`Copyright © 2002 by Annual Reviews All rights reserved
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.comon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`THALIDOMIDE: Emerging Role in Cancer Medicine
`
`Paul Richardson, Teru Hideshima, and Kenneth Anderson
`Jerome Lipper lVIyeloma Center, Division ofHematologic Oncology, Department of
`Adult Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston,
`Massachusetts 02115; e-mail.‘ paul_richardson@cb’ci.harvard. edu
`
`Key Words
`
`immunomodulation, antiangiogenesis, myeloma
`
`I Abstract Thalidomide—removed from widespread clinical use by 1962 because
`of severe teratogenicity—has antiangiogenic and immunomodulatory effects, includ-
`ing the inhibition of tumor necrosis alpha factor. It has now returned to practice as
`an effective oral agent in the management of various disease states including ery-
`thema nodosum leprosum, for which it was approved by the U.S. Food and Drug
`Administration in 1998, and more recently certain malignancies, including multiple
`myeloma. Although thalidomide’s mechanism of action remains incompletely under—
`stood, considerable insight has been generated by extensive preclinical studies in mul-
`tiple myeloma. Moreover, clinical trials have confirmed benefit in relapsed disease, and
`the role of thalidomide in treating newly diagnosed patients is currently under study. Its
`use in other tumors is under evaluation, with promise in renal cell carcinoma, prostate
`cancer, glioma, and Kaposi’s sarcoma. Activity has also been demonstrated in chronic
`graft-versus-host disease and in symptom relief as part of palliative care.
`
`INTRODUCTION
`
`A tragic event in the history of drug development occurred with the ovcr—thc-
`counter marketing of thalidomide in Europe during the late 1950s for the treat-
`ment of pregnancy-associated morning sickness. As early as 1961. reports of
`teratogenicity and dysmyelia (stunted limb growth) associated with thalidomide
`use prompted its subsequent withdrawal (1,2). The return of thalidomide as a
`therapy in certain conditions stems from its broad array of pharmacoimmuno-
`logic effects (3). This rehabilitation was reflected by its approval in 1998 by the
`U.S. Food and Drug Administration for the short-term treatment of cutaneous
`manifestations of moderate to severe erythema nodosurn leprosum (ENL), along
`with its use as maintenance therapy to prevent and suppress the cutaneous man-
`ifestations of ENL recurrence (3). Thalidomide has since became a treatment of
`choice for ENL, and its wide spectrum of activity has fostered its application
`in a variety of disease states (Table 1) (3—5). Because of its teratogenic effects,
`thalidomide is now used under strict guidelines to prevent fetal exposure to the
`drug (4).
`
`0066-4219/02/0218-0629$14.00
`
`629
`
`|PR2018—00685
`
`Celgene Ex. 2050, Page 1
`
`IPR2018-00685
`Celgene Ex. 2050, Page 1
`
`
`
`630
`
`RICHARDSON I HIDESHIMA I ANDERSON
`
`TABLE 1
`
`Potential therapeutic uses of thalidomide currently under investigation
`
`Cancer and related conditions
`
`Infectious diseases
`
`Autoimmune diseases
`
`Solid tumors (e.g., brain, breast, renal cell carcinoma);
`hematologic malignancies (e.g., multiple myeloma)
`HIV/AIDS and related conditions;
`aphthous ulcerations; wasting syndrome;
`mycobaclerial infections (e.g., tuberculosis)
`
`Discoid and systemic lupus erythematosus;
`chronic graft-versus-host disease; inflammatory
`bowel disease; rheumatoid arthritis; multiple sclerosis
`
`Dermatologic diseases
`
`Behcet’s syndrome; prurigo nodularis; pyoderma gangrenosum
`
`Other disorders
`
`Sarcoidosis; diabetic retinopathy; macular degeneration
`
`In the field of medical oncology, the discovery of thalidomide’s antiangiogenic
`properties has coincided with the emerging importance of angiogenesis in tumor
`growth and progression. Thalidomide has been shown to inhibit angiogenesis in-
`duced by basic fibroblast growth factor (,6 -FGF) in a rabbit cornea micropocket as-
`say andby vascular endothelial growth factor (VEGF) in a murine model of corneal
`vascularization (6, 7). In human studies, the drug appears to undergo activation
`to metabolites with antiangiogenic activity (8). Because of these antiangiogenic
`properties, thalidomide is currently undergoing evaluation in the treatment of var-
`ious solid tumors, multiple myeloma, and other hematologic malignancies (9—13).
`Results in multiple myeloma are particularly promising, although thalidomide’s
`antiangiogenic effects are believed to be only part of its antimyeloma activity.
