Prim l/t/(‘t um/ i’l‘tli Ill‘l’ of (ic/i/lmuiitiir’)‘ (hauling): edited by
`Derek Raghnvan. Howard l. Scher, Steven A. lxibcl. and Paul H
`Langc Lippincou-Ravcn Publishers. Philadelphia.
`t‘ N97
`
`{LiiAPTER 8
`
`iii'inciples of Chemotherapy for Genitourinary
`{lancer
`
`iti’ipllC'dthDS for Development of New Anticancer Drugs
`
`interactions.
`
`population of cells that retain the ability to grow, under stringent
`physiologic control mechanisms, and a self-renewing popula—
`tion for tissues that turn over rapidly, such as bone marrow and
`gastrointestinal epithelium.
`In this situation. balance is main—
`tained between natural attrition and replacement.
`The static. or terminally differentiated. population usually
`includes cells that do not undergo cell division after fetal life.
`such as skeletal muscle and neuronal tissue. The cells of an
`expanding population do not normally undergo continuous
`growth and division. but they may respond to stress. such as
`injury, with a period of replacement growth. For example, hepa-
`tocytes can respond to surgical resection of liver tissue by reenv
`tering the cell cycle and replacing the lost tissue. Another exam-
`ple of the expanding population is the stem cells in the bone
`marrow: these cells normally rest in the Gt. phase, but they can
`reenter the cell cycle. That they are predominantly in G” may
`protect them, in part, from the effects of cytotoxic agents.
`By contrast, the seltlrcncwing cell populations, such
`cells
`of the gastrointestinal tract. hair follicles, and bone marrow, are
`in a continuous proliferative state, with constant cell turnover.
`and are thus most commonly injured by cytotoxic chemother—
`apy: the static cell populations are the least vulnerable to the
`effects of chemotherapy.
`Malignant growth is essentially uncontrolled. occurring as a
`result of a breakdown in the mechanisms that turn off gi‘thh.
`The patterns that contribute to tumor growth may include a
`reduction in the length of the cell cycle. a decrease in the rate
`of cell death, or an increase in recruitment of cells into the
`active cell cycle. In general, malignant growth appears to follow
`a Gompertzian pattern.2 in which a period oi‘exponential growth
`is followed by a slowing of the growth rate. This process may
`occur through the tumor‘s outgrowing its vascular supply. as a
`result of the development of toxic breakdown products associ-
`ated with cellular turnover, or through other subtle cell—cell
`
`iiitt‘ick Creaven and Derek Raghavan
`
`
`is. totoxic chemotherapy has been in use for the management of
`six auced cancer for more than a century,| arising from concepts
`developed by Lissauer, Ehrlich, anti many others. The initial
`.iit‘rmpts at such treatment were characterized by a lack of speci—
`ity. with a fine balance between the toxicity to the tumor and
`mu experienced by the host. As reviewed in detail elsewhere,’
`during the past century. our application of chemotherapy to the
`s:
`‘ttment ot‘cancer has been refined, predicated on an improved
`understanding of the biochemical basis of its action and a clearer
`insight into the cellular and molecular mechanisms underlying
`2 .cmal and malignant growth.
`
`t‘ilMOR CELL BIOLOGY IN RELATION TO
`{THEMOTHERAPY
`
`The anticancer agents are a varied collection of drugs that
`at through a range ol‘ mechanisms. predominantly focused on
`interference with cell
`reproduction,
`Investigators generally
`agree that the differences between the growth characteristics of
`normal and malignant tissues form the major basis of the effec—
`tive Lise of cytotoxic chemotherapy.2 Differences among cellu—
`tar transport characteristics. with diffcrcntial uptake and efflux
`of cytotoxic agents. may also contribute to the difference in
`response to some cytotoxic agents. Moreover, important differ-
`cnces apparently exist among intracellular metabolic functions.
`such as the expression otglutathione, an intracellular Scavenger.
