Synthesis and Biological Evaluation of EC20: A New Folate-Derived

Endocyte, Inc., 1205 Kent Avenue, West Lafayette, Indiana 47906. ... performed in M109 tumor-bearing Balb/c mice confirmed that 99mTc-EC20 predominant...
4 downloads 0 Views 191KB Size
1200

Bioconjugate Chem. 2002, 13, 1200−1210

Synthesis and Biological Evaluation of EC20: A New Folate-Derived, 99mTc-Based Radiopharmaceutical Christopher P. Leamon,* Matthew A. Parker, Iontcho R. Vlahov, Le-Cun Xu, Joseph A. Reddy, Marilynn Vetzel, and Nikki Douglas Endocyte, Inc., 1205 Kent Avenue, West Lafayette, Indiana 47906. Received June 5, 2002; Revised Manuscript Received August 3, 2002

A new peptide derivative of folic acid was designed to efficiently coordinate 99mTc. This new chelate, referred to as EC20, was found to bind cultured folate receptor (FR)-positive tumor cells in both a time- and concentration-dependent manner with very high affinity (KD ∼ 3 nM). Using an in vitro relative affinity assay, EC20 was also found to effectively compete with 3H-folic acid for cell binding when presented either alone or as a formulated metal chelate. Following intravenous injection into Balb/c mice, 99mTc-EC20 was rapidly removed from circulation (plasma t1/2 ∼ 4 min) and excreted into the urine in a nonmetabolized form. Data from γ scintigraphic and quantitative biodistribution studies performed in M109 tumor-bearing Balb/c mice confirmed that 99mTc-EC20 predominantly accumulates in FR-positive tumor and kidney tissues. These results suggest that 99mTc-EC20 may be clinically useful as a noninvasive radiodiagnostic imaging agent for the detection of FR-positive human cancers.

INTRODUCTION

The field of nuclear medicine has been revitalized with the arrival of tissue-specific radiopharmaceutical targeting technologies. Ligands capable of concentrating at pathological sites have been derivatized with chelatorradionuclide complexes and then used as noninvasive probes for diagnostic imaging purposes. For example, vasoactive intestinal peptide, somatostatin analogues, monoclonal antibodies, vitamin B12, and even folic acid have all been used as ligands to localize radionuclides to tumors (1-10). Applications with monoclonal antibodies and various fragments thereof initially received the most attention because it was believed that precise, or perhaps superior, tumor-specific targeting might easily be achieved. Unfortunately, this approach was found to be technically challenging and inferior to other methods since (i) antibodies have prolonged circulation times due to their rather large molecular size (an unfavorable trait for imaging purposes), (ii) antibodies are expensive to produce, (iii) antibodies can be immunogenic, forcing their laborious humanization whenever multiple doses would be anticipated, and (iv) tumor to nontarget tissue ratios (T/NT) of antibody-linked radionuclides were essentially sub-optimal (11-13). Thus, more focus has recently been directed toward the use of smaller tumor-specific ligands that do not suffer from such limitations. Use of the vitamin folic acid for the targeting of bioactive agents to tumors has been effectively exemplified over the past eleven years (14). In fact, folate conjugates of radiopharmaceutical agents (1-6), MRI contrast agents (15), low molecular weight chemotherapeutic agents (16), antisense oligonucleotides and ribozymes (17-21), proteins and protein toxins (22-27), immunotherapeutic agents (28-31), liposomes with entrapped drugs (32-36), and plasmids (37-43) have all been successfully delivered to FR-expressing cancer cells. Chronologically, the first folic acid “conjugate” described * To whom correspondence should be addressed. Phone: (765) 463-7175. Fax: (765) 463-9271. E-mail: [email protected].

for in vivo tumor imaging was a histamine derivative containing 125-iodine (44). Although impressive tumor images were obtained with this conjugate, it was not considered a relevant clinical candidate owing to the longlived 125I radionuclide component. Subsequent reports described the synthesis and use of a deferoxamine-folate conjugate for tumor targeting (1, 2). Deferoxamine chelates 67Ga, a γ-emitting radionuclide that has a 78 h halflife. Favorable pharmacokinetic biodistribution profiles and high T/NT ratios were obtained with this agent in a tumor-bearing nude mouse model, however partial hepatobiliary clearance was also noted. Thus, further preclinical development was stopped due to anticipated problems in accurately imaging regio-abdominal locations in humans. This obstacle was easily overcome, however, by simply replacing the deferoxamine chelator with diethylenetriamine pentaacetic acid (DTPA), an efficient chelator of 111In (68 h half-life). 111In-DTPA-folate displayed similar tumor-accumulating properties to its deferoxamine predecessor, but its primary route of elimination was confirmed to be through the kidneys (3). Thus, 111In-DTPA-folate was selected for development, and phase I/II clinical trials were conducted at several US sites in 1999 and 2000 (14). There were no plans to commercialize 111In-DTPAfolate, mostly because the field of radiodiagnostic imaging had been transitioning toward using 99mTc-based probes. Nonetheless, clinical evaluation of 111In-DTPA-folate was continued for the purpose of obtaining valuable human biodistribution data while the development of a newer 99m Tc-based folate conjugate was concurrently underway. 99m Tc has been adopted as the preferred radionuclide for diagnostic imaging because (i) the radionuclide is easily obtained in a clinical laboratory from commercially available 99Mo-99mTc generators, (ii) the cost of producing large amounts of 99mTc is insignificant compared to the cost of 111In, and (iii) 99mTc has a much shorter (6 h) halflife, which allows higher radionuclide doses (∼25 mCi) to be administered, yielding higher resolution images

10.1021/bc0200430 CCC: $22.00 © 2002 American Chemical Society Published on Web 11/01/2002