`Its other potential actions include modulation of adhesion molecules, inhibition
`of tumor necrosis alpha factor (TNF-oz), downregulation of lymphocyte surface
`molecules, lowering of CD4:CD8 peripheral lymphocyte ratios, and direct effects
`on myeloma cells themselves (10, 14—18).
`This chapter presents a comprehensive review of the pharmacology of thalido-
`mide, a description of preclinical studies in multiple myeloma to illustrate the
`drug’s complex putative mechanisms of action, and a description of clinical stud-
`ies in multiple myeloma. Studies in other hematologic malignancies are also ad-
`dressed, as is the status of research in solid tumors and in other cancer-related
`applications.
`
`PHARMACOLOGY
`
`Thalidomide is a derivative of glutamic acid and is pharmacologically classi-
`fied as an immunomodulatory agent (19). Structurally, thalidomide contains two
`amide rings and a single chiral center (Figure 1); its full chemical name is alpha-
`N{phthalimido}glutarimide [C13 04 N2 H9] and its gram molecular weight is
`258.2 (19).
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.comon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`|PR2018—00685
`
`Celgene Ex. 2050, Page 2
`
`IPR2018-00685
`Celgene Ex. 2050, Page 2
`
`
`
`|PR2018—00685
`
`Celgene Ex. 2050, Page 3
`
`THALIDOMIDE
`
`631
`
`Thalidomide
`
`N
`
`47
`
`O
`
`OL-(N-phthalimido) glutarimide
`
`O A
`
`nalogs (also known as IMiDs)
`
`@waw0
`(Ebb
`
`Figure 1
`Structures of thalidomide and its potent ana-
`logues, immlmomodulatory drugs (IMiDs).
`
`The currently available formulation is a nonpolar racemic mixture present as
`the optically active S and R isomers at physiologic pH, which can effectively
`cross cell membranes (19, 20). The S isomer has been linked to thalidomide’s
`tcratogcnic cffccts, whereas the R isomcr appears to be primarily responsible for
`its sedative properties (7, 20, 21). The isomers rapidly interconvert at physiologic
`pH in vivo, and thus efforts at formulating only the R isomer have failed to obviate
`the teratogenic potential of thalidomide (20, 22).
`
`Pharmacokinetics
`
`Pharmacokinetic analysis of thalidomide in humans has been limited by the ab-
`sence of a suitable intravenous formulation, owing to the drug’s instability and
`poor solubility in water. The pharmacokinetics of thalidomide have therefore been
`determined only from animal studies and in humans receiving the oral therapy.
`Single-dose thalidomide trials have been conducted in healthy volunteers, patients
`with HIV infection, and patients with hormone-refractory prostate cancer (23—27).
`As Table 2 shows, the pharmacokinetics appear highly variable. Moreover, in
`patients with HIV infection, dose adjustments based onboth ideal body weight and
`body surface did not affect this variability (23, 24). As a result of this variability,
`the pharmacokinetic properties of thalidomide in humans have not been well char-
`acterized, and this has confounded the definition of a dose-response effect against
`human cancer.
`
`
`
`Annu.Rev.Med.2002.53:629-657.DownloadedfromWWW.annualreviews.org
`
`
`
`
`
`
`
`bycynthia.montalvo@bakerbotts.comon01/16/18.Forpersonaluseonly.
`
`
`
`
`
`IPR2018-00685
`Celgene Ex. 2050, Page 3
`
`
`
`632
`
`RICHARDSON I HIDESHIMA I ANDERSON
`
`TABLE 2
`
`Single dose pharmacokinetic parameters ot‘thalidomide in humans (27)
`
`Mean apparent
`pharmacokinetic parameters
`
`Population
`
`Elderly patients with
`hormone-refractory
`prostate cancer
`Patients with HIV
`infection
`
`Healthy female volunteers
`
`Healthy male volunteers
`
`Dose
`
`200 mg
`
`800 mg
`300 mg
`
`200 mg
`
`200 mg
`
`tnlax
`
`3.3*
`
`4.4*
`3.4
`
`5.8
`
`4.4
`
`t1/2 (L)
`
`6.5
`
`18.3
`5.7
`
`4.1
`
`8.7
`
`7d (L)
`
`66.9
`
`165.8
`78.2
`
`53.0
`
`120.7
`
`Abbreviations: lmax, time to reach maximum concentrations; 11/2, elimination half-life; Vd, volume of distribution;
`HIV. human immunodeficiency virus.
`a,
`Median value.
`
`Absorption
`
`When oral thalidomide at a dose of 100 mg/kg has been administered in animal
`studies, maximum serum concentrations were reached within 4 h (28). Absorption
`was apparently independent of the administered doses and slower than drug elim-
`ination. Recent studies in humans show a similar pattern; thalidomide at 200 mg
`per dose reaches peak concentration (tmax) in a mean of ~4 h (23—25, 29).
