOR mechanistic Target of Rapamycin
MRI Magnetic Resonance Image
NFPA Non-functioning pituitary adenoma’
NOD-SCID non-obesessive severe combined immunodeficiency
NP40 Noniodet P40
ix
PA Pituitary adenoma
PBS Phosphate Buffered Saline
PDK1 Phosphoinositide dependent kinase 1
PI3K Phosphoinositide-3-kinase
PIP2 Phosphatidylinositol 4,5-bisphosphate
Pit-1 Pituitary-specific positive transcription factor-1
PRL Prolactin
Ptd-FGFR4 pituitary tumor derived – Fibroblast Growth Factor Receptor 4
PTEN Phosphotase and Tensin homolog
RTK Receptor Tyrosine Kinase
LH Luteinizing Hormone
SDS Sodium dodecyl sulfate
SF-1 Steroidogenic Factor 1
SNP Single Nucleotide Polymorphism
TEF Thyrotroph Embryonic Factor
TMZ Temozolomide
x
Tpit T-box Transcription Factor
TSC1,2 Tuberous Sclerosis Complex
TSH Thyroid-Stimulating Hormone
TVDT Tumor Volume Doubling Time
xi
List of Tables
Table 1.2.1: Examining elements of the PI3k/Akt/mTOR in models of PA.....................20
Table 1.2.2: In vitro studies examining effect of mTOR inhibitors on PA cells..............25
Table 3.3.1: Pituitary adenoma histological subtypes ………………………………......48
Table 3.3.2: Characteristic growth patterns in pituitary adenoma ………………….......48
Table 3.3.3: Factors associated with a shorter preoperative TVDT ……………….........50
Table 3.3.4: Factors associated with residual volume following surgery …………........51
xii
List of Figures Figure 1.1.1: Pituitary cell lineage ……………………………………………….............3
Figure 1.2.1: The PI3K/Akt/mTOR pathway ……………………………………...........15
Figure 3.2.1: Screenshot of ITK SNAP …………………………………………...........41
Figure 3.2.2: Diagram of pituitary fossa depicting growth of PA into CS ………..........42
Figure 3.2.3: Recorded PA growth patterns ……………………………………............43
Figure 3.2.4: UHN classification of cavernous sinus invasion …………………............45
Figure 3.3.1: Growth rates (TVDT) of different PA histological subtypes …….............49
Figure3.3.2 Schematic summary for Aim #1....................................................................53
Figure 4.3.1: (GH), (PRL), and Ki-67 expression in tumor model ……………..............63
Figure 4.3.2: Tumor development in the pc, G388, and R388 mice ……………............64
Figure 4.3.3: tumor volume and growth rate in pc, G388, and R388 tumors …..............65
Figure 4.3.4: Immunohistochemical staining showing p-S6, p-4EBP1 expression in pc,
G388, and R388 GH4 tumors ……………………………………………………...........66
Figure 4.3.5: Growth rate of pc, G388, R388 tumors following
RAD001 treatment ………………………………………………………………...........67
Figure 4.3.6: A) Immunohistochemical and western blot analysis of p-s6 expression
following RAD001 treatment in pc, G388 and R388………………………………........69
Figure 4.3.7: A) Immunohistochemical and western blot analysis of p-4EBP1 expression
following RAD001 treatment in pc, G388 and R388………………………………........70
Figure 4.3.8: A) Immunohistochemical and western blot analysis for GH expression
following RAD001 treatment in pc, G388 and R388 ……………………………...........71
Figure 4.3.9: Circulating IGF-1 levels in pc, G388 and R388 mice. ..............................72
Figure4.3.10: Schematic summary of Aim #2..................................................................73
xiii
xiv
List of Appendices Appendix#1 Human Pituitary Adenoma Primary Culture Protocol
xv
Chapter 1
INTRODUCTION
1.1 Overview of PA and its Clinicopathologic Correlates The pituitary gland is a small endocrine gland that is located at the base of the brain and consists
of the anterior lobe and posterior lobe. Proliferations of hormone secreting epithelial cells of the
anterior lobe can sometimes lead to neoplastic transformation giving rise to pituitary adenoma
(PA). Although generally a benign neoplasm, PA can become locally invasive and infiltrate
sensitive structures surrounding the pituitary fossa such as the optic chiasm and cavernous sinus.
Furthermore, particularly aggressive PA can recur following surgery. The following subsections
provide a description of pituitary anatomy and an overview of PA and how it’s growth and
recurrence has been associated with various tumor- and patient specific characteristics.
1.1.1 Pituitary Gland
The pituitary is a small endocrine gland that sits at the base of brain in an area of the sphenoid
bone referred to as the sella turcica. The pituitary gland is surrounded superiorly by the optic
chiasm and hypothalamus, laterally on either side by the cavernous sinus, and inferiorly by the
sellar bone. It is encased in fibrous material referred to as the capsule. Outside of the capsule,
the pituitary is surrounded by dura which attaches to the inferior aspect of the diaphragma sella.
The pituitary is the primary endocrine gland for the body as it regulates many of the hormones
secreted from other endocrine glands. In turn, the hormonal activity of the pituitary gland is
regulated by hypothalamic hormones, and negative feedback mechanisms from hormones
released by the pituitary’s target organs. The pituitary can also regulate it’s own activity through
various autocrine and paracrine functions(Bilezikjian, et al. 2004).
The pituitary is composed of two general areas referred to as the neurohypophysis and
adenohypophysis. A third area called the pars intermedia is rudimentary in humans but plays a
more prominent role in rodents. The neurohypophysis consists of the infundibulum, pituitary
1
stalk, and pars nervosa and is essentially composed of axonal projections from the hypothalamus.
Consequently, hormones that are stored and secreted from the neurohypophysis, including
oxytocin and antidiuretic hormone (ADH), are produced in the hypothalamus. The
adenohyphysis is composed of epithelial tissue and consists of the pars distalis, which is the
largest component of the adenohypophysis, and the pars tuberalis which extends from the
adenohypophysis towards the pituitary stalk. The intercommunication between the
hypothalamus and adenohypophysis is achieved via a portal vein network(Asa and Ezzat 2002).
The adenohypophysis, specifically the pars distalis, is composed of acini that contain six
hormone-secreting cells within a reticulin rich stroma(Asa and Ezzat 2009). All of these cells are
regulated by hypothalamic releasing or inhibiting factors. The specific cells and the hormones
that they secrete are as follows: corticotroph cells secrete adenocorticotropic hormone (ACTH);
the somatotrophs secrete growth hormone (GH), lactotrophs secrete prolactin (PRL),
mammosomatotrophs secrete both GH and PRL, thyrotrophs secrete thyroid stimulating
hormone (TSH), and gonadotrophs secrete follicle-stimulating hormone/Luteinizing Hormone
(FSH/LH). In addition, the adenohypophysis contains the non-endocrine folliculostellate cells
which are sustenacular cells that release cytokines and growth factors important for the
maintenance of normal pituitary function((Marin, et al. 1991)).
The adenohypophysis is derived from the oral ectoderm to form Rathke’s pouch. A single
progenitor cell with the aid of specific transcription factors form the six cells of the
adenohypophysis. Essentially, each of the cells is derived from one of three lineages which
include pituitary specific positive transcription factor 1(Pit-1), the T-box transcription factor (T-
pit), and steroidogenic factor 1(SF-1)(figure1.1). The somatotrophs, lactotrophs,
mammosomatotrophs, and thyrotrophs are derived from precursors expressing Pit-1. Pit-1 is
required for GH gene expression and cooperation with ERα allows mammosomatotrophs to
develop. A GH silencing mechanism aids in lactotroph differentiation. Coexpression of
thytroph embryonic factor (TEF) and GATA-2 is needed to initiate TSH production.
Corticotroph development is dependent on T-pit. Stereodogenic factor-1 (SF-1) and GATA-2
are required for gonadotroph differentiation(Asa & Ezzat, 2009) (figure 1.1.1). In addition, there
are various transdifferentiation mechanisms that can convert one cell type of Pit1 lineage to
2
another depending, in part, on certain hormonal requirements by the body at certain times such
as pregnancy or puberty, thus a level of fluidity is maintained among Pit1 cells (somatotrophs,
lactotrophs, and mammosomatrophs). Identification of the particular lineage from which a cell is
derived is an important pathological criterion for diagnosing a specific tumor of
adenohypophyseal origin.
Figure 1.1.1: Pituitary cell lineage as adapted from Asa and Ezzat, 2009
1.1.2 Pituitary Adenomas
Pituitary tumors or adenomas (PA) are benign neuroendocrine proliferations or over growths of
adenohypophyseal differentiation. They are a common intracranial neoplasm, evident in up to
20% of the general population(Asa & Ezzat, 2009), though only a smaller percentage require
medical treatment. PA are generally derived from one of the six hormone secreting cells of the
adenohypophysis and are thought to originate from a single transformed cell and are thus
classified as monoclonal lesions((Alexander, et al. 1990). They are characterized based on the
expression of pituitary-specific transcription factors, hormones, and overall PA cell
structure(Mete & Asa, 2012). Examination of the density of hormone-secreting granules within
the PA cells leads to the designation of “densely” and “sparsely” granulated PA. A detailed
Oral ectoderm
Rathke’s Pouch Stem cell
Corticotroph
Gonadotroph
Somatotroph Stem cell
Pit-1
Somatotroph
Mammo- somatotroph
ERα
GH-Repressor
Thyrotroph
Lactotroph
ERα
ERα
GH-Repressor
3
pathological examination of PA can lead to the identification of many different histological
subtypes.
PAs are further classified based on their biochemical activity. Clinically functioning PA are
tumors that oversecrete hormone and cause corresponding hormonal syndromes within the body.
PRL-secreting PA are the most common functioning PAs accounting for up to 30% of all PA.
Due to PRL overexpression, patients harboring a prolactinoma suffer from amenorrhea,
infertility and galactorrhea in females, infertility or impotence in males. GH-secreting PAs are
often associated with acromegaly. Clinical presentation of acromegaly is characterized by acral
and facial changes, hyperhidrosis (abnormally increased perspiration), headaches, sexual
dysfunction, and soft tissue enlargement. ACTH hypersecretion and adrenal overstimulation lead
to hypercortisolism, which is responsible for the symptoms of Cushing’s disease. Cushing’s
disease gives rise to fat accumulation in the face resulting in moon face, as well as buffalo hump,
and truncal obesity. Additionally, the skin gets thinner caused by the loss of the epidermal layer
and the tension over the accumulated fat produces purple striae. The development of psychiatric
disorders such as depression and psychosis is also associated with Cushing’s disease(Bertagna,
et al. 2009). TSH-secreting PAs are rare and present with a mild increase in thyroxin levels with
abnormal TSH levels. Symptoms associated with gonadotroph PA that secrete FSH and/or LH
may manifest in sexual dysfunction and hypogonadism. Nevertheless, gonadtroph PAs that
secrete FSH/LH are very rare and this PA subtype is not usually characterized by hormonal
production(Lake, et al. 2013).
On the other hand, clinically non-functioning PA (NFPA) are generally referred to as “silent”
and do not cause hormonal abnormalities. Of the non-functioing subtype, gonadotrophs are the
most common, comprising approximately 85% of all NFPA((Greenman and Stern 2009)). Other
NFPA include the “silent” counterparts to the corresponding functional variant, and null cell
adenomas which are of adenohypophysial origin but are negative for any pituitary transcription
factor or hormone((Mete and Asa 2012)). Presenting symptoms from the NFPA subtypes are
mainly due to mass effect such as headache or visual disturbance. These tumors may also lead to
varying degrees of hypopituitarism depending on the amount of adenohypophyseal damage due
to the PA((Lindholm, et al. 2006)). The distinction between clinically functioning and
4
nonfunctioning PA is based on a biochemical assessment rather than a histopathological
examination (Mete and Asa 2012).
PAs are also classified radiologically, through examination of computed tomography (CT) or
magnetic resonance image (MRI) which provides information on the size of the PA as well as on
the degree of local invasion((Bonneville, et al. 2005)). MRI provides good resolution of the
sellar area while CT is good at depicting tumor that invades into bone. Generally, PA visualized
on MRI are categorized based on size with microadenomas referring to PA that are less than 1cm
in diameter while macroadenomas are PA that are greater than 1 cm in diameter (Asa and Ezzat
1998). The designation of ‘giant’ adenoma is sometimes reserved for PA that are
≥3cm(Juraschka, et al. 2014) As noted earlier, the sellar fossa is adjacent to several areas
including the optic chiasm, and cavernous sinus. PA that extend to or even invade these areas,
typically the larger macroadenomas, make treatment more difficult and are often associated with
a poorer outcome ((Rey-Dios, et al. 2014; Woodworth, et al. 2014)).
There are several treatment options available for patients with PA. These include medical
therapy, surgery, radiotherapy, and/or gamma knife radiosurgery. There are currently a limited
number of medications that are taken to reduce hormone secretion and/or tumor size. These
include the dopamine agonists such as bromocriptine used to treat PRL-secreting
prolactinomas(Price and Bridges 2014). Dopamine agonists target the D2 dopamine receptor to
restore tonic PRL inhibition. Another medication is the somatostatin analogue, such as long
acting octreotide, which is used to treat acromegalic patients (Zhang, et al. 2014). However,
Octreotide is only effective in particular subtypes of GH PA. The vast majority of PA do not
respond to medication. Endonasal microscopic or endoscopic transsphenoidal surgery is the
primary treatment for most PA with gross total removal accomplished in most
instances(Dallapiazza, et al. 2014).However, the more aggressive or invasive the PA, the lower
the probability for complete surgical removal. Radiotherapy and gamma knife radiosurgery are
generally reserved for recurrent PA that are not amenable to reoperation. Radiation-induced side
effects from these treatment options include visual comprise, cranial neuropathies, or cognitive
decline(Brada et al., 1993; Hahn et al., 2009).
5
Classifying a PA based on its radiological, biochemical, and histopathological features yields the
most accurate portrait of a particular PA and the complete profile of the tumor should be
incorporated into patient management. Thus, a multidisciplinary team of endocrinologists,
neurosurgeons, radiologists, and pathologists should be assembled for the treatment of PA.
1.1.3 Growth Patterns of PA
Although generally a benign brain tumor, PAs have a tendency to compress or invade structures
adjacent to the pituitary fossa. The invasive potential of PA has been correlated to tumor size
with larger macroadenomas typically engaging in more locally invasive behavior(Scheithauer,
Kovacs, Laws, & Randall, 1986). Furthermore, invasion has been thought to be associated with
histological subtype. Such aggressive macroadenomas include sparsely granulated
somatotrophs, densely granulated lactotroph adenoma, sparsely granulated corticotrophs,
thyrotroph adenoma, acidophil stem cell adenoma (PRL- and GH positive PA of Pit-1 lineage),
and crooke cell adenoma. (ACTH-positive PA of T-pit lineage)((Mete and Asa 2012) ). The
factors that are involved in aiding different growth and invasion patterns in PA need to be
clarified.
One of the most common extrasellar growth patterns exhibited by PA is superior growth towards
the suprasellar cistern and optic chiasm. It is estimated that 80% of macroadenomas present with
this pattern(Ramakrishnan et al., 2013; Zada, Lin, & Laws, 2010). It has been reported that it is
the NFPA subtype in particular that has a proclivity for suprasellar growth(Zada et al., 2010). Of
the NFPA, the silent corticotroph adenomas are more commonly associated with a larger,
irregular-shaped, lobulated suprasellar component compared to null cell/gonadotrophs, indicative
of more aggressive behavior(Nishioka, Inoshita, Sano, Fukuhara, & Yamada, 2012).
Preferential inferior growth is also evident in different histological subtypes of PA. Lundin et al.,
analysed 115 PA that included GH-, PRL- and NFPA and reported that prolactinomas are most
associated with invasion into the base of skull, and associated with a lower frequency of
suprasellar tumor extension (Lundin, Nyman, Burman, Lundberg, & Muhr, 1992). Other groups
have noted that GH-secreting adenomas frequently invade the sphenoid sinus and
6
clivus(Hagiwara et al., 2003; Marro et al., 1997; Zada et al., 2010). Silent GH-PRL, subtype III
adenomas and acidophil stem cell adenomas also possess characteristic downward growth with
significant bone invasion rather than presenting with the more common suprasellar
expansion(Mete & Asa, 2012)
Parasellar growth into the cavernous sinus (CS) represents a surgical challenge due to the
sensitive carotid artery, veins and cranial nerves located within this area. Cavernous sinus
invasion (CSI) has been reported to occur in up to 20% of cases and is more commonly observed
in clinically NFPA rather than functioning PA((Tanaka, et al. 2003)). Within the nonfunctioning
group specifically, silent corticotroph adenomas have been reported to invade the CS more
frequently than the null cell/gonadotroph subtypes and other hormone inactive subtypes
(Jahangiri et al., 2013; Nishioka et al., 2012). An examination of CSI and its extent is discussed
below.
1.1.4 Cavernous Sinus Invasion
The CS space is a venous plexus located on either side of the pituitary fossa. Each side is
outlined by four dural walls: the roof, medial, lateral and posterior walls. The CS contains the
intracavernous segment of the carotid artery (ICA) as well as the oculomotor, trochlear,
opthalamic, and maxillary cranial nerve located within the lateral wall of the CS. The abducens
nerve also runs through the middle of the CS alongside the ICA.
The infiltration of the CS space usually occurs unilaterally and is observed more frequently in
macroadenomas but can also occur in microadenomas(Cottier et al., 2000; Knosp, Steiner, Kitz,
& Matula, 1993; Vieira, Cukiert, & Liberman, 2006). CSI often precludes gross total resection of
tumor and increases intra- and postoperative complications((Sepehrnia, et al. 1991)). Therefore,
the detection of CSI preoperatively is of critical clinical importance.
One of the ways preoperative diagnosis of CSI is accomplished is by taking into account the
boundaries of the CS space. The CS space is surrounded by dura mater and the medial dural wall
that separates the CS from the pituitary fossa is an important determinant of CSI. It is
7
perforation of this medial wall by PA that determines invasion as opposed to compression of the
CS. However, localizing the medial wall on MRI is difficult and is usually only defined upon
direct surgical observation(Ceylan, Anik, & Koc, 2011). Nevertheless, there are a few studies
that were purportedly able to visualize PA induced disruption of the medial wall utilizing either
T2-weighted (T2W) images(Sol et al., 2012) or proton-density weighted images(Cao et al.,
2013). Aside from examining medial wall defects, the size and symmetry of the CS as well as
bulging of its lateral wall due to PA are also sometimes taken into account when determining
CSI(Cottier et al., 2000).
Our in-house method commonly utilized by UHN neuroradiologists uses the detection of lateral
displacement of the lateral wall of the CS as one of the criteria for CSI (See chapter 3.1.2).
The venous plexus in the CS space is also a helpful feature for determining CSI. The venous
content of the CS is divided into several compartments relative to the ICA. There is the medial
compartment which is between the ICA and the pituitary fossa, the superior compartment which
is above the ICA, the lateral compartment which is lateral to the ICA, and the inferior
compartment which is below the ICA. Cottier et al., further divided the inferior compartment
into the carotid sulcus compartment (between ICA and carotid sulcus of sphenoid bone) and the
inferolateral compartment which is between the carotid sulcus and the lateral
compartment(Cottier et al., 2000). The configuration of these compartments depend mainly on
the configuration of the carotid siphon and its position within the cavernous sinus space (Knosp,
et al. 1993)In the majority of cases, the superior and inferior compartments are the largest
(Knosp et al., 1993). Knosp reported that the superior compartment was most often affected
when PA entered the CS space(Knosp et al., 1993). However, Cottier et al., reported that PA
growth below the ICA was more common in PA that invaded the CS(Cottier et al., 2000). Also,
obliteration of the carotid sulcus venous compartment was observed most often when the CS was
invaded (Bonneville et al., 1989; Cottier et al., 2000; Vieira et al., 2006). Vieria also noted that
when the lateral venous compartment was not depicted the probability of CSI was high(Vieira et
al., 2006).
The ICA is very easily observed on MRI due it’s void of high velocity blood flow and as such it
is a critical radiological landmark for determining CSI. The diagnosis of CSI is typically
8
reserved for cases in which the ICA is completely encased by tumor, but this occurs late
((Moreau, et al. 1998)). Establishing criteria that would predict CSI before the PA encases the
ICA is of clinical importance because this may facilitate a greater probability of gross total
removal of PA and improve patient outcome. Knosp et al., assessed the extent of PA
involvement into the CS by dividing the CS into a series of lines drawn between the
supracavernous and intracavernous segments of the carotid artery. There is the medial
intercarotid line, the median intercarotid line ((Knosp et al. 1993)), as well as the lateral
intercarotid line and PA that crosses each successive line from medial to lateral represents a
greater degree of CS involvement and a greater probability of CSI((Knosp et al. 1993)). Most
studies suggest that when PA crosses the medial or lateral intercarotid line the probability of CSI
is high (Cottier et al., 2000; Moreau et al., 1998; Sol et al., 2012; Vieira et al., 2006). A number
of studies have also attempted to establish a threshold value for carotid artery encasement that
would predict CSI. Cottier reports that when 67% of ICA is encased by tumor, CSI is
likely((Cottier, et al. 2000)), Vieria reports a high probability of CSI when at least 45% of the
ICA is encased by PA((Vieira, et al. 2006)). Sol reports a threshold value of 135.5 degrees of
ICA encasement in order to make an accurate diagnosis of CSI((Sol, et al. 2012)).
The UHN neuroradiology department defines CSI when there is greater than 50% encirclement
of the ICA (See chapter 3.1.2).
Determining when the CS is not invaded is also useful clinical parameter. It is suggested that
when there is normal pituitary interposed between the PA, no tumor involvement beyond the
medial intercarotid line, and/or when there is a low percentage of carotid artery encasement, the
CS is not invaded. Several studies have noted that when 25% or less of carotid artery is encased,
there is no CSI(Connor, Wilson, & Hogarth, 2014; Cottier et al., 2000; Vieira et al., 2006).
Depiction of the medial and/or superior compartments is also helpful for excluding the
possibility of CSI although their absence does not necessarily indicate CSI. Conner noted that
when the lateral or inferolateral compartments were visible or for every 25% reduction in ICA
encasement, the probability for complete resection increased greatly.
Assessing PA invasion into the CS and its extent should be conducted by considering the
features of the CS itself, its venous compartments, and the ICA. Determination of CSI on MRI is
9
clinically relevant and aids in determining appropriate treatment strategies.
1.1.5 PA Residual and Recurrence
Despite gross total resection in a majority of cases, some PA tend to recur following surgery.
Tumor recurrence is a clinically relevant scenario observed in in up to 46% of
cases((Cappabianca, et al. 2000; Losa, et al. 2008)). Most PA tend to recur between 1-5 years
after surgery (Roelfsema, Biermasz, & Pereira, 2012). In a recent meta-analysis conducted by
Chen et al., the tumor-growth free survival (TGFSR) at 5 and 10 years was 71% and 59%,
respectively. Furthermore, they noted that the percentage of PA recurrence was higher and
TGSFR was lower in cases with residual following surgery compared to cases without
residual(Chen et al., 2012)
The postoperative outcome of PA has been shown to relate to histological subtype. It has been
reported that prolactinomas have the highest incidence of recurrence while NFPA have the
lowest rate of remission(Roelfsema et al., 2012). The prognostic value of invasion in determining
PA recurrence has been reported to be highest for PRL and ACTH tumors and lower for FSH/LH
and GH tumors(Trouillas et al., 2013). Tumor size was also shown to correlate with long-term
disease-free outcome but not with recurrence/progression-free status. According to a recent
systematic review by Roelfsema., the postoperative basal (non-stimulated) hormone levels are
the only predictor of tumor recurrence in functioning PA while no single factor is of prognostic
value for recurrence in NFPA(Roelfsema et al., 2012).
1.1.6 PA Growth Rate and Tumor Volume Doubling Time
(TVDT)
As mentioned above, PA have the potential to become locally invasive or recur following
surgery. This complicates patient management and necessitates the quantification of PA growth
as a means to predict outcome. It has been proposed that monitoring pituitary macroadenoma
10
growth using 3D volumetric assessment of MRI is more accurate than conventional 2D
interpretation of MRI(Ringstad et al., 2012).
An examination of PA volume has been shown to correlate to surgical outcome. Shrestha et al
determined that PA with a preoperative volume greater than 8cm3 were associated with a greater
risk of postoperative residual(Shrestha, Qi, Bao, & Wang, 2012). In another study, a greater risk
of postoperative residual was associated with a preoperative PA volume of over 5mL(Jain,
Gupta, Pathak, Bhansali, & Bapuraj, 2008).
It has been reported through mathematical modelling that the majority of PA grow
exponentially(Honegger et al., 2003; Honegger et al., 2008). Exponential tumor growth is
associated with a constant tumor growth rate. If growth rate is constant, PA growth velocity can
be described using a tumor volume doubling time (TVDT), which is the time needed for a tumor
to double in size.
The TVDT has been employed as a means to estimate tumor growth in a number of other solid
tumors including other benign intracranial neoplasms such as meningioma ((Nakaguchi, et al.
1999))
To date, with respect to PA, the TVDT has been reported in only a few studies, primarily for
residual NFPA. The reported growth rates of PA exhibit a broad range: the TVDT range in one
study was 200-2550 (980 days)
((Ekramullah, Saitoh, Arita, Ohnishi, & Hayakawa, 1996)and 506-537 (mean 1836 days) in a
another study(Tanaka et al., 2003)Residual NFPA growth rate has been correlated to patient age
with elderly patients exhibiting a slower tumor growth rate than younger patients.
Hsu reported the TVDT in a cohort consisting of residual functioning and nonfunctioning PA
and reported a TVDT range of 3.7 months to 879.4 months (mean 78.7 months)(Hsu, Guo,
Chien, & Ho, 2010).
It is currently unknown whether the preoperative pattern of PA growth in the context of its
11
growth rate has any predictive value in determining postoperative tumor growth rate. However,
currently, a commonly used molecular marker to predict PA growth potential is the examination
of the Ki67/MIB-1 labeling profile.
1.1.7 The Ki-67/MIB-1 Labeling Index: A Measure of Cell
Proliferation
The Ki-67/MIB-1 index is commonly used to assess the proliferative activity of tumor cells. The
nuclear protein Ki-67 is recognized by the MIB-1 monoclonal antibody and is expressed
throughout the cell cycle in G1, G2, and M phases but not G0. The percentage of cells
immunostained for the Ki-67 protein represents a marker of proliferation in many
cancers((Urruticoechea, et al. 2005)). However, despite its widespread use in PA pathological
examinations, its usefulness as predictor of PA growth behavior remains unclear(Chacko et al.,
2010; Madsen et al., 2011; Prevedello, Jagannathan, Jane, Lopes, & Laws, 2005).
Currently, there are no reliable molecular markers used to predict the growth behavior of PA.
However, the examination of the Ki-67/MiB-1 is frequently used in the pathological assessment
of PA to gain insight into the growth rate of PA. The Ki-67 has been studied in relation to PA
invasion. A study by Chacko et al., defined the biological significance of the MIB LI in 159
surgically excised PA and found that MIB1 correlated with PA invading the CS. Furthermore,
bilateral CS invasion had a significantly higher MIB1 than unilateral extension(Chacko et al.,
2010). Similarly, Pan et al., 2005 determined that expression of Ki-67 in PA has associations
with PA growth into the CS(Pan, Chen, Liu, & Zhao, 2005).
