| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Advances in Brief |
Cancer Research Campaign Targeting and Imaging Group, Department of Oncology, Royal Free and University College London Medical School, London NW3 2PF [M. P. N., S. K. S., K. D. B., A. J. G., N. C., D. O., R. H. J. B.], with the Cancer Research Campaign Phase I/II Trials Committee, and Cancer Research Campaign Centre for Cancer Therapeutics, Institute of Cancer Research, 15 Cotswold Road Sutton, Surrey SM2 5NG [C. J. S., J. M., S. M. S.], United Kingdom
 |
ABSTRACT |
|---|
 Top ABSTRACT IntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
|
Introduction |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
A convincing demonstration that such a complex system can be developed for clinical use requires evidence that each of the components of ADEPT functions by the mechanisms proposed. This can be provided by measuring antibody-enzyme conjugate concentration, enzyme activity, and prodrug and drug levels in tumor and normal tissues. We describe a clinical trial in which these parameters were measured along with conventional measurements of toxicity, efficacy, and immunogenicity.
|
Materials and Methods |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
SB43 is an IgG1 mouse monoclonal antibody against CPG2 (11) . Purified antibody samples were galactosylated (SB43-gal) according to a modification of the method described by Mattes (12 , 13) . CMDA, a synthetic benzoic acid mustard prodrug, undergoes cleavage of its terminal glutamic acid residue by CPG2 to generate the active alkylating agent 4-[(2 chloroethyl)[2(mesyloxy)ethyl]-amino] benzoic acid (14) . Quality and safety of the product were determined using the Cancer Research Campaign Operation Manual (15) .
Freeze-dried CMDA was reconstituted in 1.5 ml of DMSO, a solution that
is stable in B-D syringes for >12 h. The CMDA/DMSO solution was
injected in free running 1.26% sodium bicarbonate. The treatment
schedule is given in Table 1
.
|
HAMA and anti-CPG2 antibody (HACPG2A) response were measured by ELISA (19) .
Patients had torso SPECT gamma camera imaging performed at intervals of up to 72 h after radiolabeled conjugate injection on an IGE Gemini 700 camera. Enzyme in tumor and normal tissue was estimated by measuring radioactivity as a percentage of injected radioactivity. SPECT images were reconstructed using General Electric filtered back projection software and corrected for decay, attenuation, and Compton scatter. Estimates of radioactivity per unit mass were then made using region of interest analysis as described previously (18) .
Prodrug and active drug concentrations in plasma were determined by HPLC, and their presence was confirmed by liquid chromatography mass spectroscopy (20 , 21) and CPG2 in plasma and tumor biopsies by HPLC (21 , 22) . Fractional clearance estimates were made assuming an exponential clearance and a single-compartment model. An exponential curve fit using the least sum of squares method was used. Fractional half-life of clearance was calculated using the equation t1/2 = ln 2/K where K is slope of ln plasma concentration versus time. These results are descriptive only as the concentration of the drug at the time of administration (C0) would theoretically be zero, and drug is being continually released from the prodrug by tumor localized antibody-enzyme conjugate. The pharmacokinetic details of the prodrug/drug system generated in this trial are reported in detail elsewhere (21) . Area under the concentration versus time curve (AUC) to infinity was calculated by using the trapezoidal rule and by adding a tail by integrating the exponential curve-fit from the last measured time point to infinity.
Standard WHO criteria for response were used. Standard National Cancer Institute CTC (23) were used to evaluate toxicity. Survival times were calculated from the start of treatment. Duration of response was calculated from the onset of response to the date of disease progression.
Patients.
The trial was performed with local ethical committee, Department of
Health, United Kingdom, and Administration of Radioactive Substances
Committee approval and according to Good Clinical Practice under the
auspices of the Cancer Research Campaign Phase I/II Committee’s
Targeting Trials Group, by whose trials office the clinical data were
monitored. All patients gave written informed consent. Eligibility
criteria were unresectable or metastatic, histologically proven
colorectal carcinoma; no antitumor treatment in the previous 4 weeks;
measurable disease by plain X-ray, CT, or ultrasound scan; age >20 years; life expectancy >4 months; Eastern Cooperative
Oncology Group performance status of 0–2; and normal hematological,
renal, and cardiac indices unless abnormal due to tumor. Pretreatment
serum CEA levels were required to be raised but <1000 µg
l-1; if they were not raised, then CEA had to be
demonstrated immunohistochemically on tumor biopsy (24)
.
