REPORTS
`
`_..
`
`EXHIBIT
`
`,,,,,,,eA °'9v~ J
`) /7 I IS'-
`MARK RICHMAN, RPR
`
`Antitumor Effects of
`Doxorubicin in Combination
`With Anti-epidermal
`Growth Factor Receptor
`Monoclonal Antibodies
`
`Jose Baselga, Larry Norton,
`Hideo Masui, Aranasio
`Pandiella, Keren Coplan,
`Wilson H. Miller, Jr.,
`John Mendelsohn*
`
`Background: A variety of human
`tumors frequently express h_igh levels
`of epidermal growth factor (EGF)
`receptor and its ligand, transforming
`growth factor a (TGF-a), which in
`some tumors is associated with poor
`prognosis. Monoclonal antibodies
`(MAbs) that block the binding of
`TGF-a or EGF to the receptor can
`inhibit proliferation of tumor cells
`that express the · receptor. Studies
`suggest that these MAbs may en(cid:173)
`hance
`the antltumor effects of
`chemotherapy. Purpou: Our purpose
`was to study, in vitro and in vivo, the
`ant.itumor effects of doxorubicin in
`combination with antl-EGF receptor
`MAbs against tumor cells expressing
`high levels of EGF receptor. Our
`goal was to achieve maximum initial
`cytoreduction with high-dose doxo(cid:173)
`rubicin in association with prolonged
`blockade of EGF
`receptor with
`MAbs. Methods: Anti-EGF receptor
`MAbs 528 (isotype lgG2a) and 225
`(isotype lgG 1) were used in combina(cid:173)
`tion with doxorubicin against cells
`from human A431 squamous cell
`carcinoma and human MDA-468
`breast adenocarcinoma. Both A431
`and MDA-468 cells express high
`levels of EGF receptors and TGF-a.
`Cultured cells were
`treated with
`doxorubicin (range, 0-10 nM) in the
`presence or absence of MAb 528 or
`
`225 (range, 0-30 nM), At 48 hours,
`doxorubicin·containing medium was
`removed, and treatment with anti(cid:173)
`body was continued for 5 days, when
`cell prolifera.tion assays were per(cid:173)
`formed. The activity of the agents
`and the combinations against well(cid:173)
`establisbed xeoografts
`in BALB/c
`nude mice was also studied. In nude
`mice, doxorubicin was given at doses
`of 50-100 µg/20 g body weight on 2
`successive days, and MAbs 528 and
`225 were given at a dose range of 0-2
`mg intraperitoneally twice a week.
`Resu/Js: MAbs 528 and 225 both
`enhanced the antitumor effects of
`doxorubicin against A431 and
`MDA-468 tumor cells, producing ad·
`ditive growth suppression
`in cell
`cultures. MAb 528
`increased
`the
`antitumor effects of doxorubicin by
`32%-42%, and similar results were
`obtained with MAb 225. Io BALB/c
`athymic mice, the treatment or well(cid:173)
`established xenografts with either
`doxorubicin or anti-EGF receptor
`MAb alone
`temporarily
`inhibited
`growth, but the combination of both
`agents substantially enhanced aotl(cid:173)
`tumor activity over that or doxo(cid:173)
`rubicin alone in A431 and MDA-468
`cell xenografts. The combination
`treatment of mice bearing A431
`xenografts resulted in tumor eradlca-.
`tion of 40%-100% in the surviving
`mice in several independent experi(cid:173)
`ments. The enhanced antitumor ac(cid:173)
`tivity was dose dependent. Con(cid:173)
`clusions: Our results suggest that
`anti-EGF receptor MAbs substan·
`Hally enhance the effects of doxo(cid:173)
`ru bicin against well-established
`xenografts of tumor cells expressing
`high levels of EGF receptors. Im(cid:173)
`plications: Clinical trials with anti(cid:173)
`EGF receptor MAbs are being con(cid:173)
`ducted, and
`trials with anti-EGF
`receptor MAbs combined with doxo(cid:173)
`rubicin are planned. [J Natl Cancer
`Inst 85:1327-1333, 1993}
`
`Human epithelial tumors and tumor
`cell lines frequently express high levels
`of epidermal growth
`factor
`(EGF)
`receptor and its ligand, transfonning
`growth factor a (TGF-a) (/). In some
`tumors, this expression of high levels
`of EGF receptor and TGF-a is associ(cid:173)
`ated with a more aggressive clinical
`behavior and poor prognosis
`(2-4).