The Ki-67 may also indicate possible PA residual/recurrence. A study by Wildhalm et al., has
determined that in NFPA, residual tumors that progressed had a significantly higher MIB-1
relative to the residuals that did not progress(Widhalm et al., 2009). In another study that
consisted of GH-secreting PA, Fusco et al determined that Ki-67 was significantly lower in
tumors that were cured after surgery versus those that were not (Fusco et al., 2008).
12
The relationship between Ki-67 and the growth rate of recurrent tumor has also been examined
in a few studies. The Ki-67 LI has been shown to be inversely correlated to TVDT, where lower
TVDT were associated with higher MIB-1 (Ekramullah et al., 1996).
The impact that the Ki-67 marker has on PA invasiveness and recurrence remains controversial
with conflicting reports in the literature. Nevertheless, this marker provides a general estimate of
the proliferative potential of PA and is routinely included in PA pathology assessments.
Although PA generally possess a relatively low proliferative index in comparison to other
cancers, they still have the capacity for aggressive growth. This has prompted a number of
studies to examine molecular signaling pathways that may be involved in cell growth and other
proproliferative activities. Recently, a candidate pathway that is frequently upregulated in
various endocrine and non-endocrine cancers, including PA, is the PI3K/Akt/mTOR pathway.
This particular pathway and its relation to PA is the focus of the next section.
1.2 Overview of PI3K/Akt/mTOR Pathway in Relation to PA This section was published in Endocine Related Cancer (Monsalves et al., 2014)
The PI3K/Akt/mTOR pathway is a signal transduction cascade involved in cell growth and
metabolism. It is ubiquitously upregulated in various endocrine and non-endocrine cancers and
has recently been implicated in PA. Several studies have shown that this pathway is involved in
human and animal PA cells in vitro and that these PA cells are responsive to mTOR inhibition
manifesting in an overall PA cell reduction.
1.2.1 The PI3K/Akt/mTOR Pathway The PI3k/Akt/mTOR pathway is a signal transduction cascade involved in cell growth and
metabolism(Leslie, Biondi, & Alessi, 2001). The pathway is activated by upstream receptor
tyrosine kinases (RTK) which feed the signal through the phophotidylinositide-3-kinase (PI3K)
complex. PI3K is a lipid kinase consisting of a regulatory (p85) and catalytic (p110) subunit. It
is activated directly via the p85 subunit which interacts with phosphotytrosine residues on the
RTK. Alternatively, in the indirect route, PI3K interacts with the RTK via the IRS-1 or 2 adaptor
proteins(White, 1998). Both methods of activation lead to the conversion of PIP2 to the second
13
messenger PIP3. PIP3 recruits the kinases PDK1 and Akt to the plasma membrane. Akt is then
phosphorylated by PDK1 and mTORC2, on its threonine and serine residues respectively. These
phosphorylation events lead to the activation of Akt (figure 1.2.1).
Akt is the central mediator of the PI3K/Akt/mTOR pathway and phosphorylates several
downstream targets which ultimately lead to cell proliferation. These include phosphorylating the
apoptosis inducing factor BAD(Datta et al., 1997) and the FKHR/FOXO transcription
factors(Medema, Kops, Bos, & Burgering, 2000) which inhibit apoptosis and promote cell
survival, or phosphorylating glycogen synthase kinase-3 which removes inhibition of pro-
proliferative pathways (van Weeren, de Bruyn, de Vries-Smits, van Lint, & Burgering, 1998).
mTOR is another phosphorylation target of Akt and is the focus of this paper. mTOR is a kinase
that plays an important role in cell growth via modulation of cell cycle regulators or maintaining
nutrient supplies into the cell(Advani, 2010). It is affected by Akt through the tuberous sclerosis
complex (TSC), which is composed of two subunits: TSC1 (hamartin) and TSC2
(tuberin)(Manning, Tee, Logsdon, Blenis, & Cantley, 2002). TSC2 is a negative regulator of
mTOR and phosphorylation of TSC2 by Akt relieves TSC2’s inhibitory effect on mTOR(Zhang
et al., 2003). Once activated, mTOR phosphorylates it’s downstream effectors including p70S6K
and eIF4E binding protein 1 (4EBP1) which are both involved in protein synthesis(Harrington et
al., 2004; Jefferies et al., 1997).
Termination of the PI3K pathway is accomplished via phosphatase-mediated degradation of
PIP3 or through the negative feedback induced by the P70S6K-mediatphosphorylation of the
upstream IRS1.
14
Figure 1.2.1: The PI3K/Akt/mTOR pathway
1.2.2 PI3K/Akt/mTOR Pathway in Endocrine and Non-Endocrine
Neoplasia
Given the key role of the PI3K/Akt/mTOR pathway in cell growth and metabolism, its role in
pathological states such has neoplasia has been investigated extensively in the past decades. As
described previously, the PI3K/Akt pathway is activated by upstream ligand dependent RTKs.
One of the most widely studied RTK is the ERBB2 receptor which is frequently overexpressed
15
in breast and other cancers. In breast cancer, for example, ERBB2 is positively associated with
worse histological grade, aneuploidy, high rate of cell proliferation and poor survival (Revillion,
Bonneterre, & Peyrat, 1998). Furthermore, transgenic mice overexpressing ERBB2/HER2
develop mammary tumors and lung metastases(Guy et al., 1992; Muller, Sinn, Pattengale,
Wallace, & Leder, 1988).
The PIK3CA gene, which encodes the p110 catalytic subunit of PI3K, appears to be involved in
a number of cancers. For example, PIK3CA has been reported to have increased amplification in
ovarian carcinoma, increased PIK3CA expression, and subsequent increased PI3K activity
(Shayesteh et al., 1999). Somatic mutations in PIK3CA have also been reported in a number of
cancers, including tumors of the colon, breast, brain, lung, and parathyroid(Kasaian, et al. 2013)
(Samuels & Velculescu, 2004). Furthermore, PIK3CA mutations have been shown to upregulate
AKT and promote oncogenic transformation in vitro (Oda et al., 2008) and in vivo (Bader, Kang,
& Vogt, 2006).
The tumor suppressor gene PTEN, a negative regulator of PI3K signaling, has also been widely
studied in its association to cancer. It has been reported that somatic mutations, allelic
inactivation, or gene silencing via promoter hypermethylation are present in glioblastoma,
melanoma, endometrial, and colon cancer(Chow et al., 2011; Lahtz, Stranzenbach, Fiedler,
Helmbold, & Dammann, 2010; Mhawech-Fauceglia et al., 2014; Nassif et al., 2004).
Furthermore, the inactivation tends to be associated with poor clinical outcome. In prostate
cancer, for example, Yoshimoto et al., 2007 demonstrated that homozygous PTEN deletion was
an indicator of a more advanced disease at surgery and also associated with faster time to
biochemical recurrence of disease (Yoshimoto et al., 2007).
AKT activation is a general feature of malignancy, seen in breast, ovarian, and pancreatic
cancers (Altomare et al., 2002; Kandel & Hay, 1999; Roy et al., 2002; Tanno et al., 2004). Of
particular note, brain tumors such as meningioma and glioblastoma have increased cell
proliferation with concurrent activation of the AKT pathway (Johnson, O'Connell, Pilcher, &
Reeder, 2010; Riemenschneider, Betensky, Pasedag, & Louis, 2006) Furthermore AKT
activation has been linked with poor prognosis in many human cancers(Nam et al., 2003; Perez-
Tenorio & Stal, 2002; Yamamoto et al., 2004). In addition, AKT is associated with resistance to
16
chemo- and radiotherapy (Brognard, Clark, Ni, & Dennis, 2001; Clark, West, Streicher, &
Dennis, 2002; Tanno et al., 2004).
It has been shown that a number of cancers have increased mTOR activation. This activation is
also associated with clinicopathologic activities. For example, in gastric cancer an increase in
phosphorylated cytoplasmic mTOR was associated with depth of tumor invasion, tumor stage,
and poorer survival rates (Murayama et al., 2009). Nuclear phosphorylated mTOR expression
was also associated with poor survival in endometrial cancer (Yoshida et al., 2010). Based on the
wealth of data available in the literature on the role of PI3K/AKT/mTOR pathway, it is clear that
this pathway plays a key role in cancer development and correlates with clinical outcome.
Therefore, pharmacologic inhibition of this pathway may be an effective treatment strategy for
treating cancer.
1.2.3 The Importance of Establishing Novel Therapeutic Targets
in PA
PA are typically benign non-metastasizing lesions that may have few characteristics in common
with more aggressive cancer types (Asa & Ezzat, 2009). However, a large proportion of PA at
diagnosis are macroadenomas that have a tendency to become locally invasive (Mete & Asa,
2012) An examination of the molecular determinants of these invasive properties in PA may
reveal certain signalling patterns that are also involved in more malignant tumors.
Due to the aggressive nature of some PA, complete surgical removal may be impossible in the
clinical setting. In such cases, the presence of residual tumor may result in tumor recurrence.
Furthermore, some PA may recur several times following multiple survival procedures.
Recurrent tumors that are not amenable for reoperation undergo radiation therapy, which may
lead to adenohypophseal damage leading to hypopituitarism, optic nerve damage, or cognitive
deficits ((Brada et al., 1993; Hahn et al., 2009)).
As mentioned previously, most PA at first clinical presentation are large macroadenomas, and
these macroadenomas consist primarily of NFPA, which are unresponsive to current medications
(Mete & Asa, 2012). Among functional PA, the available medications, including somatostatin
17
analogs or dopamine agonists, are only effective in certain subtypes of GH PA and PRL PA,
respectively ((Mete & Asa, 2012)). Thus, there remains a need to identify novel targets for
therapy for a large set of patients for whom current options are limited.
The pathobiology of PAs is complex and is yet to be elucidated. It has been established that the
classical oncogenes and tumor suppressor genes are not commonly altered in PA (Asa & Ezzat,
2009). However, recent interest has focused on exploring whether elements of the
PI3K/Akt/mTOR pathway plays a role in the progression of PA, with ultimate hope that
targeting this pathway in PAs will provide a novel therapeutic target.
1.2.4 Analysis of the RTK/PI3K/Akt/mTOR Pathway in PA
Several studies have used mouse and rat pituitary cell lines in vitro to study PA tumorigenesis
(Table 1.2.1). The reader is referred to reference (Ooi, Tawadros, & Escalona, 2004) for a
summary of how current mouse and rat PA cell lines were established.
The PI3K/Akt/mTOR pathway is activated by RTK and RTK activation has been examined in
relation to the pituitary. RTKs are important for normal pituitary function and help to modulate
hormone production and cell growth ((Ezzat, 2001)). Aberrant RTK activation possesses the
ability to confer proproliferative potential and abnormal hormone production to pituitary cells
leading to hyperplastic and/or neoplastic growth. Two important RTKs in normal pituitary
development that are also implicated in PA are EGFR and FGFR. Currrently, studies examining
RTK activation in PA and subsequent downstream PI3K/Akt/mTOR signalling remain scant.
EGFR has been implicated in PA. Preclinical studies demonstrate that EGF is able to enhance
mRNA levels of prolactin in PA cell lines while gefitinib, an EGFR inhibitor, has been shown to
block serum-induced cell proliferation and prolactin gene expression(Vlotides et al., 2008).
Furthermore, Lapatinib, a dual EGFR and HER2 inhibitor has been reported to be more potent
than gefitinib at abrogating PRL expression and PA cell proliferation both in vitro and in
vivo(Fukuoka et al., 2011). The EGFR-mediated effects on PA cell hormone regulation and
proliferation have been shown to be PI3K-dependent. In the human situation, EGFR is
18
demonstrable in all types of PA \and its expression has been shown to correlate with tumor
aggressiveness especially in GH PA and NFPA (Cooper, Vlotides, Fukuoka, Greene, & Melmed,
2011; LeRiche, Asa, & Ezzat, 1996) .
The FGF receptor has also been implicated in PA. The FGFR consists of four receptors and
mRNA for prototypic isoforms of FGFR1, 2, and 3 are present in the normal pituitary(Abbass,
Asa, & Ezzat, 1997). A common alteration in the FGFR that promote PA is the presence of an N-
terminally truncated variant of FGFR4 called pituitary tumor derived FGFR4 (ptd-
FGFR4)(Ezzat, Zheng, Zhu, Wu, & Asa, 2002). N-terminal truncation of FGFR4 results in a
constituitively activated protein that is localized to the cytoplasm where it promotes PA
oncogenic transformation in vitro and in vivo. FGFR4 inhibition with a FGFR-selective
inhibitor has been shown to restore membranous FGFR4 and inhibit PA proliferation. Ptd-
FGFR4 is present in human PA where it has been shown to correlate with tumor
aggressiveness(Ezzat, Zheng, & Asa, 2004; Ezzat, Zheng, Winer, & Asa, 2006). Recently, a SNP
in the FGFR4 gene, resulting in a glycine to arginine substitution in the transmembrane domain
of the FGFR4 protein, has been identified in PA. The presence of the polymorphic variant has
been shown to promote PA cell proliferation and hormone deregulation in vitro and in vivo. In
humans, the presence of this SNP has been shown to correlate with GH levels and tumor size in
GH PA(Tateno et al., 2011). The FGFR alterations with subsequent PI3K/Akt/mTOR signalling
have not examined in PA, however, the polymorphic variant of FGFR4 and downstream mTOR
signalling has been reported in pancreatic neuroendocrine tumors(Serra et al., 2012).
Consistent with a role for PI3K activation in PA, in study by Lin et al., mutations of the PIK3CA
gene were assessed in 353 pituitary tumors. Nine percent of the invasive, but none of the non-
invasive tumors harbored PIK3CA gene mutations. Genomic PIK3CA amplifications, defined as
more than four copies, were observed in both invasive and non-invasive tumors with a
prevalence of 20-40% (Lin et al., 2009). Another study examined PIK3CA mutations and
genomic amplifications in 33 PAs including ACTH-, GH-, PRL-producing and NF-PAs and
found that PIK3CA mutations were evident in 12.1% of tumors including one non-invasive
ACTH tumor. Genomic amplification (defined as copy number ≥4) were found in 21.2% of cases
(Murat et al., 2012).
19
Additional elements of the PI3K/Akt/mTOR pathway have been examined in PA. In human
tissue, AKT, PTEN, and p27 (a target of Akt) mRNA and protein expression was analyzed in
ACTH-, GH-, PRL-, and NF-PAs, as well as in normal pituitary tissue controls. In PAs, AKT
mRNA expression and phosphorylated-AKT protein levels were increased in comparison to
normal pituitary tissue. Additionally, the levels of PTEN and p27 were lower in PA(Musat et al.,
2005). The immediate downstream effectors of mTOR, pS6 and eIF4E, have also been studied
in human PA tumor samples with more frequent activation of S6/eIF4E evident in all recorded
PA subtypes relative to normal pituitary controls (33-71% vs. 20%), with GH-PAs exhibiting the
highest frequency of overexpression (Sajjad et al., 2013). Mouse models harbouring genetic
mutations resulting in the formation of PA have also been reported to exhibit unregulated
p70S6k/S6 expression in PA tissue relative to adjacent CNS tissue(Kenerson, Dundon, & Yeung,
2005; Lu, Willingham, Furuya, & Cheng, 2008)In contrast, another study reported that there was
no difference in the expression of phosphorylated or total mTOR, TSC2, or p70S6K as compared
to controls. However, this group did report an increase in c-myc levels (a target of Akt) in all PA
subtypes, as well as a mild activation of cyclin D1 but only in NF-PAs (Dworakowska et al.,
2009). Furthermore, elevated levels of cyclin D1 staining in NF-PAs compared to other tumors
has been replicated in a number of other studies. In a study including only NF-PAs and GH-
PAs, cyclin D1 allelic imbalance, which may indicate gene amplification, was detected in one
quarter of the PAs analyzed (Hibberts et al., 1999). Table 1.2.2 provides a comprehensive list of
the different components of the PI3K/Akt/mTOR pathway examined in PA.
Table 1.2.1: Examining elements of the PI3k/Akt/mTOR in models of pituitary adenoma
Study Model Species Treatment Proteins Outcome
Vlotides et
al., 2008
GH3 cells Rat EGF EGFR
expression
Increase PRL mRNA
levels and cell
proliferation;
Romano et
al., 2006
GH4C1
cells
Rat IGF-1;
PI3K/Akt
PI3K/Akt Increased PRL
release; PI3K/Akt
20
inhibitors inhibitors reduce
PRL release
Kowarik
et al.,
2010
AtT20,
MTt/S,
TtT/GF
cells
Rat/mous
e
PDGF/LY2940
02
PDGFR
expression,
PDK1, Akt
Mixed expression of
PDGFR in cells;
PDGF-induced cell
proliferation and
VEGF-A secretion;
VEGF-A secretion
blocked by
LY294002
Banerjee
et al.,
2003
GH3 cells Rat 17-α-
estrodiol/wort
mannin
PI3k/Akt Increased VEGF-A
mRNA expression;
decreased by
wortmannin
Fernandez
2003 and
2004
Lactotroph
cells
Rat IGF-
1/LY294002
Akt, BAD,
bcl-2
Increased cell
proliferation and cell
survival; reversed by
PI3K inhibitor
LY294002
Rose et al
2004
aT3 mouse IGF-1/LHRH IRS-1,
PI3K,Akt
IGF-1 stimulates cell
proliferation and
survival; cotreatment
with LHRH reduces
cell survival
Lu et al,
2008
Genetic
mutant
Developing
TSHoma
Mouse Knock in
mutation of the
thyroid
receptor B
gene
Akt,mTOR,
p70s6k, bcl-2
Decreased apoptosis;
LY294002 reduces
pituitary growth
Kenerson Renal and Rat Germline mTOR, S6 The mTOR pathway
21
et al 2005 pituitary
Tumor in
Eker rats
mutation of
TSC2 gene
activity is critical for
tumor progression
Musat et
al., 2005
ACTH-,
GH-, PRL-
NF-Pas
Human N/A p-Akt,
PTEN, p27
Increased p-Akt
expression; lower
nuclear p27 and
PTEN expression as
compared to normal
pituitary
Sajjad et
al., 2013
GH-,
ACTH-,
NF-PAs
Human N/A mTOR(pS6/e
IF4e)
Increased mTOR
activity in all PA
subtypes vs controls
Daworako
wska et
al., 2009
ACTH-,
GH-, PRL-
NF-
adenomas
Human N/A mTOR,TSC2
,p70s6k
No difference in
mTOR, TSC2,
p70s6k expression
compared to
controls; decreased
c-myc and increased
cyclin D1 expression
only in NF adenomas
Lin et al.,
2009
Pituitary
adenomas
Human Genetic
PI3kCA
mutations
PI3K/Akt Some invasive and
no non-invasive PA
harbored mutation
1.2.5 Current mTOR inhibitors
The proposed role of aberrant PI3K/Akt/mTOR signalling in PA makes this tumor group
amenable for the use of mTOR inhibitors in the clinic. Results using mTOR inhibitors in
preclinical models of PA have provided further support for mTOR pathway involvement in PA
22
tumorigenesis and as a target for medical therapy.
1.2.6 Effects of Rapamycin (and Rapalog)-induced mTOR
Inhibition on PA Cell Proliferation and Viability
Rapamycin (sirolimus) is an immunosuppressant and antiproliferative agent that has previously
been shown to be effective at abrogating cancer-related properties in a number of tumors
(Douros & Suffness, 1981; Majumder et al., 2004). Specifically, rapamycin binds to the
mTORC1 complex affecting downstream signalling events including cell cycle arrest and protein
synthesis inhibition. Tumors that harbor upstream mutations from mTOR, such as PTEN deletion
or Akt overexpression, are ideal targets for treatment with mTOR inhibitors. Everolimus
(RAD001) is an orally available analog to rapamycin (rapalog) and has also been shown to be an
effective anticancer agent in a number of in vitro cell lines and animal models (Beuvink et al.,
2005).
PAs are typically of low proliferative potential as assessed by the cell cycle antigen, Ki-67 and of
low mitotic index. However, the ability of Ki-67 index to predict tumor growth/invasion and
recurrence is debatable (Chacko et al., 2010). The presence of large macroadenomas that have a
propensity for invasion into extrasellar structures could potentially benefit from antiproliferative
agents. The ability of rapamycin and its analogs to inhibit cell proliferation is key to its antitumor
efficacy. Several studies have examined the antiproliferative properties of rapamycin and
rapalog treatment in pituitary and PA cells (Table 1.2.2). In normal rat pituitary cells, rapamycin
was shown to inhibit basal proliferation and insulin-, cAMP-, and estradiol-induced proliferation
of cells (Kawashima, Yamakawa, & Arita, 2000). Human prolactin gene is expressed in the GH3
cell line and its transcription was shown to be inhibited by rapamycin (Wera, Belayew, &
Martial, 1995; Wera, Zheng, et al., 1995). In an animal model using rats that carry an
inactivating germline mutation of the TSC2 gene that results in pituitary tumor formation,
rapamycin induced regression in size of the pituitary tumors and a concomitant decrease in levels
of phosphorylated-S6 (the target of p70S6K) (Kenerson et al., 2005). A recent study using
human PA cells in primary culture has also demonstrated that rapamycin induces mTOR
23
inhibition in mTOR-active GH-, ACTH-, and NF-PAs, although, in this study, cell viability or
proliferation was not assessed (Sajjad et al., 2013). Nevertheless, these studies provide evidence
for rapamycin as a possible anti-PA agent through its mTOR-inhibiting effects.
Gorshtein et al., first demonstrated the antiproliferative effects of rapamycin and its orally
bioavailable analog RAD001 on GH3 and MtT/S cells as well as on human GH-PA cells in
primary culture. They reported that treatment with rapamycin or RAD001 significantly decreased
cellular viability and proliferation in a dose- and time-dependent manner, which was reflected by
decreased of phosphorylated p70S6K (Gorshtein et al., 2009). Another study used GH3 and the
prolactin secreting MMQ cell lines and also noted a time- and concentration-dependent decrease
in cellular viability following exposure to rapamycin or RAD001 (Sukumari-Ramesh, Singh,
Dhandapani, & Vender, 2011). These findings were attributed to decreased levels of mTOR
phosphorylation at the serine-2448 residue, which is a key determinant of mTOR activity. A
third study noted a small but statistically significant antiproliferative effect of 1 nM rapamycin
on mouse AtT20 cells and no effect at lower doses (Cerovac et al., 2010).
Musat et al. have reported that NFPAs have the highest levels of AKT mRNA expression of all
PA subtypes (Musat et al., 2005). Furthermore, it has been shown that high levels of
phosphorylated AKT may contribute to early recurrence of NF-PA (Noh et al., 2009). Therefore,
it is important to evaluate the antiproliferative efficacy of rapalogs in this histological subtype.
Zatelli et al. examined RAD001 sensitivity in 40 NF-PAs and reported that 70% (28/40)
displayed a dose-dependent reduction in cellular viability. Furthermore, in these 28 samples,
RAD001 blocked IGF-1 induced increases in cell proliferation and VEGF secretion, although
RAD001by itself had no effect on VEGF secretion (Zatelli et al., 2010). Of note, investigation of
patient medical records showed that those patients who responded to RAD001 were younger,
were more likely to have an invasive macroadenoma, and predominantly female. Lee et al.
examined a particular strain of rats with a MENX, a multiple endocrine neoplasia-like syndrome,
mutation. MENX is caused by a biallelic frameshift mutation in the cdkn1b gene, encoding p27,
which leads to the development of multiple endocrine tumors, including NFPA. Administration
of RAD001 to MENX PA cells in primary culture also resulted in a dose-dependent reduction in
cellular viability that reached significance at a dose of 100 nM in 45% of their sample (5 out of
24
11 cultures) (Lee et al., 2011). Finally, Cerovac et al., exposed 28 human NFPAs in primary
culture to rapamycin and reported a small (<20%) reduction in cellular proliferation in 29% (8 of
28 tumors) of their samples (Cerovac et al., 2010).
Table 1.2.2: In vitro studies examining effect of mTOR inhibitors on PA cells
Cells Species Hormone Drug Outcome
Lactotrophs(1°) Rat PRL Rapamycin Inhibit basal and growth
factor induced proliferation
of cells (Kawashima, 2000)
GH3(CL) Rat GH/PRL Rapamycin Transcription inhibition of
prolactin gene (Wera, 1995);
decreased cell viability and
proliferation (Gorshtein,
2009, Sukumari, 2011);
cotreatment with RT results
in radiosentization
(sukumari); decrease cell
proliferation/increased
apoptosis (Dai 2013)
RAD001 Decreased cell viability and
proliferation (gorshtein,
sukumari); cotreatment with
RT results in radiosentization
(sukumari)
MtT/S(CL) Rat GH Rapamycin/RAD001 Decreased cell viability and
proliferation (Gorshtein)
MMQ(CL) Rat PRL Rapamycin Decreased cell viability and
proliferation (Sukumari);
decreased cell
proliferation/increased
25
apoptosis(Dai 2013)
αT3-1(CL) Mouse FSH/LH Rapamycin Decreased cell
proliferation/increased
apoptosis (Dai 2013)
RAD001 Decreased cell viability
(Sukumari)
AtT20(CL) Mouse ACTH rapamycin Decreased cell proliferation
(Cerovac)
GH-adenoma(1°)
Human
GH
Octreotide/Rapamycin
Rapamycin/RAD001
Greater decrease in cell
proliferation than rapamycin
or octreotide alone (Cerovac)
Decreased cell viability and
proliferation (Gorshtein)
NFPA(1°) Human N/A Rapamycin Decreased cell proliferation
(Cerovac)
Octreotide/Rapamycin Greater antiprolierative
response than either
octreotide or rapamycin
exposure
RAD001 Decreased cell viability;
blockage of growth factor
induced cell proliferation and
VEGF secretion (Zatelli)
SOM230/RAD001 Greater antiproliferative
effect seen when drugs were
combined versus individually
(Zatelli)
NVPBEZ235 Decreased cell viability
(Zatelli)
MENX (1°) Rat N/A RAD001 Decreased cell proliferation;
triggers apoptosis (Lee)
26
NVPBEZ235 Greater antiproliferative
effect than RAD001; triggers
apoptosis(Lee)
CL=Cell line
1°=primary cells
1.2.7 Effects of Dual PI3K/mTOR Inhibition on PA cellular
Proliferation and Viability
One potential pitfall of mTOR inhibition is that mTOR activation normally causes negative
feedback on IRS1. O’Reilly et al. reported that inhibition of mTOR by rapamycin does indeed
cause a decrease in p70S6K, abolishing its negative feedback to IRS1 and increasing Akt
phosphorylation (O'Reilly et al., 2006). This mechanism may explain the partial response to the
inhibitor treatments mentioned above. Theoretically, it seems that co-administration of an agent
able to decrease Akt phosphorylation upstream could potentiate the antiproliferative action of
rapamycin treatment downstream.
