Review
`
`Inhibitors of the mammalian
`target of rapamycin
`
`Janet E Dancey
`Investigational Drug Branch/CTEP/DCTD, National Cancer Institute, 6130 Executive Boulevard,
`Room 7131 Rockville, MD 20852, USA
`
`The mammalian target of rapamycin (mTOR) is a downstream protein
`kinase of the phosphatidylinositol 3′-kinase–Akt signalling pathway. As a
`result of its position within this pathway and its central role in controlling
`cellular growth, mTOR is viewed as an important target for anticancer ther-
`apeutics development. Currently, the mTOR inhibitor rapamycin (sirolimus,
`Wyeth) and its derivatives temsirolimus (CCI-779, Wyeth), everolimus
`(RAD-001, Novartis Pharma AG) and AP-23573 (Ariad Pharmaceuticals) are
`being evaluated in cancer clinical trials. Preclinical studies suggest that sen-
`sitivity to mTOR inhibition may correlate with aberrant activation of the
`phosphatidylinositol 3′-kinase pathway and/or with aberrant expression of
`cell-cycle regulatory or antiapoptotic proteins. Clinical trial results show
`that mTOR inhibitors are generally well tolerated and may induce pro-
`longed stable disease and even tumour regressions in a subset of patients.
`Questions remain regarding optimal dose, schedule, patient selection and
`combination strategies for this novel class of agents.
`
`Keywords: AP-34573, everolimus, mammalian target of rapamycin, sirolimus, temsirolimus
`
`Expert Opin. Investig. Drugs (2005) 14(3):313-328
`
`1. Introduction
`
`The mammalian target of rapamycin (mTOR) is a downstream protein kinase of
`the phosphatidylinositol 3′-kinase (PI3K)–Akt signalling pathway. As a result of its
`position within this pathway and its central role in controlling cellular growth,
`mTOR is considered to be an important target for anticancer therapeutics develop-
`ment. By targeting mTOR, the immunosuppressant and antiproliferative agent
`rapamycin inhibits the signals required for cell-cycle progression, cell growth and
`proliferation in both normal and malignant cells. Currently, mTOR inhibitors
`rapamycin (sirolimus, Rapamune™, Wyeth) and its derivatives temsirolimus
`(CCI-779, Wyeth), everolimus (RAD-001, Novartis Pharma AG) and AP-23573
`(Ariad Pharmaceuticals) are being evaluated in cancer clinical trials. An additional
`agent, TAFA-93 (isotechnika), has recently entered human trials in the prevention
`of organ rejection after transplantation.
`
`2. Biochemistry of protein kinase of the phosphatidylinositol
`3′-kinase–Akt–mammalian target of rapamycin pathway
`
`mTOR is an evolutionarily conserved 290-kDa serine-threonine kinase that reg-
`ulates both cell growth and cell-cycle progression through its ability to integrate
`signals from nutrient and growth factor stimuli [1,2]. mTOR, a member of the
`PI3K-kinase-related kinase (PIKK) superfamily, is composed of 2549 amino
`acids that are grouped into highly conserved, yet functionally poorly understood,
`domains. Modelling of the tertiary structure of mTOR suggests that most of the
`protein consists of helical repeat units that may form an extended superhelical
`structure to create multiple interfaces for protein–protein interactions [3]. mTOR
`
`1. Introduction
`
`2. Biochemistry of protein kinase
`of the PI3K–Akt–mTOR pathway
`
`3. mTOR in human cancer
`
`4. Rapamycin and derivatives
`
`5. Expert opinion
`
`Ashley Publications
`www.ashley-pub.com
`
`10.1517/13543784.14.3.313 2005 Ashley Publications Ltd ISSN 1354-3784
`
`313
`
`NOVARTIS EXHIBIT 2027
`Par v. Novartis, IPR 2016-01479
`Page 1 of 16
`
`

`
`Inhibitors of the mammalian target of rapamycin
`
`Nutrients
`
`Raptor
`
`mLST8
`
`mTOR
`
`Rheb
`-GTP
`
`GF Stimulation
`
`TSC1
`TSC2
`AMPK
`LKB1
`
`PIP3
`
`Akt
`
`PDK-1
`
`PIP2
`
`PI3-K
`
`PTEN
`
`4EBP
`
`PP2A
`
`S6K
`
`eIF-4E
`
`Translation
`
`S6
`
`Transcription
`
`↑ Protein synthesis
`G1 progression
`Cell survival
`
`Figure 1. mTOR signalling. GF receptor stimulation leads to activation of PI3K and 3'-OH phosphorylation of PIP2 to generate PIP3. PIP3
`then recruits PDK-1 and Akt to the plasma membrane to be activated. The tumour suppressor phosphatase PTEN dephosphorylates PIP3
`at the D3 position of the inositol ring. Activated Akt phosphorylates and inhibits TSC, removing its inhibitory effect on Ras-related small
`GTPase Rheb, which acts as a positive upstream regulator of TOR. TSC2 is also inhibited by the presence of amino acids, thus allowing
`Rheb to activate mTOR through an unknown mechanism. Activation of mTOR in complex with other proteins such as raptor and possibly
`other proteins such as mLST8 leads to phosphorylation of eIF-4E-binding protein (4E-BP) and ribosomal protein S6 kinase 1. This
`interaction results in an increase in translation rates of a subset of mRNAs including those encoding proteins required for cell-cycle
`progression. See Section 2 for additional details.