`These observations and the fact that the
`introduction of the TGF-a gene (also
`known as TGFA) into cells bearing
`EGF receptors has transforming capac(cid:173)
`ity (5,6) suggest that the autocrine
`pathway constituted by the EGF recep(cid:173)
`tor and TGF-a may have an important
`role in human tumors. 1
`We have produced monoclonal anti(cid:173)
`bodies (MAbs) that perform the follow(cid:173)
`I) bind to
`the EGF
`ing functions:
`receptor with affinity comparable to
`that of the natural ligand, 2) compete
`with EGF binding, and 3) block EGF(cid:173)
`induced
`tyrosine kinase activation
`(7-10). When cells that express the
`EGF
`receptor, produce
`its
`ligand
`TGF-a, and depend on activation of the
`EGF receptor for growth are cultured
`in the presence of saturating concentra(cid:173)
`tions of anti-recepior MAb, binding of
`TGF-cx to the EGF receptor is blocked
`and ligand-dependent cell prolifera1ion
`is inhibited (7,8, 11-13). Treatment with
`intraperitoneal injections of anti-EGF
`receptor MAbs results in marked inhi(cid:173)
`bition of tumor growth in mice bearing
`subcutaneous xenografts of cancer cell
`lines that express high levels of EGF
`receptors (14, 15). However, this treat(cid:173)
`ment does not consistently eliminate
`xenografts
`that are well established
`(14). Since the majority of pa1ients
`with
`solid
`tumors
`thal cannot be
`surgically removed have well-estab(cid:173)
`lished tumor masses, anti-EGF recepior
`MAbs alone will probably not be cura(cid:173)
`tive
`in
`this setting. Therefore, new
`
`•su "NOICS" section following "References.'"
`
`Journal of the National Cancer Institute, Vol. 85, No. 16, August 18, 1993
`Downloaded frQm http9: I /academic.oup. com/jnci/art i cle·abst r act/ 85/16/1327 /900381
`by Mt Sinai School of Medicine, Levy Library user
`o n 09 February 2018
`
`REPORTS 1327
`
`
`
`1 of 8
`
`Celltrion, Inc. 1045
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`strategies are required to explore the
`potential clinical applications of these
`agents.
`One such approach is to use com(cid:173)
`binations of chemotherapeutic agents
`and anti-EGF receptor MAbs. We con(cid:173)
`sider this a promising strategy for a
`number of reasons. First, data from
`culture of cells that depend on EGF
`receptor activation by endogenous
`TGF-a for growth demonstrate that,
`when these cells are cultured at low
`density,
`the addition of exogenous
`growth factor is required for optimal
`proliferation (11,16). Growth of these
`cells is inhibited by anti-EGF receptor
`MAbs (1 1, 16). By reducing the number
`of viable
`tumor cells, chemotherapy
`limit the supply of endogenous
`may
`growth factor in the environment of the
`remaining tumor cells and thus increase
`susceptibility to the antitumor effects of
`anti-EGF receptor MAbs. Second, there
`is experimental evidence that anti-EGF
`receptor MAbs may synergize with
`chemotherapy {17). Third, it has been
`shown recently that both the death of
`cells exposed to chemotherapy and the
`death of cells deprived of essentiaJ
`growth
`factors
`involve programmed
`cell death (apoptosis) (18-21). These
`findi ngs suggest the possibility of a
`common cytotoxic pathway, raising the
`potential for a combined therapeutic
`approach. Doxorubicin, an anthra(cid:173)
`cycline antibiotic that has been shown
`to increase the number of EGF recep(cid:173)
`tors in several cancer cell lines (22,23),
`was the chemotherapeutic agent chosen
`for the present studies.
`Our goal was to analyze, both in
`vitro and in vivo, the antitumor effects
`of combined
`treatment with doxo(cid:173)
`rubicin and anti-EGF receptor MAb
`against
`tumor cells expressing high
`levels of EGF receptor. Our results
`show a marked rumoricidal effect of
`this combination, particularly in vivo.
`where treatment has resulted either in
`eradication of well-established tumors
`or in growth
`inhibition substantially
`enhanced over that produced by doxo(cid:173)
`rubicin alone.
`
`Materials and Methods
`MAbs Against EGF Receptor
`
`Anti -EGF receptor MAbs S28. 22S, and 4SS
`
`were used. MAbs S28 (isotype l3G2a) and 22S
`(isotype lgG I) bind to the external ponioo of the
`EGF receptor with affinity companble to that of
`the natural ligand, compete with EGF binding,
`block EGF-induced tyrosine kinase activation,
`and inhibit proliferation of cells dependent on
`EGF receptor activation for gJ0\11\h (7-9). MAb
`455 (1s01ypc lgG I) binds to the external portion
`of the EGF receptor, but, unlike MAbs S28 and
`225, docs not compete with EGF binding. docs
`no1 block EGF-induced tyrosine ltinase activa(cid:173)
`tion, and docs not inhibit proliferation of cells
`dependent on EGF receptor activation for gro\11\h
`[(8); Baselga J, Masui H, Mendelsohn J, ct al.:
`unpublished observations).