Somatostatin receptor type 2 (SSTR2) was shown to deactivate the PI3K pathway by inhibiting
p85 tyrosine phosphorylation and thereby decreasing Akt phosphorylation (Theodoropoulou et
al., 2006). Somatostatin-analogs such as octreotide are frequently used for the treatment of
neuroendocrine tumors (Lamberts, de Herder, & Hofland, 2002). Cerovac et al. demonstrated
that the addition of octreotide to rapamycin increased serine-phosphorylated IRS-1 levels but
notably decreased rapamycin induced pAkt-Ser levels (Cerovac et al., 2010). In AtT20 cells, the
addition of octreotide to rapamycin resulted in a 50% decrease in cell viability, an effect that was
significantly greater than the actions of octreotide or rapamycin alone (Cerovac et al., 2010).
Rapamycin combined with somatostatin analogues are effective at inhibition of human PA cell
growth in vitro. In one study, among human NFPAs that did not respond to rapamycin treatment
alone (20/28), all responded to concurrent rapamycin and octreotide treatment. Moreover, in the
27
cotreatment group a lower dose of rapamycin was required to maintain a significant
antiproliferative effect. This study also noted that cotreatment suppressed proliferation to a
greater extent in the cells that did respond to rapamycin alone (8/28) (Cerovac et al., 2010).
Similar results were obtained when human NFPAs were exposed to SOM230, a somatostatin
receptor multiligand, in combination with RAD001. This study showed higher response rate to
RAD001 alone (28/40 cultures responded), but cotreatment with SOM230 significantly
potentiated the antiproliferative effect of RAD001 compared to SOM230 or RAD001 alone
(Zatelli et al., 2010). Similar antiproliferative results were also obtained when ACTH PA cells in
primary culture were exposed to SOM230 and RAD001 in combination (Zatelli, Minoia, Filieri,
Tagliati et al., 2010).
The effects of RAD001 on the negative feedback loop which, by inhibiting mTOR, reduces
p70S6k phosphorylation and induces IRS-1 expression, was examined in pituitary cell lines. In
one study, GH3 and AtT20 cells were exposed to rapamycin, which reduced serine-
phosphorylated mTOR and phosphorylated p70S6K levels. In AtT20 cells, rapamycin treatment
increased phosphorylated Akt levels, but this effect was reversed by octreotide in a serine-
phosphorylated IRS1-dependent mechanism (Cerovac et al., 2010). In contrast, Gorshtein
showed that rapamycin or RAD001 treatment of GH3 and MtT/S cells abrogated
phosphorylation of p70S6K without an accompanied rise in phosphorylated Akt levels 24 hours
after treatment (Gorshtein et al., 2009).
Recently, a new class of compounds have been developed that are designed to inhibit both
mTOR and upstream PI3K pathway components. Theoretically, these drugs are designed to
inhibit mTOR without stimulating an increasing in upstream Akt activity. One of these agents,
NVPBEZ235, is a synthetic small molecule that inhibits both PI3K and mTOR kinase activity by
binding to the ATP cleft of these enzymes (Maira et al., 2008). In a study by Lee et al. in 2011,
NVPBEZ235 was reported to inhibit the viability of 100% (10 of 10) of MENX rat PAs in
primary cell culture. This result was significantly more potent that previous studies of RAD001
effects on this cell line. Furthermore, it was noted that incubation with NVPBEZ235 at
concentrations that suppress cellular viability also decreased the phosphorylation of Akt and S6,
whereas RAD001 decreased phosphorylated S6, but increased phosphorylated Akt (Lee et al.,
28
2011). In another study examining NVPBEZ235, NFPA cells in primary cell culture exposed to
the dual inhibitor exhibited a 50% decrease in cell viability in 37 of 40 cases (Zatelli et al.,
2010).
Dual PI3K/mTOR inhibitors have also been combined with more conventional therapies to yield
promising preclinical results in PA cells. One such dual-inhibitor known as XL765 used in
combination with temozolomide (TMZ), an orally-available alkylating agent previously shown
to be effective at reducing PA cell viability (Ma et al., 2011; Sheehan, Rainey, Nguyen,
Grimsdale, & Han, 2011), synergistically inhibited the proliferation of αT3-1, MMQ, and GH3
PA cell lines (Dai et al., 2013).
1.2.8 Effects of mTOR Inhibition on Apoptosis on PA Cells in vitro
Apoptosis, or programmed cell death, is characterized by a rapid sequence of events leading to
the elimination of damaged cells. It is defined by morphological changes such cell shrinkage and
nuclear demarcation. In neoplastic tissue, apoptosis is typically suppressed in favour of cell
survival.
PA are clonal lesions that are thought to derive from a single transformed cell and defects within
the apoptotic pathway may preserve the survival of the altered cell and promote PA genesis and
growth. However, the clinical utility of using indicators of apoptosis as a prognostic tool in PAs
is not consistently reported(Guzzo, Carvalho, & Bronstein, 2014). Nevertheless, induction of PA
cell apoptosis via mTOR inhibition has been an important characteristic of their anti-PA
properties. Caspases are the main protein family effectors of apoptosis and the study by Zatelli
et al study demonstrated that the effects of RAD001 on cellular viability of NFPA may be
attributed to increased levels of caspase 3/7 activity (Zatelli et al., 2010). Increases in caspase 3/7
activity were also reported by Lee et al. following NVPBEZ235 treatment of MENX rat PAs in
primary culture (Lee et al., 2011) but did not affect caspase activity in AtT20 cells (Cerovac et
al., 2010). The use of XL765 and TMZ synergistically increased caspase 3/7 levels and TUNEL-
positive T3-1, MMQ, and GH3 cells compared to either treatment alone (Dai et al., 2013).
29
1.2.9 Effects of PI3K and mTOR Inhibition on PA Cell Cycle
The cell cycle is an important regulator of cell proliferation. Many cyclins, cyclin dependent
kinases, and cyclin kinase inhibitors have been implicated in PA tumorigenesis(Saeger, 2004)
and some have been shown to be affected by mTOR inhibition.
Previous in vitro studies have reported that induction of cell cycle arrest is an important
mechanism by which mTOR inhibitors exert their antiproliferative effects on PA cells (Faivre,
Kroemer, & Raymond, 2006). It has been reported that treatment with rapamycin results in the
arrest of PA cell lines in the G0/G1 phase.
The early G1 phase is positively regulated by D-type cyclins and their corresponding cyclin-
dependent kinases (Sherr, 2000). Rapamycin and it’s analogs seem to primarily affect cyclin D3
levels while cyclin D1 and the cyclin dependent kinases (CDK4 and CDK6) remain unaffected.
The dual PI3K/mTOR inhibitor NVP may be a more potent agent at abrogating PA cell cycle
progression by attenuating both cyclin D1 and Cyclin D3.
The KIP1/CIP family of cyclin-dependent kinase inhibitors (CKDI) negatively regulate cyclins
and CDKs in the G1 phase of the cell cycle. Gorshtein et al. have demonstrated that p21/CIP
levels were reduced by rapamycin in GH3 cells (Gorshtein et al., 2009). A similar down-
regulation of p21/CIP was also noted when AtT20 cells were exposed to high concentrations of
NVPBEZ235 (Cerovac et al., 2010). On the other hand, p27/KIP1 was reported to have greater
transcriptional and protein expression following combined rapamycin/octreotide treatment in
AtT20 cells (Cerovac et al., 2010). Increased p27 also seemed to play a role in the
antiproliferative action of the dual inhibitor NVPBEZ235. Lee at al. were able to demonstrate
that increased levels of p27 positively correlate to the antiproliferative efficacy of NVPBEZ235
in GH3 cells transfected with wild type p27 and in MENX rat PA cells in primary culture (Lee et
al., 2011). However, unlike rapamycin/octreotide cotreatment, which increased protein and
transcript levels of p27 in AtT20 cells, NVPBEZ235 only increased p27/KIP1 protein levels in
GH3 cells, indicating regulation at a post-transcriptional level (Cerovac, Tichomirowa, Stalla et
30
al., 2011).
A major determinant of G1/S progression is retinoblastoma (Rb) phosphorylation. CDK4 when
associated with D-type cyclins phosphorylates Rb in the G1 phase. Gorshtein et al. showed that
rapamycin inhibits Rb phosphorylation in GH3 and MtT/S cells and reduces subsequent E2F
transcriptional activity (Gorshtein et al., 2009). Octreotide and rapamycin cotreatment of AtT20
cells seems to potentiate this Rb inhibition effect (Cerovac et al., 2010).
In response to decreased E2F activity, cell cycle components that contribute to late-G1 phase and
S-phase entry, such as the expression of E2F regulated genes cyclin E and CDK2, were also
reduced by rapamycin in GH3 cells (Gorshtein et al., 2009). In AtT20 cells, cyclin E expression
is reduced even further by rapamycin and octreotide co-treatment (Cerovac et al., 2010).
Thus, it appears that mTOR inhibitors possess the capacity to reduce proliferation of PA cells by
inducing apoptosis and restoring some of the inhibitory mechanisms involved in the cell cycle
particularly in the G1 phase.
1.2.10 mTOR Inhibition and Radiation on PA Cells
Radiotherapy is typically reserved for particularly aggressive PA that are not suitable for surgical
resection or are multirecurrent lesions that are resistant to pharmacologic therapies. To date, a
number of studies have shown that mTOR inhibition radiosensitizes cancer cells (Altmeyer et al.,
2012; Burris, 2013; Mauceri et al., 2012), Existing data suggests that mTOR inhibition may also
lead to decreased radioresistance of PA cells in vitro. Ramesh et al. report that inhibition of
mTOR radiosensitizes GH3 cells such that 2.5 Gy in combination with 0.5 nM rapamycin or
RAD001 reduced cellular viability more effectively than 2.5 Gy or 10 Gy alone (Sukumari-
Ramesh et al., 2011). Another study utilized nelfanavir, a protease inhibitor, to increase radiation
sensitivity of GH3 cells, as well as MMQ and AtT20 cells. At three days post-radiation, cellular
viability was decreased in all cell lines in a radiation-dose-dependent manner (dose range 0-6
Gy) (Zeng et al., 2011). In this study, apoptosis was induced in vitro at higher rates by co-
nelfinavir and radiation treatment compared to either intervention alone. The proposed
31
mechanism of nelfinavir action is decreased phosphorylated S6, a key downstream target of the
PI3K/Akt/mTOR pathway. Thus, these data suggest that nelfinavir action may bare similarities
to that of rapamycin and its derivatives.
An increase in the radiosensitization of PA cells by mTOR inhbitors is an important therapeutic
feature because although radiotherapy is a relatively uncommon form of treatment for PA, it is
nevertheless utilized for aiding in the arrest of tumor growth of invasive and aggressive PA.
Radiotherapy has been associated with a number of deficits in patients with PA including optic
nerve damage, cranial nerve palsies, and adenohypophseal damage. Thus, when radiotherapy is
employed, coupling this form of therapy with mTOR inhibition in PA patients may reduce the
required dose of radiation thus minimizing the subsequent radiation-induced adverse effects.
1.2.11 mTOR Inhibition Using Xenograft In Vivo Models of PA.
Xenograft models are an important tool for studying the behaviour of PA. Unlike transgenic
models that may take a prolonged period of time to develop a PA, xenograft models can be
generated quickly and reliably and are more reflective of the common sporadic PA which do not
usually possess underlying genetic mutations(Asa & Ezzat, 1998). Furthermore, some transgenic
mouse models first develop pituitary hyperplasia prior to developing a PA which is not
characteristic of the human scenario.
Two studies to date have used xenograft models to examine the effects of mTOR inhibition on
PA. Zeng et al. demonstrated that GH3 cells implanted into the flanks of nude mice treated with
a combination of nelfinavir and radiation experienced delayed tumor size quadrupling time. Co-
treatment displayed a synergistic effect compared with nelfinavir or radiation treatment alone.
Immunohistochemistry of tumor sections from this study confirmed downregulation of
phosphorylated S6 following treatment with nelfinavir (Zeng et al., 2011). In another study by
Dai et al., 2013, GH3 cells were also used and implanted into the flanks of female nude mice.
These mice were treated with a combination of XL765 and TMZ which was shown to inhibit
tumor growth and induce apoptosis, inhibit GH and PRL secretion and downregulate Akt,
mTOR, and S6 phosphorylation (Dai et al., 2013).
32
Unfortunately, there are currently no intracranial mouse models of PA in which to study mTOR
inhibition. Intracranial models may be more reflective of the human situation by mimicking the
intracranial microenvironment of PA. An intracranial xenograft model may provide a more
accurate portrait of the nature of sporadic PA growth behaviour following mTOR inhibition.
One factor hampering the establishment of an intracranial mouse model is the difficulty in
accessing the mouse pituitary fossa. Nevertheless, more robust xenograft models in general and
intracranial models specifically are needed to better characterize PI3K/Akt/mTOR pathway
signalling and how it impacts PA growth dynamics. Utilization of improved preclinical models
of PA may provide stronger evidence for the role of this oncogenic pathway in PA and pave the
way for clinical trials involving mTOR inhibition in PA patients.
33
Chapter 2
Outline and Overview of Aims
2.1 Purpose
The overall goal of my research is to gain a better understanding of pituitary adenoma (PA)
biology. I have two main aims for this project. The first aim is to correlate PA biology with
relevant clinical characteristics in order to provide a comprehensive understanding of the
mechanisms contributing to tumor growth behaviour. The second aim will be to establish an
intracranial model of PA and identify signal transduction pathways involved in PA in order to
identify appropriate targeted treatment strategies.
2.2 Hypothesis 1
To establish whether the growth and extension pattern of PA can predict postoperative growth
rate and recurrence in addition to whether the PA growth rate correlates with proliferation and
growth factor expression.
2.2.1 Aim#1: Examine growth patterns of pituitary adenomas and
histopathological correlates
To retrospectively review clinical, radiological and histopathological features of PA and
correlate them with growth patterns and rate of PA
2.2.2 Rationale for Aim#1
PA are a frequently occuring intracranial neoplasm, evident in up to 20% of the general
population(Asa and Ezzat 2009). PA comprise several histopathological subtypes each
34
displaying characteristic presenting symptoms based on specific hormonal imbalances in the
body and/or symptoms of tumor mass, such as visual disturbance. First-line therapy for the
majority of PA is typically endonasal surgery although medications such as dopamine agonists or
somatostatin analogs, are useful in certain PA subtypes.
Despite its often benign and indolent nature, the majority of PA are large macroadenomas at
diagnosis. Macroadenomas usually grow out of the sellar compartment and compress or invade
adjacent structures. Parasellar PA growth into the CS is particularly problematic as it often
precludes complete tumor removal and complicates patient management(Sepehrnia et al. 1991).
It has been suggested that different histopathological subtypes have a proclivity for different
patterns of extrasellar growth(Mete and Asa 2012). Furthermore, the rate of tumor growth has
been suggested to be different for different subtypes. Specific molecular markers that are
involved in the cell cycle, such as Ki-67, are often employed as a means to assess PA growth
potential(Chacko, et al. 2010). However, it is still a matter of debate as to how accurate this
marker is for estimating PA growth behavior.
Postoperatively, there is a probability of tumor residual especially in cases in which the CS is
involved. The extent of CS involvement often correlates to a greater probability of PA residual.
Several studies have attempted to establish preoperative criteria for assessing PA involvement
into the CS as a means to predict surgical outcome(Cottier et al. 2000; Knosp et al. 1993). Even
when the PA is completely removed, some PA have a tendency to recur. The probability of
tumor recurrence varies based on its histopathological subtype and biochemical
status(Roelfsema, et al. 2012). Particularly aggressive PA are often multirecurrent and require
reoperation and/or adjuvant therapy
The clinical and molecular factors that are involved in PA growth and residual/recurrence are not
well-understood. Thus, the objective of the first aim was to correlate patient demograpic, MRI,
histopathological parameters to the growth behavior of PA both pre- and postoperatively with the
goal of identifying predictive factors of PA growth behavior.
35
2.3 Hypothesis 2 To investigate whether the mTOR pathway is expressed in an intracranial xenograft mouse
model of PA harboring wild type or mutated FGFR4 and to determine a possible differential
response to mTOR inhibition between these tumors.
2.3.1 Aim#2: Examine preclinical intracranial mouse model of PA
for interrogating therapeutic strategies: role of mTOR inhibitors
To establish an intracranial model of PA and 2b. to use this pre-clinical model for the
investigation of targeted therapeutics against the mTOR pathway.
2.3.2 Rationale for Aim#2
The pathogenesis of PA is poorly understood. The molecular markers that contribute to PA
formation and progression remain elusive. Furthermore, many of the classical oncogenes and
tumor suppressor genes involved with other forms of neoplasia are rarely altered in PA(Asa and
Ezzat 2002).
Recently the fibroblast growth factors and their receptors have been implicated in PA pathology.
A specific subtype of the FGF receptor, FGFR4, has garnered particular interest in the
pathobiology of PA after the identification of an N-terminally truncated variant that is only
present in PA, not in the normal pituitary, and is referred to as pituitary tumor derived FGFR4
(ptd-FGFR4)(Ezzat, et al. 2002). Ptd-FGFR4 has been shown to confer oncogenic characteristics
to pituitary cells both in vitro and in vivo, an effect not seen in wild-type full length
FGFR4(Ezzat, et al. 2006).
Shortly after the discovery of ptd-FGFR4 in PA, a polymorphism in the gene encoding the
FGFR4 protein was identified(Bange, et al. 2002). This polymorphism occurs at codon 388 and
results in the substitution of an uncharged glycine residue for a charged argine residue in the
36
transmembrane domain of the FGFR4 protein (FGFR4-G388R). At least one allele of this SNP
is present in approximately 50% of the population. It is associated with poor prognosis in
several cancers including breast and prostate(Thussbas, et al. 2006; Wang, et al. 2004). There is a
paucity of data implicating this SNP in PA. However, one study has identified distinct signaling
and hormone regulatory properties that distinguish FGFR4-R388 from the prototypic FGFR4-
G388 form in GH-PA (Tateno, et al. 2011)and another in ACTH-PA(Nakano-Tateno, et al.
2014).
The activation of the PI3K/Akt/mTOR pathway signaling may serve as a viable downstream
target that may also distinguish the FGFR4-R388 SNP variant from wild type FGFR4-G388.
The PI3K/Akt/mTOR pathway is activated via upstream receptor tyrosine kinases (RTKs), which
includes FGFR4. Also, elements of this pathway have been shown to upregulated in PA in in
vitro and in vivo models, including samples of human PA of various histotypes(Musat, et al.
2005; Sajjad, et al. 2013). Moreover, PA cells respond to mTOR inhibition with a
downregulation of mTOR pathway expression resulting in a reduction of proliferation and
induction of apoptosis(Gorshtein, et al. 2009; Zatelli, et al. 2010).
A major factor impeding the study of the mechanisms contributing to the formation and
progression of PA is the lack of studies utilizing a robust intracranial model of PA. Current in
vivo models involve transgenic mutants that may or may not develop a PA even after a
prolonged period of time. Alternatively, xenograft of PA cell lines are used but these are
implanted subcutaneously and may not be characteristic of the human scenario. Thus, the first
part of aim 2 (aim 2a) is to establish an intracranial model of PA that can be conducted quickly
and reliably. The second part of aim 2 is to use this PA model to identify the mTOR pathway as
a possible therapeutic target that distinguishes wild type FGFR4-G388 from FGFR4-R388 and
helps to establish mTOR inhibitors as a viable treatment option for PA patients.
37
Chapter 3
Growth Patterns of PA and Histopathological Correlates This chapter was published in the Journal of Clinical Endocrinology and Metabolism (JCEM, 2014) IF=6.430
(Monsalves, et al. 2014)
3.1 Introduction Pituitary adenomas (PA) are common in neurosurgical practice, accounting for up to 25% of all
intracranial neoplasms(Asa and Ezzat 2009). The majority of pituitary macroadenomas are
clinically nonfunctioning or hormonally inactive and exhibit gonadotroph differentiation, (Mete
and Asa 2012). Pituitary adenomas exhibit a wide range of behaviour (Mete, et al. 2012) and some
have a tendency to compress or invade structures adjacent to the sella, such as the optic chiasm
or cavernous sinus (CS), causing significant morbidity and impeding complete surgical removal.
Furthermore, PA sometimes recurs despite gross total resection(Alahmadi, et al. 2012;
Benveniste, et al. 2005).
The factors that govern and can aid in predicting the behaviour of PA, such as clinical growth of
residual tumour following subtotal resection are not well understood.
Currently, it is thought that the most reliable tool for predicting the growth behaviour of PA is to
determine the histopathological subtype(Lundin, et al. 1992; Scheithauer, et al. 1986; Zada, et al.
2010). In addition, the MIB-1 labeling index (MIB-1 LI), a measure of cell proliferation, has
been used as the main predictor of tumor recurrence and to some extent tumor invasiveness in
PA(Chacko et al. 2010; Noh, et al. 2009; Prevedello, et al. 2005). However, our current
understanding of all of the factors involved in tumour regrowth, invasiveness and recurrence is
lacking.
There have been some recent attempts to identify other factors that may aid in the prediction of
growth patterns and recurrence rates for PAs. For instance, preoperative PA volume has been
shown to correlate to the presence of residual tumor(Jain, et al. 2008; Ringstad, et al. 2012;
Shrestha, et al. 2012). A few studies have shown a positive correlation between residual PA
growth and the MIB-1 LI (Ekramullah, et al. 1996; Honegger, et al. 2003; Hsu, et al. 2010;
Kawamoto, et al. 1995; Saeger, et al. 2008; Tanaka et al. 2003). However, overall there is limited
38
understanding of whether preoperative tumor volume and growth rate across serial radiological
examinations is predictive of postoperative growth rate of a residual PA.
In our study, we examined the association between pre- and postoperative PA growth
characteristics in association to patient demographics, MRI parameters and histopathological
factors that may predict growth rate of PA both pre- and postoperatively.
3.2 Materials and Methods
3.2.1 Patient Population
Following institutional research ethics approval, we retrospectively reviewed all patients who
had surgery for resection of a histologically proven PA between 1999-2011.All surgeries done
prior to 2004 were microscopic and thereafter purely endoscopic for the patient cohort included
in this study. All patients in this study have been reviewed and managed by our institutional
multidisciplinary clinic that includes neurosurgeons, neuro-endocrinologists and
neuropathologists. All patients routinely have a complete hormonal biochemical analysis prior to
surgery by our endocrinologist (SE), including dynamic studies when indicated by baseline
biochemical assessments.
3.2.2 Magnetic Resonance Imaging (MRI) Patients were required to have a minimum of two pre- and two postoperative MRI examinations
to allow for measurement of tumor volume doubling time (TVDT). Cases requiring urgent
surgery due to acute presentation with visual compromise were not included in this study. The
first preoperative MRI was the first recorded image and the second preoperative MRI was the
one taken immediately prior to surgery. A minimum of three months between the first and
second scan was required. The first postoperative image had to be at least three months after
surgery to allow for a reduction in postoperative changes related to scarring and inflammation,
and the last postoperative MRI was the last recorded image which was at least 12 months from
the first postoperative scan. Additionally, none of the patients had received medical therapy for
39
their PA, except for some prolactinomas that were medically refractory.
All patients included in this study underwent a sella/pituitary specific imaging protocol. Coronal
and sagittal sections of T1-weighted MR images, with and without contrast enhancement, were
used for the two preoperative scans. For postoperative assessment, coronal and sagittal sections
of T2-weighted MR images were used, as T2 scans were more sensitive than T1 at detecting
subtle postoperative changes. Slice thickness including MRI magnet strength were standardized.
For volumetric assessment, MR images were uploaded into ITK-SNAP(Insight ToolKit Snake
Automatic Partioning)(University of Pennsylvania, www.itksnap.org)(Yushkevich et al., 2006).
ITK-SNAP is free software available on-line that allows for segmentation and measurement of
target volumes on MRI and has been used by our group previously in several publications
(Hayhurst, et al. 2012; Lwu, et al. 2013)Ebinu et al., 2013). Once the MRIs are uploaded, the
tumors are manually contoured slice by slice and at the end a volume is generated along with a
3D reconstruction of the tumor(Figure 3.2.1). The reader is referred to (Yushkevich et al.,
2006)(www.itksnap.org) for a more thorough description of ITK-SNAP.
40
Figure 3.2.1: Screenshot of ITK-SNAP volumetric software displaying tumor segmentation
in axial, sagittal, and coronal planes. Target tumor volumes were identified and manually
contoured slice-by-slice and a subsequent tumor volume was generated along with a 3D
reconstruction of tumor (bottom left pane)
3.2.3 Growth Pattern and MRI Characteristics
The pattern and direction of tumor growth was recorded as superior, inferior, anterior, posterior
and lateral relative to the sellar fossa. Many tumors exhibited a combination of these growth
patterns. Superior growth was noted if the tumor extended into the suprasellar cistern and/or
compressed the optic chiasm as observed in the coronal plane. Additionally, a line was drawn in
the sagittal plane extending from the posterior clinoid to the tuberculum and if tumor grew above
this line, it was noted as superior growth. Inferior growth was recorded as tumor that appeared
to be eroding through the sellar floor and entering into the sphenoid sinus. Anterior growth was
tumor that extended over the planum sphenoidale and in some cases extended beneath the
inferior surface of the frontal lobes. Posterior growth was observed when the tumor extended
into the interpeduncular cistern/prepontine area and/or compressed the brainstem. Lateral
growth and extension towards the CS was defined according to Knosp criteria for CS
41
extension(Knosp et al., 1993). Knosp criteria are divided into five grades (0-4) based on a series
of lines drawn through the supra- and intracavernous segments of the carotid arteries as observed
in mid-sella coronal sections. In our study, PA extending to at least the lateral boundary of the
intracavernous carotid segments (Knosp grade ≥2) was considered true PA extension into the CS
as previously correlated to surgical findings.
Figure 3.2.2: midsagital coronal diagram of pituitary fossa depicting different PA extension
patterns into the Cavernous Sinus. Knosp grading is as follows: Grade 0 = tumor does not
extend past medial wall of intracavernous carotid artery. Grade 1= tumor extends midway
between supra- and intracavernous segments of carotid artery (intercarotid line); Grade 2 =
tumor extends to lateral wall of ICA Grade 3 = tumor extends beyond lateral wall of ICA.
Grade 4 – complete encasement of ICA by PA. Knosp Grade 2+ definition of CSI in my
cohort.
42
Figure 3.2.3: Recorded PA growth patterns in (A) the coronal plane and (B) the sagittal
plane as depicted in Monsalves et al., 2014. Inferior growth into sphenoid sinus (erosion of
sphenoid bone by PA), superior growth towards optic chiasm, lateral growth towards and into
CS (Knosp grading), anterior growth above the planum, and posterior growth towards
brainstem were noted.
43
As an alternative, we used an in-house UHN method for determining CS invasion as proposed by
Dr W Kucharczyk(Figure X). In this model, the CS is invaded if either of the following criteria
are met: a) at least 50% of the carotid artery is encircled by tumor, b) the lateral wall of the CS is
displaced, and/or c) there is increased amount of tissue between the lateral wall of the CS and the
carotid artery. This proposed model corresponds to approximately Knosp grade 3 or 4.
