Fax: (205) 975-7060
Targeted radiotherapy with [90Y]-SMT 487 in mice bearing human nonsmall cell lung tumor xenografts induced to express human somatostatin receptor subtype 2 with an adenoviral vector
Article first published online: 12 FEB 2002
Copyright © 2002 American Cancer Society
Supplement: Eighth Conference on Radioimmunodetection and Radioimmunotherapy of Cancer
Volume 94, Issue Supplement 4, pages 1298–1305, 15 February 2002
How to Cite
Rogers, B. E., Zinn, K. R., Lin, C.-Y., Chaudhuri, T. R. and Buchsbaum, D. J. (2002), Targeted radiotherapy with [90Y]-SMT 487 in mice bearing human nonsmall cell lung tumor xenografts induced to express human somatostatin receptor subtype 2 with an adenoviral vector. Cancer, 94: 1298–1305. doi: 10.1002/cncr.10300
- Issue published online: 12 FEB 2002
- Article first published online: 12 FEB 2002
- Manuscript Accepted: 14 NOV 2001
- Manuscript Received: 31 OCT 2001
- National Cancer Institute. Grant Numbers: CA80104, N01-CO-97110
- nonsmall cell lung cancer;
- somatostatin receptor
Novel approaches to increasing the therapeutic efficacy of targeted radiotherapy of cancer are required. One strategy to achieve this goal is to induce high-level expression of a receptor on the surface of tumor cells that can be targeted with a radiolabeled peptide. The objectives of this study were to 1) induce somatostatin receptor (SSTr2) expression in tumor cells using an adenovirus encoding the SSTr2 gene (AdSSTr2), 2) demonstrate tumor localization of [111In]-DTPA-D-Phe1-octreotide in AdSSTr2-injected tumors, and 3) show therapeutic efficacy with [90Y]-DOTA-D-Phe1-Tyr3-octreotide ([90Y]-SMT 487).
SSTr2 expression was validated in vitro by the binding and subsequent internalization of [111In]-DTPA-D-Phe1-octreotide (21.3% per mg of total protein) in A-427 cells infected with AdSSTr2. In vivo imaging confirmed 5- to 10-fold greater uptake 5.5 hours after intravenous administration of [111In]-DTPA-D-Phe1-octreotide in AdSSTr2-injected tumors relative to control tumors. For therapy studies, mice bearing established subcutaneous A-427 tumors were given two intratumoral injections of AdSSTr2 1 week apart, followed by an intravenous injection of 400 μCi or 500 μCi [90Y]-SMT 487 at 2 and 4 days after each adenoviral administration. Control animals either were not treated or were administered 500 μCi [90Y]-SMT 487 with no AdSSTr2 injection.
These studies showed that untreated animals and animals treated with no virus and 500 μCi [90Y]-SMT 487 had median tumor quadrupling times of 16 and 25 days, respectively. Mice administered AdSSTr2 and either 400 μCi or 500 μCi of [90Y]-SMT 487 demonstrated significantly longer median tumor quadrupling times of 40 and 44 days, respectively (P < 0.02).
These studies are the first to demonstrate in vivo therapeutic efficacy using a radiolabeled peptide targeted to a receptor expressed on the surface of tumor cells following gene transfer. Future studies will focus on the optimization of this approach. Cancer 2002;94:1298–1305. © 2002 American Cancer Society.
Targeted radiotherapy has been an effective form of treatment for lymphomas; however, it has been less successful with solid tumors.1, 2 In general, targeted radiotherapy has been conducted using radioisotopes attached to monoclonal antibodies (MAbs) for targeting the radiation to tumors expressing antigens or receptors recognized by the MAb. This strategy has been limited by bone marrow toxicity from the long serum half-life of the radiolabeled antibody, poor tumor penetration of the high-molecular-weight MAb, and low tumor antigen and receptor expression.3–5 Clearly, novel strategies for improving targeted radiotherapy are necessary. Strategies that have been evaluated include the use of radiolabeled, genetically engineered MAb fragments and radiolabeled peptides to decrease toxicity and increase tumor penetration; the use of radiosensitizers to increase the tumor response to radiation; and the use of cytokines to up-regulate tumor antigens and receptors.6 These combined modality treatments have had some success and should have an impact on the clinical treatment of solid tumors with targeted radiotherapy.
