`Vol. 88, pp. 2663-2667, April 1991
`Immunology
`
`Anti-Tac-H, a humanized antibody to the interleukin 2 receptor,
`prolongs primate cardiac allograft survival
`PAUL S. BROWN, JR.*, GARY L. PARENTEAU*, FREDERICK M. DIRBAS*, ROGER J. GARSIAt,
`CAROLYN K. GOLDMANt, MARIA A. BUKOWSKIt, RICHARD P. JUNGHANSt, CARY QUEEN:,
`JOHN HAKIMI§, WILLIAM R. BENJAMIN§, RICHARD E. CLARK*, AND THOMAS A. WALDMANNt¶
`*Surgery Branch, National Heart, Lung, and Blood Institute, and tMetabolism Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD
`20892; tProtein Design Labs, Inc., Mountain View, CA 94043; and §Hoffmann-La Roche Inc., Nutley, NJ 07110
`Contributed by Thomas A. Waldmann, December 27, 1990
`
`High-affinity interleukin 2 receptors (IL-2Rs)
`ABSTRACT
`are expressed by T cells activated in response to foreign
`histocompatibility antigens but not by normal resting T cells.
`To exploit this difference in IL-2R expression, anti-Tac-M, a
`murine monoclonal antibody specific for the IL-2Ra chain, was
`used to inhibit organ allograft rejection. However, the use of
`murine anti-Tac as an immunosuppressive agent was limited by
`neutralization by human anti-murine antibodies and by weak
`recruitment of effector functions. To circumvent these diffi-
`culties, a humanized antibody to the IL-2R, anti-Tac-H, was
`prepared. This molecule is human with the exception of the
`hypervariable segments, which are retained from the mouse. In
`vivo survival of anti-Tac-H is 2.5-fold longer than simultane-
`ously administered anti-Tac-M (terminal ty2, 103 hr vs. 38 hr).
`In addition, anti-Tac-H is less immunogenic than anti-Tac-M
`when administered to cynomolgus monkeys undergoing het-
`erotopic cardiac allografting. Specifically, all monkeys treated
`with anti-Tac-M developed measurable anti-anti-Tac-M levels
`by day 15 (mean onset, 11 days). In contrast, none of the
`animals receiving anti-Tac-H produced measurable antibodies
`to this monoclonal antibody before day 33. Finally, there was
`a prolongation of graft survival in the cynomolgus heterotopic
`cardiac allograft model in animals receiving anti-Tac. In
`animals that received anti-Tac-M, the allograft survival was
`prolonged compared to that of the control group (mean sur-
`vival, 14 ± 1.98 days compared to 9.2 ± 0.48 days; P < 0.025).
`Graft survival was further prolonged by anti-Tac-H with a
`mean survival of 20.0 ± 0.55 days (compared to controls, P <
`0.001; compared to anti-Tac-M, P < 0.02). There was no
`toxicity attributable to the administration of either form of
`anti-Tac. Thus, anti-Tac-H significantly prolonged allograft
`survival in primates, without toxic side effects, and may be of
`value as an adjunct to standard immunosuppressive therapy in
`humans.
`
`The development of specific, effective, and nontoxic immu-
`nosuppressive agents remains a major goal in the prevention
`of allograft rejection in humans. Many of the effective agents
`are associated with broad toxicity. Furthermore, all of the
`polyclonal, and most monoclonal, antibodies that have been
`clinically effective recognize the majority of circulating T
`cells, thus yielding broad nonspecific immunosuppression,
`leaving the transplant recipient susceptible to opportunistic
`infections.
`An ideal immunosuppressive agent would selectively tar-
`get only those T cells destined to participate in the immune
`rejection. Specifically, a strategy to achieve more specific
`immunosuppression would be to target antigens that are
`absent on resting lymphocytes but are expressed on lympho-
`cytes responding to an allograft. We have chosen to target the
`
`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.
`
`interleukin 2 receptor a chain (IL-2Ra; p55, CD25, or Tac
`protein), as its expression marks a critical step in the acti-
`vation ofalloreactive T cells (1-3). The scientific basis for this
`approach is the observation that T cells in patients rejecting
`allografts express IL-2R identified by the anti-Tac monoclo-
`nal antibody, whereas normal resting cells and their precur-
`sors do not (1, 2). In individuals receiving organ allografts,
`the T lymphocytes of the host having appropriate T-cell
`antigen receptors that recognize the foreign histocompatibil-
`ity antigens expressed on the donor organ become activated.
