7
[CANCER RESEARCH 44, 1002-1007, March 1984] Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti- Epidermal Growth Factor Receptor Monoclonal Antibodies1 Hideo Masui,2 Tomoyuki Kawamoto, J. Denry Sato,3 Bonnie Wolf, Gordon Sato,4 and John Mendelsohn5 Cancer Center, 0-058, University of California, San Diego, La Jolla, California 92093 ABSTRACT Monoclonal antibodies (MoAbs) were raised against epidermal growth factor (EGF) receptors on a human epidermoid carcinoma cell line, A431. Administration of anti-EGF receptor MoAbs inhib ited tumor formation in athymic mice by A431 cells and by another epidermal carcinoma cell line, 1222. When one of the same MoAbs was used in therapy against Li-7 (a human hepa- toma) and HeLa cells (a cervical carcinoma), tumor growth was not affected. The number of EGF receptors on A431 cells was about 100-fold higher than on T222, Li-7, and HeLa cells, sug gesting that the number of EGF receptors may not be an important determinant in suppressing tumor growth. Three anti- EGF receptor MoAbs were used in the present studies. MoAbs 528 (immunoglobulin G2a) and 225 (¡mmunoglobulin G1) are capable of competing with EGF for receptor binding and inhibit proliferation of A431 cells in culture. The other MoAb, 455 (immunoglobulin G1 ), is incapable of blocking the binding of EGF to its receptors and has no effect on the proliferation of cultured A431 cells. All three MoAbs inhibited A431 tumor growth in athymic mice, indicating that the antibody isotype and the site of binding on the EGF receptor are not the determinants of antipro- liferative activity in vivo. The observation that MoAb against the receptor for EGF is cytostatic rather than cytocidal in vitro against A431 cells, yet completely prevents tumor growth In vivo, sug gests that some host animal responses also may be involved in the antitumor effect. MoAbs against growth factor receptors could provide useful immunotherapeutic agents. INTRODUCTION There have been many attempts at passive immunotherapy of malignant tumors, but there are only a few reports in which the administration of conventional antisera inhibited tumor growth in experimental animals or in cancer patients (26). Thus, passive immunotherapy has not been established as a treatment for human cancer. The major reasons are the difficulties in identifying tumor-associated antigens and in preparing homogeneous anti body in large quantities by conventional methods. Development of the hybridoma technique by Kohler and Mil- stein (14, 15) has overcome these difficulties, and experimental cancer immunotherapy has been revived in recent years using MoAbs.6 Several recent reports demonstrated that MoAbs inhibit Received May 31, 1983; accepted December 2, 1983. 1This research is conducted in part by the Clayton Foundation for Research, California Division. 2 Recipient of Grant CA 23052. To whom requests for reprints should be addressed. 3 Present address: Molecular Genetics, City of Hope Medical Center, 1500 E. Duarte Road, Duarte, CA 91010. 4 Recipient of NIH Grants CA 33397, GM 17702, and CA 19731. 5 Recipient of Grant CA 23052. Clayton Foundation Investigator. 6 The abbreviations used are: MoAb, monoclonal antibody; EGF, epidermal the growth of human tumor xenografts in athymic mice (8, 9,16, 27). Furthermore, the results of several clinical studies on the treatment of leukemia and lymphoma using MoAbs have been published, but in these cases, clinically significant results have not been obtained (17,18, 20, 23-25), with one exception. This was a report on the treatment of a human B-cell lymphoma with anti-idiotype antibody, in which the patient has been free from disease for more than 1.5 years (19). In most of these studies, MoAbs against tumor-associated antigens have been used for treatment. Another type of MoAb which might be used for cancer immunotherapy is an antibody against plasma membrane receptors for growth factors. It is well known that the proliferation of tumor cells in culture is controlled by various growth factors, and a similar control mechanism is postulated for the control of tumor cell growth in vivo (2). Therefore, MoAbs against growth factor receptors, which could block access of growth factors to their receptors, may provide useful therapeutic agents. Recently, Trowbridge and Domingo (29) reported that treatment with anti-transferrin receptor MoAb can inhibit tumor formation by a human melanoma cell line in athymic mice. A431 cells, a human epidermoid carcinoma cell line, express an unusually large number of EGF receptors on the cell surface membrane (1 to 3 x 106/cell) (4, 7), and addition of EGF to the culture medium inhibits the proliferation of these cells in culture (1, 6). We have developed MoAbs against the EGF receptor using partially purified EGF receptors from A431 cells as antigen (13). In this paper, we report that the administration of anti-EGF receptor MoAbs inhibits tumor formation in athymic mice by 2 human epidermoid carcinomas, A431 and T222. However, growth of 2 other xenografts is not affected. MATERIALS AND METHODS Animals. BALB/c athymic mice 6 to 8 weeks of age were used in the present studies. These mice were bred and maintained at the Athymic Mouse Colony Facility at the University of California, San Diego. The procedures for breeding and maintenance have been published previ ously (22). Human Tumor Cells. A431, T222, Li-7, and HeLa cells were main tained in a mixture of DME and Ham's F-12 medium (Grand Island Biological Co.), containing 15 mw 4-(2-hydroxyethyl)-1-piperazineeth- anesulfonic acid buffer and 5% newborn calf serum, at 37°C in 5% CO2:95% air atmosphere. The Li-7 human hepatoma tumor line was established originally at the Cancer Center Research Institute, Tokyo, Japan (10). This tumor was obtained after 50 passages and has been transplanted serially in this facility for 22 passages over the past 3 years. The human lung epidermoid carcinoma (T222) was established in our laboratory by inoculating fresh tumor tissue from surgery and was used after 47 to 50 passages in the growth factor; DME, Dulbecco's modified Eagle's medium; PBS, phosphate-buff ered saline. 1002 CANCER RESEARCH VOL. 44 on March 25, 2020. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Growth Inhibition of Human Tumor Cells in Athymic Mice by ...[CANCER RESEARCH 44, 1002-1007, March 1984] Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti-Epidermal Growth