`
`Cell Lines
`
`The human A431 squamous cell carcinoma
`cell line expresses large quantities of both EGP
`receptors (2 x 106 per cell) and TGF-<i (7,8) and
`is inhibited by MAbs 22S and S28 in culture and
`in xenografts (8.9, /4). The human MDA-468
`breast carcinoma cell line was obt1incd from Dr.
`C. Aneaga at Vanderbilt University (Nashville,
`Tenn.). Like A43 I cells, MDA-468 cells express
`high levels of EGF receptors (106 per cell) and
`TGF-o and is inhibited by MAbs 22S and 528 in
`culture and in xenografts (I 5). Both cell lines
`were grown at 37 oC in monolayer culture with
`Dulbecco's Modified Eagle medium (DMEM)
`and Ham's medium F-12 (1 : 1) containing 10-.
`fetal bovine scrum.
`
`of tumor if no tumor could be detected in two
`successive tumor evaluations.
`When tumors re.ached a mean siu of greater
`than 0.3 cm' (A43 l cells) or greater
`than
`0.2 cm> (MDA-468 cells), animals were divided
`trealmcol groups, Each animal
`into different
`experiment included four treatment BJOUps: con(cid:173)
`trol, doxorubicin alone, antibody alone, and
`doxorubicin io combination with antibody. Each
`treatment group consisted of at least five animals
`with comparable tumor siu among groups from
`the same experiments, so that initial tumor sjze
`was not a variable within a particular experi·
`ment. Doxorubicin was give.a inuapcritoncally in
`0.5 mL distilled water al a dose range of 0.100
`14gf20 g body weight for 2 successive days. MAb
`S28. MAb 22S, or nonspecific polyclonal mouse
`lgG antibody (Sigma CbemicaJ Co .• St. Louis.
`Mo.) was given intraperitoneally in O.S mL of
`phosphate-buffered saline (PBS) (pH 7 .4), with a
`dose range of 0.2 mg. Antibody treatment was
`administered twice weekly, a schedule that has
`previously been shown
`to maintain st.able
`receptor-sal\lrating blood levels at doses of I mg
`or more per iojection {14). Control animals were
`treated with O.S mL of PBS twice a week.
`Finally, groups of at least five mice were
`treated with a combination of doxorubicin and
`either MAb 528, MAb 22S. or polyclonal mouse
`lgG antibody at
`the same dose levels and
`treatment intervals as described above for eacli
`separate trcaUncnL
`
`Cell Culture a nd Growth Assay
`
`Results .
`Cells were seeded m 12-well pl11cs (Coming Treatment of A43I and MDA-468
`Glass Works. Coming, N.Y.) 11 a density of Cells lo Culture
`SOOO cells per well. On the following day. cells •
`were
`treated with varying cooccnuations of
`doxorubicin (range, 0-10 nM) (Adna Laborato(cid:173)
`nes, Columbus, Ohio) either in the presence or
`in the absence of M Ab S28, 22S, or 4SS at
`different concentrations (range. 0.30 nM). At 48
`hours. doxorubicin-containing medium was re·
`moved, cells were washed twice by addition and
`decantation of phosphate-buffered saline (PBS)
`(pH 7 .4 ), and the treatment with antibody was
`continued as before. On day S. the cells were
`counted after trypsinization using a cell counter
`(Coulter Electronics Inc .. Hialeah, Fla.).
`
`Culture of A431 or MDA-468 cells
`with doxorubicin at
`increasing con(cid:173)
`(0- 10 nM)
`for 2 days
`centrations
`resulted in a dose-dependent inhibition
`of growth assayed on day 5 of culture.
`Treatment with either MAb 528 or
`MAb 225 (0-30 nM) continuously for 5
`days produced a dose--dependent inhibi(cid:173)
`tion of growth of both cell lines. The
`combined
`treatment with
`increasing
`concentrations of doxorubicin and anti(cid:173)
`EGF receptor MAb 528 resulted in an
`additive inhibition of growth in both
`A431 cells and MDA-468 cells increas(cid:173)
`ing
`the
`inhibitory effects of doxo(cid:173)
`rubicin by 32%-42% (Fig. l). Similar
`results were obtained when doxorubicin
`was used in combination with anti-EGF
`receptor MAb 225 against both cell
`lines (data not shown). The increased
`antitumor activity was not observed
`when cells were treated with doxo(cid:173)
`rubicin in combination with the non(cid:173)
`blocking anti-EGF receptor MAb 455
`(data not shown).
