H. Michael Shepard, Gail D.
`Lewis, Jay C. Sarup, Brian
`M. Fendly, Daniel Maneval,
`Joyce Mordenti, Irene
`Figari, Claire E. Kotts,
`Michael A. Palladino, Jr. ,
`Axel Ullrich, and Dennis
`Slamon
`
`Monoclonal Antibody
`Therapy of Human Cancer:
`Taking the HER2
`Protooncogene to the Clinic
`
`R. Lib/au, Ph. Gajdos, F.A.
`Bustarret, R. El Habib, J. F.
`Bach, and E. Morel
`
`Intravenous -y-Giobulin in
`Myasthenia Gravis:
`Interaction with
`Anti-Acetylcholine Receptor
`Autoan tibodies
`
`Hiroaki Ida, Akihiko Kurata,
`Katsumi Eguchi, Atsushi
`Kawakami, Kiyoshi Migita,
`Takaaki Fukuda, Tatsufumi
`Nakamura, Yukio
`Kusumoto, Jay A.
`Berzofsky, and Shigenobu
`Nagataki
`
`Different B-Cell Responses
`to Human T-Cell
`Lymphotropic Virus Type I
`(HTL V-I) Envelope
`Synthetic Peptides in
`HTLV-1-Infected Individuals
`
`Complete contents listing
`inside
`
`VOLUME 11 , NUMBER 3
`MAY 1991
`ISSN 0271-9142
`JCIMDO 11(3) 117-160 (1991)
`
`JUN 13 1991
`
`PFIZER EX. 1048
`Page 1
`
`

`

`Journal of
`Clinical
`Immunology
`
`The Journal of Clinical Immunology is devoted exclu(cid:173)
`sively to clinical immunology and its application to the
`practice of medicine_
`
`Journal of Clinical Immunology is published bi(cid:173)
`monthly by Plenum Publishing Corporation, 233
`Spring Street, New York, N.Y. 10013. Subscription
`orders should be addressed to the publisher.
`
`Journal of Clinical Immunology is abstracted or in(cid:173)
`dexed in Ad Referendum, Allergy, Beck Medical
`Information, Biological Abstracts, Chemical Ab(cid:173)
`stracts, Current Awareness in Biological Sciences,
`Current Contents, Excerpta Medica, Index Medicus,
`ISI/Biomed, Literature Scan: Transplantation, Sci(cid:173)
`ence Citation Index, and Texas Medicine.
`
`Advertising inquiries should be addressed to Joseph
`Bourgholtzer, Inc. , The ffil Building, Box 832, Mah(cid:173)
`wah, New Jersey 07430-telephone: (201) 818-1010;
`fax: (201) 818-0086.
`
`© 1990 Plenum Publishing Corporation. Journal of
`Clinical Immunology participates in the Copyright
`Clearance Center (CCC) Transactional Reporting Ser(cid:173)
`vice. The appearance of a code line at the bottom of
`the first page of an article in this journal indicates the
`copyright owner's consent that copies of the article
`may be made for personal or internal use. However,
`this consent is given on the condition that the copier
`pay the flat fee of $6.00 per copy per article (no
`additional per-page fees) directly to the Copyright
`Clearance Center, Inc., 27 Congress Street, Salem,
`Massachusetts 01970, for all copying not explicitly
`permitted by Sections 107 or 108 of the U.S. Copyright
`Law. The CCC is a nonprofit clearinghouse for the
`payment of photocopying fees by libraries and other
`users registered with the CCC. Therefore, this consent
`does not extend to other kinds of copying, such as
`copying for general distribution, for advertising or
`promotional purposes, for creating new collective
`works, or for resale, nor to the reprinting of figures,
`tables, and text excerpts. 0271-9142/90 $6.00
`
`Subscription rates: Volume 10, 1990 (6 issues) $210.00
`(outside the U.S., $245.00). Price for individual sub(cid:173)
`scribers certifying that the journal is for their personal
`use, $52.00 (outside the U.S., $65.00).
`
`Second-class postage paid at New York, N.Y., and at
`additional mailing offices.
`
`Postmaster: Send address changes to Journal of Clin(cid:173)
`ical Immunology, Plenum Publishing Corporation, 233
`Spring Street, New York, N.Y. 10013.
`
`Printed in USA.
`
`EDITOR
`
`SUDHIR GUPTA University of California, Irvine, CA
`
`ASSOCIATE EDITORS
`
`REBECCA H_ BUCKLEY Duke University School of Medicine, Durham, NC
`
`MAX D_ COOPER University of Alabama Medical Center, Birmingham, AL
`
`ROBERT A. GOOD All Children's Hospital, SL Petersburg, Florida
`
`KIMISHIGE ISHIZAKA The Johns Hopkins University School of Medicine, Baltimore, MD
`
`JOHN H _ KERSEY University of Minnesota School, Minneapolis, MN
`
`ADVISORY BOARD
`
`HARVEY COL TEN Washington University School of Medicine, St. Louis, MO
`
`ANTHONY FAUCI National Institutes of Health, Bethesda, MD
`
`LARS HANSON University of Goteborg, Goteborg, Sweden
`
`ENG TAN Scripps Clinic and Research Foundation, La Jolla, CA
`
`THOMAS WALDMANN National Institutes of Health, Bethesda, MD
`
`HANS WIGZELL Karolinska Institute, Stockholm, Sweden
`
`EDITORIAL BOARD
`
`CLARK L. ANDERSON, Columbus, OH
`ROBERT F. ASHMAN, Iowa City, lA
`KENNETH A_ AULT, Portland, ME
`ANTONY BASTEN, Sydney, Australia
`
`BENJAMIN BONA VIDA, Los Angeles,
`CA
`LAURENCE BOXER, Ann Arbor, MI
`JOHN c_ CAMBIER, Denver, CO
`
`EDWARD A. CLARK, Seattle, WA
`LORAN T_ CLEMENT, Los Angeles, CA
`JOHN M. DWYER, Sydney, Australia
`
`GEORGE S. EISENBARTH, Boston, MA
`GABRIEL FERNANDES, San Antonio,
`TX
`STEVEN GILLIS, Seattle, WA
`JANIS V. GIORGI, Los Angeles, CA
`EDMOND GOIDL, Baltimore, MD
`
`DAVID W. GOLDE, Los Angeles, CA
`
`SIDNEY H. GOLUB, Los Angeles, CA
`
`JOHN A. HANSEN, Seattle, WA
`
`EUGENIE S. KLEINERMAN,
`Houston, TX
`
`WALTER KNAPP, Vienna, Austria
`
`WILLIAM J. KOOPMAN, Birmingham,
`AL
`
`VINA Y KUMAR, Dallas, TX
`
`ROGER J. KURLANDER, Durham, NC
`
`H. CLIFFORD LANE, Bethesda, MD
`
`JEFFREY A- LEDBETTER, Seattle,
`WA
`RICHARD G_ LYNCH, Iowa City, lA
`MART MANNIK, Seattle, WA
`DEAN D. METCALFE, Bethesda, MD
`CAROL A. NACY, Washington D_c_
`JOHN R_ ORTALDO, Frederick, MD
`
`ERIC A. OTTESEN, Bethesda, MD
`
`ROBERTSON PARKMAN, Los
`Angeles, CA
`
`RONALD PENNY, Sydney, Australia
`CARL W. PIERCE, SL Louis, MO
`STEPHEN H. POLMAR, SL Louis, MO
`MARIO RICCI, Florence, Italy
`ROBERT R. RICH, Houston, TX
`STANLEY A. SCHWARTZ, Ann
`Arbor, MI
`
`GREGORY W. SISKIND, New York,
`NY
`
`HANS L. SPIEGELBERG, La Jolla, CA
`TIMOTHY A. SPRINGER, Boston, MA
`
`FRANK H . VALONE, San Francisco,
`CA
`PETER A. WARD, Ann Arbor, MI
`RALPH J.P_ WEDGWOOD, Seattle, WA
`NATHAN J. ZVAIFLER, San Diego,
`CA
`
`PFIZER EX. 1048
`Page 2
`
`