Our group has investigated another strategy, which combines a gene transfer approach with the use of radiolabeled peptides. This approach uses gene transfer techniques to increase tumor receptor expression. The induced receptor is then targeted with radiolabeled peptides, which should have good tumor penetration and reduced bone marrow toxicity in therapy studies. Previous studies demonstrated that adenoviral vectors encoding the genes for cell surface antigens or receptors could increase the in vivo tumor localization of radiolabeled antibodies and peptides specific to these receptors in the context of locoregional tumor models.7–10 The current study extends this approach, with the goal of determining whether a radiolabeled peptide can be used to achieve a therapeutic effect in vivo after infection of established tumors with an adenoviral vector encoding the gene for a cell surface receptor. In this regard, radiolabeled octreotide analogues are the most widely studied peptides for radiotherapy of SSTr2-positive tumors.11–13 In this report, it is shown that AdSSTr2 induced the expression of SSTr2 in human nonsmall cell lung cancer cells (A-427) in vitro as evidenced by binding and internalization of [111In]-DTPA-D-Phe1-octreotide. Similarly, following intravenous (i.v.) injection of [111In]-DTPA-D-Phe1octreotide in mice, AdSSTr2-injected A-427 tumors showed significantly increased accumulation relative to control tumors. Therapy studies with this animal model showed a statistically significant antigrowth effect on the A-427 xenografts for combined intratumoral AdSSTr2 and i.v. [90Y]-DOTA-D-Phe1-Tyr3-octreotide ([90Y]-SMT 487) treatments. To our knowledge, this is the first study to demonstrate therapeutic efficacy using a gene transfer strategy to induce receptor expression on tumors that are subsequently targeted with a radiolabeled peptide.
Materials and Methods
Cell Line and Adenovirus
The A-427 human nonsmall cell lung carcinoma cell line was obtained from American Type Culture Collection (Waldorf, MD) and maintained in Eagle minimum essential medium containing 10% fetal bovine serum, 1% nonessential amino acids, and 1% sodium pyruvate. The recombinant AdSSTr2 under control of the cytomegalovirus promoter was prepared using standard techniques described previously.9 The recombinant AdSSTr2 was plaque purified, validated by polymerase chain reaction (PCR), and titered within 293 cells using plaque assay techniques for direct determination of the viral plaque-forming units (pfu).
In Vitro Internalization of [111In]-DTPA-D-Phe1-Octreotide
[111In]-DTPA-D-Phe1-octreotide was a generous gift from Mallinckrodt, Inc. (St. Louis, MO) and was radiolabeled according to the manufacturer's instructions. A-427 cells at ≈80% confluency in 75-cm2 flasks were infected with AdSSTr2 at 10 pfu per cell for 2 hours at 37 °C with gentle rocking, followed by the addition of normal growth medium. After 24 hours, cells were harvested and mixed with uninfected A-427 cells at the following levels: 100% (no uninfected A-427 cells), 75%, 50%, 25%, 10%, 5%, 1%, 0%; 1.5 × 105 cells were reseeded in 24-well plates. Twenty-four hours later, the cells were washed with Hanks balanced salt solution (HBSS), and [111In]-DTPA-D-Phe1-octreotide (4 μCi, 35 nM) was added to each well in the presence or absence of an excess of unlabeled octreotide (Sandostatin, Novartis, Summit, NJ; 5 μg/well) as an inhibitor. The cells were incubated for 1 hour at 37 °C and washed twice with HBSS, followed by a third wash with HBSS containing 20 mM sodium acetate (pH 4.0, 10 minutes, 37 °C) to remove surface-bound radioactivity. A volume of 1 N NaOH equivalent to the total incubation volume (0.4 mL) was added to each well to lyse cells. The plates were imaged using an Anger 420/550 Mobile Radioisotope Gamma Camera (Technicare, Solon, OH) equipped with a low-energy, parallel hole collimator (Model 14S22022). Cell culture plates containing cell monolayers were imaged initially after the addition of [111In]-DTPA-D-Phe1-octreotide, and after 1 hour of incubation and washing. Planar imaging techniques were applied, with at least 50,000 total counts per image collected. Image acquisition time ranged from 5 to 40 minutes, with the longer time required for imaging the plates after washing. To confirm the gamma camera results, the cells were harvested into individual tubes and counted in a gamma counter to determine the amount of internally bound radioactivity. Data are presented as the percentage of internally bound radioactivity per mg of total protein.
