264
`
`Bioconjugate Chem. 1990, 1, 264-268
`
`Antitumor Activity of a Thioether-Linked Immunotoxin: OVB3-PE
`
`David FitzGerald,' Thierry Idziorek, Janendra K. Batra, Mark Willingham, and Ira Pastan
`Laboratory of Molecular Biology, National Cancer Institute, National Institutes of Health, 9000 Rockville Pike,
`Bethesda, Maryland 20892. Received April 6, 1990
`
`A thioether-linked immunotoxin was made between Pseudomonas exotoxin and the monoclonal antibody
`OVB3. This conjugate, OVB3-PE, was cytotoxic for the human ovarium cancer cell line OVCAR-3
`(ID of 2.5 X lo-'* M) and it was therefore tested for antitumor activity in a nude mouse model of ova-
`rian cancer. This model employs the injection of a lethal number of OVCAR-3 cells into the perito-
`neal cavity of nude mice. When 0.2-1 pg of OVB3-PE was injected intraperitoneally on three successive
`days beginning 3-5 days after OVCAR-3 cell implantation, the survival of the tumor-bearing mice was
`increased 2-4-fold compared to that of untreated control mice. Median survival times for control mice
`ranged from 44 to 50 days while survival times of 150 days or greater were seen in mice treated with
`OVB3-PE. When OVB3-PE administration was delayed until 2-4 weeks after tumor cell implantation,
`OVB3-PE treatment also showed antitumor activity, but the duration of survival was less than with
`the early treatments. OVB3-PE was also cytotoxic for MCF-7 breast carcinoma cells, HT-29 colon
`carcinoma cells, and A431 epidermoid carcinoma cells.
`
`INTRODUCTION
`Ovarian cancer is an important cause of death in women.
`New approaches, such as immunotoxin therapy (1,2), are
`needed because ovarian cancer is frequently resistant to
`chemotherapy and because it often spreads to the peri-
`toneal cavity before it is discovered so that surgery cannot
`remove all the tumor. In order to treat this disorder, we
`have developed a monoclonal antibody, OVB3, that has
`reacted with all adenocarcinomas of the ovary examined
`so far (25/25) as well as some adenocarcinomas of the
`breast and colon (3). This antibody was coupled previously
`to Pseudomonas exotoxin (PE)' by a disulfide bond (3,
`4 ) and the resulting conjugate, OVB&PE, was shown to
`kill ovarian cancer cells in tissue culture and to prolong
`the life of immunodeficient mice with human ovarian
`cancer implants (3). Here we extend our original findings
`and describe a method to conjugate P E to OVB3 by a
`thioether bond. Thioether conjugates may have advantages
`in animals over disulfide conjugates since the carbon-
`sulfur bond is likely to be more stable in vivo.
`In this study we have prepared a thioether conjugate
`of OVB3 and P E and evaluated its cell-killing activity in
`tissue culture and in a tumor model. In vitro it inhibited
`protein synthesis with an ID50 of 2-3 X 10-l2 M. The an-
`titumor activity of the thioether conjugate was tested in
`the OVCAR-3 nude mouse model of human ovarian cancer.
`Mice implanted with 25 X lo6 OVCAR3 cells die about
`40 days after tumor implantation with massive ascites (5),
`but mice treated 3-5 days after implantation with OVB3-
`PE survive to 100 days or longer, with a few animals living
`for over 175 days. Furthermore, administration of OVB3-
`PE at day 26 after tumor implantation, when significant
`ascites and a large tumor load is present, causes the
`disappearance of ascites and prolongation of the life of the
`tumor-bearing animals.
`
`1 Abbreviations: PE = Pseudomonas exotoxin, MST = median
`survival time, ip = intraperitoneally, H and E = hematoxylin and
`eosin, SMCC = succinimidyl4-(N-maleimidomethyl)cyclohexane-
`l-carboxylate, SDS-PAGE = sodium dodecyl sulfate-polyac-
`rylamide gel electrophoresis.
`
`Because OVB3-PE has been shown to be active against
`human ovarian cancer growing in mice, a phase I study
`using OVB3-PE has begun in women. The information
`contained in this paper was used to plan the initial protocol.