Synthesis and Biological Evaluation of EC20

Bioconjugate Chem., Vol. 13, No. 6, 2002 1201

Scheme 1a

a Reagents and conditions: (i) 20% Piperdine, DMF; (ii) Fmoc-Asp(OtBu)-OH, PyBop, DIPEA, DMF; (iii) Boc-Dap(Fmoc)-OH, PyBop, DIPEA, DMF; (iv) Fmoc-D-Glu-OtBu, PyBop, DIPEA, DMF; (v) N10-TFA-Pte-OH, DIPEA, DMSO; (vi) TFAA, HSCH2CH2SH, iPr3SiH; (vii) H4NOH, pH 10.3.

without the risk of hazardous radiation exposure to vital organs. Several folate-based 99mTc conjugates have already been described in the literature. For example, folate conjugates of 99mTc-6-hydrazinonicotinamido-hydrazido (HYNIC) (5), 99mTc-12-amino-3,3,9,9-tetramethyl-5-oxa4,8 diaza-2,10-dodecanedinoe dioxime (OXA) [6], 99mTcethylenedicysteine (4), and even 99mTc-DTPA-folate (45), have all shown promising in vivo tumor uptake qualities. The objective of this study, however, was to develop an alternative, proprietary 99mTc-folate conjugate that would be suitable for clinical development. Thus a new agent, herein referred to as EC20, was prepared and characterized. Detailed analytical and preclinical biological data were collected, and much of the data used to support an Investigational New Drug application is reported below. EXPERIMENTAL PROCEDURES

Materials. N10-Trifluoroacetylpteroic acid was purchased from Eprova AG, Schaffhausen, Switzerland. Peptide synthesis reagents were purchased from NovaBiochem and Bachem. 99mTc Sodium Pertechnetate was supplied by Syncor. [ReO2 (en)2]Cl was prepared according to Rouschias (46). Cellulose plates and DEAE ion exchange plates were purchased from J. T. Baker. Synthesis, Purification, and Analytical Characterization of EC20. EC20 is a folate-containing peptide consisting in sequence of pteroic acid (Pte), D-Glu, β-Ldiaminopropionic acid (βDpr), Asp, and Cys. EC20 was synthesized on an acid-sensitive Wang resin loaded with Fmoc-L-Cys (Trt)-OH, as detailed in Scheme 1. Benzotriazole-1-yl-oxy-tris-pyrrolidino-phosphoniumhexafluorophosphate (PyBOP) was applied as the activating reagent to ensure efficient coupling using low equivalents of amino acids. Fmoc protecting groups were removed after every coupling step under standard conditions (20% piperidine in DMF). After the last assembly step, the peptide was cleaved from the polymeric support by treatment with 92.5% trifluoroacetic acid containing 2.5% ethanedithiol, 2.5% triisopropylsilane, and 2.5% deionized water. This reaction also resulted in simultaneous removal of the t-Bu, Boc and trityl protecting groups. Finally, the trifluoroacetyl moiety was removed in aqueous ammonium hydroxide to give EC20. The crude EC20 product was purified by HPLC using an Xterra RP18 30 × 300 mm 7 µm column (Waters),

mobile phase 32 mM HCl (A) and MeOH (B), and gradient conditions starting with 99% A and 1% B and reaching 89% A and 11% B in 37 min at a flow rate of 20 mL/min. Under these conditions, EC20 monomer typically eluted at 14.4 min, whereas EC20 disulfide dimer (minor contaminant) eluted at 16.8 min. A 2 mg sample of HPLC-purified EC20 was dissolved in 0.62 mL of D2O, and a 500 MHz 1H NMR spectrum was collected. Table 1 lists the chemical shifts, signal shapes, and J values for all nonexchangeable protons of the EC20 molecule. EC20 was also analyzed by electrospray-mass spectrometry. Major positive ion peaks (m/z, relative intensity): 746.1 (M+H)+, 100; 747.1, 44; 556.8, 32; 570.8, 16. Importantly, the calculated molecular weight for the formula C29H35N11O11S is 745.2. Preparation of the Nonradioactive Reagent Vial. EC20 kits are used for preparation of the 99mTc-EC20 radioactive drug substance. Each kit contains a sterile, nonpyrogenic lyophilized mixture of 0.1 mg of EC20, 80 mg of sodium R-D-glucoheptonate, 80 mg of tin (II) chloride dihydrate, and sufficient sodium hydroxide or hydrochloric acid to adjust the pH to 6.8 ( 0.2 prior to lyophilization. The lyophilized powder is sealed in a 5 mL vial under an argon atmosphere. The kits are then stored frozen at -20 °C until use or expiration (current shelf life is >2 years). Importantly, the tin (II) component is required to reduce the added 99mTc-pertechnetate, while the sodium R-D-glucoheptonate component is necessary to stabilize the newly reduced 99mTc prior to its final chelation to the EC20 compound. Preparation of 99mTc-EC20. The procedure for chelating 99mTc to EC20 consists of the following steps. First, a boiling water bath containing a partially submerged lead vial shield is prepared. The top of an EC20 vial is swabbed with 70% ethanol to sanitize the surface, and it is placed in a suitable shielding container. Using a shielded syringe with 27-gauge needle, 1 mL of Sodium Pertechnetate 99mTc Injection (96%. Two additional 99mTc-chelating agents, 99mTc-EC14 (an EC20 analogue containing 1 additional Glu residue: PteD-Glu-D-Glu-βDpr-Asp-Cys) and 99mTc-EC28 (a non pteridine-containing control: benzoyl-D-Glu-D-Glu-βDprAsp-Cys), as well as 111In-DTPA-folate (3) were also prepared to >90% radiochemical purity and were coevaluated in this study. All solutions were diluted with either saline alone or a saline solution containing 100 equivalents of folic acid (for competition) such that the final radiopharmaceutical concentration was 10 µmol/mL. Animals received an approximate 40 µmol/kg i.v. dose of test article in 100 µL volume via a lateral tail vein during brief diethyl ether anesthesia. Four hours after injection, animals were sacrificed by CO2 asphyxiation and dissected. Selected tissues were removed, weighed, and counted to determine 99mTc distribution. CPM values were decay-corrected, and results were tabulated as % injected dose per gram of wet weight tissue. Gamma Scintigraphy. M109 tumor cells (1 × 106 per animal) were inoculated in the subcutis of the right axilla of Balb/c mice two weeks prior to the experiment. Animals received an approximate 50 µmol/kg i.v. dose of test article in 100 µL volume via a lateral tail vein during brief diethyl ether anesthesia. Four hours after injection, animals were sacrificed by CO2 asphyxiation and then placed on top of an image acquisition surface. Whole-body image acquisition was performed for 1 min at a count rate of 50-75 000 counts per min using a Technicare Omega 500 Sigma 410 Radioisotope Gamma Camera. All data were analyzed using a Medasys MS-DOS-based computer equipped with Medasys Pinnacle software. Urinary Excretion and Metabolism. The urinary HPLC speciation profile of 99mTc-EC20 was obtained using Balb/c mice. Mice (∼20 g each) were injected with 1 mCi (6.7 nmol) of 99mTc-EC20 via a lateral tail vein. Following a 1, 4, or 6 h time period, groups of two mice were euthanized by CO2 asphyxiation, and urine was collected. After filtration through a GV13 Millex filter, the radiochemical speciation was assessed using an HPLC system equipped with a Nova-Pak C18 3.9 × 150 mm column and a radiochemical detector. The system was isocratically eluted with 20% methanol containing 0.1% TFA at a flow rate of 1 mL/min. Serum Protein Binding. Fresh rat serum and commercial male human serum (type AB donors, Sigma Chemical Co.) were used to evaluate in vitro binding of 99mTc-EC20 to serum proteins. One minute after 99m Tc-EC20 was mixed with 1 mL of serum at room temperature, 0.3 mL of the serum solution was transferred to a clean Amicon Centrifree ultrafiltration device (30 000 NMWL) in triplicate. Within one minute of loading the centrifuge with the serum solution, the device was spun at 1000g for 20 min at 20 °C. Fifty microliter samples of the original solution, and of the filtrate from each device, were transferred to a clean tube and counted in an automatic γ counter. A control solution of 99mTc-EC20 mixed with 1 mL of normal saline was ultrafiltered in an identical fashion. The percentage of free 99mTc was calculated for each of the three samples. RESULTS

