Early Detection and Diagnosis
αvβ3 Integrin-targeting of intraperitoneally growing tumors with a radiolabeled RGD peptide
Article first published online: 9 NOV 2006
Copyright © 2006 Wiley-Liss, Inc.
International Journal of Cancer
Volume 120, Issue 3, pages 605–610, 1 February 2007
How to Cite
Dijkgraaf, I., Kruijtzer, J. A.W., Frielink, C., Corstens, F. H.M., Oyen, W. J.G., Liskamp, R. M.J. and Boerman, O. C. (2007), αvβ3 Integrin-targeting of intraperitoneally growing tumors with a radiolabeled RGD peptide. Int. J. Cancer, 120: 605–610. doi: 10.1002/ijc.22297
- Issue published online: 29 NOV 2006
- Article first published online: 9 NOV 2006
- Manuscript Accepted: 3 AUG 2006
- Manuscript Received: 22 MAY 2006
- ovarian cancer;
- RGD peptide;
Ovarian cancer is the fourth most common cause of cancer deaths among females in the western world after cancer of the breast, colon and lung. The inability to control the disease within the peritoneal cavity is the major cause of treatment failure in patients with ovarian cancer. The majority of ovarian carcinomas express the αvβ3 integrin. Here we studied the tumor targeting potential of an 111In-labeled cyclic RGD peptide in athymic BALB/c mice with intraperitoneally (i.p.) growing NIH:OVCAR-3 human ovarian carcinoma tumors. The cyclic RGD peptide, c(RGDfK)E, was synthesized, conjugated with DOTA and radiolabeled with 111In. The targeting potential of 111In-DOTA-E-c(RGDfK) was studied in athymic mice with i.p. growing NIH:OVCAR-3 xenografts and the optimal dose of this compound was determined (0.01 μg up to 10 μg). The biodistribution at optimal peptide dose was determined at various time points (0.5 up to 72 hr). Furthermore, the therapeutic potential of 177Lu-DOTA-E-c(RGDfK) was studied in this model. Two hours after i.p. administration, 111In-DOTA-E-c(RGDfK) showed high and specific uptake in the i.p. growing tumors. Optimal uptake in the i.p. growing tumors was observed at a 0.03–0.1 μg dose range. Tumor uptake of 111In-DOTA-E-c(RGDfK) peaked 4 hr p.i. [(38.8 ± 2.7)% ID/g], gradually decreasing at later time points [(24.0 ± 4.1)% ID/g at 48 hr p.i.]. Intraperitoneal growth of OVCAR-3 could be significantly delayed by injecting 37 MBq 177Lu-labeled peptide i.p. Radiolabeled DOTA-E-c(RGDfK) is suitable for targeting of i.p. growing tumors and potentially can be used for peptide receptor radionuclide therapy of these tumors. © 2006 Wiley-Liss, Inc.
Ovarian cancer remains the most lethal gynecological malignancy in the western world.1 About 90% of ovarian cancer originates from malignant transformation of the ovarian surface epithelium and in most patients the disease is diagnosed in an advanced stage.2 Most patients with Stage III and IV ovarian cancer achieve a clinical complete remission after cytoreductive surgery and combination chemotherapy. Unfortunately, in the majority of these patients, the disease relapses.3 Over the past 25 years the principal treatment of advanced ovarian cancer has been surgery, followed by chemotherapy. At the moment, paclitaxel combined with cisplatin or carboplatin is the standard firstline treatment in most countries,4 resulting in response rates of 70–80%. In comparison with former regimens, response rates have ameliorated and tolerability has improved. However, only small improvements in overall survival have been achieved. Further progress will depend on new treatments that eradicate residual disease after surgery and chemotherapy. The use of radiolabeled antibodies to achieve this goal has been investigated in past decades. For example, Epenetos et al. found that patients who achieve complete remission with conventional therapy had a relatively long-mean survival when treated with intraperitoneal (i.p.) radioimmunotherapy using 90Y-labeled anti MUC1 antibody.5 However, the promising results of this Phase I/II study were not confirmed in a recently completed Phase III trial.6
During the past decade, radiolabeled receptor-binding peptides have emerged as an important class of radiopharmaceuticals for diagnosis and therapy. Peptides used for tumor-targeting offer considerable advantages over antibodies, as they are not immunogenic, accumulate rapidly in the target tissue and clear rapidly from the blood and nontarget tissues.