`

`Journal of Clinical Immunology, Vol. 11 , No. 3, 1991
`
`Special Article
`
`Monoclonal Antibody Therapy of Human Cancer: Taking
`the HER2 Protooncogene to the Clinic
`
`H. MICHAEL SHEPARD, 1
`4 GAIL D. LEWIS, 1 JAY C. SARUP, 1 BRIAN M. FENDLY, 1 DANIEL
`•
`
`MANEY AL, 1 JOYCE MORDENT!, 1 IRENE FIGARI, 1' CLAIRE E. KOTTS, 1
`MICHAEL A. PALLADINO, JR., 1 AXEL ULLRICH, 2 and DENNIS SLAMON3
`
`Accepted: January 22, 1991
`
`The HER2 protooncogene encodes a 185-kDa transmem(cid:173)
`brane protein (p185HER2) with extensive homology to the
`epidermal growth factor (EGF) receptor. Clinical and
`experimental evidence supports a role for overexpression
`of the HER2 protooncogene in the progression of human
`breast, ovarian, and non-small cell lung carcinoma. These
`data also support the hypothesis that p 185HER2 present on
`the surlace of overexpressing tumor ceHs may be a good
`target for receptor-targeted therapeutics. The anti(cid:173)
`pl858ER2 murine monoclonal antibody (muMAb) 4D5 is
`one of over 100 monoclonals that was derived following
`immunization of mice with cells overexpressing
`pl85"ER2. The monoclonal antibody is directed at the
`extracellular (ligand binding) domain of this receptor
`tyrosine kinase and presumably has its effect as a result of
`modulating receptor function. In vitro assays have shown
`that muMAb 4D5 can specifically inhibit the growth of
`tumor cells only when they overexpress the HER2 pro(cid:173)
`tooncogene. MuMAb 4D5 has also been shown to en(cid:173)
`hance the TNF-a sensitivity of breast tumor cells that
`overexpress this protooncogene. Relevant to its clinical
`application, muMAb 4D5 may enhance the sensitivity of
`pl85HER2-overexpressing tumor cells to cisplatin , a
`chemotherapeutic drug often used in the treatment of
`ovarian cancer. In vivo assays with a nude mouse model
`have shown that the monoclonal antibody can localize at
`
`1Department of Developmental Biology, Genentech, Inc., 460
`Point San Bruno Boulevard, South San Francisco, California
`94080.
`2Max Planck Institute for Biochemistry , Martinsreid , Germany.
`3Department of Hematology and Oncology, University of Cali(cid:173)
`fornia, Los Angeles , California 90024.
`"To whom correspondence should be addressed.
`
`the tumor site and can inhibit the growth of human tumor
`xenografts which overexpress pl858 ER2. Modulation of
`pl85HER2 activity by muMAb 4D5 can therefore reverse
`many of the properties associated with tumor progression
`mediated by this putative growth factor receptor. To(cid:173)
`gether with the demonstrated activity of muMAb 4D5 in
`nude mouse models, these results support the clinical
`application of muMAb 4D5 for therapy of human cancers
`characterized by the overexpression of pl85HER2.
`KEY WORDS: HER2; neu; TNF-a; monoclonal antibody ther(cid:173)
`apy.
`
`BACKGROUND: THE HER2 PROTOONCOGENE
`AND HUMAN CANCER
`
`Cellular protooncogenes encode proteins that are
`thought to regulate normal cellular proliferation and
`differentiation. Alterations in their structure or am(cid:173)
`plification of their expression lead to abnormal
`cellular growth and have been associated with car(cid:173)
`cinogenesis {1-4). Protooncogenes were first iden(cid:173)
`tified by either of two approaches. First, molecular
`characterization of the genomes of transforming
`retroviruses showed that the genes responsible for
`the transforming ability of the virus in many cases
`were altered versions of genes found in the genomes
`of normal cells. The normal version is the protoon(cid:173)
`cogene, which is altered by mutation to give rise to
`the oncogene. An example of such a gene pair is
`represented by the EGF receptor and the v-erbB
`gene product. The virally encoded v-erbB gene
`product has suffered truncation and other alter-
`
`117
`
`0271 -9142/91 /0500-0117$06.50/0 <!:> 1991 Plenum Publishing Corporation
`
`PFIZER EX. 1048
`Page 3
`
`