EC20 and Related Structures. The structure of the EC20 molecule is depicted in Figure 1. The folic acid

1204 Bioconjugate Chem., Vol. 13, No. 6, 2002

Figure 1. Structure of EC20.

moiety in the EC20 molecule is present to allow its highaffinity binding to cellular FRs. The other part of the molecule consists of a bifunctional peptide-based chelator, which provides the site of 99mTc coordination. Interestingly, the most successful class of bifunctional technetium chelators to date is represented by similar amine/amide thiolate compounds that coordinate TcO3+ via an (Namine/amide)4-x (Sthiolate)x donor atom set (49-56). Radiochemical Characterization. EC20 kits are used for preparation of the 99mTc-EC20 radioactive drug substance. Each kit contains a sterile, nonpyrogenic lyophilized mixture of EC20, sodium R-D-glucoheptonate, and tin (II) chloride that is sealed under an argon atmosphere. The final 99mTc-EC20 drug product formulation is then formed by the simple reconstitution of the lyophilized powder with 99mTc pertechnetate in saline followed by a brief heating step. As shown in Figure 2, HPLC analysis of the 99mTcEC20 formulation shows four radiochemical components, designated as peaks A-D. Peak A was confirmed to be free 99mTc, and this byproduct is typically present at 30kDa serum protein fraction in solutions of rat or human serum (69% and 72%, respectively). Importantly, since 99mTc-EC20 does preferentially accumulate within FR-positive tissues (vide infra), its apparent affinity for serum proteins appears to be inconsequential. Blood Clearance. We next evaluated the pharmacokinetic blood clearance of the 99mTc-EC20 formulation. Thus, Balb/c mice received an intravenous dose of 50 µg/kg 99mTc-EC20, and blood samples were collected at various time intervals thereafter. As shown in Figure 8, 99m Tc-EC20 was rapidly removed from circulation in the Balb/c mouse. The plasma half-life of this radiopharmaceutical was estimated to be ∼4 min, and less than 0.2% of the injected 99mTc-EC20 dose remained in circulation after 4 h (assuming that blood represents 5.5% of the total body mass). These data indicates that folate con-

Figure 4. Blocking of 3H-folic acid binding to KB cells with various folate-containing competitors. KB cells were incubated for 15 min on ice with 100 nM 3H-folic acid in the presence and absence of increasing competitor concentrations. (•) Folic acid; (9) EC20; (2) EC20:Re isomer 1; (1) EC20:Re isomer 2; (0) DTPA-folate. Error bars represent 1 standard deviation (n ) 3). Table 2. Relative Affinity Estimationsa IC50 (nM)

S.D.

RA

folic acid

test article

118

( 19

1.00

EC20 EC20:Re isomer 1 EC20:Re isomer 2

128 83 86

( 25 ( 16 (3

0.92 1.42 1.37

( 0.23 ( 0.36 ( 0.23

DTPA-folate

136

( 12

0.87

( 0.16

S.D.

a

Relative affinities (RA) were defined as the inverse molar ratio of compound required to displace 50% of 3H-folic acid bound to FR-positive KB cells. The relative affinity of folic acid was set to 1. Each test article was evaluated in triplicate.

jugates are rapidly removed from circulation following intravenous administration and that valuable tissue biodistribution data can be obtained after only a few hours postinjection without the concern for nonspecific tissue uptake due to blood-borne radioactivity. Tissue Biodistribution. The ability of 99mTc-EC20 to target tumors in vivo was assessed using a FR-positive M109 model (36). These tumor cells are syngeneic for the

1206 Bioconjugate Chem., Vol. 13, No. 6, 2002

Leamon et al.

