Introduction

This study was carried out in preparation for a Phase I clinical trial of adenovirus-mediated gene therapy for locally recurrent prostate cancer. Prostate cancer is the second leading cause of cancer death in men, with an estimated 220,900 new cases and 28,900 deaths in 2005 ¹. Although current therapies are generally effective in early stage disease, patients with locally recurrent or metastatic prostate cancer often have a poor prognosis ². Recently, gene therapy has emerged as a powerful tool with the potential to be effectively used as an adjuvant therapy for prostate cancer ³. Previous studies from this laboratory have demonstrated the feasibility of the sodium–iodide symporter (NIS) as a therapeutic gene for prostate cancer ^4,5,6,7,8. NIS is primarily a thyroid protein, expressed at the basolateral membrane of thyroid follicular cells and providing for the active accumulation of iodide for the biosynthesis of thyroid hormones ⁹. It is this ability of the thyroid gland to concentrate iodide that has made possible the use of radiolabeled iodide to successfully image and treat thyroid disease for over 50 years. Radioiodide therapy is effective and noninvasive, with an extensive safety record in the treatment of thyroid cancer ¹⁰.

Cloning and characterization of the NIS gene (approved symbol SLC5A5) led to the demonstration of its expression in other organs and presented the possibility of the use of radioiodide to treat other tumor types ^11,12,13. Although there is potential for breast cancer ¹⁴, thus far, native NIS expression in tissues other than the thyroid has not proven sufficiently strong to support radioiodide therapy. This has led to a number of studies attempting either to upregulate native expression or to introduce robust NIS expression in nonthyroid tissues using viral vectors ^{15,16,17,18,19}. Introduction of NIS expression in prostate cancer cells would provide for ¹²³I imaging to confirm correct localization of the protein, followed by ¹³¹I therapy of the disease in a fashion similar to that used for thyroid cancer. To be functional, the NIS protein must be correctly targeted to the cell membrane ²⁰. Also, to be therapeutically effective, concentrated iodide must be retained within the cell for a sufficient period of time. A number of studies using a variety of promoters to drive NIS expression in prostate cancer cells have shown that it is properly membrane-targeted and functional ^4,5,7,21,22. In the case of the thyroid, iodide organification by thyroid peroxidase is known to increase residence time of the iodide within the cell ⁹. The previously mentioned studies, along with similar ones in breast cancer ²³ and multiple myeloma cells ²⁴, have proven that radioiodide ablation can take place in the absence of iodide organification, which is absent when NIS is expressed in the prostate.

In the present study a replication-deficient adenoviral vector containing the cytomegalovirus (CMV) promoter driving expression of the human NIS gene was used. Previous studies using this vector in a mouse xenograft model of prostate cancer demonstrated effective radioiodide imaging and tumor ablation ⁵.

The current study was carried out in a clinically relevant large animal model in preparation for escalation to a Phase I clinical trial in patients with locally recurrent prostate cancer who have failed external beam radiotherapy. Our ultimate goal was to reinforce the evidence from previous studies in smaller models, examining the safety and efficacy of NIS as a therapeutic gene. The canine model described here closely resembles the clinical setting, as the dog prostate is similar in size and structure to that of humans ^25,26. Furthermore, the canine model allows the use of ¹³¹I doses that are similar to those anticipated for clinical trials in humans.

Results and discussion

¹²³I Imaging

In recent years NIS gene transfer has generated much interest as a result of its potential as both a reporter/diagnostic and a therapeutic gene. This potential has not yet been realized in a clinical setting. This study, combined with extensive toxicity data in rats (unpublished) and proven efficacy of the construct in mice ⁵, represents an important step before commencement of a Phase I clinical trial of NIS gene transfer for therapy of locally recurrent prostate cancer.

Vector biodistribution studies often involve invasive techniques such as tissue biopsy followed by PCR analysis ²⁶, whereas NIS expression can be tracked in a very straightforward, noninvasive fashion using radioactive iodide ²⁷. In this study, animals received direct intraprostatic injections of virus (1 10¹² particles) at multiple sites, to increase viral dispersal. SPECT/CT fusion imaging revealed clear images of the prostate in all dogs that were injected with Ad5/CMV/NIS (Figs. 1A, 1B, and 1D), with no prostate image visible in the control animal (Fig. 1C). This confirmed functional NIS expression at the target site and provided grounds to proceed to ¹³¹I therapy.