`which interacts with some alkylating agents and the platinum
`:‘omplexes to inactivate them. More recent data suggest the
`possibility of subtle interactions between the expression of
`growth-controlling factors. such as the receptor for the epider‘
`tnal growth factor, and the impact of cytotoxic agents. with
`resulting synergistic or antagonistic effects.
`Normal tissues are composed predominantly of a static popu-
`iation of cells that rarely undergo cell division, an expanding
`
`NOVARTIS EXHIBIT 2052
`Par v. Novartis, IPR 2016-01479
`Page 1 of 22
`
`

`

`9
`
`/
`
`(.‘iur'rrik 8
`
`Cellular Kinetics and Cell Cycle Control
`
`The kinetics of tumor cell growth. both in vilro and in vivo.
`has been the subject of considerable study.2 although our con—
`cepts on this topic remain fluid. Surprisingly little information
`is available regarding the kinetics of human tumor growth,3
`although a greater body of information is available on the
`growth of animal
`tumors.Z Investigators generally agree that
`tumor cells grow through an orderly sequence of steps:
`
`. The initial growth phase ((1,) is characterized by synthesis
`of ribonucleic acid (RNA) and protein. as well as dcoxyri—
`bonucleic acid (DNA) repair: this is a period of variable
`length. and its duration determines the length of the total
`cell cycle of the individual cell.
`. This phase is followed by the synthetic (S) phase, in which
`new DNA is synthesized.
`. The cells progress through the G3 phase, in which the total
`DNA content is double. that of the normal cell.
`. The mitotic (M) phase sees the division of the chromo—
`somes and separation into two offspring cells.
`. After mitosis. the cells may spend a variable period in a
`resting state known as G”: the cells are out of active cycle
`and appear not to be affected by chemotherapy to any major
`extent.
`
`and excretion. The mathematic description of these ratc pro~
`
`A detailed description of the molecular biology of cell cycle
`control is beyond the scope of this chapter. and the principles
`are reviewed elsewhere.41 In brief. several candidate genes and
`growth factors appear to regulate the various steps of the cell
`cycle. For example. in different tissues. entry into G] appears
`to be regulated by a range of factors. including MYC and F05
`(formerly known as c—myt‘ and crfnx). platelet~derived growth
`factors. and insulin-like growth factorvl and its receptor.
`A major cell cycle controller. p34“"‘3, regulates entry into
`S phase: this protein appears to have been conserved during
`evolution from yeasts to humans. with regard to both structure
`and function.5 In turn. this kinase appears to interact with its
`targets by association with the cyclins. a family of proteins
`found in different stages of the cell cycle. The GI cyclins are
`present before S phase. and they interact with p34 to cause the
`Cells to enter S phase. Entry into S phase is the first major cell
`cycle control point. These cyclins appear to interact directly
`with oncogenes and suppressor genes. probably through a phos»
`phorylation function of the cyclinvp34 complex.
`Several oncogenes. such as [MS and SRC (formerly known
`as ms and m"). have activity in the M phase of the cell cycle:
`in fact.
`they may actually interact
`in the process leading to
`activation ofthc M-phase promoting factor, which. in turn. con-
`trols entry into mitosis (the second major cell cycle control
`point).
`'lumor progression may be a function of a series of
`events involving progressive loss of control over entry into the
`S phase and loss of regulation of M phase: such progression
`is due to genetic instability and is central to the. evolution of
`malignancy. These changes may be contingent on the loss of
`checkpoint function. a mechanism by which the cell cycle
`pauses transiently and allows checking of the accuracy of repli~
`cation."
`The concept of the cell cycle is of great importance to our
`understanding of cytotoxic action. Most agents affect some as-
`pect ofthe synthesis of DNA. RNA. or protein and act at diffcr~
`
`cnt points within the cell cycle. This finding may be important
`when adding agents in a combination regimen; for example.