Figure 5. Time-dependent association of 99mTc-EC20. KB cells were incubated with 10 nM 99mTc-EC20 for increasing periods of time at 37 °C. Following multiple washes, cells were harvested and counted for associated radioactivity. Error bars represent 1 standard deviation (n ) 3). Figure 8. Blood clearance of 99mTc-EC20 in Balb/c mice. Each animal received an intravenous dose of 50 µg/kg EC20 (67 nmol/ kg) in approximately 0.1 mL PBS during brief diethyl ether anesthesia. At the designated times postinjection, each animal was euthanized by CO2 asphyxiation, and blood was collected and counted for associated radioactivity. Error bars represent 1 standard deviation (n ) 3 animals).

Figure 6. Concentration-dependent association of 99mTc-EC20. KB cells were incubated for 2 h at 37 °C in the presence of increasing concentrations of 99mTc-EC20. Following multiple washes, cells were harvested and counted for associated radioactivity. Error bars represent 1 standard deviation (n ) 3).

Figure 7. Concentration-dependent association of 99mTc-EC20 “peak B”. KB cells were incubated for 2 h at 37 °C in the presence of increasing concentrations of “peak B” that was chromatographically isolated from the 99mTc-EC20 formulation. Following multiple washes, cells were harvested and counted for associated radioactivity. Error bars represent 1 standard deviation (n ) 3). (b) peak B; (O) peak B plus 1 mM folic acid.

Balb/c mouse, and they reproducibly form subcutaneous solid tumors within two weeks post inoculation. 99mTcEC14, which is structurally similar to 99mTc-EC20 except it contains one additional D-Glu residue (i.e., Pte-D-GluD-Glu-βDpr-Asp-Cys), 99mTc-EC28 (a nonpteroate containing control consisting of benzoyl-D-Glu-D-Glu-βDpr-

Asp-Cys), and the previously reported 111In-DTPA-folate radiopharmaceutical (3) were also evaluated in this bioassay. Importantly, the 99mTc-EC28 control agent should not bind to cell surface FRs because it lacks an essential pteridine ring moiety. Four hours after receiving an approximate 40 µmol/ kg i.v. dose, animals were euthanized, and selected tissues were removed, weighed, and counted to determine 99m Tc distribution. As shown in Table 3, the three “folate” containing radiopharmaceuticals, 99mTc-EC14, 99mTcEC20, and 111In-DTPA-folate, predominantly accumulated in the FR-positive tumor and kidneys; however, the kidneys concentrated a higher percent injected dose per gram of tissue (%ID/g) than did the tumor. Interestingly, the net tumor accumulation of 111In-DTPA-folate and 99m Tc-EC20 was nearly the same (19% and 17% ID/g, respectively), whereas the tumor uptake of 99mTc-EC14 was somewhat less at ∼10% ID/g. Regardless, all three agents displayed high tumor-to-blood ratios (>30:1). Folate-specific targeting was further demonstrated by two distinct methods. First, the accumulation of 99mTcEC14, 99mTc-EC20, and 111In-DTPA-folate in the FRpositive tumor and kidneys was effectively blocked (>94%) when these agents were co-administered with a 100-fold excess of folic acid. Second, the 99mTc-EC28 control agent failed to appreciably accumulate in the kidneys and tumor. Both observations provide evidence that an intact “folate-like” (or pteroate) moiety is required to afford targeted uptake and retention of these radiopharmaceutical agents into FR-positive tissues. Scintigraphic Imaging. The predominant uptake of 99m Tc-EC20 by the FR-positive M109 tumors and kidneys was further demonstrated using γ scintigraphy. As shown in Figure 9, a ventral image taken of a mouse 4 h after receiving a 50 µmol/kg i.v. dose of 99mTc-EC20 distinctively localizes the γ radiation to the two kidneys and the M109 tumor mass (shoulder region). No appreciable radiotracer was observed in other body tissues. Importantly, a similar image profile has been reported for the 111In-DTPA-folate radiopharmaceutical (3). Urinary Excretion and Metabolism. It was previously determined that the primary elimination route for 111In-DTPA-folate was via the urine (3). Because

Synthesis and Biological Evaluation of EC20

Bioconjugate Chem., Vol. 13, No. 6, 2002 1207

Table 3. Biodistribution of Folate Radiopharmaceuticals in Balb/c Mice Bearing Subcutaneous M109 Tumors. % injected dose per gram tissue (4 hr post intravenous injection)a 99mTc-EC14

99mTc-EC20

111In-DTPA-folate

99mTc-EC14

+ folic acid

99mTc-EC20

+ folic acid

111In-DTPA-folate

+ folic acid

99mTc-EC28

blood heart lung liver spleen intestine kidney muscle stomach tumor

0.31 ( 0.14 2.39 ( 0.64 2.08 ( 0.40 3.44 ( 2.19 2.68 ( 2.49 1.70 ( 0.55 98.0 ( 40.7 0.99 ( 0.28 1.47 ( 0.58 9.83 ( 2.77

0.19 ( 0.07 0.08 ( 0.01 0.15 ( 0.04 1.37 ( 0.98 2.99 ( 1.43 0.32 ( 0.11 5.94 ( 0.52 0.09 ( 0.11 0.10 ( 0.03 0.43 ( 0.52

0.34 ( 0.03 1.57 ( 0.26 2.22 ( 0.63 3.56 ( 0.25 0.95 ( 0.15 2.56 ( 0.61 138 ( 12.4 0.67 ( 0.20 1.45 ( 0.55 17.2 ( 1.02

0.09 ( 0.02 0.08 ( 0.01 0.31 ( 0.26 1.15 ( 0.22 0.38 ( 0.33 2.93 ( 1.49 5.64 ( 2.13 0.06 ( 0.02 3.35 ( 5.19 0.45 ( 0.18

0.21 ( 0.10 2.57 ( 0.82 1.72 ( 0.61 5.21 ( 2.63 3.30 ( 2.33 1.87 ( 0.69 191 ( 79.2 1.19 ( 0.48 1.62 ( 0.65 19.3 ( 5.86

0.09 ( 0.04 0.06 ( 0.02 0.09 ( 0.02 0.81 ( 0.03 1.46 ( 0.73 0.82 ( 0.14 3.14 ( 1.96 0.05 ( 0.04 0.25 ( 0.20 0.46 ( 0.42

