| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Immunology |
Ludwig Institute for Cancer Research, Melbourne Tumour Biology Branch, Melbourne, Victoria, 3084 Australia [A. M. S., D. G., M. R., K. C., E. C. N., F. E. S., J. R., A. K., V. K., R. M., F-T. L., Z. L., K. L., A. W. B.]; Ludwig Institute for Cancer Research, New York Branch, Memorial Sloan Kettering Cancer Center, New York, New York 10021 [E. S., E. C. R., K. K., G. R., E. H., L. J. O.]; Biovation Limited, Aberdeen AB2288GW (F. J. C.); and Department of Molecular and Cell Biology, Institute of Medical Sciences, Foresterhill, Aberdeen AB2522D [W. J. H.], Scotland; and Department of Pathology, Immunology Division, Cambridge University, Cambridge, CB212B United Kingdom [K. L. A.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
chain. The genes for the humanized 3S193 monoclonal
antibody IgG1 were transfected into mouse myeloma NS0 cells and cloned
for the establishment of high antibody-producing colonies. Humanized
3S193 antibody was subsequently produced through in
vitro culture and under good manufacturing practice
conditions using hollow-fiber bioreactors. The purified humanized 3S193
(hu3S193) was subsequently characterized and validated for use in
preliminary immunotherapy investigations. hu3S193 reacted specifically
with Ley antigen, with similar avidity to the murine form.
hu3S193 demonstrated potent immune effector function, with higher
antibody-dependent cell-mediated cytotoxicity than its murine
counterpart and potent complement-dependent cytotoxicity
(ED50, 1.0 µg/ml). The in vivo
immunotherapeutic potential of hu3S193 was assessed in a human breast
xenograft model using MCF-7, Ley-positive cells. Six i.v.
doses of up to 1 mg of hu3S193 were administered to animals bearing
established tumors (120–130 mm3) with no significant
effect on tumor growth. In contrast, in an MCF-7 xenograft preventive
model, a 1-mg hu3S193 dosage schedule was able to significantly slow
tumor growth compared with placebo and isotype-matched control IgG1
antibody. hu3S193 has promise for immunotherapy of
Ley-positive tumors and is currently entering Phase I
clinical trials. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
1
2Galß13[Fuc14]
GlcNAcß1R (1, 2, 3, 4, 5, 6, 7)
. The Ley
antigen has been shown to be expressed by 60–90% of human carcinomas
of epithelial cell origin, including breast, pancreas, ovary, colon,
gastric, and lung cancer (1, 2, 3, 4, 5
, 8)
, but does not appear to
be expressed on tumors of neuroectodermal or mesodermal origin
(1
, 5) . Immunohistochemical analysis has demonstrated some
expression of Ley antigen in normal tissues,
particularly in mucosa of the esophagus, stomach, large and small
intestine, some exocrine cells of the pancreas, some epithelial cells
in the gallbladder, ciliated epithelium of the trachea and bronchus,
and type II pneumocytes (5
, 7
, 8)
. The restriction of
Ley expression on normal tissues to epithelial
surfaces, as determined by immunohistochemical methods, suggests this
antigen normally may be relatively inaccessible to administered
antibody (5)
. Accordingly, the high frequency of
Ley-expressing tumors, its high density, and
altered expression on the surface of tumor cells and relatively
homogenous expression in primary and metastatic lesions have led to its
selection as an antigenic target for solid tumor immunotherapy
(2
, 5
, 6 , 8
, 9)
. A range of Ley antibodies have been identified, but a consistent problem with Ley antibodies has been a degree of cross-reactivity with Lex and H-type 2 structures and agglutination with human RBCs (6 , 10) , emphasizing the importance of understanding the target antigen expression and antibody specificity. A small number of Phase I clinical trials with mouse or chimeric anti-Ley antibodies have been conducted to date. In a Phase I study of murine BR55-2 in 12 patients with breast cancer at doses up to 80 mg/m2 for 5 consecutive days, one minor response was observed, transient reduction in skin disease was seen in three patients, and human antimurine antibody prevented repeated treatment cycles (11) . The murine anti-Ley mAb ABL-364 has been studied in patients with non-small cell lung cancer, at dose of 50 or 100 mg for 6 days over a 12-day period but with no responses observed (12) . A chimeric BR-96-doxorubicin construct has been evaluated in a range of patients with advanced cancers including breast cancer at doses up to 700 mg/m2 (weekly for 3 weeks; Ref. 13 ). Upper gastrointestinal toxicity was seen in doses of >200 mg/m2, and this toxicity was reduced with steroid premedication at higher BR-96-doxorubicin doses. The treatment of micrometastatic disease in bone marrow has also been reported in a study involving 19 patients with breast cancer infused with murine mAb ABL-364 at a dose of 100 mg for 6 days (over 12 days). Minor nausea and vomiting were reported, and a reduction in cytokeratin-positive cells was found in bone marrow in a majority of patients (14) . Incorporating the anti-Ley monoclonal antibody B3, the LMB-1 immunotoxin (recombinant Pseudomonas exotoxin) directed against Ley has also been studied, with responses in 5 of 38 patients studied (15) .