  • Upload
    others

  • View
    3

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Growth Inhibition of Human Tumor Cells in Athymic Mice by ...[CANCER RESEARCH 44, 1002-1007, March 1984] Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti-Epidermal Growth

[CANCER RESEARCH 44, 1002-1007, March 1984]

Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti-Epidermal Growth Factor Receptor Monoclonal Antibodies1

Hideo Masui,2 Tomoyuki Kawamoto, J. Denry Sato,3 Bonnie Wolf, Gordon Sato,4 and John Mendelsohn5

Cancer Center, 0-058, University of California, San Diego, La Jolla, California 92093

ABSTRACT

Monoclonal antibodies (MoAbs) were raised against epidermalgrowth factor (EGF) receptors on a human epidermoid carcinomacell line, A431. Administration of anti-EGF receptor MoAbs inhib

ited tumor formation in athymic mice by A431 cells and byanother epidermal carcinoma cell line, 1222. When one of thesame MoAbs was used in therapy against Li-7 (a human hepa-

toma) and HeLa cells (a cervical carcinoma), tumor growth wasnot affected. The number of EGF receptors on A431 cells wasabout 100-fold higher than on T222, Li-7, and HeLa cells, sug

gesting that the number of EGF receptors may not be animportant determinant in suppressing tumor growth. Three anti-

EGF receptor MoAbs were used in the present studies. MoAbs528 (immunoglobulin G2a) and 225 (¡mmunoglobulin G1) arecapable of competing with EGF for receptor binding and inhibitproliferation of A431 cells in culture. The other MoAb, 455(immunoglobulin G1), is incapable of blocking the binding of EGFto its receptors and has no effect on the proliferation of culturedA431 cells. All three MoAbs inhibited A431 tumor growth inathymic mice, indicating that the antibody isotype and the site ofbinding on the EGF receptor are not the determinants of antipro-

liferative activity in vivo. The observation that MoAb against thereceptor for EGF is cytostatic rather than cytocidal in vitro againstA431 cells, yet completely prevents tumor growth In vivo, suggests that some host animal responses also may be involved inthe antitumor effect. MoAbs against growth factor receptorscould provide useful immunotherapeutic agents.

INTRODUCTION

There have been many attempts at passive immunotherapy ofmalignant tumors, but there are only a few reports in which theadministration of conventional antisera inhibited tumor growth inexperimental animals or in cancer patients (26). Thus, passiveimmunotherapy has not been established as a treatment forhuman cancer. The major reasons are the difficulties in identifyingtumor-associated antigens and in preparing homogeneous anti

body in large quantities by conventional methods.Development of the hybridoma technique by Kohler and Mil-

stein (14, 15) has overcome these difficulties, and experimentalcancer immunotherapy has been revived in recent years usingMoAbs.6 Several recent reports demonstrated that MoAbs inhibit

Received May 31, 1983; accepted December 2, 1983.1This research is conducted in part by the Clayton Foundation for Research,

California Division.2Recipient of Grant CA 23052. To whom requests for reprints should be

addressed.3Present address: Molecular Genetics, City of Hope Medical Center, 1500 E.

Duarte Road, Duarte, CA 91010.4 Recipient of NIH Grants CA 33397, GM 17702, and CA 19731.5Recipient of Grant CA 23052. Clayton Foundation Investigator.6The abbreviations used are: MoAb, monoclonal antibody; EGF, epidermal

the growth of human tumor xenografts in athymic mice (8, 9,16,27). Furthermore, the results of several clinical studies on thetreatment of leukemia and lymphoma using MoAbs have beenpublished, but in these cases, clinically significant results havenot been obtained (17,18, 20, 23-25), with one exception. Thiswas a report on the treatment of a human B-cell lymphoma withanti-idiotype antibody, in which the patient has been free from

disease for more than 1.5 years (19).In most of these studies, MoAbs against tumor-associated

antigens have been used for treatment. Another type of MoAbwhich might be used for cancer immunotherapy is an antibodyagainst plasma membrane receptors for growth factors. It is wellknown that the proliferation of tumor cells in culture is controlledby various growth factors, and a similar control mechanism ispostulated for the control of tumor cell growth in vivo (2).Therefore, MoAbs against growth factor receptors, which couldblock access of growth factors to their receptors, may provideuseful therapeutic agents. Recently, Trowbridge and Domingo(29) reported that treatment with anti-transferrin receptor MoAb

can inhibit tumor formation by a human melanoma cell line inathymic mice.

A431 cells, a human epidermoid carcinoma cell line, expressan unusually large number of EGF receptors on the cell surfacemembrane (1 to 3 x 106/cell) (4, 7), and addition of EGF to the

culture medium inhibits the proliferation of these cells in culture(1, 6). We have developed MoAbs against the EGF receptorusing partially purified EGF receptors from A431 cells as antigen(13). In this paper, we report that the administration of anti-EGF

receptor MoAbs inhibits tumor formation in athymic mice by 2human epidermoid carcinomas, A431 and T222. However,growth of 2 other xenografts is not affected.