`
`Distribution
`
`Animal studies have demonstrated a wide distribution of thalidomide throughout
`most tissues and organs (28). It is present in semen following oral administration
`in rabbits, but it is not known whether the drug is present in human semen (19, 30).
`Human pharmacokinetic studies to date also indicate that thalidomide has a large
`apparent volume of distribution (Vd) (24—26). Further, studies in elderly prostate
`cancer patients suggest variability in Vd, possibly due to alterations in absorption
`and plasma protein binding (24).
`
`Metabolism
`
`Thalidomide undergoes rapid and spontaneous nonenzymatic hydrolytic cleavage
`at physiologic pH to generate up to 50 metabolites, of which five are considered
`primary metabolites (8, 20, 22, 28, 31). Research efforts to better characterize the
`biologic properties of the specific metabolites have been complicated by their in-
`stability and rapid degradation under physiologic conditions (32). Whereas in vitro
`studies suggest thalidomide induces cytochrome P-450 isoenzymes in rats, recent
`evaluation of single- or multiple-dose pharmacokinetic parameters of oral thalido-
`mide at 200 mg daily in healthy human volunteers has indicated that thalidomide
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.cornon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`|PR2018-00685
`
`Celgene Ex. 2050, Page 4
`
`IPR2018-00685
`Celgene Ex. 2050, Page 4
`
`
`
`THALIDOMIDE
`
`633
`
`does not inhibit or induce its own metabolism over a 21-day period in humans,
`and thus very little metabolism of thalidomide is thought to occur via the hepatic
`cytochrome P-450 system (26, 33, 34).
`
`Excretion
`
`Thalidomide appears to be rapidly excreted in urine as its metabolites, with the
`nonabsorbcd portion of the drug excreted unchanged in feces. However, clearance
`is primarily nonrenal; mean terminal half-lives of the R and S isomers in healthy
`male human volunteers were measured at 4.6 and 4.8 h, respectively (23, 28, 29).
`In a study report of urinary excretion data for a single dose (200 mg daily), the
`elimination half-life was ~8 h, with minimal drug excretion over a 24-h period
`(23). Both single and multiple dosing of thalidomide in older prostate cancer
`patients revealed a significantly longer half-life at a higher dose (1200 mg daily)
`than at a lower dose (200 mg daily) (24). Conversely, no effect of increased age
`on elimination half-life was identified in the age range of 55—80 years (24). Thus,
`the effects of renal or hepatic dysfunction on the clearance of thalidomide remain
`unclear, and additional studies are needed to better characterize age-related or
`physiologic effects on drug clearance.
`
`Drug Interactions
`
`The only drug that has been systematically evaluated for interaction with thalido-
`mide is oral hormonal contraceptives, which showed no significant interaction. An-
`imal studies suggest that thalidomide enhances the sedative effects of barbituates
`and alcohol as well as the catatonic effects of chlorpromazine and reserpine. Cen-
`tral nervous system stimulants (including methamphetamine and methylphenidate)
`appear to counteract the depressant effects of thalidomide (35).
`
`Potential Antitumor Effects
`
`D’Amato et 31., while evaluating thalidomide ’s mechanism of teratogenicity, found
`that it exhibited antiangiogenic properties (6,7). They postulated that thalido-
`mide inhibited angiogenesis by interrupting processes induced by fi-FGF and/or
`VEGF (6, 7, 36). Further in Vitro studies suggested that the antiangiogenic effect
`of thalidomide was due to specific metabolites and not the parent compound (37).
`Another important property of thalidomide is that it selectively inhibits TNF-oz
`production while leaving the patient’s inunune system otherwise intact (38). This
`has led to its application in various disorders characterized by abnormal TNF-oc
`activity, including ENL, mycobacterium tuberculosis infection, graft-versus—host
`disease, rheumatoid arthritis, systemic lupus erythematosis, multiple sclerosis,
`Crohn’s disease, cancer— and HIV-related cachexia, diabetes mellitus, and endo-
`toxic shock.
`
`The exact mechanism of thalidomide-induced TNF-oz inhibition is unclear, but
`it does appear to differ from other TNF-oz inhibitors such as pentoxyfylline and
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.cornon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`|PR2018-00685
`
`Celgene Ex. 2050, Page 5
`
`IPR2018-00685
`Celgene Ex. 2050, Page 5
`
`
`
`634
`
`RICHARDSON I HIDESHIMA I ANDERSON
`
`dexamethasone (39, 40). One mechanism postulated is that thalidomide inhibits
`TNF-uc synthesis by accelerating degradation of TNF-uc mRNA, resulting in a
`significant but incomplete suppression of TNF-a protein production (39, 41). Of
`particular interest is the recent demonstration that thalidomide decreases the bind-
`ing activity of NF-ICB which in turn controls activation of the TNF-ot gene (42).