`
`EXPERIMENTAL PROCEDURES
`Reagents. Purified P E was purchased from Swiss
`Serum and Vaccine Institute, Berne, Switzerland. The
`OVB3 antibody was produced by in vitro culture at Da-
`mon Biotech. Purified antiTac antibody was a gift from
`T. Waldmann (NCI). Cross-linking and protein-
`modification reagents were obtained from Pierce Chemical
`co.
`Construction of OVB3-PE by Thioether Linkage.
`To couple PE to OVB3 by a thioether bond, P E was first
`reacted with sulfo-SMPB (sulfosuccinimidyl 4 - ( p -
`maleimidopheny1)butyrate). Typically PE at 2-3 mg/
`mL in borate buffer, pH 8.0, was reacted with a 3-fold
`molar excess of sulfo-SMPB. Immediately prior to use
`sulfo-SMPB was dissolved in dimethyl formamide (DMF)
`to a final concentration of 5-10 mg/mL (as appropriate).
`Usually 5 pL or less of the sulfo-SMPB solution was added
`per milliliter of P E solution. The reaction, which
`proceeded a t 37 "C for 30 min, was quenched by the
`addition of excess glycine. PE modified in this manner
`was then resolved from low molecular weight reactants by
`HPLC gel-filtration chromatography (TSK-250, Bio-
`Rad, 21.5 X 600 mm). The running buffer for this column
`was 0.2 M NaP04, 1 mM EGTA, pH 7.0. With this
`chromatography system, both PE and chemically modified
`PE elute at approximately 140 mL. OVB3 was reacted
`for 1 h at 37 "C with a $fold molar excess of P-iminothi-
`olane in 0.2 M NaP04, 1 mM EGTA at pH 8.0. Reaction
`with 2-iminothiolane introduced approximately 0.9 mol
`of SH per mol of OVB3. This was determined with 5,5'-
`dithiobis(2-nitrobenzoic acid) (4). As with PE, excess gly-
`cine was used to quench the reaction. Chemically modified
`antibody was recovered by gel filtration on a TSK-250 (21.5
`x 600 mm) column. The running buffer was the same as
`described above and the antibody eluted at 117 mL.
`Finally, PE, having a reactive maleimide group, was mixed
`
`Not subject to U.S. Copyright. Published 1990 by American Chemical Society
`
`IMMUNOGEN 2273, pg. 1
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Antitumor Activity of Thioether-Linked
`
`Immunotoxin
`
`1
`
`2
`
`
`
`3
`
`
`
`OVB3-PE -
`'
`OVB3 -
`
`O
`
`2
`
`PE -
`
`Figure 1. SDS-PAGE run under nonreducing conditions, using
`10% acrylamide: lane l,OVB3-PE (made by thioether linkage);
`lane 2,OVB3; and lane 3, PE.
`
`with OVB3-SH. The two proteins were allowed to react
`a t room temperature overnight and then further purified
`by gel filtration. The 1:l conjugate which eluted at 105
`mL was recovered and used in all subsequent experiments
`(see Figure 1 for SDS-page analysis of the final product).
`Protein concentration for the 1:l conjugate was determined
`with the following conversion factor: absorbance of 1.3 at
`280 nm was equivalent to 1.0 mg/mL of total protein.
`AntiTac-PE was also made by thioether linkage and
`served as a control immunotoxin. AntiTaePE (thioether)
`was made by using a similar protocol to the one used for
`the thioether-linked OVB3 immunotoxin. However, the
`one-to-one conjugate was purified by using a different
`separation strategy. The final reaction mixture was applied
`to a MonoQ (Pharmacia/LKB) column and eluted with
`a linear NaCl gradient. This separated unreacted antibody
`(which eluted a t approximately 0.2 M NaC1) from a
`complex mixture that contained the immunotoxin and un-
`reacted PE (this complex mixture eluted between 0.25 and
`0.28 M NaC1). The one-to-one conjugate was then
`separated from high molecular weight material and un-
`reacted PE on a HPLC sizing column. To ensure the
`conjugate was made correctly, it was tested on target
`HUT102 cells and had an ID50 value of 0.5 ng/mL.