`These activated cells express the inducible IL-2Ra and
`participate in the rejection of the allograft. In addition, the
`serum concentration of soluble IL-2Ra released by activated
`cells is elevated before and during allograft rejection episodes
`(4, 5). Furthermore, antibodies to the IL-2Ra inhibit the
`proliferation of T cells reacting to foreign histocompatibility
`antigens expressed on the donor organ and prevent the
`generation of cytotoxic T cells in allogeneic cocultures (6).
`In light of these studies, the use of monoclonal antibodies
`directed toward the IL-2Ra chain represents a rational ap-
`proach that may allow the development of donor-specific
`immunological unresponsiveness, since only activated cells
`express this receptor subunit. In studies reported by Kirkman
`et al. (7), the survival of allografts was prolonged in rodent
`recipients treated with anti-IL-2R monoclonal antibodies. In
`collaborative studies with that group, treatment with our
`murine IgG2a monoclonal antibody to the primate IL-2Ra
`(anti-Tac-M) resulted in a significant increase in both renal
`allograft and recipient survival in a renal allograft model
`performed in Macaca fascicularis (cynomolgus) monkeys
`(8). Due to these encouraging results, human recipients of
`cadaveric-donor renal allografts were treated with various
`anti-IL-2Ra monoclonal antibodies as adjunctive immuno-
`therapy. In studies performed by Soulillou et al. (9) and
`Cantarovich et al. (10), prophylactic use of a monoclonal
`antibody (33B3.1) directed against the IL-2R was shown to be
`highly efficient in preventing renal allograft rejection in the
`first weeks after human transplantation. Similarly, in a col-
`laborative study with Kirkman and coworkers (11), a ran-
`domized prospective trial of anti-Tac-M in human renal
`transplantation was undertaken. There was a significant
`reduction in early rejection episodes in the anti-Tac-treated
`patients and the time to first rejection was delayed. Despite
`these effects, anti-Tac-M administration did not affect actual
`or actuarial graft or patient survival.
`Therapy with murine monoclonal antibodies such as anti-
`Tac-M has been hampered by two major problems that may
`contribute to their relatively limited effectiveness in preven-
`
`Abbreviations: IL-2R, interleukin 2 receptor; CDR, complementar-
`ity-determining region.
`To whom reprint requests should be addressed at: Metabolism
`Branch/National Cancer Institute, Building 10, Room 4N115, Na-
`tional Institutes of Health, Bethesda, MD 20892.
`
`2663
`
`PFIZER and SAMSUNG v. GENENTECH
`IPR2017-01489
`PFIZER EX. 1710, Page 1
`
`
`
`2664
`
`Immunology: Brown et al.
`
`Proc. Natl. Acad. Sci. USA 88 (1991)
`
`tion of allograft rejection. First, murine monoclonal antibod-
`ies are foreign proteins that are neutralized when patients
`develop antibodies to them. Second, most murine monoclo-
`nal antibodies are less effective than human antibodies at
`recruiting human host effector functions. To circumvent
`these difficulties, we used genetic engineering to prepare a
`humanized anti-Tac monoclonal antibody (anti-Tac-H) by
`combining the complementarity-determining regions (CDRs)
`of the murine anti-Tac antibody with human IgG1 K frame-
`work and constant regions (12, 13). This hyperchimeric
`anti-Tac antibody maintains high binding affinity (3 x 109
`M-1) for Tac-expressing cells and preserves the ability to
`inhibit antigen and mixed leukocyte-induced T-cell prolifer-
`ation (12, 13). Furthermore, anti-Tac-H manifests a new
`activity of antibody-dependent cellular cytotoxicity with
`human mononuclear cells that was absent in the parental
`murine anti-Tac.
`In this report, we compare the in vivo characteristics of
`unmodified murine anti-Tac and humanized anti-Tac anti-
`body in cynomolgus monkeys. Anti-Tac-H manifested longer
`in vivo survival and reduced immunogenicity when compared
`to anti-Tac-M. Furthermore, cardiac allograft survival was
`prolonged in the group treated with anti-Tac-H as compared
`to the untreated group as well as the group treated with
`murine anti-Tac.