In Vivo Imaging with [111In]-DTPA-D-Phe1-Octreotide
BALB/c nude mice (National Cancer Institute, Frederick Research Laboratory, Frederick, MD) were inoculated subcutaneously (s.c.) in each rear flank with 2 × 106 A-427 cells mixed 1:1 with Matrigel (Collaborative Biomedical Products, Bedford, MA). Eighteen days after tumor cell inoculation, one tumor was injected intratumorally with 1 × 109 pfu of AdSSTr2 and the other with 1 × 109 pfu of a control adenovirus (AdTRHr, encoding the gene for thyrotropin-releasing hormone receptor).14 [111In]-DTPA-D-Phe1-octreotide (130 μCi, 0.7 μg) was injected i.v. into the tail vein 48 and 96 hours after adenovirus injection. The mice were imaged at 3 minutes and 5.5 hours after injection of [111In]-DTPA-D-Phe1-octreotide with an Anger 420/550 Mobile Radioisotope Gamma Camera equipped with a pinhole collimator. Intrinsic resolution of the detector was 3.3 mm. During imaging, the mice were maintained with halothane anesthesia and positioned in ventral recumbency with the legs extended from the body. For single image sessions, at least 50,000 total counts per image were collected.
In a second group of animals, 2 and 4 μg of unlabeled octreotide was administered via a retro-ocular injection 72 and 96 hours after the AdSSTr2 injection. A second 1 × 109 pfu injection of AdSSTr2 or AdTRHr was given intratumorally 1 week after the first adenoviral injections. [111In]-DTPA-D-Phe1-octreotide was then administered i.v. into the tail vein 48 and 96 hours after the second adenoviral injections, and the mice were imaged as described above.
Region-of-interest (ROI) analyses were conducted on the AdSSTr2- and AdTRHr-injected tumors in each group of animals using a modified version of NIH Image (NucMed Image, Mark D. Wittry, St. Louis University, St. Louis, MO). A background ROI was drawn outside the animal image to correct the tumor ROI. Total counts and pixels were recorded for all regions. The average counts per pixel for the background region were subtracted from the average counts per pixel in the tumor ROI. The percentage of injected dose (% ID) in the tumor region was calculated as the ratio between the background corrected counts in the tumor region, divided by the initial background corrected counts in the whole animal. The % ID per g of tissue (%ID/g) was calculated based on the tumor weight at the time the animals were sacrificed after the second imaging session. Each group of animals was killed after the second imaging session (6 hours after injection of [111In]-DTPA-D-Phe1-octreotide) and the tissues were collected, weighed, and counted in a Minaxi Auto-Gamma 5000 series gamma counter (Packard, Downers Grove, IL). Biodistribution data are presented as %ID/g.
Therapy with [90Y]-SMT 487
SMT 487 binds with high affinity to SSTr213 and was a kind gift from Novartis (Basel, Switzerland). Yttrium-90 was obtained from Dupont NEN (Boston, MA). Radiolabeling of SMT 487 with [90Y] was performed according to published procedures13 to obtain the product in > 95% pure form. The animal model used for therapy studies was the same as that described above for the imaging studies. BALB/c nude mice were inoculated s.c. with 2 × 106 A-427 cells (1:1 mixture with Matrigel) in the rear flank. Twenty-four days later, the mice were administered 1 × 109 pfu AdSSTr2 intratumorally (Day 0), and the first measurement of the tumor (length × width) was made with vernier calipers. The range of median values between four treatment groups for the initial tumor areas was 47–68 mm2. Mice received a retro-ocular injection of either 400 μCi or 500 μCi [90Y]-SMT 487 on Days 2 and 4. The mice then received an additional intratumoral injection of AdSSTr2 on Day 7, followed by two more 400 μCi or 500 μCi doses of [90Y]-SMT 487 on Days 9 and 11. Control tumor-bearing mice either did not receive treatment or received four 500 μCi doses of [90Y]-SMT 487 on Days 2, 4, 9, and 11 without AdSSTr2 injections. Mice received food and water ad libitum, and tumors were measured and mice weighed every 3 days.
The difference in initial tumor size among treatment groups was evaluated using analysis of variance. The Kaplan-Meier method was utilized to estimate the distributions of time to tumor surface area quadrupling from the initial size on Day 0. The log rank test was employed to make global and pairwise comparisons of time to tumor quadrupling among the various treatment groups.