`It has also been postulated that thalidomide’s effect on angiogenesis may be
`through TNF-oz inhibition, since TNF-oz has proangiogenic effects (6). However,
`the absence of a demonstrable TNF-oz effect in experimental models of angiogen-
`esis, coupled with the inability of strong TNF-oc inhibitors to directly influence
`angiogenesis, suggests that thalidomide’s antiangiogenic activity is not related to
`TNF-oz inhibition alone (6, 7).
`
`Annu.Rev.Med.2002.53:629-657.DownloadedfromWWW.annua1reviews.orgbycynthia.montalvo@bakerbotts.cornon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`Immunomodulatory Effects
`
`Results of studies evaluating the effects of thalidomide on lymphocytes have been
`inconsistent (43—45). Emerging evidence suggests that thalidomide does not di-
`rectly suppress lymphocyte proliferation (14). However, differential effects on
`T cell stimulation, shifts in T cell responses, and inhibition of proliferation of
`already stimulated lymphocytes have been shown (17,44, 46—48). Modification
`of surface adhesion molecule on leukocytes, inhibition of neutrophil chemotaxis,
`and effects of cytokines other than TNF-oz have been demonstrated, including
`the inhibition of interleukin (IL)—12 production, enhanced synthesis of IL-2, and
`inhibition of IL-6 (14,16, 38, 49—51).
`
`Adverse Effects
`
`Common side effects reported during treatment with thalidomide are summarized
`in Table 3 (27). Sedation and constipation appear to be the most common adverse
`effects in cancer patients (9, 13, 36). The most serious adverse effect associated
`with thalidomide is peripheral neuropathy. There may also be an increased inci-
`dence of thromboembolic events, but this appears to be rare when the drug is used
`as a single agent. However, recent reports have shown thromboembolic complica-
`tions to be more frequent when the drug is combined with steroids and particularly
`
`TABLE 3 Clinical adverse events reported during thalidomide use (27)
`
`Neurologic
`
`Gastrointestinal
`
`Dermatologic
`
`Miscellaneous
`
`Sedation
`
`Dizziness
`
`Constipation
`
`Exfoliative/erythrodermic
`
`Xerostomia
`
`Nausea
`
`Cutaneous reactions
`
`Weight gain
`
`Mood changes
`
`Increased appetite
`
`Brittle fingernails
`
`Headaches
`
`Pruritis
`
`Edema of face/limbs
`
`Reduction in thyroid
`hormone secretion
`
`Hypotension
`
`|PR2018—00685
`
`Celgene Ex. 2050, Page 6
`
`IPR2018-00685
`Celgene Ex. 2050, Page 6
`
`
`
`THALIDOMIDE
`
`635
`
`with anthracycline-based chemotherapy, causing early cessation of one study ex-
`ploring a combination of thalidomide with doxil and dexamethasone (52, 53). Pos-
`sible cardiovascular effects of thalidomide include bradycardia and hypotension.
`The risk of adverse cardiovascular events appears greater in older patients with
`coronary disease taking multiple blood-pressure-lowering medications along with
`thalidomide (27).
`It is well-known that thalidomide must not be used during pregnancy, and rec-
`ommended contraceptive methods must be used by both men and women of child-
`bearing potential. The System for Thalidomide Education and Prescribing Safety
`(STEPS), implemented to ensure the safe distribution of thalidomide, requires a
`patient’s compliance with contraception guidelines and mandatory surveillance
`procedures. In addition, all healthcare providers who plan to prescribe and/or dis-
`pense thalidomide must be registered with the program (4).
`Thalidomide-induced peripheral neuropathy commonly presents with numb-
`ness of the toes and feet, muscle cramps, weakness, signs of pyramidal tract in-
`volvement, and carpal tunnel syndrome (5, 19). Clinical improvement typically
`occurs upon prompt discontinuation of the drug, but long-standing sensory loss
`has been documented (54—57). Drug-related neuropathy is characterized as asym-
`metric, painful, peripheral paresthesia with sensory loss (5, 54—57). The risk of
`developing peripheral neuropathy during thalidomide treatment increases when
`high cumulative doses are administered and especially in the elderly (19).
`In cancer patients, thalidomide should be used with caution when there is a prior
`history of neuropathy and when used in combination with other agents known to
`have neurotoxic effects (27). However, whether the incidence of thalidomide-
`induced peripheral neuropathy is increased in cancer patients with a history of
`vinca alkaloid use remains unknown. Safety data from current phase I and II
`clinical trials of thalidomide in the treatment of solid tumors, multiple myeloma,
`and hematologic malignancies suggest that peripheral neuropathy occurs in 10%—
`30% of patients (10, 12, 13, 58).