`
`MATERIALS AND METHODS
`Monoclonal Antibody Production. Anti-Tac, a murine-
`derived monoclonal antibody (anti-Tac-M) that binds to the
`IL-2Ra subunit, was produced and characterized as de-
`scribed (14). Humanized anti-Tac antibody (anti-Tac-H) was
`constructed as described by combining the CDRs of the
`anti-Tac antibody with human IgG1 K framework and con-
`stant regions (12). A computer model of murine anti-Tac was
`used to identify several amino acids that, while outside the
`CDRs, were likely to interact with the CDRs or antigen.
`These murine amino acids were also retained in the human-
`ized antibody. Anti-Tac-H was produced in a continuous
`perfusion bioreactor from SP2/0 cells transfected with the
`genes encoding the heavy and light chains of the hyperchi-
`meric antibody and purified on an IL-2R affinity column. The
`eluted antibodies were further purified on two serially linked
`Sephacryl S-300 columns. The final products were sterilized
`through a 0.2-,m Corning filter and were shown to contain
`<10 endotoxin units/mg. Anti-Tac-H was shown to be >99%
`pure as determined by SDS/polyacrylamide gel electropho-
`resis under reducing and nonreducing conditions.
`Cardiac Allograft Model. Cardiac allografts between out-
`bred cynomolgus (M. fascicularis) monkeys were performed
`by a modification of the operative technique of Michler et al.
`(15) as described (16). The donor aorta was anastomosed end
`to side to the infrarenal abdominal aorta of the recipient and
`the donor pulmonary artery was anastomosed to the adjacent
`vena cava. All procedures were carried out under standard
`aseptic conditions and were approved by the Animal Re-
`search Committee of the National Heart, Lung, and Blood
`Institute according to the guidelines of the National Institutes
`of Health for humane use and care of laboratory animals (17).
`Palpation and inspection were the methods used to determine
`graft function and to diagnose rejection. A swollen, warm, or
`boggy graft with decreased systolic function was indicative of
`impending rejection. Graft loss was determined by cessation
`of systolic function.
`Experimental Groups. There were three experimental
`groups of five animals each. The animals in group I under-
`went heterotopic, cardiac allograft transplantation as de-
`scribed above but did not receive any form of immunosup-
`pression. The animals in group II received 1 mg of anti-Tac-M
`per kg of body weight intravenously on the day before the
`
`operation and then every other day until the day of graft
`rejection. The animals in group III were treated identically to
`those in group II except that anti-Tac-H (1 mg/kg) was
`substituted for anti-Tac-M.
`Statistical Analysis. Analysis of variance was used to de-
`termine statistical significance. Significance was considered
`to occur at the P s 0.05 level. All values reported are reported
`as mean ± SEM.
`Immunogenicity Assay. Monkey antibodies to anti-Tac-M
`and anti-Tac-H were evaluated by an antigen-bridging
`ELISA. Dynatech Immulon II wells were coated with 50 ng
`of either anti-Tac-M or anti-Tac-H in 50 1ul of carbonate/
`bicarbonate buffer for 3 hr at 37TC. Wells were washed with
`PBS containing 0.05% Tween 20 (PBST), blocked for 1 hr at
`37TC with bovine serum albumin/PBST, and washed again
`prior to addition of test samples. Wells were incubated with
`either 50 ,uI of affinity-purified goat anti-anti-Tac standards or
`test samples for 18 hr at 4TC. The unbound proteins were
`washed from the wells and 50 ,ul of biotinylated anti-Tac-M
`(or -H) was added to each well for 2 hr at 37TC, washed, and
`then incubated with 50 1zl of alkaline phosphatase-conjugated
`streptavidin for 2 hr at 37TC. After washing off excess
`streptavidin, the quantity of biotin-labeled anti-Tac bound by
`monkey antibodies to the monoclonal antibodies was deter-
`mined colorimetrically by the addition of 50 ,ul of p-nitro-
`phenyl phosphate in diethanolamine buffer (pH 9.8) with
`incubation at 37°C for 1 hr (Sigma). The absorbance at 405 nm
`was determined with a Titertek microplate reader. Since the
`color intensity developed is directly proportional to the
`antibody concentration in the serum samples, the concentra-
`tions of antibodies to anti-Tac-H and anti-Tac-M in the
`monkey samples were calculated from a standard curve
`developed with the titered purified goat antibodies directed to
`the monoclonal antibodies.
`Serum Concentrations of Anti-Tac-H and Anti-Tac-M. The
`serum concentrations of anti-Tac-H and anti-Tac-M were
`determined by a solid-phase IL-2R inhibition assay (18).