All patients had negative HAMA titers and negative reactions to intradermally administered A5CP and SB43-gal. All patients had previously been treated with 5-fluorouracil-based chemotherapy regimens and had either relapsed or shown no response. All patients had a triple lumen Hickman catheter inserted under direct radiographic screening into a subclavian vein. Cyclosporin A was given starting 2 days before A5CP administration, initially as a continuous i.v. infusion at 5 mg kg-1 day for 7 days and then p.o. at 15 mg kg-1/day in divided doses for 7 additional days. Dose adjustments were made to keep blood levels in the range 150–350 Ng ml-1. Ondansetron and s.c. cyclizine were used as required. All patients were given a thyroid blockade against 131I using potassium iodide.
|
Results |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
.
Concentrations of the components of ADEPT were measured in tumor
and normal tissues to determine whether conditions for effective
therapy were met. The results were related to toxicity and of the
components of ADEPT were measured in tumor and normal tissues to
determine efficacy assessments.
|
. Indirect measurements of enzyme
activity in plasma by HPLC and spectrophotometric assay correlated with
the estimates of enzyme made from measurements of radiolabeled A5CP
until clearing antibody was given. Thereafter, active enzyme
measurements gave lower values, consistent with enzyme inactivation by
SB43-gal in addition to clearance of complexed A5CP. The concentration
of enzyme was less than the limits of detection (5 x
10-5 units of enzyme ml-1
of plasma) 48 h after A5CP administration.
|
,
reaching median peak levels of 6.8% of the injected radioactivity
kg-1 4 h after administration. Initially,
blood and other normal tissue levels of radioactivity exceeded those in
the tumor, but this trend was reversed after SB43-gal
administration (Fig. 3)
.
|
|
. This showed that the enzyme in the tumor was
inaccessible to inactivation by SB43-gal and that gamma camera
estimates of enzyme concentration in tumor were valid. Normal liver was obtained in biopsies from two patients, and no enzyme activity was detectable in them by HPLC. However, there was retention of some radioactivity, suggesting that enzyme activity was lost in the residual radiolabeled material.
Selectivity.
Tumor:plasma ratios of enzyme were >10,000:1 (based on undetectable
levels in the blood) at the time when prodrug administration was
started. The validation of gamma camera estimates of enzyme
concentration given by the biopsy studies supports the view that
effective enzyme levels persist in the tumor for at least up to 85 h. Tumor:liver ratios in the two patients biopsied also exceeded
10,000:1 (based on concentrations below the limits of detection in
normal liver).
Prodrug and Drug Levels in Plasma.
Prodrug cleared from plasma with a biological half-life of 16 min
(range, 5–27 min; Fig. 4a
).
Drug was detected in all cases within 3 min of the end of prodrug
administration; the half-life recorded for drug was longer than for the
prodrug at 46 min (range, 7–85 min; Fig. 4b
). There was
variation in the concentrations of prodrug and drug between patients
and between days in the same patient. Calculated area under the curve
to infinity correlated linearly for prodrug and drug for a given
administration and patient. This showed that the prodrug was activated
by the enzyme, and the available data on enzyme distribution suggest
that this was principally occurring in the tumor with diffusion of drug
into the circulation.
|
Immune Response.
HAMA and human anti-CPG2 antibody were found in all patients after 2
weeks, preventing further therapy.
Responses.