`
`Assay of Tumor Growth in
`Athymk Mite
`
`BALB/c athymic mice, 6-8 weeks of age, were
`used. These mice were bred and maintained in
`accordance with institutional guidelines at the
`Memorial Sloan-Kettering Cancer Center animal
`facility as described previously (24). Cultured
`A43 l and MDA-468 cells were trypsiniud and
`washed with serum-free DMEM and Ham's
`medium F-12 (I: I). Cells (107) were injected
`subcutaneously into animals on day 0. MDA-468
`cells were injecred only into female mice. Tumor
`size was measured rwicc a week using
`the
`larger diameter x
`ronnula w/6 X
`(smaller
`diametcr)1 . Animals were considered to be free
`
`1328 REPORTS
`
`Journal of the National Cancer Institute, Vol. 85, No. 16, August 18, 1993
`
`Downloaded f<OOI http•' //acadcmic.oup .c0111/jnci/article· abstract/8S/16/1327/90038l
`by Mt Sin ai School of Medicine, Levy Library ua:er
`on 0 9 February 2018
`
`
`
`2 of 8
`
`Celltrion, Inc. 1045
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`A
`
`100
`
`B
`
`100
`
`l:fu.
`
`;:c
`
`~
`lI.
`
`;:z:
`I
`
`;:c
`
`....
`
`~
`~,:
`
`80
`
`...
`..
`40 -
`
`60
`
`20
`
`80
`
`w c:
`
`8
`
`60
`
`40
`
`~
`
`20
`
`~
`
`. . ,I.
`~
`
`lJ;.
`
`:
`
`~
`
`I
`
`..
`
`~ IZ ...
`
`l.::c
`
`""
`. l:z:~
`
`0
`MAb 528 {nM) 0 5 1030
`
`0 51030
`
`0 5 10 30
`
`0 5 1030
`
`0
`MAb 528 {nM) 0 5 10 30
`
`0 5 1030
`
`0 5 10 30
`
`0 5 1030
`
`OOXO{nM)
`
`0
`
`2
`
`5
`
`10
`
`OOXO {nM)
`
`0
`
`2
`
`5
`
`10
`
`Fig. 1. Inhibitory effects of doxorubicin in combination wirh MAb 528 in cultures of A43 I cells (A) and MDA-468 cells (B). Doxorubicin was added to
`cultures a l increasing conccntra11ons (0-10 nM) for 2 days. MAb 528 was present at described concentrations for S days. Cell number was de1ennined on
`day S using a cell counrer. Rcsulls represent rhe mean :!: SE of triplicate readings.
`
`Treatment of A431 Cell Xeno~rafts
`
`We explored the in vivo effects of
`the combination therapy in a series of
`experiments. First, doxorubicin admin(cid:173)
`istered at 100 µg/20 g body weight
`intraperitoneally on 2 successive days
`was determined to be an LD 10 • that is,
`a dose killing 10% of mice. Our initial
`pilot study of combination therapy used
`a limited number of mice bearing well(cid:173)
`established A43 l xenografts more than
`0.4 cmJ
`in size. When animals, in
`groups of at least five, were treated
`with either MAb 528 alone (I mg twice
`a week for 2 weeks) or doxorubicin
`alone (100 µg/20 g body weight for 2
`days), tumor growth rate was modestly
`inhibited by less than 50% at 15 days
`(data not shown). However, treatment
`of eight animals with the combination
`of MAb 528 and doxorubicin, at the
`same dose and schedule, resulted in
`complete eradication of tumors in all
`five animals (100%) that were alive at
`1 S days. After ihe 15 days of follow(cid:173)
`up. the animals in this group were
`killed, and lack of residual tumor was
`confirmed by necropsy.
`We conducted experiments to con(cid:173)
`firm the antitumor activity of doxo(cid:173)
`rubicin and MAb 528 in larger numbers
`of athymic m ice bearing well(cid:173)
`established subcutaneous xenografts of
`A431 cells. When tumors reached a
`mean size of 0.4 cmJ, the mice were
`allocated into groups with comparable
`
`tumor sizes, and therapy with the above
`regimen was staned (Fig. 2, A). In
`untreated animals, tumors grew rapidly
`and all of the animals died within 5
`weeks, as expected. Doxorubicin alone
`initially slowed
`tumor growth, but
`within I 0 days tumors began to regrow
`rapidly. Treatment with MAb 528
`reduced tumor growth rate; however, as
`expected from previous studies with
`well-established
`tumors {14),
`the
`xenografts were not eliminated.
`In
`contrast,
`the combination
`treatment
`with doxorubicin and MAb 528 re(cid:173)
`sulted in a major antitumor effect (Fig.
`2. A).