`

`118
`
`SHEPARD ET AL.
`
`ations that render it constitutively active and endow
`it with the ability to induce cellular transformation
`(5) .
`The second method for detecting cellular trans(cid:173)
`forming genes that behave in a dominant fashion
`involves transfection of cellular DNA from tumor
`cells of various species into nontransformed target
`cells of a heterologous species. Most often this was
`done by transfection of human , avian , or rat DNAs
`into the murine NIH 3T3 cell line (1-5). Following
`several cycles of genomic DNA isolation and re(cid:173)
`transfection, the human or other species DNA was
`molecularly cloned from the murine background
`and subsequently characterized. In some cases, the
`same genes were isolated following transfection and
`cloning as those identified by the direct character(cid:173)
`ization of transforming viruses. In other cases,
`novel oncogenes were identified. An example of a
`novel oncogene identified by this transfection assay
`is the neu oncogene. It was discovered by Weinberg
`and colleagues in a transfection experiment in
`which the initial DNA was derived from a carcino(cid:173)
`gen-induced rat neuroblastoma (6,7). Characteriza(cid:173)
`tion of the neu oncogene revealed that it had the
`structure of a growth factor receptor tyrosine ki(cid:173)
`nase , had homology to the EGF receptor, and
`differed from its normal counterpart, the neu pro(cid:173)
`tooncogene, by an activating mutation in its trans(cid:173)
`membrane domain (8).
`The association of the HER2 protooncogene with
`cancer was established by yet a third approach, that
`is , its association with human breast cancer. The
`HER2 protooncogene was first discovered in eDNA
`libraries by virtue of its homology with the EGF
`receptor, with which it shares structural similarities
`throughout (5). When radioactive probes derived
`from the eDNA sequence encoding p185HER2 were
`used to screen DNA samples derived from breast
`cancer patients, amplification of the HER2 protoon(cid:173)
`cogene was observed in about 30% of patient sam(cid:173)
`ples (9). Further studies have confirmed this origi(cid:173)
`nal observation and extended it to suggest an
`important correlation between HER2 protoonco(cid:173)
`gene amplification and/or overexpression and wors(cid:173)
`ened prognosis in ovarian cancer and non-small cell
`lung cancer (10-14).
`The association of HER2 amplification/overex(cid:173)
`pression with aggressive malignancy , as described
`above, implies that it may have an important role in
`progression of human cancer; however, many tu(cid:173)
`mor-related cell surface antigens have been de(cid:173)
`scribed in the past, few of which appear to have a
`
`direct role in the genesis or progression of disease
`(15 ,16). The data which support a role of HER2
`overexpression in the basic mechanisms of human
`cancer are summarized below.
`Amplified expression of p185HER2 can lead to
`cellular transformation as assessed by morphologi(cid:173)
`cal alterations and growth of p185HER2-overex(cid:173)
`pressing cells in soft agar and in nude mice (1 7, 18).
`In addition , NIH 3T3 fibroblasts overexpressing
`pl85HER2 have an increased resistance to cytotox(cid:173)
`icity mediated by activated macrophages or recom(cid:173)
`binant human TNF-a (19) , the cytokine that ap(cid:173)
`pears to be mainly responsible for macrophage(cid:173)
`mediated tumor cell cytotoxicity (20). This
`observation extends also to breast tumor cells,
`which overexpress pl85HER2 (19) , and suggests that
`high levels of pl85HER2 expression may be related
`to tumor cell resistance to at least one component of
`the host' s antitumor surveillance armamentarium,
`the activated macrophage. This work has been
`reviewed previously (21) , and similar data have
`recently been reported for ovarian tumor cell lines
`which overexpress pl85HER2 (22). Further support
`for a role of pl85HER2 or the related neu oncogene(cid:173)
`encoded tyrosine kinase in tumorigenesis comes
`from work with transgenic mice that have been
`manipulated to overexpress one or the other of
`these two related genes. Transgenic mice express(cid:173)
`ing the activated form of the rat neu protooncogene,
`under the control of a steroid inducible promoter,
