Imaging a Genetically Engineered Oncolytic Vaccinia Virus (GLV-1h99) Using a Human Norepinephrine Transporter Reporter Gene

Malignant pleural mesothelioma and pancreatic cancer are highly aggressive diseases. The annual incidence in the United States was estimated to be ∼40,000 cases for pancreatic cancer and ∼4,000 cases for malignant mesothelioma in the year 2004 (1). The increasing incidence of mesothelioma worldwide, especially in industrialized nations, is due to the etiology of this disease from asbestos exposure (2). Both of these tumors are highly resistant to current therapy, with 5-year survival rates of only 5% for pancreatic cancer and 9% for mesothelioma (3). Even with combined surgery, chemotherapy, and radiation, only a small minority of patients are rendered disease-free for a prolonged period of time (4).

Oncolytic viral therapy has been studied and tested over the past century, and many viral types, including adenovirus, herpes simplex virus, Newcastle disease virus, myxoma virus, vaccinia virus, and vesicular stomatitis virus, are being investigated as novel agents for the treatment of human cancer (5). Adenovirus H101 was approved in 2005 for the treatment of head and neck cancer in China (6). Importantly, viruses generally kill cancer cells that are high in ribonucleotide reductase, high in DNA repair enzymes, and resistant to apoptosis, characteristics that tend to make tumor cells resistant to chemotherapy and radiation therapy (7).

Vaccinia viruses are particularly attractive agents for oncolytic therapy because versions of this virus have been given to millions of humans in the eradication of smallpox. Vaccinia virus is also an excellent vector because its large genome allows for insertion of multiple foreign genes (8). In addition, it is highly immunogenic and able to induce strong host immune responses against virus-infected cells (9). Although the acceptance of oncolytic viral therapy has been mixed in the medical community, a substantial amount of data has been reported from clinical trials with vaccinia virus in cancer patients showing good safety and promising responses (9–12).

Future human oncolytic viral therapy studies would benefit greatly from the ability to track and monitor viral distribution, tumor targeting, proliferation, and persistence by noninvasive imaging (13). It would provide important safety, efficacy, and toxicity correlations. Such real-time tracking would also provide useful viral dose and administration schedule information for optimization of therapy and would obviate the need for multiple and repeated tissue biopsies. The virus used in the current study is GLV-1h99. This is a recombinant vaccinia virus expressing transgenes for the human norepinephrine transporter (hNET) and β-galactosidase. hNET is a cell-membrane transporter that is highly expressed in many neuroendocrine tumors and can be imaged by radiolabeled metaiodobenzylguanidine (MIBG). The use of hNET reporter gene imaging is particularly attractive from a clinical investigative standpoint because (a) hNET is a human protein that should minimize immunogenicity and (b) MIBG can be radiolabeled with ¹²³I or ¹³¹I for single-photon emission computed tomography (SPECT) and γ-camera imaging and also with ¹²⁴I for positron emission tomography (PET) imaging. Currently, [¹²³I]MIBG is a clinically approved radiolabeled probe for imaging hNET expression.

In this study, we show that insertion of the hNET reporter gene into a recombinant vaccinia virus does not alter tumor killing; GLV-1h99 retains excellent tumor specificity and cytotoxic efficacy. In addition, we show the feasibility of using the hNET reporter system for in vivo noninvasive imaging of oncolytic viral therapy in an orthotopic pleural mesothelioma tumor model by [¹²³I]MIBG γ-camera imaging and [¹²⁴I]MIBG PET imaging.

Cell lines. Human pancreatic carcinoma cell lines PANC1, BxPC-3, HS766T, and MiaPaCa-2, the mesothelioma cell line MSTO-211H, and the human neuroblastoma cell line SK-N-SH, which expresses hNET, were obtained from the American Type Culture Collection. JMN cells were a kind gift from Dr. Frank Sirotnik (Memorial Sloan-Kettering Cancer Center, New York, NY). H2052 and H2373 cell lines were a kind donation from Dr. Pass (Karmanos Cancer Institute, Wayne State University, Detroit, MI).

