In vivo localization of [(111)In]-DTPA-D-Phe1-octreotide to human ovarian tumor xenografts induced to express the somatostatin receptor subtype 2 using an adenoviral vector.

Adenoviral vectors, encoding genes for cell surface antigens or receptors, have been used to induce their high level expression on tumor cells in vitro and in vivo. These induced antigens and receptors can then be targeted with radiolabeled antibodies or peptides for potential radiotherapeutic applications. The purpose of this study was to determine a dosing schema of an adenoviral vector encoding the human somatostatin receptor subtype 2 (AdCMVhSSTr2) for achieving the highest tumor localization of [(111)In]-DTPA-D-Phe1-octreotide, which binds to this receptor, in a human ovarian cancer model as a prelude to future therapy studies. AdCMVhSSTr2 was produced and used to induce hSSTr2 on A427 human nonsmall cell lung cancer cells and on SKOV3.ipl human ovarian cancer cells in vitro, as demonstrated by competitive binding assays using [125I]-Tyr1-somatostatin and [(111)In]-DTPA-D-Phe1-octreotide. Mice bearing i.p. SKOV3.ip1 tumors administered 1 x 10(9) plaque-forming units of AdCMVhSSTr2 i.p. 5 days after tumor cell inoculation, followed by an i.p. injection of [(111)In]-DTPA-D-Phe1-octreotide 2 days later, showed a range of 15.3-60.4% median injected dose/gram (ID/g) in tumor at 4 h after injection compared with 3.5% ID/g when [125I]-Tyr1-somatostatin was administered and 0.3% ID/g when the negative control peptide [125I]-mIP-bombesin was administered. Mice administered a control adenoviral vector encoding the gastrin-releasing peptide receptor did not have tumor localization of [(111)In]-DTPA-D-Phe1-octreotide (<1.6% ID/g), demonstrating specificity of [(111)In]-DTPA-D-Phe1-octreotide for the AdCMVhSSTr2 induced tumor cells. In another set of experiments, the tumor localization of [(111)In]-DTPA-D-Phe1-octreotide was not different 1, 2, or 4 days after AdCMVhSSTr2 injection (31.8, 37.7, and 40.7% ID/g, respectively; P = 0.88), indicating that multiple injections of radiolabeled peptide can be administered with equivalent uptake over a 4-day period. [(111)In]-DTPA-D-Phe1-octreotide tumor localization in animals administered AdCMVhSSTr2 on consecutive days or 2 days apart was 22.4% ID/g and 53.2% ID/g, respectively (P = 0.009) when [(111)In]-DTPA-D-Phe1-octreotide was given 1 day after the second AdCMVhSSTr2 injection. There was no difference in [(111)In]-DTPA-D-Phe1-octreotide localization after a single AdCMVhSSTr2 injection (40.7% ID/g) or two injections of AdCMVhSSTr2 given 1 (45.9% ID/g) or 2 (53.2% ID/g) days apart, where [(111)In]-DTPA-D-Phe1-octreotide was given in each case 4 days after the first AdCMVhSSTr2 injection (P = 0.65). Therefore, two AdCMVhSSTr2 injections did not increase [(111)In]-DTPA-D-Phe1-octreotide tumor localization compared with one injection, which eliminates concerns about an immune response to a second dose of AdCMVhSSTr2. This will be the basis for a therapeutic protocol with multiple administrations of an octreotide analogue labeled with a therapeutic radioisotope.

Radiosensitization mediated by a transfected anti-erbB-2 single-chain antibody in vitro and in vivo.