We have previously reported the development of the murine anti-Ley mAb 3S193 (6) . To develop an anti-Ley antibody with minimal immunogenicity for use in clinical trials, we describe the construction and production and subsequent in vitro and in vivo characterization of a humanized CDR-grafted construct of 3S193 (hu3S193).
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
Expression of the hu3S193 Antibody
We have briefly described previously the humanization of mu3S193
(6)
. In an attempt to create an antibody suitable for
application in immunotherapy, the antibody m3S193 (IgG3) was humanized
using the variable region frameworks of the KOL heavy chain and
REI chain. Several IgG1 humanized antibodies containing
different combinations of the heavy and light chains were produced, and
detailed analyses were performed to determine the preferred construct.
These antibodies were tested for binding to synthetic
LeY oligosaccharide-keyhole limpet hemocyanin
(Chem-Biomed, Edmonton, Alberta, Canada) and to intact MCF-7 cells by
ELISA. The preferred construct was determined, and the genes for the
humanized 3S193 monoclonal antibody IgG1 were transfected into mouse
myeloma NS0 cells (European Collection of Animal Cell Cultures
number 85110503) and cloned for the establishment of
high-antibody-producing colonies.
Production of hu3S193
Hybridoma Culture.
hu3S193/NS0 was initially cultivated in RPMI 1640 (Life Technologies,
Gaithersburg, MD) plus additives (0.6 µg/ml insulin, 1.0 µg/ml
hydrocortisone, 10-5 M
-thioglycerol, 50 µg/ml penicillin, and 50 µg/ml streptomycin),
supplemented with 0.1 M hypoxanthine, 0.016 M
thymidine, and 10% fetal bovine serum. Small-scale production of
antibody for evaluation of culture conditions and initial
characterization was performed under these conditions. The culture was
then weaned to serum-free conditions and maintained at 37°C, 5%
CO2 in HyQ-CCM1 medium (HyClone, Logan, UT)
supplemented with 0.2% Ex-Cyte VLE, a bovine growth enhancement
supplement (Bayer, Pittsburg, PA).
Large-Scale Production.
Large-scale production of humanized 3S193 IgG1 was achieved by three
production runs using an AcuSyst Maximizer 1000 instrument equipped
with two 1.2-m2 cellulose acetate hollow-fiber
bioreactors (Cellex Biosciences, Minneapolis, MN). Large-scale
production was performed in the Biological Production Facility of the
Ludwig Institute for Cancer Research.
Production runs were commenced from frozen ampoules of the hu3S193 NS0 transfectoma with seeding of 2 x 105 cells/ml as a stationary culture in T175 tissue culture flasks (Nunclon; Nunc, Roskilde, Denmark). The flasks were incubated at 37°C and 5% CO2 in HyQ-CCM1 medium (HyClone), containing bovine transferrin (56.0 µg/ml) and bovine albumin (1.1 g/ml), supplemented with 0.2% Ex-Cyte VLE (Bayer). The preculture was maintained until a suitable number of cells were obtained to inoculate roller bottles. The roller bottle cultures were further incubated until a suitable number of cells (>1.0 x 108) were available for the inoculation of the bioreactor. The culture was prepared for inoculation and pumped into two 1.2-m2 cellulose acetate hollow-fiber cartridges per production run, according to the manufacturer’s specification.
The process temperature was kept at 37°C. Daily samples of culture medium, tested for glucose and lactate concentration, as well as pH and dissolved oxygen, determined the perfusion rate of the system. Whenever possible, the intracellular compartment glucose concentration was kept at >0.8 g/liter, lactate at <1.8 g/liter, and the dissolved oxygen concentration at >80 mm Hg. The continuously perfused cultures in the serum-free conditions described above were harvested daily, monitored, and maintained for up to 55 days.
Antibody-containing medium was harvested continuously into depyrogenated, sterile glass bottles once the culture showed a consistent metabolic pattern. The collected supernatant was centrifuged to clarify the medium from cells and debris and then stored at -20°C. Harvest samples were tested by ELISA to monitor the antibody production rate of the run. Biosafety testing was performed on the master and working cell banks as well as on production lots.
|
Purification of hu3S193 |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
Antibody content at all steps after the anion exchange resin was estimated by absorbance at 280 nm, using the calculated absorbance value of 1.36 absorbance units/mg of protein. In addition to monitoring the protein profile of the column eluant, the chromatographic purification was monitored by SDS-PAGE and IEF gel analyses. Proteins were visualized by colloidal Coomassie blue or silver staining analyses.
|
Protein Characterization of Purified hu3S193 |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
|
Biological Characterization of Purified hu3S193 |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
Competitive Cell Binding Assay.