`the use of a spindle poison (such as a vinca alkaloid) may hold
`tip cells from entry into the (51 phase and thus may reduce the
`impact of an agent that acts predominantly at that point of the
`cell cycle. This effect is limited by the finding that many agents
`act at multiple points in the cell cycle. Inhibition of checkpoint
`function may explain the way in which tumor cells can be more
`vulnerable to the effects of cytotoxic agents than are normal
`cells: thus. the vulnerability of cancer cells to agents that target
`the S phase or the M phase may occur because the malignant
`cells proceed unchecked through the cell cycle despite a series
`of errors. whereas the normal cells stop at finite checkpoints
`while needed repairs occur.
`Cell cycle characteristics can be measured in several ways,
`including the use of labeling of mitoses3 and flow cytometry.
`When considering the biology of tumor growth as assessed by
`flow cytometry, the proportion of cells in the G, and S phases
`is thought to be most important. although the level of ancuploidy
`(proportion of cells that do not have a normal or diploid DNA
`content) appears to be an important prognostic determinant in
`some tumors. Another. more direct parameter of the cell cycle
`is measurement of tumor doubling time.3
`Also of importance is the growth fraction (proportion ofcells
`within a tumor that are in active proliferative phase). which can
`range from 25% to over 90% in human tumors. The rate of cell
`loss is also important; in most tumors, it
`is high. ranging from
`70% to over 9 Ct? In general. the length of the GI phase is
`one of the primary determinants of proliferative behavior; thus,
`if G,
`is short. the duration of the cell cycle is usually rapid.
`whereas cells with a long G, or those that spend considerable
`time in (it, have a much longer cell cycle and are less sensitive
`to the impact of chemotherapy.
`
`Clonality of Tumor Cell Populations
`
`In animal tumors. which tend to be clonal. first—order kinetics
`appears to apply in response to chemotherapy: that is, a dose
`of single—agent chemotherapy kills a fixed proportion of tumor
`cells. For example,
`if a tumor mass containing ll)7 cells is
`treated with an agent that kills 90% of the cells. 9 X It)" cells
`will be killed by a single dose. leaving behind 10“ viable cells
`(or 10% of the original tumor mass). A second dose kills 9 X
`It)5 cells and leaves 105 cells still alive. If treatment
`is not
`repeated.
`these cells will regrow. and the mass will rapidly
`return to its former size. This process is also influenced by the
`proportion ofcells that undergo spontaneous cell death. as well
`as the pl‘OlifCt'dthC rate of any remaining viable cells.
`Clinically.
`the situation is much more complex. because
`many human tumors appear not
`to be purely uniclonal. but
`rather are composed of multiple subpopulations of cells with
`different characteristics.7 Whether this phenomenon is due to
`the evolution from single clonal populations (stem cells) or is
`caused by initial evolution of multiple clones in response to an
`initial carcinogenic stimulus is not clear.
`
`PHARMACOLOGY ()F ANTICANCER AGENTS
`
`The time course of the sojourn of drugs in the body is deter-
`mined by the rates ofdrug absorption. distribution. metabolism,
`
`NOVARTIS EXHIBIT 2052
`Par v. Novartis, IPR 2016-01479
`Page 2 of 22
`
`

`

`l’Rthdl’li-ZSUI"(itll‘iklti‘l'llltRAl’YFOR (ii-.Nl‘l'()l'l€l.\.\l{\’(l‘tNLl-ZR
`
`91
`
`the data can
`uses is referred to as pharmacokinetics. Often.
`be fitted to mathematic models that are simplified descriptions
`of the complex physiologic realities. Many of these processes
`are so—called first order: that
`is. the rate at which the process
`occurs is proportional to the drug concentration. although some
`processes that depend on enzymes or carriers follow Michaelis‘
`‘slenten kinetics.
`in which the process is first order at low coir
`ccntrations and zero order (i.e.. occurring at a fixed rate) at high
`concentrations of the drug.