PURPOSE: The erbB-2 receptor is overexpressed in several human cancers, including ovarian, prostate, and breast. We have developed plasmid and adenoviral vectors expressing an anti-erbB-2 single chain antibody (sFv), directed to the endoplasmic reticulum (ER) of target cells, that is cytotoxic to tumor cells overexpressing erbB-2 through induction of apoptosis. The anti-erbB-2 sFv also sensitizes erbB-2 overexpressing cells to the cytotoxic effects of cisplatin. On this basis, we hypothesized that human ovarian cancer cells expressing anti-erbB-2 sFv with downregulated erbB-2 product, p185erbB-2, also would be sensitized to ionizing radiation. Therefore, we designed experiments to test the ability of the anti-erbB-2 sFv to radiosensitize human ovarian cancer cells in vitro and in vivo. METHODS AND MATERIALS: To test our hypothesis, we established subcutaneous (s.c.) tumors in the flanks of nude mice with SKOV3.ip1 human ovarian cancer cells and SKOV3 cells stably expressing the ER directed anti-erbB-2 sFv (SKOV3/pGT21). The tumors were treated with 10 Gy 60Co, or received no radiation. We then determined the regression rate, delay in regrowth, and time to tumor doubling of the tumors treated with radiation in the transfected group and controls. In addition, SKOV3.ip1 and SKOV3/pGT21 tumors were dissected from the irradiated animals and assayed for differences in p185erbB-2 expression at 12 weeks after irradiation by immunohistochemistry. Further, in vitro clonogenic survival assays were performed on the parental SKOV3.ip1 and SKOV3/pGT21 cell lines. RESULTS: A statistical analysis of the combined data was done for two in vivo experiments. The analysis of the combined data showed that animals with irradiated tumor SKOV3/pGT21 had a significantly higher regression rate (p = 0.0055), longer delay in regrowth (p = 0.0001) and time to tumor doubling (p = 0.0004), than those animals with tumor SKOV3.ip1 that received radiation. We observed a similar significant effect for the same parameters in the unirradiated tumor SKOV3/pGT21 compared to unirradiated tumor SKOV3.ip1. Immunohistochemical analysis of the SKOV3/pGT21 tumor cells demonstrated focal accumulation of p185erbB-2 in scattered clumps of cells and less p185erbB-2 membrane expression than cells of SKOV3.ip1 tumors. However, SKOV3.ip1 and SKOV3/pGT21 cells had similar in vitro sensitivity to radiation. CONCLUSIONS: These data support the hypothesis that tumors with reduced p185erbB-2 expression mediated by the anti-erbB-2 sFv are rendered more susceptible in vivo to the cytotoxic effects of ionizing radiation than tumors that maintain their normal expression of p185erbB-2. However, a similar effect was not observed with the same tumor cells in vitro. Thus, as has been described by others (1, 2), in vitro and in vivo results do not always correlate. Therefore, appropriate assays to assess clinical relevance need to be determined for each particular system studied.

Tumor localization of a radiolabeled bombesin analogue in mice bearing human ovarian tumors induced to express the gastrin-releasing peptide receptor by an adenoviral vector.

BACKGROUND: The adenoviral vector, AdCMVGRPr, has been used to induce the expression of the murine gastrin-releasing peptide receptor (GRPr) both in vitro and in vivo. A bombesin analogue ([125I]-mIP-bombesin) has been shown to bind with high affinity to GRPr and to localize to intraperitoneal (i.p.) ovarian tumors 2 days after induction of GRPr in an athymic nude mouse model. The present study was conducted to determine the level of localization of [(125/131)I]-mIP-bombesin in the tumors at 2, 4, and 7 days after AdCMVGRPr administration and to determine the feasibility of giving multiple doses of [131I]-mIP-bombesin for therapy. METHODS: Human ovarian cancer cells (SKOV3.ip1) were infected in vitro with AdCMVGRPr and were assayed for receptor expression at 2, 4, and 7 days after infection by using a radiolabeled bombesin-binding assay. Biodistribution studies utilized athymic nude mice inoculated i.p. with SKOV3.ip1 cells. The tumors were induced to express GRPr with an i.p. injection of AdCMVGRPr followed by administration of [125I]-mIP-bombesin 2 days later (AdCMVLacZ or saline was used for negative controls). In addition, the tumor localization of [125I]-mIP-bombesin was determined 4 and 7 days after AdCMVGRPr administration. The tumor localization of [131I]-mIP-bombesin was compared with [125I]-mIP-bombesin in this in vivo model. RESULTS: SKOV3.ip1 cells infected with AdCMVGRPr resulted in 80.3 +/- 5.9% binding of [125I]-Tyr4-bombesin at 2 days after infection, which decreased to 46.8 +/- 0.4% at 4 days and to 17.7 +/- 0.1% at 7 days. The biodistribution study showed that the tumor localization (14.9 +/- 8.2% injected dose/gram; ID/g) of [125I]-mIP-bombesin 2 days after administration of AdCMVGRPr was significantly greater than its localization in other organs (P < 0.003) and was significantly greater than in AcCMVLacZ- and saline-treated mice (P < 0.003). Injections of [125I]-mIP-bombesin at 4 and 7 days after a single AdCMVGRPr administration showed tumor localization of 4.5 +/- 3.0% ID/g at Day 4 and 3.9 +/- 3.5% ID/g at Day 7. The decreased localization at longer times after AdCMVGRPr infection correlated with in vitro results. The tumor uptake of [125I]-mIP-bombesin was comparable to the uptake of [131I]-mIP-bombesin (21.2 +/- 8.3% ID/g versus 15.4 +/- 5.6% ID/g, respectively), as was the normal tissue biodistribution. CONCLUSIONS: The expression of GRPr in human ovarian cancer cells can be accomplished both in vitro and in vivo by using AdCMVGRPr, with the in vivo tumor localization of [125I]-mIP-bombesin being significantly greater than in control animals. The tumor localization of [125I]-mIP-bombesin and [131I]-mIP-bombesin at 2 days after AdCMVGRPr was comparable in a mouse model of human ovarian carcinoma. Injections of [125I]-mIP-bombesin at Days 4 and 7 after AdCMVGRPr infection resulted in tumor localization of [125I]-mIP-bombesin but at a level lower than 2 days. Thus, the total amount of radioactivity delivered to the tumor should be increased by multiple injections of [131I]-mIP-bombesin, which would be required for a therapeutic effect.