MATERIALS AND METHODS

Animals. BALB/c athymic mice 6 to 8 weeks of age were used in thepresent studies. These mice were bred and maintained at the AthymicMouse Colony Facility at the University of California, San Diego. Theprocedures for breeding and maintenance have been published previ

ously (22).Human Tumor Cells. A431, T222, Li-7, and HeLa cells were main

tained in a mixture of DME and Ham's F-12 medium (Grand Island

Biological Co.), containing 15 mw 4-(2-hydroxyethyl)-1-piperazineeth-anesulfonic acid buffer and 5% newborn calf serum, at 37°C in 5%

CO2:95% air atmosphere.The Li-7 human hepatoma tumor line was established originally at the

Cancer Center Research Institute, Tokyo, Japan (10). This tumor wasobtained after 50 passages and has been transplanted serially in thisfacility for 22 passages over the past 3 years. The human lung epidermoidcarcinoma (T222) was established in our laboratory by inoculating freshtumor tissue from surgery and was used after 47 to 50 passages in the

growth factor; DME, Dulbecco's modified Eagle's medium; PBS, phosphate-buff

ered saline.

1002 CANCER RESEARCH VOL. 44

on March 25, 2020. © 1984 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Page 2: Growth Inhibition of Human Tumor Cells in Athymic Mice by ...[CANCER RESEARCH 44, 1002-1007, March 1984] Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti-Epidermal Growth

Immunotherapy of Human Tumor Xenografts

present studies. Cell lines were established from these tumors in ourlaboratory and were cultured as described above.

Anti-EGF Receptor MoAbs. In the present studies, we used 3 anti-

EGF receptor MoAbs, referred to as Nos. 528, 225, and 455, raisedagainst the EGF receptor on A431 cells, derived from a epidermoidcarcinoma of the vulva. The development and characterization of thesemurine MoAbs have been published elsewhere (Footnote 7; Ref. 13).MoAb 528 has the isotype lgG2a and competes with EGF for binding toits receptors. When MoAb 528 is added to culture medium, proliferationof A431 cells is inhibited. MoAb 225 of the isotype lgG1 also competeswith EGF binding to receptors and inhibits proliferation of cultured A431cells. In contrast, MoAb 455, also of the isotype lgG1, is incapable ofblocking EGF from binding to its receptors and does not affect growthof A431 cells in culture.

Hybridoma cells producing MoAbs were cultured in a 2:1:1 RPMI-1640:DME:Ham's F-12 mixture containing 15 mM 4-(2-hydroxyethyl)-1 -

piperazineethanesulfonic acid, insulin (10 ^g/ml), 10 ^M mercaptoethanol,and 10 UM ethanolamine. To raise large amounts of MoAbs, cells wereharvested and injected i.p. into athymic mice (107 cells/animal). After 2

to 3 weeks, ascites fluid was collected, centrifuged to remove floatingcells, and kept frozen until purification. MoAbs were purified by ammonium sulfate fractionation followed by DEAE-cellulose column chroma-

tography. The purity of immunoglobulins in these preparations was morethan 90%, as judged by sodium dodecyl sulfate:polyacrylamide gelelectrophoresis. Ascites fluid from one animal provided 3 to 5 mg ofpurified MoAb.

A murine-murine hybridoma (ARB229), which produces MoAb against

a haptene (tyrosyl arsenate), obtained from Hybritech, Inc., was used asa control.

Assay of Tumor Growth in Athymic Mice. A431 cells, T222 cells, Li-

7 cells, or HeLa cells from stock cultures were trypsinized and washedwith serum-free DME:Ham's F-12 (1:1) medium by centrifugation. Tumor

cells were injected s.c. into athymic mice at the specified doses in 0.2ml of medium. Tumor size was measured twice weekly using the formula,'/2 x x x length x width x height (cm). In some experiments with Li-7

hepatoma, T222 epidermoid carcinoma, and A431 epidermoid carcinoma, cells for injection into mice were obtained from tumor xenografts.in these experiments, tumors of 2- to 4-g weight were removed, mincedfinely with scissors, and washed with DME:Ham's F-12 medium, and a

portion was injected s.c. into athymic mice through an 18-gauge needle.

For treatment, varying quantities of MoAb were dissolved in 0.5 ml ofPBS, pH 7.4, and injected ¡.p.into mice. Initially, injections were given 3times per week, but identical results were obtained in later studies usinga twice weekly schedule. Control animals received 0.5 ml of PBS. Eachexperimental group consisted of 5 or 6 athymic mice unless otherwisedescribed.

125I-EGFBinding Assay. The number of EGF receptors on A431 cells,

T222 cells, Li-7 cells, and HeLa cells was determined by the method

described previously (13). Briefly, receptor grade EGF (CollaborativeResearch) was labeled with 125I(Amersham) by the chloramine T method(11). The specific activity was 2.5 x 105 cpm/ng. Cells fixed with 0.2%

p-formaldehyde were plated in wells of microtiter plates containing glass

fiber filters (V & P Enterprises, San Diego, CA) and washed with PBScontaining 0.25% gelatin and 1% albumin. Varying concentrations of 126I-

labeled EGF plus unlabeled EGF were added to the wells. After incubationfor 2 hr at 37°C, the cells were washed with PBS containing 0.25%

gelatin and 5% newborn calf serum, and radioactivity bound to filterswas measured with a y counter. Nonspecific binding was determined byadding 100-fold excess unlabeled EGF. Scatchard analysis was used to

determine the number of EGF receptors and «,,values.Clearance of 125l-labeled MoAb in Athymic Mice Carrying A431

Tumors. 125l-labeledMoAb 528 was prepared from purified MoAb 528

by the procedures described previously (13). A431 cells were injecteds.c. into athymic mice. When small tumors had formed. 125l-labeledMoAb

528 (10 iiCi/0.6 jig) was injected i.p. into a group of 5 animals. The sameamount of 125l-labeled MoAb 528 plus 1.0 mg of unlabeled MoAb 528

was injected into a second group of 2 animals. Ten n\ of blood were

taken from the tail veins on successive days and mixed withl ml of 0.1%albumin solution, and trichloroacetic acid was added. Acid-precipitable

radioactivity was counted in a 7 counter. The first group of mice werekilled, tumors and muscle were removed from each animal and weighed,and the tissue radioactivity was measured in a 7 counter. The specificradioactivity of tumor and muscle was calculated for each animal.

