Oncogene (2000) 19, 6680 – 6686
`ª 2000 Macmillan Publishers Ltd All rights reserved 0950 – 9232/00 $15.00
`
`www.nature.com/onc
`
`The rapamycin-sensitive signal transduction pathway as a
`target for cancer therapy
`
`Manuel Hidalgo*,1 and Eric K Rowinsky1
`
`1The University of Texas Health Science Center at San Antonio, The Institute for Drug Development, Cancer Therapy and
`Research Center, San Antonio, Texas, USA
`
`The high frequency of mutations in cancer cells which
`result in altered cell cycle regulation and growth signal
`transduction, conferring a proliferative advantage,
`in-
`dicates that many of these aberrant mechanisms may be
`strategic targets for cancer therapy. The macrolide
`fungicide rapamycin, a natural product with potent
`antimicrobial, immunosuppressant, and anti-tumor prop-
`erties, inhibits the translation of key mRNAs of proteins
`required for cell cycle progression from G1 to S phase.
`Rapamycin binds intracellularly to the immunophilin
`FK506 binding protein 12 (FKBP12), and the resultant
`complex inhibits the protein kinase activity of a protein
`kinase termed mammalian target of rapamycin (mTOR).
`The inhibition of mTOR, in turn, blocks signals to two
`separate
`downstream pathways which
`control
`the
`translation of specific mRNAs required for cell cycle
`traverse from G1 to S phase. Blocking mTOR a€ects the
`activity of the 40S ribosomal protein S6 kinase (p70s6k)
`and the function of the eukaryotic initiation factor 4E-
`binding protein-1 (4E-BP1), leading to growth arrest in
`the the G1 phase of the cell cycle. In addition to its
`actions on p70s6k and 4E-BP1,
`rapamycin prevents
`cyclin-dependent kinase activation, inhibits retinoblasto-
`ma protein (pRb) phosphorylation, and accelerates the
`turnover of cyclin D1 that leads to a deficiency of active
`cdk4/cyclin D1 complexes, all of which can inhibit cell
`cycle traverse at
`the G1/S phase transition. Both
`rapamycin and CCI-779, an ester analog of rapamycin
`with improved pharmaceutical properties and aqueous
`solubility, have demonstrated impressive activity against
`a broad range of human cancers growing in tissue culture
`and in human tumor xenograft models, which has
`supported the development of compounds
`targeting
`rapamycin-sensitive signal-transduction pathways. CCI-
`779 has completed several phase I clinical evaluations
`and is
`currently undergoing broad disease-directed
`e(cid:129)cacy studies. The agent appears to be well tolerated
`at doses that have resulted in impressive anti-tumor
`activity in several
`types of
`refractory neoplasms.
`Important challenges during clinical development include
`the definition of a recommended dose range associated
`with optimal biological activity and maximal therapeutic
`indices, as well as the ability to predict which tumors will
`be sensitive or resistant to CCI-779. Oncogene (2000)
`19, 6680 – 6686.
`
`Keywords: rapamycin; CCI-779; signal
`clinical development
`
`transduction;
`
`*Correspondence: M Hidalgo, Department of Medicine, Division of
`Medical Oncology, The University of Texas Health Science Center at
`San Antonio, 7703 Floyd Curl Dr. Mail code 7884. San Antonio,
`Texas, TX, 78229, USA
`
`Cell proliferation is a complex multifaceted process
`that
`requires
`the synthesis of essential
`regulatory
`proteins involved in the transduction of extracellular
`and autocrine proliferative stimuli. Since several of
`these highly regulated processes are aberrant in many
`types of cancers, conferring a proliferative advantage,
`they are potential strategic targets for therapeutic
`development against cancer
`(Sherr, 2000).
`Indeed,
`several novel classes of therapeutics that interfere with
`discrete essential elements of aberrant signal transduc-
`tion and cell cycle regulation, such as inhibitors of
`various receptor tyrosine kinases, oncogenes, critical
`proteins involved in signal transduction (e.g. Ras, Raf),
`and cyclin-dependent kinases, are being developed as
`anti-cancer agents (Rowinsky et al., 1999, Senderowicz
`and Sausville 2000). One such agent,
`rapamycin
`(sirolimus; Rapamune1; Wyeth-Ayerst, PA, USA), a
`macrolide fungicide isolated from the bacteria Strepto-
`myces hygroscopicus, possesses potent antimicrobial,
`immunosuppressant, and antitumor properties (Baker
`et al., 1978; Sehgal et al., 1975; Vezina et al., 1975).