For peptide receptor-targeted radiotherapy, peptides with specific affinity for tumor-associated receptors on cancer cells, labeled with cytotoxic radionuclides can be used. Because of the restricted expression of the αvβ3 integrin in tumors, αvβ3 is considered a suitable receptor for tumor-targeting. RGD peptides contain the amino acid sequence Arg-Gly-Asp and preferentially bind to the αvβ3 integrin receptor. In previous studies, we have shown that the radiolabeled cyclic RGD peptides specifically accumulated in subcutaneously growing αvβ3-expressing tumors in athymic mice.7 Here we studied the tumor-targeting potential of an 111In-labeled cyclic RGD peptide in athymic BALB/c nude mice with i.p. growing NIH:OVCAR-3 ovarian carcinoma tumors. Furthermore, in this mouse model, the therapeutic potential of the 177Lu-labeled cyclic RGD peptide was investigated.
Material and methods
Radiolabeling of DOTA-E-c(RGDfK)
111In-DOTA-E-c(RGDfK) was prepared by adding 18.5 MBq 111InCl3 (Mallinckrodt, Petten, The Netherlands) to 5 μg (4.4 nmol) DOTA-E-c(RGDfK) (Fig. 1) dissolved in 300 μL 0.5 M ammonium acetate buffer, pH 6, containing 0.6 mg/mL gentisic acid. 177Lu-DOTA-E-c(RGDfK) was prepared by adding 1.3 GBq (35.5 mCi) 177LuCl3 (NRG, Petten, The Netherlands) to 9 μg (8.0 nmol) DOTA-E-c(RGDfK) dissolved in 300 μL 0.5 M ammonium acetate buffer, pH 5, containing 0.6 mg/mL gentisic acid.
The reaction mixtures were degassed and the mixtures were heated at 100°C for 15 min. The radiochemical purity was determined by reversed-phase high-performance liquid chromatography (RP-HPLC) (HP 1100 series, Hewlett Packard, Palo Alto, CA) using a C18 column (RX-C18, 4.6 × 250 mm2, Zorbax) eluted with a gradient mobile phase (8–20% B over 25 min, solvent A = 25 mM ammonium acetate buffer, solvent B = acetonitrile) at 1 mL/min. The radioactivity of the eluate was monitored using an inline radiodetector (Flo-One Beta series, Radiomatic, Meriden, CT).
Nude mouse tumor model
To ensure a reproducible growth of tumors in the mice, NIH:OVCAR-3 ovarian carcinoma cells were serially transplanted i.p. in female BALB/c nude mice. On day 0 OVCAR-3 cells were harvested from a mouse with ascitic tumor, and 6- to 8-week-old mice were inoculated i.p. with 0.2 mL of a cell suspension of NIH:OVCAR-3 cells (5 × 107 cells/mL). The radiolabeled DOTA-E-c(RGDfK) was administered 10–11 days after tumor inoculation.
Ten or eleven days after tumor inoculation, mice were randomly divided in 7 groups and received 0.01, 0.03, 0.1, 0.3, 1, 3 (n = 4–5) or 10 μg (n = 2) DOTA-E-c(RGDfK) i.p. Therefore, 111In-labeled DOTA-E-c(RGDfK) was prepared as described above (140 kBq/nmol) and peptide dose was adjusted by adding nonradiolabeled DOTA-E-c(RGDfK). Mice were killed by CO2 asphyxiation 2 hr postinjection (p.i.) and the OVCAR-3 cells were harvested by rinsing the abdominal cavity twice with 5 mL 0.9% NaCl. Cells were spun down (1,200 rpm, 5 min), and the activity in pellet and the supernatant was determined. In addition, blood and the major organs and tissues were collected, weighed and counted in a γ-counter (1480 Wizard, Wallac, Turku, Finland). The percentage of injected dose per gram (% ID/g) was determined for each sample. At 7 time points p.i. (0.5, 2, 4, 8, 24, 48 and 72 hr), the biodistribution of the optimal dose of 111In-DOTA-E-c(RGDfK) was determined in 5 mice/group. In addition, to investigate whether the localization of 111In-DOTA-E-c(RGDfK) is receptor-mediated the biodistribution of the optimal dose of 111In-DOTA-E-c(RGDfK) in the presence of an excess of unlabeled RGD peptide was determined at each time point (2 mice/ group). Furthermore, at 2 hr p.i. the effect of the route of administration was studied in 1 group of mice (n = 5). This group received 0.1 μg 111In-DOTA-E-c(RGDfK) intraveneously (i.v.).
Radionuclide therapy study
One group of mice with i.p. OVCAR-3 tumors (n = 10) received 37 MBq 177Lu-DOTA-E-c(RGDfK). A second group of mice with i.p. OVCAR-3 tumors (n = 12) did not receive any treatment and served as control group. Twice a week the body weight of the mice in each group was recorded. When i.p. ascitic tumor growth was apparent, mice were killed by CO2 asphyxiation. Differences in survival between groups were compared using a logrank test for survival analysis. The level of significance was set at p < 0.05.
