\
`Immunotoxins
`
`edited by
`
`ARTHUR E. FRANKEL
`Duke University Medical Center
`.
`Department of Medicine
`Division of Hematology/Oncology
`Box 3898 DUMC
`Durham, North Carolina 27710
`
`J
`i
`
`1988 KLUWER ACADEMIC PUBLISHERS
`BOSTON I DORDRECHT I LANCASTER
`
`IMMUNOGEN 2313, pg. 1
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Distributors
`
`for the United States and Canada: Kluwer Academic Publishers, 101 Philip
`Drive, Assinippi Park, Norwell, MA 02061, USA
`.
`for the UK and Ireland: Kluwer Academic Publishers, Falcon House, Queen
`Square, Lancaster LA1 lRN, UK
`for all other countries: Kluwer Academic Publishers Group, Distribution
`Centre, P.O. Box 322, 3300 AH Dordrecht, The Netherlands
`
`Library of Congress Cataloging in Publication Data
`lmmunotoxins.
`
`(Cancer treatment and research)
`Includes bibliographies and index.
`1. Antibody-toxin conjugates. 1. Frankel, ArtJ1ur E.
`JJ . Serie. [DNLM : l. Antibody-Toxin Conj4gate .
`WJ CA693 I QW 630 133J
`QR185.8.A58146 1988
`I BN 0-89838-984-4
`
`87-28173
`
`616.07'9
`
`Copyright
`
`© 1988 by Kluwer Academic Publishers.
`All rights reserved. No part of this publication may be reproduced, stored in
`a retrieval system, or transmitted in any form or by any means, mechanical,
`photocopying, recording, or otherwise, without the prior written permission
`of the publishers, Kluwer Academic Publishers, 101 Philip Drive, Assinippi
`Park, Norwell, MA 02061, USA .
`
`PRINTED IN THE UNITED STATES OF AMERICA
`
`IMMUNOGEN 2313, pg. 2
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`28. Clinical studies: Solid tumors
`
`Lynn E. Spitler
`
`Introduction
`
`Immunotoxins (ITs) permit delivery of a therapy at the tumor site specifical(cid:173)
`ly. The use of ITs as therapy of patients with solid tumors presents problems
`which require unique solutions. These include stability in vivo, cellular heter(cid:173)
`ogeneity, access to tumor, biodistribution, and the immune response to the
`immunoconjugate (Table 1). It is necessary to address some of these issues
`before contemplating entry into clinical trials, whereas others are more
`appropriately undertaken after clinical trials have been initiated, using the
`results of the clinical observations as a focus for planning improvements as
`part of a second generation effort. This involves both optimizing the admin(cid:173)
`istration of the currently available products and developing new, improved
`products.
`It is important to know that the conjugates can be expected to have
`acceptable stability in vivo. Without this, it would not be reasonable to
`expect the antibody to achieve targeting of the toxin to tumor cells. Such
`stability can be demonstrated at the preclinical level in vitro by incubation
`of the IT in serum at 37ac and in vivo by administration to experimental
`animals. Once a conjugate with reasonable stability has been achieved, one
`could initiate clinical trials with this agent while proceeding, if appropriate,
`with second generation efforts to enhance stability through new or improved
`conjugation techniques.
`Similarly, the question of cellular heterogeneity should be considered
`before initiation of clinical trials. It would only be reasonable to proceed in
`the trials if it was like ly that the antibody used for targeting had reactivity to
`a high proportion of cellf~ in the patient's tumor. Thi could be achieved by
`1) preselecting an antibody having broad cro ·-reactivity with tumors of a
`particular histologic type, 2) selecting the patients to be treated on the basis
`of demon trated reactivity of the antibody with biopsies of their tumor or
`3) custom con truclion of an antib dy having reactivity to the patient's
`tumor. T he u e of c cktail of IT to attack a highe r proportion of cells in
`the population could be con ide red a a seco nd generation effort. Similarly,
`second generation efforts could involve lhe use of agents uch as interf ron ,
`
`Frankel , A .E .. (cd.), lmmunoloxin s.
