`tJA
`11~ I DISPLAY
`
`JANUARY 1984
`VOLUME 81
`NUMBER 1
`
`LIBRARY
`t: E.B
`1qR4
`u~WERSIT'f OF
`l~l c;'lll lr,Tf\1\l
`
`. ! .
`
`..... .
`"-
`
`Proceedings
`
`OF THE
`
`National Acaden1y
`of Sciences
`
`OF THE UNITED STATES ()F AMERICA
`
`BIOLOGICAL SCIENCES
`
`PFIZER EX. 1604
`Page 1
`
`
`
`Proceedings
`OF THE
`National Academy
`of Sciences
`OF THE UNITED STATES OF AMERICA
`
`Officers
`of the
`Academy
`
`Editorial Board
`of the
`Proceedings
`
`FRANK PRESS, President
`JAMES D. EBERT, Vice President
`BRYCE CRAWFORD, JR., Home Secretary
`WALTER A . ROSENBLITH, Foreign Secretary
`ELKAN R. BLOUT, Treasurer
`
`ROBERT A . ALBERTY
`K . FRANK AUSTEN
`PAUL BERG
`JOHN R . BORCHERT
`JOHN E. DOWLING
`
`DANIEL E . KOSHLAND , JR., Chairman
`PETER D . LAX
`ROBERT E . MARSHAK
`PETER H . RAVEN
`JOHN RODGERS
`NEVIN S . SCRIMSHAW
`
`J. EDWIN SEEGMILLER
`ROBERT L. SINSHEIMER
`CHARLES H. TOWNES
`FRANK H. WESTHEIMER
`JAMES B. WYNGAARDEN
`
`Managing Editor: FRANCES R. ZWANZIG
`Senior Associate Editor: GARY T . COCKS
`Associate Editor: CAY BUTLER
`Associate Editor: TERRY W . PEARSON
`Associate Editor: DOROTHY P. SMITH
`Assistant Managing Editor: JOANNE D'AMICO
`
`Senior Copy Editor: LYNN A. BROWN
`Copy Editors: BARBARA A. BACON , MARYANNE CUNNINGHAM-AGNEW, RUTH E. ESCH,
`P . J. GROFF, JANET L. OVERTON , ERIC R . WASSYNG
`Administrative Assistants: DELORES BANKS, JACQUELINE BODE, BEULAH EDWARDS ,
`CYNDY MATHEWS
`Circulation : JULIA LITTLE, VIRGINIA TREADWAY
`
`Editorial correspondence: PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, 2101 Constitution Avenue, Washing(cid:173)
`ton, DC 20418.
`
`Business correspondence: Circulation Office of the PROCEEDINGS , National Academy of Sciences, 2101 Constitution Ave(cid:173)
`nue, Washington, DC 20418 .
`
`Information for Contributors : See issue Number 1, January 1984.
`
`Copyright: The National Academy of Sciences has copyrighted this journal as a collective work and does not own copy(cid:173)
`rights for individual articles. Requests for permission to reproduce parts of individual articles or for reprints of individual
`articles should be addressed to the authors . Microforms of complete volumes are available to regular subscribers only and
`may be obtained from University Microfilms , Xerox Corporation , Ann Arbor, MI 48103.
`
`Subscriptions: All correspondence concerning subscriptions should be addressed to the Circulation Office of the PROCEED(cid:173)
`INGS. Subscriptions are entered on a calendar year basis only. For 1984, subscription rates are $185 in the United States and
`$220 elsewhere. Subscribers are requested to notify the Circulation Office of the PROCEEDINGS 6 weeks in advance of any
`change of address ; also the local postmaster. The Academy is not responsible for nonreceipt of issues because of an
`improper address unless a change of address is on file. The notice of address change should list both the old and new
`addresses . Claims for replacement copies will not be honored more than 60 days after the issue date for domestic subscrib(cid:173)
`ers and not more than 90 days after the issue date for foreign subscribers.
`
`: .. '
`
`Back Issues: Volumes 76-81 , January 1979 and thereafter, are available from the Circulation Office of the PRocEEDINGS .
`The price of a single issue is $21.00 for Volumes 76-78 or $11.00 for Volumes 79-81.
`
`Second class postage paid at Washington, DC, and at additional mailing offices.
`
`PRINTED IN THE USA
`
`PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (ISSN-0027-8424) is pub(cid:173)
`lished semimonthly by THE NATIONAL ACADEMY OF SCIENCES, 2101 Constitution Avenue, Washington , DC 20418.