`
`ibsorption
`
`shows two components of the plasma decay: an initial rapid
`component and a subsequent slower component. both of which
`have the characteristics of log—linear. that
`is. firsteorder. pro—
`cesses. Mathematically. such a plasma decay can be fitted to a
`twoecompartmcnt model in which the body is conceptualized
`as consisting of two compartments. a central compartment into
`which the drug is introduced and a peripheral or tissue compart—
`ment into which it diffuses. ultimately to equilibrium. The sec-
`ond component of the plasma decay consists of the elimination
`processes of metabolism or excretion. The rate of this second
`process gives the half-life of the drug in the body and is an
`important pharmacokinetic characteristic of all drugs (tl/g/j’). For
`some drugs such as the anthracyclincs. a third component of
`the plasma decay is seen indicating a so—called deep tissue com—
`partment usually corresponding to the binding of the drug to
`some tissue component such as nucleic acid from which the
`drug is slowly released. An even simpler model in which the
`body is regarded as a single compartment can sometimes be
`used. but for many drugs it can lead to major errors in computing
`the important pharmacokinetic parameters. These "compart-
`ments” are mathematic constructs that usually have little or
`no correspondence with actual physiologic compartments. The
`total area under the plasma concentration >< time curve (AUC
`or C X t) is an important measure of the total exposure of the
`tissues to the drug. Other important pharmacokinetic parameters
`that can be calculated from the indices of plasma decay are the
`total body clearance and the apparent volume of distribution.
`the theoretic volume required to dissolve the total body content
`of the drug if it were uniformly distributed in the concentration
`found in plasma.
`in general. the drug is distributed in intravascular. extracellue
`tar. and intracellular water. but
`it must cross a membrane to
`pass from one of these locations to another. in addition. certain
`sites are protected from easy drug access. Such sanctuaries are
`usually characterized by lower drug concentration than other
`tissues.
`
`li'U) to dihydro—S-fluorouracil and converting cytosine arabino-
`
`Cytotoxic agents may be administered directly into the circu-
`!.trion (intravenous or intraarterial administration) or by the ex—
`uavascular approach, which includes oral.
`intramuscular.
`in~
`irathecal. intravesical. and intraperitoneal routes. The route of
`cxtravascular delivery influences absorption. Factors that deter—
`mine the uptake characteristics of a drug include the structure
`and size of the molecule and its negative log of dissociation
`constant (pKa) and. thus. its solubility characteristics.
`The clinical activity of specific agents may vary with the
`nature of the route and schedule of administration and conse—
`quent absorption. For example. cyclophosphamide can be ad-
`ministered orally in a dose of 100 mg/mZ/d for 14 days to pa
`tients with advanced prostate cancer and is well
`tolerated,
`causing only modest myelosuppression and gastrointestinal tox—
`icity.K When the drug is administered to similar populations of
`patients by intravenous bolus injection leg. 750 to 1000 mg/
`it]: every 3 weeks). the side effects may be more substantial."
`with no apparent improvement in therapeutic outcome
`Successful
`intravesical chemotherapy is predicated on the
`desire for cytotoxic agents to be active locally nit/mitt systemic
`absorption, thus protecting the patient from systemic side ef~
`t'ects while maximizing the concentration at the tumor surface.
`Thus. thiotepa. a small. readily absorbed molecule. is potentially
`less useful in this context than larger molecules. such as doxoru-
`bicin or mitomycin C'” Furthermore. the level of systemic ab-
`sorption of thiotepa can be increased if the agent is administered
`soon after transrtrethral tumor resection in the presence of a
`residual denuded bladder epithelium.
`Ultimately. the key to therapeutic effectiveness of any cyto-
`toxic agent is a function of the product of its concentration and
`the time available at the tumor site (C X t). Most cytotoxic
`agents are administered by intravenous or intraarterial routes.
`and calculations of the actual plasma C X t equation are made
`accordingly.
`
`Distribution and Transport
`
`The amount of cytotoxic agent available at the tumor target
`and the length of time during which it is present determine its
`level of efficacy. Several factors are influential. including the
`lipid solubility of the drug.
`its binding to protein and other
`carriers. and the mechanisms available to allow entry into the
`tumor (such as passive diffusion or active transport). A major
`factor is the plasma level of the drug. major determinants of
`which are its distribution characteristics. After rapid intrave-
`nous injection. the plasma concentration of the drug initially
`falls rapidly.