`Because of its profound immunosuppressive actions,
`rapamycin was initially developed and received reg-
`ulatory approval for the indication of prevention of
`allograft
`rejection following organ transplantation
`(Sehgal, 1995). The antiproliferative actions of rapa-
`mycin have been demonstrated to be due to its ability
`to modulate critical signal transduction pathways that
`link mitogenic stimuli
`to the synthesis of proteins
`required for
`cell
`cycle
`traverse
`from G1
`to S
`(Wiederrecht et al., 1995). Impressive antiproliferative
`activity has been demonstrated following treatment of
`a diverse types of experimental tumors with rapamycin
`(Eng et al., 1984, Muthukkumar
`et al., 1995;
`Seu€erlein and Rozengurt, 1996). However, the poor
`aqueous solubility and chemical stability of rapamycin
`precluded its clinical development as an anti-cancer
`agent. Recently, a series of rapamycin analogs with
`improved aqueous solubility and stability have been
`synthesized and evaluated. CCI-779 (Wyeth Ayerst,
`PA, USA), a soluble ester analog of rapamycin, was
`selected for development as an anti-cancer agent based
`on its prominent anti-tumor profile and favorable
`pharmaceutical and toxicological characteristics
`in
`preclinical studies (Gibbons et al., 2000). Several phase
`I studies of CCI-779 have been completed and disease-
`directed e(cid:129)cacy evaluations in a number of tumor
`types are being performed (Raymond et al., 2000;
`Hidalgo et al., 2000). This review will summarize the
`principal mechanisms of anti-tumor action of rapamy-
`cin, specifically its e€ect on rapamycin-sensitive signal
`transduction pathways, and will discuss the preliminary
`results of experimental and clinical studies with this
`novel class of anti-cancer agents.
`
`Roxane Labs., Inc.
`Exhibit 1028
`Page 001
`
`

`
`Mechanism of action of rapamcyin and
`rapamycin analogs
`
`Rapamycin, and its ester analog, CCI-779, uniquely
`interfere with cell cycle progression from G1 to S phase
`in response to proliferative stimuli by blocking the
`translation of mRNAs of essential cell cycle proteins
`(Wiederrecht et al., 1995). The principal mechanisms
`responsible
`for
`these
`actions, which have been
`elucidated only over
`the
`last
`several years, are
`graphically depicted in Figures 1 and 2.
`
`Upstream actions and the target of rapamycin
`
`Rapamycin binds intracellularly to members of the
`immunophilin family of FK506 binding proteins
`(FKBPs), inhibiting their enzymatic activity as prolyl
`isomerases (Heitman et al., 1991; Koltin et al., 1991;
`Fruman et al., 1995). Although there are many
`members of
`the FKBP family, a large body of
`biochemical and genetic studies suggest that FKBP12
`is the most important binding protein with respect to
`the rapamycin-sensitive signal
`transduction pathway
`(Heitman et al., 1991; Koltin et al., 1991; Fruman et
`al., 1995). The resultant rapamycin-FKBP12 complex
`interacts with and inhibits the activity of a 290 kd
`kinase,
`termed mammalian target of
`rapamycin
`(mTOR) (also known as FRAP, RAFT1, and RAP1)
`(Figure 1) (Sabatini et al., 1994; Sabers et al., 1995;
`Brown et al., 1994; Chiu et al., 1994). mTOR is a
`member of a recently identified family of protein
`kinases
`termed phosphoinositide
`3-kinase
`related
`kinases (PIKKs), which are involved in many critical
`regulatory cellular functions pertaining to cell cycle
`progression, cell cycle checkpoints that govern cellular
`responses to DNA damage, DNA repair, and DNA
`recombination (Sarkaria et al., 1998).
`cells
`In response
`to growth stimuli, quiescent
`increase the translation of a subset of mRNAs whose
`protein products are required for traverse through the
`G1 phase of the cell cycle. mTOR regulates essential
`signal transduction pathways and is involved in the
`
`transduction pathway.
`signal
`Figure 1 Rapamycin-sensitive
`Rapamycin and CCI-779 bind to the immunophilin FK506
`binding protein-12 (FKBP-12). The rapamycin-FKBP12 complex
`blocks the kinase activity of the mammalian target of rapamycin
`(mTOR). The inhibition of mTOR kinase activity inhibits the
`downstream translational regulators 4E-BP1/PHAS and p70s6k.