`three
`total of
`We performed a
`independent experiments
`(the pilot
`study above plus
`the experiments
`shown in Fig. 2, A and B) to determine
`in A43 l cell xenografts the antitumor
`activity of a combination of doxo(cid:173)
`rubicin, I 00 µg/20 g weight given on 2
`successive days and I mg MAb 528
`given twice a week. The data from
`these experiments have been pooled in
`Table I. A total of 26 animals were
`treated with the combination therapy.
`After 15 days of treatment, 2 1 animals
`(81 %) were alive. Eighteen of these
`mice were free of tumor by day 15,
`and the other three had no tumors by
`day 30. Therefore, by day 30, all 2 1
`surviving animals in the combination
`group (100%) had their tumors eradi(cid:173)
`cated. After 100 days from the start of
`treatment, only two of the 16 surviving
`
`the combined
`treated with
`animals
`therapy had tumors, which were re(cid:173)
`growing slowly.
`To demonstrate that the antitumor
`effects of the combination were not due
`to a nonspecific action ·of mouse IgG
`given in combination with doxorubicin,
`we treated mice bearing comparable
`A43 I xenografts with doxorubicin in
`combination With a nonspecific' mouse
`IgG. We used
`the same dose and
`schedule as with MAb 528. There were
`no differences
`in mean
`tumor size
`between animals
`treated with doxo(cid:173)
`rubicin alone or in combination with
`mouse IgG (Fig. 2, B).
`Additional experiments were de(cid:173)
`signed to determine the dose depend(cid:173)
`ency of the response to combination
`therapy. Our previous phannacologic
`studies indicated that an MAb treat(cid:173)
`ment schedule of I mg given intra(cid:173)
`peritoneally twice weekly would pro(cid:173)
`duce stable blood levels of antireceptor
`MAb capable of saturating xenograft
`EGF receptors (more than I 0 times the
`equilibrium dissociation constant, Kd)
`(14). The results observed with MAb
`528, given alone or in combination
`with doxorubicin and at a higher dose
`(2 mg intraperitoneally), twice a week
`for 5 weeks on a schedule similar to
`in Fig. 2, A. were
`the experiment
`comparable to those observed with the
`I -mg dose (Table I ). When xenografts
`were treated with combination therapy
`utilizing MAb doses of less than I mg,
`
`Journal of lhe National Cancer Institute, Vol. 85, No. 16, August 18, 1993
`
`Downloaded from hctps ://academic .oup.cocn/jnci/article-abstract/85/16/1327 /900381
`by Mt Sinai School of Medicine, Levy Library user
`on 09 February 2018
`
`REPORTS 1329
`
`
`
`3 of 8
`
`Celltrion, Inc. 1045
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`MAb528
`
`l l l l l l l l H
`uooxo
`
`ooxo
`
`A
`
`14
`
`12
`
`10
`
`(')
`
`E
`.£,. 8
`Cl>
`.~ en
`~
`E 6
`:a
`
`4
`
`2
`
`Fig. 2. Antitumor activity of MAb
`528
`in combination with doxo(cid:173)
`rub i c:i n
`(DOXO) on wcll (cid:173)
`established A43 I squamous cell
`carcinoma xenografts in athynuc
`mice. Trca1mcnt was s1arted when
`tumors reached a mean siu or 0.4
`cmJ on day 11 in Fig. 2. A or day
`9 in Fig. 2, B. Each 1rea1mcnt
`group consisted of a1
`leas1 five
`animals. A 1olal of 10 mice were
`treated in the combination group in
`1he experimenl ploned m Fig. 2, A,
`and 8 animals in Fig. 2, B. Resuhs
`are given in mean 1umor size :!:
`SE. Error bars arc not present
`when less than three animals re(cid:173)
`mained alive in a cenain treatment
`group. Doxorubicin (100 µg/20 g
`body weigh1) was given
`in1ra(cid:173)
`peritoneally on days I and 2 of
`trca1ment (day 11 and 12 or day 9
`and 10 aflcr tumor cell inocula-
`1ion). MAb 528 (I mg) was given
`DOXO + MAb 528
`inlrapentoncally on day I of treat(cid:173)
`o.__~-L--""::::o..io--o-o.i-o-ci~1--~-'
`ment and twice a week thereafter
`30
`40
`50
`0
`10
`20
`for a
`total of
`I 0 doses. A)
`days
`Treatment wnh either doxorubicin
`alone or MAb alone partially in(cid:173)
`hibited tumor growth. Doitorubicin in combination with MAb 528 completely eradicated all tumors in the animals surviving on day 30 (N = 8). 8)
`Doxorub1cin m combination with a nonspecific mouse lgG did not result m a g.reater antitumor effect than doxorubicin alone, while doxorub1cm in
`combination with MAb 528 resulted in disappearance of all tumors 1n 1he animals surviving on day 30 (N = 6). Arrows show days on which 1reatmen1 was
`administered.