Comparison of the antibody-antigen affinity of the humanized and murine
forms of 3S193 and another murine anti- Ley
antibody, BR55-2 (IgG3; Z. Steplewski, Wistar Institute, Philadelphia,
PA) was performed. The ability of these antibodies to compete with
125I-hu3S193 for binding to
Ley-positive MCF-7 cells was assessed using a
published method (17)
. Increasing amounts of each
unlabeled antibody were added to MCF-7 cells in triplicate, followed by
20 ng of 125I-hu3S193, mixed, and incubated at
room temperature for 45 min. Cells were washed three times in RPMI 1640
and centrifuged, and the supernatants and pellets were counted in a
gamma counter (Cobra II, model 5002, Auto-gamma; Packard Instruments,
Canberra, Australia). IgG1 and IgG3 control antibodies huA33 and R24
(Ludwig Institute for Cancer Research, New York, NY) were assessed in
parallel.
Biosensor Analysis.
Biosensor analyses were performed using a BIAcore 2000 (BIAcore AB,
Uppsala, Sweden) on a carboxymethyldextran-coated sensor chip (CM5).
The chip was derivatized with synthetic Lewis Y tetrasaccharide,
coupled to BSA (Alberta Research Council, Edmonton, Alberta, Canada),
using standard amine-coupling chemistry
(N-hydroxysuccinimide and
N-ethyl-N'-dimethylaminopropyl-carbodiimide).
Samples of hu3S193 and m3S193 were diluted in HBS buffer (10
mM HEPES, pH 7.4, 150 mM NaCl, 3.4
mM di-Na-EDTA, and 0.005% Tween 20), and aliquots (30
µl) were injected over the sensor chip surface at a flow rate of 5
µl/min. After the injection phase, dissociation was monitored by
flowing HBS buffer over the chip surface for 300 s. Bound antibody
was eluted, and the chip surface was regenerated between samples by
injection of 20 µl of 10 mM NaOH. For kinetic analyses of
binding, varying concentrations of hu3S193 and m3S193 were injected
over the sensor chip surface. Apparent
Ka and
Kd rate constants were calculated by
nonlinear least squares regression analysis, using BIAevaluation
version 3.0 software (Pharmacia Biosensor, Uppsala, Sweden).
Immune Effector Function.
Both CDC and ADCC assays used MCF-7 breast carcinoma cells as target
cells. An isotype-matched, but unrelated, antibody huA33
(17)
was used as a control, as was a control
(Ley negative) cell line, SW-1222.
CDC Assay.
Target cells were plated at a density of 
100 cells/well, with
varying amounts of hu3S193 (0.001–10 µg/ml), in microtiter plates
(Nunclon; Nunc). Diluted human complement, prepared from the blood of
normal, healthy volunteers, was added to each well. Tests were done in
triplicate with medium, antibody, and complement controls. After a 4-h
incubation at 37°C, cells were fixed with methanol, and viable cells
were visualized with Giemsa stain. The percent cytotoxicity of the
antibody was calculated as follows: number of viable cells in well
treated with antibody + complement ÷ number
of viable cells in untreated well x 100. The percent
cytotoxicity was plotted versus concentration of antibody
(µg/ml), and the concentration of antibody causing 50% cytotoxicity
was determined.
ADCC Assay.
Target cells were incubated for 18 h with 100 µCi of
51Cr, washed, and plated at a density of 10,000
cells per well into a 96-well microtiter plate. Effector cells
(peripheral blood mononucleocytes) were freshly prepared and added to
the target cells to achieve E:T ratios of 50:1. Aliquots of hu3S193
(0.003–10 µg/ml) were added to appropriate wells. After incubation
at 37°C for 4 h, samples of cell supernatant were collected and
counted for released 51Cr. Controls included in
the assay corrected for spontaneous release (medium alone) and total
release (detergent). The percentage of cell lysis (cytotoxicity) was
calculated according to the formula: percentage cytotoxicity
= (sample counts - spontaneous release)
÷ (total release - spontaneous release)
x 100. The percent cytotoxicity was plotted versus
concentration of antibody (µg/ml), and the concentration of antibody
causing 50% cytotoxicity was determined.
|
Therapeutic in Vivo Studies with hu3S193 |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
Prevention of MCF-7 Tumor Growth in Mice.