`Tissue Culture. OVCAR-3, MCF-7, A431, and HT-
`29 cells were maintained in DMEM, 10% FCS. HUT-
`102 cells were maintained in RPMI 1640,10% FCS. For
`cytotoxicity studies, cells were plated at 1-2 x lo5 cells/
`well in 24-well Costar tissue-culture plates (6). Adherent
`cells were seeded approximately 24 h before the addition
`of immunotoxin. HUT-102 cells were washed 3 or 4 times
`with tissue-culture medium and used the same day.
`Cytotoxic Activity. Cytotoxic activity was determined
`by measuring inhibition of protein synthesis. Immuno-
`toxins were added to cells in culture for 20 h. At the end
`of this period, [3H]leucine a t 10 pCi/mL was added for a
`further hour. Cells were washed with PBS, solubilized with
`0.1 M NaOH, precipitated with excess TCA, and counted.
`Experiments were done with triplicate samples and usually
`repeated on 3 or 4 separate occasions.
`Animal Experiments. B75 female mice 6-9 weeks old
`were used to grow OVCAR-3 ascites tumors. Usually, 25
`X lo6 washed cells were injected into recipient mice on day
`1. Untreated, these mice died from their tumor burden
`40-50 days later (5,7). To prevent undue suffering, mice
`beginning to show distress from their tumor burden were
`killed prematurely. Tumor-bearing mice were treated by
`
`Bioconjugate Chem., Vol. 1, No. 4,
`OVCAR-3 CELLS
`
`1990 265
`
`lo0l
`
`I
`100 -
`
`80 80 -
`
`OVB3-PE
`
`+ XS OVB3 + XS OVB3
`AntiTAC-PE
`
`20
`
`0 1- . . ......, . := ...., . .......,
`
`. . . J
`
`I
`
`I
`
`100
`
`1000
`
`IO
`ng/ml
`Figure 2. Inhibition of protein synthesis by OVB3-PE. Various
`concentrations of OVB3-PE (two individual experiments are
`shown) or antiTac-PE shown as total protein in ng/mL were
`added to OVCAR-3 cells for approximately 20 h. Following this,
`[3H]leucine was added to individual wells to a final concentration
`of 10 pCi/mL for 1 h. Inhibition of protein synthesis was
`determined by measuring the radioactivity in TCA-precipita-
`ble material in immunotoxin-treated wells compared to that of
`untreated wells. Parallel experiments were conducted where
`excess OVB3 (75 pg/mL) was added to cells immediately prior
`to the addition of the OVB3-PE.
`Table I. Cytotoxic Activity of OVB3-PE for Various
`Adenocarcinomas
`OVB3-PE
`OVB3-PE
`IDm,= ng/mL
`cell line
`IDm,= ng/mL
`cell line
`A431
`OVCAR-3
`0.5
`0.4
`HT-29
`0.3
`MCF-7
`0.5
`a ID& were assessed after an overnight incubation with immu-
`notoxin.
`
`ip administration of immunotoxins. The early-treatment
`protocols (experiments 1-5) involved giving injections of
`OVB3-PE beginning on days 3-5 and various numbers of
`individual injections were administered. For the late-
`treatment protocol (experiment 6), immunotoxin
`treatments were initiated 19-33 days after the injection
`of the tumor cells. In the case of late treatments, five
`injections were given approximately every other day. All
`immunotoxin preparations were formulated in normal
`saline/human serum albumin (10 mg/mL), filter sterilized
`(0.22 pM, GV-Millipore), and injected ip in a volume of
`0.5 mL.
`
`RESULTS
`OVBB-PE (Thioether): Evaluation of Cytotoxic
`Activity. To assess the cytotoxic potential of a thioether-
`linked immunotoxin, a one-to-one conjugate was made
`between the monoclonal antibody OVB3 and Pseudomo-
`nus exotoxin (PE). OVB3-PE was purified by gel-
`filtration chromatography and evaluated for its cyto-
`toxic activity by adding various concentrations of the im-
`munotoxin to OVCAR-3 cells. After a 20-h incubation,
`inhibition of protein synthesis was determined. As shown
`in Figure 2 and Table I, the ID50 was 0.5 ng/mL (2.5 X
`10-l2 M). When the control immunotoxin, antiTac-PE,
`which does not bind OVCAR-3 cells was added, an ID50
`of greater than 1000 ng/mL was noted. Also excess OVB3
`competed a t least 100-fold for the cytotoxic activity of
`OVB3-PE. OVB3-PE was also assessed for cytotoxic
`activity against other cancer cell lines and for its antitu-
`mor activity against OVCAR-3 tumors growing in nude
`mice (see below).