`
`PRECLINICAL STUDIES OF THALIDOMIDE
`
`AND ITS ANALOGS IN MULTIPLE MYELOMA
`
`Although thalidomide was initially used to treat multiple myeloma because of
`its antiangiogenic effects, the mechanism of its antimyeloma activity appears to
`be more complex (Figure 2). Preclinical studies of thalidomide and its potent
`analogs (also known as immunomodulatory drugs, IMiDs) suggest that these drugs
`act against myeloma in several ways. First, there appears to be a direct effect on
`the myeloma cell and/or bone marrow stromal cell, which inhibits tumor growth
`and survival. Second, adhesion of myeloma cells to bone marrow stromal cells
`(BMSCs) triggers secretion of cytokines, which augment myeloma cell growth
`and survival (59—61) and confer drug resistance (62); importantly, thalidomide
`modulates adhesive interactions (14) and thereby may alter tumor cell growth,
`
`Annu.Rev.Med.2002.53:629-657.DownloadedfromWWW.annualreviews.orgbycynthia.montalvo@bakerbotts.cornon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`|PR2018—00685
`
`Celgene Ex. 2050, Page 7
`
`IPR2018-00685
`Celgene Ex. 2050, Page 7
`
`
`
`636
`
`RICHARDSON I HIDESHIMA I ANDERSON
`
`C. Thalidomide
`
`MM cells A
` lL-G
`
`
`Th I'domid
`
`A. Thalidomide-u)-
`3'
`
`TNFa+
`
`BIL-1B
`
`"
`
`
`
`Bone
`marrow
`Stromal
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.cornon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`VEGF+
`
`
`
`
`
`WV)6 lymphocytes IFN1‘
`°
`D Thalidomide
`Us”
`1 '
`CDB+ T cells
`E. Thalidomide
`NK cells
`
`Figure 2 Possible effect of thalidomide on myeloma cells’ and bone marrow stro-
`mal cells’ (BMSCs’) microenvironment in vivo. (A) Thalidomide directly inhibits
`myeloma cell growth. (B) Thalidomide inhibits myeloma cell adhesion to BMSCs.
`(C) Thalidomide blocks IL-6, TNF-oc, and IL-1 13 secretion from BMSCs. (D) Thalido-
`midc blocks thc ability of VEGF and fi-FGF to stimulatc ncovascularization of bonc
`marrow. (E) Thalidomide induces lL-Z and INF-y secretion from '1' cells.
`
`survival, and drug resistance. Third, cytokines secreted into the bone marrow mi-
`croenvironment by myeloma cells and/or BMSCs, such as IL-6, IL-lfi, IL-10,
`and TNF-oz may augment myeloma cell growth and survival (61, 63), and thalido-
`mide may alter their secretion and bioactivity (64). Fourth, thalidomide decreases
`the secretion of VEGF, IL-6 (65), and ,BFGF by myeloma and/or BMSCs.
`Since lL-6 is known to promote myeloma cell growth and survival, and VEGF
`has been demonstrated to induce myeloma cell migration (66), thalidomide may
`directly block tumor cell growth and migration, as well as inhibiting bone mar-
`row angiogenesis (67, 68). Recent studies have demonstrated the direct antitumor
`activity of thalidomide and IMiDs on myeloma cells (18). IMiDs induce growth
`inhibition of myeloma cells in a dose-dependent fashion, with an IC50 as low as
`0.1—1 MM. IMiDs are effective against myeloma cell lines resistant to conven-
`tional chemotherapeutic agents such as doxorubicin, melphalan, or mitoxantrone,
`as well as dexamethasone. Moreover, thalidomide and the IMiDs enhanced the
`antitumor activity of dexamethasone (Figure 3a); conversely, these effects were
`partially inhibited by IL-6 (Figure 3b), a potent myeloma growth and antiapop-
`totic factor. IMiDs also showed significant activity against dmg—resistant patient
`myeloma cells (Figure 4). The observed clinical activity of thalidomide in patients
`
`|PR2018—00685
`
`Celgene Ex. 2050, Page 8
`
`IPR2018-00685
`Celgene Ex. 2050, Page 8
`
`
`
`THALIDOMIDE
`
`637
`
`
` fl[III/Illa
`
`"Ill/III/A
`
`L =
`|MiD1
`
`1 =
`|MiD2
`
`|MiD3
`
`0
`
`control
`
`Thal
`
`
`
`0.1
`
`|MiD1 (pM)
`
`0.1
`|MiD3 (uM)
`
`1
`
`1
`
`
`
`5
`
`é4
`63
`
`E 2
`O
`
`0
`
`5
`g4
`5 3
`
`E 2
`o 1
`o
`
`o
`
`o
`
`1
`
`1
`
`
`
`0.1
`
`Thal (uM))
`
`0.1
`|MiD2 (uM)
`
`
`
`0
`
`0
`
`‘24
`353
`
`E 2
`O
`
`1 0
`
`5
`g4
`35, 3
`
`E 2
`0 1
`o
`
`Figure 3 Effect of Dex and ll.-6 on response of mu] iple myeloma cells to thalido-
`mide and the lMiDs. (a) lVlM. l S cells were cultured with 1.0 MM thalidomide,
`lMiDl, lMiDZ, or IMiD3 in control media alone (D) or with 0.001 (I), 0.01 (E), and
`0.1 MM (I) Dex. (b) l\/[M. l S cells were cultured in control media alone and with 0.1
`and 1.0 MM thalidomide, IMiDl, IMiDZ, or IMiD3 either in the presence (D) or ab-
`sence (I) of lL-6 (50 ng/ml). In each case, 3H-TdR uptake was measured during the
`
`last 8 h of 48-h cultures. Values represent the mean (:: SD) 3H-TdR (cpm) of triplicate
`cultures.