`Plates were coated with 16 ng of the soluble form of IL-2Ra
`(sIL-2Ra) overnight at 4°C and then blocked with 1% bovine
`serum albumin. Wells were washed and incubated with 200 ,pl
`of serum sample overnight at 4°C. Without washing the
`samples from the wells, 50 ,ul of horseradish peroxidase-
`labeled IL-2 was added to a final concentration of 1:2000 for
`3 hr. During this reaction, IL-2 binds to the immobilized
`sIL-2Ra that was not previously occupied by anti-Tac. The
`unbound proteins were washed from the wells and the
`quantity of bound horseradish peroxidase-labeled IL-2 was
`determined in a colorimetric enzymatic assay with 2,2'-
`azinobis(3-ethybenzthiazoline sulfonate) (Sigma) in 0.1 M
`citrate buffer/0.03% H202, pH 4.2, for 30 min. The color
`intensity in this reaction was inversely proportional to the
`antibody concentration in the serum samples. The concen-
`trations of anti-Tac-H and anti-Tac-M were calculated from
`standard curves developed with purified anti-Tac-H and
`anti-Tac-M titrated on each plate.
`Pharmacokinetics. The pharmacokinetics of anti-Tac-H
`and anti-Tac-M were assessed by using 125I-labeled human-
`ized anti-Tac (125I-anti-Tac-H) and 13'l-labeled murine anti-
`Tac (131I-anti-Tac-M). lodination of each of the above prep-
`arations was performed by the iodine monochloride tech-
`nique of McFarlane (19). Five to 20 ,uCi (1 Ci = 37 GBq) of
`1311-anti-Tac-M and 125I-anti-Tac-H was administered in a
`single syringe to five cynomolgus monkeys. A 10-min blood
`sample was obtained for plasma volume determination. Ad-
`ditional blood samples were collected for counting at 2-hr
`intervals and then daily for 10 days after the administration
`of the labeled protein. The time course of decline of radio-
`activity from the plasma was plotted semilogarithmically.
`The terminal biological half-life (ti,2) of each labeled protein
`was determined graphically. The fraction of the intravascular
`
`PFIZER and SAMSUNG v. GENENTECH
`IPR2017-01489
`PFIZER EX. 1710, Page 2
`
`
`
`Immunology: Brown et al.
`
`Proc. Natl. Acad. Sci. USA 88 (1991)
`
`2665
`
`'3'l-anti-Tac-M
`Fraction of
`group IV pool
`catabolized
`per day
`0.49
`0.85
`0.60
`0.48
`0.46
`
`Survival
`t1/2, hr
`41
`25
`42
`43
`41
`
`Survival
`t1/2' hr
`108
`84
`104
`85
`133
`
`Metabolism of 125I-anti-Tac-H and '311-anti-Tac-M in
`Table 1.
`cynomolgus monkeys
`1251-anti-Tac-H
`Fraction of
`group IV pool
`catabolized
`per day
`0.25
`0.35
`0.24
`0.22
`0.18
`
`Animal
`1
`2
`3
`4
`5
`Mean +
`SEM
`0.58 ± 0.07
`0.25 ± 0.03
`103 ± 9
`38 ± 3
`Animals 1-3 received 0.5 mg of anti-Tac-H per kg and 0.5 mg of
`anti-Tac-M per kg in association with the radiolabeled monoclonal
`antibodies, whereas animals 4 and 5 received 2 mg of anti-Tac-H per
`kg on the day prior to the radiolabeled anti-Tac turnover studies.
`
`-2.5-fold longer survival in cynomolgus monkeys than did
`the simultaneously administered anti-Tac-M.
`Function. In functional studies, anti-Tac-M was inactive in
`antibody-dependent cellular cytotoxicity assays with either
`human or cynomolgus mononuclear cells. In contrast, anti-
`Tac-H was able to support antibody-dependent cellular cy-
`totoxic activity with either human or cynomolgus monkey
`mononuclear cells (data not shown).