RESULTS

Tumorigenicity of Human Tumor Cells in Athymic Mice. Ininitial experiments, the tumorigenicity of human tumor cells inBALB/c athymic mice was tested by injecting varying numbersof cells s.c. (Table 1). Inoculation of more than 106 A431 cellswas needed to produce tumors consistently, and 107 cells were

selected for the experiments conducted in this study. HeLa cellswere less tumorigenic than were A431 cells, requiring an inoculation of more than 5 x 106 cells for tumor formation. The

tumorigenicity of T222 human epidermoid carcinoma cells andLi-7 human hepatoma cells was also tested, and it was foundthat inoculation of 108 cells was required for tumor formation.

Furthermore, tumor growth was too slow for a useful growthassay. In feasible conduction of immunotherapy experimentsusing T222 and Li-7 cells, an inoculum of finely minced tissue

was used in subsequent experiments. The inoculum consistedof 0.2 ml of minced tumor tissue. Since therapy against T222cells and Li-7 cells was analyzed in a system using minced tumor

and HeLa cells in a system using cultured cells, A431 experiments were carried out with both systems.

Growth Inhibition of A431 Cells in Athymic Mice by Administration of Anti-EGF Receptor MoAbs. Three monoclonal an

tibodies against the EGF receptor were assayed for their capacityto suppress the growth of A431 cells in athymic mice (Chart 1).Treatment 3 times weekly with ascites fluid containing eitherMoAb 528 or 225 completely inhibited tumor formation whentreatment was started on the day of tumor cell inoculation. Afterstopping the MoAb treatment, the mice were observed for 2months, and no tumors formed in these animals. Administrationof EGF (30 /ug/animal, 3 times weekly) did not affect the tumorgrowth. Treatment with ascites fluid containing control MoAbfrom ARB229 did not affect the growth of A431 cells in athymicmice (Chart 2). Also, treatment with mouse serum immunoglob-ulin did not affect the tumor growth (data not shown).

Next, the capacity of MoAb treatment to modify the growth ofdeveloped tumors was tested. Seven days after injection of

Table1Tumorigenicityof A431 and HeLacells in athymic mice

Cells were inoculateds.c. as described in "Materials and Methods." Tumor size

was measured2 weeks after inoculation.Eachgroup given injectionsof A431 cellsconsisted of 6 animals, and each group that received HeLa cells consisted of 2animals.

CellsA431HeLaNo.of cells

inoculated1073.3

x10°10«3.3

x10s5

x1071075

X10»10«Tumors

formed86/66/62/60/62/22/21/20/2Tumorsize6

(cm3)2.2

±0.61.3±0.30.3

±0.20.01.40.40.20.0

a Numberof animalsin which tumors formed per total numberof animalsstudied." Average ±S.E. of tumor size for A431 cells and the average size for HeLa

cells.

MARCH 1984 1003

on March 25, 2020. © 1984 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Page 3: Growth Inhibition of Human Tumor Cells in Athymic Mice by ...[CANCER RESEARCH 44, 1002-1007, March 1984] Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti-Epidermal Growth

H. Masui et al.

02 •

0013

DAYS

16 20

Chart 1. Growth inhibition of A431 cells in athymic mice by anti-EGF receptorMoAb. A431 cells were injected s.c. into BALB/c athymic mice (107 cells/animal).

For the treatment with MoAbs, 0.5 ml of ascites (which contained 2 to 3 mg ofanti-EGF receptor immunoglobulin) was injected i.p. into animals simultaneouslywith the tumor cell inoculation and thereafter as indicated by the arrows. One groupof animals was treated with 30 ^g of EGF in 0.2 ml of PBS. The same volume ofPBS was injected into the control animals. Each group consisted of 5 or 6 animals.The tumor size was measured as described in Table 1. Ascites which containedanti-EGF receptor MoAb was prepared in the following way. Hybridoma cells (2 to3 •10' cells/animal) were injected i.p. into BALB/c athymic mice. Ascites was

collected 2 to 3 weeks after inoculation and centrifuged, and the supematantswere frozen until usage. •,control group; O, EGF-treated group; A and •.grouptreated with anti-EGF receptor MoAbs 528 and 225, respectively. Bars, S.E.

12 14

DAYS

Chart 2. Effect of MoAb ARB229 against the tyrosyl arsenale haptene on thegrowth of A431 cells in athymic mice. Experimental procedures were described inthe legend of Chart 1. Ascites fluid containing MoAb ARB229 was injected astreatment. •,control group; O, ascites fluid (0.5 ml). Bars, S.E.

A431 cells, tumors had grown to an approximate volume of 1 cucm. At this juncture, the tumor-bearing animals were divided into

2 groups displaying similar size distributions of tumors, andtreatment with ascites fluid containing MoAb 528 was startedon a twice weekly schedule. As shown in Chart 3, this treatmentregimen reduced tumor growth significantly. As previously, thesame treatment abolished tumor formation completely whenstarted at the time of tumor cell inoculation (data not shown).

In the experiments described above, ascites fluid was used asthe source of MoAbs. To quantitate the amount of antibodyneeded for inhibition of tumor formation, experiments were performed with purified immunoglobulin.

Cultured A431 cells were inoculated into athymic mice; theanimals were treated with various amounts of purified MoAb528, 225, or 455, and the effects upon tumor formation weretested. Because experiments with T222 cells and Li-7 cells used

minced tumor, rather than a suspension of cultured cells, theeffect of MoAb treatment on growth of A431 minced tumor wasexamined also (Table 2). Treatment with increasing concentrations of MoAbs resulted in progressive reduction in mean tumorsize in each case.