`inhibition of 4E-BP1/PHAS and p70s6k decrease
`the
`The
`translation of mRNA of specific proteins essential for cell cycle
`progression from G1 to S phase
`
`Targeting rapamycin-sensitive pathway
`M Hidalgo and EK Rowinsky
`
`GROWTH
`FACTOR
`
`mTOR
`
`6681
`
`Figure 2 Rapamycin and CCI-779 inhibits the phosphorylation
`of 4E-BP1/PHAS, preventing the release of the eIF-4E and the
`activation of the eIF4F complex
`
`coupling of growth stimuli with cell cycle progression.
`Phosphatidylinositol 3-kinase (PI3K)/protein kinase B
`(Akt) (PI3K/Akt) appears to be the key modulatory
`factor in the upstream pathway by which growth
`factor-growth factor receptor interactions a€ect
`the
`phosphorylation state of mTOR (Figure 1) (Downward
`1998; Scott et al., 1998; Nave et al., 1999). PI3K plays
`a central
`role
`in cellular proliferation, motility,
`neovascularization, viability, and senescence and is
`upregulated in cancer cells (Shayestech et al., 1999;
`Cantley et al., 1991). Its main physiological function is
`the phosphorylation of the D3 portion of membrane
`phosphoinositols (Cantley et al., 1991; Carpenter et al.,
`1990). Although the role of PI3K and its lipid products
`in signal transduction processes is not clear, the activity
`of this enzyme on tyrosine kinases induces mitogenesis,
`cellular growth, and cellular transformation (Carpenter
`et al., 1990; Varticovski et al., 1994; Hu et al., 1995).
`Recently, several studies have investigated the role of
`small molecule-inhibitors of PI3K as potential tumor
`suppressor agents. For example, the flavonoid deriva-
`tive, LY294002 (Eli Lilly, Indianapolis, IN, USA), a
`potent PI3K inhibitor,
`is a competitive, reversible
`inhibitor of
`the ATP binding site of
`the enzyme
`(Vlahos et al., 1994; Hu et al., 2000). The agent
`induces G1 arrest
`in proliferating cells,
`leading to
`almost complete inhibition of melanoma cell prolifera-
`tion, partial
`inhibition of MG-63 osteosarcoma cell
`growth, and inhibitor of OVCAR-3 ovarian carcinoma
`inducing prominent apoptotic e€ects (Hu et al., 2000;
`Casagrande et al., 1998; Thomas et al., 1997). The
`inhibitor also completely inhibits the retinoblastoma
`protein (pRb) hyperphosphorylation that normally
`occurs during G1 progression and induces up-regula-
`tion of
`the cyclin-dependent kinase inhibitor p27
`(Casagrande et al., 1998).
`There are ample experimental data indicating that
`mTOR functions downstream of the PI3K/Akt path-
`way and is phosphorylated in responses to stimuli that
`activate the PI3K/Akt pathway (Scott et al., 1998;
`Nave et al., 1999; Hu et al., 1995; Sekulic et al., 2000).
`PI3K and Akt are considered proto-oncogenes, and the
`pathway is inhibited by the tumor suppressor gene
`PTEN (Wu et al., 1998). There are other signaling
`pathways that are activated downstream of PI3K, but
`the Akt pathway is of particular interest because of its
`role
`in inhibiting apoptosis and promoting cell
`proliferation by a€ecting the phosphorylation status
`
`Oncogene
`
`Roxane Labs., Inc.
`Exhibit 1028
`Page 002
`
`

`
`Targeting rapamycin-sensitive pathway
`M Hidalgo and EK Rowinsky
`
`6682
`
`of cell-survival and apopotosis-inducing proteins like
`BAD (Downward, 1998).
`
`Downstream effects
`
`Following activation via phosphorylation, mTOR
`modulates two separate downstream pathways that
`control
`translation of
`specific subsets of mRNAs
`including the eukaryotic initiation factor 4E binding
`protein-1 (4E-BP1), also known as PHAS-1 (phos-
`phorylated heat- and acid-stable protein), and the 40S
`ribosomal protein S6 kinase (p70s6k) (Figure 1) (Brunn
`et al., 1997; Gingras et al., 1998; Hara et al., 1997).