`BP: Binding protein; elF: Eukaryotic initiation factor; GF: Growth factor; mLST8: Mammalian orthologue of LST8; mTOR: Mammalian target of rapamycin;
`PDK: Phosphatidylinositol-dependent kinase; PDK-1: Phosphatidylinositol-dependent kinase-1; PI3K: Phosphatidylinositol 3′-kinase; PIP2: Phosphatidylinositol-4,5-
`bisphosphate; PIP3: Phosphatidylinositol-3,4,5-triphosphate; PP2A: Protein phosphatase 2A; PTEN: Phosphatase and tensin homologue; Raptor: Regulatory-associated
`protein of mTOR; Rheb: Ras homologue enriched in brain; SC: Tuberous sclerosis complex.
`
`functions in a protein complex that integrates signals from
`a variety of sources, including growth factors, energy stores
`and hypoxia, with the protein translation apparatus [4].
`There remains a number of unresolved questions regarding
`the function and the mechanism of action of mTOR. For
`example, it is not clear whether the intrinsic kinase activity
`of mTOR is sufficient for its full activity in vivo or if some
`mTOR functions may be mediated through protein interac-
`tions independent of its kinase function [5]. Furthermore, it
`is not clear whether mTOR may also serve as a scaffold for
`other proteins with catalytic activity, such as kinases and
`phosphatases that may regulate its overall activity in vivo [6].
`However, it does seem clear that not all the cellular activi-
`ties of mTOR are sensitive to inhibition by rapamycin [5].
`Thus, mTOR function in normal and malignant cells
`requires further elucidation, and a further understanding of
`these functions is likely to lead to optimal strategies for
`therapeutic interventions.
`The pathway from growth factor receptor stimulation to
`mTOR activation proceeds through and in parallel to PI3-K
`and Akt (Figure 1). In response to extracellular stimuli, PI3K
`phosphorylates the 3′-hydroxyl of phosphatidylinositol-4,5-
`bisphosphate (PIP2) to generate phosphatidylinositol-3,4,5-
`triphosphate (PIP3). The formation of PIP3 leads to the bind-
`ing through their pleckstrin homology (PH) regions and
`
`kinase-1
`phosphatidylinositol-dependent
`of
`activation
`(PDK-1) and Akt at the plasma membrane [1]. The tumour
`suppressor phosphatase PTEN (phosphatase and tensin
`homologue deleted on chromosome 10) dephosphorylates
`PIP3 at the 3′ position of the inositol ring, reversing the action
`of PI3K. Activated Akt phosphorylates and inhibits the tuber-
`ous sclerosis complex (TSC). In mammals, TSC1 (hamartin)
`and TSC2 (tuberin) associate to form a heterodimer that
`inhibits cell-cycle progression and cell proliferation [7,8] that, at
`least in part, is mediated through mTOR inhibition. A current
`model suggests that PI3K-dependent activation of Akt results
`in the phosphorylation and inactivation of TSC2 in mamma-
`lian cells [9]. Following its phosphorylation, TSC2 destabilises,
`disrupting the formation of functional TSC1/2 complexes and
`removing their inhibitory effect on mTOR [10-12]. Conversely,
`energy deprivation activates the tumour suppressor gene prod-
`uct LKB1, which in turn phosphorylates and activates TSC2
`[13,14]. TSC2 acts as a GTPase-activating protein (GAP)
`toward the Ras-related small GTPase Rheb (Ras homologue
`enriched in brain) [15], a positive upstream regulator of
`mTOR. Therefore, activation of TSC2 by LKB1, as may
`occur in a nutrient-deprived state, inhibits Rheb and results in
`the downregulation of mTOR. In contrast, inhibition of