`
`with myeloma that is refractory to conventional therapies (69), coupled with our
`in vivo studies, suggests that thalidomide can overcome resistance to conventional
`treatments. In addition, our laboratory work suggests that dexamethasone can add
`to the antiproliferative effect of thalidomide and its analogs in Vitro, suggesting
`potential utility of coupling these agents in novel therapeutics.
`
`|PR2018—00685
`
`Celgene Ex. 2050, Page 9
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.comon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`IPR2018-00685
`Celgene Ex. 2050, Page 9
`
`
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.comon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`638
`
`RICHARDSON I HIDESHIMA I ANDERSON
`
`I
`E?
`Ex
`9
`i :6
`n.
`E/
`
`E/
`=/
`e?
`E?
`4
`Thal
`
`|MiD1
`
`|MiD2
`
`h\\\\\\\\\\\\\\‘€€
`
`1
`0
`
`O-‘wa‘fim‘l
`
`CPM(x103)
`
`
` 6'? 4
`
`
`
`
`
`:
`:7
`2?
`5/
`
`
`
`i/
`:/
`e?
`e?
`1
`|MiD3
`
`Thal
`
`|MiD1
`
`|MiD2
`
`|MiD3
`
`Figure 4 Effect of thalidomide and its analogs on DNA synthesis of patient multi-
`ple myeloma cells cultured with control media (D), or with 0.1 MM (I), 1.0 MM (E),
`10 MM (I), and 100 MM (I) thalidomide, IMiDl, IMiDZ, or IMiD3. In each case,
`3H-TdR uptake was measured during the last 8 h of 48-h cultures. Values represent the
`
`mean (:: SD) 3H-TdR (cpm) of triplicate cultures.
`
`The mechanism of growth inhibition induced by thalidomide and its analogs is
`not totally understood. IMiDs, and to a lesser extent thalidomide, induce apopto-
`sis of myeloma MM. IS cells, as evidenced both by increased sub-G1 cells on P1
`staining and by increased annexin V—positive cells. In these cells, which character-
`istically have wild-type p53, exposure to these agents (and dexamethasone) down-
`regulates p21, thereby facilitating G1 -to-S transition and enhanced susceptibility
`to apoptosis. Moreover, p27 is also upregulated by treatment with both thalido-
`mide and IMiDs (T. Hideshima, et al., unpublished data). This profound apop-
`totic effect may correlate with the fact that complete responses to thalidomide are
`occasionally observed. The mechanism of t11alidomide/IlVIiDs-induced apoptosis
`
`|PR2018—00685
`
`Celgene EX. 2050, Page 10
`
`IPR2018-00685
`Celgene Ex. 2050, Page 10
`
`
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.cornon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`THALIDOMIDE
`
`639
`
`is also not entirely understood. However, our preliminary data indicate the par-
`ticipation of caspase-8 activation in thalidomide/IMiDs-induced apoptosis. IL-6
`overcomes the downregulation of p21 induced by these agents, consistent with the
`increase in DNA synthesis triggered by IL-6 even in the presence of these drugs. In
`contrast, in Hs Sultan cells and AS cells, derived from a multiple myeloma patient,
`(which are both wild-type and mutant p53), the IMiDs and thalidomide induce
`p21 and related G1 growth arrest (Figure 5), thereby conferring protection from
`apoptosis, as has been observed in other systems (70, 71). Prior studies showed that
`p21 was constitutively expressed in the majority of myeloma cells and inhibited
`proliferation in both a p53-dependent and -independent fashion (72).