`© 1988 Kluwcr Academic Publishers. ISBN 0·8983~-9R~ · 4 .
`All rights reserved.
`
`493
`
`IMMUNOGEN 2313, pg. 3
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Table 1. Unique problems in consideration of the use of immunotoxins in therapy of solid
`tumors
`
`I. Stability of conjugates in vivo
`
`2. Cellular heterogeneity
`-
`inter tumor
`-
`intra tumor
`-
`cell cycle/ploidy
`-
`antigen expression
`
`3. Access and localization in the tumor.
`
`4. Biodistribution
`-
`uptake by the reticuloendothelial system via carbohydrate receptors
`-
`internalization into cell
`intracellulm distribution
`-
`
`5. Immune response to immunoconjugate
`- murine antibody
`-
`ribosomal inhibiting protein
`
`to increase antigenic representation of the tumor cells. Custom construction
`of an antibody to a patient's tumor is impractical because 1) the time
`involved generally precludes this approach and 2) for most tumors, anti(cid:173)
`bodies are already available so that all that is necessary is to screen the
`patient's tumor and select the antibody having appropriate reactivity.
`The issue of access of the IT to the tumor and localization can best be
`addressed after the clinical trials have been initiated. Clinical trials are
`necessary in order to determine whether or not the IT reaches the tumor
`and the extent of localization. It is important to determine the optimal
`dosing regimen for delivery of IT to the tumor. In addition, there are a
`number of other ways which may improve localization. These include ,
`among other things, increasing vascular permeability, antigen representa(cid:173)
`tion, or altering the binding affinity of the antibody.
`Some ribosomal inhibiting proteins, such as the A chain of ricin, contain
`carbohydrates which bind to carbohydrate receptors in the reticuloendothe(cid:173)
`lial system. Immunoglobulins also have carbohydrates which, if exposed,
`could also bind to such carbohydrate receptors. Clinical trials are necessary
`to determine the clearance and side effects of the IT io assess the relevance
`of such carbohydrate binding in therapy. If relevant, efforts could be aimed
`at deglycosylation/hypoglycosylation and/or use of agents to block the car(cid:173)
`bohydrate receptors, with the realization that greater toxicity or a different
`spectrum of toxicity might result because of greater availability of IT.
`Finally, entry into clinical trials is necessary to determine if the patients
`mount an immune response to the components of the IT and, if so, to
`determine a means to abrogate the immune response. This could be done
`through second generation efforts using 1) agents to modulate the immune
`response, 2) induction of tolerance, or 3) modifying the immunoconjugate
`
`494
`
`IMMUNOGEN 2313, pg. 4
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`to make it less immunogenic. It should be noted that the possibility of the
`occurrence of an immune response is not unique to IT therapy. It will be a
`consideration with all pr duct · involving monoclonal antibodies since they
`all represent foreign proteins. The problem may not be circumvented by the
`u e of human monoclonal antibodies since the idiotype is foreign to the
`patient ancl an immune re. ponse may still occur. Preliminary evidence
`suggests that this, in~leed, is the case.
`It is clear from this discussion that there are many considerations in
`judging the appropriate time for entry into clinical trials. It is important that
`any agent used in clinical trials be safe and have a reasonable chance for
`therapeutic efficacy. On the other hand, there is only so much information
`which can be gained through preclinical evaluation, and then it is necessary
`to turn to sludie
`in patients in order to gain further information. It is likely
`that in upcoming years we will sec additional clinical testing of ITs and that
`the information gained from the ·e trials will be used in second generation
`efforts to improye the effica.cy of these products.
`
`FDA review
`
`ITs present special pr blems in preclinical evaluation Lhat are unlike those
`presented by other cancer therapeutics [I]. These arc shown in Table 2 and
`are discus eel b low. These considemlions follow concept propo ed by the
`Fo d <lJld Drug Administration of the United
`tate in its document entitled
`Points to Consider in the Mwwfacture of Monoclonal Antibody Products for
`Human Use.