The inhibitory effect of MoAbs 225 and 455 was somewhatless than that of MoAb 528 on tumors from A431 cells. Complete

inhibition of tumor formation was obtained by treatment withMoAb 528 at a 0.2-mg dose, whereas the same effect withMoAb 225 and 455 therapy required a 2-mg dose. A significantsuppression of tumor formation was evident at 0.2-mg doses of

MoAbs 225 and 455 (Table 2). When these experiments wererepeated, there was some variation in the rate of tumor growth,but similar responses to MoAb treatment are always obtained.The results of these experiments demonstrate clearly that anti-

EGF receptor MoAbs, with different characteristics in terms ofigG isotypes and capacity to block EGF binding, inhibit thegrowth of A431 tumors in athymic mice.

In a parallel experiment, minced A431 tumor tissue was inoculated into athymic mice, and animals were treated with MoAb528. Inhibition required higher antibody concentrations withminced tumor than with cultured cells. Complete inhibition of

4.0

3.0

2.0

1.0

0.0

i I I

14

DAYS

Charts. Effect of anti-EGF receptor MoAb on the growth of A431 cells in

athymic mice. A431 cells were injected s.c. in athymic mice. Seven days after thetumor cell inoculation, the treatment with 0.5 ml of MoAb 528 ascites fluid wasstarted. The treatment schedule is shown by the arrows. •,control group; O,treated group. Bars, S.E.

Table2Inhibition of tumor formation by A431 cells in animals treated with anti-EGF

receptor MoAbsIn Experiments 1, 2, and 3, A431 cells (107/animal) were injected s.c. Treatment

was started simultaneously, and thereafter, animals were treated twice a week.Sixteen days after tumor cell inoculations, treatment was stopped, and tumor sizewas measured. In Experiment 4, minced tumor was inoculated s.c. instead of asingle-cell suspension, and treatment with MoAb was the same as above. After 20days, treatment was stopped, and tumor size was measured. When minced tumorwas used, a small residue always remained at the site of injection (see text forexplanation).

ExperimentTreatment1

528IgG2

225IgG3

455IgG4

528 IgGDose

(mg)00.00020.0020.020.22.000.0020.020.22.000.020.22.000.020.22.0Tumorfor

mation86/66/66/64/60/60/66/66/65/53/50/55/56/63/50/66/64/46/60/6Tumorsize

(cm3)3.8

±0.8"2.4

±0.42.4±0.71.1±0.5(0.01<p<0.05)001

.9 ±0.43.7±0.52.8±0.60.3±0.2(0.05 <p<0.1)02.4

±0.55.6±1.30.8

±0.4(0.02 <p<0.05)08.4

±1.75.6±2.21.0±0.3(0.01<p<0.02)0.1±0.0 (p< 0.001)

" Numberof animalsinwhich tumors formed per total numberof animalsstudied.5Average ±S.E.

1004 CANCER RESEARCH VOL. 44

on March 25, 2020. © 1984 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Page 4: Growth Inhibition of Human Tumor Cells in Athymic Mice by ...[CANCER RESEARCH 44, 1002-1007, March 1984] Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti-Epidermal Growth

Immunotherapy of Human Tumor Xenografts

tumor formation was obtained with the 2.0-mg dosage, whereas

with tumors from cultured cells, growth was completely suppressed with only 0.2 mg. We estimate that an inoculum of 0.2ml of minced tumor tissue contains, at most, 1 to 2 x 107 cells,

which is comparable to the inoculum of cultured A431 cells. Ourestimate is based on the following considerations: a volume of0.2 ml of packed cultured A431 cells contains about 4 x 107

cells; and microscopic examination shows that approximately 50to 70% of the A431 minced tumor consists of connective tissue,necrotic tumor tissue, and dead cells. We reason that the weakerinhibitory response with tumors from minced tisssue may beattributable to the fact that the minced tumor retains some ofthe preexisting histológica! structure of tumor tissue and may becapable of producing tumors more easily than a single-cell sus

pension. When minced tumor was inoculated s.c., measurableresidual tissue always remained even when tumor formation wasinhibited, probably due to the connective tissue present in theinoculum. The residual s.c. mass never exceeded 0.3 cu cm. Itshould be noted that growing tumors did not appear in adequately treated animals, which were observed for 2 months afterMoAb treatment was discontinued.

Clearance of 125l-labeled Anti-EGF Receptor MoAb in

Athymic Mice Carrying A431 Tumors. To determine the clearance of anti-EGF receptor MoAbs, 125l-labeled MoAb 528 was

injected i.p. into athymic mice carrying A431 tumors, and thespecific radioactivity in the blood was measured on successivedays as described in "Materials and Methods" (Chart 4). When0.6 ¿igof 125l-MoAb 528 was administered, most of the acid-

precipitable radioactivity was cleared from the blood in 2 to 4days. When 1 mg of unlabeled MoAb 528 was added as a carrierat the time of 125l-MoAb 528 injection, the radioactivity remained

in the blood for up to 8 days. For the former group of mice, thespecific radioactivities were determined in the muscle and in thetumor 6 days after injection. The ratio of the specific radioactivity(tumormuscle) was 3.46 ±0.43. The results suggest that anti-

EGF receptor MoAb is concentrated in the tumors, and theclearance data indicate that treatment on a twice weekly schedule is resonable. When mice were given injections of 125l-MoAb

528 plus carrier MoAb 528, the specific activity of blood remainedmuch higher than that of tumor and muscle even 9 days afterinjection.