Figure 1.

Whole-body planar image of a dog (No. 385) that had received an intraprostatic injection of Ad5/CMV/NIS at (A) 5 h and (B) 24 h following ¹²³I administration. (C) The control animal (No. 384) received intraprostatic injections of saline, followed 3 days later by iv administration of 3 mCi ¹²³I. In each photo, a clear image was seen of the thyroid, salivary glands (where included), and stomach, which are known to express NIS. The urinary bladder is also apparent as a result of ¹²³I excretion. In the animal that was injected with Ad5/CMV/NIS (A, B), there is also a clear image of the prostate. The prostate is not apparent in the control animal (C). Image (D) is a SPECT/CT fusion image showing ¹²³I concentration in the prostate of a dog that had been injected with Ad5/CMV/NIS.

Full figure and legend (118K)

We also saw images of the thyroid and stomach, which express NIS, and the bladder, into which ¹²³I was being excreted. We carried out serial imaging 1–48 h after iodide administration. From the reconstructed tomographic images, we drew regions of interest (ROIs) around all visible organs and the whole body. We also drew background ROIs using appropriate regions adjacent to the organs of interest. After correction for background activity, we adjusted all ROI counts for any differences in acquisition time and for iodide decay (¹²³I half-life is 13.2 h). We then expressed the level of uptake in each organ as a percentage of the total activity in the dog and determined the dose of radioiodide delivered to each organ. This was done by adapting the Medical Internal Radiation Dose tables for humans ²⁸, relative to the size of the canine organs (thyroid 0.5 g, prostate 4 g, stomach 85 g, liver/spleen 330 g). This permitted calculation of the estimated mean dose to each organ per millicurie ¹³¹I administered (Table 1), corrected for acquisition time, background, and iodide decay (¹³¹I half-life is 192 h). We calculated the average absorbed dose to the prostate to be 23 2 cGy after an intravenous administration of 1 mCi ¹³¹I. This indicated that an 85-mCi dose of ¹³¹I would be sufficient to obtain a target dose of 2000 cGy to the prostate. This is substantially less than the 200 mCi maximum dose we have proposed. We predict that the levels of NIS expression in the prostate of patients will be even higher as viral delivery will be guided by transrectal ultrasound allowing up to 25 injection points, permitting improved viral dispersal within the gland. Also, patients can be maintained on lower iodine intake by dieting efforts that were not achievable in this study in dogs.

TABLE 1 - Estimated centigrays (cGy) of ¹³¹I delivered to each organ demonstrating uptake of radioiodine following a dose of 116 mCi/m².

Full table

As a result of the native ability of the thyroid gland to accumulate iodide efficiently, hypothyroidism is a clear possibility following therapy of this kind. This can be easily treated through the use of thyroid hormones. Patients will be informed of this before therapy. However, significant inhibition of thyroidal uptake is routinely achieved through the use of T₃ supplementation prior to and during radioiodide therapy ^10,14. This reduces thyroidal NIS expression by suppressing TSH production and so has a protective effect on the thyroid. A recent study by Wapnir et al. ¹⁴ reported an 75% reduction in thyroidal iodide uptake with the use of T₃. Since TSH regulation of NIS is exclusive to the thyroid, and not a characteristic of CMV promoter-driven NIS expression, this will not affect iodide uptake in the target tissue.

As a result of the 3D imaging techniques used, it was possible to separate reliably the intense image of the stomach from that of the liver. The estimated radioiodide dose to the liver in this model was extremely low (0.89 0.63 cGy/mCi) and, when scaled up for a dose of 86 mCi, would still be well within the known radiation dose tolerance of the liver routinely administered in many types of external radiation therapy treatments that primarily or secondarily include liver radiation. Consequently, such a dose would not be associated with any adverse toxicity ²⁹.