There was evidence of antitumor activity. One patient had a partial
response lasting 4 months, six patients had stable disease after
previous tumor progression with a median progression-free survival of 4
months (range, 2–16 months), and one of these patients had a decline
in serum CA19/9 levels sustained for 4 months. Three patients had
progressive disease. The tumor response in the patient with a partial
response is shown in Fig. 5, a–d
.
|
|
Discussion |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
was in the range that produced an
antitumor effect in animal models (3)
. The very favorable
tumor:plasma and tumor:liver ratios of enzyme recorded (>10,000:1)
make it unlikely that there is a significant amount of enzyme
remaining in normal tissues at the time of prodrug administration. The levels of radioactivity in the tumor gave a reasonably accurate indication of the enzyme level in the tumor at the time of prodrug administration, as judged by comparison of enzyme measurements in biopsies and SPECT gamma camera data. Although there was only approximately 1.5% and 1% of injected radioactivity kg-1 in the tumor after 52 and 85 h, respectively, this represented a satisfactory level of active enzyme for prodrug activation, as indicated above. Blood and other normal tissue levels of radioactivity remained close to those in the tumor until 85 h after administration. However, enzyme activity measurements in plasma and liver biopsies after the second antibody administration showed no measurable enzyme activity. This was consistent with inactivation of enzyme activity in normal tissues by the antienzyme antibody SB43-gal, whereas inactivation did not occur in the tumor. Thus, the levels of radioactivity in blood and normal tissues overestimated enzyme activity after the administration of SB43-gal.
The drug present in plasma was probably generated in the tumor, and it is interesting that the two patients with the highest plasma drug levels had large burdens of tumor. Treating smaller tumors may be advantageous because the potential for release of the drug into the circulation would be reduced in proportion to the targeted tumor volume, whereas the drug concentration in the tumor would be similar.
The characteristics of the CPG2 enzyme in prodrug activation are known
(25)
and indicate a Km
for prodrug conversion of 3 µM. This optimal
level of prodrug concentration was maintained for approximately 2 h in the plasma (Fig. 4b
). Also the
IC50 of the drug is 200 µM(25
), and this value was never exceeded in the
blood in any of the study patients (Fig. 4b
). Despite this,
drug in the circulation appeared to cause myelosuppression. Given that
no active enzyme was found in plasma by HPLC assay at any of the time
points when prodrug was given, it is likely that the presence of drug
in the circulation was the result of "leakback" from tumor. It is
also probable that higher concentrations of drug were present in the
tumor than were measured in the blood, and it is possible that a change
in prodrug regimen (for instance, to an infusion over 3 or more days)
would prevent plasma drug concentration from rising to a toxic level.
The possibility that the prodrug itself caused myelosuppression was
excluded in a previous study in which the prodrug was given alone
(26)
.
The prodrug CMDA is converted to the cytotoxic parent drug 4-[(2-chloroethyl)[2-(mesyloxy)ethyl]amino]benzoic acid (14) , which was found to be most effective in vivo in ADEPT experimental models, leading to complete regressions in transplanted human tumor xenografts that were resistant to all conventional cytotoxic agents (2) . Relevant antibody-enzyme conjugates (2000 units/kg, i.v. injection) were administered to animals with transplanted choriocarcinoma xenografts, followed 72, 94, and 99 h later by CMDA (400 mg/kg, i.v. injection). Control groups of animals received saline alone, CMDA alone, or irrelevant antibody-enzyme conjugate followed by CMDA at the same doses and time points as the test groups. All control animals were dead by day 110, whereas 9 of 12 of the ADEPT test animals were still tumor free at day 300 (P < 001).
It is not known how the relatively small amount of drug leads to bone marrow suppression 25–35 days after injection. However, we have noticed that after CMDA incubation with cells in culture, the cells do not die immediately but appear to be primed for death and die later.4
We appreciated that the CMDA-derived drug might have a long half-life from the outset, but CMDA was a practical prodrug for synthesis and testing of the principles of ADEPT. The relatively long half-life of the drug found in plasma is consistent with it causing the dose-limiting myelosuppression. ADEPT with the same antibody and enzyme but with a drug with a shorter half-life has been shown to be an effective therapy for colorectal cancer in an animal model and is being developed for clinical use (27 , 28) .