`
`© 1984 by THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA.
`
`POSTMASTER: Send address changes to: PROCEEDINGS, 2101 Constitution Ave., Washington, DC 20418.
`
`PFIZER EX. 1604
`Page 2
`
`
`
`Proc. Nat/ . Acad. Sci . USA
`Vol. 81 , pp . 216-219, January 1984
`Immunology
`
`Human anti-idiotype antibodies in cancer patients: Is the
`modulation of the immune response beneficial for the patient?
`
`(internal image of antigen/gastrointestinal cancer/monoclonal antibody)
`
`HILARY KoPROWSKI *, DoROTH EE HERL YN*, MICHAEL LuBECK *, ELAINE DEFREITAs*, AND HENRY F. S EARst
`
`*The Wistar Institute, 36th and Spruce Streets, Philadelphia, PA 19104; and tAmerican Oncologic Hospital , Fox Chase Cancer Center, Central and Shell mire
`Avenues, Philadelphia, PA 19111
`
`Communicated by John R . Brobeck , August31, 1983
`
`Inoculation of human subjects with mouse
`ABSTRACT
`monoclonal antibody results in the production of anti-idiotype
`antibody that reacts with the binding site of the monoclonal
`antibody. This reaction is hapten-inhibited, suggesting that an
`internal image of the antigen is produced by the anti-idiotype
`response. The anti-idiotype antibody isolated from sera of
`three patients showed significant crossreactivity. Patients who
`developed the anti-idiotype antibody improved clinically and
`had long remission from their disease. The possible presence of
`the internal image of cancer antigen on the human immuno(cid:173)
`globulin molecule may change the conditions under which the
`immune system reacts to the tumor antigen and may open new
`approaches to the control of tumor growth.
`
`The treatment of human leukemias with mouse monoclonal
`antibody (mAb) directed against antigen(s) present on leuke(cid:173)
`mia cells results in a transient destruction of the leukemia
`cells (1), observed almost immediately after administration
`of the mAb (2). Sera of some patients who received anti-Leu
`1 mAb for treatment ofT-cell leukemia inhibited binding of
`fluoresceinated mAb to T leukemia cells and were consid(cid:173)
`ered to contain an anti-idiotype (anti-Id) antibody. This was
`also observed in sera of renal transplant patients who re(cid:173)
`ceived an OKT3 anti-T-cell mouse mAb (3). It has been sug(cid:173)
`gested that binding of anti-Id to the mAb may prevent the
`antitumor activity of the mAb (1). This may be true in cases
`when direct action of mAb on cancer cells occurs (2). How(cid:173)
`ever, when no immediate antitumor effect of mAb is ob(cid:173)
`served and when induction of tumor cell destruction occurs
`only a long time after the administration of mAb (4), it may
`be that an immune response of the host to his tumor is in(cid:173)
`duced by mAb treatment.
`Ierne has postulated a network of interacting antibody
`molecules and lymphocytes in the immune system in which
`idiotypes of antibody molecules are recognized by anti-idio(cid:173)
`typic antibodies and cells (5). Within the set of anti-idiotypic
`antibodies could be expected to be those that were directly
`complementary to the parotope-i.e ., the antigen binding
`site of the idiotypic antibody. The binding of such anti-idio(cid:173)
`typic antibodies to the idiotypes has in fact been shown to be
`inhibitable by the antigen . This has led to the concept that
`such anti-idiotypic antibodies are internal images of the anti(cid:173)
`gen. It was therefore predicted that priming an animal with
`anti-Id rather than with an antigen would be possible, and
`this, in fact , has been successfully achieved (6). The first
`physiological evidence that anti-Id could mimic the action of
`an antigen was obtained with anti-Ids against insulin (7). Di(cid:173)
`rect evidence for the beneficial effect of an anti-Id antibody
`in infectious diseases has recently been found in mice primed
`with anti-Id in the absence of antigen that appeared immu(cid:173)
`nized against Trypanosoma rhodesiense after such treatment
`
`(8). When these animals were challenged with T. rhode(cid:173)
`siense, specific idiotype appeared in all animals, some of
`which had the specific idiotype even before challenge.
`In the present paper, we describe the development of anti(cid:173)
`Id in sera of patients with gastrointestinal cancer who were
`treated with mouse mAb. Furthermore , we show that bind(cid:173)
`ing between the anti-Id and mAb can be inhibited by hapten.
