| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
I11
Laboratory of Molecular Biology, Charité Children’s Hospital, Humboldt-University, 10117 Berlin [K. K., W. W., G. G., P. F.], and Medical Department II, Division of Renal Diseases, Hospital München-Harlaching, 81545 München [J. E. S.], Germany
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSStatistical AnalysisRESULTS AND DISCUSSIONREFERENCES |
|---|
I1
(Cam) in vitro and in vivo.
Previously, mAb 138H11, produced against human renal
-glutamyltransferase, stained over 99% clear cell and papillary RCC
on frozen sections, showing a membranous expression of the target
antigen. In contrast, in normal kidneys GT was restricted
to the brush-border in the lumen of proximal tubules and not accessible
to the circulation. Thus, human tumor-bearing kidneys
perfused in an extra-corporeal system with 99mTc-138H11
revealed a high, specific uptake into the tumor. In this study,
fluorescence-activated cell sorting analysis showed binding of
mAb 138H11 to RCC cell lines, whereas squamous cell carcinoma lines,
fibroblasts, and the murine RENCA were negative. XTT cell
proliferation assays revealed efficient killing of the Caki-1 cell line
by the 138H11-Cam conjugate using SPDP
(EC50 = 5 x 10-11
M) as a covalent linker. For in vivo
testing, five groups of eight nude mice each were injected with
2.5 x 106 Caki-1 cells s.c. and treated with the
following: (a) PBS; (b) 138H11;
(c) Cam; (d) a mixture of 138H11 and
Cam; and (e) 138H11-Cam conjugate. Treatment
started on day 1 after tumor induction and was repeated three times.
The data show a highly significant inhibition of tumor growth with the
138H11-Cam conjugate versus PBS
(P = 0.004). Only mice treated with
138H11-Cam showed a tumor shrinkage to minimal residues. In a second
experiment, lower doses of the 138H11-Cam conjugate were compared
with an antineuroblastoma mAb (ch14.18), confirming targeted killing of
RCC by the 138H11-Cam conjugate at tolerable toxicity in
vivo. In conclusion, these combined results encourage further
studies for targeted therapy of metastatic RCC with mAb 138H11
conjugates. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSStatistical AnalysisRESULTS AND DISCUSSIONREFERENCES |
|---|
3% of all human tumors. In
the United States alone, approximately 30,000 new cases and 12,000
deaths each year attributable to RCC have been reported from 1996
through 1998 (1, 2, 3, 4)
. In Germany, RCC belongs among the 12
most frequently diagnosed malignant tumors with an estimated
6,612 deaths in 1995, according to the tumor register in
Munich.4
The incidence is drastically increasing, and for the year 2000, a
worldwide mortality of 100,000 is expected (2, 3, 4)
. Once
the tumor has spread beyond the kidney, patients have a very poor
prognosis. The 5-year survival rate for the Robson states III and IV
are 15–35% and 0–10%, respectively (4)
. When local
recurrence occurs, most patients die within a few months. As tumors
restricted to the kidney rarely cause symptoms, up to 30% of patients
are diagnosed with metastatic disease, and another 40% of patients
experience a relapse after nephrectomy (3)
. For these
patients, there is no effective treatment available currently.
Conventional chemotherapy is ineffective due to several mechanisms of
high multidrug resistance of RCC, and radiotherapy is only useful for
palliation of tumor symptoms (2
, 5)
.
It has been known for a long time that RCC sometimes can regress
spontaneously. This has been attributed to immune mechanisms
(6, 7, 8)
. From this experience, systemic immunotherapy using
IFN-
, Interleukin-2, or a combination of both was developed.
However, results of large clinical trials have shown overall objective
responses <20% and of short duration, accompanied with severe
toxicities (8, 9, 10, 11)
.