`Previous reports that cells overexpressing p21 protein demonstrate chemore-
`sistance (73) further support the notion that a protective effect of G1 growth ar-
`rest is induced by thalidomide and IMiDs in Hs Sultan— and AS—patient—derived
`myeloma cells. Conversely, the frequent regrowth of progressive myeloma noted
`
`£13
`
`A
`
`E
`3:
`g
`
`a
`5
`
`“9
`=“
`+
`e
`
`S
`a
`8
`a
`x
`j;
`
`fi
`E
`S
`3
`E
`8
`(91)
`
`4*
`IBM mm W Wifi
`le—tubuiin
`'
`‘
`’
`
`
`MM.1Seell
`
`lazpz-I m.-
`lea—tubulin
`
`.. m m w
`
`
`HS Sultan cell
`
`iBipE'i
`
`m- a“ *
`
`'Bi‘l‘tubU'i”M
`MM patient cell
`
`B
`
`MM1S
`
`Hs Sultan
`
`MM patient
`
`wt
`[Pzp53
`”33953 m
`
`mt
`
`wt m
`mt
`wt
`We”
`
`Figure 5 Effect of thalidomide and its analogs on p21 expression in multiple mye-
`loma (MIVI) cell lines and patient cells. (a) MIVI. l S cells were cultured with 10 MM
`of thalidomide, IMiDl, IMiDZ, and IMiD3 for 48 h. MIVLIS cells were also cultured
`with IL-6 (50 ng/ml) alone and with IMiDl, 10 MM Dex, and Dex plus IL-6. Cells
`were lysed, subjected to SDS PAGE, transferred to PVDF membrane, and blotted
`with anti-p21 Ab. The membrane was stripped and reprobed with anti-a-tubulin Ah.
`(1)) 1\/[M. 1 S, Hs Sultan, and patient 1\/[M cells were lysed and immunoprecipitated with
`wt—p53 and mt—p53 Ab, transferred to PVDF membrane, and blotted with anti-p53 Ab.
`
`|PR2018—00685
`
`Celgene EX. 2050, Page 11
`
`IPR2018-00685
`Celgene Ex. 2050, Page 11
`
`
`
`640
`
`RICHARDSON I HIDESHIMA I ANDERSON
`
`A
`
`an
`
`non-stimulated
`
`stimulated
`
`a
`a
`8 ~ N
`e - w m 72
`E
`63> Q Q
`on g Q Q E
`*5
`E g 2
`E 2
`e z a
`8
`lephospho-MAPK W
`IBzERKZ
`
`E
`“‘
`5:” —E:XE
`EEEE See
`Sink-QB...
`
`|BzRAl=TK “bu“ m
`
`
`
`MM.1S ceil
`massif
`
`Figure 6 Effect of thalidomide and its analogs on growth and apoptotic signaling
`in myeloma cell lines, l\/[M. 1 S and l\/1M.1R cells. (a) lVlM. l S cells were cultured for
`48 h in media, with 50 MM of PD98059, and with 10 MM of lMiDl, lMiDZ, or
`lMiD3. Cells were then triggered with 50 ng/ml of lL-6 for 10 min, lysed, transferred
`to PVDF membrane, and blotted with anti-phospho MAPK Ab. Blots were stripped
`and reprobed with anti-ERKZ Ab. (b) MlVllS and l\/ll\/l.lR cells were treated with
`thalidomide (100 MM), lMiD l (100 MM), or Dex (10 MM) and harvested at l 2 h. Total
`cell lysates were subjected to immunoprecipitation with anti-RAFTK Ab and analyzed
`by immunoblotting with anti-P—Tyr Ab or anti-RAFTK Ab.
`
`clinically upon discontinuation of thalidomide treatment may correlate with re-
`lease of drug-related G1 growth arrest. We have previously shown that related
`adhesion focal tyrosine kinase (RAFTK, Figure 6) mediated dexamethasone-
`induced myeloma cell apoptosis. Thalidomide and lMiDs, like dexamethasone,
`induced tyrosine phosphorylation of RAF"1K, suggesting a key role of RAFTK in
`thalidomide- and HVliDs-induced apoptosis. Activation of RAFTK was also found
`in dexamethasone-resistant, IMiDs-sensitive MM. 1R myeloma cells.
`Recent studies have also demonstrated that myeloma cell lines and patient
`myeloma cells produce TNF-oz (63). TNF-oz in turn triggers E-6 secretion from
`BMSCs in a dose-dependent fashion;
`importantly, TNF-oz is a more potent
`stimulus of lL-6 secretion than TGF-fil and VEGF (Figure 7). Remarkably, TNF-oz
`also induces NF-ICB activation and expression of LFA-l, ICAM-l, VCAM-l,
`and MUC-l on myeloma cell lines, as well as ICAM-1 and VCAM—l on BMSCs.