`This suggests that the anti-ld is the internal image of the anti(cid:173)
`gen expressed by the cancer cells. Finally, we discuss these
`findings in relation to the outcome of mAb treatment of gas(cid:173)
`trointestinal cancer.
`
`MATERIALS AND METHODS
`mAbs. Mouse mAbs 17-1A, C42032, and C41472, which
`bind to human gastrointestinal cancer cells, have been de(cid:173)
`scribed elsewhere (9, 10). mAb C42032 is specific for colo(cid:173)
`rectal carcinoma (CRC)-associated antigen(s) of Mr 180,000,
`160,000, 50,000, and 40,000. mAb C41472 (IgG2a) is specific
`for a CRC-associated antigen of Mr 50,000. The A5C3 mAb
`against hepatitis virus has been described (11) . mAb 17-1A
`(lgG2a, K light chain) and C42032 (lgG2a, K light chain) were
`purified from ascites obtained from hybridoma-bearing mice
`by affinity chromatography on a protein A-Sepharose col(cid:173)
`umn (Pharmacia) as described by Ey et a/. (12).
`Colorectal Cancer Extract. The 3 M KCl extract of SW-
`1222 cells , a cell line derived from colorectal cancer, has
`been described (13). Since this preparation binds mAb 17-
`1A, it seems likely that the material containing the antigen is
`in soluble form.
`Patients. All patients had metastatic or recurrent gastroin(cid:173)
`testinal adenocarcinoma and were injected systemically with
`one dose of a purified sterile pyrogen-free preparation of
`mAb 17-1A (9) concentrated from ascites fluid of BALB/c
`mice (14). Of 9 patients who received 192 mg or less of mAb,
`7 developed anti-mouse globulin antibodies (10). Of 20 pa(cid:173)
`tients who received 200-1,000 mg of mAb, 3 developed anti(cid:173)
`mouse globulin antibodies (10). Sera of 3 patients of the first
`group who developed anti-mouse globulin antibodies (nos.
`07, 08, and 09) (10) and of 2 patients of the second group
`(nos. 14 and 23) were either screened or processed for isola(cid:173)
`tion of anti-ld antibodies. Sera used for isolation of anti-Id
`antibodies from subjects 07, 08, and 23 were obtained at the
`time when all 3 showed the highest concentration of anti(cid:173)
`mouse globulin antibodies (see Fig. 1). Patient 08 received a
`second injection of 130 mg of mAb 20 months after the first
`injection, and serum obtained after this second injection was
`used in a screening test for the presence of anti-Id (Table 1).
`Preparation of Rabbit Anti-ldiotypic Antibodies. New Zea(cid:173)
`land White rabbits were injected subcutaneously at multiple
`sites with 300 p.g of purified mAb 17-lA emulsified in
`Freund's complete adjuvant and , 30 days later, were injected
`intramuscularly with 100 p.g of mAb 17-1A. Sera were col-
`
`The publication costs of this article were defrayed in part by page charge
`payment. This article must therefore be hereby marked "advertisement"
`in accordance with 18 U.S.C. §1734 solely to indicate this fact.
`
`Abbreviations: anti-ld, anti-idiotype; mAb , monoclonal antibody;
`CRC, colorectal carcinoma.
`
`216
`-------------------------------------------------------------------------------
`
`.....
`
`PFIZER EX. 1604
`Page 3
`
`
`
`Immunology: Koprowski et a/.
`
`Proc. Nat/. Acad. Sci. USA 81 (1984)
`
`217
`
`on day 10 of the secondary response. Antisera were
`on mAb C42032 and on mAb 17-1A immunoadsor-
`as described below .
`Jso11ation of Anti-Idiotypic Antibodies from Human Sera.
`were prepared by coupling purified
`(30 mg each) to 2 ml of Affi-Gel 10 (Bio-Rad). Sera
`obtained from patients injected with one dose of mAb
`(10, 14). Serum samples previously shown by radio-
`11111~'"'a~"UJ to contain anti-murine lgG antibody were se(cid:173)
`adsorbed on mAb C42032 and 17-1A immunoad-
`to remove anti-isotypic and anti-idiotypic antibod(cid:173)
`ly. Adsorbed antibodies were eluted with 0.1
`buffer (pH 2.8), immediately neutralized with
`·-~••h<>rP buffer, and dialyzed against phosphate-buffered
`; the protein was quantitated by absorption at 280 nm
`= 14).