Growth Inhibition of Other Human Tumor Cells by Administration of Anti-EGF Receptor MoAb. As described above,

. T-0 2 4 6 8 10

DAYS

Chart 4. Clearance of radioactivity from the blood of mice given injections of0.6 jig (10 tiCi) of '25l-labeledMoAb 528. The proceduresare described in "Materialsand Methods." Each line shows the radioactivity in the blood of each animal:

, mice given injections without camer MoAb 528; , micegiven injectionsof 1.0 mg of unlabeledMoAb 528.

Table 3Growth inhibition of human tumor cells in athymic mice by anti-EGF receptor

MoAb

In vivo tumor growth assays were performed according to the proceduresdescribed in "Materials and Methods." The number of HeLa cells inoculated was 5x 107in this experiment. For Li-7 cells and T222 cells, minced tumor was used

(see text). MoAb treatment was started on the day of tumor cell inoculation, andthereafter treatment was twice weekly. When minced tumor was used, a smallresidue always remainedat the site of injection (see text for explanation).

Experiment1234TumorcellsHepatoma

(U-7)HeLaLung

epidermoidcarcinoma(T222)Lung

epidermoidcarcinoma(T222)No.

528igGtreat

ment00.22.000.22.000.22.000.0020.020.2Tumorformation86/65/56/66/66/65/56/60/60/66/65/50/60/6Tumorsize(cm3)4.5

±0.4*5.1

±1.04.9±1.30.9

±0.10.6±0.10.8

±0.11.6

+0.20.2±0.0 (p<0.001)0.1±0.0 (p<0.001)1.4±

0.40.6±0.2 (0.001<p<0.01)0.2±0.1(p<0.001)0.0(p< 0.001)

Numberof animalsin which tumors formed per total numberof animalsstudied.6 Average ±S.E.

administration of anti-EGF receptor MoAbs inhibited the growth

of A431 cells in athymic mice. To ascertain whether the effect ofthis MoAb treatment upon tumor xenograft growth is generalized, we tested the effects of MoAb 528 on the growth of Li-7

human hepatoma cells, HeLa cells, and T222 human lung epidermoid carcinoma cells. The results are illustrated in Table 3.The growth of Li-7 human hepatoma cells and HeLa cells was

not affected by this treatment. However, the growth of mincedT222 tumors was inhibited significantly, even at a 0.002-mg

dose. The inhibitory effect of MoAb 528 on this tumor growthwas greater than that with minced A431 tumors (Table 2, Experiment 4).

125I-EGF Binding Assay. One reason for selectivity in the

antitumor activities of MoAb treatments might be differences inligand binding to the various target cell populations. One important variable is the number of receptors for EGF on these cells,because it is known that A431 cells have a number of EGFreceptors which is far higher than that reported for HeLa cells.7

We compared the number of receptors and the apparent dissociation constants for 125I-EGFon the cells used in these studies,using the procedures described in "Materials and Methods." The

numbers of receptors on A431, T222, Li-7, and HeLa cells (andtheir K«,values) were: 2.6 x 106/cell («<,1.6 x IO'9 M); 2.9 x104/cell (K,, 4.2 x 1Q-9 M); 2.8 x 104/cell (K^ 2.3 x 10~9 M); and2.0 x 104/cell (K«,7.1 x 10~9 M), respectively. Thus, A431 cells

have a greater number of receptors for this growth factor thando T222, Li-7, and HeLa cells, by a factor of 2 logs.

Using saturating amounts of 125l-MoAb 528, the number of

binding sites for MoAb was compared between A431 cells andT222 cells. In this experiment, the numbers of sites on A431cells and T222 cells were 2.0 x 106/cell and 4.0 x 10"/cell,

respectively, demonstrating that A431 cells had 50 times more

7J. D. Sato, A. D. Le, J. Mendelsohn,J. Polikoff, G. H. Sato, and T. Kawamoto.

Characterizationof mousemonoclonalantibodies to EGFreceptors of A431 humanepidermoidcarcinomacells, submitted for publication.

MARCH 1984 1005

on March 25, 2020. © 1984 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Page 5: Growth Inhibition of Human Tumor Cells in Athymic Mice by ...[CANCER RESEARCH 44, 1002-1007, March 1984] Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti-Epidermal Growth

H. Masui et al.

binding sites for MoAb 528 than did T222 cells. These datasuggest that the number of EGF receptor sites on tumor cellsurface membranes does not account for differences in MoAbantitumor activity against the various cell lines examined in thesestudies.

DISCUSSION

The results of our present studies show that the growth ofA431 and T222 tumor xenografts in athymic mice is remarkablyinhibited by anti-EGF receptor MoAbs. When MoAb 528 is usedin therapy against Li-7 hepatoma or HeLa cells, under identical

conditions, tumor growth remains unaffected. Both A431 cellsand T222 cells were derived from epidermoid carcinomas. HeLacells were established from a cervical cancer and were considered to be epidermoid carcinoma cells (5). However, a reexami-nation of the original slides revealed that the tumor is an ade-

nocarcinoma (12). The results suggest that tumor growth suppression by MoAbs against EGF receptors may be limited toepidermoid carcinomas. The fact that A431 cells (vulva source)and T222 cells (lung source) both respond to MoAb therapyindicates that the tissue origin of the EGF receptors used toraise MoAbs is not a relevant factor in the antitumor effect.Presently, we are testing the effect of these MoAbs on thegrowth of a variety of human tumor xenografts in athymic miceto clarify these points.

The number of EGF receptors on A431 cells is approximately100-fold higher than that on Li-7 and HeLa cells, but the receptornumber on T222 cells is comparable to that on Li-7 and HeLa

cells. This suggests that the EGF receptor number on the cellsurface membranes is not an important determinant in tumorgrowth inhibition.