The level of iodide uptake in the stomach of this dog model is not a cause for concern as it was significantly higher than that seen routinely in humans ²⁸. This may be as a result of increased NIS expression in the canine gastric mucosa and may also be as a result of the pooling of gastric juices, as these animals were anesthetized for a prolonged period for serial imaging. It was not possible to perform vector distribution analysis at the time of imaging, as we did not perform necropsy until 13 days following virus injection, to permit ¹³¹I therapy and subsequent animal monitoring. Previous experience has shown that vector expression would not persist for this long, and extensive toxicity studies carried out in a rat model revealed no significant vector expression in the rat stomach. Further, gastric toxicity is not a reported side effect of radioiodide therapy of cancer ¹⁰ nor is gastric uptake readily observed in nuclear scans of patients undergoing iodide therapy. Although the current study employed a nonspecific promoter to drive NIS expression, the proposed clinical trial will pave the way for use of more recently developed prostate-specific promoters ^4,8,26, or the systemic administration of targeted vectors, to reduce even further the possibility of extratumoral NIS expression ^30,31. It should also be noted that in the anticipated clinical trial involving humans, pretherapy scans will be conducted to ensure that doses to such organs are minimal and as insignificant as those routinely employed in radioiodide therapy of thyroid cancer.

Toxicology monitoring

This study demonstrated the successful introduction of NIS expression into the prostate of dogs, with no vector-related toxicity observed. None of the animals experienced any surgical complications and there were no significant weight changes or signs of distress observed in any of the animals studied. One animal (No. 385) developed a fever after surgery, presumably as a result of an infection apparent at the incision site, but this was resolved after treatment with antibiotics.

We performed blood chemistry panels on samples taken on day 0 (laparotomy) and day 13 (necropsy). These results are shown in Table 2 and demonstrate that the liver function tests and white blood counts obtained both before and after vector administration remained within the normal range for these animals. Although a small selection of values obtained were outside the normal range (marked *), none of these were significantly changed.

TABLE 2 - Results from a panel of blood chemistries performed on whole-blood samples taken from each dog on day 0 (laparotomy) and day 13 (necropsy).

Full table

The histopathology of multiple sections from each organ was examined (Fig. 2 and Table 3). The prostate sections from the control dog (No. 384) differed considerably from that of the treated dogs (Nos. 382, 383, and 385). Neutrophils and macrophages predominated the inflammatory response in the control prostate sections (Fig. 2A), while lymphocytes and macrophages were more predominant in prostate sections from the treated dogs (Fig. 2B). Additional changes noted in prostate sections from treated dogs included fat necrosis (2/3) (Fig. 2C), hemorrhage (3/3), and thrombosis and vasculitis (1/3). The granulomatous inflammation in the urinary bladder sections from the control animal is likely a result of saline delivery into the adjacent prostate. Urinary bladder changes for the treated dogs (Fig. 2D) may be associated with the inflammatory changes in the prostate and/or with virus delivery/leakage.

Figure 2.

Immunohistochemical staining of tissue sections harvested at animal necropsy. (A) Prostate section from control animal (No. 384), with neutrophil and macrophage infiltration (original magnification 200). (B) Prostate section from treated animal (No. 382), with lymphocytes and macrophages more predominant (original magnification 200). (C) Fat necrosis and concomitant inflammatory response in prostate section from treated dog (original magnification 100). (D) Bladder thrombi observed in treated dog (original magnification 100). (E) Glycogen deposition in liver sections from treated dog (original magnification 100) (was also observed in control animal).

Full figure and legend (258K)

TABLE 3 - Major histopathologic findings in sections from organs harvested on day 13.

Full table

The changes noted in the liver consisted of focal nonsuppurative inflammation and hepatocellular vacuolation consistent with glycogen deposition (Fig. 2E). These are considered normal for nonfasted dogs of this strain and age.

The lack of changes in blood chemistries, including liver enzymes, and the absence of any significant findings on histopathological examination of liver sections from treated dogs support the view that this treatment will not be hepatotoxic. Also, a previously reported Phase I clinical trial using direct intraprostatic injection of a replicating adenovirus followed by prodrug therapy ³² reported little or no hepatotoxicity in humans.

The extensive experience and safety record of ¹³¹I therapy in the treatment of thyroid disease are major advantages for this approach to prostate cancer therapy. The side effects of ¹³¹I therapy are generally mild and short-lived, with serious side effects remaining extremely rare ¹⁰. Although previous studies have reported the successful imaging and therapy of prostate tumor xenografts in small-animal models, this is the first study of its kind in a large animal that included administration of therapeutic doses of ¹³¹I similar to those used in the clinical setting. None of the animals in this study experienced any surgical complications or vector-related toxicity. Blood chemistry panels showed no significant change following therapy, with only minimal inflammation observed in H&E-stained sections. The results presented here are very encouraging for NIS-based gene therapy and relevant directly to the planned Phase I clinical trial. They provide further evidence of the safety and efficacy of NIS as a therapeutic gene and support translation of this work into the clinical setting.