`IOJ•mtJetiitio,n Assay for Detection of Anti-Id. Polystyrene
`in, 6.35 mm) (Precision Plastic Ball, Chicago;
`) were washed three times with 95% ethanol. The
`beads were incubated overnight at 4°C with gentle
`with a dilution of either rabbit or human anti-ld anti-
`in 0.02 M sodium tetraborate (pH 8.2), then washed
`times with phosphate-buffered saline and further incu(cid:173)
`for at least 3 hr at room temperature with phosphate-
`saline/2% bovine serum albumin/ 0.04% NaN 3. The
`were then exposed to 1251-labeled mAb 17-1A as the
`idiotype and incubated with the potential source of
`anti-ld-i.e., human sera diluted to 25% concentra(cid:173)
`in Ca2+ -, Mg2+ -free phosphate-buffered saline/ 2% bo(cid:173)
`serum albumin/ 0.04% NaN 3. After an additional over(cid:173)
`incubation, the beads were washed and the radioactiv-
`was measured in a gamma counter.
`1~tlimulatiion of Peripheral Blood Mononuclear Cells. Peri ph(cid:173)
`blood mononuclear cells were stimulated with F(ab'h
`at 10 ng/ ml of mAb 17-1A in vitro as described
`to activate antigen-specific T cells. During the following
`, aliquots of cells were separated into T- and B-cell
`by rosetting with sheep erythrocytes treated
`2-aminoethylisouronium bromide (16). Both popula(cid:173)
`were stained with F(ab'h fragments of mAb 17-1A or
`mAb and goat anti-mouse fluorescein-conju-
`lg and subsequently analyzed in a cytofluorograph. In
`, peripheral blood mononuclear cells from the same
`were stimulated with F(ab'h fragments of mAb 17-
`or anti-influenza mAb for 9 days in a modified Mishell(cid:173)
`culture for specific human lg production (15). Super(cid:173)
`from these cultures were assayed in a solid-phase
`immunoabsorbent assay for specific human
`(KPL Laboratories, Gaithersburg, MD).
`
`RESULTS
`screen serum samples for the presence ofanti-ld, a com(cid:173)
`assay was performed using rabbit anti-Id and four
`sera previously incubated with 1251-labeled mAb 17-
`The results (Table 1) indicate that the three sera (nos. 23,
`and 14) obtained after one injection of mAb showed in hi(cid:173)
`of binding of mAb 17-1A that was higher than that of
`injection samples. Binding inhibition values ob(cid:173)
`for post-mAb injection serum of patient 14 were low
`compared with those of the other two sera but higher than
`the pre-mAb exposure serum of the same subject. Inhibi(cid:173)
`values for serum obtained from patient 08, 7 days after
`r~>r·pn' '"rl a second injection of mAb were already high .
`crude rapid screening assay with polyclonal rabbit anti(cid:173)
`was followed by isolation and purification of indi(cid:173)
`anti-lds by the immunoadsorbent technique described
`and Methods. Anti-Ids isolated from serum of
`23, who received 750 mg of mAb , and from sera of
`08 and 07, who received 133 and 125 mg of mAb,
`(obtained after the first mAb injection at times
`
`Inhibition of binding of 1251-labeled mAb 17-1A to rabbit
`Table 1.
`anti-1d 17-1A by patients ' sera
`
`Serum
`no.
`
`Days
`after mAb
`injection
`
`23
`
`09
`
`14
`
`08
`
`0*
`108
`184
`0*
`28
`0
`49
`0*
`7t
`
`% inhibition of
`binding of 1251-labeled
`17-1A to rabbit anti-
`Id 17-1A
`
`10%
`
`25%
`
`50%
`
`0
`25
`17
`NT
`NT
`0
`10
`0
`31
`
`21
`96
`NT
`20
`46
`NT
`NT
`21
`64
`
`2
`73
`87
`NT
`NT
`0
`17
`NT
`NT
`
`Rabbit anti-1d 17-1A (7.5 ~-tglml) was coupled to polystyrene
`beads. 1251-labeled 17-1A (12 ,500 cpm) was first incubated on ice
`with 10%, 25%, or 50% serum for 1 hr and then added to the coupled
`beads. Percentage of inhibition of binding of 1251-labeled 17-1A to
`rabbit anti-1d 17-1A was calculated by the following formula: %
`[(100 x cpm bound in presence of ser(cid:173)
`binding inhibition = 100 -
`um)/ cpm bound without serum] . NT, not tested .