`uniformly develop mammary carcinoma (23). In
`another transgenic mouse model the HER2 pro(cid:173)
`tooncogene product, "activated" by point mutation
`analogous to the rat neu oncogene product, or an
`unaltered form of the HER2 protooncogene, has
`been expressed in mice (24). The main malignancies
`induced in this model were either lung adenocarci(cid:173)
`noma or lymphoma but not mammary carcinoma.
`While it is not known why the different transgenic
`mouse models give such distinct results , the latter
`model may be of particular significance given the
`.
`.
`b
`185HER2
`recent report o an assoctahon etween p
`f
`overexpression and poor prognosis in nonsmall cell
`lung cancer (14). These differing results suggest
`some difference in the activity of activated neu and
`JIER2-encoded tyrosine kianses, although effects
`due to mouse strain differences cannot be excluded.
`The structural similarities between pl85HER2 and
`.
`HER2
`the EGF receptor suggest that function of p185
`may be regulated similarly to the EGF receptor. In
`particular, one expects that the tyrosine kinase
`activity associated with the cytoplasmic domain of
`
`Journal of Clinical Immunology, Vol. 11 , No . 3, 1991
`
`PFIZER EX. 1048
`Page 4
`
`

`

`TNF R
`
`TNF
`
`HER2
`
`HER2
`Ligand
`II
`
`....
`
`119
`
`Tumor
`Cytotoxic
`Response
`
`, .... ~ .... I
`
`Suppression of
`Cytotoxic Response
`
`....
`
`....
`
`Cellular
`..,. Proliferation
`
`Fig. 1. Suppression of the TNF cytotoxic response by activation ofpl85HER2
`• The schematic shows both the TNF
`cytotoxic pathway (top) and the pl85HER2-stimulated cell proliferation/transformation pathway (bottom). Signaling
`from the TNF receptor following interaction with TNF has not been characterized. Binding of ligand to pl85HER2
`is shown to activate the receptor-associated tyrosine kinase activity, resulting in stimulation of cellular
`proliferation and suppression of the tumor cell cytotoxic response to TNF.
`
`receptor would be ligand activated. This pro(cid:173)
`receives support from recent work describing
`ligand for p185HER2 (25). These data lead to a
`(Fig. 1) wherein antagonists that downregu(cid:173)
`the function of p185HER2 should have the effect
`inhibiting growth of tumor cells dependent upon
`HER2 function and of increasing the sensitivity
`such tumor cells to TNF-a. By analogy with
`'"'"'''"''"' work done with two related tyrosine ki(cid:173)
`' the EGF receptor (26) and the activated neu
`OOilcogerte product (27), we hypothesized that
`antibodies targeted to the extracellular
`- ..... ,,u of p185HER2 may have the desired proper-
`
`family of monoclonal antibodies focused
`the extracellular domain of p185HER2 were
`(28). To do this, an NIH 3T3 fibroblast cell
`that overexpresses p185HER2 [NIH 3T3/HER2-
`(18)] was used to immunize BALB/c mice.
`were subsequently boosted with NIH
`3-400 and, finally, with a preparation
`
`of Clinical Immunology, Vol. II , No.3, 199/
`
`enriched for p185HER2 by wheat germ agglutinin
`chromatography of membrane extracts of this cell
`line. Following splenocyte fusion with a mouse
`myeloma partner, the hybridomas were cultured in
`96-well microtiter plates. Hybridomas positive for
`an ti-p 185HER2 activity, but with little or no anti(cid:173)
`EGFR activity, were detected by ELISA (Fig. 2). A
`critical property of an anti-p185HER2 monoclonal
`antibody with potential for therapy would be its
`lack of cross-reactivity with the closely related
`EGF receptor, which is expressed at elevated levels
`in multiple tissues (29). To select further monoclo(cid:173)
`nal antibodies with this characteristic, a number of
`assays were performed, including immunoprecipita(cid:173)
`tion assays utilizing in vivo labeled EGF receptor
`and p185HER2 (Fig. 3A) and FACS analysis of
`antibody binding to tumor cells overexpressing ei(cid:173)
`ther p185HER2 or the EGFR (Fig. 3B). The screen(cid:173)
`ing results are summarized in Table I. Based upon
`these results, nine of the p185HER2 monoclonal
`antibodies were chosen for further characterization,
`including a cell growth inhibition assay utilizing the
`SK-BR3 human breast adenocarcinoma cell line,
`which greatly overexpress p185HER2. The monoclo-
`
`PFIZER EX. 1048
`Page 5
`
`