`
`Binding activity and specificity
`
`For olid tumors, it is important that the antibody to be used in the
`con truct of th
`IT shows specitic reactivity to a high percentage of tumors
`f the same histological type obtained from various individuals. Techniques
`often used to evaluate binding include enzyrne-Jjnked immunoassay (EIA)
`radioimmunoassay (RlA) flow cytometry, immunoperoxidase staining, and
`immunofluorescence.
`f cells in
`The antibody must also how reactivity with a high perce11tage
`the population. At the present Lime, it i thought that ITs kill only the cells
`to which tbe antibody component of the IT binds because internalization of
`the A chain by each cell i necessary for subsequent cell killing. This is
`unlike the ·ituation with chemotherapeutic and radiotherapeutics conju(cid:173)
`gated to moqoclona l antibodies in which cells surrounding the bound conju(cid:173)
`gate would al o be killed.
`Because it has been reported that one ricin A chain entering the cytosol is
`sufficient to kill the cell [2], it is essential that antibody u ed in conjugates
`with each material not have any imp rtanl cr s -r a tivity with normal
`
`495
`
`IMMUNOGEN 2313, pg. 5
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Table 2. Key points in preclinical evaluation of immunotoxin
`
`I . Binding activity and specificit y of a ntibody and IT
`a. Reactivity with tumors of a particular histologica l type
`I. Percentage of tum ors
`2. Percentage of cell s in population
`b. Lack of reactivity with normal tissues
`
`2. Product fre e of contamina tion
`a. Murine viruses
`b. Pare ntal hybridoma DNA and RNA
`
`3. T oxicology
`
`4. Efficacy
`a. In vitro
`b. In vivo
`
`Reproduced, with permission , fro m Spitle r, L.E. (1987) Phase I Clinical Trials with Immuno(cid:173)
`toxins . In: lmmunoconjugates: Antibody Conjugates in Radioimaging and Therapy of Cancer.
`(C. W . Vogel, ed .) Oxford University Press, New York.
`
`tissues. If there were unrecogni zed cross-reactivity with membrane antigens
`on normal tissues, . vere toxicity could result. Preclinical toxicology testing
`might not reveal pote ntial toxicity in humans if the membrane antigens are
`not represented on th e surfac of the cells of the species being tested. One
`reasonable approach to this problem is to use the immunoperoxidase tech(cid:173)
`nique to examine the antibody carefully for cross-reactivity with frozen sec(cid:173)
`tions of normal tissues from several autopsies [3]. Frozen sections are used .
`unless it is known beforehand that formalin fixation does not destroy the
`relevant antigen. Another useful approach is to determine whether the
`antigen is represented on th e tissue of the species used for the toxicology
`evaluation. If the antigen is present in these animals, it would suggest that
`documented safety in the subhuman species would correlate with safety in
`clinical trials .
`
`Contamination testing
`
`Contamination testing is an important element of preclinical testing to
`ensure the safety of administration of the product to patients . It is necessary
`to show that the product is free from contamination with viruses and with
`parental hybridoma DNA and RNA. Electron micro c py and o ther assay
`procedures are u e ful way. to screen for viral co ntaminatio n. Special testing
`for lymphocytic cho ri meniJlgitis virus by intracerebral inocul ation in wean(cid:173)
`ling mice is necessary . In addition
`the product must be evaluated for con(cid:173)
`tamination with C-type particles . Procedure validation can also be used to
`show that the purification process effectively removes murine viruses con(cid:173)
`taminating the hybridoma cell line (4] .
`
`496
`
`IMMUNOGEN 2313, pg. 6
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`The product must also be shown to have minimal or no contamination
`with parental hybridoma DNA or RNA because of the concern regarding
`the oncogenic potential of nucleic acids from the malignant cell line used as
`the fusion partner. Procedure validation can be used to show that the
`purification process removes parental hybridoma nucleic acids [4] . In addi(cid:173)
`tion, the product can be assayed for nucleic acids by hybridization tech(cid:173)
`niques.