`Efficacy of Anti-Tac Monoclonal Antibodies in Prolonging
`Cardiac Allograft Survival. The efficacy of the two forms of
`the anti-Tac monoclonal antibody in prolonging allograft
`survival was assessed by using a primate heterotopic cardiac
`allograft model. The animals were either untreated (group I;
`n = 5) or treated with 1 mg of anti-Tac-M per kg (group II;
`n = 5) or the same dose of anti-Tac-H (group III; n = 5) by
`bolus intravenous infusion every other day until graft rejec-
`tion. The five animals that received no immunosuppression
`(group I) rejected their grafts on or about day 9 posttransplant
`(mean survival, 9.20 ± 0.48 days; Table 2). In the animals that
`received unmodified murine anti-Tac (group II), graft sur-
`vival was prolonged when compared to that of the control
`group: mean survival in anti-Tac-M group 14 ± 1.98 days (P
`< 0.025 compared to the untreated group). Graft survival was
`further prolonged with anti-Tac-H with a mean survival of
`20.0 ± 0.55 days (P < 0.001 compared to control). Graft
`survival after treatment with anti-Tac-H was also prolonged
`over that observed in the group treated with anti-Tac-M (P <
`0.02). There was no evidence of toxicity attributable to the
`administration of either form of anti-Tac.
`Immunogenicity of Anti-Tac-M and Anti-Tac-H. The im-
`mune response to anti-Tac-M and anti-Tac-H was assessed
`by determining the serum concentrations of monkey anti-
`bodies to murine and humanized anti-Tac. All five animals
`receiving anti-Tac-M developed measurable antibody levels
`directed toward the murine monoclonal antibody. In the four
`animals with a graft survival of <9 days, monkey anti-anti-
`Tac-M antibodies were demonstrable 1-10 days (mean, 4
`Efficacy of anti-Tac monoclonal antibodies in
`Table 2.
`prolonging cardiac allograft survival
`Graft
`survival,
`days
`8, 9, 9, 9, 11
`9, 12, 12,
`17, 20
`19, 19, 20
`20, 22
`
`Group
`I
`II
`
`Antibody
`(dose)*
`None
`Anti-Tac-M
`(1 mg/kg)
`Anti-Tac-H
`(1 mg/kg)
`*Given every other day.
`
`III
`
`Mean
`survival ±
`SEM, days
`9.2 ± 0.48
`14 ± 1.98
`
`P vs.
`control
`
`<0.025
`
`20.0 ± 0.55
`
`<0.001
`
`pool catabolized per day (fractional catabolic rate) was
`determined according to the method of Matthews (20) as
`discussed (21).
`Function and Assay. The ability of the monoclonal anti-
`bodies to function in antibody-dependent cellular cytotoxic-
`ity was assessed with human and cynomolgus mononuclear
`cells as described (13).
`
`RESULTS
`Pharmacokinetics. To determine their pharmacokinetic
`properties 13ll-anti-Tac-M and 1251I-anti-Tac-H were admin-
`istered simultaneously in a single syringe to cynomolgus
`monkeys. These animals also received either (t) unlabeled
`anti-Tac [0.5 mg of anti-Tac-M per kg of body weight and 0.5
`mg of anti-Tac-H per kg of body weight (three animals)] in
`association with the radiolabeled monoclonal antibodies or
`(ih) 2 mg of anti-Tac-H per kg (two animals) on the day prior
`to the radiolabeled anti-Tac injection. The unlabeled anti-Tac
`was administered to saturate the antigenic target of the
`monoclonal antibodies-that is, the Tac protein soluble in the
`circulation as well as that on the surface of T cells. Initially,
`there was a rapid decline in the fraction of the administered
`radioactivity remaining intravascular, predominantly reflect-
`ing distribution of the radiolabeled antibody from the intra-
`vascular to the extravascular compartments (Fig. 1 and Table
`1). This phase was followed by a slower terminal exponential
`decline in plasma radioactivity that predominantly reflects
`catabolism of the monoclonal antibody. In the five animals
`studied, the ti,2 of the terminal exponential of 12-5I-anti-Tac-M
`decline was 38
`3 hr with a mean fraction ofthe intravascular
`pool of anti-Tac-M catabolized per day of 0.58
`0.07. The
`survival t~l2 of 125I-anti-Tac-H was prolonged when compared
`to radiolabeled anti-Tac-M. The mean t,!2 of the terminal
`exponential of the serum 125I-anti-Tac-H die-away curve was
`9 hr with a fraction of the intravascular pool catabo-
`103
`lized per day of 0.25 + 0.03. Thus, anti-Tac-H had an
`
`A
`
`14
`
`8
`
`7
`DAYS
`
`0.4
`
`0.2
`
`0.1
`
`zZ
`
`c:w
`cnw
`
`C,)
`0
`
`aw
`
`w z u
`
`-
`
`0 z 0 u
`
`a 0.051
`
`FIG. 1. The metabolism of 125I-anti-Tac-H and '1ll-anti-Tac-M
`(each at 0.5 mg/kg) in a single cynomolgus monkey. The fractions of
`the injected dose remaining in the serum for 125I-anti-Tac-H (A) and
`for '311-anti-Tac-M (o) are indicated. The ti/2 of the terminal expo-
`nential with anti-Tac-H (t1/2, 104 hr) was 2.5-fold longer than that
`observed with anti-Tac-M (t1/2, 42 hr).