`

`120
`
`SHEPARD ET AL.
`
`1.8
`
`1.6
`
`1.4
`
`1.2
`
`1.0
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`0 0
`
`E
`r:::
`N en
`
`'<T
`
`0
`0
`
`2C4
`
`7C2
`
`7F3
`
`2H11
`
`3H4
`
`3E8
`
`405
`
`598
`
`6E9
`
`703
`
`596 40 .1 H1
`
`MAb (1 ug/ml)
`Fig. 2. ELISA screening of anti-pi85HER2 monoclonal antibodies. Results shown measure the relati ve
`reactivities of the purified anti-pi85HER2 monoclonal antibodies (added to I IJ.g/ml) with membrane extracts
`enriched in EGF receptor (open bars; from A431 squamous carcinoma cells) or enriched in pi85HER2 (filled
`bars ; from NIH 3T3/HER2-3-400).
`
`nal antibody, muMAb 405 , was clearly the most
`effective of the group in this assay (Table II).
`The initial results characterizing the growth in(cid:173)
`hibitory activity of these monoclonal antibodies
`were extended by comparing them for activity
`against a battery of human breast and ovarian tumor
`cell lines that expressed varying levels ofp185HERz.
`These results reveal that the monoclonal antibodies
`can be growth inhibitory, they may have no affect
`on cell proliferation, or they may stimulate the
`proliferation of breast tumor cells. Growth inhibi(cid:173)
`tion appears to depend upon overexpression (Table
`III). This property, in particular, is shared by the
`monoclonal antibodies 405 and 3H4. These mono(cid:173)
`clonal antibodies may exert their effects on cell
`growth by similar mechanisms since they compete
`for binding to the receptor (Tables I and Ill) (28)
`and, therefore , may recognize the same or overlap(cid:173)
`ping epitopes. The other monoclonal antibodies
`vary in their ability to inhibit proliferation, but 7C2
`and 6E9 are consistently less active in this respect.
`
`The potent growth inhibitory activity of 2C4 for
`MOA-MB-175 breast tumor cells is not understood
`at present but may represent cross-reactivity with
`another receptor expressed on these cells. Simi(cid:173)
`larly , the properties that distinguish 7C2 from the
`other antibodies with regard to its ability to stimu(cid:173)
`late the proliferation of several of the tumor cell
`lines shown in Table III has not been determined.
`The 6E9 monoclonal antibody has been shown to
`bind to the extracellular domain of pl85HERZ, al(cid:173)
`though only to a subset of receptors on the surface
`of SK-BR-3 tumor cells (30). The functional signif(cid:173)
`icance of this subset of receptors is unclear. In
`addition to its activity on breast tumor cells, which
`overexpress pl85HERz, muMAb 405 is also clearly
`the most active of the monoclonal antibodies with
`respect to its ability to inhibit growth of SKOV-3, a
`human ovarian adenocarcinoma cell line that over(cid:173)
`expresses pl85HERz (Table III). Experiments are
`currently planned to try to understand in more
`detail how these monoclonal antibodies may exert
`
`Journal of Clinical Immunology, Vol. II , No. 3, J99l
`
`PFIZER EX. 1048
`Page 6
`
`