The BALB/c MCF-7 xenograft model described above was used in these
studies. However, in this study, mice received the first dose of
antibody treatment retro-orbitally at the same time as inoculation of
20 x 106 MCF-7 cells and
insertion of the estrogen pellet. The groups of five animals
subsequently received five additional antibody doses retro-orbitally at
48-h intervals. There were four treatment groups; hu3S193 in PBS at 0.1
or 1.0 mg and control groups receiving 1.0 mg of isotype-matched,
unrelated antibody huA33 or PBS alone. Tumor volume was measured
[(length x width2)/2] for 56
days from commencement of treatment. Results from groups of mice were
assessed for statistical significance by Student’s paired t
test.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
chain. The huVH and huVHT were less efficient at binding
than the chimeric heavy chain. The inclusion of Thr-24 slightly
improved the performance and reduced the cross-reactivity to
Ley-related structures. Four further humanized antibodies were produced. These were huVHA/huVKF, huVHAS/huVKF, and huVHASY/huVKF, in which the heavy chain included murine residues selected from Ala-27, Ser-76, and Tyr-79, and huVHTAS/huVKF, which combines the incorporation of Ala-74 and Ser-76 with that of Thr-24. These antibodies were tested for binding to synthetic LeY oligosaccharide-keyhole limpet hemocyanin and to intact MCF-7 cells by ELISA. The results indicated the huVHASY/huVKF antibody offered an improvement in binding over the huVHT/huVK antibody. Comparison of the huVHT, huVHAS, and huVHTAS antibodies suggested that combining the two changes was not beneficial.
huVHASY/huVKF showed a little cross-reactivity to
Lex and H type 2 antigens in ELISAs, which,
although the antibody did not cause hemolysis, meant that huVHT/huVKF
(previously described as clone 11 in Ref. 6
) was the
preferred construct. The amino acid sequences of the light- and
heavy-chain regions of the translated protein are shown in Fig. 1
.
|
A master cell bank was established and after passing the requisite
biosafety testing; a manufacturer’s working cell bank was then
established. Large-scale production was performed using the working
cell bank. Three separate bioreactor production runs were performed to
produce the humanized 3S193 large-scale clinical batch. The bioreactors
produced hu3S913 at a rate of 114.2–166.5 mg/day, with an average
production rate of 144.4 mg/day for the three runs (Fig. 2
). The first bioreactor run lasted 50 days. Daily harvests over 40 days
yielded 6569 ml of extracapillary medium. The bulk antibody
concentration was estimated by ELISA and indicated a total of 8325.6 mg
of hu3S193 IgG. The second bioreactor run lasted 55 days (51 harvest
days). A total of 5175 ml of extracapillary medium was harvested, which
was estimated by ELISA to contain a total of 6279 mg of antibody. The
final run was cultured over 43 days. The 37 harvest days produced a
total volume of 3907 ml, containing 6554 mg of hu3S193 IgG.
|
Characterization of hu3S193
Protein Analysis.
The A280 nm trace from SEC analysis
of purified hu3S193 showed a single peak with a elution time of 8.62
min, corresponding to a calculated molecular weight of 157,000. No
evidence of dimers or other high molecular weight species was observed,
in agreement with nonreducing gel analysis. Further analysis by
SDS-PAGE of hu3S193, under reducing conditions, indicated the expected
molecular masses of 25 and 50 kDa for the hu3S193 light and heavy
chains, respectively. No other significant protein bands were
demonstrated, attesting to the purity of the preparation. IEF analysis
showed a distinct single band for the hu3S193 sample visible with a
determined pI of 9.0 (data not shown).
Scatchard Analysis.
Hu3S193 was analyzed to determine the apparent binding affinity of the
purified antibody for its antigen. Samples of the large-scale GMP
antibody were found to have an identical apparent binding affinity of
5.33 x 106/M to
small-scale-produced hu3S193. The binding site number per MCF-7 cell
was calculated to be 7 x
106/cell.
Competitive Cell Binding Assay.
There was a slight loss of apparent binding affinity in hu3S193
compared with m3S193; however, hu3S193 was comparable and slightly
superior to another murine anti- Ley antibody
(BR55-2; Fig. 3
).
|
, and measured parameters are shown in Table 1
. The data were fitted globally to a 1:1 Langmuir binding model with
compensation for mass transfer. The presence of mass transfer effects
was inferred from changes in the shape of the sensorgram at one analyte
concentration but with varying flow rate (data not shown). Data and
parameters were compared with those of the equivalent murine antibody.
The humanized antibody demonstrated faster on and off rates compared
with its murine counterpart, resulting in a reduction in apparent
binding affinity. The Ka of
4.19 x
106/M for hu3S193 is highly
comparable with the specific binding affinity determined by the
cell-based Scatchard analysis.
|
|
).
|
). A similar potency was determined for
hu3S193 purified from small-scale production runs (data not shown).