`There is ample evidence indicating that the activation
`of either PI3K or Akt
`is su(cid:129)cient
`to induce the
`phosphorylation of both 4E-BP1/PHAS-1 and p70s6k
`through mTOR (Chung et al., 1994; Petrisch et al.,
`1995). Furthermore, treatment of activated PI3K- or
`Akt-expressing cells with rapamycin blocks the phos-
`phorylation of p70s6k and 4E-BP1/PHAS-1, suggesting
`that mTROR is required for these activities (Gingras et
`al., 1998; Burgering and Co€er, 1995).
`The first downstream regulator modulated by the
`phosphorylation status of mTOR, 4E-BP1/PHAS-1, is
`a low molecular-weight protein that
`inhibits
`the
`initiation of translation through its association with
`eIF-4E,
`the mRNA cap-binding subunit of
`the
`eukaryotic initiation factor-4 (eIF-4F) complex (Figure
`2). The binding of 4E-BPs to eIF-4E is dependent on
`the phosphorylation status of 4E-BP. In the unpho-
`sphorylated state that predominates in quiescent cells,
`4E-BP1/PHAS-1 binds to eIF-4E which inhibits its
`activity (Sonenberg et al., 1998).
`In response to
`proliferative stimuli, 4E-BP1/PHAS-1 becomes phos-
`phorylated through the action of mTOR and other
`kinases which decrease its binding a(cid:129)nity for eIF-4E.
`These actions promote the dissociation of the 4E-BP1/
`PHAS-1 complex, increasing the availability of eIF-4E,
`which can then bind to eIF-4G, -4B, and -4A, forming
`the multisubunit eIF-4F complex. These interactions
`lead to an increase in the translation of mRNAs with
`regulatory elements in the 5’-untranslated region such
`as cyclin D1 and ornithine decarboxylase (Sonenberg
`and Gingras, 1998; Rosenwald et al., 1995; Shantz and
`Pegg 1995). In contrast, growth factor deprivation or
`treatment with rapamycin results in the dephosphor-
`ylation of 4E-BP1/PHAS-1, an increase in eIF-4E
`binding, and a concomitant decrease in the translation
`of mRNAs for cell cycle progression from G1 to S
`phase, as shown in Figure 2 (Brunn et al., 1997;
`Gingras et al., 1998).
`The second downstream target modulated by mTOR
`is the kinase p70s6k. Upon activation by proliferative
`stimuli mediated by the PI3K/Akt signal transduction
`pathway, mTOR phosphorylates/activates
`p70s6k,
`which,
`in turn, phosphorylates the 40S ribosomal
`protein S6 (Hu et al., 2000). The phosphorylation of
`S6 leads to the recruitment of
`the 40S ribosomal
`subunit into actively translating polysomes,
`thereby
`enhancing the translation of mRNAs with a 5’ terminal
`oligopolypyrimidine including those that encode for
`ribosomal proteins, elongation factors, and insulin
`growth factor – II. Rapamycin treatment results in a
`rapid and profound dephosphorylation of p70s6k,
`suppressing its activity (Seu€erlein and Rozengurt
`1996; Grewe et al., 1999).
`
`Oncogene
`
`In addition to its well characterized inhibitory
`actions on p70s6k and 4E-BP1/PHAS-1,
`rapamycin
`interferes with other intracellular processes involved
`in cell cycle progression which undoubtedly contribute
`to its antiproliferative actions. These other actions are
`particularly important in exponentially growing cells
`were inhibition of p70s6k phosphorylation by rapamycin
`does not result in G1 cell cycle arrest (Kawamata et al.,
`1998). Rapamycin increases the turnover of cyclin D1
`at the mRNA and protein level (Hashemolhosseini et
`al., 1998). This e€ect,
`in addition to the decrease
`translation of cyclin D1 mRNA subsequently to 4E-
`BP1/PHAS inhibition, results in a relative deficiency of
`cyclin D1 in the cdk4/cyclin D1 complexes required for
`retinoblastoma protein phosphorylation (Morice et al.,
`1993; Nourse et al., 1994). Rapamycin also blocks the
`elimination of
`the cyclin dependent kinase (cdks)
`inhibitor p27 and facilitates the formation of
`the
`cyclin/cdks-p27 complexes (Nourse et al., 1994; Luo
`et al., 1996). Furthermore,
`in exponentially growing
`cells, rapamycin upregulates p27 at both the mRNA
`and protein level and inhibits cyclin-A-dependent
`kinase activity (Kawamata et al., 1998). It appears
`that, although rapamycin treatment results in cell cycle
`arrest at the G1/S transistion, the precise mechanism
`responsible for this e€ect is both cell cycle- and cell
`type-specific.