0.06 ( 0.04 0.03 ( 0.01 0.05 ( 0.01 0.50 ( 0.26 0.60 ( 0.38 0.47 ( 0.19 0.62 ( 0.14 0.02 ( 0.01 0.21 ( 0.19 0.11 ( 0.06

tumor/blood

34.1 ( 7.41

2.00 ( 2.00

51.0 ( 8.20

4.70 ( 1.30

102 ( 43.4

5.00 ( 4.60

2.00 ( 0.50

a

Values shown represent the mean ( s.d. of data from three animals per cohort.

duration of this experiment. The amount of free 99mTc in the standard (peak A) was ∼2%. Importantly, peak B within this radiochemical profile is believed to be EC20 chelated to 99mTc at an unconventional, less stable position. However, the radioactivity measured in this fraction was not included in the overall radiochemical purity estimation for 99mTc-EC20. Collectively, these data indicate that the 99mTc-EC20 formulation remained stable in saline solution throughout this 6 h investigation. After 1 and 4 h postinjection into Balb/C mice, the radiochemical speciation profile of 99mTc-EC20 in the mouse urine did not change. The radioactivity present in the urine at 6 h postinjection, however, was too low to accurately assay by HPLC. The proportion of parent drug among radioactive species recovered in urine remained relatively constant at approximately 90% throughout the 4 h during which it could be quantitated. This value is very similar to the ∼93% purity of the standard, indicating that 99mTc-EC20 is predominately excreted into the urine in an un-modified form. DISCUSSION

Figure 9. Whole-body γ images (ventral view). Images were obtained 4 h following intravenous administration of 99mTc-EC20 to a Balb/c mouse bearing a subcutaneous folate receptorpositive M109 tumor. Only the kidneys (K) and tumor (T) exhibit significant accumulation of this radiotracer. 99m Tc-EC20 is a related folate-based radiopharmaceutical, we elected to evaluate its urinary excretion and metabolic profile using a radiochemical HPLC method. Thus, following a 1 mCi injection (6.7 nmol per mouse) of 99mTcEC20, mice were euthanized at various time intervals, and urine samples were collected for HPLC analysis. Similar to the HPLC profile shown in Figure 2, both the 99mTc-EC20 standard and the urine samples exhibited four radioactive peaks. As shown in Table 4, the radiochemical purity of the standard (sum of peaks C and D presumably corresponding to the syn and anti 99m Tc-EC20) remained constant at ∼93% over the 6 h

The primary goal of this investigation was to identify a folate-based 99mTc chelator candidate for clinical development. This project required the design, synthesis, analytical, and radiochemical characterization of a new chemical entity, as well as in vitro and in vivo biological evaluations. We adopted an efficient solid-phase synthetic procedure to produce EC20, a small molecular weight peptide derivative of folate that contains a D-γ-Glu peptide linkage (Figure 1). In its natural form, folate (or pteroylglutamate) has a single glutamyl residue present in an L configuration. However, a D-Glu enantiomer residue was incorporated into the EC20 molecule for the purpose of providing additional metabolic protection against tissue-resident γ-glutamyl hydrolases (62). Importantly, similar to EC20, we have found that substitution of the L-Glu residue with a D-Glu residue does not alter folic acid’s ability to bind to the high affinity FR (unpublished data). EC20 was found to efficiently chelate 99mTc when in the presence of R-D-glucoheptonate and tin (II) chloride. When analyzed by radiochemical HPLC, >95% of the resulting 99mTc-EC20 formulation consisted of a mixture of syn and anti stereoisomers, each equally capable of binding to FR with high affinity (Figure 3). Approximately 3% of the 99mTc in the formulation was chelated to EC20 at some other site on the EC20 molecule besides the expected Dap-Asp-Cys moiety. Although this component was not isolated in sufficient quantity for optimal

1208 Bioconjugate Chem., Vol. 13, No. 6, 2002 Table 4. Excretion and Metabolism of

Leamon et al.

99mTc-EC20

from the Balb/c Mousea area percent

99mTc-EC20

Standard

urine samples (two mice/timepoint)

peak

rt (min)

0 hr

1 hr

6 hr

A (pertechnetate) B (unknown) C (isomer 1) D (isomer 2) sum C and D

1.4 3.4 5.5 18.5

2 4.5 15.5 78 93.5

2.1 4.5 15.7 77.7 93.4

1.8 4.8 15.9 77.5 93.4

1 hr 8.3 2.5 20.4 68.8 89.2

4 hr 6.3 2.6 18.1 73 91.1

9.4 5.4 7.3 77.9 85.2

10.2 0 11.1 78.7 89.8

a Mice were injected with 1 mCi (6.7 nmol) of 99mTc-EC20 via a lateral tail vein. At the indicated times, groups of two mice were euthanized, and urine was collected. The radiochemical speciation was then determined by HPLC. The area percent sum of peaks C and D (syn and anti isomers) is used to calculate the overall purity of intact 99mTc-EC20.