44
Figure 3.2.4: UHN classification of cavernous sinus invasion
The presence of a heterogeneous mass on T1 or a hyperintense signal on T2, suggesting either
cystic or hemorrhagic changes was also recorded. Growth patterns and tumor morphology were
only determined preoperatively. Analysis of postoperative growth patterns was limited to
alteration of CS extension, as other growth patterns could not definitively be ascertained given
the postoperative scarring and changes seen as a consequence of nasal-septal flap reconstruction.
An increase in size of image characteristics from the very first postoperative scan to a scan one
year apart was taken as a sign of tumor and not postoperative changes/scar. Furthermore, we
ensured that the MRIs were reviewed by three independent observers: our senior neuro-
radiologist (WK), first author (EM) and neurosurgeon (BG) and an interobserver consistency
45
score was recorded to ensure no bias in review of imaging.
3.2.4 Tumor Volumetry and Growth Rate
The TVDT was calculated using the previously established formula for tumor growth rate
measurements: TVDT= t*log22/log2(Vf/V0), where t is the time interval between the first and
second MRI examinations, V0 is the initial tumor volume, and Vf is the final tumor
volume(Nakaguchi et al., 1999). TVDT was calculated both for the pre- and postoperative MRIs.
To minimize measurement error, postoperative residual were grouped into one of 3 categories: 1.
Growth (a ≥25% increase in volume), 2.Reduction (≥25% reduction in volume), or 3.Stability
(±24.9% change in volume).
For the subgroup of patients who had more than two pre and two post-operative images, the
TVDT was calculated for all available imaging to ensure the rate was not significantly different.
We had compared the TVDT at different intervals and since there was no significant variability
we only reported the TVDT for two time points to standardize the results.
3.2.5 Correlating Tumor Growth Rate to MIB-1, FGFR4, P27 Age,
Sex, Functional Status, and Histopathology of PA Each patient in our cohort was reviewed in a systematic manner by two expert endocrine
pathologists (Drs. Ozgur Mete and Sylvia L. Asa) at the time of initial diagnostic assessment.
The status of the MIB-1 labelling index, p27 and N-terminally truncated FGFR4 were extracted
from standardized diagnostic synoptic pathology reports. The MIB-1 LI was assessed in the hot
spot regions by counting preferably 1000 tumor cells; however, given the nature of some
specimens less than 1000 tumor cells were assessed for this evaluation. Cytoplasmic positivity
for FGFR4 was considered to be positive. The extent or the intensity of staining patterns were
not scored. The loss of p27 or positivity for p27 is recorded. The extent of positive nuclei was
not scored.
Correlations were made between preoperative TVDT, patient age, gender, functional status,
histopathology, and PA growth patterns. The preoperative TVDT was also correlated to the
postoperative TVDT. PA were analysed in multiple different ways: based on the entire group,
46
stratified based on functional status (clinically functioning vs. non-functioning) and stratified
based on histopathological subtype.
3.2.6 Statistical Analysis
A Cronbach’s alpha was generated for PA volumes in order to determine interobserver
reliability. Correlations were made using Pearson’s correlation coefficient and Spearman rank
coefficient. Mean comparisons were made using t-tests or ANOVAs. Categorical variables were
examined using Chi-square, Fisher’s Exact test, and Odds Ratios (OR). All statistical
calculations were performed using SPSS v.20.0 (Chicago, IL). Significance values that were
p<0.05 were considered statistically significant.
3.3 Results
3.3.1 Clinical Demographics and Radiological Characteristics
We reviewed all patients treated between 1999 and 2011 at our institution and identified 500
consecutive patients who had complete follow-up data to be analyzed for this study. Of the 500
we identified 153 patients that satisfied our inclusion criteria. Of the 153 patients, 82(53.6%)
were female and 71(46.4%) were male. The mean±SD age was 53±15years(range=25-87). The
preoperative imaging interval was variable and to eliminate some variability, cases with at least
three month imaging intervals were selected. The mean preoperative interscan imaging interval
was 17.3 months (range 3-113.9 months). Postoperative imaging intervals were more uniform as
patients were followed closely by our multidisciplinary PA clinic with regular scans at 3, 6, 12,
24 etc months. The mean postoperative PA interscan interval was 29.2 months (range=12-84
months). All recorded PA were considered macroadenoma(diameter >1cm). PA histological and
growth characteristics are listed in tables 3.3.1 and 3.3.2.
47
Table 3.3.1: Pituitary adenoma histological subtypes. All major subtypes of PA represented in my cohort. Nonfunctioning gonadotrophs were the most common as expected. Plurihormonal PA were defined as those secreting two or more hormones.
Subtype Number(%)
Gonadotroph 63(41.2)
Null Cell 24(15.7)
Prolactinoma* 10(6.5)
Somatotroph 16(10.5)
Corticotroph
Silent 26(17.0)
5(19.2)
Plurihormonal 13(8.5)
Thyrotroph 1(0.7)
Total 153(100)
*prolactinomas were medically refractory
Table 3.3.2: Characteristic growth patterns in pituitary adenoma. Most PA presented with superior extension towards the optic chiasm and this was the most frequent indication for surgical referral. A significant proportion also extended into the CS. All included PA were macroadenomas (≥1 cm in diameter)
Growth Pattern Number
Anterior 22(14.4%)
Posterior 46(30.1%)
Suprasellar 123(80.4%)
Sphenoid Sinus 37(24.2%)
Cavernous Sinus (Knosp≥2) 78(51%)
Total 153(100%)
3.3.2Preoperative PA Volume and Tumor Growth Rate (TVDT) Interobserver consistency for measuring tumor volume on ITK-SNAP was measured and the
respective alpha levels for the first preoperative, second preoperative, first postoperative, and
second postoperative scans were: α=0.996, 0.997, 0.914, 0.909. An alpha level above 0.7 is
generally considered acceptable as a measure of good internal consistency(A. Christmann, 2006).
48
A subset of patients had multiple preoperative and postoperative scans.
The mean±SD preoperative TVDT for the study population was calculated to be 1147±870 days
(range=60-3478days), with functioning tumors demonstrating a significantly shorter TVDT at
747±564days compared with clinically non-functioning PA at1334±926 days (p=0.0001). The
mean±SD TVDT based on histopathological subtypes were null cell 1579±1235, gonadotrophs
1097±668, prolactinomas 1222±1223, somatotrophs 895±612, corticotrophs2215±2098, and
plurihormonal tumors were, 941±502 days (Figure 3.3.1).The presence of a cyst/haemorrhage
was evident in 22% of the cohort. However, the cystic/haemorrhagic components remained small
(5-20% of tumor volume) and did not grow over time.
Figure 3.3.1: Growth rates (TVDT) of different PA histological subtypes included in cohort
The preoperative TVDT was positively correlated to age(r=0.361; p<0.0001). In addition, a
shorter mean TVDT was significantly associated with suprasellar extension in somatotrophs (330
vs 1146 days, p=0.003) and the presence of a cyst/hemmorhage in both somatotrophs(32%
cystic/hemorrhagic)(307 vs 1002 days, p=.004) and null cell tumors(33%
49
cystic/hemorrhagic)(831 vs 1927 days, p=0.01). The preoperative TVDT was also inversely
correlated to the MIB-1 LI(r=-0.256, p=0.005). Gonadotrophs were the only histological
subtype to display a significantly shorter mean TVDT with both FGFR4 positivity(983 days vs
1379 days, p=0.047) and p27 negativity(559 vs 1386 days, p=0.007)(Table 3.3.3). No other
factors were significantly related to preoperative TVDT.
Table 3.3.3: Factors associated with a shorter preoperative TVDT identified by whole group analysis,
stratification by histological subtype, and stratification by hormonal status.
Factor Entire cohort Tumor
Histotype
Hormonal Status
Age P<0.001 Gonadotrophs** Functioning***,
Nonfunctioning***
S Growth NS Somatotrophs** NS
C/H Component NS Somatotrophs**,
Null Cell*
NS
MIB-1 LI P<0.01 NS Nonfunctioning*
Negative p27 status NS Gonadotrophs** Nonfunctioning*
Positive FGFR4 status NS Gonadotrophs* NS
*p<0.05 NS=Nonsignificant; S = Suprasellar; C/H = Cystic/Hemorrhagic
**p<0.01
***p<0.001
3.3.3 Association of Residual Volume with Growth Patterns and
Patient Characteristics Fifty three cases (34.6%) exhibited postoperative residual volumes while the remaining 100 did
not. Of the 53 residuals, the majority (80.8%) were clinically nonfunctioning. The presence of a
residual volume was associated with older patient age (57 vs 51, p=0.038). The presence of
residual was also associated with several preoperative growth patterns: 28.3% of residual had
anterior (p=0.001)(OR=5.24), 54.7% had posterior (p<0.0001)(OR=5.90), 94.2% had suprasellar
(p=.002)(OR=5.73), and 71.7% had CS extension (p<0.0001)(OR=3.80), 42.4% of which
continued to have definitive extension into the CS (p<0.0001)(OR=14.6) post-operatively (Table
50
3.3.4). The presence of a residual or its subsequent growth behavior (growth, stability,
regression) was not associated with any other factor.
Table 3.3.4: Factors associated with residual volume following surgery
Factors Significance
Age P=0.038
Cavernous Sinus Invasion P<0.0001
Suprasellar Extension P=0.002
Posterior Growth P<0.0001
Anterior Growth P=0.001
3.3.4 Postoperative Tumor Volume and Tumor Growth Rate
(TVDT)
There were a total of 23(43.3%) tumors with postoperative residuals that demonstrated growth,
with 17(73.9%) nonfunctioning and 6(26%) functioning PA. The mean±SD age was 51±12;
13(56.5%) females and 10(43.5%) males. The overall mean±SD postoperative TVDT for both
the functioning and nonfunctioning PA was 1164±615 days with a mean radiological follow-up
period of 29.2 months(range=12-84 months). The mean preoperative TVDT of PA that
demonstrated postoperative growth was 1165±178 days. This was not significantly different
from postoperative TVDT (p=0.106). The TVDT for nonfunctioning PA was 1272±637 and
857±461 days for functioning PA. The difference between functioning and non-functioning
TVDT was not significant (p=0.089).
The postoperative TVDT was positively correlated to preoperative TVDT(r= 0.497, p=0.026)
and to patient age in non-functioning(r=0.596, p=0.015) but not in functioning PA. A shorter
mean TVDT was associated with female gender in functioning PA only(689 vs 1695 days,
p=0.017) but not in any particular subtype. No other factors were associated with postoperative
TVDT. However, due to the fact that there are only 6 functioning tumor postoperatively, these
results should be interpreted with caution.
51
3.3.5 Correlation of Growth Rate with Expression of MIB-1,
FGFR4, and p27 92.6% of cases had sufficient material for detailed examination of MIB-1 LI. The overall mean
MIB-1 LI was 3.19%(range=7.5). Clinically functioning PA had a mean MIB-1 LI of 3.88% and
non-functioning tumors had a mean LI of 2.86% and this difference was statistically
significant(p=0.007). In terms of subtype, the mean MIB-1 LI for gonadotroph, null cell,
prolactinoma, somatotroph, corticotroph, and plurihormonaltumors were 2.53%, 3.07%, 3.79%,
2.72%, 3.99%, 5.1%, respectively. The single thyrotroph adenoma in the series had a MIB-1 LI
of 5.0%. Furthermore, the mean MIB-1 LI was marginally but nonsignificantly greater for those
residual masses that grew compared to those that remained stable or regressed (3.15%, 2.71%,
2.66%, respectively).
The MIB-1 was positively correlated with age (r=0.215, p=0.011) and was significantly higherin
the female gender, specifically inthose with thecorticotroph subtype(5.07%vs 1.97%, p<0.0001).
The MIB-1 was also higher in gonadotrophs that exhibited anterior extension (3.29%vs 2.41%,
p=0.025) and in null cell tumors with CS extension(3.90%vs 2.23%, p=0.016). In functioning
PA, a higher MIB-1 was associated with suprasellar(4.55%vs 3.07%, p=0.019) and posterior
extension(5.9%vs 3.9%, p=.019) as well as with the presence of hemorrhage or cystic change
(5.04%vs 3.49%, p=0.026).
FGFR4 positive PA (53.9% of cases) wereassociated with patients of older age (60 vs 48 years
old, p=.016) and 1.83 times more likely to be non-functioning than functioning PA(p<0.0001).
FGFR4 positive PA were also 2.16 times more likelyto have a cystic change or a hemorrhagic
component(p=0.05).
In contrast to FGFR4, p27 negative PA (5.2% of cases)were associated with patients of younger
age(44 vs 57 years old, p=.003) and 4.13 more likely to be functioningthan
nonfunctioningtumors(p=0.042). The p27 expression profile was inversely related to the MIB-
1(r=-0.187, p=.032) but not related to FGFR4.
52
Figure 3.3.2: Schematic summary of Aim 1
53
3.4 Discussion
To gain insight into PA growth behavior, we opted to examine the preoperative growth direction
and rate of PA and correlate it to postoperative growth rate . To date, attempting to correlate PA
growth rate pre- and postoperatively has never been reported in the literature. The postoperative
TVDT has, however, been examined in relation to PA in a limited number of studies, primarily
in residual NFPA(Ekramullah et al. 1996; Tanaka et al. 2003). Our cohort which included both
functioning PA and NFPA revealed that the preoperative and postoperative growth rate as
assessed by TVDT were correlated. Previous studies suggested that preoperative PA volumes
were able to predict postoperative outcome but growth rates were not assessed in these
studies(Jain et al. 2008; Shrestha et al. 2012). Our data suggests that an understanding of
preoperative PA growth dynamics will provide insight into how these tumors will behave
postoperatively in terms of radiological growth which will help enable the selection of
appropriate strategies in a more timely fashion and may even be able to preempt future PA
recurrence.
Additionally, our data suggest that the growth rate of PAs are influenced by various patient and
tumor-specific characteristics including the age and sex of the patient, the specific histologic
subtype of PA, it’s hormonal activity, it’s preponderance for different growth directions relative
to the pituitary fossa. Some of these factors have been reported in connection to PA growth
behavior previously. For example, it has been shown that PA that invade into the CS or
sphenoid sinus are associated with a more aggressive phenotype with increased likelihood of
postoperative residual and recurrence(Trouillas, et al. 2013). The current study aimed to provide
a more complete growth profile of PA and use this information to create a model that may help
predict PA residual, growth, and recurrence.
The immunohistological profile has also been correlated to growth rate in our study. This
includes the MIB-1 LI, a marker of cell proliferation, which has been previously correlated to
postoperative PA TVDT in some studies where a higher MIB-1 is associated with a faster
(lower) TVDT(Hsu et al. 2010; Tanaka et al. 2003). In our study, the MIB-1 was inversely
correlated with TVDT consistent with what other have reported but our correlation was only seen
with respect to preoperative TVDT and not to postoperative TVDT. The status of biomarkers,
54
including ptd-FGFR4 and the early cell cycle inhibitor p27 were also included in our study as
gleaned from institutional synoptic PA pathology reports. ptd-FGFR4 is an N-terminally
truncated variant of normal FGFR4; it is constitituitively active and is localized to the cytoplasm.
It has been previously reported to be prevalent in 40-50% of all PA, with a specific tendency for
expression in null cell and gonadotroph PA(Ezzat, et al. 2004; Ezzat et al. 2002). ptd-FGFR4 has
been correlated previously to PA invasiveness(Morita, et al. 2008). Although positivity for ptd-
FGFR4 was seen in approximately 50% of our cohort primarily in null cell and gonadotroph PA,
it was not associated with growth or invasiveness. The status of p27 has been shown previously
to be positive in most PA suggesting that the early cell cycle is not adversely affected in the
majority of cases. Additionally, consistent with other studies(Hewedi, et al. 2011; Zhao 1999),
p27 negativity was not shown to correlate with PA growth or recurrence in our cohort in
general. However, gonadotrophs that were both p27 negative and FGFR4 positive were
associated with a faster rate of growth implicating the prognostic value of these biomarkers in
this subtype specifically . The concurrent examination of p27 and FGFR4 has not reported
previously in the PA literature.
55
CHAPTER 4
Preclinical Model of PA for Interogating Therapeutic Strategies:
Role of mTOR inhibitors
4.1 Introduction
Identification of molecular mechanisms that are involved in PA initiation and progression are not
clearly defined. Some molecular initiators of pituitary tumorigenesis have been identified. For
example, somatic mutations resulting in G-protein abnormalities and subsequent increases in
cAMP levels have been identified in a subset of GH-PA(Vallar, et al. 1987). RAS mutations also
occur but these are limited to the rare metastatic pituitary carcinoma(Pei, et al. 1994). Other more
common molecular events that modulate PA growth in general have not been well characterized.
Modulation of growth factors and their receptors are necessary for normal pituitary growth and
function. Dysregulation of these factors may lead to abnormal pituitary growth and tumor
development. For example, FGF-2 mRNA has been shown to be overexpressed in PA((Ezzat, et
al. 1995)). Recent interest has focused on FGFR4, which is a 110 kDa transmembrane kinase
that is involved in mitogenesis and angiogenesis(Abbass, et al. 1997). A 60kDa N-terminally
truncated pituitary-tumor derived variant of FGFR4 (ptd-FGFR4)has been identified in 40-50%
of PA but not in normal pituitary(Ezzat et al. 2002). Preclinical evidence suggests that ptd-
FGFR4 but not full-length FGFR4 confers oncogenic properties to pituitary tumor cells. In
human PA tissue, positivity for ptd-FGFR4 has been associated with increased PA
aggressiveness(Ezzat et al. 2004).
Germline allelic alterations in FGFR4 have also been identified in PA. An adenine to guanine
SNP at codon 388 of the FGFR4 gene results in a glycine to arginine substitution in the
transmembrane domain of the FGFR4 protein (FGFR4-G388R)((Bange et al. 2002). This SNP
results in increased growth and growth hormone production in GH4 cell lines and human GH PA
tissue(Tateno et al. 2011). In ACTH PA, the FGFR4-R388 variant is associated with the
56
presence of silent ACTH macroadenomas while wild-type FGFR4 (G388) is associated with
ACTH production(Nakano-Tateno et al. 2014).
Downstream molecular mechanisms from FGFR4 have not been examined in detail. Ezzat et al.,
2006 have reported that ptd-FGFR4 promotes oncogenic transformation by disturbing the
NCAM/b-catenin complex thereby disrupting normal cell to cell interactions. In their study, an
FGFR4 inhibitor was used to restore the NCAM complex and reduce cell and tumor
growth(Ezzat et al. 2006). In the FGFR4-G388R SNP src and STAT3 serine phosphorylation
was shown to promote growth while STAT3 tyrosine phosphorylation modulated hormone
secretion(Tateno et al. 2011).
Other signaling mechanisms in PA need to be identified. Recent interest has focused on the
PI3K/mTOR pathway which is a signaling cascade important for maintaining growth and
homeostasis within the cell (see chapter 3). It is ubiquitously overexpressed in cancer and has
been shown to be overexpressed in PA, particularly in NFPA and GH-PA(Musat et al. 2005;
Sajjad et al. 2013). It has been demonstrated that the mTOR inhbitors are effective in PA cells
by reducing cell viability and proliferation and also arresting cells early in the cell
cycle(Gorshtein et al. 2009; Zatelli et al. 2010).
The PI3K/mTOR pathway is frequently initiated by activation of RTKs which may include
FGFR4. The aim of this study is to determine a differential expression pattern of the mTOR
pathway between FGFR4-G388 and R388 tumors and how this expression may influence
sensitivity to mTOR inhibition using RAD001 and affect tumor growth rate and hormone levels.
57
4.2 Material and Methods
4.2.1 Cell lines and cultures
As there are no human-derived hormone-producing pituitary cell lines, we used rat pituitary GH4
mammosomatotroph cells. These cells were obtained from Dr Ezzat's lab from his senior research
fellow, Dr Toru Tateno, and propagated in Ham F10 medium 12.5% horse serum (HS)(Sigma,
Oakville, ON) and 2.5% fetal bovine serum (FBS; Sigma, Oakville, ON) 2mM glutamine, 100
IU/ml penicillin, 100 ug/ml streptomycin (37C, 95% humidity, 5% C02 atmosphere incubation).
The cell lines were maintained as described previously by Tateno et al., 2011 (Tateno et al. 2011)
4.2.2 Plasmids and Transfections
Plasmids encoding human prototypic FGFR4 (G388) or the polymorphic form FGFR4- R388
were generated and stably transfected into GH4 as previously described (Tateno et al. 2011).
Plasmids were obtained in collaboration with Dr T Tateno (PDF) in my co- supervisors lab (Dr S
Ezzat ). Construct fidelity was confirmed by DNA sequencing after introduction into pcDNA 3.1
as described in Tateno et al., 2011. Cells were transfected using Lipofectamine 2000 (Life
Technologies, Rockville, MD) according to the manufacturer’s instructions. Stable clones were
selected using neomycin (G418) at a concentration of 0.7 ug/ml and maintained in this
conditioned media throughout the experiments
4.2.3 Animals
All animal experiments and protocols were approved by the University Health Network (UHN)
Animal Care and Use Committee. Immunocomprimised, non-obesesevere combined
immunodeficiency (NOD-SCID) male mice were purchased from Jackson Laboratory
(Sacramento, CA,USA) and used for intracranial studies. Cell lines used were GH4 cells, used as
58
parental control, G388, and R388 representing the SNP variant FGRF4 expressing cells. GH4
empty vector (PC) transfected cells were also used as controls.
4.2.4 In vivo Tumor Models
Six week old NOD-SCID mice were anesthetized with intraperitoneal injections of Avertin
[1.25% solution, 0.2 mL/10 g body weight approximately 0.45 mL for 6 week old mice]. Mice
were used at age 6-8 weeks. Intracranial xenografts were generated as described previously (Burrell et
al., 2014). Briefly, the head was cleaned with 70% ethanol, the scalp was opened with a 1 cm
incision midline and the usual s tereo tac t ic landmarks were used for injections. A burr hole
was created in the skull with a dental drill and cells were injected using a Hamilton syringe. 5*105
were injected using a hamilton syringe via stereotactic coordinates 3mm deep into the frontal cortex on
the right side of the brain. All animals were injected on the same day, with assistance of senior research
associate Dr Shahrzad Jalali. In order to establish the model that allowed for tumor growth and
sufficient time for growth to permit MRI examination and drug studies we did a serial dilution of
105 and 106 cells injected intracranially to optimize the best model for PA xenografts as previous
such studies do not exist.
Injection of 106 cells resulted in rapid tumor development precluding imaging and
detailed analysis of growth rate studies or response to therapy for alteration in tumor growth.
Injection of 105 cells resulted in tumors with very long growth latency periods. We decided to
inject 5*105 cells for our next set of experiments. Each mouse received one injection containing
5*105 cells suspended in 10 ul PBS. Mice were monitored closely by serial magnetic resonance
imaging (MRI) starting at week 2 and sacrificed at the end of drug treatment period or sooner if
signs of illness developed, typically characterized by hunched or abnormal posture, decreased
movement, lethargy, paralysis or weight loss.
4.2.5 Magnetic Resonance Imaging (MRI) MRI was performed with a 7 Tesla Biospec 70/30 (Bruker Corporation), using the B- GA12
gRTient coil insert and 7.2 cm inner diameter linearly polarized volume resonator coil for
59
radiofrequency (RF) transmission as detailed previously (Chung et al., 2012).
The protocol provided a stack of transverse 2D slices with shared geometric prescription (16 x
16 mm field-of-view or FOV) testing multiple contrast mechanisms, as follows: i) T2-
weighted-RARE anatomical imaging (echo time or TE=72 ms; repetition time or TR=5000
ms; rapid acquisition relaxation enhancement or RARE factor=16; readout bandwidth or
RBW=50 kHz; 25×125×500 μm voxels) for; 1 min and 20 sec; ii) Contrast enhanced T1-
weighted RARE (TE=8 ms; TR=1200 ms, RARE factor=4; RBW=81.5 kHz; 125×125×500
µm voxels; start imaging at 5 min post-contrast) for 1 min and 20 sec.
4.2.6 Treatment Schedule.
After intracranial injection of cells, the mice were monitored by weekly MRI. After tumor
development was noted, a s de t e rmined by T 2 MRI, mice were randomly stratified into
control or treatment arms, ensuring approximate equal tumor volume distrbution into each arm.
Treatment consisted of daily intraperiotoneal I.P. injections of 5mg/kg RAD001 (Cedarlane,
Burlington, ON) or DMSO dissolved in ddH20 (control). Treatment lasted for three weeks at
which point the mice were sacrificed and their brains extracted and placed into formalin for
histological analysis. For 5 mice in each of the three cell groups, tumor was isolated using a
surgical microscope and preserved in liquid nitrogen for western blot analysis (see below for
details below). MRI was performed weekly on all animals while on treatment to allow tumor
volume analsysis.
4.2.7 Tumor Size Measurement Serial MR images were used to establish the tumor volumes and growth rates. ITK-SNAP was
used to analyze the MRI images. Tumor region of interest (ROI) on T2-weighted (T2w)
images was manually defined and used to measure the tumor volume. Growth rate studies using
both T1 and T2 images were correlated and found to show good correlation. Interobserver
correlat ions were generated s tat is t ical l y using Cronbach 's alpha and an
excel lent rel iabi l i ty score was achieved (α =0.89) Furthermore, T1 and T2
images were compared to di fferent iate tumor from post-surgical changes.
60
These studies were done by two independent examiners (EM) and (SJ). Since, T2 MR
images paralleled the T1 images closely, were primarily used for calculating tumor volume and
growth rate in this study for longitudinal studies using T2 images.
4.2.8 Immunohistochemisty
Immunostaining was conducted on formalin-fixed paraffin embedded on lesional tissue in brain
sections of control and treated animals. The sections were deparaffinized in xylene and
rehydrated in graded ethanol and rinsed in dH20. Heat-induced antigen retrieval was used by
pressure cooking the sections for 20 minutes in Antigen Retrieval Solution (low pH). Slides
were incubated with primary antibodies with appropriate conditions. Detection was performed
using vectastain ABC reagent and DAB chromagen (Vector Labs, Burlingame, CA). Slides were
counterstained in Meyer’s Haematoxylin for 5 minutes, dehydrated with ethanol (70%, 95%,
100%) and coverslipped. Primary antibodies were MIB-5/Ki-67 (Dako), p-S6 (Cell Signalling),
p-4EBP1 (Cell Signalling), growth hormone and prolactin antibodies were kindly provided by
the Dr. Ezzat’s lab. All slides were scanned using Zeiss Mirax Scanner and analysed using
Mirax Viewer software.