To study the intratumoral distribution, 3 mice with i.p. growing OVCAR-3 tumors received 1.1 MBq (0.3 μg) 111In-DOTA-E-c(RGDfK) i.p. Mice were killed by CO2 asphyxiation 2 hr p.i. One solid tumor deposit per mouse was removed and snap-frozen in liquid isopentane and cryofixed (OCT Tissue-Tek; SAKURA Finetek, CA) The tumor was sectioned (5 μm) using a cryostatic microtome. After air drying, tumor sections were exposed to a storage phosphor imager screen overnight. The screen was scanned in a phosphor imaging system (Molecular Imager GS363, BioRad Laboratories, Hercules, CA) at a pixel size of 100 × 100 μm. Images were processed with Quantity One software (version 4.5.2, BioRad Laboratories, Hercules, CA).
Radiolabeling of DOTA-E-c(RGDfK)
RP-HPLC analysis indicated that the radiochemical purity of 111In-DOTA-E-c(RGDfK) and 177Lu-DOTA-E-c(RGDfK) preparations used in these experiments was at least 93%. The elution profile of 111In-DOTA-E-c(RGDfK) and 177Lu-DOTA-E-c(RGDfK) showed a single peak for both compounds with an elution time of 14.0 min for the 111In-labeled compound (Fig. 2) and 14.2 min for the 177Lu-labeled compound.
The results of the peptide dose escalation study of 111In-DOTA-E-c(RGDfK) are summarized in Figure 3. The optimal tumor uptake was observed when a peptide dose of 0.03 μg and 0.1 μg was administered [(20.6 ± 9.7)% ID/g and (18.5 ± 5.9)% ID/g, respectively). At 2 hr p.i., the tumor-to-blood ratio at a peptide dose of 0.1 μg was 133 ± 46 (Fig. 4). At higher peptide doses the uptake in the tumor was significantly lower, probably due to saturation of the αvβ3 integrin receptors in the tumor.
The optimal dose of 0.1 μg per mouse was used to study the biodistribution of the radiolabeled peptide at 0.5, 2, 4, 8, 24, 48 and 72 hr p.i. The results are shown in Figure 5. Tumor uptake peaked at 4 hr p.i. [(38.8 ± 2.7)% ID/g] and gradually decreased with time to (19.3 ± 1.9) %ID/g at 72 hr p.i. Blood levels were (0.98 ± 0.20)% ID/g 0.5 hr p.i. and rapidly decreased to (0.006 ± 0.001)% ID/g at 72 hr p.i. Coinjection of an excess unlabeled RGD peptide (50 μg) along with 0.1 μg 111In-DOTA-E-c(RGDfK) resulted in a significantly lower radioactivity concentration in the tumor, indicating that uptake of the major fraction of 111In-DOTA-E-c(RGDfK) in the tumor is αvβ3-mediated (Fig. 6).
The route of administration of 111In-DOTA-E-c(RGDfK) in the i.p. OVCAR-3 model was clearly in favor of the i.p. route (Fig. 7). At 2 hr p.i., tumor uptake after i.p. administration of 0.1 μg 111In-DOTA-Ec(RGDfK) was (35.2 ± 3.8)% ID/g whereas after i.v. administration, the tumor uptake was (0.98 ± 0.26)% ID/g.
The results of the survival study are given in Figure 8. Mice that received 37 MBq 177Lu-DOTA-E-c(RGDfK) i.p. showed a statistically significant longer survival than the mice that received no treatment (p = 0.017). The median survival for the untreated mice was 5 weeks, whereas the median survival for the treated mice was 21 weeks.
In Figure 9, tumor sections from 3 tumor deposits are shown. In all tumor sections, 111In-DOTA-E-c(RGDfK) showed a heterogeneous distribution throughout the tumor 2 hr p.i., mainly accumulating in the periphery of the tumor deposit.
111In-labeled cyclic RGD peptide showed tumor targeting potential in i.p. growing ovarian carcinoma. Uptake of the peptide in the i.p. growing ascitic tumor was specific, dose dependent and saturable. Optimal tumor uptake of 111In-DOTA-E-c(RGDfK) was achieved at a peptide dose in the range of 0.03–0.1 μg. At higher peptide dose, the uptake in the tumor was reduced due to saturation of the αvβ3 integrin receptors. At the lowest peptide dose, 0.01 μg, the blood levels were relatively high, which may be caused by a relevant fraction of the peptide associated with plasma proteins.