`*Sample obtained immediately before mAb injection.
`tserum obtained 7 days after second administration of mAb.
`
`indicated in Fig. 1), were shown to be human immunoglob(cid:173)
`ulins by binding to 1251-labeled anti-human F(ab'h frag(cid:173)
`ments. The yield of anti-ld protein from serum samples var(cid:173)
`ied: from no. 08, 13 JLg/ml; from no. 07, 8.9 JLg/ml; from no.
`23, 43 JLg/ml. The largest amount was obtained from no. 23,
`
`' : ~
`H
`...
`7 w
`,.., f\
`
`8
`
`6
`
`Q
`)( 5
`~
`a..
`(.)
`
`4
`
`~
`
`• t
`
`: .
`)
`) .
`
`•
`
`PAT IENTS •
`• 23
`• 07
`• 08
`
`•
`
`•
`
`3 ~ 2
`J
`l
`
`.
`
`•
`
`20
`
`80
`100 200 300 400 500
`40
`60
`DAYS AFTER MAb TREATMENT
`
`FIG. 1. Presence of human anti-mouse globulin antibody in sera
`of three subjects as determined by binding to mouse mAb 17-1A in
`the presence of 1251-labeled rabbit anti-human F(ab'h immunoglob(cid:173)
`ulin (14) . Dotted arrows indicate days after treatment when serum
`samples used for isolation of anti-idiotypic antibody were taken (one
`sample from patient 23 , pooled sera from patients 07 and 08). Solid
`arrows indicate the last day of circulation of mAb 17-1A in patient
`blood.
`
`PFIZER EX. 1604
`Page 4
`
`
`
`218
`
`Immunology : Koprowski et a/.
`
`Proc. Nat/. Acad. Sci. USA 81 ( 1984)
`
`Patient
`08
`07
`23
`
`Table 2. Binding of human anti-idiotype 17-1A antibodies to monoclonal antibodies
`Binding of 1251-labeled mAb to anti-ld, cpm
`Anti-ld ,
`A5C3
`C41472
`C42032
`17-1A
`1-Lg/ ml
`166 ± 18
`308 ± 44
`1,693 ± 131
`214 ± 31
`< 5.0
`1,023 ± 91
`251 ± 29
`168 ± 26
`363 ± 21
`6
`813 ± 101
`427 ± 24
`512 ± 59
`21,339 ± 156
`18
`427 ± 23
`NT
`7,956 ± 293
`208 ± 15
`7
`cpm bound to normal human lgG (40 1-Lg/ ml) of 1251-labeled 17-1A, C42032, and A5C3 were 151 ± 25,
`139 ± 32, and 247 ± 32, respectively. Results represent mean ± SD of triplicate determinations. NT,
`not tested.
`
`which also showed the highest levels of anti-mouse globulin
`antibodies (Fig. 1).
`Binding of the isolated anti-Id to mAb 17-1A and to three
`other mAbs is shown in Table 2. The results indicate that the
`binding of anti-Ids from all three sera to mAb 17-1A was sig(cid:173)
`nificantly higher than to the three other mAbs, two of which
`(C42032 and C41472) also detect antigenic sites on CRC cells,
`though these sites are different from those recognized by
`mAb 17-1A (10). Immunoglobulin isolated from sera of all
`three subjects prior to exposure to mAb 17-1A was adjusted
`to approximately 2.5 1-tg/ ml and coupled to pol ystyrene
`beads. These preparations did not bind any of the mAbs list(cid:173)
`ed in Table 1, indicating the absence of anti-ld in pre-expo(cid:173)
`sure serum.
`These results indicate that anti-ld could be detected by the
`screening assay in four sera of patients injected with mouse
`mAb and that anti-Id could be isolated and purified from
`three sera reacting only with the idiotype injected into the
`patient.
`Crossreactivity Between Human Anti-Ids. The results of
`the competition assay (Table 3) indicate significant cross(cid:173)
`reactivity between the anti-Ids of sera of patients 07 and 23
`and slightly less but still significant crossreactivity between
`anti-ld sera of patients 08 and 23 . Similar crossreactivity was
`found between anti-Id 07 and post-mAb serum obtained from
`no. 08 (results not shown). These results suggest that the
`anti-Ids in different patients are directed against the same
`site.