`
`PFIZER and SAMSUNG v. GENENTECH
`IPR2017-01489
`PFIZER EX. 1710, Page 3
`
`
`
`2666
`
`Immunology: Brown et al.
`
`Proc. Natl. Acad. Sci. USA 88 (1991)
`
`antibodies directed toward anti-Tac-M. Such antibodies were
`demonstrable within 2 days after the first demonstration of
`rapid clearance of anti-Tac-M. In contrast, the animals re-
`ceiving anti-Tac-H generally maintained measurable levels of
`this monoclonal antibody throughout the period of adminis-
`tration of the monoclonal antibody-that is, from initiation of
`therapy until after graft rejection (Fig. 2B). In all cases,
`anti-Tac-H was demonstrable in the trough sample obtained
`within 48 hr of graft rejection. Taken together, these results
`indicate that anti-Tac-H is less immunogenic in cynomolgus
`monkeys than is the anti-Tac-M monoclonal antibody.
`
`tion. In these four animals, anti-
`days) before allograft reject
`bodies first appeared on c
`lays 6-15 (mean, day 11) after
`initiation of anti-Tac-M as
`Iministration. Exceedingly high
`antibodies were still present 17
`levels of anti-anti-Tac-M z
`months after anti-Tac-M th
`erapy.
`-ss immunogenic when adminis-
`Anti-Tac-H was much le
`ys. None of the animals produced
`tered to cynomolgus monke,
`antibodies to anti-Tac-H c
`luring the 19- to 22-day period
`between initiation of anti-T,
`ac administration and the time of
`animals, no anti-anti-Tac-H anti-
`allograft rejection. In two z
`during the 21 and 26 days of
`bodies were demonstrable
`g three animals developed anti-
`observation. The remaining
`anti-Tac-H antibodies on de
`ays 33, 33, and 42 after initiation
`tn. Furthermore. the hicihest con-
`of anti-Tac-H administratio
`centrations of anti-Tac-H ar
`itibodies observed 4 months after
`rg/ml) in three animals
`.6, and 23
`initiation of therapy (0.47, 3
`a00-fold lower than the highest
`producing antibodies were
`anti-anti-Tac-M antibodies (300
`concentrations of monkey
`00ntig/ml) observed in animals
`331, 1675, 2510, and 14,0
`body. During the first week of
`receiving the murine antit
`therapy, all 10 animals reck
`eiving either anti-Tac-M or anti-
`im levels of the monoclonal anti-
`Tac-H had measurable seru
`ir administration as well as 48 hr
`body both immediately afte
`-rior to the subsequent dose of
`afterwards-that is, just p
`that adequate anti-Tac antibody
`antibody-thus indicating I
`tturate the receptors and to leave
`had been administered to se
`ly (Fig. 2). Specifically, the geo-
`residual circulating antibod
`h levels for anti-Tac-M during the
`metric mean peak and troug
`tg/ml, respectively.
`1.03 and 3.9
`initial week of study were I
`)eriod of administration but prior
`However, after this initial p
`3als receiving anti-Tac-M demon-
`to graft rejection, the anim
`the administered monoclonal an-
`strated rapid clearance of 1
`vas no longeredemonstrable (i.e.
`tibody so that anti-Tac-M v
`wn prior to the subsequent admin-
`<40 ng/ml) in the circulatic
`his precipitous fall in measurable
`istration of the antibody. TI
`with the development of monkey
`anti-Tac-M was associated
`
`.2
`
`co
`a)e
`o-f
`0) u
`encc
`
`60 A
`48
`
`36
`24
`
`12
`
`0
`
`8
`
`,o0 B
`6
`W0 B
`0
`36
`24
`12
`
`1
`
`%O0
`
`8
`
`DISCUSSION
`The IL-2Ra chain has been used as a target for relatively
`specific immunosuppression to exploit the difference in re-
`ceptor expression between resting T cells that do not display
`IL-2Ra and T cells that express this receptor when stimulated
`by the specific interaction with foreign histocompatibility
`antigens on an allograft. Anti-IL-2R-directed therapy has
`been incorporated into combination immunosuppression pro-
`tocols involving human recipients of cadaveric renal al-
`lografts. In each of these studies, there was a reduction in the
`number of early rejection episodes without evident toxicity.