`
`MAb528
`Hl~lHlli
`uooxo
`
`B 10
`
`8
`
`C')
`
`E 6
`.£,.
`Cl>
`N
`·c;;
`....
`~ 4
`.a
`
`2
`
`reduced
`the antitumor effects were
`(data not shown). This result suggests
`that saturation of EGF receptors may
`be required to achieve optimal results
`with doxorubicin plus anti-EGF recep(cid:173)
`tor blockade therapy. We also tested
`the antitumor response to the combina(cid:173)
`tion
`therapy with a
`lower dose of
`doxorubicin
`(50 µg given
`intra-
`
`peritoneally on 2 successive days).
`Doxorubicin treatment alone had no
`effect, but combination therapy with
`doxorubicin and MAb 528 was again
`seen to have a strong inhibitory effect.
`However, in contrast to the results with
`higher doses of doxorubicin, no tumor
`xenografts were completely eradicated
`(data not shown).
`
`Table J. Eradication of tumors by high-dose doxorubicm and an11-EGF receptor MAbs•
`
`Trca1men1t
`
`None
`Doxorubicin
`Doxorubicin + mg lgG
`1 mg MAb 528
`Doxorubicin + I mg MAb 528
`2 mg MAb 528
`Doxorubicin + 2 mg MAb 528
`I mg MAb 225
`Doxorubicin + I mg MAb 225
`
`Initial No.
`of mice
`
`31
`29
`9
`12
`26t
`6
`6
`10
`10
`
`Mice alive
`on day 15
`
`Mice tumor-free
`on day 15
`
`No.
`
`27
`23
`6
`12
`21
`6
`6
`10
`10
`
`%
`
`87
`79
`67
`100
`81
`100
`100
`100
`100
`
`No.
`
`I
`0
`I
`18§
`0
`4
`
`2
`4
`
`%
`
`3
`3
`0
`8
`69
`0
`67
`20
`40
`
`The above experiments were per(cid:173)
`formed with MAb 528 lgG2a, which
`has
`the
`capacity
`to
`activate
`complement-mediated cytotox1c1ty as
`well as cell-mediated toxicity (25) . We
`decided to repeat the treatment pro(cid:173)
`tocols with MAb 225
`lgG l, which
`bears an Fe portion that is immuno(cid:173)
`logically nonreactive and has been
`shown in earlier studies to be unable to
`mediate toxicity via the complement or
`cellular immune pathways (25). Again,
`strong anticancer effects were observed
`with combined therapy; tumors were
`eradicated in four of the I 0 animals
`(40%)° surviving on day 15 after
`treatment (Fig. 3, A; T_able I). When
`doxorubicin was administered at a 50%
`reduced dose (50 µ.g/20 g body weight
`on 2 successive days) in combination
`with MAb 225 (I mg intraperitoneally
`twice a week). results similar to those
`obtained with reduced-dose doxorubicin
`and MAb 528 were observed (data not
`shown).
`
`treated with doxorubicin, MAb, or
`•Well-<.Stablished (6-12 days) A431 xenografts were
`combination therapy. The stalus of animals and xenografts 15 days after the initiaiion of trca1men1 is
`shown. Results of five experiments are given. Data on animals that received the same lrealmenl and
`. dose have been pooled.
`tDoxorubicin was given inllaperitoneally al 100 µg/20 8 twice weekly. MAb or lgG was
`administered inuape.ritoneally twice weekly al the smed doses.
`·
`·
`t: Animals from the pilol study and experiments shown in Fig. 2, A and B. arc included.
`§The remaining three animals alive with lumors on day 15 had no 1umors by day 30.
`
`Treatment of MDA-468
`Cell Xenografts
`
`the effects of
`We also examined
`combination
`therapy on MDA-468
`breast adenocarcinoma cells in nude
`
`1330 REPORTS
`
`Journal of the National Cancer Institute, Vol. 85, No. 16, Augus1 18, 1993
`
`Downloade d from h t t ps , I /aeadomic .oup. eom/jnei/artiele- abotract/85/16/1327/900381
`by Ht Sinai School of Medicine. Levy Library user
`on 09 February 2018
`
`
`
`4 of 8
`
`Celltrion, Inc. 1045
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`MAb225
`+ ~ + ~ ~ + + + +
`uooxo
`
`B 1.2
`
`1.0
`
`... ~.s
`E
`~
`1! .6
`VI
`0
`E
`~ .4
`
`A
`
`3.5
`
`3.0
`--- 2.5
`M
`E
`.£.
`2.0
`QI
`N
`'iii
`.....