`in time.
`the rate of decline decreases. A plot of
`the natural logarithm (in) against time (semilog plot) generally
`
`The presence of sanctuary sites may be of real importance;
`for example. the blood-brain barrier appears to protect the brain
`against the local uptake of cytotoxic agents. and thus the brain
`may be the site of first relapse in tumors otherwise responsive to
`chemotherapy. Similarly. the testis apparently may functionally
`constitute a sanctuary site against the effect of chemotherapy;
`up to a third of patients treated for metastatic testis cancer before
`surgical removal of the affected testis have residual cancer
`within the testis at subsequent orchiectomy. despite an extrates—
`ticular complete response.
`
`Metabolism
`
`Two important types of metabolisrn'of antitumor agents are
`known. Antitumor agents that resemble normal metabolites are
`often metabolized by the same mechanisms as the normal me-
`tabolites. Most purine and pyrimidine antimetabolites require
`activation to a nucleotide. usually the triphosphatc. to be active.
`and these reactions are carried out by the mechanisms in the
`cell used to metabolize the corresponding normal preformed
`purines and pyrimidines (the so—called salvage pathways). Some
`of the antifolates undergo polyglutatnation by the mechanisms
`used for folates. Degradative pathways are also active in the
`cell. such as those responsible for reducing S-tluorouracil (5—
`
`NOVARTIS EXHIBIT 2052
`Par v. Novartis, IPR 2016-01479
`Page 3 of 22
`
`

`

`MECHANISMS ()li DRUG RES "I‘ANCE
`
`Several mechanisms of resistance to cytotoxic chemotherapy
`are known (Table is l).
`in general. these mechanisms can be
`classified on the basis of cellular distribution. intracellular fac
`tors include those that act at the cell surface. others within the
`cytoplasm. and those functioning at the level of the nucleus. ln
`addition are extracellular factors. such as those that affect the
`
`distribution and metabolism of the drugs. includng competitors
`for cellular transport mechanisms.
`Some cytotoxic agents can be exported from tumor cells
`through a mechanism based on the cellular surface. the so—called
`mtiltidrug efflux pump. which ischaracteriicd by the expression
`of a specific l70~kd protein complex. P glycoprotcin.33 investi—
`gators initially demonstrated that response to the cellular effects
`of agents as diverse as the vinca alkaloids. actinomycin D. and
`doxorubicin is reduced in normal and malignant cells that ex-
`press a protein complex on the cell surface. coded by a series
`of multidrug resistance genes (PGY. formerly known as mdr).
`This occurs as a result of reduced intracellular concentrations
`
`of the agents because of increased cellular efflux.23 Expression
`of the PGY phenotype appears to be an example of induced
`resistance. because previous exposure to colcliicine or to one
`of these agents can induce PG Y»based resistance to the entire
`group.
`Expression of PGY phenotype. has been identified in renal
`carcinoma. although its significance has been difficult to define
`because most rcnal carcinomas are resistant to the available
`
`Excretion of cytotoxic agents occurs predominantly in the
`kidneys and liver. and abnormalities in the function of either
`or both organs may influence the pattern of toxicity.H Renal
`dysfunction particularly affects the disposition of the platinum
`complexes. methotrexate. and bleomycin.
`
`Factors Modifying Pharmacokinetics
`
`Absorption of drugs may be affected by diseases of the gas—
`trointestinal
`tract. by previous surgical procedures. by corn
`pounds that change the pH of the gut. by coadministration of
`other drugs. and by other factors.”"“ Distribution can be influ-
`enced by disease states. such as cardiac failure. ascites. pleural
`effusion. and edema. and by the coadministi‘ation ofsubstances
`that can displace compounds from binding to serum albumin.H
`Age and amount of adipose tissue may have an impact on the
`clearance and toxic effects of cytotoxic agents. For example.
`obesity appears to reduce the clearance of doxorubicin in
`adults.” although the impact on toxicity of the drug is unclear.