We developed the mAb 138H11 against human GT (EC 2.3.2.2) for
possible use in differential diagnosis and targeted therapy of RCC
(12, 13, 14)
. This mAb stained over 99% of primary clear cell
and papillary RCCs on frozen sections. These two tumor types represent
over 90% of the malignant RCC. Oncocytomas (considered benign),
chromophobe, and Duct Bellini carcinomas were all negative
(12)
. mAb 138H11 has also proven useful for the
differential diagnosis of metastases when the primary tumor is not
known (15)
. In immunoscintigraphy,
99mTc-labeled 138H11 demonstrated specific uptake
into RCC in ex vivo perfused human tumor-bearing kidneys
(13)
. The enrichment of the labeled mAb in the tumor was
up to 20-fold compared with the kidney cortex tissue.
This specific tumor uptake can be explained by the localization
of the target antigen. In a normal kidney, GT is confined mostly to
the brush border membrane of the proximal tubules (12)
.
Similarly, in the liver, GT is located at the luminal part of the
bile canaliculi. Thus, in normal organs, the 138H11-antigen is not
accessible via the circulation. In contrast, in primary and
metastatic RCC, the enzyme is expressed over the entire tumor cell
surface, thus becoming accessible to mAb 138H11 in the circulation
(15
, 16) . In conclusion, a targeting effect with mAb
138H11 is attributable to a different antigen localization on tumor
cells compared with normal organs. This is unlike most other antibody
targeting strategies, which rely mostly on a higher antigen
concentration in tumor cells compared with normal tissues
(17
, 18)
.
Although tumor therapy with mAbs alone may result in impressive
clinical responses (17
, 19)
, increasing the response rate
requires the improvement of the antibody-effector function. This can be
done, for example, by creating immunotoxins, bispecific antibodies, and
cytokine fusion proteins or by the chemical coupling of cytotoxic drugs
(20, 21, 22)
. We decided to use the latter strategy and
coupled Cam to mAb 138H11 (21)
. Cam was rationally
designed on the basis of calicheamicin
I1 (23)
. Both
compounds belong to the chemical group of enediyne antibiotics:
naturally occurring substances that are able to damage DNA by
double-strand cleavage (22
, 24)
. Cam is several orders
of magnitude more potent than other chemotherapeutics (21
, 25)
, including the natural calicheamicin
I1 (20
, 23)
. In
this study, we evaluated for the first time the potential of a novel
chemoimmunoconjugate of mAb 138H11 and Cam in vivo.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSStatistical AnalysisRESULTS AND DISCUSSIONREFERENCES |
|---|
FACS
Trypsinized cells (5 x 104
in PBS) were incubated with 50 µg of purified mAb for 60 min at
4°C, washed with 1 ml of PBS containing 5% FCS, and then pelleted at
1000 x g for 5 min. This was followed by
incubation for 60 min at 4°C with FITC-labeled goat antimouse-IgG
(GAM-FITC; Becton Dickinson, Heidelberg, Germany). Cells were washed
once and fixed with 1% paraformaldehyde/PBS prior to FACS analysis
(FACScan, Becton Dickinson). Cells incubated with irrelevant mAbs,
ch14.18 (17)
, or secondary antibody only were used
as negative controls.
Conjugation of Cam to mAb 138H11 and ch14.18
The principle used to conjugate Cam to 138H11 involved the
activation of the amino group with SPDP (Pierce), followed by
disulfide exchange with iminothiolane-modified mAb 138H11 (Fig. 1)
. For activation, Cam (500 µg in 500 µl of DMF) was
reacted with 10 µl of SPDP stock (2.2 mg in 200 µl DMF) at 4°C
for 6 h and stored in small aliquots at -80°C until further
use. mAb 138H11 [10 mg in 2 ml of PBS (pH 8.3)] was reduced by adding
10 µl of iminothiolane (Pierce; 4.4 mg dissolved in 1 ml DMF) for
1 h under nitrogen. The reduced antibody was purified by size
exclusion chromatography using a NAP-5 column (Pharmacia)
preequilibrated with PBS. The conjugation of activated Cam
(Mr 1,464) with thiolated mAb
138H11 (Mr 155,000) was done at RT
for 1 h under nitrogen using a 2:1 molar ratio. Finally, solutions
were filter-sterilized. The purification grade was checked by HPLC
analysis. The number of drug molecules/mAb were two on the average as
calculated from the ratio of
A303/A280 in the HPLC
spectrum. The dosage of the 138H11-Cam conjugate for cytotoxicity
and in vivo experiments was calculated according to the
total amount of Cam. The conjugation of Cam with ch14.18 was done
as described above.