Materials and methods

Animals

All animal experiments were approved and carried out according to the guidelines of the Animal Care and Use Committee of the Mayo Clinic (Rochester, MN, USA). Four small adult male beagles (Nos. 382–385) were purchased from Marshall Farms (North Rose, NY, USA). Dogs were given oral T₃ (0.7 g/kg) beginning 8 days before viral infection, in an effort to reduce thyroidal uptake of iodide. An iodine-deficient diet was also initially used to enhance radioiodide uptake, but was poorly accepted by the animals and later replaced with a nutritionally complete canine diet (Lab Diet 5L18; PMI Nutrition International, Brentwood, MO, USA). The timeline of the experiment is shown in Fig. 3. Imaging and therapy were carried out in sets of two animals each (2 2), to meet equipment (imaging/anesthesiology) and radiation safety (dosage/exposure) requirements. Following virus/radioiodide administration, animals were housed in metabolic cages to facilitate cleaning and handling according to radiation safety and biosafety standards.

Figure 3.

Time course of experiment. Animals were given T₃ tablets (0.7 g/kg) for 8 days prior to undergoing a laparotomy and intraprostatic injection of virus. Blood samples were also collected prior to surgery. Three days later animals received an iv dose of 3 mCi ¹²³I followed by serial imaging 1, 3, 5, 24, and 48 h later. One day following completion of imaging, animals were transferred to a dedicated facility for administration of a therapeutic dose of ¹³¹I followed by 7 days of observation. On day 13, blood samples were collected, necropsy was performed, and tissues were harvested for histopathological analysis.

Full figure and legend (51K)

Adenovirus production

A replication-deficient human recombinant adenovirus serotype 5 construct containing human NIS under the control of the CMV promoter was produced in collaboration with the Mayo Clinic Vector Production Facility using previously described methods ⁴. Following plaque purification, the recombinant adenovirus Ad5/CMV/NIS was expanded in 293 cells and purified by banding on CsCl density gradients, followed by dialysis.

In vivo adenoviral delivery

Injections were carried out under general anesthesia (induction, 6.25 mg/kg Telazol, 0.125 mg/kg Butorphanol, 0.75 mg/kg Xylazine; maintenance, 0.5–2% isoflurane). A small midline incision laparotomy was performed (day 0), followed by direct visualization of the prostate and injection of virus (1 10¹² particles) using an insulin syringe. Three animals received intraprostatic injections of Ad5/CMV/NIS (Nos. 382, 383, 385), and one control animal was given intraprostatic injections of saline (No. 384). Injections were carried out at multiple sites to increase viral dispersal within the prostate, and cauterizing was used to limit leakage.

In vivo ¹²³I imaging

SPECT/CT fusion imaging was carried out 3 days following viral infection (days 3–5) using a SPECT/CT system (Hawkeye System; GE Medical Systems, Milwaukee, WI, USA). For SPECT, a total of 60 images were acquired using a 64 64 matrix, zoom 1.5, 360° circular orbit per study. For CT, 1-cm-thick slices were acquired over a 40-cm axial length. SPECT/CT images were acquired at 1, 3, 5, 24, and 48 h after administration of the ¹²³I. The SPECT images were reconstructed using an iterative reconstruction algorithm, which incorporated corrections for scatter and attenuation. Dogs were anesthetized (induction, 6.25 mg/kg Telazol, 0.125 mg/kg Butorphanol, 0.75 mg/kg Xylazine; maintenance, 0.5–2% isoflurane) and an iv catheter and urinary catheter placed before ¹²³I administration and imaging. Oral antibiotics (Cephalexin, 20 mg/kg) were given to the animals to prevent infection from the urinary catheter. Each animal received an iv injection of 3 mCi ¹²³I followed by serial image acquisitions at approximately 1, 3, 5, 24, and 48 h after injection.