Among the 3 MoAbs utilized in our studies, 2 are capable ofcompeting with EGF for receptor binding and of inhibiting A431cells in culture; the other is incapable of blocking EGF binding toits receptors and has no significant effect on the growth ofcultured A431 cells.7 Despite the different in vitro characteristics,

when administered to mice all 3 MoAbs inhibited the tumorformation by A431 cells. This suggests that binding of MoAbs toEGF receptors, and not blocking of EGF from binding to itsreceptors, is the essential process leading to suppression oftumor growth in the animals. However, it is also possible thatbinding of MoAbs to EGF receptors could alter the actual functionof the receptors, resulting in tumor growth inhibition.

Recently, Koprowski and coworkers (8,28) proposed that onlyMoAbs of isotype lgG2a can inhibit the growth of human tumorsin athymic mice. As described in our studies, we found that notonly MoAb of isotype lgG2a (No. 528), but also MoAbs of lgG1isotype (Nos. 225 and 455) were capable of inhibiting tumorgrowth. This difference may be due to the fact that MoAbs usedin their studies were raised against tumor-associated antigens,

whereas the MoAbs in our studies were raised against a growthfactor receptor. These 2 types of MoAbs might exert antitumoractivity by different mechanisms.

Data from cell culture studies in our laboratory, as well asmany others, suggest that malignant cells have specific requirements for cell proliferation. These include some growth-promot

ing agents, such as hormones, growth factors, and transportproteins, that often are needed or may be essential for selectedcell types (i.e.. EGF, transferrin, and insulin) (2). Furthermore,studies in which essential hormones, growth factors, and nu

trients are withheld from cultured cells have suggested thattumor cells are more likely to die when deprived of substancesessential for growth than are normal cells (21). EGF is known tostimulate proliferation of various types of cells in culture, butwhen added to the culture medium of A431 cells, it results ingrowth inhibition. In comparison to the studies with MoAbsconducted in athymic mice inoculated with A431 cells, EGFtherapy has no effect on tumor growth. This failure to inhibitgrowth of A431 cell tumors by EGF is possibly attributed to theshort half-life of EGF in animals (3). In contrast, anti-EGF receptorMoAbs have a longer half-life of about 3 days, under the exper

imental conditions in our studies (Chart 4). Thus, MoAbs mayprovide an important method for controlling access of malignantcells to growth factors and hormones required for proliferation.

In cell culture, MoAbs 528 and 225 inhibit proliferation of A431cells, but they do not kill the cells, since the cell number increasesslightly in the presence of these MoAbs.7 When mice that re

ceived injections of A431 cells were treated with MoAb, morethan 90% of the inoculated cells probably were killed by theantibody. The data in Tables 1 and 2 suggest that, if 106 or morecells of the 107 cell inoculum had survived the in vivo MoAb

treatment, tumors would have formed. The mechanism by whichMoAbs suppress tumor growth in animals remains unclear, butannihilation of A431 tumors by complement-mediated lysis is

unlikely, because cultured A431 cells are not lysed in the presence of these MoAbs and complement. Since cytotoxic T-cellsare absent in athymic mice, natural killer cells, antibody-depend

ent killer cells (K), and/or macrophages may have contributed tothe inhibition of tumor formation in the animals. Further experiments should clarify the mechanisms by which anti-receptor

MoAbs prevent tumor cell growth and explore refinements whichcould lead to direct application in the treatment of human cancer.

ACKNOWLEDGMENTS

We thank M. A. Zurbach for processing the manuscript and H. Matsunaga andhis colleagues of the Athymic Mouse Colony Facility for provision and maintenanceof athymic mice. We are grateful to Dr. W. Desmond (Hybritech, La Jolla, CA) forproviding ARB229 hybridoma.

REFERENCES

1. Barnes, D. W. Epidermal growth factor inhibits growth of A431 human epidermoid carcinoma in serum-free cell culture. J. Cell Biol., 93: 1-4, 1982.

2. Barnes, D., and Sato, G. Methods for growth of cultured cells in serum-freemedium. Anal. Biochem., 102: 255-270, 1980.

3. Carpenter, G., and Cohen, S. Epidermal growth factor. Annu. Rev. Biochem.,48: 193-216, 1979.

4. Fabricant, R. N., DeLarco, J. E., and Todaro, G. J. Nerve growth factorreceptors on human melanoma cells in culture. Proc. Nati. Acad. Sei. U. S. A.,74:565-569, 1977.

5. Gey, G. O., Coffman, W. D., and Kubicek, M. T. Tissue culture studies of theproliferative capacity of cervical carcinoma and normal epithelium. CancerRes., 72:264-265,1952.

6. Gill, G. N., and Lazar, C. S. Increased phosphotyrosine content and inhibitionof proliferation in EGF-treated A431 cells. Nature (Lond.), 293:305-307,1981.

7. Haigler, H., Ash, J. F., Singer, S. J., and Cohen, S. Visualization by fluorescenceof the binding and intemalization of epidermal growth factor in human carcinoma cells A431. Proc. Nati. Acad. Sei. U. S. A., 75:3317-3321,1978.

8. Herlyn, D., and Koprowski, H. lgG2a monoclonal antibodies inhibit humantumor growth through interaction with effector cells. Proc. Nati. Acad. Sei. U.S. A., 79. 4761-4765, 1982.

9. Herlyn, D. M., Steplewski, Z., Herlyn, M. F., and Koprowski, H. Inhibition ofcolorectal carcinoma in nude mice by monoclonal antibody. Cancer Res., 40:714-721,1980.

10. Hirohashi, S., Shimosato, Y., Kameya, T., Koide, T., Mukojima, T., Taguchi,Y., and Kageyama, K. Production of a-fetoprotein and normal serum proteinsby xenotransplanted human hepatomas in relation to their growth and morphology. Cancer Res., 39: 1819-1828, 1979.