|
-SPDP; (b) the mixture of mAb
138H11 and Cam (both in concentrations corresponding to the
conjugate; (c), the 138H11-Cam conjugate; (d)
the Cam-ch14.18 conjugate; (e) 138H11; (f)
ch14.18; or (g) PBS, each in a total volume of 200 µl,
were added to the wells at 1:10 serial dilutions. The plates were
incubated for 72 h at 37°C and 5% CO2.
Visual scoring of cell viability was performed at 24, 48, and 72 h
after drug application. Cytotoxicity assays were performed using a
vital stain XTT assay. 10 µl of a 1.5-mg Phenazin
(Sigma-Aldrich, Deisenhofen, Germany)/ml aqua bidest solution
were added to 10 ml of a 5-mg XTT (Sigma-Aldrich)/ml RPMI stock
solution. This freshly prepared mixture was sterile-filtrated, and 40
ml were added to each well with the cells to be tested. Absorption was
measured at 450 nm after 2 h of incubation at 37°C with 5%
CO2. Absorption/concentration profiles were
obtained using the following 4-parameter fitting equation:
 |
where A and D describe the left and right asymptote of the sigmoid curve, respectively. B describes the slope, and C describes the inflection point of the curve. The inflection point corresponds to approximately the half-maximal effective dose, EC50 (21) .
Treatment of Human RCC Xenografts in Nude Mice
High Dose.
Five groups of eight female NMRI-nu/nu mice ("Swiss"; M & B, Ry,
Denmark) each were implanted with 2.5 x 106 Caki-1 cells s.c. The mice were treated with:
(a) PBS; (b) 138H11; (c) Cam (1
µg/kg); (d) the mixture of 138H11 and Cam (1 µg/kg);
and (e) the 138H11-Cam conjugate corresponding to a 20
µg/kg dose of Cam.
Low Dose.
Three groups of six mice each xenografted as above were treated with:
(a) PBS; (b) the 138H11-Cam (10 µg/kg)
conjugate; and (c) the ch14.18-Cam (10 µg/kg)
conjugate.
In both experiments (Fig. 4
and 5)
, treatment started one day after
tumor inoculation in agreement with many accepted experimental models
(20
, 27
, 28)
and was repeated every other day for three
times. Mice were observed, weighed, and measured individually
throughout the experiment. Each tumor was measured by a caliper in two
dimensions, and the tumor volume was calculated using an established
formula:
 |
|
|
|
Statistical Analysis |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSStatistical AnalysisRESULTS AND DISCUSSIONREFERENCES |
|---|
) or the Mann-Whitney
rank-sum test (SigmaStat, SPSS SCIENCE; Fig. 5
). |
RESULTS AND DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSStatistical AnalysisRESULTS AND DISCUSSIONREFERENCES |
|---|
120 mg of mAb 138H11 were purified.