`Is the Anti-Id Directed to the Framework or Antigenic Site
`of the ldiotype Determinants? To answer this question , 125I(cid:173)
`Iabeled mAb 17-1A was incubated with a 3 M KCI cell ex(cid:173)
`tract of a human colon carcinoma that binds mAb 17-1A and
`with 3 M KCI extract of melanoma cell s that does not bind
`mAb 17-1A. The antibody-antigen mixture was then added
`to no. 23 anti-Id-coated beads and binding was compared
`with binding of the radiola beled mAb alone . These experi(cid:173)
`ments were performed with non saturating amounts of iodin(cid:173)
`ated mAb to detect changes in binding with small amounts of
`competitive haptens. For control purposes, the iodinated
`mAb 17-1A was mixed with an extract from melanoma cells
`that had previously been shown to not bind mAb 17-1A. The
`CRC cell extracts in concentrations of 0.1 or 0.5 mg/ ml were
`found to inhibit the binding of anti-ld to iodinated mAb 17-
`1A by 39% and 68% , respectively , whereas the extract from
`melanoma cells in concentrations up to 0.5 mg of protein/ ml
`
`Table 3. Crossreactivity between human anti-Ids in competition
`assay for detection of anti-ld
`
`First
`Anti-ld 23
`
`Antibody
`Second
`None
`07 Pre-mAb
`Post-mAb
`08 Pre-mAb
`Post-mAb
`
`cpm bound
`4,297
`4,595
`1,231
`4,097
`2,585
`
`% inhibition
`of mAb 17-1A
`binding
`
`0
`71
`5
`40
`
`did not significantly affect mAb binding (Table 4) . Thi s hap.
`ten inhibition of the binding reaction suggests the presence
`of an " internal image " of the CRC epitope on the anti-ld mol(cid:173)
`ecule , as supported by the finding that the 3M KCI extract of
`CRC cells did not bind to the anti-Id but did bind, as expect(cid:173)
`ed, to mAb 17-1A.
`Stimulation of B Cells by mAb 17-IA. Buffy coat cells were
`obtained from patient 23, 5 months after injection of mAb,
`and from patient 08 , 20 month s after injection of mAb. The
`percentage of lymphocytes that specificall y bound 17-1A
`F(ab'h of patient 08 was 1.2 and that of patient 23 was 0.2.
`During 7 days in culture with mAb 17-1A, the percentage of
`lymphocytes of patient 23 that specifically bound 17-1A
`F(ab'h increased from 0.2 to 13 . All the 17-1A-binding cells
`were present in the B-cell population . In addition , after 9
`days , human anti-17-1A mAb IgG was detected . Incubation
`of lymphocytes from the same patient with anti-influenza
`mAb under identical conditions produced no detectable hu(cid:173)
`man Ig to either mAb 17-1A or anti-influenza mAb .
`Patients Who Responded with a Remission After mAb
`Treatment. Table 5 gives data on five patients , four of whom
`had a small to medium tumor burden of metastatic or recur(cid:173)
`rent colorectal adenocarcinoma (14) and one, no . 14, who
`had a pancreatic carcinoma with small liver metastases . The
`tumors in all five patients regressed after a single admi nistra(cid:173)
`tion of mAb 17-1A. Anti-Id was detected in the sera of four
`patients in the screening test , and the anti-Id was isolated
`and purified from sera of three other patients. Three patients
`(nos. 07, 09, and 14) are currently without evidence of tumor
`growth. Patient 08 had a local recurrence of tumor 18 months
`after mAb 17-1A treatment without any additional therapeu(cid:173)
`tic measures. It is too soon to evaluate the long-term thera(cid:173)
`peutic results in patient 23 .
`
`DISCUSSION
`mAb 17-1A has been used successfully to diagnose gastroin(cid:173)
`testi nal cancers in humans by binding to tumor tissue in situ
`
`Table 4. Hapten inhibition of binding of human anti- ld to
`mAb 17-1A
`
`Hapten
`
`Source
`CRC SW-1222
`
`MEL SK-MEL-21
`
`% inhibition of binding
`of anti-Id 23 to
`Cone.,
`1251-labeled 17-1A
`mg/ ml
`68.5
`0.5
`39.4
`0.1
`6.6
`0.5
`0.1
`0
`CRC = colorectal cancer tissue culture line. MEL = melanoma
`tissue culture line. 1251-Labeled mAb 17-1A was mixed with 3 M KCl
`cell extracts to final extract concentrations of 0.5 and 0.10 mg/ml
`and 62,500 cpm of iodinated mAb per mi. After a 1-hr incubation on
`ice, 200 I-Ll of the mixture or 200 I-Ll of iodinated mAb alone was
`added per polystyrene bead coupled with anti-ld 17-1A (15 1-Lgf ml;
`patient 23) and this mixture was incubated overnight at 4°C. Results
`were calculated by the followi ng formula : % inhibition of binding "'
`[(100 x cpm bound in the presence of extract)/ cpm bound
`100 -
`without extract].