`By inhibiting the translation of critical mRNAs
`involved in the G1 to S phase transition in response to
`mitogenic stimuli, and by interfering with the balance
`of cyclin/cyclin-dependent-kinase/cyclin-dependent ki-
`nase inhibitors in the early phases of the cell cycle,
`rapamycin inhibits
`the
`growth of
`cancer
`cells.
`Rapamycin exerts concentration-dependent inhibition
`of cell proliferation and tumor growth in a variety of
`murine and human cancers growing in both cell culture
`and xenograft models including B16 melanoma, P388
`leukemia, MiaPaCa-2 and Panc-1 human pancreatic
`carcinoma and tumors derived from B-cell lymphoma,
`small cell
`lung cancer carcinoma, and childhood
`rhapdomyosarcoma (Muthukkumar et al., 1995; Seuf-
`ferlein and Rozengurt 1996; Hosoi et al., 1999).
`Rapamycin also induces p53-independent apoptosis in
`childhood rhabdomyosarcoma cell lines and enhances
`the apoptotic-inducing e€ects of cisplatin in murine T-
`cell, human HL-60 promyelocytic leukemia, and hu-
`man ovarian SKOV3 cancer cell
`lines (Hosoi et al.,
`1999; Shi et al., 1995).
`
`Clinical development
`
`The unfavorable pharmaceutical properties of rapamy-
`cin, particularly its poor aqueous
`solubility and
`instability, preclude its clinical development as an
`anti-cancer agent, and, therefore, soluble ester analogs
`were synthesized and evaluated as a collaborative e€ort
`between investigators at Wyeth-Ayerst and the Na-
`tional Cancer Institute. CCI-779, a water soluble ester
`analog of rapamcyin, was selected for further develop-
`ment based on favorable pharmaceutical, toxicologic,
`and antitumor profiles in preclinical evaluations. In the
`National Cancer
`Institute human tumor cell
`line
`screen, CCI-779 and rapamycin demonstrated similar
`anti-tumor profiles and potencies (Pearson correlation
`coe(cid:129)cient, 0.86), with IC50 values frequently less than
`
`Roxane Labs., Inc.
`Exhibit 1028
`Page 003
`
`

`
`Targeting rapamycin-sensitive pathway
`M Hidalgo and EK Rowinsky
`
`6683
`
`1078 M (Gibbons et al., 2000). Platelet-derived growth
`factor stimulation of
`the human glioblastoma line
`T98G was markedly inhibited (IC50, 1 pM), consistent
`with its proposed mechanism of action as an inhibitor
`of signal transduction, and growth-inhibited cells were
`arrested in G1 (Gibbons et al., 2000). In studies
`involving cancer cell
`lines growing in tissue culture,
`human prostate, breast, and small cell lung carcinomas,
`glioblastoma, melanoma, and T-cell
`leukemia were
`among the most sensitive cancers to CCI-779, with
`IC50 values in the nanomolar range (Gibbons et al.,
`2000). Significant growth inhibition was also observed
`following treatment of a variety of human tumor
`xenografts with CCI-779, but the preponderance of
`tumor growth inhibition,
`in contrast to overt tumor
`regression, suggests that subsequent disease-directed
`trials
`should be designed to assess
`this potential
`outcome. In addition, several
`intermittent CCI-779
`dosing regimens were
`e€ective
`in human tumor
`xenograft studies, which is important in view of the
`possibility that prolonged immunosuppression can
`result from both rapamycin and CCI-779 administered
`on continuous dose-schedules and since the immuno-
`suppressive e€ects of rapamycin analogs have been
`demonstrated to resolve in approximately 24 h follow-
`ing treatment (Gibbons et al., 2000).
`Thus far, CCI-779 has been evaluated in two phase I
`studies, in which the agent has been administered as a
`30-min IV infusion weekly and as a 30-min IV infusion
`daily for 5 days every 2 weeks (Raymond et al., 2000;
`Hidalgo et al., 2000). These schedules were selected for
`initial clinical evaluation to avoid prolonged drug-
`induced immunosuppression which resolves
`shortly
`after drug treatment. However, similar to the tradi-
`tional paradigm used to develop nonspecific cytotoxic
`agents,
`these phase
`I
`studies were designed to
`determine the maximum tolerated dose based on
`dose-limiting toxicities as classically defined. To date,
`16 patients have been treated with CI-779 at doses
`ranging from 7.5 to 220 mg/m2/week on the weekly
`schedule and 35 patients have received doses ranging
`from 0.75 to 24 mg/m2/day on the daily for-5-days
`every-2-week schedule. The principal toxicities of CCI-
`779 have included dermatologic toxicity, myelosuppres-
`sion, reversible elevations in liver function tests, and
`asymptomatic hypocalcemia. The cutaneous e€ects,
`which have been relatively more common with CCI-
`779 on the weekly schedule, have been multifaceted.