`TSC2 as occurs in the presence of amino acids, as well as
`binding to 14-3-3 proteins [16-18], with Akt phosphorylation
`
`314
`
`Expert Opin. Investig. Drugs (2005) 14(3)
`
`NOVARTIS EXHIBIT 2027
`Par v. Novartis, IPR 2016-01479
`Page 2 of 16
`
`

`
`or through loss of TSC2 function through mutation, as occurs
`in TS patients, allows Rheb to activate mTOR and leads to the
`phosphorylation of the downstream mTOR targets [8]. The
`molecular mechanism by which Rheb activates mTOR is cur-
`rently unknown. The net result of these signalling interactions
`suggests a model in which growth factor signalling through
`PI3K–Akt is coordinated with nutrient availability signalling
`through LKB1–TSC1/2 to Rheb and mTOR.
`mTOR functions in a complex with at least two other pro-
`teins: regulatory associated protein of mTOR (raptor) [19,20]
`and mammalian orthologue of LST8 (mLST8, also known as
`G-protein β-subunit-like protein [GβL]) [20,21]. The function
`of raptor is not completely understood; however, it appears
`to be essential for mTOR signalling in vivo and is critical for
`mTOR substrate phosphorylation in vitro [6]. One model
`that has been proposed to explain the role of mTOR–raptor–
`mLST8 suggests that a change in the configuration of the
`mTOR–raptor complex, which is mediated by nutrient con-
`ditions such as amino acid availability, affects the ability of
`mTOR to interact with and phosphorylate its substrates [6].
`In the absence of amino acids, the mTOR–raptor–mLST8
`complex prevents mTOR from binding to its substrates and/
`or prevents the access of mTOR (or mTOR-associated
`kinases) to the substrates [6]. In the presence of amino acids, a
`conformational change promotes the interaction between
`raptor and mTOR substrates and/or increases access of the
`substrates to mTOR and its associated kinases [6]. Further
`studies are required to address how amino acids elicit these
`proposed changes in mTOR–raptor complex and verify this
`model [6]. However, current evidence suggests that activated
`mTOR, in complex with raptor, and possibly other proteins,
`leads to phosphorylation of two key proteins; eukaryotic ini-
`tiation factor-4E (eIF-4E) binding protein 1 (4E-BP1) and
`protein S6 kinase 1 (S6K1) [3,22]. S6K1 or 4E-BP1 phosphor-
`ylation is often used as in vitro and in vivo readouts of
`mTOR activity (described in this section) in laboratory and
`clinical research studies.
`Unphosphorylated 4E-BP1 binds to RNA cap-binding
`protein eIF-4E, inhibiting its coupling to mRNA methyl-7
`GpppN cap and the multi-protein translational–initiation
`complex, required for initiating translation of cap-dependent
`mRNAs, a subset of mRNAs with regulatory elements located
`in the 5′-untranslated regions (UTRs) [23]. Stimulation of cells
`by hormones, mitogens, growth factors, cytokines and G-pro-
`tein-coupled agonists results in the activation of mTOR, lead-
`ing to multi-site phosphorylation of 4E-BP1, release of eIF4E
`to bind to cap mRNA transcripts and other initiation com-
`plex proteins and the initiation of cap-dependent translation.
`This interaction results in an increase in translation rates of
`cap-dependent mRNAs, which include those encoding a
`number of proteins required for cell-cycle progression, such as
`mRNAs that encode cell-cycle regulator proteins cyclin D and
`ornithine decarboxylase [24]. This effect on translation of cer-
`tain regulatory mRNAs may be one means by which mTOR
`regulates cell growth.