`
`PFIZER EX. 1604
`Page 5
`
`
`
`Immunology: Koprowski et a/.
`
`Proc. Nat/. Acad. Sci. USA 81 (1984)
`
`219
`
`Anti-Id response in patients who had a disease
`after one injection of mAb 17-1A
`
`17-1A mAb treatment
`
`Anti-ld response
`
`Date
`
`Amount , mg Screening*
`
`December 1981
`December 1981 t
`March 1982
`September 1982
`February 1983
`
`125
`92
`133
`433
`845
`
`ND
`+
`+
`+
`+
`
`Isolationt
`+
`+
`ND
`ND
`+
`
`by detection of the radiolabeled antibody by gamma
`(10). This does not preclude the possibility that
`antigen defined by mAb 17-1A may act as an im(cid:173)
`to elicit host immune mechanisms for the control
`growth. As several patients benefited from treat-
`with mAb 17-1A, a functional involvement of immune
`"""''~~<""" specifically connected with the 17-1A-defined
`sted.
`~u"v''"'" mAb may directly destroy tumor cells by activa(cid:173)
`of complement or by interaction with macrophages or
`cells (17, 18), it is possible that other mechanisms may
`·
`in the present study because of the need for
`large amounts of mAb to interact with all antigenic sites
`cells of a solid tumor mass and because of the long
`period to measure the antitumor effect. In fact , the
`in size and eventual disappearance of lung metasta-
`which could be measured by radiography , occurred over
`months in one patient (10). Thus, although one may ex(cid:173)
`to observe an antitumor effect a long time after the ad-
`.
`of mAb , binding of mAb to metastatic tumor
`samples was detected only up to 1 wk after injection
`(results not shown).
`mechanism might involve a network of in(cid:173)
`anti-idiotypic T and B cells directed against the
`Among them may be those that react with the mAb
`site. Such anti-Id would be considered internal im(cid:173)
`the tumor antigen. The hapten inhibition binding de(cid:173)
`in the present paper suggests that an internal image
`the 17-1A CRC antigen was produced by the anti-Id re(cid:173)
`. Although it is still difficult to correlate the presence
`with the outcome of the immunotherapy, it should
`•·mtenlllOlled that three patients who produced anti-Id have
`tumor regression and currently have no detectable dis(cid:173)
`One had remission lasting 18 months, and it is still too
`to evaluate the outcome in the fifth patient. Sera of
`other patients who are apparently in remission were
`for the presence of anti-ld but they did not seem to
`anti-mouse globulin antibodies. The assay for the
`does not reflect the immune status of the individual
`B cells of one of the patients are stimulated in vitro by
`17-1A similarly to patients who had had circulating
`globulin antibodies (results not shown). In such
`repeated injection of mouse mAb could disturb the
`c--·•n•,><u' '"immune state, boosting the production of anti-ld
`may or may not be reactive with other determinants on
`immunoglobulin. Because the test for detection of
`••uumclr antibody in cancer patients is rather insensitive,
`determination of an anti-Id response , which detects small
`(<5.0 J.Lg) of the antibody, may be the only indica(cid:173)
`of functional changes in the network system.
`As all human anti-Ids showed crossreactivity, it seems
`that the inoculation of mAb 17-1A in other subjects
`induce production of crossreacting anti-Ids and possi-
`
`bly identical internal images of the cancer antigen. The pres(cid:173)
`ence of an internal image of an antigen on a human immuno(cid:173)
`globulin molecule, in contrast to the presence of the antigen
`on a tumor cell, may change the conditions under which the
`immune system reacts to the tumor antigen. Modulation of
`the immune response of the subject as a result of such anti(cid:173)
`gen presentation may explain a successful outcome of immu(cid:173)
`notherapy.