`Age may alter disposition of doxorobucin: for example. Robert
`and Hoerni" demonstrated reduced clearance in older patients.
`compared with younger cohorts (Chap. 29). Numerous factors
`affect the rate at which drugs are metabolized by cytochrome
`P450. ' 7 Many compounds can induce cytochrome P45tl. include
`ing phenobarbitol and hydantoin. chlorinated hydrocarbon in—
`secticides, food additives. and carcinogens present in tobacco
`I smoke. Some antincoplastic agents may inhibit drug metabo~
`lism.Ix as may the presence of hepatic discach Excretion of
`compounds by the kidneys depends heavily on renal function.
`Coincidental administration of compounds that can compete
`for tubular reabsorption may affect renal clearance of certain
`compounds?“ as may urinary pH if the compound is able to
`become ionized.“
`
`(ill.\l’lltl\’ 8
`
`side to the cori‘csg'tonding uracil arabinosidc by dcaniination.
`These reactions occur in the cells ofthe tumor and in the cells of
`normal tissues. In addition to these specific metabolic reactions.
`compounds that do not show resemblance to physiologic sub-
`strates are inetaboli/ed primarily in the liver by the pathways
`used for detoxification ofxenobiotics. Most important is oxida—
`tion, often followed by conjugation. Oxidation is carried out
`by cytochrome F450. a family of enzymes located primarily in
`the microsomal or smooth endoplasmic reticulum fraction of
`the liver. This pathway is nonspecific in terms of structural
`requirements and oxidizes most lipid—soluble compounds. This
`pathway is responsible for the initial oxidation of the oxazapho
`sphorine ring of the oxazaphosphorines cyclophosphamide and
`ifosfamide. a reaction leading to the conversion of these com—
`pounds to their active metabolites.
`Knowledge of the metabolism of cytotoxic agents is impor—
`tant in designing treatment strategies: for example. intravesical
`delivery of cyclophosphamide would make no sense. because
`the drug requires hepatic metabolism to its active form to be
`effective (see later). In the patient with hepatic dysfunction or
`hepatic failure. impaired hepatic conjugation alters the metabo
`lism of doxorubicin and of the vinca alkaloids. whereas the
`microsomal activation of cyclophosphamide may be impaired
`in this clinical setting.
`
`Excretion
`
`cancer to the effects ot'cisplatin. GSH is found in most mamma-
`
`cytotoxic agents. and the presence or absence of this phenotype
`does not correlate with outcome. The study of the multidrug
`phenotype in bladder cancer cells has also proved difficult: the
`expression of P glyoprotein in bladder cancer has been variable
`and inconstantfl‘zs Expression ofP glycoprotein may be upreg—
`ulated in resistant populations of bladder cancer cells after treat-
`ment with the rhethotrexate. vinblastine. Adriamycin (doxoru—
`bicin). cisplatin (MVAC)
`l‘t:;llti’l€ll.2h
`in other tumor types.
`multidrug resistance cart occur in the absence of expression
`of the l7tl-kd P glycoprotein. whereas other proteins may be
`associated with similar patterns of resistance.27 a finding that
`perhaps explains this phenomenon in the absence of expression
`of P glycopmtcin.
`Ultimately. apart from its predictive function. this work is
`unlikely to be of great importance unless the multidrug resis—
`tance phenotype can be overcome at a functional
`level. For
`example. the calcium channel blockers. such as vcrapamil. have
`been shown to reverse multidrug resistance.28 although the toxic
`side effects of this approach have precluded routine use. Al-
`though clinical trials have not yet been published in bladder
`cancer. work initiated in our laboratories suggests that vera~
`painil can overcome the impact of the PG)’ phenotype. at least
`in bladder cancer cell lines in vitro.”
`The mechanisms of resistance to the platinum coordination
`complexes have been studied in detail. particularly in relation
`to ovarian cancer and malignant melanoma (Chap. 9). Althouin
`several mechanisms have been identified. including factors that
`influence cellular accumulation.