`
`Dancey
`
`The second target of mTOR is the phosphorylation and
`activation of S6K1. Previously, activation of S6K1 had been
`correlated with increased translation of 5′-terminal oligopyri-
`midine tract (TOP) mRNAs, which encode components of
`the translational apparatus [25]. However, the requirement for
`S6K1 activity in translation of TOP-containing mRNAs has
`been disputed. Recent studies indicate that the translation of
`TOP mRNAs may occur independently of S6K1 function, as
`S6K1 activation is insufficient to relieve translational repres-
`sion of TOP RNAs and complete inhibition of mTOR by
`rapamycin had only a slight repressive effect on translation of
`TOP mRNAs [24,26]. These recent studies led to the conclu-
`sion that the regulation of TOP translation by growth factors
`and mitogens is primarily through the PI3K pathway with lit-
`tle role for mTOR [24]. Instead, S6K1 has been implicated in
`glucose homeostasis and regulation of eukaryotic elongation
`factor 2 kinase [24]. The exact mechanism(s) by which mTOR
`regulates translation and cell growth are complex and require
`further study.
`Consistent with its role as a central controller of cellular
`growth, mTOR activation leads to the phosphorylation of
`several downstream signalling effectors and transcription fac-
`tors in addition to its effects on 4E-BP1 and S6K1, which in
`turn influence cell proliferation, survival and angiogenesis. Of
`particular note for therapeutics development is that many,
`although not all, of the protean functions of mTOR appear to
`be sensitive to inhibition by rapamycin. The additional cellu-
`lar effects of mTOR include its direct phosphorylation of sig-
`nal transducer and activator of transcription-3 (STAT3) [27].
`Similarly, S6K1 phosphorylation of the transcription factor
`cAMP response element modulator (CREM) [27,28] is reported
`to be rapamycin sensitive. mTOR has been reported to regu-
`late autophagy [29,30]. mTOR signalling may also provide an
`antiapoptotic function; Akt-mediated protection from apop-
`tosis is mediated, at least in part, by mTOR-dependent stabi-
`lisation of cell surface amino acid transporters [31], and the
`proapoptotic protein BAD has been reported to be a substrate
`of S6K1 [32]. In addition, microarray analysis of RNA isolated
`from cells deprived of nutrients or treated with rapamycin has
`demonstrated an important role for mTOR in controlling the
`expression of genes involved in many metabolic and biosyn-
`thetic pathways [33]. Clearly, the biochemical effects of mTOR
`signalling are multiple, incompletely catalogued, poorly
`understood and potentially context specific [24]. However, the
`multitude of cellular signalling processes in which mTOR
`participates in normal and malignant cells and the inhibition
`of some of these processes by pharmacological inhibition has
`contributed to the keen interest in mTOR inhibition as a
`strategy for development of therapeutics.
`
`3. Mammalian target of rapamycin in human
`cancer
`
`Although mutations of mTOR have not been reported in
`human cancers, both aberrant PI3K-dependent signalling and
`
`Expert Opin. Investig. Drugs (2005) 14(3)
`
`315
`
`NOVARTIS EXHIBIT 2027
`Par v. Novartis, IPR 2016-01479
`Page 3 of 16
`
`

`
`Inhibitors of the mammalian target of rapamycin
`
`aberrant protein translation have been identified in a wide
`variety of malignancies and may contribute to the oncogenesis
`and malignant progression. For example, components of the
`PI3K pathway that are mutated in different human tumours
`include activation mutations of growth factor receptors,
`amplification and/or overexpression of PI3K and Akt. The
`resultant aberrant pathway signalling not only leads to a
`growth advantage during carcinogenesis and stimulates cancer
`cell proliferation but also contributes to treatment resistance
`due to a high PI3K–Akt-mediated survival threshold [1]. If
`such cancer cells are ‘addicted’ to the growth and survival-sig-
`nalling effects of the PI3K–Akt pathway, it is possible that this
`dependency will
`result
`in cancer-cell
`sensitivity
`to
`mTOR inhibition [1].
`In addition to cancer-cell dependency on aberrant PI3K
`signalling for proliferation and survival, endothelial cell pro-
`liferation may also be dependent on mTOR signalling.