For peptide receptor-targeted radiotherapy (PRRT), high uptake in the tumor is important, but retention of the radiolabel in the tumor is also crucial. With a peak in tumor uptake at 4 hr p.i., followed by a slow decrease to (24.0 ± 4.1)% ID/g after 48 hr, this cyclic RGD peptide fulfilled this requirement. The long retention of 111In-DOTA-E-c(RGDfK) in the tumor deposits may be due to internalization of the radiolabeled ligand. The internalization of radiolabeled RGD peptides and RGD-containing macromolecular conjugates by αvβ3-expressing cells has been described.8, 9
Coinjection of an excess of unlabeled RGD peptide resulted in a significant decrease of radioactivity in the tumor, clearly demonstrating that the uptake of 111In-DOTA-E-c(RGDfK) in the tumor was αvβ3-mediated. It is noteworthy that the uptake of 111In-DOTA-E-c(RGDfK) in the i.p. growing tumor after i.p. administration was much higher compared to the uptake in subcutaneously (s.c.) growing OVCAR-3 tumors after i.v. administration. The uptake of 111In-DOTA-E-c(RGDfK) at 2 hr p.i. in the i.p. growing tumor was (35.2 ± 3.8)% ID/g, which is more than 15-fold higher than the uptake in s.c. tumors [(2.04 ± 0.3)% ID/g].
Radiolabeled antibodies directed against the mucin-1 antigen and other tumor-associated glycoproteins, such as TAG-72 and gp-38 have been used to target ovarian cancer.10 Several clinical trials on radioimmunotherapy in patients with ovarian cancer have been published.11, 12, 13, 14, 15, 16, 17, 18 However, the use of monoclonal antibodies (MAbs) for tumor targeting show some disadvantages like the high molecular weight which hinders rapid pharmacokinetics resulting in slow diffusion into the target tissue and comparatively high blood concentration. Another complicating factor for the application of radiolabeled antibodies is that MAbs are immunogenic. Although antibody fragments have been developed to overcome these problems, peptides have more favorable properties for tumor targeting. Peptides have a small size and a rapid clearance from blood and nontarget tissues compared to proteins and antibodies. In addition, peptides have a low immunogenicity.
The effect of the route of administration has been the subject of various studies. These studies showed that for radiolabeled antibodies, the route of administration, i.v. vs. i.p. only has limited effect on the targeting of i.p. growing tumors.19, 20, 21, 22, 23 Colcher et al. investigated the efficacy of intracavitary radiolabeled MAb administration and demonstrated the advantage of the concomitant use of intracavitary and i.v. administered MAbs for tumor targeting.19 Furthermore, Koppe et al. showed that in nude mice with i.p. LS174T tumors the biodistribution of the monoclonal antibody MN-14 labeled with 131I, 186Re and 88Y after i.v. and i.p. administration was not significantly different.20 In a study in patients suspected of having primary or recurrent ovarian carcinoma were simultaneously injected i.v. and i.p. with 125I/131I-labeled chimeric antibody MOv18.21 In this study, antibody uptake in i.p. tumor deposits was independent of the route of administration. The present study demonstrates that for radiolabeled peptides, the route of administration has a marked influence on tumor targeting. Tumor uptake after i.v. administration of 0.1 μg 111In-DOTA-E-c(RGDfK) at 2 hr p.i. was (0.98 ± 0.26)% ID/g, whereas after i.p. administration tumor uptake was (35.2 ± 3.8)% ID/g (Fig. 7). Because of the fast blood clearance of peptides compared to antibodies, the i.p. route of administration is preferred for peptides.
Compared to other studies investigating the tumor targeting potential of peptides, the i.p. tumor model showed excellent tumor uptake of 111In-DOTA-E-c(RGDfK) and low uptake in nontarget tissues. Especially, the kidney concentration of 111In-DOTA-E-c(RGDfK) is low compared with that of other 111In-labeled peptides. Apparently, this peptide is not efficiently reabsorbed in the renal tubular cells.
In this mouse model of ovarian cancer the therapeutic potential of the cyclic RGD peptide was also determined. In this i.p. OVCAR-3 model, the tumor burden consisted of cell clusters in ascitic fluid and small omental solid tumor depositions. High energy particles such as 90Y (βmax 2.3 MeV, t1/2 64 hr) and 188Re (βmax 2.1 MeV, t1/2 17 hr) are considered more appropriate for the treatment of larger tumors, whereas low energy particles such as 177Lu (βmax 0.5 MeV, t1/2 161 hr) and 131I (βmax 0.6 MeV, t1/2 192 hr) could be more effective for the treatment of tumor cell clusters and small tumor lesions.24 Therefore, DOTA-E-c(RGDfK) was radiolabeled with the β-emitting radionuclide 177Lu. In mice with i.p. growing OVCAR-3 tumors i.p. treatment with 177Lu-DOTA-E-c(RGDfK) resulted in a statistically significant better survival as compared to untreated controls (p = 0.017).
In conclusion, we demonstrated that 111In-DOTA-E-c(RGDfK) has high and specific uptake in nude mice with i.p. growing OVCAR-3 tumors. PRRT experiments in this model of ovarian cancer indicated that i.p. tumor growth can be inhibited significantly by a therapeutic dose of 177Lu-DOTA-E-c(RGDfK).
The authors thank Mr. Gerry Grutters and Mr. Hennie Eikholt for technical assistance during animal experiments.