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.cornon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`|PR2018—00685
`
`Celgene EX. 2050, Page 12
`
`IPR2018-00685
`Celgene Ex. 2050, Page 12
`
`
`
`100
`90
`
`A 80
`£370
`0.60
`B,
`u. 50
`z
`'_ 40
`30
`20
`10
`0
`
`C
`
`e
`
`5
`
`2 4
`U)
`E
`5 3
`“P
`=|
`
`N
`
`_\
`
`IM-9
`
`BMSC
`
`IM-9/BMSC
`
`THALlDOlVHDE
`
`641
`
`lM-9
`
`BMSC
`
`lM-9/BMSC
`
`3.5
`
`3 0
`E 2.5
`c»
`5 2.0
`o 1
`_'I
`5
`_
`1.0
`
`0.5
`
`0
`
`D3.0
`
`2.5
`
`c
`0')
`E 2-0
`5
`“P 1.5
`='
`_‘ O
`
`.001
`
`media
`
`VEGF
`
`TNFa
`
`TGFB1
`
`o
`
`01
`000010001001
`TNFu(ng/m|)
`
`1
`
`1o
`
`Figure 7 TNF-a secretion from IM-9 multiple myeloma (lVlM) cells stimulates IL-6
`secretion from bone marrow stronial cells (BMSCs). (a) IM-9 l\/[M cells, BMSCs, and both
`together were cultured for 24 h (D) and 48 h (I). (b) lM-9 myeloma cells, BMSCs, and
`both together were cultured for 48 h. (c) BMSCs were cultured for 48 h in the presence of
`media alone, VEGF (10 ng/ml), TNF-oc (10 ng/ml), or TGF-fil (10 ng/ml). (d) Two multi-
`ple myeloma patient BMSCs (I, 0) were cultured for 24 h with TNF-oc (0.0001—10 ng/ml).
`TNF—a (a) or IL—6 (1), c, d ) levels were measured in culture supematants by ELISA. Values
`
`represent the mean (:: SD) of triplicate cultures.
`
`As a result, adherence of myeloma cells to BMSCs is significantly upregulated
`by TNF-oz—related induction of these adhesion molecules. Adherence of myeloma
`cells to BMSCs induces drug resistance of myeloma cells, induces IL-6 secre-
`tion in BMSCs, and activates p44/42 MAPK in myeloma cells, thereby promoting
`tumor cell growth. Because thalidomide and its analogs are potent inhibitors of
`TNF-oz production they may not only act directly on myeloma cells but also
`act indirectly by inhibiting cytokine cross-talk in the myeloma/BMSC microen—
`vironment. In conclusion, TNF-oz induces an increase in proliferation, as well as
`MAPK/ERK activation, in myeloma cells, and it induces lL-6 secretion, as well
`as NF-ICB activation, in BMSCs. TNF-oc induces adhesion molecules on myeloma
`
`Annu.Rev.Med.2002.53:629-657.Downloadedfromwww.annualreviews.orgbycynthia.montalvo@bakerbotts.comon01/
`
`
`
`
`
`16/18.Forpersonaluseonly.
`
`|PR2018-00685
`
`Celgene EX. 2050, Page 13
`
`IPR2018-00685
`Celgene Ex. 2050, Page 13
`
`
`
`642
`
`RICHARDSON I HIDESHIMA I ANDERSON
`
`cells and BMSCs, increasing IL-6 secretion and the binding of myeloma cells to
`BMSCs; conversely, blockade of TNF-uc—induced NF-ICB activation inhibits these
`sequelae. These studies confirm a central role for TNF-oz in the growth and sur-
`vival of myeloma cells in the bone marrow milieu and suggest the utility of novel
`therapeutics targeting TNF-oz in multiple myeloma.
`Keifer et al. recently reported that thalidomide inhibited NF-ICB activation trig-
`gered by TNF-oc in Jurkat T cells (74). This inhibitory effect of thalidomide on
`NF-ICB activation was mediated by suppression of IKB kinase (IKK) activity.
`Phosphorylation of IKBa, inhibitory molecule of NF-ICB, by IKK is essential for
`degradation of IKBoc. Since adherence of myeloma cells to BMSCs induces IL-6
`secretion from BMSCs via NF-ICB activation (60), and TNF-oz induces adhesion
`molecules on both myeloma cells and BMSCs also via NF-KB activation, NF-KB
`may be a novel therapeutic target.
`A recent study demonstrated that thalidomide and lMiDs augment natural
`killer (NK) cell cytotoxicity in multiple myeloma (75). This report showed that
`thalidomide and IMiDs do not induce T cell proliferation alone but act as co-
`stimulators to trigger proliferation of anti-CD3-stimulated T cells from multiple
`myeloma patients, accompanied by an increase in interferon-y and IL-12 secre-
`tion. Treatment of patient peripheral blood mononuclear cells with thalidomide
`or IMiDs triggered increased lysis o