`0 1.5
`E
`.a
`
`1.0
`
`0.5
`
`Fig. 3. A) Antilumor act1v1ty of
`MAb 225
`in combination with
`doxorubici n
`(DOXO) on well-
`established A431 squamous cell
`carcinoma xenografts. Treatment
`was su111ed when tumors rc.lched a
`mean sii:.e of 0.3 cm3 . A total of
`I 0 mice were treated in the com-
`bination group. Results are gi,•en
`in mean tumor sii:.e :!: SE. Doxo-
`rubicin (I 00 µg/20 g body weight)
`was given intraperitoneally on days
`I and 2. MAb 225 (I mg) was
`given intraperitonc.llly on day I
`and twice a week thereafter for a
`total of 10 doses. Treatment with
`either doxorubicin alone or MAb
`alone resulted in transient inhibi-
`tion of tumor growth. Doxorubicin
`in combination with MAb 225 had
`a pronounced an1i1umor activity.
`Arrows show days on which treat(cid:173)
`ment was administered. B) Anti·
`tumor activity of MAb 528 in
`combination with doxorubicin
`(DOXO) on well-established
`MDA-468 breast adcnocarcmoma xcnografts. Treatment was staned when tumors reached a mean sii:.c of 0.2 cm'. A total of 9 mice were treated in the
`combination group. Results are given in mean tumor size :!: SE. Doxorubicin (100 µg/20 g body weight) was given intrapcritoncally on days I and 2.
`MAb 528 (2 mg) was given mtraperitoncally on day I and 1w1ce a week thereafter for a total of 10 doses. Treatment with either doxorubicin alone or MAb
`alone resulted in transient inhibition of tumor growth. Doxorubicin in combination with MAb 528 confirmed the antitumor activity observed in A43 I cells.
`Arrows show days on which treatment was administered.
`
`MAb528
`
`+ l + ~ + + + + +
`uooxo
`
`DOXO
`
`control
`
`.2
`
`o.__~-'-~~.._~--'-~~-'-~--'
`0
`10
`20
`30
`40
`50
`days
`
`in
`tumors
`the
`I 00% of
`to
`mouse xenografts. To treat animals, we· 40%
`used
`the same schedule as
`in
`the
`surviving animals
`in 5 experiments.
`experiment showed in Fig. I, A, except Lower doses of doxorubicin or anti(cid:173)
`in combination were
`less
`chat the dose of MAb 528 was 2 mg body
`given
`intraperitoneally
`twice weekly effective.
`based on our previous studies with this
`cell line (/ 5). Doxorubicin alone at the Discussion
`maximally tolerated dose (I 00 µg/20 g
`body weight on 2 successive days) did
`not have a noticeable antitumor ac(cid:173)
`tt v1ty,
`and
`treatment of
`these
`established tumors with MAb 528 only
`slowed tumor growth, which was pre(cid:173)
`dicted on the basis of our previous
`srudies (/ 5). In mice bearing MDA-468
`xenografts, combination therapy with
`doxorubicin and MAb confirmed the
`in A43 I
`antitumor activity observed
`xenografts (Fig. 3, B). One of nine
`( 11 %) treated mice surviving on day 30
`was tumor free.
`in mice
`In summary, our studies
`demonstrate that the treatment of well·
`established A43 l cell and MDA-468
`cell xenografts with a high dose of
`doxorubicin (I 00 µgl 20 g weight on 2
`successive days) in combination with
`either MAb 528 or 225 at a dose of I
`mg twice a week, or higher, results in a
`major antitumor effect. The
`results
`were most impressive in our A43 I cell
`xenograft model, where the combina(cid:173)
`tion treatment resulted in eradication of
`
`These studies demonstrate that anti(cid:173)
`EGF receptor MAbs are capable of
`enhancing the effects of doxorubicin
`against well-established xenografts of a
`squamous cell carcinoma line and an
`adenocarcinoma that express high lev(cid:173)
`els of EGF receptor. Based on consid(cid:173)
`erations presented earlier in the report,
`our goal was to achieve a maximum
`initial cytoreduction by chemotherapy
`in association with a complete receptor
`blockade. Therefore, our in vivo ex(cid:173)
`periments were designed
`to give a
`high-dose intensity of doxorubicin ac(cid:173)
`companied by a prolonged blockade of
`EGF receptors with MAbs. We ob(cid:173)
`served comparable results in multiple,
`independently conducted experiments,
`employing two different anti-EGF re(cid:173)
`ceptor MAbs with distinct isotypes and
`tumors with
`two different histologic
`types. The inhibition was dose depend(cid:173)
`ent for both doxorubicin and MAb.
`Two studies (/7,26) have shown that
`combination treatment with the chem-
`
`otherapeutic agent cisplatin and anti~
`growth factor receptor MAb inhibited
`tumor xenograft growth in nude mice.