`These prior attempts to use the murine anti-Tac monoclonal
`antibodies in allograft transplantation protocols were limited
`by the relatively short survival of the murine monoclonal
`antibody, weak recruitment of effector functions, and neu-
`tralization by human antibodies directed to the mouse mono-
`clonal antibodies. To circumvent these difficulties, human-
`ized anti-Tac-H was constructed by combining the CDRs of
`the murine anti-Tac antibody with human yl heavy and K
`light-chain framework and constant regions (12). In the
`present study, we demonstrate that these humanized anti-
`bodies have a longer in vivo survival in primates compared to
`the murine antibody, are more effective in recruiting effector
`functions, have reduced immunogenicity, and display greater
`effectiveness in preventing allograft rejection in the primate
`heterotopic cardiac allograft model used.
`The pharmacokinetics of radiolabeled anti-Tac-H differ
`substantially from those of radiolabeled anti-Tac-M when
`administered to normal cynomolgus monkeys. In animals
`receiving unlabeled anti-Tac to block the antigenic target, the
`tl/2 of the terminal exponential of humanized anti-Tac was 103
`hr compared to 38 hr for anti-Tac-M. Furthermore, the
`percentage of the intravascular pool of radiolabeled antibody
`catabolized per day was 25% with radiolabeled anti-Tac-H,
`24
`32
`40
`markedly lower than the 58% observed with anti-Tac-M. The
`H observations of prolonged survival of a humanized monoclo-
`°-
`-H Trough Levels
`nal antibody in the present study are in accord with the
`& Anti -Tac-
`observations of LoBuglio and coworkers (22), who noted
`O
`longer survivals of murine/human chimeric monoclonal an-
`tibodies in humans when compared to the parent murine
`monoclonal antibody. In studies defining the metabolism of
`immunoglobulin subunits, it was shown that the rate of
`catabolism of an immunoglobulin is controlled by the Fc
`region of the immunoglobulin, specifically the CH2 domain
`(C, constant region; H, heavy chain) (21, 23). Thus, the
`longer survival of anti-Tac-H when compared to anti-Tac-M
`probably reflects the replacement of the murine IgG2a CH2
`domain with the IgG1 human CH2 domain. The survival tl'2
`of the IgG1 humanized anti-Tac molecule in cynomolgus
`monkeys noted in the present study (103 hr) is shorter than
`the mean t1/2 of >20 days we reported previously for human
`IgG1 immunoglobulins administered to humans (24). This
`disparity may reflect the differences in the recipient species
`dinath parelctitifferences Ince recient human
`used in the pharmacokinetic studies since polyclonal human
`IgG was shown to have a ti12 of survival of >20 days in
`humans, 12 days in baboons, and only 6.6 days in rhesus
`monkeys (21). In light of these observations in subhuman
`
`16
`
`Cr
`
`M
`
`>
`
`P O
`En
`
`1
`DAYS Pi
`
`16
`24
`32
`OST TRANSPLANT
`
`40
`
`anti-Tac-H antiserum concentration
`Anti-Tac-M and;
`FIG. 2.
`receiving anti-Tac-M (A) and anti-
`profile in individual monkeys
`obtained at the times indicated (A)
`Tac-H (B). The samples were
`lody infusion. The last infusion of
`immediately before the antib
`er transplantation (12 days after first
`anti-Tac-M was on day 11 aft(
`H was on day 17 after transplantation
`infusion) and that of anti-Tac-J
`since the animals rejected their al-
`(18 days after first infusion),
`spectively. Seroconversion is defined
`lografts on days 12 and 19, res
`s to the monoclonal antibodies were
`as the first time that antibodie
`detected.
`
`PFIZER and SAMSUNG v. GENENTECH
`IPR2017-01489
`PFIZER EX. 1710, Page 4
`
`
`
`Immunology: Brown et al.
`
`Proc. NatL. Acad. Sci. USA 88 (1991)
`
`2667
`
`primates and humans, anti-Tac-H may have a much longer
`survival in humans than the 103 hr observed for this mono-
`clonal antibody in cynomolgus monkeys.