`Dermatologic manifestations have included aseptic
`folliculitis, erythematous maculopapular rashes, ezce-
`matoid reactions, dry skin, vesicular lesions, and nail
`disorders. The principal hematologic toxicity has been
`thrombocytopenia, whereas anemia,
`leukopenia, and
`neutropenia have generally been less common and
`severe. Other toxicities and biochemical abnormalities,
`which have generally been mild to moderate in severity,
`reversible, and noted over wide dosing ranges include
`mucositis, hypertriglyceridemia, hypercholesterolemia,
`and reversible decrements in serum testosterone. The
`maximum tolerated doses of CCI-779 on the daily-for-
`5-day-every-2-week schedule are projected to be 15 and
`24 mg/m2/day in patients with minimal or extensive
`prior myelotoxic therapy, respectively, whereas the
`maximum tolerated dose has not been determined for
`CCI-779 administered on a weekly schedule. The
`preliminary results of pharmacokinetic studies indicate
`
`dose-dependent pharmacokinetics, elimination half-life
`values of approximately 15 – 17 h, and preferential
`partitioning of CCI-779 in red blood cells (Raymond
`et al., 2000; Hidalgo et al., 2000). Major tumor
`responses (partial responses; 450% reduction in the
`sum of the bidimensional product of all measurable
`lesions) have been noted in previously-treated patients
`with renal cell carcinoma and non-small cell
`lung
`(550%
`carcinoma,
`and minor
`tumor
`responses
`reduction in the sum of the bidimensional product of
`all measurable lesions) have been observed in pre-
`viously-treated patients with soft tissue sarcoma, serous
`papillary carcinoma of
`the
`endometrium, breast
`carcinoma, squamous cell carcinoma of the skin, and
`non-Hodgkins
`lymphoma. The fact
`that CCI-779
`consistently induced tumor regressions at relatively
`nontoxic doses in the phase I studies is particularly
`noteworthy, since this observation suggests that the
`optimal therapeutic dose of CCI-779 may be lower
`than the maximum tolerated dose.
`Disease-directed e(cid:129)cacy studies of CCI-779 in a
`broad range of tumor types will be initiated following
`the completion of phase I studies. Based on the results
`of experimental studies involving malignant gliomas
`that rely on paracrine or autocrine stimulation of
`receptors that trigger the PI3/Akt pathway, studies
`directed at assessing the anti-tumor activity of CCI-779
`in patients with glioma are planned. However, since
`CCI-779 is principally metabolized by cytochrome
`P450 mixed function oxidases that are induced by
`many types of anticonvulsant agents commonly co-
`administered to patients with malignant gliomas, the
`toxicities, pharmacokinetics, and optimal dose-schedule
`of CCI-779 are being evaluated in patients concurrently
`receiving treatment with P450 mixed function oxidase-
`inducing anticonvulsant agents. In addition, phase I
`studies evaluating the feasibility of administering CCI-
`779 in combination with various cytotoxic chemother-
`apeutics such as 5-fluorouracil and gemcitabine are
`underway. An oral
`formulation of CCI-779, which
`would increase the feasibility of protracted drug
`administration, is also in early clinical development.
`For CCI-779 and other rationally developed agents,
`in which tumor growth inhibition and delay are
`predominant therapeutic e€ects in preclinical studies,
`traditional phase II tumor screening paradigms that
`focus solely on tumor regression may be suboptimal
`since clinically significant inhibitory e€ects on tumor
`growth may not be detected (Gelmon et al., 1999).
`Although regressions of well established tumors have
`been noted in preclinical studies of CCI-779, tumor
`growth delay is the principal therapeutic e€ect of the
`agent. In addition to the traditional screening end-
`point,
`tumor
`regression, disease-specific
`screening
`studies or phase II clinical trials, at which time ‘go
`or no go’ decisions are made about the subsequent
`development of CCI-779 and other antirproliferative
`agents, must be designed to evaluate inhibition of
`tumor growth. Possible clinical endpoints that have
`been proposed for the evaluation of antiproliferative
`agents include time to progression, the proportion of
`patients without progressive disease as
`their best
`response, symptomatic benefit, and reduction in tumor
`markers. Potential surrogate endpoints include inhibi-
`tion at the target, particularly if this e€ect can be
`validated to reflect
`tumor growth delay and/or
`
`Oncogene
`
`Roxane Labs., Inc.