1006 CANCER RESEARCH VOL. 44

on March 25, 2020. © 1984 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Page 6: Growth Inhibition of Human Tumor Cells in Athymic Mice by ...[CANCER RESEARCH 44, 1002-1007, March 1984] Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti-Epidermal Growth

11. Hunter, W. M., and Greenwood, F. C. Preparation of iodine-131 labelled humangrowth hormone of high specific activity. Nature (Lond.), 194:495-496,1962.

12. Jones, H. W. Jr., McKusick, V. A., Harper, P. S., and Wuu, K-D. The HeLa celland a reappraisal of its origin. Obstet. Gynecol., 38: 945-949, 1971.

13. Kawamoto, T., Sato, J. D., Le, A., Polikoff, J., Sato, G. H., and Mendelsohn,J. Growth stimulation of A431 cells by epidermal growth factor: identificationof high-affinity receptors for epidermal growth factor by an anti-receptormonoclonal antibody. Proc. Nati. Acad. Sei. U. S. A., 80:1337-1341,1983.

14. Kohler, G., and Milstem, C. Continuous cultures of fused cells secretingantibody of predefined specificity. Nature (Lond.), 256: 495-497,1975.

15. Kohler, G., and Milstein, C. Derivation of specific antibody-producing tissueculture and tumor lines by cell fusion. Eur. J. Immund., 6: 511-519,1976.

16. Koprowski.H., Steptewski, Z., Hertyn, D., and Heriyn, M. Study of antibodiesagainst human melanoma produced by somatic cell hybrids. Proc. Nati. Acad.Sei. U. S. A., 75: 3405-3409, 1978.

17. Miller, R. A., and Levy, R. Response of cutaneous T-cell lymphoma to therapywith hybridoma monoclonal antibody. Lancet, 2: 226-230,1981.

18. Miller, R. A., Maloney, D. G., McKillop. J., and Levy, R. In vivo effects of murinehybridoma monoclonal antibody in a patient with T-cell leukemia. Blood, 58:78-86,1981.

19. Miller, R. A., Maloney, D. G., Wamke, R., and Levy, R. Treatment of B-celllymphoma with monoclonal anti-idiotype antibody. N. Engl. J. Med., 306: 517-522, 1982.

20. Nadler, L. M., Stashenko, P., Hardy. R., Kaplan, W. D., Button, L. N., Kufe, D.W., Antman, K. H., and Schlossman, S. F. Serotherapy of a patient with amonoclonal antibody directed against a human lymphoma-associated antigen.

Immunotherapy of Human Tumor Xenografts

Cancer Res., 40: 3147-3154,1980.21. Pardee, A. B., Dubrow, R., Hamlin, J. L., and Kletzien, R. F. Animal cell cycle.

Annu. Rev. Biochem., 47: 715,1978.22. Reid, L., Colburn, P., Sato, G., and Kaplan, N. 0. Approaches to chemotherapy

using the athymic nude mice. In: D. P. Houchens, and A. A. Ovejera (eds.),Proceedings of the Symposium on the Use of Athymic (Nude) Mice in CancerResearch, pp. 123-131. New York: Gustav Fisher, 1978.

23. Ritz, J., Pesando, J. M., Notis-McConarty, J., Clavell, L. A., Sallen, S. E., andSchlossman, S. F. Use of monoclonal antibodies as diagnostic and therapeuticreagents in acute lymphoblastic leukemia. Cancer Res., 41:4771-4775,1981.

24. Ritz, J., Pesando, J. M., Sallen, S. E., Clavell, L. A., Notis-McConarty. J.,Rosenthal, P., and Schlossman, S. F. Serotherapy of acute lymphoblasticleukemia with monoclonal antibody. Blood, 58: 141-152,1981.

25. Ritz, J., and Schlossman, S. F. Utilization of monoclonal antibodies in thetreatment of leukemia and lymphoma. Blood, 59:1 -11,1982.

26. Rosenberg, S. A., and Terry, W. E. Passive immunotherapy of cancer inanimals and man. Adv. Cancer Res., 25: 323-388,1977.

27. Shouval, D., Shafritz, D. A., Zurawski, V. R., Jr., Isselbacher, K. J., and Wands,J. R. Immunotherapy in nude mice of human hepatoma using monoclonalantibodies against hepatitis B virus. Nature (Lond.), 298: 567-569,1982.

28. Steplewski, Z., Heriyn, D., Maul, G., and Koprowski, H. Hypothesis: macrophages as effector cells for human tumor destruction mediated by monoclonalantibody. Hybridoma, 2: 1-5,1983.

29. Trowbridge, I. S., and Domingo, D. L. Anti-transferrin receptor monoclonalantibody and toxin-antibody conjugates affect growth of human tumor cells.Nature (Lond.), 294. 171-173,1981.

MARCH 1984 1007

on March 25, 2020. © 1984 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Page 7: Growth Inhibition of Human Tumor Cells in Athymic Mice by ...[CANCER RESEARCH 44, 1002-1007, March 1984] Growth Inhibition of Human Tumor Cells in Athymic Mice by Anti-Epidermal Growth

1984;44:1002-1007. Cancer Res   Hideo Masui, Tomoyuki Kawamoto, J. Denry Sato, et al.   Anti-Epidermal Growth Factor Receptor Monoclonal AntibodiesGrowth Inhibition of Human Tumor Cells in Athymic Mice by

  Updated version

  http://cancerres.aacrjournals.org/content/44/3/1002

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/44/3/1002To request permission to re-use all or part of this article, use this link

on March 25, 2020. © 1984 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from