4.2.9 Western Blot Analysis
To quantify protein expression in tumor tissue, brains were extracted, tumors immediately
isolated and placed in liquid nitrogen for snap-freezing. Lysis buffer (0.5% sodium
deoxycholate, 0.1% sodium dodecyl sulfate, 1% Nonidet P-40 and 1X PBS) containing
proteinase and phosphatase inhibitors was added to tumors which were then homogenized using
a mechanical tissue homogenizer. Total cell lysates were incubated on ice for 30 mins followed
by micro-centrifugation at 10 000g for 10 min at 4°C, and the supernatent was collected. Protein
concentration was determined by BCA (bicinchonic) assay as per manufacturers specifications
(Pierce Chemicals Co., Rockford, IL). A total of 40ug was separated on a 10% SDS
polyacrylimide gel electrophoresis (SDS PAGE) and blotted onto nitrocellulose membranes (Bio
RAD, Hercules CA) using a semi-dry turbo transfer apparatus (Bio-Rad, Hercules, CA). After
blocking the membrane with PBS (or TBS) containing 5% nonfat dry milk and 0.05% Tween 20
61
(Sigma-Aldrich) for 1 hour the membranes were incubated with primary antibodies overnight at
4 degrees. The following antibodies were used: 1:1000 p-4EBP1 (Cell Signaling); 1:2000 p-S6
(Cell Signaling); 1: 1000 Growth Hormone (Dr. Ezzat lab); 1:1000 Prolactin (Dr Ezzat lab);
1:500 ki-67 (Dako). Membranes were washed and later incubated with IRDye secondary
antibodies against the species the primary antibody was derived from (Bio-RAD, Hercules, CA).
Protein bands were detected using and quantified using LI COR Software.
4.2.10 ELISA Mouse whole blood was extracted from the right ventricle of the heart and placed in Microvette
CB 300z serum extraction tubes. Whole blood was centrifuged for 5 minutes at 5000 RPM at
4C. Serum was extracted and stored at -80C. ELISA was used to measure serum IGF-1
(Quantikine) levels according to manufacturer’s protocol.
4.2.11 Statistical Analysis Nonparametric tests (Kruskall-Wallis, Mann Whitney U) were used to examine median
differences in tumor growth rates, IGF-1 levels. H-Scoring was used to semi-quantify IHC data.
Intensity of Western Blot protein bands were quantified using LICOR software.
4.3 Results
4.3.1 In vivo tumor formation assay
To examine the biologic impact of the FGFR4 SNP, we introduced FGFR4-G388 or FGFR4-
R388 into pituitary adenoma GH4 cells. Initially, we injected 106 GH4 cells but we noted rapid
tumor growth that precluded our ability to track tumor growth rate so 5*105 cells were used in
later experiments. Targeting the frontal intraparenchymal region was chosen as the model to
investigate for the purposes of my masters thesis. Evidence of a proliferative GH4 tumor co-
expressing GH and PRL was noted (Figure 4.3.1) in pc, G388, and R388 mice. The R388 tumors
62
expressed more GH relative to the G388 and pc tumors which is in line with what has been
reported previously using these cells (Tateno et al. 2011)
Figure 4.3.1: The FGFR4-R388 polymorphism deregulates pituitary hormone production intracranial
xenograft mouse tumor. Growth hormone (GH), prolactin (PRL), and Ki-67 protein expression were
examined in pituitary GH4 mammosomatotroph intracranial tumors stably expressing empty vector(pc)
(control), the prototypic FGFR4 receptor (G388), or the human polymorphic FGFR4 variant (R388). G388
produce a proliferating tumor with enhanced PRL production and diminshed GH production whereas injection of
R388 cells produce a proliferating tumor with diminished PRL expression and enhanced GH expression.
(n=5)(magnification 40X; scale: 50um)
4.3.2 MRI analysis
We first established the growth rate of the PA grown as frontal ic xenografts. MRI examination
of tumor growth was performed using T1W MR images with gadolinium enhancement (Figure
4.3.2). However, given that T1W images with contrast on each animal every week is associated
with high number of animal loss, we opted for T2W MR images without contrast for subsequent
measurements of tumor volume and growth rate.Correlative studies were conducted between
63
T2W and T1W images and were found to have a good correlation. Tumor development was
noted on MRI after 4 weeks in all 3 tumor groups (Figure 4.3.2
Figure 4.3.2: Injection of GH4 cells into the frontal intraparenchymal area in NOD-SCID mice results in
radiologically determined tumor growth. T1W MR images with gadolinium contrast enhancement in the
axial plane demonstrating tumor development three weeks post intracranial injection of 5*105 GH4 cell
stably transfected with either pc, G388, and R388 mice.
4.3.3 Tumor volume and growth rate.
Weekly MR images allowed for an examination of tumor growth rate. Tumor volumes were
found to increase exponentially over time with pc tumors exhibiting a significantly longer
latency for tumor development relative to the G388 or R388 tumors (p=0.034)(figure 4.3.3A)
and was also associated with a slower growth rate (increased TVDT) (Figure 4.3.3B)
64
A
Figure 4.3.3: FGFR4-G388R cells result in intracranial tumors that exhibit short latency periods
followed by rapid tumor expansion A. Analysis of serial MRI in axial plane using ITK-SNAP for
volumetric assessment revealed revealed that intracranial xenograft mouse models of GH4
mammosomatotroph tumors transfected with either G388 or R388 resulted in significantly larger tumors
than those transfected with PC at week 4. B. Volumetric data from the intracranial GH4 tumors transfected
with either pc, G388, or R388 were incorporated into the formula for estimating tumor volume doubling time
(TVDT) and a growth rate was generated (in days). pc had a significantly longer TVDT (slower growth
rate)than G388 or R388 tumors. *p<0.05. N=8 in each group (pc, G388, R388)
4.3.4 mTOR pathway expression
We next wanted to determine mTOR pathway expression in our GH4 tumors using the
immediate downstream effectors, S6 and 4EBP1 as a readout for mTOR expression. There was
abundant staining of both p-S6 and p-4EBP1 particularly in the G388 and R388 tumors(Figure
4.3.4) suggesting that our model is ideal for mTOR inhibition therapy
65
Figure 4.3.4: Intracranial GH4 tumors transfected with either prototypic FGFR4 (G388) or
polymorphic FGFR4 (R388) exhibit abundant mTOR signaling. Immunostaining of p-S6 and p-
4EBP1 in intracranial xenograft mouse GH4 tumors transfected with either G388 or R388 reveal
widespread expression. p-S6 and p-4EBP1 are the immediate downstream effectors of mTOR and are
therefore the most typically used readouts for mTOR (n=5) (magnification: 40X; scale: 50um)
4.3.5 mTOR Inhibition Treatment Using RAD001
Validation of our initial experiments was conducted. A total of 20 mice were randomly stratified
into the RAD001 (mTOR inhibitor) treatment arm (n=10) or the DMSO control arm (n=10) once
tumor development was noted. This procedure was repeated sequentially for the mice GH4
xenografts transfected with empty vector (PC), FGFR4-G388, and FGFR4-R388.
4.3.6 The Growth Rate of GH4 Tumors A significant reduction in tumor volume and growth rate was noted in mice that were treated
with RAD001. This significant reduction was seen in PC and the G388 tumors and R388
tumors(Figure 4.3.5 a, b, c). The differences in growth rates for untreated tumors in the three
groups did not differ significantly. However, the magnitude of the growth rate reduction was of
similar magnitude in all three cell groups: 2.1X, 2.2X, 2.4X in pc, G388, and R388 tumors, and
66
these differences were not statistically significant.
67
Figure 4.3.5: The mTOR inhibitor RAD001 decreases intracranial GH4 tumor volume regardless of FGFR4 status a) Intracranial GH4 tumor in control (DMSO) group exhibiting representative large, hemorrhagic tumors (top) as indicated on both hematoxylin and eosin (H&E) stained coronal formalin fixed paraffin embedded (FFPE) brain tissue sections and axial T2W MRI. Intracranial GH4 tumors treated with 5mg/kg RAD001 were typically smaller than in control as evident in representative bottom coronal H&E FFPE section and T2W MRI. B) The growth rates for intracranial GH4 tumors transfected with either pc, G388 or R388 as determined by serial MRI and reported as TVDT are displayed as boxplot. The median growth rates are reported. The median growth rates in all three control groups were not significantly different from each other. However, the growth rates in all 3 groups decreased following RAD001 treatment although the magnitude of the decrease was similar in all groups.
4.3.7 mTOR Expression Patterns Following RAD001 Treatment
Following a examination of growth rate after treatment with RAD001, we determined mTOR
signalling patterns and found that there was a substantial decrease in p-S6 expression particularly
in the G388 and R388 tumors(Figure 4.3.6a,b,c). There appeared to be no treatment effects with
p-4EBP1 as the expression pattern of this protein did not change following treatment
(Figure4.3.7a,b,c).
Figure 4.3.6 (below): RAD001 decreases p-S6 expression in intracranial xenograft GH4 mouse tumor transfected with either prototypic FGFR4 (G388) or polymorphic FGFR4 (R388)A) Immunohistochemical analysis of p-S6 expression in GH4 tumors transfected with either pc, G388 or R388 shows a decrease following treatment with RAD001 (n=5)(magnification =40X; scale=50um) B. Equal amounts of tumor lysates were resolved by SDS-PAGE and analyzed by immunoblotting with p-S6 (32 kDa) and Tubulin (50kDA). Quantification of protein was conducted using Li-Cor software and the intensity of the p-S6 band was compared to intensity of the loading control(tubulin). Subsequent results are reported in 'relative fluorescence units' and displayed in boxplot format. Median p-S6 band intensity was significantly attenuated in both G388 and R388 tumors following mTOR inhibition with RAD001(n=5).
68
69
Figure 4.3.7 : RAD001 has no effect on p-4EBP1 expression in intracranial GH4 tumors transfected with FGFR4-G388R A) Immunohistochemical analysis of p-4EBP1 expression in GH4 tumors transfected with either pc, G388 or R388 shows no change in expression following treatment with RAD001 (n=5)(magnification =20X; scale=50um) B. Equal amounts of tumor lysates were resolved by SDS-PAGE and analyzed by immunoblotting with p-4EBP1 (20 kDa) and Tubulin (50kDA). Quantification of protein was conducted using Li-Cor software and the intensity of the p-4EBP1 band was compared to intensity of the loading control(tubulin). Subsequent results are reported in 'relative fluorescence units' and displayed in boxplot format. Median p-4EBP1 band intensity was not significantly altered in either G388 or R388 tumors following mTOR inhibition with RAD001(n=5).
4.3.8 Hormone Expression Patterns Following RAD001
Treatment
The hormone expression profile of our GH4 tumors following RAD001 treatment was
examined next. Growth hormone is expressed in tumors with greatest intensity of
expression in R388 tumors( Figure 4.3.8 a,b,c). Expression significantly decrease in the
R388 group following inhibition with RAD001 (Figure 4.3.8.8c)
70
Figure 4.3.8: RAD001 decreases GH expression in intracranial GH4 xenograft tumors harboring the FGFR4 polymorphism (R388) A) Immunohistochemical analysis of GH expression in GH4 tumors transfected with either pc, G388 or R388. R388 tumors exhibit enhanced GH expression with decreased expression levels following treatment with RAD001 (n=5)(magnification =40X; scale=50um) B. Equal amounts of tumor lysates were resolved by SDS-PAGE and analyzed by immunoblotting with GH (25kDa) and Tubulin (50kDA). Quantification of protein was conducted using Li-Cor software and the intensity of the GH band was compared to intensity of the loading control(tubulin) band. Subsequent results were reported in 'relative fluorescence units' and displayed in boxplot format. Median GH protein band intensity was enhanced in R388 groups and demonstrated a significant reduction in GH protein levels following mTOR inhibition with RAD001(n=5).
4.3.9 Mouse Serum IGF-1 Levels Following RAD001
Treatment Serum measurements of IGF-1 (systemic hormone stimulated by GH) were analysed
following mTOR inhibition. Untreated control IGF-1 levels were significantly higher in
71
R388 relative to pc and G388 tumors. Following RAD001 treatment, all groups
experienced a decrease in IGF-1 levels but the decrease was only significant in the R388
group(Figure 4.3.9).
Figure 4.3.9:RAD001 decreases serum IGF-1 levels in intracranial xenograft GH4 tumors
harboring the FGFR4 polymorphism (R388) Circulating IGF-1 levels as a measure of GH4
tumor output are noted in pc, G388 and R388 mice. Median IGF-1 levels are enhanced in the R388
group and display a significant decrease in IGF-1 levels following mTOR inhibiton with RAD001in
the R388 group only(p=0.005).
72
Figure 4.3.10: Schematic summary of Aim 2
4.4 Discussion
Our preclinical in vivo mouse model of PA has enabled us to delineate some of the
pathophysiological mechanisms contributing to the PA disease process. Our target
pathway was the PI3K/Akt/mTOR pathway as this pathway has been shown to be
upregulated in PA both in human and animal PA cells at the in vitro level. Some in vivo
experiments examining mTOR pathway expression in subcutaneous PA models have also
been conducted. In our study, we opted to grow tumor cells in the intracranial
environment which, in our opinion, reasonably recapitulated the actual PA mileu.
Differential mTOR pathway expression was examined with respect to cells and tumors
transfected with either wild-type FGFR4 or the SNP variant of FGFR4. FGFRs are
important for normal pituitary growth and hormone modulation. Prototypic
FGFR4(G388) is a 110 KDa membrane-bound protein consisting of three extracellullar
Ig-like domain, a transmembrane domain, and an intracellular tyrosine kinase
domain(Bange et al. 2002) . In our mouse models, FGFR4 harbored a polymorphism at
codon 388 resulting in the transmembrane domain possesing an arginine residue instead
of wild-type glycine (R388). The presence of this SNP has been shown to correlate with
increased aggressive behavior in various cancers and in GH and ACTH secreting PA cell
lines has shown to phosphorylate src and the serine residue of STAT3 to increase cell
proliferation(Nakano-Tateno et al. 2014; Tateno et al. 2011). In the in vivo mouse
harboring the FGFR4 SNP we found increased expression of GH relative to wild-type
73
FGFR4 and empty vector controls (pc). R388 tumors were also associated with increased
serum IGF-1 levels compared to pc, and G388 tumors. G388 cells and tumors were
associated with higher PRL levels compared to pc, and R388 cells. The GH and PRL
expression patterns found in our tumor models have been reported previously in these
cells(Tateno et al. 2011). Downstream mTOR pathway expression as determined by p-S6
and p4EBP1 was also noted. Finally, positive Ki-67 staining indicated a proliferating
GH4 tumor model.
Once we established and characterized our intracranial model, we were able to target the
mTOR pathway using an mTOR inhibiting agent called RAD001. This agent has been
used previously in human and animal PA cells to reduce cell proliferation and viability
while arresting cells in the early stages of the cell cycle and downregulating
mTOR(Gorshtein et al. 2009; Zatelli et al. 2010). We did see a reduction in tumor growth
rate following mTOR inhibition. However, the magnitude of tumor inhibition was similar
in all groups (pc, G388, and R388). The decrease in tumor growth was associated with
lower expression in p-S6. One unexpected finding was the concommitant reduction in
GH levels in R388 tumors following treatment with RAD001. The reduction was also
noted in serum IGF-1 levels. These findings suggest that patients with a GH-secreting
tumor would benefit from RAD001 treatment. RAD001 has been shown by the current
study to be efficacious in reducing tumor growth rate regardless of the FGFR4 genotype.
74
Chapter 5
Discussion
There were two aims to my project, one was a clinical retrospective review of PA patients
with the goal of identifying factors that would aid in predicting radiologically determined
PA growth. The second aim was preclinical and sought to establish an intracranial mouse
xenograft model of PA and then to use that model to identify novel targeted therapies,
namely inhibitors targeting the mTOR pathway. Although these two aims were under the
umbrella of PA pathology, they were conducted independently from one another with no
interconnecting components. Therefore, the two aims will be discussed below
sequentially.
In our clinical retrospective study we have attempted to establish a correlation between
preoperative and postoperative growth rate for PA and investigate factors that may
predict growth rates of PA and residual postoperative growth rates of PA.
To date, only two studies report on postoperative TVDT for non-functioning adenomas
with 930±180 days in one study and 1836±3445 days in the other study (Ekramullah et
al., 1996; Tanaka et al., 2003). Another study included both functioning and non-
functioning tumors together and reported a postoperative TVDT of 2361±4449 days (Hsu
et al., 2010). Our recorded postoperative TVDT, which included both functioning and
nonfunctioning PA, was 976±382 days. Our study is the only one to date that attempts to
make a correlation between pre- and postoperative TVDT.
The discrepancy in TVDT values across studies may be due to differences in methods for
measuring tumor volume. For example, in previous studies, PA volume was calculated
using a formula based on tumor diameter. In our study, in contrast, a program was used to
manually contour the tumor borders which, in our opinion, is more accurate in capturing
differences in tumor volume due to the heterogeneity of tumor shape.The discrepancies in
TVDT may also be due, in part, to true biological differences. For example, the
75
composition of histological subtypes in our study was different from the study by Hsu et
al. Although our study included a majority of gonadotrophs and null cells as reported
previously, other subtypes were more represented in our sample, for example,
corticotroph and somatotroph composition were 16.9 and 10.4% in our study compared to
3% and 0% in the study by Hsu et al. Furthermore, corticotrophs comprised the third
largest proportion of our cohort.
In our study, the TVDT correlated to specific patient demographics. Preoperative TVDT
was related to age, specifically in non-functioning adenomas. This significant trend was
also evident for postoperative TVDT. A similar finding has been reported by Tanaka et
al, who report patients under 61 harbouring a gonadotroph PA displayed a faster TVDT
of their residual than patients over 61 years of age. Gender was also related to the
postoperative TVDT only, where females with functioning tumors had a faster TVDT
than did males(Tanaka et al., 2003).
The MIB-1 LI is currently used as a predictor of aggressive behaviour in PA with the
general understanding that higher MIB-1 LI indicates a faster recurrence (Chacko et al.,
2010; Noh et al., 2009; Prevedello et al., 2005). However, to date no studies have
examined the correlation of pre-operative growth rate to MIB-1 LI in PA. Several studies
have found significant inverse correlations between TVDT and MIB-1 LI for residual PA
indicating that a higher cell proliferation index is associated with a faster PA growth rate.
In our study, the preoperative TVDT was inversely related to the MIB-1 LI in non-
functioning PA but this correlation was not seen between postoperative TVDT and MIB-
1 LI. Also, MIB-1 LI was higher in clinically functioning tumors than in non-functioning
PAs in keeping with what has been reported previously (Thapar et al., 1996). The MIB-1
LI was also associated with preoperative presence of CS growth. Other growth patterns
that exhibited a significantly higher MIB-1 LI include PA with preoperative suprasellar,
anterior, and/or posterior extension. There was an inverse correlation between MIB-1 LI
and age, and other authors have found similar inverse correlations in PA. The MIB-LI is
also significantly higher in females with functioning adenomas relative to males.
Wolfsberger et al. found similar results in females with PA exhibiting a higher MIB-1 LI
76
than males(Wolfsberger et al., 2004).
It should be noted that in our study there was nodirect relation between pre- and
postoperative CS extension: only 42% of preoperative cases had continued postoperative
CS extension. Furthermore, a small percentage of cases (10.4%) developed postoperative
CS extension that was not evident preoperatively. These discrepancies can in part be as a
consequence of how accurately CS invasion is defined on MR imaging and how well
Knosp criteria correlates to intraoperative presence of true invasion.
Besides MIB-1 LI, other biological markers may predict PA growth characteristics. One
such candidate protein is FGFR4.The FGFs and their receptors comprise several different
isoforms that help regulate mitogenesis and angiogenesis. FGFR4 expression has been
previously studied in relation to PA clinical characteristics in a few studies (Morita et al.,
2008; Qian et al., 2004; Zhao, Tomono, & Nose, 1999). Qian et al. examined 137 PA
patients and reported that the majority of their cohort expressed positive cytoplasmic ptd-
FGFR4 staining (59.1%), which is similar to our findings (53.9% FGFR4 positivity) and
congruent with preclinical studies on ptd-FGFR4 positivity in PA(Qian et al., 2004). Ptd-
FGFR4 mRNA was previously reported to correlate with CS invasiveness (but not tumor
size) in somatotrophs(Morita et al., 2008), and its protein expression was also shown to
correlate with PA invasiveness, primarily in gonadotroph and null cell tumors(Ezzat et
al., 2006). Our results confirm that the majority of gonadotrophs and null cell tumors
express cytoplasmic ptd-FGFR4, however, in this series FGFR4 positivity was not related
to invasiveness. In the study by Morita et al., age was not significant correlated to
FGFR4(Morita et al., 2008), whereas in our cohort we found a significant association of
older age and FGFR4 positivity in null cell tumors, a subtype which was not included in
their study. It has been reported that FGFR4 is correlated to the MIB-1 LI. In the study by
Qian et al., which included 137 PA patients, it was demonstrated based on
immunohistochemical analysis that there was a significant positive correlation between
FGFR4 and MIB-1 LI expression (Qian et al., 2004). This finding was not replicated by
our study. The difference in our results may be explained by variation in PA subtype
distribution between the two studies (66% nonfunctioning PA in our study vs 39%
nonfunctioning PA in their study) or differences in staining and quantification techniques,
77
since their mean MIB-1 LI was much lower (1.50%vs 3.99%), even though the
proportions of FGFR4 positivity were similar (53.9% vs 59.1%). On the other hand, we
report an inverse correlation between p27 and MIB-1 LI which has been demonstrated
previously (Korbonits et al., 2002), and provides some confidence in the consistency of
our data more generally.
The expression of p27, which blocks early cell cycle progression, is another candidate for
use as prognostic tool in PA. The p27 expression is reduced in a number of cancers and is
associated with poor clinical outcome(Chu, Hengst, & Slingerland, 2008). It has been
shown that nuclear p27 expression is attenuated in human PA relative to normal pituitary
tissue(Musat et al., 2005). However, the p27 status of most PA in our study were positive
suggesting that the early stages of cell cycle regulation (G1 to S phase) are not adversely
affected in most PA, at least in this patient population. With respect to the expression
profiles of p27 among functioning and nonfunctioning PA, corticotrophs have been
reported to express the lowest amount of p27(Komatsubara et al., 2001). In our study, the
majority of corticotrophs were positive for p27. When growth characteristics of PA were
examined, we found no correlations to p27 expression. To date, several studies have also
failed to show correlation between p27 expression and PA invasiveness or
recurrence(Hewedi, Osman, & El Mahdy, 2011; Zhao et al., 1999) and currently no
studies have attempted to correlate PA growth rate with p27 expression.
To date, no study has examined how p27 and FGFR4 expression may correlate to one
another in PA. We did not find a significant association between FGFR4 and p27
expression. However, gonadotrophs that were both FGFR4 positive and p27 negative
were significantly associated with a faster rate of growth. Thus, in these subtypes
specifically, both the FGFR4 and p27 biomarkers may have prognostic value.
Histopathological subtyping of PA is another means of predicting PA growth
behaviour.In our cohort, somatotroph and null cell subtypes with a cystic or hemorrhagic
component had a faster TVDT. Tanaka et al.recorded the presence intratumoral cyst
which occurred with a similar prevalence as in our cohort (27.5% vs 22% in our study)
but found no correlations to TVDT. Also in our study, only somatotroph subtypes that
exhibited suprasellar extension with optic chiasm compression had significant
78
associations with a faster preoperative TVDT. Our findings suggest that growth towards
the optic chiasm may be an indication for monitoring somatotrophs more closely than for
other subtypes with similar growth patterns. The tendency for somatotrophs to invade the
sphenoid sinus as reported by Zada et al. was not present in our study.
Another factor relevant to predicting PA growth is the presence of postoperative residual
tumor. A residual mass after surgery can herald potential tumor re-growth and so it is
important to determine predictive factors that may be associated with its occurrence,
which may help to establish guidelines for postoperative tumor surveillance. We report
residual volumes in 34.6% of our cohort, with 41.5% of these cases showing evidence of
actual tumor growth. This amounts to tumor recurrence in 14.5% of total cases in our
study. Due to the heterogeneity of PA subtypes and surgical techniques, rates of
recurrence reported in the literature are variable and range from approximately 7-
46%(Alahmadi, Dehdashti, & Gentili, 2012; Cappabianca et al., 2000; Chen et al., 2012;
Losa et al., 2008; Rudnik et al., 2006). In our study, a number of preoperative growth
patterns including suprasellar, posterior, and anterior growth as well as pre- and
postoperative cavernous sinus extension patterns were predictive of the presence of a
residual after surgery. However, neither the recorded growth patternsnor the preoperative
growth rate were predictive of the growth rates of the residual PA, whether it istumor
growth, regression, or stability. We did not find the MIB-1 LI to significantly correlate
with growth characteristics of the residual in our study though MIB-1 LI has previously
been correlated to postoperative residual tumor progression (Widhalm et al., 2009), the
difference in our results and literature reports can be explained by how one defines
growth of residual PA post-operatively, which varies widely between studies.
Due to the heterogeneity of PA, no single predictor of PA growth behaviour can be taken
in isolation as a means to predict its outcome. These predictors must be combined in
order to formulate the most accurate estimation of PA growth which in turn will inform
sound clinical management. A recent study by Trouillas has taken multiple
clinicopathological variables such as tumor proliferation rate, size, and degree invasion
and used this information to predict postoperative outcome(Trouillas et al., 2013). They
found that invasive (into the sphenoid sinus and/or cavernous sinus), and proliferative
79
tumors (Ki-67 >1%, p53 positivity and/or the presence of mitotic nuclei) had a poorer
prognosis, with increased probability of tumor persistence or recurrence compared to
non-invasive tumors. Similarly, our article which has examined various
clinicopathological factors for PA based on their histology and functional status can also
serve as a guide to make a more informed prognosis for PA.
Dysregulated signaling cascades that subserve the clinicopathological observations
outlined above have yet to be identified. Consistent molecular aberrations that underlie
the formation and progression of sporadic PA are not clearly defined. In order to identify
and characterize the biological impact of a proposed pathophysiologic mechanism
involved in PA, a robust in vivo model of PA must be established. To date, a number of
models of PA have been generated but these have primarily harbored genetic mutations.
For example, transgenic mice possessing a dopamine D2 receptor knockout or p27
deletion do develop PA but these are uncharacteristic of the human situation in that, in
humans, these genes are not generally altered(Friedman et al., 1994; Ikeda, Yoshimoto, &
Shida, 1997). Furthermore, these PA models sometimes progress through a hyperplastic
stage before becoming neoplastic(Asa, Kelly, Grandy, & Low, 1999; Friedman et al.,
1994) which is not a common phenomenon that is seen clinically. In our study we have
reliably generated an intracranial xenograft model of PA which may more closely
approximate the sporadic nature of the vast majority of PA. Subcutenous xenograft
models of PA have been established previously but these may not mimic the intracranial
microenvironment in which PA resides.
The tumor growth rate, as assessed by TVDT, was examined in our cell lines as described
previously. Our preliminary experiments showed that the GH4 cells transfected with
empty vector (pc) grew the slowest over time relative to the wild-type FGFR4-G388 or
the FGFR4-R388 tumors. Validation of our initial experiments revealed that the pc
tumors tended to grow the slowest (longest TVDT) but the growth rates between the
untreated tumors of the pc, G388 and R388 tumors were not significantly different from
each other. Our in vivo data does not parallel in vitro findings where the R388 cells have
been shown to grow significantly faster than the pc and G388 cells(Tateno et al. 2011).