`
`Toxicology
`
`Because ITs represent new agents in cancer the rapy , careful toxicology
`studies are essential to evaluate the afe ty of the ·e products. Appropriat
`toxicology testing includes acute toxicity testing in mice and subacute le ting
`in rats and primates; the detaiis are determined by the pr pos d do ing
`regimen. As experience from preclinical toxicity te. ling and clinical trials
`grows, it may no longer be necessary to perform primate testing on each
`new IT in which only the antibody component of the product is changed if
`binding studies document that there is no cross-reactivity with normal tis(cid:173)
`sues.
`For products involving an A chain separated from a B chain moiety, it is
`essential to demonstrate that there is no contamination of the A chain
`preparation by the whole toxin by the most sensitive means available.
`In evaluating the results of preclinical toxicity testing, it must be kept in
`mind that there is a marked variation in species sensitivity to the toxic
`effects of plant lectins. The reference work on this subject was done by
`Balint [5], who
`tudi d furm animals feeding on whole castor beans . A
`greater than 100-fold liffcr · nee in sensitivity between species was observed,
`with th e I thai dose of beans in hens being 14 g(kg and in horses, 0.1 g/kg .
`Abrin, a toxin related to ricin al o has a 'significant species variation in
`toxicity as bown by Jan en
`r al. [6]. The lethal dose in rats was reported
`to be 0.35- 0.5 mg/kg wherea that for rabbits was 0.03-0.06 mg/kg. Accord(cid:173)
`ingly, it is appropriate that clinical. trials of IT be initiated at doses substan(cid:173)
`tially lower than the doses shown to be safe in preclinical toxicity testing.
`
`Efficacy
`
`With ITs, as with other potential therapeutic agents, it is essential that a
`reasonable probability of efficacy be demonstrated before proceeding with
`clinical trials. This can be done by demonstrating specific cytotoxicity to
`tumor cells in vitro and/or inhibition of tumor growth in vivo. In add ition ,
`radioimmunoimaging studies with radio labelcd antibody of the s~une type to
`be used in IT preparation can show that the an tibody does localiz
`in tumors
`in human subjects.
`Several methods can be used to assess cytotoxicity in vitro . These include
`
`497
`
`IMMUNOGEN 2313, pg. 7
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`inhibition of incorporation of radioactive precursors of DNA (e.g.,
`
`e4CJthymidine), of protein (e.g., eHJleucine or e5S]methionine), dono(cid:173)
`
`genic assays, or trypan blue dye exclusion . It is essential that specificity also
`be confirmed by the use of a control cell line with which the antibody
`component of the IT does not react to evaluate the possibility that any
`observed cytotoxicity might be nonspecific. It is also necessary to evaluate
`both the test and control cell lines for susceptibility to the toxin component
`of the IT to judge whether any observed cytotoxicity is a result of differ(cid:173)
`ences in the antibody binding to the cell line or just to differences in
`sensitivity of the cells to the toxic moiety.
`For evaluation of in vivo efficacy, the animal model will be somewhat
`artificial because the antibody component of the IT is often specific to
`human tumors and does not cross-react with tumor-associated antigens or
`tumors of subhuman species. In many cases, this necessitates the use of the
`nude mouse model for evaluation of efficacy in vivo. In this model, consid(cid:173)
`eration should be given to testing the IT in a manner relevant to the human
`situation. In some studies, IT has been mixed with the tumor cells prior to
`implantation, injections have been started simultaneously with implantation,
`or ITs have been injected directly into the tumor. While these studies may
`give some indication about activity of the ITs, they are not analogous to the
`situation in human subjects with malignancy and do not provide substantial
`information about potential efficacy in the clinical setting. It is more
`appropriate to initiate IT therapy when the tumors have become established
`in order to evaluate the potential of the IT for clinical efficacy.
`
`Preclinical evaluation of Xomazyme®-Mef, a ~urine monoclonal
`antimelanoma antibody-ricin A chain immunotoxin
`
`I
`
`-
`
`.. '
`
`Taking these points into consideration, we conducted the preclinical evalua(cid:173)
`tion of an antimelanoma IT for subsequent initiation of clinical trials [7].
`Key elements of this preclinical evaluation are described below.