`Endothelial cell proliferation
`is stimulated by vascular
`endothelial cell growth factor (VEGF) activation of the
`PI3K–Akt–mTOR pathway and, furthermore, VEGF pro-
`duction may be partly controlled by mTOR signalling
`through mTOR effects on the expression of hypoxia-induci-
`ble factor-1α (HIF-1α) [34-36]. HIF-1α is a heterodimeric
`transcription
`factor containing an
`inducibly expressed
`HIF-1α subunit and a constititutively expressed HIF-1β sub-
`unit. Under hypoxic conditions, the HIF-1α subunit accu-
`mulates due to a decrease in the rate of proteolytic
`degradation through the ubiquitin–proteasome pathway. The
`resulting HIF-1α–HIF-1β heterodimers lead to the transcrip-
`tion of certain gene products, including VEGF. Recent studies
`suggest that amplified signalling through PI3K and its down-
`stream target, mTOR, enhances HIF-1-dependent gene
`expression in vitro and that this expression is partially sensitive
`to mTOR inhibition [37]. Therefore, tumour angiogenesis
`may depend on mTOR signalling in two ways: through
`hypoxia-induced production of VEGF by tumour and stro-
`mal cells, and through VEGF stimulation of endothelial cell
`proliferation and survival through PI3K–Akt–mTOR path-
`way. Furthermore, the antitumour effects noted by inhibiting
`mTOR may be related to antiproliferative effects within
`tumour cells as well as endothelial cells.
`More directly related to mTOR effects on protein transla-
`tion in cancer cells, aberrantly high rates of protein biosynthe-
`[38], and certain tumour
`sis are observed in tumours
`suppressors and proto-oncogenes may regulate malignant pro-
`gression by altering the protein synthesis machinery [39]. For
`example, signalling through Ras and Akt acts rapidly to
`increase the association of mRNA transcripts with polyribo-
`somes, and, therefore, may
`immediately and broadly
`influence protein translation [40]. In addition, dysregulation of
`cap-dependent translation through overexpression of eIF-4E
`confers malignant characteristics and induces cancer by sup-
`pressing apoptosis in a breast-cancer model, underscoring the
`potential of therapeutics that selectively target the Akt–
`mTOR–eIF-4E pathway [41,42]. Further investigation will be
`
`needed to clarify to what extent deregulation of translation of
`total or specific mRNAs contributes to tumorigenesis. How-
`ever, deregulated components of the translational machinery,
`or specific oncoproteins that are overexpressed in cancer cells
`and that are under mTOR translational control, could be sen-
`sitive to mTOR targeted therapy. Thus, mTOR inhibition
`may lead to the inhibition of malignant progression by alter-
`ing the translation of multiple proteins including those that
`control cell size, cell-cycle progression and cell survival as well
`as angiogenesis.
`
`4. Rapamycin and derivatives
`
`Rapamycin is a macrocyclic lactone produced by Streptomy-
`ces hygroscopicus, a soil bacterium native to Easter Island
`(Rapa Nui). Rapamycin possesses fungicidal, immunosup-
`pressive and antiproliferative properties [43]. Because of its
`ability to suppress lymphocyte activation, rapamycin was
`developed and received regulatory approval as an immuno-
`suppressant for the prophylaxis of renal allograft rejection.
`Rapamycin’s immunosuppressant effects are due to its inhi-
`bition of the biochemical events required for the progression
`of IL-2-stimulated T cells from G1 to S phase of the cell
`cycle. However, rapamycin and derivatives temsirolimus,
`everolimus and AP-23573 also inhibit cellular proliferation
`in a variety of tumour models, and are currently under clini-
`cal evaluation as potential cancer therapeutics. These clinical
`studies will help to validate mTOR as an anticancer drug tar-
`get. A review of the mechanisms of action and resistance and
`a summary of clinical trial results for these agents is
`described in Section 4.1 – 4.8.
`
`4.1 Mechanism of action
`For many years, rapamycin has been used as a pharmacological
`probe; thus, much is known about its mechanism of action and
`by inference, if not actual experimentation, about the other
`agents of this pharmaceutical class. Rapamycin targets the ubiq-
`uitously expressed peptidyl-prolyl cis-trans isomerase FK506-
`binding protein of 12 kDa (FKBP12) and the rapamycin–
`FKBP12 complex binds to the FKBP12–rapamycin-binding
`(FRB) domain adjacent to the kinase domain of mTOR. The
`mechanism by which rapamycin inhibits mTOR is unclear.