¹³¹I administration

Following completion of imaging, animals were transported to a dedicated facility for administration of a therapeutic dose of ¹³¹I (day 6). The walls and floors of this facility were covered with plastic, as well as absorbent paper, to avoid radioiodide contamination from the cages. Personnel wore radiation detection badges and rings to monitor exposure and followed a tailored protocol for care of the animals following iodide administration. Radiation safety personnel carried out daily monitoring of radiation levels. Each animal received an iv dose of 116 mCi/m² (equivalent to 200 mCi/1.72 m², a common dose of radioiodine administered for thyroid cancer and the maximum dose proposed in our Phase I trial in humans). The weight of the animals was converted to body surface area according to the following equation:

Residual activity in the syringe after injection was taken into account to determine the precise dose received by each animal (Table 4). All radioisotope handling was carried out under the direct supervision of the Radiation Safety Department at the Mayo Clinic. The animals were sacrificed 7 days following ¹³¹I administration (100 mg/kg Pentobarbital, iv).

TABLE 4 - Dose of ¹³¹I administered to each animal, equivalent to 116 mCi/m².

Full table

Toxicology monitoring

The animals were routinely monitored three times daily and general observations were noted on their respective charts. Body weight was monitored and blood samples were taken on day 0 (laparotomy) and day 13 (necropsy) to detect any vector-related changes in blood chemistry. At time of necropsy, all organs that had demonstrated ¹²³I uptake upon imaging were harvested for histopathology.