`signal
`transduction.
`ionic
`fluxes. and intracellular enzyme function,m the function of the
`intracellular scavenger. glutathione (GSH). has been the focus
`of particular attention in the context of the resistance of bladder
`
`NOVARTIS EXHIBIT 2052
`Par v. Novartis, IPR 2016-01479
`Page 4 of 22
`
`

`

`l’mxrnrrtcs ()l‘ (Hit-HUI lilaKAl’Y run Lirxt'tiu'rttxxm (I.\.\'r.t<.i<
`
`93
`
`Table 8-1. Targets of modulation of resistance
`Effector
`
`Hyperthermia
`Dipyridamole
`Tamoxifen
`Epidermal growth factor
`Cyclosporin A
`
`Agent
`
`Cisplatin (CDDP)
`Cisplatin
`Cisplatin
`Cisplatin
`Cisplatin
`
`Cisplatin. alkylating agents
`Cisplatin
`
`Cisplatin
`Cisplatin
`Cisplatin
`Methotrexate
`Vinca alkaloids
`5AF|uorouracil
`S-Fluorouracil
`Methotrexate
`
`Another model of resistance. proposed after a reanalysis of
`
`to cells and has many functions. including regulation of pro-
`n and DNA synthesis and detoxification. it appears to react
`ith cisplatin to reduce the drug's intracellular availability. in—
`»ttitors of GSH synthesis. such
`buthionine sulfoximine. have
`ran shown to cause a decrease in intracellular levels ot'GSH.
`uh a concomitant increase in the cytotoxicity of some antican~
`‘r agents. such as the alkylating agents. cisplatin,3| and pacli-
`:xcl. ‘3 Although many of the experimental data regarding the
`unificance of glutathione in cisplatin resistance have been de-
`'v’c‘d from models of ovarian cancer. we have demonstrated
`‘iztt high levels of glutathionc are present in cell lines derived
`om bladder cancer. and this finding may correlate with cis-
`itttin resistance;U
`Our understanding of these mechanisms is relatively crude,
`«d other factors appear to influence the responsiveness of blad—
`Wcr cancer to cytotoxic chemotherapy, The expression of several
`:ncogene products seems to influence resistance to cytotoxic
`' tents, The exact nature ofthis interaction is not yet clear and is
`ourticularly difficult to define because several of these products
`. ode for specific aspects of cellular growth control. irrespective
`if exposure to cytotoxic agents. For example. the interaction
`:l‘ epidermal growth factor and its specific receptor (epidermal
`trowth factor receptor {EGFR]) is involved in the regulation
`tit growth of bladder cancer. In vitro treatment with epidermal
`growth factor can increase cellular sensitivity of epithelial tu—
`.nors to cisplatin. however, presumably by an effect on a signal
`transduction pathway}4 Investigators have also reported that
`specific monoclonal antibodies can block the function of the
`EGFR,35 and further, treatment with these anti—EGFR mono«
`glonal antibodies plus cisplatin can cause a synergistic antitu-
`mor effect.“ These data are particularly difficult to interpret in
`view of the previously documented impact of expression of
`EGFR on the natural history of bladder cancer. as well as the
`demonstration that ER82 (formerly known as vrbB-Z) gene am-
`plification and overexpression are adverse prognostic determi»
`nants in bladder cancer.