Immunohistochemistry, gelelectrophoresis, and HPLC gel chromatography
revealed biological activity and a single band or peak, respectively
(not shown). Specific binding of purified mAb 138H11 as well as the
138H11-Cam conjugate to living RCC cells was determined in FACS
analysis. All analyzed human RCC cell lines such as Caki-1, Caki-2, and
A498 were positive (Fig. 2)
, whereas controls such as squamous cell carcinoma lines C33-a (Fig. 2)
and D36-1, fibroblasts, and the mouse RCC cell line RENCA were all
negative.
|
conjugate was active with an
EC50 of 4.5 x 1011 in the cytotoxicity assay using the Caki-1
cell line. This cytotoxicity was nearly two logs higher than that
observed for free or activated Cam and a mixture of mAb and Cam
as shown by a strong left shift of the curve in Fig. 3
. The latter control suggests that the increased toxicity of the
conjugate is due to a synergistic targeting effect and not just
additive. In contrast, on a neuroblastoma cell line (SKN-ML), the
cytotoxicity of the 138H11-Cam conjugate was comparable with
unconjugated Cam (Table 1)
. Used as a negative control, the chimeric antineuroblastoma antibody
ch14.18-Cam conjugate (20)
showed the opposite
reaction, demonstrating a strong increase in cytotoxicity on its
target cell line SKN-ML but not on RCC (Table 1)
.
|
|
-antibody Conjugates in Nude
Mice
, the 138H11-Cam conjugate was applied at a dose corresponding to 20
µg Cam/kg mouse body weight. Free Cam was given at a dose of 1
µg/kg only because of its known high toxicity (20)
. The
mice treated with the conjugate showed a strong reduction in total
tumor volume with a shrinkage to minimal residues (Fig. 4)
. This
shrinkage was highly significant in the t test on day 26
compared with the PBS control (P = 0.004).
Cam, mAb 138H11, and a mixture of both also showed some initial
tumor reduction, but here the tumors continued growth shortly after the
treatment was stopped. The reduction of tumor volume measured during
the first days is at least partially due to the fact that the
monitoring started 1 day after injection of the tumor cells. Thus, the
calculated initial tumor volume includes local swelling and the
injected buffer. However, when treatment started, the tumors already
represented palpable masses similar to established tumors.
In a second experiment (Fig. 5)
, the 138H11-Cam conjugate was applied at a 50% lower dose
corresponding to 10 µg Cam/kg and compared with a group treated
with a ch14.18-Cam conjugate. After an initial tumor shrinkage in
all mice similar to that in the first experiment, the tumors of the
PBS- and ch14.18-Cam conjugate-treated mice started to grow
strongly. In contrast, in the group of mice treated with the
138H11-Cam conjugate, the tumors continued to shrink. They did not
start regrowing before day 15. At this time point, the 138H11-Cam
group was highly statistically different from the PBS-treated group
using the t test (P = 0.004 on day
14). The 138H11-conjugate group continued to be statistically different
in the t test from the ch14.18-conjugate group until the end
of the experiment on day 25 (P = 0.024).
However, in the PBS-treated control group, three mice showed an
exorbitant tumor growth starting on days 3, 9, and 11, respectively,
with a tumor volume of 1598 mm3
on day 26 in the
first case. Therefore we plotted the medians of tumor volumes instead
of the means in Fig. 5
and used the nonparametric Mann-Whitney rank-sum
test for statistical evaluation on day 26. The 138H11-Cam group was
significantly different from both the ch14.18-Cam-treated
(P = 0.011) as well as the PBS-treated
(P = 0.035) group. In contrast,
ch14.18-Cam conjugate-treated mice were not significantly different
from the PBS control group (P = 0.945).
This clearly shows the significant difference between the
138H11-conjugate and the control antibody-conjugate in the treatment of
experimental RCC, confirming the targeting effect due to 138H11 as
already observed in vitro (Table 1)
and in Fig. 4
. Although
some tumors in the 138H11 conjugate group started to grow again around
days 15 to 20, until this point in time there was a clear reduction in
tumor size and a stoppage of tumor growth, in contrast to the
controls. In future experiments, the regrowth of tumors may be
prevented by differing schedules of treatments, e.g.,
another injection of the 138H11-Cam conjugate at day 15. In
conclusion, the second experiment (Fig. 5)
strongly supports the
results of the first experiment (Fig. 4)
.