characterization, it was determined to bind to FR with high affinity (Figure 6). Finally, the remaining ∼ 2% of the radioactivity in the 99mTc-EC20 formulation was attributed to free 99mTc. This radiopharmaceutical demonstrated both time- and concentration-dependent association with FR-positive cells, as expected (22, 48). 99mTc-EC20 was rapidly cleared from the blood (t1/2 ∼ 4 min), which is a great quality for diagnostic imaging agents, and it preferentially accumulated within FR-positive tumors in large amounts. The performance of 99mTc-EC20 was directly compared to that of a similar FR targeting agent, 111In-DTPA-folate, using two different methods. First, both folate-based radiopharmaceuticals were found to equally compete with folic acid for binding to FRs on KB cells (Figure 3 and Table 1). Second, the biodistribution of each agent in tumor-bearing mice was nearly identical (Table 2). High tumor uptake and tumor-to-blood ratios were measured for 99mTc-EC20 and 111In-DTPA-folate. Taken together, these results suggest that, like 111In-DTPA-folate (63), 99m Tc-EC20 has the potential to effectively localize in FRpositive tumors when clinically administered to patients. Several folate-based 99mTc conjugates have previously been described in the literature. Limited biodistribution data are available on a 99mTc-12-amino-3,3,9,9-tetramethyl-5-oxa-4,8 diaza-2,10-dodecanedinoe dioxime (OXA) folate conjugate; however, moderate levels (∼7% ID/g) of tracer uptake in a KB tumor was reported (6). Studies involving the biodistribution of a 99mTc-ethylenedicysteine-folate conjugate in mammary tumor-bearing rats were also reported. Unfortunately, the rats in that study were fed a folate-rich diet. Thus, low tumor uptake and low tumor-to-blood ratios were obtained presumably because of competition with endogenous folate ligands (4). Last, a 99mTc-6-hydrazinonicotinamido-hydrazido (HYNIC) folate derivative (HYNIC-folate) was shown to accumulate in large amounts within 24JK-FBP tumors (5). Interestingly, 99mTc-EC20 accumulated within M109 tumors to nearly identical levels as that of HYNIC-folate in 24JK-FBP tumors (∼17% ID/g) (5; Table 2). These two agents also displayed a roughly 50:1 tumor-to-blood ratio at 4 h post intravenous injection. Thus, folate clearly has the ability to deliver many different types of 99mTc chelates to FR-positive tumors, and EC20 has proven to be among the best. The high-affinity FR is expressed at high levels mainly on cancer cells. For example, epithelial cancers of the ovary, mammary gland, colon, lung, nose, throat, and brain have all been reported to express elevated levels of the FR (64-72). In fact, >90% of all human ovarian tumors are known to express large amounts of this receptor (66, 70, 73). Overexpression of FR by cancer cells may well be selected for to provide the cells with a growth advantage relative to neighboring normal tissue. Indeed, highly de-differentiated metastatic cancers express con-

siderably more FR than their localized, low-grade counterparts (71). Each year ∼26 000 women in the United States are diagnosed with ovarian cancer, and less than 50% of those women survive more than five years. One reason for the low survival rate is the difficulty in diagnosing this form of cancer. Because of the fear of rupturing an unidentified abdominal mass and the potential for spreading cancer throughout the abdominal cavity, fine needle biopsy is not often performed. Rather, the diagnoses and staging of suspicious ovarian masses is typically done through surgical laparotomy, which is an invasive and expensive procedure. Since 99mTc-EC20 binds tightly to FR which is present in large amounts on ovarian and other cancers, this radiopharmaceutical may provide an inexpensive, noninvasive, but reliable method for the early diagnosis of cancer. Importantly, 99mTc-EC20 may also help guide the clinical decision process with these patients by making more definitive and earlier diagnoses of recurrent or residual disease. In summary, a new peptide derivative of folate was created to efficiently chelate 99mTc. This new compound, referred to as 99mTc-EC20, avidly binds to FR-positive tumor cells in vitro and in vivo despite its moderately high propensity to bind to serum protein. 99mTc-EC20 displayed rapid pharmacokinetic distribution out of the blood and into FR-positive tumors and kidney, and it was eliminated via the urine in a nonmetabolized form. These results indicate that 99mTc-EC20 should be clinically investigated as a noninvasive radiodiagnostic imaging agent for the detection of FR-positive cancers. ACKNOWLEDGMENT

This work was supported in part by a Phase I and II SBIR grant from the National Institutes of Health. The authors also wish to thank Dr. Mark Green and Carla Mathias for conducting the serum protein binding study and Dr. Philip S. Low for his valuable comments. LITERATURE CITED (1) Wang, S., et al. (1996) Synthesis, purification, and tumor cell uptake of 67Ga-deferoxamine-folate, a potential radiopharmaceutical for tumor imaging. Bioconjugate Chem. 7, 5662. (2) Mathias, C. J., et al. (1996) Tumor-selective radiopharmaceutical targeting via receptor-mediated endocytosis of Gallium-67-deferoxamine-folate. J. Nuc. Med., 37 (6), 1003-1008. (3) Mathias, C. J., et al. (1998) Indium-111-DTPA-folate as a potential folate-receptor-targeted radiopharmaceutical. J. Nucl. Med., 39, 1579-1585. (4) Ilgan, S., et al. (1998) 99mTc-Ethylenedicysteine-folate: A new tumor imaging agent. Synthesis, labeling and evaluation in animals. Cancer Biother. Radiopharm. 13 (6), 427-435. (5) Guo, W., G. H. Hinkle, and Lee, R. J. (1999) 99mTc-HYNICfolate: A novel receptor-based targeted radiopharmaceutical for tumor imaging. J. Nucl. Med. 40 (9), 1563-1569.