`

`MONOCLONAL ANTIBODY THERAPY OF HUMAN CANCER
`
`121
`
`A.
`
`B.
`
`CVN
`
`....
`I
`0
`It)
`ci
`0
`~ ~
`2
`1
`
`c:o
`0
`
`3
`
`HER1
`....
`::x;
`It)
`ci
`0
`~ ~
`4
`5
`
`c:o
`0
`
`6
`
`HER2
`....
`::x;
`It)
`ci
`0
`~ ~
`7
`8
`
`c:o
`0
`
`9
`
`I
`
`~
`)\
`
`: \
`
`J.
`
`\
`
`\
`
`185
`-
`-170
`
`-0
`~
`E
`::I z
`-~ -a
`Q) a:
`
`10 2
`104
`103
`10
`1
`0.1
`Relative Fluorescence Intensity
`Fig. 3. MuMAb 4D5 does not cross-react with the EGFR. (A) Immunoprecipitation of metabolically labelelled NIH 3T3 cells
`transfected by control plasmid (CVN), by a plasmid encoding the EGFR (HER!) or a plasmid encoding pi85HER2 (HER2).
`MuMAb 40.1.HI is directed against hepatitis B surface antigen (lanes I, 4, 7); muMAb 4D5 is directed against pi85HER2
`(lanes 2, 5, 8) ; muMAb 108 is directed against the EGFR (lanes 3, 6, 9). (B) Fluorescence-activated cell sorter histograms
`of muMAb 40.1 .H I (solid lines) or muMAb 4D5 (dotted lines) reacted with SK-BR-3 tumor cells (approx. 2 x 106 receptors
`per cell; upper panel) or the same antibodies reacted with the A431 squamous carcinoma cell line (approx . 2 x 106 EGFR
`per cell; lower panel).
`
`-. ....... ~. effects on tumor cell proliferation. The in
`results summarized in Table III clearly show
`when the monoclonal antibodies are compared
`efficacy, as measured by their abilities to inhibit
`1.,..,,u,irh of breast and ovarian tumor cells overex(cid:173)
`'lll•:ssml! p185HER2, muMAb 4D5 is usually the most
`and is therefore a good candidate for further
`· n in other models that may be pre-
`of its efficacy in human clinical trials. Inter- ·
`, the most dramatic activity of the antibody
`seen in cell lines that overexpress greater than
`the level observed in MCF-7 breast tumor
`lines [a low expressor control cell line; Table III
`19)]. Patients who overexpress greater than five(cid:173)
`the normal level of pl85HER2 have been shown
`have a very poor prognosis (10). These results
`aid in choosing patients who are most likely to
`""'"'"""' ... in clinical trials.
`The model depicted in Fig. 1 predicts that down(cid:173)
`-.. ~ ..... ,·,., of p185HER2 by a monoclonal antibody or
`
`other reagent should result in decreased cellular
`proliferation, as shown in Table III, but also in(cid:173)
`creased sensitivity to TNF-a. Results of experi(cid:173)
`ments in which tumor cells overexpressing
`pl85HER2 were treated with muMAb 4D5 or a
`control monoclonal antibody, alone and in combi(cid:173)
`nation with TNF-a, suggest the validity of this
`model (Fig. 4) (31). MuMAb 4D5 treatment of
`breast tumor cells overexpressing pl85HER2 re(cid:173)
`sulted in enhanced sensitivity of these cells to
`TNF-a. The growth and the TNF-a sensitivity of
`normal cells or tumor cells that do not overexpress
`the receptor were unaltered.
`In addition to the relationship between TNF-a
`resistance and pl85HER2 overexpression, a possible
`relationship between protoncogene expression and
`resistance to the chemotherapeutic drug cisplatin
`has been investigated. A correlation between HER2
`protooncogene overexpression and resistance to
`chemotherapeutic drugs rests on the grounds that
`
`of Clinical Immunology, Vol. 11, No.3 , /99/
`
`PFIZER EX. 1048
`Page 7
`
`