`The monoclonal antibody is of the IgG2a subclass. it is reactive with
`melanoma-associated antigens having molecular weights of 220,000 daltons
`and over 500,000 daltons. Binding activity was evaluated by several
`methods. EIA and RIA demonstrated that the antibody reacted with the
`majority of melanoma cell lines tested but did not show reactivity with a
`matching lymphoblastoid line derived from the donor of one of the melano(cid:173)
`ma cell lines . Flow cytometry confirmed these observations and provide.d the
`additional information that the antibody reacted with over 99% of the cells
`in the population. By immunoperoxidase testing, the antibody was shown to
`react with all melanomas tested (> 50), with multiple metastases from the
`same donor, and with 70% to 100% of cells in the population in 19 of 20
`melanomas evaluated. There was no cross-reactivity with the majority of
`normal tissues evaluated. With the exception of nevus cells and vascular
`
`498
`
`IMMUNOGEN 2313, pg. 8
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`e ndothelium, any observed rcacllvtty was substantially less than that
`observed with melanoma cells. There was no important reactivity with the
`pigmented layer of the retina, a significant observation because of potential
`antigenic similarity of these cells with melanoma cells. Reactivity was
`observed with vascular e ndothelium , but to the extent that it could be
`evaluated by
`the
`immunoperoxidase technique, was cytoplasmic. This
`observation is significant because the antibody component of the IT would
`not be excepted to react with cytoplasmic antigen following in vivo adminis(cid:173)
`tration because it would not reach antigens inside cells. The antibody also
`lacked reactivity with blood group substances A and B, and with mino r
`blood group substances.
`T he hybridoma cell line producing the antime lanoma antibody was tested
`for contamination with 12 murine viruses. No viral contamination was de(cid:173)
`tected . However, the hybridoma cell line was s hown to contain C-type
`particles by electron microscopy. The xenotropic (S+L- ) test was positive
`for the cell line, as would be expected on the basis of the electron micros(cid:173)
`copy , but was negative for the purified antibody, indicating that the purifica(cid:173)
`tion procedLtre was effective in removing this contamination. Furthe r eva l(cid:173)
`uation was carried out in collaboration with Dr. Jay Levy of the University
`of Cali fornia Medical Center, San Francisco [4) . The hybridoma line con(cid:173)
`tained five infectious xenotropic virus particles and six infectious ectropic
`·particles per 106 cells. To validate the purification procedure, the ascites
`containing the antibody was spiked wi th extra xenotropic and ectropic virus.
`Antibody was purified from the ascities and was shown to be free of
`contamination with these viruses.
`To evaluate potential contamination of the product with parental hybrid(cid:173)
`oma nucleic acids, the hybridoma cell line was c.:ultured with f14C) thymi(cid:173)
`dine or C4CJ uridine to label the DNA and RNA. The radiolabeled nucleic
`acids were isolated and added to ascites. The antibody was purified by
`protein A affinity chromatography, and it was found that all radioactivity
`eluted in the void volume of the column. No radioactivity eluted with the
`antibody, thereby showing that the purification procedure effectively elimin(cid:173)
`ated parental hybridoma nucleic acid contamination. This was verified by
`DNA hybridization, which showed minimal or undetectable contamination
`of the product with nucleic acids.
`E fficacy was demonstrated both in vitro and in vivo. The LT was shown to
`inhibit protein synthesis speci fically in a melanoma cell line and not in
`appropriate control cell lines as determined by inhibition of radiothymidine
`incorporation (Figure 1). The nude mouse model was used to evaluate in
`vivo efficacy. Melanomas were transpla nted to nude mice and allowed to
`become established . Weekly i.v. or daily i.p. injections of lTs resulted in
`significant inhibition of the growth of the melanoma as compared to that
`observed in control animals o r those given a single injectio n of IT (Figure
`2}.