Toxicity of Free and Conjugated Cam
Toxicity was monitored as a function of change in the individual
and total body weight of mice in each treatment group (Fig. 6)
. Initially, a mouse given 138H11-Cam doses equivalent to 20
µg Cam/kg every other day to determine the MTD did not show
symptoms of toxicity before the sixth injection. However, in the first
experiment with four injections of the 138H11-Cam conjugate at the
high dose (20 µg/kg), this treatment group showed an 23% weight
reduction within 12 days after treatment (Fig. 6a)
. Five of eight mice
died on days 9, 9, 20, 20, and 26, respectively. Comparing these data
with the apparent MTD and those published previously (20)
,
the high death rate was unexpected. In the neuroblastoma model, even at
a 50% weight loss (30 µg/kg dose) only 2 of 4 mice died, indicating
that the 30 µg/kg dose was just beyond the MTD (20)
. In
contrast, at a 10 µg/kg dose, mice with an initial decrease of
20% in body weight comparable with our experiment started recovery
2 weeks after the last injection. In the neuroblastoma model, the
20% weight reduction was very well tolerated by the animals
(20)
, encouraging us to continue the first experiment
until day 26.
|
, although the other mice of this treatment group did not
show signs of severe toxicity (Fig. 6a). This indicates that
a 1-µg/kg dose of free Cam is close to the MTD as shown previously
(20)
. At 10 µg/kg of free Cam, all mice had died in
the murine neuroblastoma study (20)
.
It should be noted also that all other mice, including the PBS-group in
experiment 1, showed a weight reduction between days 7 and 11, with
rapid recovery (Fig. 6a)
. This may be indicative of an
unrelated event, such as a general infection, leading to a higher
sensitivity for free and conjugated Cam just during the phase when
the treated-mice groups experienced the fastest weight reduction (Fig. 6a)
.
This conclusion is supported by the second experiment (10 µg/kg
dose), where both conjugate-treated mice groups showed two phases of a
small weight reduction of less than 10% between days 0–4 and 6–10,
respectively (Fig. 6b)
. These weight reductions were very
well tolerated and each time immediately followed by a quick recovery.
Thus, the 10 µg/kg dose resulted in a much lower toxicity for both
the 138H11-Cam and the ch14.18-Cam conjugates in the RCC model
compared with the study published previously (20)
.
Conclusion
The effective treatment of metastasized RCC remains one of the
major challenges in urological oncology, because RCC has been resistant
to all conventional as well as experimental therapeutics thus far. In
the current study, we evaluated the cytotoxic potential of a novel
chemoimmunoconjugate of mAb 138H11 and Cam for targeted therapy of
RCC in vitro and in vivo.
We demonstrated that the conjugate of 138H11 and Cam is highly toxic
for the Caki-1 RCC cell line in vitro, with an
EC50 of 4.5 x 10-11. This is 2 logs more effective than
that of free Cam or the ch14.18-Cam conjugate in the same
experimental setting (Table 1)
, indicating a stable construct on one
hand and a specific targeting effect already in vitro. This
is in contrast to a previous study (21
, 25)
, where we used
a pH degradable 3,3'-dithiobis[sulfosuccinimidyl
propionate] (DTSSP) linker for coupling Cam to mAb 138H11.
With this linker, the cytotoxicity in vitro was reduced by
approximately 1 log compared with free Cam.
In vivo, the 138H11-Cam conjugate was very effective in
reducing the tumor size and preventing (Fig. 4)
or significantly
delaying (Fig. 5)
the regrowth of residual tumor cells, in contrast to
the controls. That this tumor-inhibitory effect is due to specific
targeting by mAb 138H11 to the RCC was confirmed by comparing this
conjugate with an antineuroblastoma antibody conjugate, ch14.18-Cam.
The 138H11-Cam conjugate demonstrated a strong tumor inhibition that
was statistically significant from that of ch14.18-Cam (Fig. 5)
.