This cytotoxicity was 3-fold better than that achieved by the murine
counterpart. Control antibody huA33 had no evidence of CDC or ADCC
activity against MCF-7 cells, and hu3S193 had no cytotoxicity against
control SW-1222 cells (data not shown). |
Therapeutic in Vivo Studies MCF-7 Mouse Xenograft Model |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
, where day 0 equals the first day of antibody treatment. No significant
difference was observed between the placebo control group and the
antibody treatment groups. Tumor growth continued unabated by the
administration of antibody, and animals were culled due to large tumor
burdens 35 days from commencement of treatment.
|
. The concurrent administration of 1 mg of hu3S193 i.v. and s.c. tumor
cell inoculation with four subsequent antibody doses at 48-h intervals
was able to significantly slow tumor growth. Statistical analysis was
performed with Student’s paired t test. No difference was
observed among 0.1 mg of hu3S193, control huA33 treatment, and placebo
treatment. Significant inhibition of tumor growth was observed with
treatment with 1 mg of hu3S193 versus 1 mg of huA33 at day
56 (P = 0.044). Treatment with 1 mg of
hu3S193 versus placebo showed significant reduction in tumor
growth at day 46 (P = 0.042), day 50
(P = 0.021), and day 56
(P = 0.044).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
The advent of DNA recombinant technology has allowed the synthesis of "reshaped" antibodies. Humanized antibodies combine murine antigen binding (CDR) with human V-region framework determinants. CDR grafting is one method to include human framework regions and aims to make the surface of the antibody seem as fully human as possible but retain the murine antigen binding packaging and interface reactions determining high-affinity binding (18) . Although other methods of humanization of murine antibodies have been developed, including veneering of potential B- and T-cell binding sites, and the production of "human" antibodies through recombinant and transgenic techniques (19) , there are limited data available on the in vivo behavior of these constructs in human studies.
The humanized CDR-grafted construct of 3S193 (hu3S193) was optimized for antigen binding specificity and immune effector function properties. Subsequently the NS0 transfectoma was cloned for stability, high antibody productivity, and clonality.
The retention of antigen binding affinity after humanization is essential to allow optimal binding to tumor in vivo. The humanization of 3S193 involved careful analysis of a number of candidate constructs for affinity and specificity for Ley in ELISA tests with synthetic Ley and Ley-containing glycoproteins and glycolipids. The huVHT/huVKF construct (clone 11), containing the murine Thr-24 heavy-chain residue, although not demonstrating as high affinity to Ley as other constructs, retained the best specificity with minimal cross-reactivity to other related antigen structures, including Lex and H type 2 antigens. Our detailed specificity analysis has been reported previously (6) . The minor differences in affinity between constructs, compared with the importance of specificity for Ley antigen in vivo, resulted in the huVHT/huVKF clone being chosen for subsequent development.
Of particular importance in the humanization strategy was that hu3S193 (IgG1) displays enhanced immune effector function, particularly in mediating cellular cytotoxicity (ADCC), compared with the murine form. Co et al. (20) compared murine, chimeric, and humanized versions of the anti-Lewis Y antibody ABL 364 and showed that the humanized antibody had an ADCC (IC50, 10 µg/ml on MCF-7 cells) that was superior to the murine form but less effective CDC activity. Pharmacokinetics of the humanized form was prolonged compared with the murine form. Although a direct comparison was not made, hu3S193 is one of the most cytotoxic antibodies yet reported in terms of IC50 for both CDC (ED50, 1 µg/ml), and ADCC (IC50, 0.32 µg/ml). This has immediate relevance to its potential for passive immunotherapy of Ley-expressing tumors.
The production of hu3S193 under small- and large-scale GMP conditions was achieved without loss of affinity and retention of biological function. The scale up of production under serum-free conditions is often problematic; however, antibody production yields, stability, and consistency of production were attained during GMP bioreactor runs of up to 55 days. Bioreactor production of antibody expressed in mammalian cells under large-scale conditions is well suited to batch runs in the range of 10–100 g, has the advantage of reduced infrastructure requirements of fermenters, and has been used for antibodies approved for clinical use (21) . The hu3S193 produced under large-scale conditions was determined to be homogeneous by column and SDS-PAGE analyses. Furthermore, the final product safety tests demonstrated appropriate clearance of DNA, endotoxin, and protein A and sterility. The ability to scale up production of hu3S193 with demonstration of equivalence in affinity and potent immune effector function will allow clinical evaluation of hu3S193 to be optimally performed.
As part of the preclinical development of the hu3S193, in vitro investigations were performed to assess the affinity of the antibody for its antigen. Excellent correlation was found between the apparent binding affinities calculated from the MCF-7 cell-based scatchard analysis and BIAcore analyses of hu3S193 apparent affinity for its Ley antigen (Ka, 5.33 x 106 for cell binding assay compared with BIAcore KA of 4.19 x 106/M). A reduction in avidity was observed with humanization of 3S193; however, this was found to be primarily a result of an increased off rate compared with murine 3S193. The significance of this high off rate in vivo is uncertain, because the concentration of antibody surrounding tumor, combined with the high number of binding sites available per cell, would contribute to high antibody localization to tumor cells, as we have demonstrated in an MCF-7 xenograft BALB/c model.3
A disparity was observed between the Biosensor calculated avidity of the mu3S193 and hu3S193 and that determined by the cell binding analysis, in which only a 2–3-fold reduction in avidity was apparent after humanization. The differences in apparent binding of mu3S193 may be attributable to the murine antibody having been raised against cultured human tumor lines expressing Ley on the cell surface, whereas the Biosensor binding surface was a synthetic Ley tetrasaccharide coupled to BSA in a multivalent configuration. The humanization of mu3s193 involved framework alterations, which may also account for the higher BIAcore off rate compared with the murine, without affecting specificity. The exact conformation of the synthetic antigen is unknown, whereas the cell-bound antigen is probably in its native conformation. However, it is not uncommon to observe variations in apparent avidity when using different techniques. This has been noted by ourselves in other antigen systems (22) and by others with the Ley antigen system (23) .