`Antitumor Activity Mediated by Early Treatment
`with OVBB-PE. The thioether conjugate of OVB3-PE
`was tested for antitumor activity in the same nude mouse
`model of human ovarian cancer previously used to test the
`
`IMMUNOGEN 2273, pg. 2
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`266 Bioconjugate Chem., Vol. 1, No. 4, 1990
`
`FitzGerald et al.
`
`group
`
`day of
`injection
`
`Table 11. Median Survival Times (MST) of
`Tumor-Bearing Mice Treated with OVBI-PE
`OVB3-PE, MST,
`no. of
`rglinjection
`days
`injections
`Experiment 1 (n = 10)
`50
`3
`121
`3
`158
`3
`193
`3
`8"
`3
`Experiment 2 (n = 10)
`44
`3
`87
`3
`87
`3
`42"
`3
`Experiment 3 (n = 10)
`48
`3
`56.5
`3
`70.0
`3
`64.5
`3
`69.5
`3
`91.5b
`3
`Experiment 4 (n = 7)
`53
`3
`81
`3
`83
`3
`1 P
`3
`83
`7
`118
`7
`7
`10"
`Experiment 5 ( n = 10)
`2, 3, 5, 7, 10
`46
`5
`1
`0
`1.0
`2
`2, 3, 5, 7, 10
`90c
`5
`1.0
`3
`5, 7, 10, 12, 14
`90
`5
`a At least 50% of mice in this group died from dose-related toxicity.
`Three mice alive past day 160. Four mice alive past day 140.
`Table 111. Survival of Tumor-Bearing Mice Receiving
`Early and Late Treatments with OVBI-PE
`experiment 6 OVB3-PE, MST,
`no. of
`(n = 10); group diniection daw
`iniections
`1
`37
`5
`0
`2
`1.0
`156
`5
`3
`1.0
`63
`5
`4
`1.0
`63
`5
`
`3, 4, 5
`3, 4, 5
`3, 4, 5
`3, 4, 5
`3, 4, 5
`
`5, 6, 7
`5, 6, 7
`5, 6, 7
`5 , 6 , 7
`
`5, 6, 7
`5, 6, 7
`5, 6, 7
`5, 6, 7
`5, 6, 7
`5, 6, 7
`
`5, 6, 7
`5, 6, 7
`5, 6, 7
`5, 6, 7
`5, 6, 7, 9, 10, 11, 12
`5, 6, 7, 9, 10, 11, 12
`5, 6, 7, 9, 10, 11, 12
`
`day of
`iniection
`5,6,7,10,11
`5, 6, 7, 10, 11
`19, 20, 21, 24, 25
`26, 27, 28, 31, 32
`
`1
`2
`3
`4
`5
`
`1
`2
`3
`4
`
`1
`2
`3
`4
`5
`6
`
`1
`2
`3
`4
`5
`6
`7
`
`0
`0.2
`0.5
`1.0
`2.0
`
`0
`0.2
`0.5
`2.0
`
`0
`0.05
`0.1
`0.2
`0.5
`1.0
`
`0
`0.2
`0.5
`2.0
`0.2
`0.5
`2.0
`
`0
`200
`io0
`DAYS
`Figure 3. Inhibition of tumor growth by OVB3-PE. OVCAR-3
`(25 X 1Oe) cells were injected into the peritoneal cavity of B75
`athymic nude mice on day 1. On days 3,4, and 5, various amounts
`of OVB3-PE (ranging from 0.2 to 2.0 pg) were injected ip in 0.5
`mL of sterile normal saline, 10 mg/mL of human serum albumin.
`The survival of each group of mice was then followed. There were
`10 mice in each group.
`antitumor activity of the OVB3-PE disulfide conjugate
`(3) and various other immunotoxins (7,B). In this model,
`25 X 106 OVCAR-3 cells were injected intraperitoneally
`into athymic nude mice on day 1, and immunotoxin
`treatment was begun on day 3 or later. The immuno-
`toxin was also given by the ip route. The mice were then
`monitored for immunotoxin-mediated antitumor effects.