`
`499
`
`IMMUNOGEN 2313, pg. 9
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`120
`
`100
`
`~
`:J
`a:i
`<1: >
`
`~ 0
`
`80
`
`60
`
`40
`
`20
`
`0
`MEDIUM
`
`-11
`
`-10
`
`-9
`
`-8
`
`-7
`
`-6
`
`MOLAR CONCENTRATION (LOG-10)
`-o- CONTROL CELLS
`-
`MELANOMA CELLS
`
`Figure 1. Inhibition of radothymidine incorporation in melanoma cell line by antimelanoma IT.
`Reproduced, with permission, from Spitler, L.E. (1987) Phase I Clinical Trials with Immunotox(cid:173)
`ins. In: lmmunoconjugates: Antibody Conjugates in Radioimaging and Therapy of Cancer.
`C.W. Vogel. ed. Oxford University Press, New York.
`
`6
`
`5
`
`~ 4
`0
`
`X
`~ 3
`0
`
`2
`
`0
`
`5
`
`10
`
`15
`
`20
`
`30
`
`35
`
`. 40
`
`45
`
`50
`
`25
`DAYS
`-o- NO TREATMENT (N = 20)
`---+- SINGLE INJECTION CONJUGATE I.V. (N = 4)
`-o- CONJUGATE INJECTION DAILY I.P. (N = 10)
`- - CONJUGATE INJECTION WEEKLY I.V. (N=9)
`
`Figure 2. Inhibition of growth of melanoma in nude mice receiving weekly i.v. or daily i.p.
`injection of antimelanoma immunotoxin. Reproduced, with permission, from Spitler, L.E.,
`(1987) Phase I Clinical Trials with Immunotoxins. In: lmmunoconjugates: Antibody ConJugates
`in Radioimaging and Therapy of Cancer. C. W. Vogel. ed. Oxford University Press, New York.
`
`500
`
`IMMUNOGEN 2313, pg. 10
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Melanoma
`
`Phase I trial
`
`In a radioimmunoimaging study, c nductecl under U.S. Food a nd Drug
`Administration (FDA) Investigationa l New Drug (rND) exemptio n notice
`we showed that the monoclonal antimclan ma a ntibody used in the con(cid:173)
`f the IT localizes in me lanomas following in vivo admi nistration to
`struct
`patient . [8]. Thee e ncouraging re ults led to a tri a l of the IT in 22 patient
`with metastatic malignant me la noma. The do ·e of IT admini te red ranged
`fr m 0.01 mglkg daily for five days to I mglkg daily for fo ur day· (total
`dose: 3.2 mg to 300 mg) (9).
`Side effect a:socia ted with IT administration are summarized in Table 3.
`The severity of the
`·ide
`ffccts wa genera ll y relat~d to the d se of IT
`administered. Jn <.lose up to and including 0.5 mg/kgld·ty [or five day ·, these
`side effect. were transient and r v rsible in all patient eva lua ted. The
`f this rr was a fa ll in erum albumin with a
`major d se-limiting side effect
`conco mit a nt clccrca e in e nun to ta l protein values. Weight gain and fluid
`shift which resulted in edema were noted .
`The fall in serum albumin was observed in all patients receiving a do, e of
`at I ast 0.2 mg/kg/day of IT. This decrease u ua lly
`curred a fter the. econd
`dose of IT. There wa no associated pro te inuria. Serum albumin leve ls
`stabilized and began to return to normal within 48 hours of the last treat(cid:173)
`ment with IT.
`This side effect may be a genen1l manifestation of IT therapy rather than
`one directly related to thi particular IT ince we have observed similar side
`effect· in preclinical nod clinical te tin g of other ITs.
`All pat ient· receiving a dose fat least 0.2 mg/kg/day of IT ga ined weight
`during the cour c of ho ·pitalization for IT treatment. Asso iated with this
`f edema in some patie nt . There were no sign · of
`were clinical sign
`pulmonary e dema in any patient. The patient ' weights dccrea eel to base(cid:173)
`the ·erum albumin returned to prethe rapy leve l .
`line leve ls a
`Many of the patient ma nife ted fever
`everal hours after infu io n of IT
`and they generally ranged between 38° and 39°C. In addition mo t patients
`experienced maJaise, fatigue, and decrease in appetite during. the course of
`ho pitalizatio n. Some patiems had as ·ociarecl myalgia . The e side effect.