An in vivo therapeutic effect was not observed after i.v. administration of 0.3–1.0 mg of naked hu3S193 to mice with established Ley-positive xenografts. The lack of effect is not due to poor tumor localization, because our previous studies have shown excellent tumor targeting of 125I- and 90Y-labeled hu3S913 in this model.3 These experiments determined the mean half-life of hu3S193 to be 43.3 h in this murine model, suggesting that the 72-h dosing schedule was appropriate. The inability of the humanized antibody to have any antitumor effect in this murine model may be due to restricted murine serum complement activity. The potent in vitro ADCC of the hu3S193 would suggest that, in conjunction with potent CDC activity in the human, these humoral effects in a clinical setting would augment the therapeutic effects of hu3S193.
In the preventive tumor model, however, significant reduction in tumor growth was observed with the 1-mg hu3S193 dose level. The dosage schedule was selected on the basis of published literature and the known mean half-life of hu3S193 in an MCF-7 BALB/c model.3 Animals receiving placebo exhibited rapid, unchecked growth of xenograft. Control isotype matched antibody at 1.0-mg doses displayed similar growth curves to 0.1-mg doses of hu3S193. The experiment was not extended past 56 days because of the size of tumors in animals not receiving 1-mg hu3S193 treatment and the fact that the estrogen implants were only able to release estrogen for 60 days, after which the model becomes less reliable. The data suggest that the differences in tumor growth would have become more marked if further observation had been possible. These results demonstrate the potential efficacy of hu3S193 in patients with small-volume disease. The differences between mouse and human models of cancer are well known, but the potent complement-mediated activity of hu3S193, in conjunction with other Fc effector mechanisms (including ADCC), does provide evidence of an attractive therapeutic mechanism of tumor killing.
The high frequency of Ley-expressing tumors, the high density and altered expression of Ley on the surface of tumor cells, and the relatively homogenous expression in primary and metastatic lesions have led to its selection as an antigenic target for cancer immunotherapy (2 , 5 , 6 , 8 , 9) . The expression of Ley antigen in normal tissues (including the gastrointestinal tract) is of concern in targeting strategies against this antigen; however, the accessibility of normal tissues to anti-Ley antibodies remains unclear. Results of studies with murine and chimeric antibodies against Ley suggest gastric toxicity may be relevant in proposed clinical trials; however, significant toxicity has only been observed at protein doses of >200 mg (14) . The optimal protein dose for effective targeting of Ley tumors (tumor:blood ratios and normal tissue uptake) is yet to be defined with careful biodistribution and biopsy-based trials. In this context, high protein doses of anti-Ley antibodies reported to cause toxicities may not be required for effective tumor targeting.
The introduction of humanized antibodies into the clinic allows the immunogenicity of these constructs to be determined. The recombinant human mAb HER2 has had minimal HAHA responses seen in hundreds of patients entered into clinical trials to date, including patients receiving weekly infusions for >1 one year (24 , 25) . Other humanized antibodies, including huCTMO1 and huM195, have been administered to patients with minimal toxicity and no evidence of HAHA (26 , 27) . In comparison, humanized antibody A33, administered to patients with advanced colorectal carcinoma, has been shown to induce HAHA in a subset of patients after prolonged treatment (28) . Thus, although there has been a consistent observation of markedly reduced incidence of immune responses to humanized antibodies in patients, the humanization process does not totally remove the potential for HAHA. The development of a humanized form of 3S193 has been undertaken to minimize the risk of immune recognition and to allow repeated infusions over time. The potent immune effector functions of hu3S193 in vitro and in vivo, combined with its high specificity for Ley antigen, and the humanization strategy used in production suggest that hu3S193 has promise in the treatment of Ley-expressing tumors in cancer patients.
|
FOOTNOTES |
|---|
1 To whom requests for reprints should be
addressed, at Tumour Targeting Program, Ludwig Institute for Cancer
Research, Austin and Repatriation Medical Centre, Austin Campus,
Heidelberg, Victoria 3084, Australia. 
2 The abbreviations used are: Ley,
Lewis Y; mAb, monoclonal antibody; m3S193, murine 3S193; hu3S193,
humanized 3S193; MCF-7, Ley-positive human breast carcinoma
cell line; huA33, humanized A33 mAb; CDC, complement-dependent
cytotoxicity; ADCC, antibody-dependent cellular cytotoxicity; GMP, good
manufacturing practice; CDR, complementarity binding region; SEC, size
exclusion chromatography; IEF, isoelectric focusing; HAHA, human
anti-human antibodies.