In Vitro Internalization of [111In]-DTPA-D-Phe1-Octreotide
The AdSSTr2 vector was used for these studies and validated in vitro. Gamma camera imaging and gamma counting of internalized [111In]-DTPA-D-Phe1-octreotide in A-427 cells infected with 10 pfu per cell of AdSSTr2 are shown in Figure 1. The image immediately after the addition of [111In]-DTPA-D-Phe1-octreotide is shown in Figure 1A and the amount of internalized radioactivity after a 1-hour incubation and wash is shown in Figure 1B. The internalized radioactivity could be visualized when the percentage of infected cells in the mixture of uninfected and infected A-427 cells was > 25%. Gamma counting results of the internalized radioactivity for 100%, 75%, 50%, 25%, 10%, 5%, 1%, and 0% infected A-427 cells were (mean ± standard deviation [SD]) 21.3 ± 0.2, 13.9 ± 0.2, 9.2 ± 0.1, 5.1 ± 0.3, 2.2 ± 0.2, 1.1± 0.1, 0.4 ± 0.03, and 0.3 ± 0.03% dose/mg total protein, respectively (Fig. 1C). Internalization was inhibited to < 0.2% dose/mg total protein by the addition of excess unlabeled octreotide, which demonstrates the specificity of [111In]-DTPA-D-Phe1-octreotide binding and internalization. These studies demonstrated the expression and internalization of SSTr2 on A-427 cells following AdSSTr2 infection and binding of [111In]-DTPA-D-Phe1-octreotide. Dilution of the AdSSTr2-infected cells with uninfected cells shows the sensitivity of the assay. Expression of SSTr2 could be detected by gamma camera imaging in as few as 25% of infected cells, while it could be detected by gamma counter analysis in as few as 5% of infected cells. The internalization of SSTr2 is important for the intracellular accumulation and retention of radiolabeled somatostatin analogues.
In Vivo Imaging with [111In]-DTPA-D-Phe1-Octreotide
Induction of SSTr2 expression in A-427 tumors in vivo was evaluated by imaging tumor accumulation of [111In]-DTPA-D-Phe1-octreotide. These studies utilized [111In]-DTPA-D-Phe1-octreotide to image SSTr2 expression noninvasively, since [90Y] is a pure beta-emitter and [90Y]-SMT 487 would not be suitable for this purpose. Gamma camera images of a representative mouse from the first group 5.5 hours after i.v. injection of [111In]-DTPA-D-Phe1-octreotide at 48 and 96 hours after a single intratumoral AdSSTr2 (right side) or AdTRHr (left side) injection are shown in Figure 2A and B, while images of a mouse from the second group 48 and 96 hours after a second intratumoral injection of AdSSTr2 or AdTRHr are shown in Figure 2C and D. These results show that imaging could detect accumulation of [111In]-DTPA-D-Phe1-octreotide in tumors injected with AdSSTr2 compared with much lower levels in tumors injected with AdTRHr. The retention of radioactivity in the kidneys and excretion through the bladder is observed. The localization of [111In]-DTPA-D-Phe1-octreotide in AdSSTr2-infected tumors was similar 48 and 96 hours after one or two AdSSTr2 injections. This tumor uptake was important to validate the dosing regimen that was used in the therapy studies.
The ROI analysis and tissue counts after the second imaging session for each group of animals are shown in Figure 3. The ROI analysis in Figure 3A shows tumor uptake of [111In]-DTPA-D-Phe1-octreotide of 2.8 ± 0.4% ID/g 48 hours after a single intratumoral AdSSTr2 injection and 3.1 ± 0.6% ID/g at 96 hours. By comparison, the tumor uptake was to 1.5 ± 0.3% ID/g 48 hours after the second intratumoral AdSSTr2 injection and 1.5 ± 0.4% ID/g at 96 hours. Uptake of [111In]-DTPA-D-Phe1-octreotide in AdTRHr-injected tumors was < 0.3% ID/g at both time points. Figure 3B shows the biodistribution of [111In]-DTPA-D-Phe1-octreotide. The uptake in tumors after a single or double injection of AdSSTr2 (1.3 ± 0.7% and 1.7 ± 1.0% ID/g, respectively) was greater than in normal tissues (< 0.5% ID/g, except kidney) and in the AdTRHr-injected tumors (0.08% ID/g) for both groups of animals. This uptake of [111In]-DTPA-D-Phe1-octreotide in AdSSTr2-injected tumors was similar to that previously reported in rats bearing pancreatic tumors that natively expressed SSTr2.15, 16 The kidneys had ≈18% ID/g in both groups of animals, which agrees with the results of other studies.9 These studies showed a high level of SSTr2 expression in tumors up to 4 days following one or two direct intratumoral AdSSTr2 injections.