References

1. Jemal, A., et al. (2005). Cancer statistics, 2005. CA Cancer J. Clin. 55: 10–30. | PubMed | ISI |
2. Prostate Cancer Trialists' Collaborative Group (2000). Maximum androgen blockade in advanced prostate cancer: an overview of the randomised trials. Lancet 355: 1491–1498.
3. Foley, R., Lawler, M. and Hollywood, D. (2004). Gene-based therapy in prostate cancer. Lancet Oncol. 5: 469–479. | Article | PubMed | ISI | ChemPort |
4. Kakinuma, H., et al. (2003). Probasin promoter (ARR₂PB)-driven, prostate-specific expression of the human sodium iodide symporter (h-NIS) for targeted radioiodine therapy of prostate cancer. Cancer Res. 15: 7840–7844.
5. Spitzweg, C., et al. (2001). In vivo sodium iodide symporter gene therapy of prostate cancer. Gene Ther. 8: 1524–1531. | Article | PubMed | ChemPort |
6. Spitzweg, C., et al. (2003). Retinoic acid-induced stimulation of sodium iodide symporter expression and cytotoxicity of radioiodine in prostate cancer cells. Endocrinology 144: 3423–3432. | Article | PubMed | ChemPort |
7. Spitzweg, C., et al. (1999). Prostate-specific antigen promoter-driven androgen-inducible expression of sodium iodide symporter in prostate cancer cell lines. Cancer Res. 59: 2136–2141. | PubMed | ISI | ChemPort |
8. Spitzweg, C., et al. (2000). Treatment of prostate cancer by radioiodine therapy after tissue-specific expression of the sodium iodide symporter. Cancer Res. 60: 6526–6530. | PubMed | ChemPort |
9. Carrasco, N. (1993). Iodide transport in the thyroid gland. Biochim. Biophys. Acta 1154: 65–82. | Article | PubMed | ISI | ChemPort |
10. Mazzaferri, E. and Kloos, R. (2001). Current approaches to primary therapy for papillary and follicular thyroid cancer. J. Clin. Endocrinol. Metab. 86: 1447–1463. | Article | PubMed | ISI | ChemPort |
11. Smanik, P., et al. (1996). Cloning of the human sodium iodide symporter. Biochem. Biophys. Res. Commun. 226: 339–345. | Article | PubMed | ISI | ChemPort |
12. Dai, G., Levy, O. and Carrasco, N. (1996). Cloning and characterization of the thyroid iodide transporter. Nature 379: 458–460. | Article | PubMed | ISI | ChemPort |
13. Spitzweg, C., Harrington, K., Pinke, L., Vile, R. and Morris, J. (2001). The sodium iodide symporter and its potential in cancer therapy. J. Clin. Endocrinol. Metab. 86: 3327–3335. | Article | PubMed | ChemPort |
14. Wapnir, I. L., et al. (2004). The Na⁺/I^- symporter mediates iodide uptake in breast cancer metastases and can be selectively down-regulated in the thyroid. Clin. Cancer Res. 10: 4294–4302. | Article | PubMed | ChemPort |
15. Chung, J.-K. (2002). Sodium iodide symporter: its role in nuclear medicine. J. Nucl. Med. 43: 1188–1200. | PubMed | ChemPort |
16. Dadachova, E. and Carrasco, N. (2004). The Na⁺/I^- symporter (NIS): imaging and therapeutic applications. Semin. Nucl. Med. XXXIV: 23–31.
17. Faivre, J., et al. (2004). Long-term radioiodine retention and regression of liver cancer after sodium iodide symporter gene transfer in Wistar rats. Cancer Res. 64: 8045–8051. | Article | PubMed | ISI | ChemPort |
18. Scholz, I., et al. (2004). Radioiodine therapy of colon cancer following tissue-specific sodium iodide symporter gene transfer. Gene Ther. 12: 272–280.
19. Kogai, T., et al. (2004). Systemic retinoic acid treatment induces sodium/iodide symporter expression and radioiodide uptake in mouse breast cancer models. Cancer Res. 64: 415–422. | Article | PubMed | ChemPort |
20. Dohan, O., Baloch, Z., Banrevi, Z., Livolsi, V. and Carrasco, N. (2001). Predominant intracellular overexpression of the Na⁺/I^- symporter (NIS) in a large sampling of thyroid cancer cases. J. Clin. Endocrinol. Metab. 86: 2697–2700. | Article | PubMed | ChemPort |
21. La Perle, K. M., et al. (2002). In vivo expression and function of the sodium iodide symporter following gene transfer in the MATLyLu rat model of metastatic prostate cancer. Prostate 50: 170–178. | Article | PubMed | ChemPort |
22. Mitrofanova, E., et al. (2004). Rat sodium iodide symporter for radioiodine therapy of cancer. Clin. Cancer Res. 10: 6969–6976. | Article | PubMed | ISI | ChemPort |
23. Dwyer, R., Bergert, E., O'Connor, M., Gendler, S. and Morris, J. (2005). In vivo radioiodide imaging and treatment of breast cancer xenografts following MUC1-driven expression of the sodium iodide symporter (NIS). Clin. Cancer Res. 11: 1483–1489. | Article | PubMed | ISI | ChemPort |
24. Dingli, D., et al. (2003). Genetically targeted radiotherapy for multiple myeloma. Gene Ther. 102: 489–496. | ChemPort |
25. Barton, K. N., et al. (2003). GENIS: gene expression of sodium iodide symporter for noninvasive imaging of gene therapy vectors and quantification of gene expression in vivo. Mol. Ther. 8: 508–518. | Article | PubMed | ChemPort |
26. Steiner, M., Zhang, Y., Carraher, J. and Lu, Y. (1999). In vivo expression of prostate-specific adenoviral vectors in a canine model. Cancer Gene Ther. 6: 456–464. | Article | PubMed | ChemPort |
27. Barton, K. N., et al. (2004). A quantitative method for measuring gene expression magnitude and volume delivered by gene therapy vectors. Mol. Ther. 9: 625–631. | Article | PubMed | ChemPort |
28. (1975). MIRD dose estimate report No. 5: summary of current radiation dose estimates to humans from ¹²³I, ¹²⁴I, ¹²⁶I, ¹³⁰I, ¹³¹I and ¹³²I as sodium iodide. J. Nucl. Med. 16: 857–860.
29. Emami, B., et al. (1991). Tolerance of normal tissue to therapeutic irradiation. Int. J. Radiat. Oncol. Biol. Phys. 21: 109–122. | PubMed | ChemPort |
30. Hart, I. R. (1996). Tissue specific promoters in targeting systemically delivered gene therapy. Semin. Oncol. 23: 154–158. | PubMed | ISI | ChemPort |
31. Krasnykh, V., Douglas, J. T. and Van Beusechem, V. W. (2000). Genetic targeting of adenoviral vectors. Mol. Ther. 1: 391–405. | Article | PubMed | ISI | ChemPort |
32. Freytag, S. O., et al. (2002). Phase I study of replication-competent adenovirus-mediated double suicide gene therapy for the treatment of locally recurrent prostate cancer. Cancer Res. 62: 4968–4976. | PubMed | ISI | ChemPort |

Acknowledgements

This work was supported by a Prostate Cancer SPORE grant (CA91956). The authors are grateful to Elton Mosman (Department of Radiology) for his assistance with SPECT/CT imaging. We also thank Krista Thompson (Department of Comparative Medicine) for her assistance with animal care and Diane Soeffker (Molecular Medicine Program) for assistance with animal surgery and necropsy.