`Exhibit 1028
`Page 004
`
`

`
`Targeting rapamycin-sensitive pathway
`M Hidalgo and EK Rowinsky
`
`regression in preclinical models, and assessments of cell
`proliferation by functional
`imaging (e.g. positron
`emission tomography, nuclear magnetic
`resonance
`scanning). However, randomized clinical trials are the
`optimal means to evaluate any agent’s e€ect on tumor
`growth and/or
`related surrogate endpoints Never-
`theless,
`tumor
`regression has been observed with
`CCI-779 and other signal transduction inhibitors in
`both preclinical studies and phase I clinical
`trials,
`which may occur when the inhibition of tumor growth
`is of su(cid:129)cient magnitude so that tumor cell death,
`possibly through apoptosis, exceeds the rate of tumor
`cell proliferation (Adjei et al., 2000; Ferry et al., 2000).
`Therefore, the traditional endpoint of tumor regres-
`sion, which is used to screen for the activity of
`nonspecific cytotoxic agents, may be an appropriate
`endpoint
`in nonrandomized phase
`II
`studies of
`antiproliferative agents that have the potential
`for
`inducing tumor regression as discerned from preclinical
`studies.
`Another important issue pertaining to the develop-
`ment of CCI-779 and other
`rationally designed
`inhibitors of signal transduction is the need to assess
`relevant target e€ects in order to guide dose selection
`and facilitate the assessment of benefit in clinical trials.
`The selection of an appropriate dose for broad disease-
`directed trials of CCI-779 is especially challenging since
`objective anti-tumor activity has been observed in
`patients treated with a wide range of doses. Unlike the
`situation with nonspecific cytotoxic agents,
`in which
`the relationship between dose and response is roughly
`linear and the maximum tolerated dose is sought,
`preclinical data suggest
`that
`therapeutic activity
`generally plateaus above a ‘threshold’ dose.
`In
`contrast,
`since the relationship between dose and
`toxicity is still likely to be linear, the selection of an
`optimal biological dose for clinical trials is desirable.
`However, e€orts to precisely assess target e€ects during
`clinical investigations are extremely complex, requiring
`an understanding of the agent’s precise mechanism of
`activity, and the availability of assays that measure
`relevant drug e€ects at the tumor target or in surrogate
`tissues that reflect inhibition at the target. Still, the
`availability of
`feasible biological assays capable of
`assessing inhibition at the target does not ensure that
`such e€ects will relate to clinical benefit. Since the
`intracellular target of CCI-779 and its downstream
`signaling pathways have been well characterized, the
`development of assays that measure the phosphoryla-
`tion status of either 4E-BP1/PHAS and/or p70s6k,
`enabling evaluation of whether patients are receiving
`pharmacodynamically-active doses, is feasible (Yatsco€
`and Aspeslet, 1998). Nevertheless, a critical
`issue is
`whether these downstream e€ects correlate with the
`anti-tumor activity of CCI-779, particularly since
`malignant
`cells
`can traverse
`the
`cell
`cycle and
`proliferate despite the dephosphorylation of 4E-BP1/
`PHAS and the inactivation of p70s6k by rapamycin
`(Kawamata et al., 1998; Yatsco€ and Aspeslet, 1998).
`These observations suggest that either these pathways
`are not
`the only mechanisms by which cell cycle
`progression is
`regulated, or
`that mTOR-4E-BP1/
`PHAS-1 and mTOR-p70s6k pathways are redundant.
`Although assessing drug e€ects with such assays may
`facilitate the determination of whether a delivered dose
`is pharmacologically-active, the assay may not neces-
`
`sarily predict drug e(cid:129)cacy since the assays may be
`assessing targets that are not related to the e€ects of
`the agent on proliferation. Alternatively, downstream
`factors may render the tumor cells resistant. Therefore,
`optimal assessments of target inhibition in tumor or
`surrogate tissues may require additional
`information
`about the relevance of target inhibition in terms of cell
`proliferation,
`cell
`cycle arrest, apoptosis, and/or
`angiogenesis. Such assays will ultimately be validated
`following the acquisition of late-stage clinical outcome
`data, particularly information pertaining to tumor
`regression, time to progression, clinical benefit, and
`survival, in large numbers of patients. At this juncture,
`the utility and feasibility of assays measuring the
`inactivation of p70s6k in peripheral blood mononuclear
`cells are being assessed in the course of early disease-
`directed e(cid:129)cacy studies of rapamycin.