`

`122
`
`MAb
`
`4D5
`2C4
`2Hll
`3E8
`3H4
`SB8
`6E9
`7C2
`703
`7F3
`
`Isotype
`
`lgGl ,k
`lgGl,k
`lgG2a,k
`lgG2a,k
`IgGI ,k
`IgGI ,k
`lgG1 ,k
`IgG1 ,k
`IgGI ,k
`lgGI ,k
`
`SHEPARD ET AL.
`
`Table I. Summary Table of Monoclonal Antibodies Described
`
`ELISA a
`
`RIPb
`
`EGFR
`
`pl85HER2
`
`EGFR
`
`pl85HER2
`
`Epitopec
`
`FACSd
`
`++
`+++
`+
`+
`+
`+
`++
`++
`++
`+++
`
`++
`++
`++
`+++
`+
`++
`+
`++
`++
`+++
`
`I(p/c)
`F(p/c)
`H(p/c)
`H(p/c)
`I(p)
`nd(p)
`nd(p)
`G(p)
`F(p/c)
`G/F(p/c)
`
`++ +
`++ +
`++
`++ +
`+
`+
`
`+++
`++ +
`++ +
`
`a summary of OD 492 nm : (-) < 0. 1; ( +) 0.1 1-0.50; ( + +) 0.51-1.0 ; ( + + +) > 1.0.
`bSummary of autoradiography from immunoprecipitations: (-)bands equal to negative control; (+)weak bands but darker than negative
`control; ( + +) moderately exposed bands; ( + + +) strongly exposed bands .
`<Letters were assigned to represent individual epitopes A through I (nd, not done). MAbs were considered to share an epitope if each
`blocked binding of the other by 50% or greater in comparison to an irrelevant MAb control. The epitope composition recognized by
`immunoprecipitations with each MAb from tunicamycin-treated cells is shown. The letters p, c, or p/c in parentheses indicate that the
`monoclonal antibody binds only to the polypeptide (p), the carbohydrate (c), or both (p/c) moieties in the extracellular domain of
`pJ85HER2.
`dFluorescence staining of SK-BR-3 cells by the anti-p185HER2 monoclonal antibodies: (-) MAbs equal to the negative control MAbs; ( +)
`1- to 9-fold higher than the negative controls;(++) 10- to 99-fold higher than the negative controls;(+++) >100-fold higher than the
`negative controls .
`
`patients exhibiting overexpression appear to have a
`worsened prognosis, especially in ovarian cancer
`(10, 13). In addition, recent work with the EGF
`receptor (32) has indicated that when the anti(cid:173)
`EGFR monoclonal antibody 108.4 was added to(cid:173)
`gether with cisplatin, the antitumor effect of the
`antibody was greatly enhanced. Because the 108.4
`monoclonal antibody and muMAb 4D5 appear to
`share the ability to inhibit soft agar growth of tumor
`
`Table U. Inhibition of SK-BR-3 Proliferation by Anti-p185HER2
`Monoclonal Antibodiesa
`
`Monoclonal
`antibody
`
`405
`7C2
`2C4
`703
`3E8
`6E9
`7F3
`3H4
`2HII
`40. 1 HJ C
`4F4
`
`Relative cell
`proliferationb
`
`44.2 ± 4.4
`79.3 ± 2.2
`79.5 ± 4.4
`83.8 ± 5.9
`66.2 ± 2.4
`98.9 ± 3.6
`62.1 ± 1.4
`66.5 ± 3.9
`92.9 ± 4.8
`105.8 ± 3.8
`94.7 ± 2.8
`
`asK-BR-3 breast tumor cells were plated at a density of 4 X 104
`cells per well into 96-well microdilution plates, allowed to
`adhere, and then treated with monoclonal antibody (10 tJ.g/ml) .
`bRelative cell proliferation was determined by crystal violet
`staining of the monolayers after 72hr. Values are expressed as a
`percentage of results with untreated control cultures (100%).
`<control monoclonal antibodies 40.1Hl and 4F4 are directed
`against hepatitis B surface antigen and human interferon--y,
`respectively (27) .
`
`cells overexpressing their respective receptors, it
`seemed possible that such an interaction may also
`occur in the HER2 protooncogene system. The in
`vitro results (Fig. 5) show that treatment of SK(cid:173)
`BR-3 breast tumor cells with muMAb 4D5 enhances
`their sensitivity to cisplatin.
`
`IN VIVO PRECLINICAL EFFICACY
`
`A critical part of the rationale supporting the
`application of muMAb 4D5 to human cancer ther(cid:173)
`apy is its ability to inhibit the growth of tumor cells
`overexpressing p185HER2 in vivo. A human tumor
`xenograft model was used to test this property of
`muMAb 4D5 and to compare its activities with
`those of the other monoclonal antibodies in a rele(cid:173)
`vant model of human disease. In this model, a
`human breast tumor, characterized with respect to
`HER2 protooncogene amplification and expression,
`was grafted into the sub renal capsules of nude mice.
`Therapy was initiated 1 week postimplantation. In
`order to be active in this model, the monoclonal
`antibody must be able to localize to the overex(cid:173)
`pressing tumor cells in the lesion and subsequentlY
`exert a growth regulatory effect mediated through
`p185HER2 • Growth inhibition occurs only with tu(cid:173)
`mors that overexpress the receptor. Heterotrans(cid:173)
`plants (approximately 1 mg) of Murray breast tumor
`[a high expresser of the HER2 gene product (10)]
`were implanted into the subrenal capsule of 48
`
`Journal of Clinical Immunology, Vol. 11 , No .3, 1991
`
`PFIZER EX. 1048
`Page 8
`
`