`The most direct evidence for a beneficial effect of anti-Id
`would be the administration of anti-Id produced in a patient
`to a nonresponsive cancer patient and the triggering of an
`antitumor response , which could be measured in vivo and in
`vitro. This may not be feasible because of the paucity of hu(cid:173)
`man anti-Id. One must turn , therefore, to the production of
`xenogeneic anti-Id, which can be obtained in animals with(cid:173)
`out difficulty. The slight advantage of this approach lies in
`the possibility of using a variety of mAbs to produce anti-Id
`antibodies and, if internal images are produced , they may
`"mimic" a large number of heterogeneous antigens of the
`tumor cells.
`
`We wish to express our thanks to Dr. Fritz Melchers for his help(cid:173)
`ful critical evaluation of the data and to Dr. Louis Staudt for the
`review of the paper. This research was supported in part by National
`Institutes of Health Grants CA-33491, CA-25874 , CA-10815, and Al-
`19607.
`
`3.
`
`5.
`
`1. Levy , R . & Miller, R. A. (1983) Fed. Proc. Fed. Am . Soc. Exp.
`Bioi. 42, 2650-2656.
`2. Ritz, J ., Pesando, J . M., Sallan, S . E., Clavell, L. A., Notis(cid:173)
`McConarty , J. , Rosenthal , P. & Schlossman, S. F. (1981)
`Blood 58, 141-152.
`Jaffers , G. J. , Colvin , R. B. , Cosimi, A. B., Giorgi , J . V.,
`Goldstein , G., Fuller, T . C. , Kurnick , J. T ., Lillehie, C. &
`Russell , P. S. (1983) Transplant . Proc. 15, 646-648.
`4. Miller, R. A., Maloney , D . G., Warnke, R. & Levy , R. (1982)
`N. Engl. J. Med. 306, 517-522.
`Jerne , N. K . (1980) in ldiotypes-What They Said At The
`Time , Discussion at Les Baux-de-Provence, April 2-3, 1976,
`ed. Schnurr, l. (Basel Institute of Immunology, Basel, Swit(cid:173)
`zerland) , pp. 1-8.
`6. Eichmann , K. & Rajewsky , K. (1975) Eur. J. lmmunol. 5, 661-
`666.
`7. Sege , K . & Peterson , P. A. (1978) Proc. Nat/. Acad. Sci. USA
`75, 2443-2447 .
`.
`8. Sacks , D. L. , Esser, K. M. & Sher, A. (1982) J. Exp. Med.
`155, 1108-1119.
`9. Herlyn, M. , Steplewski, Z. , Herlyn , D. & Koprowski, H .
`(1979) Proc . Nat!. Acad. Sci. USA 16, 1438-1442.
`10. Koprowski , H ., in Monoclonal Antibodies and Cancer, Pro(cid:173)
`ceedings of the Fourth Armand Hammer Cancer Symposium,
`eds . Boss, B. D., Langman , R. E. , Trowbridge , I. S. & Dul(cid:173)
`becco, R. (Academic, New York), in press .
`11. Wands , J . R., Carlson, R. I. , Schoemaker, H., lsselbacher,
`K. J. & Zurawski , V. R. (1981) Proc. Nat/. Acad. Sci. USA 78,
`1214-1218.
`12. Ey, P. L., Prowse , S. J. & Jenkin, C. R. (1978) Immunoch em(cid:173)
`istry 15, 429-436.
`13. Herlyn , M. , Sears, H . F. , Steplewski, Z. & Koprowski , H .
`(1982) J. Clin . lmmuno/. 2, 135-140.
`14. Sears , H . F., Mattis , J., Herlyn , D., Hayry, P., Atkinson, B.,
`Ernst, C., Steplewski , Z. & Koprowski , H. (1982) Lancet i,
`762-765.
`15. DeFreitas, E., Vella, S., Linnenbach, A. , Zmijewski, C. & Ko(cid:173)
`prowski, H. (1982) Proc. Nat/. Acad. Sci. USA 19, 6646-6650.
`16. Pellogrino , M., Ferrone, S., Dierich , M. & Reisfeld , R. (1975)
`Clin. lmmunol. lmmun opathol. 3, 324-333.
`17. Herlyn, D. & Koprowski , H. (1982) Proc. Nat/. Acad. Sci.
`USA 19, 4761-4765.
`18. Steplewski, Z., Herlyn , D. , Maul, G. & Koprowski , H. (1983)
`Hydridoma 2, 1-5.
`
`PFIZER EX. 1604
`Page 6
`
`