`The rapamycin–FKBP12 complex may act by altering the com-
`position and/or conformation of the multi-protein mTOR
`complexes. By disrupting these protein complexes, rapamycin
`may impair either upstream signalling leading to mTOR activa-
`tion or kinase access to downstream substrates [44,45]. Rapamy-
`cin and its derivatives share the following features: inhibition of
`cellular proliferation by inducing G1 phase arrest, induction of
`apoptosis in selected models and limited normal tissue toxicity.
`The antiproliferative effects of rapamycins have been evalu-
`ated in numerous in vitro and in vivo tumour models. Results
`from these experiments indicate that these agents may inhibit
`tumour and endothelial cell proliferation in picomolar to
`nanomolar concentrations, and may add to the cytotoxicity of
`
`316
`
`Expert Opin. Investig. Drugs (2005) 14(3)
`
`NOVARTIS EXHIBIT 2027
`Par v. Novartis, IPR 2016-01479
`Page 4 of 16
`
`

`
`other chemotherapeutic agents and radiation [46-51]. The
`means by which rapamycin induces antiproliferative effects are
`not completely understood but are associated with the inhibi-
`tion of transition from G1 to S phase of the cell cycle. Among
`the reported cellular alterations in the presence of rapamycin
`include reduction of cyclins, particularly cyclin D [52,53], which
`would lead to inhibition of cyclin-dependent kinase activity
`and phosphorylation of retinoblastoma protein (pRb) as well
`as increases in cyclin-dependent kinase inhibitors p21cip1 and
`p27kip1 [54,55]. These observed effects would be consistent with
`the ability of rapamycin to impede G1 progression.
`In most instances, inhibition of mTOR by rapamycin leads
`to an antiproliferative response. However, there are examples
`in which rapamycin induces apoptosis of certain cell lines
`[56,57]. The molecular determinants of the antiproliferative ver-
`sus apoptotic response of cells after exposure to rapamycins are
`poorly understood. One proposed trigger for rapamycin-
`induced apoptosis may depend on the functions of p53,
`p27cip1 and p27kip1. In a series of studies, Huang and colleagues
`reported that rapamycin induced a cellular stress response
`characterised by rapid and sustained activation of the apoptosis
`signal-regulating kinase 1 (ASK1) signalling pathway. Selective
`apoptosis was seen in cells lacking functional p53 or p27kip1,
`and the apoptotic response correlated with and was dependent
`on 4E-BP1 expression. In contrast, wild-type p53 or p27cip1
`suppressed the apoptotic response to rapamycin [56-58]. Thus,
`the antitumour activity of rapamycin may depend on the func-
`tion of cell-cycle regulatory proteins as well as upstream cellu-
`lar signalling through mTOR. Based on these laboratory
`models, the expected clinical activity of rapamycin would be
`delayed tumour progression rather than tumour regression in
`most patients with sensitive disease. However, tumour regres-
`sion through rapamycin-induced apoptosis could occur.
`In addition to the effects on tumour cell proliferation, inhi-
`bition of mTOR by rapamycin also potently inhibits angio-
`genesis and endothelial cell proliferation in vitro and in vivo.
`Humar and colleagues showed that hypoxia directly enhances
`dose-dependent induction of DNA synthesis and cellular pro-
`liferation by platelet-derived growth factor (PDGF) and basic
`fibroblast growth factor (bFGF) and rapamycin specifically
`blocked the increase in proliferation observed under hypoxia
`in mouse and rat vascular smooth muscle and endothelial cell
`angiogenesis models [36]. The antiangiogenic properties of
`rapamycin are also associated with a decrease in VEGF pro-
`duction and a reduction in the response of vascular endo-
`thelial cells to stimulation by VEGF [36,59]. In vitro studies
`revealed that rapamycin is capable of blocking the activation
`of HIF-1 through enhanced degradation of HIF-1α [37]. In
`summary, there is considerable evidence that rapamycin is
`antiangiogeneic, and these antiangiogenic effects may be
`multifold in that the agent inhibits endothelial cell prolifera-
`tion in the presence of hypoxia, inhibits endothelial cell pro-
`liferation to VEGF stimulation through inhibition of mTOR,
`and decreases VEGF synthesis through enhanced HIF-1α
`degradation [59-61].