`Our studies demonstrate, however, for
`the first time that treatment of well(cid:173)
`established tumors with the combina(cid:173)
`therapy and a
`tion of antireceptor
`chemotherapeutic agent has resulted in
`tumor-free animals. as assessed either
`by a necropsy examination or by a 100-
`day follow-up period.
`Our results lead us to question the
`mechanism
`that causes
`the marked
`antitumor effects of chemotherapeutic
`agents when used in combination with
`antigrowth
`factor
`receptor MAbs.
`Doxorubicin has been shown to in·
`crease EGF
`receptor expression
`in
`some cell lines (22,23). In experiments
`riot presented here, we found
`that
`doxorubicin administered to A43 I cells
`in noncytocidal concentrations
`in(cid:173)
`creased the levels of EGF receptor and
`TGF-a messenger RNA by more than
`threefold within hours, followed by a
`2.6-fold increase in EGF receptor num(cid:173)
`ber and a 2.4-fold
`increase
`in
`the
`amount of TGF-a released into the
`conditioned culture medium (27). These
`results, together with the strong anti(cid:173)
`tumor activity of combined therapy, led
`us to entertain the hypothesis that, in
`cells
`recovering
`from noncytocidal
`
`Journal of the National Cancer Institute, Vol. 85, No. 16, August 18, 1993
`Downloaded trm https' //academic .oup .com/j nci/arti clo-ilbstr act/85/16/1327 /900381
`by Mt Sinai School of Medicine. Levy Library user
`on 09 February 2018
`
`REPORTS 1331
`
`
`
`5 of 8
`
`Celltrion, Inc. 1045
`Celltrion v. Genentech
`IPR2017-01122
`
`

`

`increased ac(cid:173)
`treatment,
`doxorubicin
`tivity
`in
`the EGF
`receptor- TGF-a
`autocrine pathway could be a compen(cid:173)
`satory cellular response to lhe effects
`of doxorubicin. Therefore, blockage of
`the EG F
`receptor-TGF-a autocrine
`loop with antireceptor antibodies during
`and after doxorubicin treatment might
`deprive these cells of the supportive
`effects of the EGF receptor-mediated
`signal
`transduction pathway and,
`thereby, increase the opportunity for
`doxorubicin-mediated cytotoxicity. Al(cid:173)
`though blockade of this pathway with
`antireceptor MAb also potentiated the
`efficacy of doxorubicin treatment in
`xenografts of MDA-468 breast ade(cid:173)
`nocarcinoma cells, we did not detect an
`increase in EGF receptor or TGF-a
`expression in these cells after exposure
`to doxorubicin. We are currently ex(cid:173)
`ploring the hypothesis that MDA-468
`cells, despite the lack of an increase in
`EGF receptor or TGF-a, appear to have
`an increased dependence on activation
`of the EOF receptor signal transduction
`pathway when they are treated with
`doxorubicin. It is also possible that the
`antitumor effects of chemotherapy and ·
`antigrowth factor receptor MAbs are
`exerted by related mechanisms that are
`synergistically affected by combination
`treatment, since
`it has been shown
`experimentally that both chemotherapy
`factor deprivation can
`and growth
`activate programmed cell death
`in
`cancer cells (18-21).
`Additional factors may have contrib(cid:173)
`uted to the marked antitumor activity
`of the combination
`therapy against
`tumor xenografts. First, our MAb could
`be blocking interactions between the
`injected tumor cells and nearby stromal
`cells that may produce TGF-a or EGF,
`since it has been shown that stromal
`celJs have the ability to produce growth
`factors that stimulate cancer cells in a
`paracrine fashion (28). Second, a role
`for immune mechanisms in the marked
`antitumor effect seen in nude mouse
`,;enografts cannot be ruled out, since
`treatment has been
`doxorubicin
`re(cid:173)
`ported to increase complement suscep(cid:173)
`tibility in cell culture (29). This pos(cid:173)
`sibility was partially addressed by our
`use of two anti-EGF receptor anti(cid:173)
`bodies, MAbs 528 and 225, only one of
`which (MAb 528, IgG2a) has demon(cid:173)
`strated partial complement-mediated
`
`cytotoxicity and macrophage-activating
`cytotoxic capacity against cultured
`A43 I cells (25). Although the slightly
`higher antitumor activity observed with
`the administration of doxorubicin plus
`MAb 528 could be due to the activa(cid:173)
`tion of mediators of inflammation,
`excellent antitumor activity was also
`obtained with doxorubicin plus MAb
`225. Therefore,
`it
`is unlikely
`that
`complement or other immune mecha(cid:173)
`nisms played a major role
`in
`the
`observed
`results. A nonspecific
`antibody-mediated effect on drug
`cytotoxicity was ruled out by substitut(cid:173)
`ing nonspecific mouse immunoglobulin
`for specific anti-EGF recep