This study shows how measurement of the parameters required for the function of ADEPT gives insight into its substantial ability for selective delivery of cancer therapy. The tumor responses demonstrate that generation of an alkylating agent at the tumor site can overcome the drug resistance usually seen with this class of drug in colorectal cancer (29) . Although it was not possible to measure drug directly in the tumor because of the small size of the biopsies, the plasma drug levels and the absence of enzyme in normal tissues suggest that this was the result of generation of high concentrations of drug in the tumor. Conventional Phase I clinical trials in which tumor response and toxicity are studied in dose escalation studies are unlikely to be adequate for investigating such a complex system. For instance, a poor antitumor effect at the maximum tolerated dose of the prodrug could be caused by inadequate levels of enzyme in the tumor, excess enzyme in normal tissues, failure to continue administering prodrug while favorable tumor and normal tissue enzyme levels persist, an inappropriate prodrug regimen, or primary drug resistance. Little could be done about the latter, but knowledge of the first four components could lead to re-design of a clinical protocol to overcome the problems.
The immunogenicity of CPG2 is not surprising, considering its bacterial origin. It has the important advantage over most mammalian enzymes that it has no human equivalent that could activate the prodrug endogenously. Administration of cyclosporin A delays the production of a human antibody response to A5B7 antibody (30) . With ADEPT, production of human antibodies directed against CPG2 and A5B7 antibody limits therapy to two or three doses, but this does not appear to prevent useful antitumor activity, as shown here and in a previous study (26) . The proteins used in ADEPT studies were expected to be immunogenic in patients. In a previous clinical trial of ADEPT (26) with SB43gal and CMDA, 11 patients received antibody-enzyme conjugate without any immunosuppressive agent. All patients had detectable HAMA and anti-CPG2 antibodies in serum within 10 days after a single treatment with antibody-enzyme conjugate (19) . Six patients received cyclosporin A 48 h before the ADEPT regimen (31) . Two patients, who had very large hepatic metastases, received cyclosporin A p.o. and developed fatal hepato-renal failure after the first cycle of therapy. Subsequent patients receiving cyclosporin A by continuous i.v. infusion showed temporary increases in creatinine and urea levels. Two patients had antibody responses after a second cycle of therapy. Two patients had no detectable antibody responses until cyclosporin A was discontinued and were able to receive three cycles of ADEPT during a 21-day period.
Other immunosuppressive or tolerizing agents may also be considered for their potential to delay the antibody response (32 , 33) . Humanization of the antibody may reduce immunogenicity, and it is possible that less immunogenic enzymes can be identified.
The data presented here support the proposed mechanism of action of ADEPT and justify additional studies to develop the system for treatment for colorectal and other cancers.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by the Cancer Research Campaign, The
Ronald Raven Chair in Clinical Oncology Trust, and the Royal Free
Hampstead NHS Trust. Zeneca PLC (Macclesfield, Cheshire,
United Kingdom) supplied a grant funding technical and nursing support
during this trial. 
2 To whom requests for reprints should be
addressed, at Department of Oncology, Royal Free and University College
Medical School, University College London, Royal Free Campus, Rowland
Hill Street, London NW3 2PF, United Kingdom. Phone: 44-171-794-0500,
ext. 5488; Fax: 44-171-794-3341; E-mail: rjhb{at}rfhsm.ac.uk
3 The abbreviations used are: ADEPT,
antibody-directed enzyme prodrug therapy; CEA, carcinoembryonic
antigen; CPG2, carboxypeptidase G2; A5CP, A5B7 F(ab')2
antibody to CEA conjugated to CPG2; CMDA, benzoic acid
mustard-glutamate prodrug; HAMA, human antimouse antibody; SPECT,
single photon emission computerized tomographic; HPLC, high-pressure
liquid chromatography; CTC, Common Toxicity Criteria; CT, computerized
tomogram.
Received 6/23/99; revised 11/16/99; accepted 11/24/99.