`
`Dancey
`
`4.2 Determinants of sensitivity and resistance to
`mammalian target of rapamycin inhibition
`Considerable research is underway to identify markers associ-
`ated with tumour cell sensitivity and resistance to rapamycin.
`Laboratory studies suggest that genetic mutations and/or
`compensatory changes leading to aberrant signal transduc-
`tion, both upstream and downstream of mTOR, influence
`tumour cell sensitivity to rapamycins [62,63]. Among the
`described abnormalities in cellular molecules that correlated
`with rapamycin resistance are mutations of mTOR or
`FKBP12 that prevent rapamycin from binding to mTOR
`[62,63]. Although such mutations have been induced in labora-
`tory models, these mutations have not been described in
`human cancers. Mutations or defects of mTOR-regulated
`proteins, including S6K1 and 4E-BP1, which impair their
`interaction with mTOR, also render cells insensitive to
`rapamycin [62]. Although inhibition of phosphorylation of
`S6K1, its target ribosomal S6 protein, and 4E-BP1 correlates
`with rapamycin sensitivity in laboratory models, inhibition
`of S6K1 and 4E-BP1 is not sufficient to guarantee sensitivity
`to rapamycin as hypophosphorylation of these mTOR targets
`has been seen in rapamycin-resistant as well as -sensitive cells
`[62,64]. Thus, additional factors must mediate cellular depend-
`ency on mTOR signalling. Expression and function of
`ataxia-telangiectasia gene product ATM, as well as 14-3-3,
`p53, PI3K–Akt and PTEN have been reported to correlate
`with rapamycin sensitivity [62,63]. Data supporting these last
`two potential mechanisms of sensitivity will be discussed in
`greater detail.
`There is considerable evidence that aberrant stimulation of
`the PI3K–Akt signalling pathway in cancer cells may increase
`the dependency of such tumours on mTOR signalling func-
`tions and their sensitivity to signal modulation by inhibiting
`mTOR [1]. Studies have reported that rapamycin and related
`compounds exert selective cytostatic/cytotoxic effects on
`PTEN-/- tumours in vivo [65,66]. In vitro and in vivo studies of
`isogenic PTEN+/+ and PTEN-/- mouse cells, as well as human
`cancer cells with defined PTEN status. showed that the
`growth of PTEN null cells was blocked preferentially by
`mTOR inhibition [65]. However, the loss of PTEN function
`does not correlate with rapamycin sensitivity in all models. In
`a series of breast cancer cell lines, overexpression of S6K1 and
`phosphorylated Akt, independent of PTEN status, was associ-
`ated with rapamycin sensitivity [64]. As noted in other studies,
`the differential sensitivity to mTOR inhibition was not
`explained by differences in biochemical blockade of the
`mTOR pathway because S6 phosphorylation was inhibited in
`sensitive and resistant cell lines. Thus, aberrant stimulation of
`the PI3K–Akt pathway, which may occur through hyper-
`action growth factor receptor, PI3K, Akt or loss of PTEN may
`be markers of tumour cell responsiveness to rapamycins.
`Loss of functional TSC, as occurs in patients with tuber-
`ous sclerosis syndrome (TS), may also confer sensitivity to
`rapamycins. Although TS-associated tumours are usually
`benign, they can affect almost every major organ system and
`
`Expert Opin. Investig. Drugs (2005) 14(3)
`
`317
`
`NOVARTIS EXHIBIT 2027
`Par v. Novartis, IPR 2016-01479
`Page 5 of 16
`
`

`
`Inhibitors of the mammalian target of rapamycin
`
`are thus responsible for considerable morbidity of affected
`individuals [7]. TSC is rarely associated with malignant
`tumours, although patients with TSC are at increased risk
`for clear-cell renal carcinoma [7,67]. Loss or reduction of the
`function of the TSC1–TSC2 protein complex in individuals
`with TSC leads to hyperactive mTOR signalling in hamar-
`tomatous tissues, implying that the loss of TSC1–TSC2 and
`dysregulation of mTOR signalling is a causative factor in
`TSC disease [67,68]. In addition, TSC2-/- p53-/- cells, as well
`as tumours from TSC2+/- mice, display an mTOR-activation
`signature with constitutive activation of S6K. Both the acti-
`vation of S6K and the growth advantage of TSC2-/- p53-/-
`cells are reverted by treatment with rapamycin [69]. Activa-
`tion of mTOR was associated with downregulation of
`PDGF receptor–PI3K–Akt signalling in cells lacking TSC1
`or TSC2. The low rate of malignant tumour development in
`TS patients may result from the negative feedback of the
`PDGF receptor–PI3K–Akt signalling pathway to balance
`activation of mTOR [69]. The observation that mTOR sig-
`nalling is upregulated in TS-associated hamartomas suggest
`that rapamycin and its derivatives may prove to be effective
`in the treatment of these patients.