`were self-limiting and well tolerated.
`A transient change to decreased voltage on EKG without other ST-T
`wave changes or clinical ympl ms was noted. Sinus tachycardia was also
`clinicaJJy observed in mo t patients. Serial echoca rdi ogram a nd creatine
`pho phokinase (CPK) leve l , including muscle brain (MB) isozyme determin(cid:173)
`ations, were obtained in f ur patients who showed low voltage o n EKG.
`There was no e vidence of . ignificant pericardia! effusions or chang
`in
`ventricular function. Levels of PK MB i zyme did not increa. e.
`Three patient experie nced possible allergic reactions. Two of these had
`
`501
`
`IMMUNOGEN 2313, pg. 11
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Table 3. Clinical and laboratory findings in patients undergoing immunotoxin therapy
`
`Clinical Observations
`
`Malaise/fatigue
`Fever
`Tachycardia
`Decreased appetite
`Nausea
`Weight gain (> 5%)
`Myalgia
`Flush
`Death
`Pruritus
`
`Laboratory Observations
`
`Albumin decrease (.10%)
`Total serum protein decrease
`Low voltage on EKG
`Fibrinogen increase
`ESR increase
`Leukocytosis
`Calcium decrease
`SGOT increasec
`LDH increasec
`Thrombocytopenia ( < 50,000 platelets/cu mm)
`PTT prolongation
`C-Reactive protein increase
`Eosinophilia
`Creatinine increase
`Hemoconcentration
`Metabolic acidosis
`
`Number of Patients
`
`15
`14
`14
`12
`6
`6
`5
`2
`1"
`1
`
`20
`20
`16
`12/13b
`8/11
`8
`7
`6d
`2d
`2d
`2d
`2/2
`1 d
`ld
`1
`
`• Relationship to IT therapy not established (see text).
`b Denominator represents number of patients evaluated, if observations were not made in all
`22 patients.
`c Increase to two or more times the baseline value.
`d Some of the abnormal values were attributed to disease progression or other causes
`(see text).
`
`Reproduced with permission from Spitler, L.E. eta!. (1987) Therapy of Patients with Malignant
`Melanoma a Monoclonal Antimelanoma Antibody-Ricin A Chain Immunotoxin, Cancer
`Research .
`
`received prior murine antibody as a part of the previously mentioned
`radioimmunoimaging study, and one experienced the reaction after a 10-day
`course of infusions. In two patients, the reactions occurred during the
`infusion and consisted of facial flush and slight nausea. There was associated
`pruritus i.n one of these patient . In another patient , the reaction resembled
`ickness with eosinophilia. All reactions were mild.
`an atypical serum
`T hree of 22 patients in this study expired within two months of receiving
`JT therapy. Two patient had rapidly progressive meta ·tatic melanoma and
`
`502
`
`IMMUNOGEN 2313, pg. 12
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`· died as a result of their widespread disease. The third patient, who had
`a prior history of coronary artery di ea e with apparent unstable angina,
`demonstrated hypotension hemoconcentration , and a metabolic acidosis
`within 12 hours after the fourth do.e of IT (0.75 mg/kg/day). The patient
`initially responded to intravenou hydration , but he developed atrial fibrilla(cid:173)
`tion and had a cardiopulmonary arrest within 36 hours of his last dose of IT.
`Autopsy findings confirmed his past history of coronary artery disease but
`did not clarify the cause of death. The immediate cause of death was
`thought to be due to an arrhythmia. The relationship between IT adminis(cid:173)
`tration and this patient' ci ath is unknown.
`Encouraging clinical results were observed in this study. One patient had
`a complete response with disappearance of a pulmonary metastasis. The
`response is ongoing at 26 months. In addition, four patients had mixed
`responses, defined as a 50% or· greater reduction in area of one or more
`metastases, while concurrently one or more lesions increased in size or new
`le ion appeared after treatment. Although these do not meet the standard
`oncologic definition of an objective response, they are noteworthy in view of
`the proposed mechanism of action of the ITs, and they are presented as
`evidence of biologic effect. The duration of the tumor regression ranged
`from 2 to 17 m nths. The observed tumor regression were unusual in that
`lesions continued regressing for prolonged intervals following a single course
`of IT without any additional therapy. Five patients have shown stabilization
`of pulmonary metastases of 3 to 14 months duration.