Therapy with [90Y]-SMT 487
Therapy experiments were conducted using [90Y]-SMT 487 based on the high level of SSTr2 tumor expression as determined by [111In]-DTPA-D-Phe1-octreotide imaging. Previous studies have shown favorable biodistribution of [90Y]-SMT 487 in rats bearing pancreatic tumors that natively express SSTr213 and relatively low estimates of radiation doses to normal organs in humans by measuring the kinetics of [86Y]-SMT 487 in baboons using positron-emission tomography.17 No significant animal weight loss was observed throughout the experiment. The change in median A-427 tumor size relative to the size on the day of the first AdSSTr2 injection is shown in Figure 4A. Mice that received two intratumoral injections of AdSSTr2 and four doses of 400 μCi or 500 μCi [90Y]-SMT 487 had median tumor quadrupling times of 40 and 44 days, respectively (Fig. 4B). The log rank test revealed a statistically significant difference in time to tumor quadrupling between the AdSSTr2 + [90Y]-SMT 487 treatment groups and the control groups (P < 0.02). The median tumor quadrupling times of the untreated group and the group that received no virus plus four doses of 500 μCi [90Y]-SMT 487 were 16 and 25 days, respectively. The difference in time to tumor quadrupling between the untreated and no virus plus [90Y]-SMT 487 control groups was significant (P = 0.0008), which is likely due to nonspecific irradiation of the tumor, since the A-427 tumors are negative for SSTr2.
To our knowledge, this is the first study to demonstrate therapeutic efficacy in an animal model using gene transfer to induce high levels of a receptor on tumors that are then targeted with a radiolabeled peptide capable of delivering a high radiation dose to the tumor cells. This represents a new paradigm in the treatment of cancer. The model system was first evaluated in vitro and in vivo to confirm expression of SSTr2 and localization of [111In]-DTPA-D-Phe1-octreotide. Previous studies showed that [111In]-DTPA-D-Phe1-octreotide bound with high affinity (1.6 nM) to membrane preparations of A-427 cells infected with AdSSTr2 at 10 pfu/cell.9 The current in vitro study was designed to demonstrate that the SSTr2 produced by infection of A-427 cells with AdSSTr2 would internalize following radiolabeled octreotide binding. Internalization is important with regard to accumulation and retention of the radiolabeled peptide in order to maximize the therapeutic effect. Figure 1 shows that [111In]-DTPA-D-Phe1-octreotide was internalized by AdSSTr2 infected cells; this was detected by gamma camera imaging and gamma counter analysis. Mixing of AdSSTr2-infected A-427 cells with uninfected A-427 cells showed that internalization of [111In]-DTPA-D-Phe1-octreotide could be detected with imaging down to 25% infected A-427 cells (Fig. 1B). This corresponds to detection of ≈38,000 SSTr2 positive cells in a well containing ≈150,000 total cells, and gives an indication of the sensitivity for imaging in the subsequent in vivo studies.
Expression of SSTr2 in A-427 tumors after intratumoral injection of AdSSTr2 was demonstrated by localization of [111In]-DTPA-D-Phe1-octreotide (Fig. 2). Gamma camera imaging was used to demonstrate that expression of SSTr2 could be evaluated noninvasively. This could be important in future clinical trials; biopsies might not be necessary to determine the level of gene transfer. Figure 3 shows that the radiolabeled octreotide uptake in AdSSTr2-injected tumors measured by tissue counting was similar to the ROI analysis for two injections of AdSSTr2, but not in the case of a single injection of AdSSTr2. This is likely due to an overestimation of the ROI after a single AdSSTr2 injection. Since the tumors that received the single injection of AdSSTr2 were smaller, it is hypothesized that some normal tissue surrounding the tumor was induced to express SSTr2 and included in the tumor ROI analysis. This could account for the discrepancy between the ROI analysis and the tissue counting and will be investigated in future studies.