`
`Buthionine sulfoximine
`
`7 P
`
`olymerase inhibitors
`Thymidine trtphosphate inhibitors
`DFMO
`Hyperthermia
`Dihydrofotate reductase
`Tubulin
`Thymidylate synthase
`Low uridine kinase
`Low polyglutamylation
`
`Mechanism
`
`Accumulation
`
`Membrane activity
`
`Detoxification
`Glutathione
`Metallothioneins
`DNA repair
`
`Access to DNA
`
`Gene amplification
`Target alteration
`
`Poor activation
`
`found in human cancer and appear to have a broad range of
`postulated roles in cell growth control. including involvement
`in cellular repair and apoptosis. Apoptosis. or programmed cell
`death. is regarded as a form of physiologic cell death because
`it represents a genetically determined cellular sequence that is
`part of the normal tissue homeostatic mechanism. Investigators
`have shown that P53, which is normally present only tran-
`siently, can be induced to accumulate within the cell by expo—
`sure to cytotoxic agents. such as cisplatin and mitomycin C.717
`and conversely. P53—dependent apoptosis modulates the cyto-
`toxicity of radiotherapy, S-FU. and doxorubicin.38 In our labora-
`tory. however. we have demonstrated that cisplatin cytotoxicity
`appears not to be mediated by apoptosis in bladder cancer cell
`lines.” These issues may be of particular importance because
`investigators have already postulated that P53 expression may
`constitute an independent prognosticator of response to the
`MVAC regimen. Because P53 may be induced by cytotoxic
`exposure. the timing ot‘tissue sampling may possibly be critical
`in determining the expression ofthis potential prognostic factor.
`especially if intravesical or systemic chemotherapy has been
`used previously.
`
`Models for Overcoming Drug Resistance of Tumor
`Populations: Clinical Implications
`
`Several models have been proposed to explain the varying
`levels of resistance to the impact of chemotherapy seen in
`human tumors. For example. Goldie and Coldmarfl“ proposed
`that tumors have a spontaneous mutation rate and the larger
`the number oftumor cells. the greater the chance ofspontaneous
`mutation. As a consequence. these investigators proposed that
`the most effective mechanism for cancer killing would he to
`initiate chemotherapy early (with a small cellular burden) and to
`introduce multiple agents in an attempt to overcome the various
`mechanisms of resistance. To date, this hypothesis has not been
`validated in clinical trials, although most ofthe studies reported
`have been flawed and have not truly evaluated the principles
`of this hypothesis.
`
`Another complex relationship has been demonstrated among
`the expression of P53 (a suppressor gene product), growth regu-
`lation, and cytotoxic response in bladder cancer. Alterations of
`thc P5} gene are among the most frequent genetic abnormalities
`
`NOVARTIS EXHIBIT 2052
`Par v. Novartis, IPR 2016-01479
`Page 5 of 22
`
`

`

`94
`
`/
`
`(ill‘tl’l‘tilx‘ H
`
`the model presented by Goldie and Coldntan. leads to the CHIP
`clttsron that cytotoxic agents can be used most effectively in
`sequence. rather than as combination schedules.“| Thus. an ini-
`tial series of treatments with the less effective of two drugs
`would eliminate a proportion of the tumor cells present. leaving
`behind a resistant population that may respond to several
`courses of treatment with the more effective drug: in this way.
`investigators have postulated that the impact ofth'e less effective
`therapy can be maximized.
`
`who have advanced disease and for whom no standard treatment
`
`has a significant chance of controlling their disease. Although
`anticancer response is recorded. this is not a primary end point
`ofa phase l trial. and patients must understand that the chance of
`significant tumor shrinkage is small. Patients often participate in
`such trials in the hope that they may secure the small percentage
`chance of success. or occasionally they participate for reasons
`ofaltruism. “l Regrettably. it has become increasingly clear that
`patients often have a poor understanding of the true purpose of
`phase I trials and of the small likelihood of individual patient
`benefit.“ Phase I trials are explanatory and should be designed
`to minimize the expenditure of patient resources. usually with
`only 3 to 4 patients entered at each dose level.45 These numbers
`allow relatively accurate identification and description of toxic—
`ity at each dose level. without the necessity to treat unreasonably
`large numbers of patients at potentially subtherapeutic doses.
`Unless the toxicity from a phase I trial is prohibitive or truly
`unpredictable and dangerous. the new agent automatically pro-
`ceeds to phase ll testing. In most cases. art absence of antitumor
`activity in phase l testing does not preclude initial phase I]
`assessment. Phase I] trials are also explanatory and attempt to
`define whether a new approach has antitnmor activity. These
`studies usually involve treatment with a predetermined dosage
`of a novel agent in defined groups of