`The history of a patient is described to illustrate the response observed.
`The patient is a 39-year-old white female who had a primary melanoma of
`the back in 1977. In 1983, the patient was found to have a right axillary
`lymph node metastasis which was resected. She was subsequently treated
`with Newcastle disease viral oncolysate. In July of 1984, she developed a
`metastasis to the breast that was excised. A recurrence of the breast lesion
`was excised one month later. She then underwent two cycles of chemother(cid:173)
`apy with bleomycin ulfate, vinblastine ulfate, hydroxyurea, and procarba(cid:173)
`zine hydrochloride. The last dose of chemotherapy wa administered on
`10/10/84. During the course of chemotherapy, a retrocardiac metastasis
`increased in size despite the quadruple drug chemotherapy. In November
`1984, she received IT therapy . The patient had radiographic evidence of
`tumor regression that continued over the ensuing months until the tumor
`became undetectable approximately eight months after IT administration.
`No new lesions have appeared, and she remains in complete remission 26
`months after a single course of IT (Figure 3).
`Survival in the 22 patients included in thi phase I study was compared
`with survival in a histork control group of comparable patients. The control
`group consisted of patients. who had bee n treated by the Eastern Coopera(cid:173)
`tion Oncology Group (ECOG). All were patients who had received previous
`chemotherapy, failed, and were then treated with a second therapy. The
`groups were comparable in having good performance status (0-1) at study
`
`503
`
`IMMUNOGEN 2313, pg. 13
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`a.
`
`Figure 3a. Chest CT scan of patient number 3 obtained after quadruple drug chemotherapy and
`immediately before IT therapy showing a retrocardiac mass that increased in size in the .face of
`chemotherapy.
`
`Figure 3b. Chest CT scan of patient number 3 obtained one year after a single course of IT
`therapy showing complete regression of the retrocardiac mass. Other studies showed no
`evidence of disease elsewhere. Reproduced, with permission, from Spitler, L.E. et al. (1987)
`Therapy of Patients with Malignant Melanoma Using a Monoclonal Anti-melanoma Antibody-
`·
`Ricin A Chain Immunotoxin. Cancer Research.
`
`504
`
`IMMUNOGEN 2313, pg. 14
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`... I
`
`'
`
`I \
`
`--
`
`--+-
`
`XOMAZYME-MEL (n = 21)
`HISTORIC CONTROL (n::: 30)
`p = .78
`
`~~
`~i
`'4
`•+ •+
`A- - - ·--- ~ ·- .. .. .. ...,. ........
`"'· ~ . --. -. -
`
`(!)
`z
`>
`> a:
`
`::::>
`(/)
`z
`0
`i=
`a:
`0
`a.
`0
`a:
`a.
`
`1.0
`
`0 .8
`
`0.6
`
`0.4
`
`0.2
`
`0 .0
`
`0
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`Figure 4. Survival of 22 patients included in the phase I trial of JT as compared with that of a
`similar historic control group of patients treated by ECOG.
`
`entry and in having lactic dehydrogenase valu s which were normal or nol
`m'ore than 25% above normal. Survival is compared from the time of the
`initiation fIT th erapy or the time of initiation of the econd chemotherapy.
`The results show th~tt there is no difference in surviva l between the e two
`thernpy i as
`groups, indicating that I
`ffective as currently available
`chemotherapy (Figure 4). Moreover, since the IT therapy is administered a
`a single five-day course, it i
`less disruptive of the patient' · life than i
`standard chemothe rapy.
`Tissue samples of metastatic lesions w re obtained from five patients
`within 24 hours of the IT infu ion. ln all these specimens, immunoperox(cid:173)
`idase. staining revealed inten e antiricin A chain reactivity with the