`

`123
`
`Ill. Inhibition of Human Breast and Ovarian Tumor Cell Growth by Monoclonal Antibodies Directed Against the Extracellular
`Domain of pi85HER2
`
`Relative
`pJ85HER2
`expression"
`
`I
`3
`4
`7
`17
`20
`33
`17
`
`4D5c
`
`3H4c
`
`2C4d
`
`7F3d
`
`7C2e
`
`Cell proliferation (% control)b
`
`94
`106
`61
`62
`60
`23
`42
`77
`
`101
`113
`84
`68
`68
`25
`56
`85
`
`101
`104
`24
`91
`65
`53
`66
`87
`
`97
`100
`48
`84
`73
`20
`64
`91
`
`106
`149
`87
`78
`113
`74
`92
`97
`
`6E9'
`
`110
`113
`103
`101
`113
`94
`105
`99
`
`on FACS assay using muMAb 4D5 and fluorescence-labeled goat anti·murine lgG I polyclonal antibody .
`no·,,v,._,,,.v assay with 10 fLg/ml of indicated monoclonal antibody (SE , - 10%). Other methods as described in the footnotes to Tabl~ II.
`and 3H4 define epitope "I."
`and 7F3 will partially block one another, 2C4 is assigned epitope "F," and 7F3 is assigned epitope "FIG."
`defines epitope " G" and will partially. block 7F3 binding.
`epitope determination not done.
`
`mice on day 0. Groups of eight animals
`injected intravenously with tissue culture(cid:173)
`muMAb 4D5 (36.4 mg/kg), PBS, or control
`oclonal antibody, muMAb 5B6 (directed
`gpl20; 36.4 mg/kg), as single agents in
`divided doses on days 7, 10, and 13. Four
`from each group were sacrificed on day 20,
`the remainder of the animals were sacrificed on
`34. Tumor sizes were measured using ocular
`and gravimetric techniques. A sum(cid:173)
`of the Jumor weights (mean ± SD) from
`,.. .. l,"a'" sacrificed on days 20 and 34 is shown in
`IV. On day 20, average tumor weights of
`s receiving muMAb 4D5 were significantly
`than those receiving the same dose of the
`antibody muMAb 5B6. Interactive effects
`muMAb 4D5 and cisplatin have also been
`LJ[J~;en,,.n in this model (33). These studies in athy-
`mice bearing human breast tumor xenografts
`demonstrated efficacy and suggested an en(cid:173)
`effect when muMAb 4D5 is given in combi(cid:173)
`with cisplatin.
`
`OF ACTION
`
`4D5 having receptor antagonist activity.
`, however, muMAb 4D5 treatment of
`tumor cells stimulates receptor tyrosine
`activity (Table V) (30, 34). In addition , it can
`the phosphorylation of intracellular sub(cid:173)
`by pl85HERZ (34). Consistent with its ability
`stimulate receptor activity, muMAb 4D5 treat-
`of SK-BR-3 or SK-OV-3 tumor cells results in
`
`of Clinical Immunology, Vol. 11 , No.3 , 1991
`
`a modulation of intracellular second messengers,
`including diacylglycerol. Diacylglycerol (DAG) is a
`product of phospholipase C breakdown of phos(cid:173)
`phatidylinositol-4,5-bisphosphate. It is a cofactor
`
`A. SK-BR-3
`
`B. SK-BR-3
`
`C. MDA-MB-175-VII
`
`D. MDA-MB-231
`
`E. HBL-100
`
`F. T24
`
`1.5
`
`1.0
`
`c
`0
`-~ 0.5
`.....
`~
`0 .....
`0...
`Q)
`()
`Q)
`>
`-~ 1.0
`Q) a:
`
`0.0
`1.5
`
`0.5
`
`0.0
`Fig. 4. Monoclonal antibody 4D5 sensitizes breast tumor cells to
`the cytotoxic effects of TNF-a. Filled bars, cell number at
`initiation of the assay; dark cross-hatching, untreated control;
`dark stipples , TNF-a alone; light cross-hatching, MuMAb 4D5;
`open bars , MuMAb 4D5 combined with TNF-a . (B) Lack of
`growth inhibition of SK-BR3 tumor cells by muMAb 40.1 H1
`(anti-hepatitis B antigen; light stipples) and failure of the 40.1 H
`I to enhance SK-BR-3 tumor cell sensitivity to TNF-a (broken
`cross-hatching). SK-BR-3 and MDA-MB-175-VII overexpress
`pi85HER2 (see Table III). MDA-MB-231 and HBL-100 are
`breast cell lines which do not overexpress p185HER2, and T24 is
`a nonoverexpressing human bladder carcinoma cell line . The
`assay was performed as described in Ref. 31.
`
`PFIZER EX. 1048
`Page 9
`
`

`

`124
`
`SHEPARD ET AL.
`
`-
`
`140 A
`~ 120
`:.0
`11)
`>
`c:
`Q)
`.._
`(..)
`Q)
`0...
`Q)
`>
`~
`Qi
`a:
`
`100
`
`o-1
`
`~
`
`~
`
`_.)
`
`80
`
`60
`
`40
`
`20
`
`8
`
`~
`
`10
`
`100
`
`0
`
`.01
`
`10
`1
`.1
`Cisplatin (Jlg/ml)
`
`100
`
`0
`
`0
`
`.01
`
`1
`.1
`Cisplatin (Jlg/ml)
`
`(
`Fig. 5. Treatment of SK-BR-3 breast tumor cells with muMAb 4D5 enhances sensitivity to
`cisplatin. MuMAb 4D5 (A) or muMAb 6E9 (control; B) and cisplatin were added at the indicated
`concentrations to SK-BR-3 breast tumor cells . The plate culture