`A major goal in the generation of humanized antibodies is
`the production of agents that lack the epitopes recognized by
`T-helper cells that may be required for the effective produc-
`tion of anti-monoclonal antibody responses. The chimeric
`antibodies (variable region retained from the mouse and
`constant region derived from human) to different antigens
`studied by Hale and LoBuglio and their coworkers (22, 25)
`elicited modest antiglobulin responses. In the present study,
`humanized anti-Tac-H was less immunogenic than the parent
`murine monoclonal anti-Tac-M. With anti-Tac-M, all animals
`produced antibodies to the monoclonal antibody with anti-
`bodies evident prior to allograft rejection in all animals with
`an allograft survival of >9 days. Antibodies directed toward
`the murine monoclonal antibody were detectable in the
`circulation by the 15th day in all animals studied. In contrast,
`only three of the five animals receiving anti-Tac-H produced
`anti-monoclonal antibody responses during the study period.
`Furthermore, in the remaining three animals, antibody was
`first detected on days 33-42 after initiation of therapy, in all
`cases after allograft rejection had occurred. In more exten-
`sive studies, this lowered immunogenicity of the humanized
`anti-Tac was demonstrated in groups of cynomolgus mon-
`keys injected with 0.05, 0.5, and 5 mg/kg doses of anti-Tac-H
`or anti-Tac-M daily for 14 days (J.H., unpublished observa-
`tions).
`The development of antibodies to the monoclonal antibody
`drastically altered the pharmacokinetics of the radiolabeled
`monoclonal antibody. For example, the til2 of radiolabeled
`anti-Tac-M was reduced from the normal value of 38 hr to 9
`hr and to <10 min in the two animals studied that manifested
`high-titer antibodies to this monoclonal antibody 17 months
`after initial administration of anti-Tac-M (T.A.W., unpub-
`lished observations). In the present therapeutic trial, the
`development of antibodies to the anti-Tac-M also led to
`accelerated catabolism of the antibody and thus reduced the
`amount of anti-Tac-M available for blocking IL-2 binding to
`Tac-expressing cells. Specifically, after the development of
`antibodies to the murine monoclonal antibody, anti-Tac-M
`was no longer demonstrable in the plasma 2 days after its
`infusion-that is, in the plasma sample obtained immediately
`before the antibody infusion. Thus, the development of
`antibodies to anti-Tac-M may be one of the factors leading to
`the failure of this monoclonal antibody preparation to prevent
`allograft rejection beyond day 14 of the study.
`A major goal in the production of humanized antibodies is
`to increase effector functions by recruiting immune effector
`cells. Mouse monoclonal antibodies are frequently unable to
`promote antibody-dependent cellular cytotoxicity in assays
`with human effectors and nucleated target cells. In some, but
`not all, cases, murine/human chimeric antibodies mediated
`antibody-dependent cellular cytotoxicity with human effec-
`tor cells more efficiently than murine antibodies (22). In
`previous studies, we demonstrated that anti-Tac-H mani-
`fested antibody-dependent cellular cytotoxicity with human
`mononuclear cells, a function absent in the parental murine
`anti-Tac (13). In the present study, we extend these obser-
`vations by indicating that anti-Tac-H can mediate antibody-
`dependent cellular cytotoxicity with cynomolgus monkey
`mononuclear cells. Thus, anti-Tac-H manifests new capabil-
`ities to mediate cytotoxicity with primate mononuclear cells
`absent with the murine monoclonal antibody.
`In the present study, cardiac allograft survival was mod-
`estly but significantly increased in the group of animals
`treated with anti-Tac-M, with an increase of mean survival
`from 9.2 to 14 days. As noted above, the development of
`
`antibodies to anti-Tac-M in monkeys may have contributed to
`the modest efficacy of anti-Tac-M in preventing allograft
`rejection. Allograft survival in anti-Tac-H-treated animals
`was further prolonged to 20.0 days, a survival significantly (P
`< 0.001) longer than that observed with anti-Tac-M. In
`contrast to the situation with anti-Tac-M, the rejection of the
`cardiac allografts by day 20 in animals receiving anti-Tac-H
`does not appear to be due to the development of antibodies
`to the infused monoclonal antibody since such antibodies
`were not demonstrable prior to graft rejection. Therefore, it
`would appear that, although anti-Tac-H can significantly
`prolong graft survival in primates without toxic side effects,
`it is not sufficient as a single agent to prevent allograft
`rejection. Nevertheless, it may complement the efficacy of
`other immunosuppressive agents in allograft protocols. Thus,
`our developing understanding of structure and function ofthe
`IL-2R on the surface of activated T lymphocytes taken
`togethe