`The OVCAR-3 tumor was lethal and untreated mice
`usually died between days 40 and 50. At death these mice
`had one or two large tumor masses, studding of the
`abdominal cavity with small tumor implants, and 5-7 mL
`of ascites fluid containing approximately 500 X lo6
`unattached tumor cells.
`When OVB3-PE was injected on days 3,4, and 5 after
`the injection of the tumor cells, there was a large increase
`in the duration of survival compared to that of the un-
`treated animals. The duration of increased survival was
`generally related to the amount of immunotoxin given
`(Figure 3 and Table 11, experiment 1). In this experiment,
`the median survival time (MST) was increased from 50
`days in control mice to 121 days for mice receiving three
`injections of 0.2 pg, to 158 days for mice receiving 0.5 pg,
`and to 193 days for mice receiving 1.0 pg of immuno-
`toxin. The injection of three doses of 1.0 pg of OVB3-
`P E was the maximum tolerated daily dose. Injection of
`2.0 pg per day for 3 days was lethal for at least 50% of
`the mice.
`Two experiments (nos. 2 and 3 of Table 11) were carried
`out where OVB3-PE was first administered on day five
`and injections were given on days 5,6, and 7. In both cases
`a significant antitumor effect was seen (Table 11).
`Additional experiments were then carried out to determine
`(1) whether a greater number of immunotoxin injections
`would increase median survival and (2) whether beginning
`OVB3-PE treatment 2 days after tumor inoculation would
`be more effective than beginning 5 days after inoculation.
`The results of experiment 4 indicated that seven injections
`were no more effective than three when 0.2 pg of immu-
`notoxin were given but were marginally better when 0.5
`pg were administered. Experiment 5 indicated that at early
`times after tumor inoculation there was little or no
`differences in survival when treatment was initiated 2 days
`after inoculation or 5 days after inoculation.
`Antitumor Activity Mediated by Late Treatment
`with OVBS-PE. To determine if it were possible to
`achieve antitumor effects even when the OVCAR-3 tumors
`had reached a large size and significant ascites was present,
`the first injection of OVB3-PE was delayed by 2-4 weeks
`after the injection of tumor cells. In one experiment (Table
`
`111, experiment 6), five daily injections of 1 pg of OVB3-
`PE were administered to tumor-bearing mice beginning
`on day 19 or 26. As a control for early treatment, OVB3-
`P E (1 pg) was also administered starting on day 5. The
`results indicated that the administration of OVB3-PE
`beginning on day 19 or 26 resulted in both antitumor
`activity and increased survival compared to control mice
`receiving only diluent (Figure 4 and Table 111). The
`increase in survival time for mice beginning treatment on
`day 19 or 26 was approximately 26 days. In contrast, the
`increase in survival for mice beginning treatment on day
`5 was greater than 100 days.
`To follow tumor growth, each mouse was weighed and
`scored for abdominal girth on a weekly basis. An increase
`in body weight began two weeks after injection of the tumor
`cells. By day 26, most of the mice had gained 10 g and
`had an abdominal girth score of 3-4+ (scale of 0-4+). The
`increase in body weight was due to the build-up of as-
`cites fluid within the abdominal cavity. The body weight
`of the untreated mice continued to increase until death
`(Figure 4A). In the group of mice that received OVB3-
`PE (1 pglinjection X 5 between days 26 and 32) there was
`a rapid decrease in body weight (Figure 4B), a reduction
`in abdominal girth, and a 26-day prolongation of survival
`(Figure 4C). Similar survival data to that shown in Figure
`
`IMMUNOGEN 2273, pg. 3
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Antitumor Activity of Thioether-Linked Immunotoxin
`A
`
`50
`
`45
`
`.
`
`, " ,
`.10
`
`I
`0
`
`.
`
`I
`10
`
`, I ,
`.10
`
`.
`
`I
`0
`
`.
`
`I
`10
`
`.
`
`,
`
`I
`
`30
`
`I
`
`40
`
`'
`
`I
`
`5 0
`
`,
`
`,
`
`I
`
`30
`
`I
`
`40
`
`1
`
`I
`
`5 0
`
`.
`
`I
`
`20
`DAYS
`
`.
`
`I
`
`20
`DAYS
`
`"
`
`,
`0
`
`.