3 K. Clarke, F. T. Lee, M. W. Brechbiel,
F. E. Smyth, E. Richards, E. Stockert, L. J. Old, and A.
M. Scott, unpublished results.
Received 11/17/99. Accepted 4/13/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSPurification of hu3S193Protein Characterization of...Biological Characterization of...Therapeutic in Vivo Studies...RESULTSTherapeutic in Vivo Studies...DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. A. Herbertson, N. C. Tebbutt, F.-T. Lee, D. J. MacFarlane, B. Chappell, N. Micallef, S.-T. Lee, T. Saunder, W. Hopkins, F. E. Smyth, et al. Phase I Biodistribution and Pharmacokinetic Study of Lewis Y-Targeting Immunoconjugate CMD-193 in Patients with Advanced Epithelial Cancers Clin. Cancer Res., November 1, 2009; 15(21): 6709 - 6715. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hatakeyama, K. Sugihara, J. Nakayama, T. O. Akama, S.-M. A. Wong, H. Kawashima, J. Zhang, D. F. Smith, C. Ohyama, M. Fukuda, et al. Identification of mRNA splicing factors as the endothelial receptor for carbohydrate-dependent lung colonization of cancer cells PNAS, March 3, 2009; 106(9): 3095 - 3100. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Kelly, F. T. Lee, K. Tahtis, F. E. Smyth, M. W. Brechbiel, and A. M. Scott Radioimmunotherapy with {alpha}-Particle Emitting 213Bi-C-Functionalized trans-Cyclohexyl-Diethylenetriaminepentaacetic Acid-Humanized 3S193 Is Enhanced by Combination with Paclitaxel Chemotherapy Clin. Cancer Res., September 15, 2007; 13(18): 5604s - 5612s. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Scott, N. Tebbutt, F.-T. Lee, T. Cavicchiolo, Z. Liu, S. Gill, A. M.T. Poon, W. Hopkins, F. E. Smyth, C. Murone, et al. A Phase I Biodistribution and Pharmacokinetic Trial of Humanized Monoclonal Antibody Hu3s193 in Patients with Advanced Epithelial Cancers that Express the Lewis-Y Antigen Clin. Cancer Res., June 1, 2007; 13(11): 3286 - 3292. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Scott, F.-T. Lee, N. Tebbutt, R. Herbertson, S. S. Gill, Z. Liu, E. Skrinos, C. Murone, T. H. Saunder, B. Chappell, et al. A phase I clinical trial with monoclonal antibody ch806 targeting transitional state and mutant epidermal growth factor receptors PNAS, March 6, 2007; 104(10): 4071 - 4076. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Farhan, C. Schuster, M. Klinger, E. Weisz, G. Waxenecker, M. Schuster, V. Sexl, G. C. Mudde, M. Freissmuth, and R. Kircheis Inhibition of Xenograft Tumor Growth and Down-Regulation of ErbB Receptors by an Antibody Directed against Lewis Y Antigen J. Pharmacol. Exp. Ther., December 1, 2006; 319(3): 1459 - 1466. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Kelly, F. T. Lee, F. E. Smyth, M. W. Brechbiel, and A. M. Scott Enhanced Efficacy of 90Y-Radiolabeled Anti-Lewis Y Humanized Monoclonal Antibody hu3S193 and Paclitaxel Combined-Modality Radioimmunotherapy in a Breast Cancer Model J. Nucl. Med., April 1, 2006; 47(4): 716 - 725. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Westwood, M. J. Smyth, M. W. L. Teng, M. Moeller, J. A. Trapani, A. M. Scott, F. E. Smyth, G. A. Cartwright, B. E. Power, D. Honemann, et al. Adoptive transfer of T cells modified with a humanized chimeric receptor gene inhibits growth of Lewis-Y-expressing tumors in mice PNAS, December 27, 2005; 102(52): 19051 - 19056. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. T. Lee, A. J. Mountain, M. P. Kelly, C. Hall, A. Rigopoulos, T. G. Johns, F. E. Smyth, M. W. Brechbiel, E. C. Nice, A. W. Burgess, et al. Enhanced Efficacy of Radioimmunotherapy with 90Y-CHX-A''-DTPA-hu3S193 by Inhibition of Epidermal Growth Factor Receptor (EGFR) Signaling with EGFR Tyrosine Kinase Inhibitor AG1478 Clin. Cancer Res., October 1, 2005; 11(19): 7080s - 7086s. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Vearing, F.-T. Lee, S. Wimmer-Kleikamp, V. Spirkoska, C. To, C. Stylianou, M. Spanevello, M. Brechbiel, A. W. Boyd, A. M. Scott, et al. Concurrent Binding of Anti-EphA3 Antibody and Ephrin-A5 Amplifies EphA3 Signaling and Downstream Responses: Potential as EphA3-Specific Tumor-Targeting Reagents Cancer Res., August 1, 2005; 65(15): 6745 - 6754. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Scott, F.-T. Lee, R. Jones, W. Hopkins, D. MacGregor, J. S. Cebon, A. Hannah, G. Chong, P. U, A. Papenfuss, et al. A Phase I Trial of Humanized Monoclonal Antibody A33 in Patients with Colorectal Carcinoma: Biodistribution, Pharmacokinetics, and Quantitative Tumor Uptake Clin. Cancer Res., July 1, 2005; 11(13): 4810 - 4817. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Chong, F. T. Lee, W. Hopkins, N. Tebbutt, J. S. Cebon, A. J. Mountain, B. Chappell, A. Papenfuss, P. Schleyer, P. U, et al. Phase I Trial of 131I-huA33 in Patients with Advanced Colorectal Carcinoma Clin. Cancer Res., July 1, 2005; 11(13): 4818 - 4826. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. R. Boghaert, L. Sridharan, D. C. Armellino, K. M. Khandke, J. F. DiJoseph, A. Kunz, M. M. Dougher, F. Jiang, L. B. Kalyandrug, P. R. Hamann, et al. Antibody-Targeted Chemotherapy with the Calicheamicin Conjugate hu3S193-N-Acetyl {gamma} Calicheamicin Dimethyl Hydrazide Targets Lewisy and Eliminates Lewisy-Positive Human Carcinoma Cells and Xenografts Clin. Cancer Res., July 1, 2004; 10(13): 4538 - 4549. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bauer, N. Adrian, B. Williamson, C. Panousis, N. Fadle, J. Smerd, I. Fettah, A. M. Scott, M. Pfreundschuh, and C. Renner Targeted Bioactivity of Membrane-Anchored TNF by an Antibody-Derived TNF Fusion Protein J. Immunol., March 15, 2004; 172(6): 3930 - 3939. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Ritter, L. S. Cohen, C. Williams Jr., E. C. Richards, L. J. Old, and S. Welt Serological Analysis of Human Anti-Human Antibody Responses in Colon Cancer Patients Treated with Repeated Doses of Humanized Monoclonal Antibody A33 Cancer Res., September 1, 2001; 61(18): 6851 - 6859. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Lovqvist, J. L. Humm, A. Sheikh, R. D. Finn, J. Koziorowski, S. Ruan, K. S. Pentlow, A. Jungbluth, S. Welt, F. T. Lee, et al. PET Imaging of 86Y-Labeled Anti-Lewis Y Monoclonal Antibodies in a Nude Mouse Model: Comparison Between 86Y and 111In Radiolabels J. Nucl. Med., August 1, 2001; 42(8): 1281 - 1287. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Tahtis, F.-T. Lee, F. E. Smyth, B. E. Power, C. Renner, M. W. Brechbiel, L. J. Old, P. J. Hudson, and A. M. Scott Biodistribution Properties of 111Indium-labeled C-Functionalized trans-Cyclohexyl Diethylenetriaminepentaacetic Acid Humanized 3S193 Diabody and F(ab')2 Constructs in a Breast Carcinoma Xenograft Model Clin. Cancer Res., April 1, 2001; 7(4): 1061 - 1072. [Abstract] [Full Text]
|
|
|
|
|
|
K. Clarke, F.-T. Lee, M. W. Brechbiel, F. E. Smyth, L. J. Old, and A. M. Scott In Vivo Biodistribution of a Humanized Anti-Lewis Y Monoclonal Antibody (hu3S193) in MCF-7 Xenografted BALB/c Nude Mice Cancer Res., September 1, 2000; 60(17): 4804 - 4811. [Abstract] [Full Text]
|
|
|
|
|
|
K. Clarke, F.-T. Lee, M. W. Brechbiel, F. E. Smyth, L. J. Old, and A. M. Scott Therapeutic Efficacy of Anti-Lewisy Humanized 3S193 Radioimmunotherapy in a Breast Cancer Model: Enhanced Activity When Combined with Taxol Chemotherapy Clin. Cancer Res., September 1, 2000; 6(9): 3621 - 3628. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
),
40 harvest days; run 2 (B00297; 
), 51 harvest days; run 3 (B00331;
), 37 harvest days.
), and BR55-2
(
) for Ley-positive MCF-7 cells in the presence of 20 ng
of 125I-hu3S193. Class-specific controls, huA33 as IgG1 and
R24 as IgG3 control, were not inhibited by addition of cold antibody
(data not shown).
) and m3S193
(
) ADCC activity over a range of mAb concentrations. %
Cytotoxicity is a measure of antibody-mediated cell lysis as
measured by release of 51Cr from labeled target cells.
. Day 0 refers to day of first antibody
dose. Tumor growth results for the five treatments are presented for
the study period: PBS placebo control (•) and four escalating dose
levels of hu3S193, 0.03 mg (
), 0.32 mg (
).
) and no
antibody, PBS placebo (