Synthesis and Biological Evaluation of EC20 (6) Linder, K. E., et al. (2000) In vitro & in vivo studies with Rand γ-isomers of 99mTc-OXA-Folate show uptake of both isomers in folate-receptor (+) KB cell lines. Soc. Nucl. Med. Proc. 47th Annual Meeting 41 (5), 119P. (7) Olsen, J. O., et al. (1995) Somatostatin receptor imaging of neuroendocrine tumors with indium-111 pentetreotide (Octreoscan). Semin. Nucl. Med. 25 (3), 251-61. (8) Virgolini, I., et al. (1994) Vasoactive intestinal peptidereceptor imaging for the localization of intestinal adenocarcinomas and endocrine tumors. N. Engl. J. Med. 331 (17), 1116-1121. (9) van Zanten-Przybysz, I., et al. (1999) 131I-labeled chimeric monoclonal antibody MOv18 in ovarian cancer patients: a pilot study. Tumor Targeting 4, 179-188. (10) Collins, D. A., Hogenkamp, H. P. , and Gebhard, M. W. (1999) Tumor imaging via indium 111-labeled DTPA-adenosylcobalamin. Mayo Clin. Proc. 74 (7), 687-91. (11) Modorati, G., et al. (1994) Immunoscintigraphy with three step monoclonal pretargeting technique in diagnosis of uveal melanoma: preliminary results. Br. J. Ophthalmol. 78 (1), 19-23. (12) Seccamani, E., et al. (1989) A simple qualitative determination of human antibodies to murine immunoglobulins (HAMA) in serum samples. Int. J. Radiat. Appl. Instrum. B, 16 (2), 167-170. (13) Colcher, D., et al. (1990) In vivo tumor targeting of a recombinant single-chain antigen-binding protein. J. Natl. Cancer Inst. 82 (14), 1191-1197. (14) Leamon, C. P., Low, P. S. (2001) Folate-mediated targeting: from diagnostics to drug and gene delivery. Drug Discov. Today, 6 (1), 44-51. (15) Konda, S. D., et al. (2000) Development of a tumortargeting MR contrast agent using the high-affinity folate receptor. Invest. Radiol. 35 (1), 50-57. (16) Ladino, C. A., et al. (1997) Folate-maytansinoids: Targetselective drugs of low molecular weight. Int. J. Cancer 73, 859-864. (17) Citro, G., et al. (1994) Inhibition of leukaemia cell proliferation by folic acid-polylysine-mediated introduction of c-myb antisense oligodeoxynucleotides into HL-60 cells. Br. J. Cancer 69, 463-467. (18) Li, S., Huang, L. (1997) Targeted delivery of antisense oligodeoxynucleotides by LPDII. J. Liposome Res. 7 (1), 6375. (19) Li, S., Huang, L. (1998) Targeted delivery of antisense oligodeoxynucleotides formulated in a novel lipidic vector. J. Liposome Res. 8 (2), 239-250. (20) Li, S., Deshmukh, H. M. , and Huang, L. (1998) Folatemediated targeting of antisense oligonucleotides to ovarian cancer cells. Pharm. Res. 15 (10), 1540-1545. (21) Leopold, L. H., et al. (1995) Multi-unit ribozyme-mediated cleavage of bcr-abl mRNA in myeloid leukemias. Blood 85 (8), 2162-2170. (22) Leamon, C. P., Low, P. S. (1991) Delivery of Macromolecules into Living Cells: A Method that Exploits Folate Receptor Endocytosis. Proc. Natl. Acad. Sci. U.S.A. 88, 55725576. (23) Ward, C. M., Acheson, N. , and Seymour, L. M. (2000) Folic acid targeting of protein conjugates into ascites tumor cells from ovarian cancer patients. J. Drug Targeting 8 (2), 119123. (24) Lu, J. Y., et al. (1999) Folate-targeted enzyme prodrug cancer therapy utilizing penicillin-V amidase and a doxorubicin prodrug. J. Drug Targeting 7 (1), 43-53. (25) Leamon, C. P.; Low, P. S. (1992) Cytotoxicity of MomordinFolate Conjugates in Cultured Human Cells. J. Biol. Chem. 267, 24966-24971. (26) Leamon, C. P., Pastan, I., and Low, P. S. (1993) Cytotoxicity of Folate-Pseudomonas Exotoxin Conjugates Toward Tumor Cells. J. Biol. Chem. 268, 24847-24854. (27) Leamon, C. P., Low, P. S. (1994) Selective Targeting of Malignant Cells with Cytotoxin-Folate Conjugates. J. Drug Targeting 2, 101-112. (28) Kranz, D. M., et al. (1995) Conjugates of folate and antiT-cell-receptor antibodies specifically target folate-receptor-

Bioconjugate Chem., Vol. 13, No. 6, 2002 1209 positive tumor cells for lysis. Proc. Natl. Acad. Sci. U.S.A. 92, 9057-9061. (29) Cho, B. K., et al. (1997) Single-chain Fv/folate conjugates mediate efficient lysis of folate-receptor-positive tumor cells. Bioconjugate Chem. 8, 338-346. (30) Kranz, D. M., et al. (1998) Targeting tumor cells with bispecific antibodies and T cells. J. Controlled Release 53, 7784. (31) Rund, L. A., et al. (1999) Bispecific agents target endogenous murine T cells against human tumor xenografts. Int. J. Cancer 83, 141-149. (32) Lee, R. J., Low, P. S. (1994) Delivery of Liposomes into Cultured KB Cells via Folate Receptor- mediated Endocytosis. J. Biol. Chem. 269, 3198-3204. (33) Lee, R. J., Low, P. S. (1995) Folate-mediated tumor cell targeting of liposome-entrapped doxorubicin in vitro. Biochim. Biophys. Acta 1233, 134-144. (34) Vogel, K., et al. (1996) Peptide-mediated release of folatetargeted liposome contents from endosomal compartments. J. Am. Chem. Soc. 118 (7), 1581-1586. (35) Rui, Y., et al. (1998) Diplasmenylcholine-folate liposomes: An efficient vehicle for intracellular drug delivery. J. Am. Chem. Soc. 120 (44), 11213-11218. (36) Gabizon, A., et al. (1999) Targeting folate receptor with folate linked to extremities of poly (ethylene glycol)-grafted liposomes: in vitro studies. Bioconjugate Chem. 10 (2), 28998. (37) Gottschalk, S., et al. (1994) Folate receptor mediated DNA delivery into tumor cells: potosomal disruption results in enhanced gene expression. Gene Ther. 1, 185-191. (38) Mislick, K. A., et al. (1995) Transfection of folate-polylysine DNA complexes: Evidence for lysosomal delivery. Bioconjugate Chem. 6, 512-515. (39) Douglas, J. T., et al. (1996) Targeted gene delivery by tropism-modified adenoviral vectors. Nat. Biotech. 14, 15741578. (40) Leamon, C. P., Weigl, D., and Hendren, R. W. (1999) Folate copolymer-mediated transfection of cultured cells. Bioconjugate Chem. 10, 947-957. (41) Guo, W., Lee, R. J. (1999) Receptor-targeted gene delivery via folate-conjugated polyethylenimine. PharmSci 1 (4), 19. (42) Reddy, J. A., et al. (1999) Optimization of folate-conjugated liposomal vectors for folate receptor-mediated gene therapy. J. Pharm. Sci. 88 (11), 1112-1118. (43) Reddy, J. A., Low, P. S. (2000) Enhanced folate receptor mediated gene therapy using a novel pH-senistive lipid formulation. J. Controlled Release 64, 27-37. (44) Antich, P., et al. (1994) Imaging of folate receptors with I-125 labeled folate using small animal imaging system built with plastic scintillating optical fibers. J. Nucl. Med. 35, 222 (abstract). (45) Mathias, C. J., et al. (2000) Synthesis of [ (99m)Tc]DTPAfolate and its evaluation as a folate-receptor-targeted radiopharmaceutical. Bioconjugate Chem. 11 (2), 253-7. (46) Rouschias, G. (1974) Chem. Rev. 74, 531. (47) Westerhof, G. R., et al. (1995) Carrier- and receptormediated transport of folate antagonists targeting folatedependent enzymes: Correlates of molecular structure and biological activity. Mol. Pharm. 48, 459-471. (48) Leamon, C. P., Low, P. S. (1993) Membrane Folate-binding Proteins are Responsible for Folate- protein Conjugate Endocytosis into Cultured Cells. Biochem. J. 291, 855-860. (49) Eisenhut, M., et al. (1991) J. Labeled Compd. Radiopharm. 29, 1283. (50) Liu, S., et al. (1996) Labeling cyclic glycoprotein IIb/IIIa receptor antagonists with 99mTc by the preformed chelate approach: effects of chelators on properties of [99mTc]chelator-peptide conjugates. Bioconjugate Chem. 7 (2), 196202. (51) Liu, S., et al. (1996) Labeling a hydrazino nicotinamidemodified cyclic IIb/IIIa receptor antagonist with 99mTc using aminocarboxylates as coligands. Bioconjugate Chem. 7 (1), 63-71. (52) Fritzberg, A. R., et al. (1988) Approaches to radiolabeling of antibodies for diagnosis and therapy of cancer. Pharm. Res. 5 (6), 325-334.