`
`.
`
`.
`
`.
`
`,
`20
`
`,
`4 0
`
`.
`
`.
`
`I
`.
`60
`
`.
`
`,
`.
`80
`
`.
`
`,
`
`100
`
`Days
`Figure 4. Antitumor effect of OVBB-PE on mice bearing large
`tumor burden with ascites. OVCAR-3 (25 X 106) cells were
`injected into the peritoneal cavity on day 1. Beginning on day
`26, five individual injections (see Table I11 for exact days) of
`OVB3-PE were given to mice with prominent distention from
`ascites accumulation. As a control for early treatment, one group
`of mice was treated beginning on day 5. A shows the progressive
`increase in weight of control mice that were not treated with im-
`munotoxin (data for individual mice are shown). B shows the
`weight of mice treated with OVBB-PE beginning on day 26. The
`arrow indicates the day when immunotoxin treatment was
`initiated. C shows the survival curves for mice treated with
`OVBB-PE (1 pg/injection) beginning on day 26 (m) or day 5 ( e )
`or untreated (a). The arrows indicate the days when immuno-
`toxin treatment was initiated. Additional details are provided
`in Table I11 (experiment 6).
`4C were obtained for the mice that were treated with
`OVB3-PE between days 19 and 25 (Table 111).
`An additional experiment was performed to determine
`the fate of large tumors treated with cytoreductive doses
`of OVB3-PE. In this experiment, the tumors grew slower
`than in the previous experiment so treatment was delayed
`
`Bioconjugate Chem., Vol. 1, No. 4, 1990 267
`until day 33. To estimate the tumor burden at the time
`when OVB3-PE was injected, three mice were sacrificed
`on day 33. By aspirating the peritoneal cavity an average
`of 4 X los OVCAR-3 tumor cells per mouse was recovered
`from ascites fluid. In addition, there was a single solid
`tumor mass in each mouse (approximately 1.0 X 1.0 cm
`in size). A comparison that was made with the tumor
`burden present on days 0-5 showed there were roughly 50-
`100 times more tumor cells present on day 33 than on days
`0-5. Five immunotoxin injections were given beginning
`on day 33. To assess the effect of this treatment, two mice
`from this treatment group were killed and examined for
`evidence of viable tumor 4 days after the final injection
`of immunotoxin. Gross anatomical examination revealed
`that the immunotoxin treatment eliminated the ascites
`fluid and unattached tumor cells. However, the solid tumor
`mass remained evident in both mice and the tumor had
`similar dimensions to the pretreatment solid tumor. This
`posttreatment tumor mass was examined microscopically
`following fixation and H and E staining. The tumor cells
`appeared viable.
`Cell Killing of O V B I P E on Breast and Colon
`Cancer Cells. The reactivity of OVB3 for various human
`tumor samples was reported previously and has been
`extended here (3). With use of immunocytochemistry on
`frozen sections of cancer specimens, OVB3 has been found
`to react with 25/25 ovarian carcinomas and approximately
`25% of breast cancers and colon tumors. Because of this
`reactivity with tumors other than ovarian, we investigated
`the cytotoxicity of OVB3-PE on cell lines derived from
`breast and colon tumors. OVB3-PE was found to have
`an ID50 of less than 1 ng/mL for both a breast cancer cell
`line, MCF-7, and a colon cancer cell line, HT-29 (Table
`I). Thus OVB3 has potential for use in the treatment of
`other adenocarcinomas besides ovarian cancers. OVB3-
`PE was also tested on epidermoid carcinomas and found
`to have potent cell-killing activity for A431 cells (Table
`I) and to a lesser extent, KB cells (data not shown).
`Recently, OVB3-PE was shown to have antitumor activity
`against HT-29 tumors (9).
`
`DISCUSSION
`We show here that it is possible to make a very potent
`immunotoxin by conjugating native P E to OVB3 by a
`thioether linkage.