`Another
`important
`issue
`regarding the optimal
`development of CCI-779 is whether
`tumors with
`specific molecular abnormalities are particularly sensi-
`tive or resistant to the antiproliferative e€ects of CCI-
`779. Tumors
`that
`rely on paracrine or autocrine
`stimulation of receptors that constitutively stimulate
`the PI3/Akt/mTOR pathway or tumors with mutations
`that activate the PI3K/Akt signal transduction path-
`way may depend on rapamycin-sensitive pathways for
`growth and thus may be
`especially sensitive
`to
`rapamycin analogs. For example, mutations of
`the
`PTEN tumor suppresor gene which encode a lipid
`phosphatase that inhibits PI3K-dependent activation of
`PKB/Akt, occur commonly in a wide variety of tumor
`types (Hosoi et al., 1998; Besson et al., 1999; Teng et
`al., 1997). The PI3K/Akt pathway is constitutively
`activated in tumors with PTEN mutations and these
`neoplasms are generally resistant
`to apoptosis. As
`predicted,
`the preliminary results of experimental
`studies in our laboratory indicate that breast cancer
`cell
`lines with PTEN mutations are extraordinarily
`sensitive to rapamycin (Hidalgo M, personal commu-
`nication).
`In addition to increased sensitivity to
`rapamycin conferred by mutations of
`the PTEN
`suppressor oncogene, abnormalities of regulators of
`the G1 checkpoint such as pRB, p16, p27, and cyclin
`D1 may also increase the sensitivity of
`tumors to
`rapamycin and may predict for drug e(cid:129)cacy (Sausville
`et al., 1999). It is clear that defining the molecular
`characteristics of tumors of patients enrolled in clinical
`trials of CCI-779 may help to identify which patients
`may benefit from treatment.
`
`Conclusion
`
`The natural product rapamycin, and its water-soluble
`ester analog, CCI-779, have potent immunosuppressant
`and antiproliferative properties stemming from their
`ability to modulate
`signal
`transduction pathways
`linking growth stimuli
`to the synthesis of specific
`proteins required for cell cycle progression from G1 to
`S phase. Both rapamycin and CCI-779 profoundly
`inhibit the proliferation of a broad range of human
`tumors both in vitro and in vivo, which have supported
`the clinical development of this class of compounds. At
`this juncture, CCI-779, which possesses more favorable
`pharmaceutical properties than rapamcyin, is complet-
`ing several phase I evaluations and is broad disease-
`
`6684
`
`Oncogene
`
`Roxane Labs., Inc.
`Exhibit 1028
`Page 005
`
`

`
`directed e(cid:129)cacy studies are planned. CCI-779 is well
`tolerated at doses that have resulted in impressive anti-
`tumor activity against
`several
`types of
`refractory
`malignancies However, developmental challenges in-
`clude the definition of recommended doses range for
`
`clinical use, the optimal disease-directed clinical trial
`paradigm enabling the appreciation of optimal biolo-
`gical activity and maximal therapeutic indices, as well
`as
`the ability to predict which tumors will be
`particularly sensitive to CCI-779.
`
`Targeting rapamycin-sensitive pathway
`M Hidalgo and EK Rowinsky
`
`6685
`
`References
`
`Adjei AA, Erlichman C, Davis JN, Cutler DL, Sloan JA,
`Marks RS, Hanson LJ, Svingen PA, Atherton P, Bishop
`WR, Kirschmeier P and Kaufmann SH. (2000). Cancer
`Res., 60, 1871 – 1877.
`Baker H, Sidorowicz A, Sehgal SN and Vezina C. (1978). J.
`Antibiot. (Tokyo), 31, 539 – 545.
`Besson A, Robbins SM and Yong VW. (1999). Eur. J.
`Biochem., 263, 605 – 611.
`Brown EJ, Albers MW, Shin TB, Ichikawa K, Keith CT,
`Lane WS and Schreiber SL. (1994). Nature, 369, 756 – 758.
`Brunn GJ, Hudson CC, Sekulic A, Williams