1210 Bioconjugate Chem., Vol. 13, No. 6, 2002 (53) Liang, F. H., Virzi, F., and Hnatowich, D. J. (1987) The use of diaminodithiol for labeling small molecules with technetium-99m. Int. J. Radiat. Appl. Instrum. B 14 (1), 63-67. (54) Chianelli, M., et al. (1992) Eur. J. Nucl. Med. 19, 625. (55) Baidoo, K. E., Lever, S. Z., and Scheffel, U., (1994) Bifunctional chelator for facile preparation of neutral technetium complexes. Bioconjugate Chem. 5 (2), 114-118. (56) Eisenhut, M., et al. (1996) Bifunctional NHS-BAT ester for antibody conjugation and stable technetium-99m labeling: conjugation chemistry, immunoreactivity and kit formulation. J. Nucl. Med. 37 (2), 362-70. (57) Wong, E., et al. (1997) Rhenium (V) and Technetium (V) Oxo Complexes of an N (2)N′S Peptidic Chelator: Evidence of Interconversion between the Syn and Anti Conformations. Inorg. Chem. 36 (25), 5799-5808. (58) Francesconi, L. C., et al. (1993) Synthesis and characterization of neutral MVO (M ) technetium, rhenium) aminethiol complexes containing a pendant phenylpiperidine group. Inorg. Chem. 32 (14), 3114-3124. (59) Marzilli, L. G., et al. (1994) Linking Deprotonation and Denticity of Chelate Ligands. Rhenium (V) Oxo Analogues of 99mTechnetium Radiopharmaceuticals Containing N2S2 Chelate Ligands. Inorg. Chem. 33 (22), 4850-4860. (60) Leamon, C. P., DePrince, R. B., and Hendren, R. W. (1999) Folate-mediated drug delivery: Effect of alternative conjugation chemistry. J. Drug Targeting 7 (3), 157-169. (61) Antony, A. C. (1992) The Biological Chemistry of Folate Receptors. Blood 79 (11), 2807-2820. (62) Jodrell, D. I., et al. (1993) The in vivo metabolic stability of dipeptide analogues of the quinazoline antifolate, ICI 198583, in mice. Biochem. Pharm. 46 (12), 2229-2234.

Leamon et al. (63) Mathias, C. J., Green, M. A. (1998) A kit formulation for preparation of [111In]In-DTPA-Folate, a folate-receptortargeted radiopharmaceutical. Nuc. Med. Biol. 25, 585-587. (64) Coney, L. R., et al. (1991) Cloning of a Tumor-associated Antigen: MOv18 and MOv19 Antibodies Recognize a Folatebinding Protein. Cancer Res. 51, 6125-6132. (65) Weitman, S. D., et al. (1992) Distribution of the folate receptor GP38 in Normal and Malignant Cell Lines and Tissues. Cancer Res. 52, 3396-3401. (66) Ross, J. F., Chaudhuri, P. K., and Ratnam, M. (1994) Differential regulation of folate receptor isoforms in normal and malignant tissues in vivo and in established cell lines. Cancer 73, 2432-2443. (67) Weitman, S. D., et al. (1992) Cellular Localization of the Folate Receptor: Potential Role in Drug Toxicity and Folate Homeostasis. Cancer Research 52, 6708-6711. (68) Weitman, S. D., Frazier, K. M., and Kamen, B. A. (1994) The folate receptor in central nervous system malignancies of childhood. J. Neurol. Oncol. 21, 107. (69) Mattes, M. J., et al. (1990) Patterns of antigen distribution in human carcinomas. Cancer Res. Suppl. 50, 880S. (70) Garin-Chesa, P., et al. (1993) Trophoblast and Ovarian Cancer Antigen LK26. Am. J. Pathol. 142 (2), 557-567. (71) Toffoli, G., et al. (1997) Overexpression of folate binding protein in ovarian cancers. Int. J. Cancer 74, 193-198. (72) Holm, J., et al. (1991) High-affinity folate binding in human choroid plexus. Biochem. J. 280, 267-271. (73) Campbell, I. G., et al. (1991) Cancer Res. 51, 5329-5338.

BC0200430