80
The discrepancy between our in vivo growth observations and in vitro data reported by
others may stem from biological factors that are introduced once the cells are injected
into the intracranial location. For example, the tumor microenvironment exposes the
cells to various growth factors and cytokines that may have changed the morphology and
growth dynamics of the cells. Another factor that may have contributed to the discrepant
growth rates may have been the presence of tumor hemorrhage. A number of the GH4
tumors had an extensive hemorrhagic component especially in the pc control group and
thus tumor growth may have been somewhat overestimated.
In our model, we examined the expression of the mTOR pathway. The mTOR pathway,
which is frequently upregulated in various cancers, has been shown to be a viable
molecular marker for PA(Musat et al., 2005). Several papers have shown that various
upstream and downstream components from mTOR, including mTOR itself, are
upregulated in animal and human PA tissue(Kenerson et al., 2005; Lu et al., 2008; Musat
et al., 2005). This has indeed been confirmed in our GH4 tumors, particularly with
respect to the abundant expression patterns of mTOR’s two target proteins, S6 and
4EBP1, which are the typical readouts for mTOR expression and are both critical for the
the positive regulation of protein synthesis..
The mTOR inhibiting agents rapamycin and its analogs including RAD001 are potent
immunosuppressants which are also used for their antiproliferative properties.
Rapamycin/RAD001 have been shown to be effective in various endocrine and non-
endocrine cancers and have also demonstrated efficacious preclinical treatment effects in
PA. Our data demonstrates mTOR pathway expression in our GH4 xenograft mouse
models and these models, which harbour different FGFR4 genotypes, are equally
susceptibile to mTOR inhibition. There is currently one other study implicating mTOR
signaling with respect to the FGFR4-G388R SNP in pancreatic neuroendocrine cells. In
their study, it was revealed that the wild-type FGFR4-G388 tumors responded to
RAD001 while the tumors harboring the SNP were resistant to mTOR inhibition(Serra et
al., 2012). In our model, RAD001 treatment resulted in reduced tumor volume and
growth rate (as assessed by a TVDT) but these reductions were of similar magnitude in
81
all cell groups (~2x reduction in growth for pc, G388, R388 tumors). Reduced growth
rate was associated with a downregulation of p-S6 expression. It has been demonstrated
previously by other groups that, following mTOR inhibition, rapamycin resistance is
promoted by the reduction of S6 leading to an elimination of the aforementioned negative
feedbackloop to IRS-1 resulting in Akt activation. It has been reported that this
phenomenon is not seen in the GH cell lines(Gorshtein et al. 2009). This was also not
seen in our GH4 tumors as we initially used octreotide, which abrogates Akt expression,
in addition to RAD001 and the combination did not reveal any synergistic
antiproliferative effects. Furthermore Akt expression in our tumor model was weak
suggesting that mTOR activation may have occurred by other signaling mechanisms.
The parallel downstrem effector of mTOR, 4EBP1, was upregulated but the expression
pattern was not affected by RAD001 in our animals. There is evidence supporting this
phenomenon where long term treatment promotes a rebound effect of p-4EBP1 and
reinitiation of protein translation(Choo, et al. 2008). It has been demonstrated that p-
4EBP1 may become rapamycin (or rapalog) insensitive over time. The treatment period
we implemented lasted for a period of three weeks which may have been long enough to
promote rapamycin insensitivity. The rebound effect of p-4EBP1 re-stimulates mRNA
translation and may provide a possible explanation for our GH4 tumors exhibiting a
slower growth rate but not a complete cessation of growth or regression. Other studies
have shown a reduced expression pattern of mTOR pathway following mTOR inhibition
but correlating this attenuation with a quantified lower growth rate has not been
performed previously.
The cells that we used to generate our model were transfected with FGFR4. Prototypic
FGFR4 is a 110 KDa membrane protein consisting of three extracelluar Ig like domains,
a transmembrane domain, a split tyrosine kinase cytoplasmic domain, and a COOH-
terminal tail. It is important to note that the FGFR4 examined in our clinical cohort was
not the same as FGFR4 examined in our mouse GH4 tumors. In our clinical study, ptd-
FGFR4 was examined which was truncated on the extracellular side and localized to the
82
cytoplasm. The FGFR4 used in our GH4 models was membrane-bound and harbored a
SNP in its transmembrane domain (discussed below). Our GH4 tumors transfected with
wild-type FGFR4 were quite aggressive but did respond to mTOR inhibition with
reduced mTOR signaling and a lower growth rate. Previous work with human data
demonstrates that wild-type FGFR4 was not associated with tumor size in patients
harbouring a somatotroph adenoma. However, in ACTH- PA, the FGFR4 allele was
associated with small hormonally active ACTH-secreting microadenomas. Another study
showed that a correlation exists between homozygous wild-type FGFR4 genotype and
postoperative recurrence in patients with Cushing’s disease((Brito, et al. 2010)) Thus,
the wild-type FGFR4 may play a more integral role in the pathology of human ACTH
tumors relative to GH PA(Nakano-Tateno et al. 2014).
In addition to prototypic FGFR4, we also used an FGFR4 variant that harboured a
germline SNP at codon 388 resulting in the presence of a glycine to arginine substitution
in the transmembrane domain of the FGFR4 protein. At least one allele for this SNP is
present in approximately half of the general population indicating that this is a clinically
relevant polymorphism. The literature regarding the role of this FGFR4-G388R SNP in
cancer has largely been correlative and suggests that is associated with poorer clinical
outcome. Data regarding downstream mechanisms from the FGFR4 SNP remain unclear.
There are two studies to date implicating this SNP in ACTH- and GH PA and suggests
that src and serine STAT3 phosphorylation is responsible for cell proliferation while
tyrosine STAT3 phosphorylation modulates hormone secretion(Nakano-Tateno et al.,
2014; Tateno et al., 2011). Clinically, the SNP allele is associated with increased tumor
size and GH levels as well as the presence of large hormonally-inactive corticotroph
macroadenomas in humans.
In our GH4 tumors, GH expression and IGF-1 secretion was noted but PRL expression
and secretion was absent. This is contrary to PRL expression and secretion noted in other
mouse models of PA tumors. The reason for this discrepancy may be that in other
models transgenic mice were used where pituitary tumors develop in the correct location
and can thus utilize the pituitary-hypothalmic portal system to sustain hormone
83
expression and secretion. Since our PA were intracranial but ectopic, the cells may have
lost this capability due to the lack of a portal system. Interestingly, following mTOR
inhibition with RAD001, GH expression in GH4 tumors experienced a reduction with a
significant attenuation in IGF-1 levels seen only in the R388 mice. It maybe that
RAD001 disrupts a connection where GH is unable to stimulate IGF-1 secretion,
specifically in tumors with aberrant FGFR4. Recently, it has been demonstrated that
tyrosine phosphorylation of STAT3 helps to regulate GH levels (Tateno et al. 2011)and it
may be through this pathway that RAD001 exerts it’s hormone attenuating effects. There
is one study to date that has reported, at the in vitro level, mTOR inhibition abrogates
IGF-1 mediated effects in human PA cells(Zatelli et al. 2010). In sum, our finding is
meaningful and warrants further investigation as it has important therapeutic implications
in patients with GH tumors who harbour the FGFR4-R388 allele.
5.1 Clinical applications for mTOR inhibitors in PA
Despite a promising role for mTOR inhibitors in preclinical models of PA, data regarding
mTOR inhibitors as efficacious therapeutic agents remains limited in clinical
management of PA. To date, there is one case report published on the combined use of
everolimus and octreotide in a patient with an ACTH pituitary carcinoma resistant to
temozolomide treatment(Jouanneau, et al. 2012). Pituitary carcinomas are extremely rare
malignant pituitary tumors which are differentiated from PA by their potential for
metastatic spread. The combined everolimus/octreotide treatment was unable to control
tumor growth in this patient who eventually died shortly after treatment was initiated. It
should be noted that the studies attesting to the efficacy of mTOR inhibition in the PA
preclinical setting were conducted using human NFPA or GH-secreting PA(Gorshtein et
al. 2009; Zatelli et al. 2010). At this stage, though preclinical results are promising, their
enthusiasm for their efficacy in clinical practice has to be moderated by the fact that
objective clinical data is lacking to confirm the therapeutic value of mTOR inhibitors in
PA patients.
84
5.2 Conclusions
Current understanding of PA pathology is limited. Our data has identified several
preclinical and clinical factors that may be involved in PA growth behavior. Clinically,
our data suggest that the growth rate of PAs are influenced by various patient and tumor-
specific characteristics including the age and sex of the patient, the specific subtype of
PA, it’s hormonal activity, it’s immunohistochemical profile including the MIB-1 LI
status, and it’s preponderance for different growth directions relative to the pituitary
fossa. Furthermore, the pre- and postoperative PA growth rates were correlated
suggesting that postoperative PA growth rates can be predicted, in part, by preoperative
growth rates thus better informing postoperative outcome.
Our preclinical in vivo model of PA has enabled us to identify the mTOR pathway as a
possible treatment target, specifically in GH4 tumors with altered FGFR4 genotype, =
We have shown that the mTOR pathway is differentially expressed in the R388 tumors
compared to the G388 tumors. Both G388 and R388 tumors display a reduction in tumor
volume, growth rate, and mTOR expression following treatment with RAD001 with a
similar reduction in all groups. The R388 group had the highest levels of GH and serum
IGF-1 levels and is the only group that experienced a significant reduction in these levels
following RAD00ptreatment. These data suggest that PA, specifically GH PA can
benefit from mTOR inhibition therapy with a overall reduction in growth rate and a
reduction in hormone levels in those GH PA harboring an FGFR4 polymorphism.
5.3 Limitations
In our first aim, we conducted a retrospective review. There are a number of limitations
inherent in such a study design. For instance, when retrieving archival data, missing
values or data that is not reported is an issue. It would have been ideal to correlate our
radiological finding of PA growth to surgical outcome, as this would have provided
definitive evidence for a particular PA growth pattern. However, often times
intraoperative findings of PA growth were not reported and thus our study, based solely
85
on radiological data, may have been an over- or underestimation of the true incidence of
PA growth. Our inclusion of pathological markers MIB-1, p27, and FGFR4 were
obtained from historical pathology reports and these markers were not always reported
for each case. Furthermore, the p27 and the FGFR4 status were only examined
qualitatively. Ideally, the tissue should have been collected for each case and reassessed
for the purposes of this study. However, for the majority of tissue samples there was a
limited amount of archival material which precluded reinvestigation of samples. Another
issue may have been the introduction of selection bias into the study. Patients were
excluded if they did not have adequate preoperative and postoperative radiological data
for comparison, even if their tumors were large. Our definition of tumor recurrence was
based on radiological outcome and we did not consider biochemical recurrence in our
definition, which also excluded a few patients.
In our in vivo study, the main question remains, as with all other preclinical models of
brain tumor the validity and relevance of this model to human disease. For example is
injection of frontal intraperhancyma an accurate model for PA, and most likely
injectionof the pit fossa is a more reliable model. However given technical limitations we
chose to proceed the initial step of this project with frontal lobe injection. This raises the
issue of how well our model recapitulates the PA microenvironment. We believe that our
ectopic intracranial model of PA does mimic the microenvironment as the brain consists
of necessary cytokines and growth factors required to facilitate PA development. There
was also an issue concerning the heterogeneity of tumor growth rates. Although cells
were injected in equal numbers, tumors of the same cell type grew at different rates thus
complicating treatment initiation and endpoints. The variable tumor growth rates had to
do with the exponential growth patterns of these tumor models, which consisted of a
growth latency period shortly after intracranial injection followed by rapid tumor
expansion. In some mice, the tumor was much more aggressive than in other mice and
this lack of uniformity sometimes hampered our ability to extract meaningful treatment
data.
86
5.4 Future Directions
With respect to our retrospective review of PA patients, increasing the sample size by
including additional patients treated after 2011 may be an option to expand the project.
Also, including a multi-centric or even nationwide database of PA parameters may
increase the validity of the study findings. Comparing biochemical relapse to our finding
of radiological tumor recurrence within the functioning PAs may be of clinical value and
should also be considered.
The xenograft mouse project would benefit from the incorporation of human in vitro data
to complement the in vivo findings. Human PA surgical samples could be dispersed in
primary culture and mTOR expression could be examined following mTOR inhibition
therapy. There are currently no human PA cell lines in which to study RAD001 effects on
human PA cells so developing such a model would be a task in itself. In fact, the original
aim of this study was to generate a stable human cell line. After several attempts this
proved to be very difficult and we were ultimately unsuccessful (See Appendix 1 for
protocol). The cells would remain viable for one passage and die shortly thereafter.
Perhaps to mimic the necessary microenvironment in which PA grow, several growth
factors, hormones, and cytokines should have been added to the culture media.
Nevertheless, if human PA cell lines are established, exposure to RAD001 and
subsequent mTOR expression could be examined. Colony formation assays could be
performed to assess cell proliferation and viability following drug treatment. RAD001 is
a known cytostatic agent so FAC sorting should be performed for cell cycle analysis
using either human PA cells in primary culture or PA cell lines. All of these experiments
should be performed in the context of the FGFR4 polymorphism.
87
REFERENCES 1. Christmann, A. V. A. (2006). Robust estimation of Cronbach's alpha. Journal of Multivariate Analysis, 97(7), 1660-1674. doi: 10.1016/j.jmva.2005.05.012 Abbass, S. A., Asa, S. L., & Ezzat, S. (1997). Altered expression of fibroblast growth factor receptors in human pituitary adenomas. J Clin Endocrinol Metab, 82(4), 1160-1166. 2. Advani, S. H. (2010). Targeting mTOR pathway: A new concept in cancer therapy. Indian J Med Paediatr Oncol, 31(4), 132-136. doi: 10.4103/0971-5851.76197 [doi] IJMPO-31-132 [pii] 3. Alahmadi, H., Dehdashti, A. R., & Gentili, F. (2012). Endoscopic endonasal surgery in recurrent and residual pituitary adenomas after microscopic resection. World Neurosurg, 77(3-4), 540-547. doi: 10.1016/j.wneu.2011.07.012 4. Altmeyer, A., Josset, E., Denis, J. M., Gueulette, J., Slabbert, J., Mutter, D., . . . Bischoff, P. (2012). The mTOR inhibitor RAD001 augments radiation-induced growth inhibition in a hepatocellular carcinoma cell line by increasing autophagy. Int J Oncol. doi: 10.3892/ijo.2012.1583 [doi] 5. Altomare, D. A., Tanno, S., De Rienzo, A., Klein-Szanto, A. J., Skele, K. L., Hoffman, J. P., & Testa, J. R. (2002). Frequent activation of AKT2 kinase in human pancreatic carcinomas. J Cell Biochem, 87(4), 470-476. 6. Asa, S. L., & Ezzat, S. (1998). The cytogenesis and pathogenesis of pituitary adenomas. [Research Support, Non-U.S. Gov't Review]. Endocr Rev, 19(6), 798-827. doi: 10.1210/edrv.19.6.0350 7. Asa, S. L., & Ezzat, S. (2009). The pathogenesis of pituitary tumors. Annu Rev Pathol, 4, 97-126. doi: 10.1146/annurev.pathol.4.110807.092259 [doi] 8. Asa, S. L., Kelly, M. A., Grandy, D. K., & Low, M. J. (1999). Pituitary lactotroph adenomas develop after prolonged lactotroph hyperplasia in dopamine D2 receptor-deficient mice. [Research Support, Non-U.S. Gov't Research Support, U.S. Gov't, P.H.S.]. Endocrinology, 140(11), 5348-5355. doi: 10.1210/endo.140.11.7118 9. Bader, A. G., Kang, S., & Vogt, P. K. (2006). Cancer-specific mutations in PIK3CA are oncogenic in vivo. Proc Natl Acad Sci U S A, 103(5), 1475-1479. doi: 0510857103 [pii] 10.1073/pnas.0510857103 [doi] 10. Beuvink, I., Boulay, A., Fumagalli, S., Zilbermann, F., Ruetz, S., O'Reilly, T., . . . Thomas, G. (2005). The mTOR inhibitor RAD001 sensitizes tumor cells to DNA-damaged induced apoptosis through inhibition of p21 translation. Cell, 120(6), 747-759. doi: S0092-8674(05)00087-5 [pii]10.1016/j.cell.2004.12.040 [doi] 11. Bonneville, J. F., Cattin, F., Racle, A., Bouchareb, M., Boulard, D., Potelon, P., & Tang, Y. S. (1989). Dynamic CT of the laterosellar extradural venous spaces. AJNR Am J Neuroradiol, 10(3), 535-542.
88
12. Brada, M., Rajan, B., Traish, D., Ashley, S., Holmes-Sellors, P. J., Nussey, S., & Uttley, D. (1993). The long-term efficacy of conservative surgery and radiotherapy in the control of pituitary adenomas. [Research Support, Non-U.S. Gov't]. Clin Endocrinol (Oxf), 38(6), 571-578. 13. Brognard, J., Clark, A. S., Ni, Y., & Dennis, P. A. (2001). Akt/protein kinase B is constitutively active in non-small cell lung cancer cells and promotes cellular survival and resistance to chemotherapy and radiation. Cancer Res, 61(10), 3986-3997. 14. Burris, H. A., 3rd. (2013). Overcoming acquired resistance to anticancer therapy: focus on the PI3K/AKT/mTOR pathway. Cancer Chemother Pharmacol. doi: 10.1007/s00280-012-2043-3 [doi] 15. Burrell, K., Singh, S., Jalali, S., Hill, RP., Zadeh, G. (2014). VEGF regulates region-specific localization of perivascular bone marrow-derived cells in glioblastoma. Cancer Res, (74)14, 3727-3239. doi: 10.1158/0008-5472.CAN-13-3119 16. Cao, L., Chen, H., Hong, J., Ma, M., Zhong, Q., & Wang, S. (2013). Magnetic resonance imaging appearance of the medial wall of the cavernous sinus for the assessment of cavernous sinus invasion by pituitary adenomas. J Neuroradiol, 40(4), 245-251. doi: 10.1016/j.neurad.2013.06.003 17. Cappabianca, P., Alfieri, A., Colao, A., Cavallo, L. M., Fusco, M., Peca, C., . . . de Divitiis, E. (2000). Endoscopic endonasal transsphenoidal surgery in recurrent and residual pituitary adenomas: technical note. Minim Invasive Neurosurg, 43(1), 38-43. doi: 10.1055/s-2000-8814 18. Cerovac, V., Monteserin-Garcia, J., Rubinfeld, H., Buchfelder, M., Losa, M., Florio, T., . . . Theodoropoulou, M. (2010). The somatostatin analogue octreotide confers sensitivity to rapamycin treatment on pituitary tumor cells. Cancer Res, 70(2), 666-674. doi: 0008-5472.CAN-09-2951 [pii] 10.1158/0008-5472.CAN-09-2951 [doi] 19. Ceylan, S., Anik, I., & Koc, K. (2011). A new endoscopic surgical classification and invasion criteria for pituitary adenomas involving the cavernous sinus. Turk Neurosurg, 21(3), 330-339. doi: 10.5137/1019-5149.JTN.4149-11.0 [doi] 20. Chacko, G., Chacko, A. G., Kovacs, K., Scheithauer, B. W., Mani, S., Muliyil, J. P., & Seshadri, M. S. (2010). The clinical significance of MIB-1 labeling index in pituitary adenomas. Pituitary, 13(4), 337-344. doi: 10.1007/s11102-010-0242-7 [doi] 21. Chen, Y., Wang, C. D., Su, Z. P., Chen, Y. X., Cai, L., Zhuge, Q. C., & Wu, Z. B. (2012). Natural history of postoperative nonfunctioning pituitary adenomas: a systematic review and meta-analysis. [Research Support, Non-U.S. Gov't]. Neuroendocrinology, 96(4), 333-342. doi: 10.1159/000339823 22. Chow, L. M., Endersby, R., Zhu, X., Rankin, S., Qu, C., Zhang, J., . . . Baker, S. J. (2011). Cooperativity within and among Pten, p53, and Rb pathways induces high-grade astrocytoma in adult brain. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't]. Cancer Cell, 19(3), 305-316. doi: 10.1016/j.ccr.2011.01.039 23. Chu, I. M., Hengst, L., & Slingerland, J. M. (2008). The Cdk inhibitor p27 in human cancer: prognostic potential and relevance to anticancer therapy. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't
89
Review]. Nat Rev Cancer, 8(4), 253-267. doi: 10.1038/nrc2347 24. Chung, C., Jalali, S., Foltz, W., Burrell, K., Wildgoose, P., Lindsay, P., . . . Menard, C. (2012). Imaging Biomarker Dynamics in an Intracranial Murine Glioma Study of Radiation and Antiangiogenic Therapy. International journal of radiation oncology, biology, physics. doi: 10.1016/j.ijrobp.2012.07.005 25. Clark, A. S., West, K., Streicher, S., & Dennis, P. A. (2002). Constitutive and inducible Akt activity promotes resistance to chemotherapy, trastuzumab, or tamoxifen in breast cancer cells. Mol Cancer Ther, 1(9), 707-717. 26. Connor, S. E., Wilson, F., & Hogarth, K. (2014). Magnetic resonance imaging criteria to predict complete excision of parasellar pituitary macroadenoma on postoperative imaging. J Neurol Surg B Skull Base, 75(1), 41-46. doi: 10.1055/s-0033-1353362 27. Cooper, O., Vlotides, G., Fukuoka, H., Greene, M. I., & Melmed, S. (2011). Expression and function of ErbB receptors and ligands in the pituitary. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't Review]. Endocr Relat Cancer, 18(6), R197-211. doi: 10.1530/ERC-11-0066 28. Cottier, J. P., Destrieux, C., Brunereau, L., Bertrand, P., Moreau, L., Jan, M., & Herbreteau, D. (2000). Cavernous sinus invasion by pituitary adenoma: MR imaging. Radiology, 215(2), 463-469. 29. Dai, C., Zhang, B., Liu, X., Ma, S., Yang, Y., Yao, Y., . . . Wang, R. (2013). Inhibition of PI3K/AKT/mTOR Pathway Enhances Temozolomide-Induced Cytotoxicity in Pituitary Adenoma Cell Lines in Vitro and Xenografted Pituitary Adenoma in Female Nude Mice. Endocrinology, 154(3), 1247-1259. doi: en.2012-1908 [pii] 10.1210/en.2012-1908 [doi] 30. Datta, S. R., Dudek, H., Tao, X., Masters, S., Fu, H., Gotoh, Y., & Greenberg, M. E. (1997). Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell, 91(2), 231-241. doi: S0092-8674(00)80405-5 [pii] Douros, J., & Suffness, M. (1981). New natural products under development at the National Cancer Institute. Recent Results Cancer Res, 76, 153-175. 31. Dworakowska, D., Wlodek, E., Leontiou, C. A., Igreja, S., Cakir, M., Teng, M., . . . Grossman, A. B. (2009). Activation of RAF/MEK/ERK and PI3K/AKT/mTOR pathways in pituitary adenomas and their effects on downstream effectors. Endocr Relat Cancer, 16(4), 1329-1338. doi: ERC-09-0101 [pii] 10.1677/ERC-09-0101 [doi] 32. Ebinu, J. O., Lwu, S., Monsalves, E., Arayee, M., Chung, C., Laperriere, N. J., . . . Zadeh, G. (2013). Gamma knife radiosurgery for the treatment of cystic cerebral metastases. [Evaluation Studies]. Int J Radiat Oncol Biol Phys, 85(3), 667-671. doi: 10.1016/j.ijrobp.2012.06.043 33. Ekramullah, S. M., Saitoh, Y., Arita, N., Ohnishi, T., & Hayakawa, T. (1996). The correlation of Ki-67 staining indices with tumour doubling times in regrowing non-functioning pituitary adenomas. Acta Neurochir (Wien), 138(12), 1449-1455. 34. Ezzat, S. (2001). The role of hormones, growth factors and their receptors in pituitary
90
tumorigenesis. [Research Support, Non-U.S. Gov't Review]. Brain Pathol, 11(3), 356-370. 35. Ezzat, S., Zheng, L., & Asa, S. L. (2004). Pituitary tumor-derived fibroblast growth factor receptor 4 isoform disrupts neural cell-adhesion molecule/N-cadherin signaling to diminish cell adhesiveness: a mechanism underlying pituitary neoplasia. [Research Support, Non-U.S. Gov't]. Mol Endocrinol, 18(10), 2543-2552. doi: 10.1210/me.2004-0182 36. Ezzat, S., Zheng, L., Winer, D., & Asa, S. L. (2006). Targeting N-cadherin through fibroblast growth factor receptor-4: distinct pathogenetic and therapeutic implications. [Research Support, Non-U.S. Gov't]. Mol Endocrinol, 20(11), 2965-2975. doi: 10.1210/me.2006-0223 37. Ezzat, S., Zheng, L., Zhu, X. F., Wu, G. E., & Asa, S. L. (2002). Targeted expression of a human pituitary tumor-derived isoform of FGF receptor-4 recapitulates pituitary tumorigenesis. [Research Support, Non-U.S. Gov't]. J Clin Invest, 109(1), 69-78. doi: 10.1172/JCI14036 38. Faivre, S., Kroemer, G., & Raymond, E. (2006). Current development of mTOR inhibitors as anticancer agents. Nat Rev Drug Discov, 5(8), 671-688. doi: nrd2062 [pii] 10.1038/nrd2062 [doi] 39. Friedman, E., Adams, E. F., Hoog, A., Gejman, P. V., Carson, E., Larsson, C., . . . Nordenskjold, M. (1994). Normal structural dopamine type 2 receptor gene in prolactin-secreting and other pituitary tumors. [Research Support, Non-U.S. Gov't]. J Clin Endocrinol Metab, 78(3), 568-574. doi: 10.1210/jcem.78.3.7907340 40. Fukuoka, H., Cooper, O., Mizutani, J., Tong, Y., Ren, S. G., Bannykh, S., & Melmed, S. (2011). HER2/ErbB2 receptor signaling in rat and human prolactinoma cells: strategy for targeted prolactinoma therapy. Mol Endocrinol, 25(1), 92-103. doi: me.2010-0353 [pii] 10.1210/me.2010-0353 [doi] 41. Fusco, A., Zatelli, M. C., Bianchi, A., Cimino, V., Tilaro, L., Veltri, F., . . . De Marinis, L. (2008). Prognostic significance of the Ki-67 labeling index in growth hormone-secreting pituitary adenomas. J Clin Endocrinol Metab, 93(7), 2746-2750. doi: jc.2008-0126 [pii]0.1210/jc.2008-0126 [doi] 42. Gorshtein, A., Rubinfeld, H., Kendler, E., Theodoropoulou, M., Cerovac, V., Stalla, G. K., . . . Shimon, I. (2009). Mammalian target of rapamycin inhibitors rapamycin and RAD001 (everolimus) induce anti-proliferative effects in GH-secreting pituitary tumor cells in vitro. Endocr Relat Cancer, 16(3), 1017-1027. doi: ERC-08-0269 [pii] 10.1677/ERC-08-0269 [doi] 43. Guy, C. T., Webster, M. A., Schaller, M., Parsons, T. J., Cardiff, R. D., & Muller, W. J. (1992). Expression of the neu protooncogene in the mammary epithelium of transgenic mice induces metastatic disease. Proc Natl Acad Sci U S A, 89(22), 10578-10582. Guzzo, M. F., Carvalho, L. R., & Bronstein, M. D. (2014). Apoptosis: its role in pituitary development and neoplastic pituitary tissue. Pituitary, 17(2), 157-162. doi: 10.1007/s11102-013-0481-5 44. Hagiwara, A., Inoue, Y., Wakasa, K., Haba, T., Tashiro, T., & Miyamoto, T. (2003). Comparison of growth hormone-producing and non-growth hormone-producing pituitary adenomas: imaging characteristics and pathologic correlation. Radiology, 228(2), 533-538. doi:
91
10.1148/radiol.2282020695 45. Hahn, C. A., Zhou, S. M., Raynor, R., Tisch, A., Light, K., Shafman, T., . . . Marks, L. B. (2009). Dose-dependent effects of radiation therapy on cerebral blood flow, metabolism, and neurocognitive dysfunction. Int J Radiat Oncol Biol Phys, 73(4), 1082-1087. doi: 10.1016/j.ijrobp.2008.05.061 46. Harrington, L. S., Findlay, G. M., Gray, A., Tolkacheva, T., Wigfield, S., Rebholz, H., . . . Lamb, R. F. (2004). The TSC1-2 tumor suppressor controls insulin-PI3K signaling via regulation of IRS proteins. J Cell Biol, 166(2), 213-223. doi: 10.1083/jcb.200403069 [doi]jcb.200403069 [pii] 47. Hewedi, I. H., Osman, W. M., & El Mahdy, M. M. (2011). Differential expression of cyclin D1 in human pituitary tumors: relation to MIB-1 and p27/Kip1 labeling indices. J Egypt Natl Canc Inst, 23(4), 171-179. doi: 10.1016/j.jnci.2011.11.003 48. Hibberts, N. A., Simpson, D. J., Bicknell, J. E., Broome, J. C., Hoban, P. R., Clayton, R. N., & Farrell, W. E. (1999). Analysis of cyclin D1 (CCND1) allelic imbalance and overexpression in sporadic human pituitary tumors. Clin Cancer Res, 5(8), 2133-2139. 49. Honegger, J., Prettin, C., Feuerhake, F., Petrick, M., Schulte-Monting, J., & Reincke, M. (2003). Expression of Ki-67 antigen in nonfunctioning pituitary adenomas: correlation with growth velocity and invasiveness. J Neurosurg, 99(4), 674-679. doi: 10.3171/jns.2003.99.4.0674 [doi] 50. Honegger, J., Zimmermann, S., Psaras, T., Petrick, M., Mittelbronn, M., Ernemann, U., . . . Dietz, K. (2008). Growth modelling of non-functioning pituitary adenomas in patients referred for surgery. Eur J Endocrinol, 158(3), 287-294. doi: 10.1530/EJE-07-0502 51. Hsu, C. Y., Guo, W. Y., Chien, C. P., & Ho, D. M. (2010). MIB-1 labeling index correlated with magnetic resonance imaging detected tumor volume doubling time in pituitary adenoma. Eur J Endocrinol, 162(6), 1027-1033. doi: EJE-09-1100 [pii] 10.1530/EJE-09-1100 [doi] 52. Ikeda, H., Yoshimoto, T., & Shida, N. (1997). Molecular analysis of p21 and p27 genes in human pituitary adenomas. Br J Cancer, 76(9), 1119-1123. 53. Jahangiri, A., Wagner, J. R., Pekmezci, M., Hiniker, A., Chang, E. F., Kunwar, S., . . . Aghi, M. K. (2013). A comprehensive long-term retrospective analysis of silent corticotrophic adenomas vs hormone-negative adenomas. [Research Support, Non-U.S. Gov't]. Neurosurgery, 73(1), 8-17; discussion 17-18. doi: 10.1227/01.neu.0000429858.96652.1e 54. Jain, A. K., Gupta, A. K., Pathak, A., Bhansali, A., & Bapuraj, J. R. (2008). Endonasal transsphenoidal pituitary surgery: is tumor volume a key factor in determining outcome? Am J Otolaryngol, 29(1), 48-50. doi: 10.1016/j.amjoto.2007.01.006 55. Jefferies, H. B., Fumagalli, S., Dennis, P. B., Reinhard, C., Pearson, R. B., & Thomas, G. (1997). Rapamycin suppresses 5'TOP mRNA translation through inhibition of p70s6k. EMBO J, 16(12), 3693-3704. doi: 10.1093/emboj/16.12.3693 [doi]
92
56. Jalali, S., Chung, C., Foltz, W., Burrell, K., Singh, S., Hill, RP., Zadeh, G. (2014). MRI biomarkers identify the differential response of glioblastoma multiforme to anti-angiogenic therapy. Neuro Oncol, 16(6), 868-879. doi: 10.1093/neuonc/nou040 57. Johnson, M. D., O'Connell, M. J., Pilcher, W., & Reeder, J. E. (2010). Fibroblast growth factor receptor-3 expression in meningiomas with stimulation of proliferation by the phosphoinositide 3 kinase-Akt pathway. J Neurosurg, 112(5), 934-939. doi: 10.3171/2009.7.JNS09726 [doi] 58. Kandel, E. S., & Hay, N. (1999). The regulation and activities of the multifunctional serine/threonine kinase Akt/PKB. Exp Cell Res, 253(1), 210-229. doi: 10.1006/excr.1999.4690 [doi] 0014-4827(99)94690-6 [pii] 59. Kawashima, K., Yamakawa, K., & Arita, J. (2000). Involvement of phosphoinositide-3-kinase and p70 S6 kinase in regulation of proliferation of rat lactotrophs in culture. Endocrine, 13(3), 385-392. doi: ENDO:13:3:385 [pii] 10.1385/ENDO:13:3:385 [doi] 60. Kenerson, H., Dundon, T. A., & Yeung, R. S. (2005). Effects of rapamycin in the Eker rat model of tuberous sclerosis complex. Pediatr Res, 57(1), 67-75. doi: 01.PDR.0000147727.78571.07 [pii] 10.1203/01.PDR.0000147727.78571.07 [doi] 61. Knosp, E., Steiner, E., Kitz, K., & Matula, C. (1993). Pituitary adenomas with invasion of the cavernous sinus space: a magnetic resonance imaging classification compared with surgical findings. Neurosurgery, 33(4), 610-617; discussion 617-618. 62. Komatsubara, K., Tahara, S., Umeoka, K., Sanno, N., Teramoto, A., & Osamura, R. Y. (2001). Immunohistochemical analysis of p27 (Kip1) in human pituitary glands and in various types of pituitary adenomas. Endocr Pathol, 12(2), 181-188. Korbonits, M., Chahal, H. S., Kaltsas, G., Jordan, S., Urmanova, Y., Khalimova, Z., . . . Grossman, A. B. (2002). Expression of phosphorylated p27(Kip1) protein and Jun activation domain-binding protein 1 in human pituitary tumors. [Research Support, Non-U.S. Gov't 63. Research Support, U.S. Gov't, P.H.S.]. J Clin Endocrinol Metab, 87(6), 2635-2643. Lahtz, C., Stranzenbach, R., Fiedler, E., Helmbold, P., & Dammann, R. H. (2010). Methylation of PTEN as a prognostic factor in malignant melanoma of the skin. [Letter Research Support, Non-U.S. Gov't]. J Invest Dermatol, 130(2), 620-622. doi: 10.1038/jid.2009.226 64. Lamberts, S. W., de Herder, W. W., & Hofland, L. J. (2002). Somatostatin analogs in the diagnosis and treatment of cancer. Trends Endocrinol Metab, 13(10), 451-457. doi: S1043276002006677 [pii] 65. Lee, M., Theodoropoulou, M., Graw, J., Roncaroli, F., Zatelli, M. C., & Pellegata, N. S. (2011). Levels of p27 sensitize to dual PI3K/mTOR inhibition. Mol Cancer Ther, 10(8), 1450-1459. doi: 1535-7163.MCT-11-0188 [pii]10.1158/1535-7163.MCT-11-0188 [doi] 66. LeRiche, V. K., Asa, S. L., & Ezzat, S. (1996). Epidermal growth factor and its receptor (EGF-R) in human pituitary adenomas: EGF-R correlates with tumor aggressiveness. [Research
93
Support, Non-U.S. Gov't]. J Clin Endocrinol Metab, 81(2), 656-662. doi: 10.1210/jcem.81.2.8636285 67. Leslie, N. R., Biondi, R. M., & Alessi, D. R. (2001). Phosphoinositide-regulated kinases and phosphoinositide phosphatases. Chem Rev, 101(8), 2365-2380. doi: cr000091i [pii] 68. Lin, Y., Jiang, X., Shen, Y., Li, M., Ma, H., Xing, M., & Lu, Y. (2009). Frequent mutations and amplifications of the PIK3CA gene in pituitary tumors. Endocr Relat Cancer, 16(1), 301-310. doi: ERC-08-0167 [pii] 10.1677/ERC-08-0167 [doi] 69. Losa, M., Mortini, P., Barzaghi, R., Ribotto, P., Terreni, M. R., Marzoli, S. B., . . . Giovanelli, M. (2008). Early results of surgery in patients with nonfunctioning pituitary adenoma and analysis of the risk of tumor recurrence. J Neurosurg, 108(3), 525-532. doi: 10.3171/JNS/2008/108/3/0525 70. Lu, C., Willingham, M. C., Furuya, F., & Cheng, S. Y. (2008). Activation of phosphatidylinositol 3-kinase signaling promotes aberrant pituitary growth in a mouse model of thyroid-stimulating hormone-secreting pituitary tumors. Endocrinology, 149(7), 3339-3345. doi: en.2007-1696 [pii]10.1210/en.2007-1696 [doi] 71. Lundin, P., Nyman, R., Burman, P., Lundberg, P. O., & Muhr, C. (1992). MRI of pituitary macroadenomas with reference to hormonal activity. [Research Support, Non-U.S. Gov't]. Neuroradiology, 34(1), 43-51. 72. Ma, S., Liu, X., Yao, Y., Cai, F., Dai, C., Bao, X., . . . Wang, R. (2011). Effect of temozolomide on cell viability in gonadotroph adenoma cell lines. Oncol Rep, 26(3), 543-550. doi: 10.3892/or.2011.1317 [doi] 73. Madsen, H., Borges, T. M., Knox, A. J., Michaelis, K. A., Xu, M., Lillehei, K. O., . . . Kleinschmidt-DeMasters, B. K. (2011). Giant pituitary adenomas: pathologic-radiographic correlations and lack of role for p53 and MIB-1 labeling. Am J Surg Pathol, 35(8), 1204-1213. doi: 10.1097/PAS.0b013e31821e8c96 [doi] 00000478-201108000-00015 [pii] 74. Maira, S. M., Stauffer, F., Brueggen, J., Furet, P., Schnell, C., Fritsch, C., . . . Garcia-Echeverria, C. (2008). Identification and characterization of NVP-BEZ235, a new orally available dual phosphatidylinositol 3-kinase/mammalian target of rapamycin inhibitor with potent in vivo antitumor activity. Mol Cancer Ther, 7(7), 1851-1863. doi: 1535-7163.MCT-08-0017 [pii] 10.1158/1535-7163.MCT-08-0017 [doi] 75. Majumder, P. K., Febbo, P. G., Bikoff, R., Berger, R., Xue, Q., McMahon, L. M., . . . Sellers, W. R. (2004). mTOR inhibition reverses Akt-dependent prostate intraepithelial neoplasia through regulation of apoptotic and HIF-1-dependent pathways. Nat Med, 10(6), 594-601. doi: 10.1038/nm1052 [doi] nm1052 [pii] 76. Manning, B. D., Tee, A. R., Logsdon, M. N., Blenis, J., & Cantley, L. C. (2002). Identification of the tuberous sclerosis complex-2 tumor suppressor gene product tuberin as a target of the phosphoinositide 3-kinase/akt pathway. Mol Cell, 10(1), 151-162. doi: S1097276502005683 [pii]
94
77. Marro, B., Zouaoui, A., Sahel, M., Crozat, N., Gerber, S., Sourour, N., . . . Marsault, C. (1997). MRI of pituitary adenomas in acromegaly. Neuroradiology, 39(6), 394-399. Mauceri, H. J., Sutton, H. G., Darga, T. E., Kocherginsky, M., Kochanski, J., Weichselbaum, R. R., & Vokes, E. E. (2012). Everolimus exhibits efficacy as a radiosensitizer in a model of non-small cell lung cancer. Oncol Rep, 27(5), 1625-1629. doi: 10.3892/or.2012.1666 [doi] 78. Medema, R. H., Kops, G. J., Bos, J. L., & Burgering, B. M. (2000). AFX-like Forkhead transcription factors mediate cell-cycle regulation by Ras and PKB through p27kip1. Nature, 404(6779), 782-787. doi: 10.1038/35008115 [doi] 79. Mete, O., & Asa, S. L. (2012). Clinicopathological correlations in pituitary adenomas. [Review]. Brain Pathol, 22(4), 443-453. doi: 10.1111/j.1750-3639.2012.00599.x 80. Mhawech-Fauceglia, P., Wang, D., Kim, G., Sharifian, M., Chen, X., Liu, Q., . . . Pejovic, T. (2014). Expression of DNA repair proteins in endometrial cancer predicts disease outcome. Gynecol Oncol, 132(3), 593-598. doi: 10.1016/j.ygyno.2014.02.002 81. Monsalves, E., Juraschka, K., Tateno, T., Agnihotri, S., Asa, SL., Ezzat, S., Zadeh, G. (2014). The PI3K/Akt/mTOR pathway in the pathophysiology and treatment of pituitary adenomas. Endocr Relat Cancer, 21(4), R331-344. doi: 10.1530/ERC-14-0188 82. Monsalves, E., Larjani, S., Loyola Godoy, B., Juraschka, K., Carvalho, F., Kucharczyk, W., ... Zadeh, G. (2014). J Clin Endocrinol Metab, (99)4, 1330-1338. doi: 10.1210/jc.2013-3054 83. Moreau, L., Cottier, J. P., Bertrand, P., Destrieux, C., Jan, M., Sonier, C. B., . . . Rouleau, P. (1998). [MRI diagnosis of sinus cavernous invasion by pituitary adenomas]. J Radiol, 79(3), 241-246. 84. Morita, K., Takano, K., Yasufuku-Takano, J., Yamada, S., Teramoto, A., Takei, M., . . . Fujita, T. (2008). Expression of pituitary tumour-derived, N-terminally truncated isoform of fibroblast growth factor receptor 4 (ptd-FGFR4) correlates with tumour invasiveness but not with G-protein alpha subunit (gsp) mutation in human GH-secreting pituitary adenomas. Clin Endocrinol (Oxf), 68(3), 435-441. doi: CEN3062 [pii] 10.1111/j.1365-2265.2007.03062.x [doi] 85. Muller, W. J., Sinn, E., Pattengale, P. K., Wallace, R., & Leder, P. (1988). Single-step induction of mammary adenocarcinoma in transgenic mice bearing the activated c-neu oncogene. Cell, 54(1), 105-115. doi: 0092-8674(88)90184-5 [pii] 86. Murat, C. B., Braga, P. B., Fortes, M. A., Bronstein, M. D., Correa-Giannella, M. L., & Giorgi, R. R. (2012). Mutation and genomic amplification of the PIK3CA proto-oncogene in pituitary adenomas. Braz J Med Biol Res, 45(9), 851-855. doi: S0100-879X2012007500115 [pii] 87. Murayama, T., Inokuchi, M., Takagi, Y., Yamada, H., Kojima, K., Kumagai, J., . . . Sugihara, K. (2009). Relation between outcomes and localisation of p-mTOR expression in gastric cancer. Br J Cancer, 100(5), 782-788. doi: 6604915 [pii] 10.1038/sj.bjc.6604915 [doi] 88. Musat, M., Korbonits, M., Kola, B., Borboli, N., Hanson, M. R., Nanzer, A. M., . . . Grossman, A. B. (2005). Enhanced protein kinase B/Akt signalling in pituitary tumours. Endocr Relat Cancer, 12(2), 423-433. doi: 12/2/423 [pii]
95
10.1677/erc.1.00949 [doi] 89. Nakaguchi, H., Fujimaki, T., Matsuno, A., Matsuura, R., Asai, A., Suzuki, I., . . . Kirino, T. (1999). Postoperative residual tumor growth of meningioma can be predicted by MIB-1 immunohistochemistry. Cancer, 85(10), 2249-2254. doi: 10.1002/(SICI)1097-0142(19990515)85:10<2249::AID-CNCR21>3.0.CO;2-5 [pii] 90. Nakano-Tateno, T., Tateno, T., Hlaing, M. M., Zheng, L., Yoshimoto, K., Yamada, S., . . . Ezzat, S. (2014). FGFR4 Polymorphic Variants Modulate Phenotypic Features of Cushing Disease. Mol Endocrinol, 28(4), 525-533. doi: 10.1210/me.2013-1412 91. Nam, S. Y., Lee, H. S., Jung, G. A., Choi, J., Cho, S. J., Kim, M. K., . . . Lee, B. L. (2003). Akt/PKB activation in gastric carcinomas correlates with clinicopathologic variables and prognosis. APMIS, 111(12), 1105-1113. 92. Nassif, N. T., Lobo, G. P., Wu, X., Henderson, C. J., Morrison, C. D., Eng, C., . . . Segelov, E. (2004). PTEN mutations are common in sporadic microsatellite stable colorectal cancer. [Research Support, Non-U.S. Gov't]. Oncogene, 23(2), 617-628. doi: 10.1038/sj.onc.1207059 93. Nishioka, H., Inoshita, N., Sano, T., Fukuhara, N., & Yamada, S. (2012). Correlation between histological subtypes and MRI findings in clinically nonfunctioning pituitary adenomas. Endocr Pathol, 23(3), 151-156. doi: 10.1007/s12022-012-9208-9 94. Noh, T. W., Jeong, H. J., Lee, M. K., Kim, T. S., Kim, S. H., & Lee, E. J. (2009). Predicting recurrence of nonfunctioning pituitary adenomas. J Clin Endocrinol Metab, 94(11), 4406-4413. doi: jc.2009-0471 [pii] 10.1210/jc.2009-0471 [doi] 95. O'Reilly, K. E., Rojo, F., She, Q. B., Solit, D., Mills, G. B., Smith, D., . . . Rosen, N. (2006). mTOR inhibition induces upstream receptor tyrosine kinase signaling and activates Akt. Cancer Res, 66(3), 1500-1508. doi: 66/3/1500 [pii] 10.1158/0008-5472.CAN-05-2925 [doi] 96. Oda, K., Okada, J., Timmerman, L., Rodriguez-Viciana, P., Stokoe, D., Shoji, K., . . . McCormick, F. (2008). PIK3CA cooperates with other phosphatidylinositol 3'-kinase pathway mutations to effect oncogenic transformation. Cancer Res, 68(19), 8127-8136. doi: 68/19/8127 [pii] 10.1158/0008-5472.CAN-08-0755 [doi] 97. Ooi, G. T., Tawadros, N., & Escalona, R. M. (2004). Pituitary cell lines and their endocrine applications. Mol Cell Endocrinol, 228(1-2), 1-21. doi: S0303-7207(04)00333-8 [pii] 10.1016/j.mce.2004.07.018 [doi] 98. Pan, L. X., Chen, Z. P., Liu, Y. S., & Zhao, J. H. (2005). Magnetic resonance imaging and biological markers in pituitary adenomas with invasion of the cavernous sinus space. J Neurooncol, 74(1), 71-76. doi: 10.1007/s11060-004-6150-9 [doi] 99. Perez-Tenorio, G., & Stal, O. (2002). Activation of AKT/PKB in breast cancer predicts a worse outcome among endocrine treated patients. Br J Cancer, 86(4), 540-545. doi: 10.1038/sj.bjc.6600126 [doi]
96
100. Prevedello, D. M., Jagannathan, J., Jane, J. A., Jr., Lopes, M. B., & Laws, E. R., Jr. (2005). Relevance of high Ki-67 in pituitary adenomas. Case report and review of the literature. Neurosurg Focus, 19(5), E11. doi: 190511 [pii] 101. Qian, Z. R., Sano, T., Asa, S. L., Yamada, S., Horiguchi, H., Tashiro, T., . . . Ezzat, S. (2004). Cytoplasmic expression of fibroblast growth factor receptor-4 in human pituitary adenomas: relation to tumor type, size, proliferation, and invasiveness. J Clin Endocrinol Metab, 89(4), 1904-1911. 102. Ramakrishnan, V. R., Suh, J. D., Lee, J. Y., O'Malley, B. W., Jr., Grady, M. S., & Palmer, J. N. (2013). Sphenoid sinus anatomy and suprasellar extension of pituitary tumors. J Neurosurg, 119(3), 669-674. doi: 10.3171/2013.3.JNS122113 103. Revillion, F., Bonneterre, J., & Peyrat, J. P. (1998). ERBB2 oncogene in human breast cancer and its clinical significance. Eur J Cancer, 34(6), 791-808. doi: S0959-8049(97)10157-5 [pii] 104. Riemenschneider, M. J., Betensky, R. A., Pasedag, S. M., & Louis, D. N. (2006). AKT activation in human glioblastomas enhances proliferation via TSC2 and S6 kinase signaling. Cancer Res, 66(11), 5618-5623. doi: 66/11/5618 [pii] 10.1158/0008-5472.CAN-06-0364 [doi] 105. Ringstad, G. A., Emblem, K. E., Holland, D., Dale, A. M., Bjornerud, A., & Hald, J. K. (2012). Assessment of pituitary adenoma volumetric change using longitudinal MR image registration. [Comparative Study]. Neuroradiology, 54(5), 435-443. doi: 10.1007/s00234-011-0894-7 106. Roelfsema, F., Biermasz, N. R., & Pereira, A. M. (2012). Clinical factors involved in the recurrence of pituitary adenomas after surgical remission: a structured review and meta-analysis. [Meta-Analysis Review]. Pituitary, 15(1), 71-83. doi: 10.1007/s11102-011-0347-7 107. Roy, H. K., Olusola, B. F., Clemens, D. L., Karolski, W. J., Ratashak, A., Lynch, H. T., & Smyrk, T. C. (2002). AKT proto-oncogene overexpression is an early event during sporadic colon carcinogenesis. Carcinogenesis, 23(1), 201-205. 108. Rudnik, A., Zawadzki, T., Galuszka-Ignasiak, B., Bazowski, P., Duda, I., Wojtacha, M., . . . Krawczyk, I. (2006). Endoscopic transsphenoidal treatment in recurrent and residual pituitary adenomas--first experience. Minim Invasive Neurosurg, 49(1), 10-14. doi: 10.1055/s-2006-932126 109. Saeger, W. (2004). Proliferation markers and cell cycle inhibitors in pituitary adenomas. [Review]. Front Horm Res, 32, 110-126. 110. Sajjad, E. A., Zielinski, G., Maksymowicz, M., Hutnik, L., Bednarczuk, T., & Wlodarski, P. (2013). mTOR is Frequently Active in GH-Secreting Pituitary Adenomas without Influencing their Morphopathological Features. Endocr Pathol. doi: 10.1007/s12022-012-9230-y [doi]
97
111. Samuels, Y., & Velculescu, V. E. (2004). Oncogenic mutations of PIK3CA in human cancers. Cell Cycle, 3(10), 1221-1224. doi: 1164 [pii] 112. Scheithauer, B. W., Kovacs, K. T., Laws, E. R., Jr., & Randall, R. V. (1986). Pathology of invasive pituitary tumors with special reference to functional classification. J Neurosurg, 65(6), 733-744. doi: 10.3171/jns.1986.65.6.0733 [doi] 113. Serra, S., Zheng, L., Hassan, M., Phan, A. T., Woodhouse, L. J., Yao, J. C., . . . Asa, S. L. (2012). The FGFR4-G388R single-nucleotide polymorphism alters pancreatic neuroendocrine tumor progression and response to mTOR inhibition therapy. [Research Support, Non-U.S. Gov't]. Cancer Res, 72(22), 5683-5691. doi: 10.1158/0008-5472.CAN-12-2102 114. Shayesteh, L., Lu, Y., Kuo, W. L., Baldocchi, R., Godfrey, T., Collins, C., . . . Gray, J. W. (1999). PIK3CA is implicated as an oncogene in ovarian cancer. Nat Genet, 21(1), 99-102. doi: 10.1038/5042 [doi] 115. Sheehan, J., Rainey, J., Nguyen, J., Grimsdale, R., & Han, S. (2011). Temozolomide-induced inhibition of pituitary adenoma cells. J Neurosurg, 114(2), 354-358. doi: 10.3171/2010.4.JNS1024 [doi] 116. Sherr, C. J. (2000). The Pezcoller lecture: cancer cell cycles revisited. Cancer Res, 60(14), 3689-3695. 117. Shrestha, R., Qi, L., Bao, G., & Wang, M. D. (2012). Prospective study of transsphenoidal pituitary surgery: is tumor volume a predictor for the residual tumor? Chin Med J (Engl), 125(14), 2444-2448. 118. Sol, Y. L., Lee, S. K., Choi, H. S., Lee, Y. H., Kim, J., & Kim, S. H. (2012). Evaluation of MRI Criteria for Cavernous Sinus Invasion in Pituitary Macroadenoma. J Neuroimaging. doi: 10.1111/j.1552-6569.2012.00710.x 119. Sukumari-Ramesh, S., Singh, N., Dhandapani, K. M., & Vender, J. R. (2011). mTOR inhibition reduces cellular proliferation and sensitizes pituitary adenoma cells to ionizing radiation. Surg Neurol Int, 2, 22. doi: 10.4103/2152-7806.77029 [doi] 120. Tanaka, Y., Hongo, K., Tada, T., Sakai, K., Kakizawa, Y., & Kobayashi, S. (2003). Growth pattern and rate in residual nonfunctioning pituitary adenomas: correlations among tumor volume doubling time, patient age, and MIB-1 index. J Neurosurg, 98(2), 359-365. doi: 10.3171/jns.2003.98.2.0359 [doi] 121. Tanno, S., Yanagawa, N., Habiro, A., Koizumi, K., Nakano, Y., Osanai, M., . . . Kohgo, Y. (2004). Serine/threonine kinase AKT is frequently activated in human bile duct cancer and is associated with increased radioresistance. Cancer Res, 64(10), 3486-3490. doi: 10.1158/0008-5472.CAN-03-1788 [doi] 64/10/3486 [pii] 122. Tateno, T., Asa, S. L., Zheng, L., Mayr, T., Ullrich, A., & Ezzat, S. (2011). The FGFR4-G388R polymorphism promotes mitochondrial STAT3 serine phosphorylation to facilitate pituitary growth hormone cell tumorigenesis. [Research Support, Non-U.S. Gov't]. PLoS Genet, 7(12), e1002400. doi: 10.1371/journal.pgen.1002400
98
123. Thapar, K., Kovacs, K., Scheithauer, B. W., Stefaneanu, L., Horvath, E., Pernicone, P. J., . . . Laws, E. R., Jr. (1996). Proliferative activity and invasiveness among pituitary adenomas and carcinomas: an analysis using the MIB-1 antibody. Neurosurgery, 38(1), 99-106; discussion 106-107. 124. Theodoropoulou, M., Zhang, J., Laupheimer, S., Paez-Pereda, M., Erneux, C., Florio, T., . . . Stalla, G. K. (2006). Octreotide, a somatostatin analogue, mediates its antiproliferative action in pituitary tumor cells by altering phosphatidylinositol 3-kinase signaling and inducing Zac1 expression. Cancer Res, 66(3), 1576-1582. doi: 66/3/1576 [pii]10.1158/0008-5472.CAN-05-1189 [doi] 125. Trouillas, J., Roy, P., Sturm, N., Dantony, E., Cortet-Rudelli, C., Viennet, G., . . . Tabarin, A. (2013). A new prognostic clinicopathological classification of pituitary adenomas: a multicentric case-control study of 410 patients with 8 years post-operative follow-up. [Multicenter Study Research Support, Non-U.S. Gov't]. Acta Neuropathol, 126(1), 123-135. doi: 10.1007/s00401-013-1084-y 126. van Weeren, P. C., de Bruyn, K. M., de Vries-Smits, A. M., van Lint, J., & Burgering, B. M. (1998). Essential role for protein kinase B (PKB) in insulin-induced glycogen synthase kinase 3 inactivation. Characterization of dominant-negative mutant of PKB. J Biol Chem, 273(21), 13150-13156. 127. Vieira, J. O., Jr., Cukiert, A., & Liberman, B. (2006). Evaluation of magnetic resonance imaging criteria for cavernous sinus invasion in patients with pituitary adenomas: logistic regression analysis and correlation with surgical findings. Surg Neurol, 65(2), 130-135; discussion 135. doi: S0090-3019(05)00379-4 [pii] 10.1016/j.surneu.2005.05.021 [doi] 128. Vlotides, G., Siegel, E., Donangelo, I., Gutman, S., Ren, S. G., & Melmed, S. (2008). Rat prolactinoma cell growth regulation by epidermal growth factor receptor ligands. Cancer Res, 68(15), 6377-6386. doi: 68/15/6377 [pii] 10.1158/0008-5472.CAN-08-0508 [doi] 129. Wera, S., Belayew, A., & Martial, J. A. (1995). Rapamycin, FK506 and cyclosporin A inhibit human prolactin gene expression. FEBS Lett, 358(2), 158-160. doi: 0014-5793(94)01416-X [pii] 130. Wera, S., Zheng, L., Hooghe-Peters, E. L., Belayew, A., Martial, J. A., & Velkeniers, B. (1995). Cyclosporin A, rapamycin and FK506 decrease prolactin release from rat pituitary cells in primary culture. Endocr Res, 21(3), 623-633. White, M. F. (1998). The IRS-signalling system: a network of docking proteins that mediate insulin action. Mol Cell Biochem, 182(1-2), 3-11. 131. Widhalm, G., Wolfsberger, S., Preusser, M., Fischer, I., Woehrer, A., Wunderer, J., . . . Knosp, E. (2009). Residual nonfunctioning pituitary adenomas: prognostic value of MIB-1 labeling index for tumor progression. J Neurosurg, 111(3), 563-571. doi: 10.3171/2008.4.17517 [doi] 132. Wolfsberger, S., Wunderer, J., Zachenhofer, I., Czech, T., Bocher-Schwarz, H. G., Hainfellner, J., & Knosp, E. (2004). Expression of cell proliferation markers in pituitary
99
adenomas--correlation and clinical relevance of MIB-1 and anti-topoisomerase-IIalpha. Acta Neurochir (Wien), 146(8), 831-839. doi: 10.1007/s00701-004-0298-0 133. Yamamoto, S., Tomita, Y., Hoshida, Y., Morooka, T., Nagano, H., Dono, K., . . . Aozasa, K. (2004). Prognostic significance of activated Akt expression in pancreatic ductal adenocarcinoma. Clin Cancer Res, 10(8), 2846-2850. 134. Yoshida, Y., Kurokawa, T., Horiuchi, Y., Sawamura, Y., Shinagawa, A., & Kotsuji, F. (2010). Localisation of phosphorylated mTOR expression is critical to tumour progression and outcomes in patients with endometrial cancer. Eur J Cancer, 46(18), 3445-3452. doi: S0959-8049(10)00869-5 [pii] 10.1016/j.ejca.2010.09.004 [doi] 135. Yoshimoto, M., Cunha, I. W., Coudry, R. A., Fonseca, F. P., Torres, C. H., Soares, F. A., & Squire, J. A. (2007). FISH analysis of 107 prostate cancers shows that PTEN genomic deletion is associated with poor clinical outcome. Br J Cancer, 97(5), 678-685. doi: 6603924 [pii] 10.1038/sj.bjc.6603924 [doi] 136. Yushkevich, P. A., Piven, J., Hazlett, H. C., Smith, R. G., Ho, S., Gee, J. C., & Gerig, G. (2006). User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage, 31(3), 1116-1128. doi: S1053-8119(06)00063-2 [pii]10.1016/j.neuroimage.2006.01.015 [doi] 137. Zada, G., Lin, N., & Laws, E. R., Jr. (2010). Patterns of extrasellar extension in growth hormone-secreting and nonfunctional pituitary macroadenomas. Neurosurg Focus, 29(4), E4. doi: 10.3171/2010.7.FOCUS10155 [doi] 138. Zatelli, M. C., Minoia, M., Filieri, C., Tagliati, F., Buratto, M., Ambrosio, M. R., . . . Degli Uberti, E. C. (2010). Effect of everolimus on cell viability in nonfunctioning pituitary adenomas. J Clin Endocrinol Metab, 95(2), 968-976. doi: jc.2009-1641 [pii] 10.1210/jc.2009-1641 [doi] 139. Zeng, J., See, A. P., Aziz, K., Thiyagarajan, S., Salih, T., Gajula, R. P., . . . Lim, M. (2011). Nelfinavir induces radiation sensitization in pituitary adenoma cells. Cancer Biol Ther, 12(7), 657-663. doi: 17172 [pii] 140. Zhang, Y., Gao, X., Saucedo, L. J., Ru, B., Edgar, B. A., & Pan, D. (2003). Rheb is a direct target of the tuberous sclerosis tumour suppressor proteins. Nat Cell Biol, 5(6), 578-581. doi: 10.1038/ncb999 [doi] ncb999 [pii] 141. Zhao, D., Tomono, Y., & Nose, T. (1999). Expression of P27kip1 and Ki-67 in pituitary adenomas: an investigation of marker of adenoma invasiveness. Acta Neurochir (Wien), 141(2), 187-192. 142. Abbass SA, Asa SL & Ezzat S 1997 Altered expression of fibroblast growth factor receptors in human pituitary adenomas. J Clin Endocrinol Metab 82 1160-1166.