The therapy study (Fig. 4) showed significantly different tumor quadrupling times between the treatment groups and the control groups; however, the tumor doubling times (7–9 days) were not significantly different among groups. The median tumor areas at the time of treatment were 47–68 mm.2 The differences in median tumor surface area between groups began to appear at the beginning of the second week of treatment. A second experiment showed differences in the median time to tumor doubling when the initial median tumor areas of the groups were 58–132 mm2 (data not shown). Possible reasons for the differences between these experiments could be that a smaller fraction of the radiation dose was deposited in the smaller tumors due to the long path length of [90Y] emissions, or that the smaller tumors were still in an exponential phase of growth that was not inhibited by the first week of treatment. Anderson et al. showed therapeutic efficacy with a [64Cu]-labeled octreotide derivative in rats bearing s.c. pancreatic tumors that natively expressed SSTr2.12 This study suggested that [64Cu] may be more appropriate for therapy of smaller tumors than [90Y] due to the differences in the path lengths of their particulate emissions.12
Stolz et al. showed that [90Y]-SMT 487 could be used to achieve complete remissions in rats bearing SSTr2-positive CA 20948 rat pancreatic tumors.13 In this study, a single i.v. dose of approximately 3 mCi yielded complete remissions in five of seven animals, and a dose of 1.5 mCi yielded a complete remission in one of five animals. No complete remissions were observed in rats that were administered an equivalent unlabeled dose of SMT 487. It is difficult to compare these results with the ones observed in this study because different tumor models were employed. However, it is clear that our model could be improved. It should be emphasized that this is an initial study to demonstrate the therapeutic utility of radiolabeled peptides targeted to receptors that were expressed on tumors utilizing a gene therapy vector. Future studies will investigate the use of higher doses of [90Y]-SMT 487, additional doses of AdSSTr2 and [90Y]-SMT 487, and improved AdSSTr2 vectors. These are all feasible, since no overt toxicity was observed in this study. Further studies will investigate blood chemistries and histologies in detail to determine the dose-limiting toxicities of this novel approach of targeted radiotherapy.
This initial proof-of-principle study was limited to intratumoral injections of the adenoviral vector followed by systemic administration of the radiolabeled peptide. Current limitations of adenoviral vectors as gene therapy vectors include infection of normal tissues (particularly the liver) in vivo; the inability of the replication-incompetent adenovirus to infect all of the tumor cells; and immunogenicity of the adenovirus, precluding multiple administrations.18 Thus, the current strategy would be limited to the treatment of locoregional disease; however, it is anticipated that improvements in adenoviral vectors will eventually lead to their use in the treatment of metastatic disease. In particular, adenoviral vectors that are conditionally replicative, contain tumor-specific promoters, or are genetically modified to infect only tumor cells may be useful for the treatment of disseminated disease.19–21
In conclusion, this is the first report (to our knowledge) to demonstrate the therapeutic efficacy of a radiolabeled peptide targeted to a receptor that was expressed on tumor cells in vivo with a gene therapy vector. This represents a novel paradigm for targeted radiotherapy of cancer. In particular, targeting of [90Y]-SMT 487 to A-427 tumors induced to express SSTr2 via intratumoral injections of AdSSTr2 produced significant inhibition of tumor growth. This approach has potential clinical applications, since both adenoviral vectors and [90Y]-SMT 487 are being used for cancer therapy in clinical trials. Further optimization of this system is necessary to determine its ultimate utility.
Sally Lagan is gratefully acknowledged by the authors for helping prepare the manuscript. The authors thank Debbie Della Manna, Richard Kirkman, Sheila Bright, Christine Olsen, Wu Qi, Zhu Min, Barbara Krum, and Sakib Hassan for their technical assistance.
- 6Experimental radioimmunotherapy and methods to increase therapeutic efficacy. In: GoldenbergDM, editor. Cancer therapy with radiolabeled antibodies. Boca Raton: CRC, 1995: 115–40..
- 17Uptake kinetics of the somatostatin receptor ligand [86Y]DOTA-DPhe1- Tyr3-octreotide ([86Y]SMT487) using positron emission tomography in non-human primates and calculation of radiation doses of the 90Y-labelled analogue. Eur J Nucl Med 1999; 26: 358–66., , , , , , et al.
- 20Recent advances in replicative viral vectors for cancer gene therapy. Curr Res Mol Ther 1998; 1: 339–45.