Overall, the 138H11-Cam conjugate demonstrated it usefulness for
targeted therapy of RCC. One problem that needs to be addressed
carefully in dose-finding studies is the potential systemic toxicity of
Cam conjugates. The first clinical trial with calicheamicin
I1 suggested that systemic
toxicity can be controlled in humans (29)
. However, the
very high efficiency in killing RCC cells with an
EC50 of 4.5 x 10-11 several logs above conventional
cytostatics (21)
may overcome the multidrug resistance of
RCC. In addition, the mAb itself may also stimulate the local immune
response, an effect that is important in controlling any surviving
tumor cells. As the primary tumor can be easily removed by surgery in
most patients, a targeted approach with 138H11-Cam holds promise for
treatment of RCC in a minimal residual disease setting. High local
doses of the 138H11-Cam conjugate may kill small metastases and
residual tumor cells.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by a grant of the Dr. Mildred
Scheel-Stiftung/Deutsche Krebshilfe to P. F. The study contains data
from the dissertation of K. K. 
2 To whom requests for reprints should be
addressed, at Molecular Biology, Charité Children’s Hospital,
Ziegelstr. 5-9, 10117 Berlin, Germany. Phone: 49-30-2802-6583/6508;
Fax: 49-30-2802-6528, E-mail: peter.fischer{at}charite.de
3 RCC, renal cell carcinoma; mAb, monoclonal
antibody; GT, -glutamyltransferase; Cam, Calicheamicin
I1; HPLC, high-performance liquid
chromatography; FACS, fluorescence-activated cell sorting; SPDP,
N-succinimidyl 3-[2-pyridylithio]propionate; XTT,
2,3-bis[2-Methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxanilide;
MTD, maximal tolerated dose.
4 Internet address: www.krebsinfo.de.
Received 12/ 1/99. Accepted 8/28/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSStatistical AnalysisRESULTS AND DISCUSSIONREFERENCES |
|---|
-1b compared with placebo in metastatic renal-cell carcinoma. N. Engl. J. Med., 338: 1265-1271, 1998.-glutamyltransferase. Cancer Immunol. Immunother., 33: 382-388, 1991.[Medline]
-glutamyltransferase. Cancer Immunol. Immunother., 35: 283-288, 1992.[Medline]
-glutamyltransferases from human kidney and renal cell carcinoma applying monoclonal antibodies. Clin. Chim. Acta, 191: 185-200, 1990.[Medline]
-glutamyltransferase isoforms from human kidney and renal cell carcinoma. Tumor Biol., 17: 369-377, 1996.
(I)1 effectively suppresses growth and dissemination of liver metastases in a syngeneic model of murine neuroblastoma. Cancer Res., 58: 2925-2928, 1998. to renal cell carcinomas. Tumor Target, 4: 271-277, 1999.
I: rationally designed caliceamicin with extremely potent and selective DNA cleaving properties and apoptosis inducing activity. Angew. Chem. Int. Ed., Engl.,, 33: 183-186, 1994.
-glutamyltransferase provides a possible tool for targeting calicheamicin to renal cell carcinomas. Adv. Exp. Med. Biol., 451: 431-436, 1998.[Medline]
This article has been cited by other articles:
|
|
 |
|
  M. M. Uttenreuther-Fischer, J. A. Kruger, and P. Fischer Molecular Characterization of the Anti-Idiotypic Immune Response of a Relapse-Free Neuroblastoma Patient following Antibody Therapy: A Possible Vaccine against Tumors of Neuroectodermal Origin? J. Immunol., June 15, 2006; 176(12): 7775 - 7786. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Wu, K. Esuvaranathan, and R. Mahendran Monitoring the Response of Orthotopic Bladder Tumors to Granulocyte Macrophage Colony-Stimulating Factor Therapy Using the Prostate-Specific Antigen Gene as a Reporter Clin. Cancer Res., October 15, 2004; 10(20): 6977 - 6984. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