|
REFERENCES |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. Fang and D. Sun Predictive Physiologically Based Pharmacokinetic Model for Antibody-Directed Enzyme Prodrug Therapy Drug Metab. Dispos., June 1, 2008; 36(6): 1153 - 1165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. F. Fidarova, E. El-Emir, G. M. Boxer, U. Qureshi, J. L.J. Dearling, M. P. Robson, R. H.J. Begent, K. R. Trott, and R. B. Pedley Microdistribution of Targeted, Fluorescently Labeled Anti-Carcinoembryonic Antigen Antibody in Metastatic Colorectal Cancer: Implications for Radioimmunotherapy Clin. Cancer Res., May 1, 2008; 14(9): 2639 - 2646. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kogelberg, B. Tolner, S. K. Sharma, M. W Lowdell, U. Qureshi, M. Robson, T. Hillyer, R. B. Pedley, W. Vervecken, R. Contreras, et al. Clearance mechanism of a mannosylated antibody-enzyme fusion protein used in experimental cancer therapy Glycobiology, January 1, 2007; 17(1): 36 - 45. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Mayer, R. J. Francis, S. K. Sharma, B. Tolner, C. J. Springer, J. Martin, G. M. Boxer, J. Bell, A. J. Green, J. A. Hartley, et al. A Phase I Study of Single Administration of Antibody-Directed Enzyme Prodrug Therapy with the Recombinant Anti-Carcinoembryonic Antigen Antibody-Enzyme Fusion Protein MFECP1 and a Bis-Iodo Phenol Mustard Prodrug. Clin. Cancer Res., November 1, 2006; 12(21): 6509 - 6516. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. A. Harding, A. D. Liu, M. Stickler, O. J. Razo, R. Chin, N. Faravashi, W. Viola, T. Graycar, V. P. Yeung, W. Aehle, et al. A {beta}-lactamase with reduced immunogenicity for the targeted delivery of chemotherapeutics using antibody-directed enzyme prodrug therapy Mol. Cancer Ther., November 1, 2005; 4(11): 1791 - 1800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Schepelmann, P. Hallenbeck, L. M. Ogilvie, D. Hedley, F. Friedlos, J. Martin, I. Scanlon, C. Hay, L. K. Hawkins, R. Marais, et al. Systemic Gene-Directed Enzyme Prodrug Therapy of Hepatocellular Carcinoma Using a Targeted Adenovirus Armed with Carboxypeptidase G2 Cancer Res., June 15, 2005; 65(12): 5003 - 5008. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Sharma, R. B. Pedley, J. Bhatia, G. M. Boxer, E. El-Emir, U. Qureshi, B. Tolner, H. Lowe, N. P. Michael, N. Minton, et al. Sustained Tumor Regression of Human Colorectal Cancer Xenografts Using a Multifunctional Mannosylated Fusion Protein in Antibody-Directed Enzyme Prodrug Therapy Clin. Cancer Res., January 15, 2005; 11(2): 814 - 825. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Michael and M.M. Doherty Tumoral Drug Metabolism: Overview and Its Implications for Cancer Therapy J. Clin. Oncol., January 1, 2005; 23(1): 205 - 229. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Cortez-Retamozo, N. Backmann, P. D. Senter, U. Wernery, P. De Baetselier, S. Muyldermans, and H. Revets Efficient Cancer Therapy with a Nanobody-Based Conjugate Cancer Res., April 15, 2004; 64(8): 2853 - 2857. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. F. Medzihradszky, D. I.R. Spencer, S. K. Sharma, J. Bhatia, R. B. Pedley, D. A. Read, R. H.J. Begent, and K. A. Chester Glycoforms obtained by expression in Pichia pastoris improve cancer targeting potential of a recombinant antibody-enzyme fusion protein Glycobiology, January 1, 2004; 14(1): 27 - 37. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Friedlos, L. Davies, I. Scanlon, L. M. Ogilvie, J. Martin, S. M. Stribbling, R. A. Spooner, I. Niculescu-Duvaz, R. Marais, and C. J. Springer Three New Prodrugs for Suicide Gene Therapy Using Carboxypeptidase G2 Elicit Bystander Efficacy in Two Xenograft Models Cancer Res., March 1, 2002; 62(6): 1724 - 1729. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Xu and H. L. McLeod Strategies for Enzyme/Prodrug Cancer Therapy Clin. Cancer Res., November 1, 2001; 7(11): 3314 - 3324. [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 |
).
The arrow indicates that the level was below the limits
of detection of this assay (- - - -). Levels in tumor (
) were
estimated from SPECT gamma camera imaging. Each point is the median of
10 patients. The range includes 60% of data points. The enzyme level
in tumor obtained from tumor biopsy and measured by HPLC (
) at
50.3 h is the median of five patients, and the range includes 60%
of data points.