`OVB3 was originally selected as an ovarian-specific mon-
`oclonal antibody. We have examined many tumors for
`reactivity with OVB3 and found t h a t OVB3 has
`preferential reactivity for ovarian cancer (25/25), but it
`also reacts with a significant percentage of adenocarci-
`nomas of the breast and colon. As shown in Table I, OVB3
`is also active against a breast (MCF-7), a colon (HT-29),
`and an epidermoid carcinoma (A431) cell line. While
`OVB3-PE inhibited these cell lines with an ID50 of 0.5 ng/
`mL, a 200-fold higher concentration was needed to kill
`HUT-102 cells (data not shown). Thus, the cytotoxicity
`of OVB3-PE was relatively specific for cells bearing the
`OVB3 antigen.
`Immunotoxin therapy has been shown to be effective
`in treating intraperitoneal tumor models of human cancer
`(7-10). Here we show in some instances that OVB3-PE
`can increase survival of nude mice bearing OVCAR-3
`tumors by greater than 100 days. The injection of
`irrelevant immunotoxins such as antiTac-PE has no an-
`titumor activity in this model (7). For OVB&PE, 1 pg
`per injection gave the most pronounced antitumor effects.
`The median survival times of 193 days in experiment 1
`and 156 days in experiment 6 represent some of the longest
`
`IMMUNOGEN 2273, pg. 4
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Bioconjugate Chem., Vol. 1, No. 4, 1990
`
`268
`survival times reported to date for an immunotoxin-
`mediated antitumor effect. We have also shown that
`OVB3-PE prolongs the life of tumor-bearing mice when
`given 19-33 days after OVCAR-3 implantation at a time
`when ascites had developed and large tumor implants were
`present.
`In every experiment where OVB3-PE was administered
`at a dose of 0.5 or 1.0 pg/injection, there was a significant
`increase in the survival of mice bearing OVCAR-3 tumors.
`However, from experiment to experiment, we have noted
`some variability in the median survival times of the various
`treatment groups. For example, the MST for the 0.5 pg
`per injection dose ranged from 158 days in experiment 1
`to 69 days in experiment 3 and for 1.0 pg from 193 days
`in experiment 1 to 91 days in experiment 3. We have noted
`intermediate MSTs in other experiments. Various lots of
`immunotoxins were checked for changes in cell-killing
`activity upon storage. No systemic loss of activity was
`noted. We do not have an explanation for the variability
`in the mouse experiments except to suggest that animal-
`to-animal variation may be greater than the small
`variations seen in tissue-culture systems.
`The administration of OVB3-PE 2-4 weeks after
`initiating tumor growth was carried out to test for anti-
`tumor activity against a large established tumor. OVB3-
`PE was effective when injections were begun either 19,26,
`or 33 days after the injection of tumor cells. Antitumor
`activity was characterized by increased survival time,
`decreased body weight, diminished abdominal girth, and
`a reduced number of tumor cells recovered immediately
`posttreatment. The antitumor effect was most pronounced
`against ascites tumor cells. In two mice that were au-
`topsied 5 days posttreatment, ascites tumor cells were not
`evident. However, in those same mice, the solid tumor was
`of similar size to tumors examined immediately prior to
`treatment. It would appear that the large solid tumor was
`poorly permeable to the immunotoxin.
`Bjorn et al. had reported that it was possible to produce
`active immunotoxins when P E was thioether-linked to
`various monoclonal antibodies (11). In this report, we have
`described a novel method to make thioether-linked PE im-
`munotoxins which is different from that of Bjorn et al. By
`first reacting PE with sulfo-SMPB, and OVB3 with 2-im-
`inothiolane we have made a very potent immunotoxin with
`an IDSO of 2.5 X 10-l2 M. In more recent experiments we
`have used SMCC in place of sulfo-SMPB and shown no
`difference in conjugate potency (data not shown). In this
`report, only data using native PE coupled to OVB3 is
`presented. We have made OVB3 conjugates with PE40,
`a recombinant form of P E lacking the toxin’s binding
`domain, but these showed little or no cytotoxicity for cells
`and were not pursued further (data not shown).
`The dosing schedule used in these experiments has been
`designed with certain clinical parameters in mind. Patients
`receiving PE conjugated to antibodies have been found to
`develop neutralizing antibodies to PE 10-12 days after the
`initial immunotoxin injection. Because of this, our
`injection protocol did not span more than 2 weeks. Thus,
`we achieved antitumor responses with an injection schedule
`that could be directly applied to patient treatment.
`Likewise, after the initial evidence that thioether conjugates
`
`FitzGerald et at.