`Studies evaluating the addition of rapamycin to standard
`anticancer therapy suggest that rapamycin may add to the
`cytotoxic effects of standard cancer drugs; however, tumour
`cell responsiveness to these combinations will be determined
`by the molecular phenotype of cancer cells. For example,
`Wendel and colleagues [70] used a mouse model of B-cell
`lymphoma to explore the consequences of inhibiting Akt
`signalling. They showed that Akt promotes tumorigenesis
`and cancer drug resistance by disrupting apoptosis. Lym-
`phomas expressing Akt but not those expressing Bcl-2 were
`sensitised by rapamycin to chemotherapy-induced apop-
`tosis. In another example, apoptosis induced by the c-myc
`gene is suppressed by mTOR downstream target eIF-4E [71].
`Although eIF-4E recapitulates Akt’s action in tumorigenesis
`and drug resistance, it renders cells insensitive to concomi-
`tant administration of rapamycin and chemotherapy. These
`results suggest that Akt signalling through mTOR and
`eIF-4E may be an important mechanism of oncogenesis and
`in vivo cancer drug resistance, and that targeting the apop-
`totic programme may restore drug sensitivity [70,72]. How-
`ever, cells overexpressing Akt that are exposed to rapamycin
`may be rendered sensitive to standard cancer drugs and be
`triggered to undergo apoptosis when exposed to the combi-
`nation. However, those tumour cells overexpressing eIF-4E
`or Bcl-2 may be insensitive to rapamycin and remain
`resistant to standard cancer therapies.
`Collectively, these data suggest that tumour cells sensitive to
`mTOR inhibition are
`likely to be those with aberrant
`signalling through the PI3K–Akt pathway, such as those with
`activating mutations and/or amplification of growth factor
`receptors, PI3K, Akt, or through the loss of PTEN or TSC1/2.
`Sensitive cells may be triggered to undergo G1 arrest or apopto-
`sis. The apoptotic effect may be determined, in part, by p53,
`
`p21cip1, p27kip and Bcl-2 function. Conversely, cells with amplifi-
`cation and/or overexpression of downstream target eIF-4E may
`be insensitive to rapamycins. In addition to the presence and
`function of these pathway proteins, microarray and proteomic
`analyses have identified many proteins whose levels were
`changed in response to rapamycin [33]. Interpreting the data and
`investigating the leads generated by these screens may lead to a
`clearer understanding of the control of gene expression by
`mTOR and patterns that are predictive of the sensitivity or resist-
`ance to its inhibition. Thus, predicting sensitivity or resistance of
`tumours to rapamycins will probably require the assessment of
`multiple, rather than individual, molecular markers.
`
`4.3 Preclinical anticancer activity of mammalian target
`of rapamycin inhibitors
`All the rapamycins under clinical development have anti-
`proliferative activity in a variety of haematological and solid
`tumour systems as single agents and in combination with stand-
`ard cancer therapeutic agents and radiation [43,47-49,51,73-80].
`Because of their high lipophilic indices, rapamycins easily cross
`the blood–brain barrier [47] and, thus, may be of value for the
`treatment of primary and secondary malignancies of the central
`nervous system. These studies suggest that rapamycins may have
`a role as single agents and in combination with standard
`therapies in a variety of malignancies.
`Relatively few studies have assessed various rapamycins
`under identical experimental conditions. Two studies sug-
`gest that antitumour effects of rapamycin and temsirolimus
`or everolimus are similar. Exposure of IGROV1 ovarian car-
`cinoma cells to rapamycin and everolimus in vitro was asso-
`ciated with substantial cellular apoptosis with both agents
`[81]. Co-treatment of rapamycin in vitro or temsirolimus
`in vivo inhibited mTOR activity and restored sensitivity of
`the MCF7–Akt transfected cell line to t