100
143. Alahmadi H, Dehdashti AR & Gentili F 2012 Endoscopic endonasal surgery in recurrent and residual pituitary adenomas after microscopic resection. World Neurosurg 77 540-547. 144. Alexander JM, Biller BM, Bikkal H, Zervas NT, Arnold A & Klibanski A 1990 Clinically nonfunctioning pituitary tumors are monoclonal in origin. J Clin Invest 86 336-340. 145. Asa SL & Ezzat S 1998 The cytogenesis and pathogenesis of pituitary adenomas. Endocr Rev 19 798-827. 146. Asa SL & Ezzat S 2002 The pathogenesis of pituitary tumours. Nat Rev Cancer 2 836-849. 147. Asa SL & Ezzat S 2009 The pathogenesis of pituitary tumors. Annu Rev Pathol 4 97-126. 148. Bange J, Prechtl D, Cheburkin Y, Specht K, Harbeck N, Schmitt M, Knyazeva T, Muller S, Gartner S, Sures I, et al. 2002 Cancer progression and tumor cell motility are associated with the FGFR4 Arg(388) allele. Cancer Res 62 840-847. 149. Benveniste RJ, King WA, Walsh J, Lee JS, Delman BN & Post KD 2005 Repeated transsphenoidal surgery to treat recurrent or residual pituitary adenoma. J Neurosurg 102 1004-1012. 150. Bertagna X, Guignat L, Groussin L & Bertherat J 2009 Cushing's disease. Best Pract Res Clin Endocrinol Metab 23 607-623. 151. Bilezikjian LM, Blount AL, Leal AM, Donaldson CJ, Fischer WH & Vale WW 2004 Autocrine/paracrine regulation of pituitary function by activin, inhibin and follistatin. Mol Cell Endocrinol 225 29-36. 152. Bonneville JF, Bonneville F & Cattin F 2005 Magnetic resonance imaging of pituitary adenomas. Eur Radiol 15 543-548. 153. Brito LP, Lerario AM, Bronstein MD, Soares IC, Mendonca BB & Fragoso MC 2010 Influence of the fibroblast growth factor receptor 4 expression and the G388R functional polymorphism on Cushing's disease outcome. J Clin Endocrinol Metab 95 E271-279. 154. Cappabianca P, Alfieri A, Colao A, Cavallo LM, Fusco M, Peca C, Lombardi G & de Divitiis E 2000 Endoscopic endonasal transsphenoidal surgery in recurrent and residual pituitary adenomas: technical note. Minim Invasive Neurosurg 43 38-43. 155. Chacko G, Chacko AG, Kovacs K, Scheithauer BW, Mani S, Muliyil JP & Seshadri MS 2010 The clinical significance of MIB-1 labeling index in pituitary adenomas. Pituitary 13 337-344. 156. Choo AY, Yoon SO, Kim SG, Roux PP & Blenis J 2008 Rapamycin differentially inhibits S6Ks and 4E-BP1 to mediate cell-type-specific repression of mRNA translation. Proc Natl Acad Sci U S A 105 17414-17419.
101
157. Cottier JP, Destrieux C, Brunereau L, Bertrand P, Moreau L, Jan M & Herbreteau D 2000 Cavernous sinus invasion by pituitary adenoma: MR imaging. Radiology 215 463-469. 158. Dallapiazza RF, Grober Y, Starke RM, Laws ER, Jr. & Jane JA, Jr. 2014 Long-term Results of Endonasal Endoscopic Transsphenoidal Resection of Nonfunctioning Pituitary Macroadenomas. Neurosurgery. 159. Ekramullah SM, Saitoh Y, Arita N, Ohnishi T & Hayakawa T 1996 The correlation of Ki-67 staining indices with tumour doubling times in regrowing non-functioning pituitary adenomas. Acta Neurochir (Wien) 138 1449-1455. 160. Ezzat S, Smyth HS, Ramyar L & Asa SL 1995 Heterogenous in vivo and in vitro expression of basic fibroblast growth factor by human pituitary adenomas. J Clin Endocrinol Metab 80 878-884. 161. Ezzat S, Zheng L & Asa SL 2004 Pituitary tumor-derived fibroblast growth factor receptor 4 isoform disrupts neural cell-adhesion molecule/N-cadherin signaling to diminish cell adhesiveness: a mechanism underlying pituitary neoplasia. Mol Endocrinol 18 2543-2552. 162. Ezzat S, Zheng L, Winer D & Asa SL 2006 Targeting N-cadherin through fibroblast growth factor receptor-4: distinct pathogenetic and therapeutic implications. Mol Endocrinol 20 2965-2975. 163. Ezzat S, Zheng L, Zhu XF, Wu GE & Asa SL 2002 Targeted expression of a human pituitary tumor-derived isoform of FGF receptor-4 recapitulates pituitary tumorigenesis. J Clin Invest 109 69-78. 164. Gorshtein A, Rubinfeld H, Kendler E, Theodoropoulou M, Cerovac V, Stalla GK, Cohen ZR, Hadani M & Shimon I 2009 Mammalian target of rapamycin inhibitors rapamycin and RAD001 (everolimus) induce anti-proliferative effects in GH-secreting pituitary tumor cells in vitro. Endocr Relat Cancer 16 1017-1027. 165. Greenman Y & Stern N 2009 Non-functioning pituitary adenomas. Best Pract Res Clin Endocrinol Metab 23 625-638. 166. Hayhurst C, Monsalves E, Bernstein M, Gentili F, Heydarian M, Tsao M, Schwartz M, van Prooijen M, Millar BA, Menard C, et al. 2012 Predicting nonauditory adverse radiation effects following radiosurgery for vestibular schwannoma: a volume and dosimetric analysis. Int J Radiat Oncol Biol Phys 82 2041-2046. 167. Hewedi IH, Osman WM & El Mahdy MM 2011 Differential expression of cyclin D1 in human pituitary tumors: relation to MIB-1 and p27/Kip1 labeling indices. J Egypt Natl Canc Inst 23 171-179. 168. Honegger J, Prettin C, Feuerhake F, Petrick M, Schulte-Monting J & Reincke M 2003 Expression of Ki-67 antigen in nonfunctioning pituitary adenomas: correlation with growth velocity and invasiveness. J Neurosurg 99 674-679.
102
169. Hsu CY, Guo WY, Chien CP & Ho DM 2010 MIB-1 labeling index correlated with magnetic resonance imaging detected tumor volume doubling time in pituitary adenoma. Eur J Endocrinol 162 1027-1033. 170. Jain AK, Gupta AK, Pathak A, Bhansali A & Bapuraj JR 2008 Endonasal transsphenoidal pituitary surgery: is tumor volume a key factor in determining outcome? Am J Otolaryngol 29 48-50. 171. Jouanneau E, Wierinckx A, Ducray F, Favrel V, Borson-Chazot F, Honnorat J, Trouillas J & Raverot G 2012 New targeted therapies in pituitary carcinoma resistant to temozolomide. Pituitary 15 37-43. 172. Juraschka K, Khan OH, Godoy BL, Monsalves E, Kilian A, Krischek B, Ghare A, Vescan A, Gentili F & Zadeh G 2014 Endoscopic endonasal transsphenoidal approach to large and giant pituitary adenomas: institutional experience and predictors of extent of resection. J Neurosurg 121 75-83. 173. Kasaian K, Wiseman SM, Thiessen N, Mungall KL, Corbett RD, Qian JQ, Nip KM, He A, Tse K, Chuah E, et al. 2013 Complete genomic landscape of a recurring sporadic parathyroid carcinoma. J Pathol 230 249-260. 174. Kawamoto H, Uozumi T, Kawamoto K, Arita K, Yano T & Hirohata T 1995 Analysis of the growth rate and cavernous sinus invasion of pituitary adenomas. Acta Neurochir (Wien) 136 37-43. 175. Knosp E, Steiner E, Kitz K & Matula C 1993 Pituitary adenomas with invasion of the cavernous sinus space: a magnetic resonance imaging classification compared with surgical findings. Neurosurgery 33 610-617; discussion 617-618. 176. Lake MG, Krook LS & Cruz SV 2013 Pituitary adenomas: an overview. Am Fam Physician 88 319-327. 177. Lindholm J, Nielsen EH, Bjerre P, Christiansen JS, Hagen C, Juul S, Jorgensen J, Kruse A, Laurberg P & Stochholm K 2006 Hypopituitarism and mortality in pituitary adenoma. Clin Endocrinol (Oxf) 65 51-58. 178. Losa M, Mortini P, Barzaghi R, Ribotto P, Terreni MR, Marzoli SB, Pieralli S & Giovanelli M 2008 Early results of surgery in patients with nonfunctioning pituitary adenoma and analysis of the risk of tumor recurrence. J Neurosurg 108 525-532. 179. Lundin P, Nyman R, Burman P, Lundberg PO & Muhr C 1992 MRI of pituitary macroadenomas with reference to hormonal activity. Neuroradiology 34 43-51. 180. Lwu S, Goetz P, Monsalves E, Aryaee M, Ebinu J, Laperriere N, Menard C, Chung C, Millar BA, Kulkarni AV, et al. 2013 Stereotactic radiosurgery for the treatment of melanoma and renal cell carcinoma brain metastases. Oncol Rep 29 407-412. 181. Marin F, Stefaneanu L & Kovacs K 1991 Folliculostellate cells of the pituitary. Endocr Pathol 2 180-192.
103
182. Mete O & Asa SL 2012 Clinicopathological correlations in pituitary adenomas. Brain Pathol 22 443-453. 183. Mete O, Ezzat S & Asa SL 2012 Biomarkers of aggressive pituitary adenomas. J Mol Endocrinol 49 R69-78. 184. Monsalves E, Larjani S, Loyola Godoy B, Juraschka K, Carvalho F, Kucharczyk W, Kulkarni A, Mete O, Gentili F, Ezzat S, et al. 2014 Growth patterns of pituitary adenomas and histopathological correlates. J Clin Endocrinol Metab 99 1330-1338 185. Moreau L, Cottier JP, Bertrand P, Destrieux C, Jan M, Sonier CB, Herbreteau D & Rouleau P 1998 [MRI diagnosis of sinus cavernous invasion by pituitary adenomas]. J Radiol 79 241-246. 186. Morita K, Takano K, Yasufuku-Takano J, Yamada S, Teramoto A, Takei M, Osamura RY, Sano T & Fujita T 2008 Expression of pituitary tumour-derived, N-terminally truncated isoform of fibroblast growth factor receptor 4 (ptd-FGFR4) correlates with tumour invasiveness but not with G-protein alpha subunit (gsp) mutation in human GH-secreting pituitary adenomas. Clin Endocrinol (Oxf) 68 435-441. 187. Musat M, Korbonits M, Kola B, Borboli N, Hanson MR, Nanzer AM, Grigson J, Jordan S, Morris DG, Gueorguiev M, et al. 2005 Enhanced protein kinase B/Akt signalling in pituitary tumours. Endocr Relat Cancer 12 423-433. 188. Nakaguchi H, Fujimaki T, Matsuno A, Matsuura R, Asai A, Suzuki I, Sasaki T & Kirino T 1999 Postoperative residual tumor growth of meningioma can be predicted by MIB-1 immunohistochemistry. Cancer 85 2249-2254. 189. Nakano-Tateno T, Tateno T, Hlaing MM, Zheng L, Yoshimoto K, Yamada S, Asa SL & Ezzat S 2014 FGFR4 Polymorphic Variants Modulate Phenotypic Features of Cushing Disease. Mol Endocrinol 28 525-533. 190. Noh TW, Jeong HJ, Lee MK, Kim TS, Kim SH & Lee EJ 2009 Predicting recurrence of nonfunctioning pituitary adenomas. J Clin Endocrinol Metab 94 4406-4413. 191. Pei L, Melmed S, Scheithauer B, Kovacs K & Prager D 1994 H-ras mutations in human pituitary carcinoma metastases. J Clin Endocrinol Metab 78 842-846. 192. Prevedello DM, Jagannathan J, Jane JA, Jr., Lopes MB & Laws ER, Jr. 2005 Relevance of high Ki-67 in pituitary adenomas. Case report and review of the literature. Neurosurg Focus 19 E11. 193. Price AK & Bridges RS 2014 The effects of bromocriptine treatment during early pregnancy on postpartum maternal behaviors in rats. Dev Psychobiol 56 1431-1437. 194. Rey-Dios R, Payner TD & Cohen-Gadol AA 2014 Pituitary macroadenoma causing symptomatic internal carotid artery compression: surgical treatment through transsphenoidal tumor resection. J Clin Neurosci 21 541-546.
104
195. Ringstad GA, Emblem KE, Holland D, Dale AM, Bjornerud A & Hald JK 2012 Assessment of pituitary adenoma volumetric change using longitudinal MR image registration. Neuroradiology 54 435-443. 196. Roelfsema F, Biermasz NR & Pereira AM 2012 Clinical factors involved in the recurrence of pituitary adenomas after surgical remission: a structured review and meta-analysis. Pituitary 15 71-83. 197. Saeger W, Ludecke B & Ludecke DK 2008 Clinical tumor growth and comparison with proliferation markers in non-functioning (inactive) pituitary adenomas. Exp Clin Endocrinol Diabetes 116 80-85. 198. Sajjad EA, Zielinski G, Maksymowicz M, Hutnik L, Bednarczuk T & Wlodarski P 2013 mTOR is frequently active in GH-secreting pituitary adenomas without influencing their morphopathological features. Endocr Pathol 24 11-19. 199. Scheithauer BW, Kovacs KT, Laws ER, Jr. & Randall RV 1986 Pathology of invasive pituitary tumors with special reference to functional classification. J Neurosurg 65 733-744. 200. Sepehrnia A, Samii M & Tatagiba M 1991 Management of intracavernous tumours: an 11-year experience. Acta Neurochir Suppl (Wien) 53 122-126. 201. Shrestha R, Qi L, Bao G & Wang MD 2012 Prospective study of transsphenoidal pituitary surgery: is tumor volume a predictor for the residual tumor? Chin Med J (Engl) 125 2444-2448. 202. Sol YL, Lee SK, Choi HS, Lee YH, Kim J & Kim SH 2012 Evaluation of MRI Criteria for Cavernous Sinus Invasion in Pituitary Macroadenoma. J Neuroimaging. 203. Tanaka Y, Hongo K, Tada T, Sakai K, Kakizawa Y & Kobayashi S 2003 Growth pattern and rate in residual nonfunctioning pituitary adenomas: correlations among tumor volume doubling time, patient age, and MIB-1 index. J Neurosurg 98 359-365. 204. Tateno T, Asa SL, Zheng L, Mayr T, Ullrich A & Ezzat S 2011 The FGFR4-G388R polymorphism promotes mitochondrial STAT3 serine phosphorylation to facilitate pituitary growth hormone cell tumorigenesis. PLoS Genet 7 e100240 205. Thussbas C, Nahrig J, Streit S, Bange J, Kriner M, Kates R, Ulm K, Kiechle M, Hoefler H, Ullrich A, et al. 2006 FGFR4 Arg388 allele is associated with resistance to adjuvant therapy in primary breast cancer. J Clin Oncol 24 3747-3755. 206. Trouillas J, Roy P, Sturm N, Dantony E, Cortet-Rudelli C, Viennet G, Bonneville JF, Assaker R, Auger C, Brue T, et al. 2013 A new prognostic clinicopathological classification of pituitary adenomas: a multicentric case-control study of 410 patients with 8 years post-operative follow-up. Acta Neuropathol 126 123-135. 207. Urruticoechea A, Smith IE & Dowsett M 2005 Proliferation marker Ki-67 in early breast cancer. J Clin Oncol 23 7212-7220.
105
208. Vallar L, Spada A & Giannattasio G 1987 Altered Gs and adenylate cyclase activity in human GH-secreting pituitary adenomas. Nature 330 566-568 209. Vieira JO, Jr., Cukiert A & Liberman B 2006 Evaluation of magnetic resonance imaging criteria for cavernous sinus invasion in patients with pituitary adenomas: logistic regression analysis and correlation with surgical findings. Surg Neurol 65 130-135; discussion 135. 210. Wang J, Stockton DW & Ittmann M 2004 The fibroblast growth factor receptor-4 Arg388 allele is associated with prostate cancer initiation and progression. Clin Cancer Res 10 6169-6178. 211. Woodworth GF, Patel KS, Shin B, Burkhardt JK, Tsiouris AJ, McCoul ED, Anand VK & Schwartz TH 2014 Surgical outcomes using a medial-to-lateral endonasal endoscopic approach to pituitary adenomas invading the cavernous sinus. J Neurosurg 120 1086-1094. 212. Zada G, Lin N & Laws ER, Jr. 2010 Patterns of extrasellar extension in growth hormone-secreting and nonfunctional pituitary macroadenomas. Neurosurg Focus 29 E4. 213. Zatelli MC, Minoia M, Filieri C, Tagliati F, Buratto M, Ambrosio MR, Lapparelli M, Scanarini M & Degli Uberti EC 2010 Effect of everolimus on cell viability in nonfunctioning pituitary adenomas. J Clin Endocrinol Metab 95 968-976. 214. Zhang L, Wu X, Yan Y, Qian J, Lu Y & Luo C 2014 Preoperative somatostatin analogs treatment in acromegalic patients with macroadenomas. A meta-analysis. Brain Dev. 215. Zhao D, Tomono, Y., Nose, T. 1999 Expression of p27kip1 and Ki-67 in Pituitary Adenomas: An Investigation of Marker of Adenoma Invasiveness. Acta Neurochirurgica 141 187-192.
106
107
Appendix 1: Human Primary Culture Protocol Human Pituitary Adenoma Primary Culture Protocol Day 1
1. Obtain autoclaved scissors and forceps; one 60 mm plate; one 30 mm
plate; syringe
2. Take tumor falcon tube (FT) and record information (tissue bank # and
MRN)
3. Aspirate most of the old media from the tumor FT and pour onto 60 mm
plate
4. Cut tumor with scissors holding with forceps
5. Obtain 10 mL FT and transfer minced tissue from plate to FT using a
pipette
6. Centrifuge FT for 5 min at 1000rpm
7. While centrifuging, obtain collagenase type v and poly-L-lysine (sigma) as
well as 3 chamber slides
8. After centrifugation, aspirate old media ; flick tube to break pellet
9. Put 1 mL collagenase into FT and use pipette to transfer onto 30 mm plate
10. Incubate for 30 min at 37C
11. While incubating coat each well of the chamber slides with lysine (keep for
at least 10 min and then remove)
12. Get out of incubator and put a little more of media onto plate (refer to Lei’s
protocol for making media)
13. Suck up all media and tissue and put into cell strainer with a 30 mm plate
underneath and mash the tissue
14. Transfer the strained liquid into a FT and centrifuge the tube for 5 min
15. Aspirate and then add 1.4 mL of media
16. Add 200 microL of media to each well and 50 microL of cells to each well
17. Incubate ON
Day 2 – Drug Treatment
RAD001 cat # E4040 LOT BDE103 (LC Laboratories)
TMZ – (Sigma Aldrich)
108
1. Obtain 8 falcon tubes 1(0), 2(1), 3(10), 4(100) – TMZ; 1(0), 2(1), 3(10),
4(100) – RAD001
2. For TMZ- Add 1.5 mL of DMEM to each tube and 1.5microL of drug to
each tube
3. For RAD001 add 0.9mL media:0.9 microL drug to each tube for 1, 10
microM; for 100 microM add 2.1 mL media: 2.1 microL drug
4. Aspirate old media from chamber; add pbs; aspirate PBS
5. Add 300MicroL of respective drug to each chamber; just add straight
media to insulin and IGF1 chambers
6. Incubate at 37C ON
7. Top vs side: TMZ vs P73;1/50; 1/200 vs pmTOR; TMZ and RAD001 vs
MIB1
0 1 - -
10 100 - -
C IGF1 C IGF1
100 ins 100 Ins
0 1 0 1
10 100 10 100
Day 3- Fixing the cells
1. Make 1% formalin (in PBS)
2. Obtain 2 FT (one for Ins another for IGF1)
3. Obtain IGF1 and Ins from 4C and spin them down before use
4. Put 10mL of DMEM into each tube
5. Add 4microL IGF1; 1microL of ins
6. Take out chamber slide from incubator
7. Aspirate old media with drug in it
8. Add 300 microL C; IGF1; Ins to respective well
9. Incubate for 10 min at 37C
10. Aspirate after incubation
109
11. Add cold PBS to each well; aspirate; add formalin to each well
12. Put in 4C for 3 hr
13. Aspirate then add PBS; put in 4C ON
110