`
`mediated an antitumor response when the immunotoxin
`was given 3-5 days after tumor implantation, we decided
`to treat more advanced tumors. While it is difficult to
`relate the staging of human disease to mouse tumor load,
`it was clear that we could achieve a significant antitu-
`mor effect when the tumor burden had increased by a
`minimum of 50-fold. OVB3-PE gave an antitumor effect
`in the dose range of 5-50 pglkg in mice when administered
`a t early times after tumor implantation. When
`administered after 3 weeks of tumor growth, 50 pg/kg also
`gave a significant antitumor response. At 100 pg/kg
`OVB3-PE was often lethal. Mice died 24-72 h after
`injection of severe liver toxicity.
`ACKNOWLEDGMENT
`Thanks to E. Lovelace, A. Rutherford, and M. Hazen
`for excellent technical assistance, t o S. Neal for
`photography, and to A. Schombert, J. Evans, and A. Gad-
`dis for typing the manuscript.
`
`LITERATURE CITED
`(1) Vitetta, E. S., Fulton, R. J., May, R. D., Till, M., and Uhr,
`J. W. (1987) Redesigning nature’s poisons to create anti-
`tumor reagents. Science 238, 1098-1104.
`(2) Pastan, I., Willingham, M. C., and FitzGerald, D. J. (1986)
`Immunotoxins. Cell 47, 641-648.
`(3) Willingham, M. C., FitzGerald, D. J., and Pastan, I. (1987)
`Pseudomonas exotoxin coupled to a monoclonal antibody
`against ovarian cancer inhibits the growth of human ovarian
`cancer cells in a mouse model. Proc. Natl. Acad. Sci. U.S.A.
`84,2474-2478.
`(4) FitzGerald, D. J. P. (1987) Construction of immunotoxins
`using Pseudomonas exotoxin A. Methods Enzymol. 151,139-
`145.
`(5) Hamilton, T. C., Young, R. C., Louie, K. G., Behrens, B. C.,
`McCoy, W. M., Grotzinger, K. R., and Ozols, R. F. (1984)
`Characterization of a xenograft model of human ovarian
`carcinoma which produces ascites and intraabdominal carci-
`nomatosis in mice. Cancer. Res. 44, 5286-5290.
`(6) Pirker, R., FitzGerald, D. J. P., Hamilton, T. C., Ozols, R.
`F., Willingham, M. C., and Pastan, I. (1985) Anti-transferrin
`receptor antibody linked to Pseudomonas exotoxin as a model
`immunotoxin in human ovarian carcinoma cell lines. Cancer
`Res. 45, 751-757.
`(7) FitzGerald, D. J., Willingham, M. C., and Pastan, I. (1986)
`Antitumor effects of an immunotoxin made with Pseudomo-
`nas exotoxin in a nude mouse model of human ovarian cancer.
`Proc. Natl. Acad. Sci. U.S.A. 83, 6627-6630.
`(8) FitzGerald, D. J., Bjorn, M. J., Ferris, R. J., Winkelhake, J.
`L., Frankel, A. E., Hamilton, T. C., Ozols, R. J., Willingham,
`M. C., and Pastan, I. (1987) Antitumor activity of an immu-
`notoxin in a nude mouse model of human ovarian cancer.
`Cancer Res. 47, 1407-1410.
`(9) Pearson, J. W., FitzGerald, D. J. P., Willingham, M. C., Wil-
`trout, R. H., Pastan, I., and Longo, D. L. (1989) Chemoim-
`munotoxin therapy against a human colon tumor (HT-29) xe-
`nografted into nude mice. Cancer Res. 49,3562-3567.
`(10) Griffin, T. W., Richardson, C., Houston, L. L., LePage, D.,
`Bogden, A., and Raso, V. (1987) Antitumor activity of in-
`traperitoneal immunotoxins in a nude mouse model of human
`malignant mesothialoma. Cancer Res. 47, 4266-4270.
`(11) Bjorn, M. J., Groetsma, G., and Scalapino, L. (1986)
`Antibody-Pseudomonas exotoxin A conjugates to human breast
`cancer cells in vitro. Cancer Res. 46, 3262-3267.
`
`IMMUNOGEN 2273, pg. 5
`Phigenix v. Immunogen
`IPR2014-00676
`
`