All cells were grown in appropriate medium, maintained in a humidified incubator at 37°C supplied with 5% CO₂, and subcultured twice weekly.

Virus strains. GLV-1h68 (non–hNET-containing virus) is a replication-competent, recombinant vaccinia virus derived from the LIVP strain (Lister strain from the Institute for Research on Virus Preparations, Moscow, Russia), and its construction was previously described (14).

GLV-1h99 (hNET-expressing virus) was derived from GLV-1h68 by replacing the Renilla luciferase-green fluorescent protein expression cassette at the F14.5L locus with a hNET expression cassette through in vivo homologous recombination.

Cytotoxicity assay. Cells were plated at 2 × 10⁴ per well in 12-well plates in 1 mL of medium per well. After incubation for 6 h, cells were infected with GLV-1h99 or GLV-1h68 at multiplicities of infection (MOI) of 1.0, 0.10, 0.01, and 0 (control wells). Viral cytotoxicity was measured daily for 7 d. Cells were washed with PBS and lysed in 200 μL per well of 1.5% Triton X-100 (Sigma) to release intracellular lactate dehydrogenase, which was quantified with a CytoTox 96 kit (Promega) on a spectrophotometer (EL321e, Bio-Tek Instruments) at 490 nm. Results are expressed as the percentage of surviving cells. This percentage was determined by comparing the measured lactate dehydrogenase of each infected sample with that in uninfected control cells. All samples were analyzed in triplicate.

Immunoblot analysis. To evaluate the level of hNET protein expression in cells (H2052, MSTO-211H, and PANC1) infected with virus (GLV-1h99 or GLV-1h68) and in the neuroblastoma cell line SK-N-SH at 12, 24, 48, and 72 h after infection, immunoblot analysis was done. A purified mouse antibody against hNET (NET17-1; MAb Technologies, Inc.) was used at a final dilution of 1:500 and incubated for 12 h at +4°C. The secondary antibody (peroxidase-conjugated anti-mouse IgG; Vector Laboratories, Inc.) exposure was for 1 h at a 1:2,000 dilution. Peroxidase-bound protein bands were visualized using the enhanced chemiluminescence method (Amersham Pharmacia Biotech).

In vitro radiotracer assay. [¹²³I]MIBG radiotracer uptake studies were done in MSTO-211H and PANC1 cells after infection with virus (GLV-1h99 or GLV-1h68) as well as in the neuroblastoma cell line SK-N-SH using previously described methods (15). Briefly, cells were plated at 1 × 10⁶ per well in six-well plates in 2 mL of medium per well. After incubation for 6 h, cells were infected with GLV-1h68 or GLV-1h99 at MOIs of 1.0 and 0 (control wells). Following 12-, 24-, 48-, and 72-h incubation periods with virus at 37°C and 5% CO₂, the medium was aspirated and the cells were washed with PBS (pH 7.4). [¹²³I]MIBG uptake was initiated by adding 2 mL of DME containing 0.0185 MBq/mL (0.5 μCi/mL) carrier-free [¹²³I]MIBG. Cells were harvested after a 60-min incubation period, and the cell pellet-to-medium activity ratio (cpm/g of pellet/cpm/mL of medium) was calculated from the radioactivity measurements assayed in a gamma counter (Packard, United Technologies). All studies were done in triplicate.

Malignant pleural mesothelioma xenograft model. Athymic nu/nu female mice were purchased from the National Cancer Institute (Bethesda, MD) and were housed five per cage and allowed food and water ad libitum in the Memorial Sloan-Kettering Cancer Center Vivarium for 1 wk before tumor cell implantation. All animal studies were done in compliance with all applicable policies, procedures, and regulatory requirements of the Institutional Animal Care and Use Committee, the Research Animal Resource Center of Memorial Sloan-Kettering Cancer Center, and the NIH “Guide for the Care and Use of Laboratory Animals.” All animal procedures were done under anesthesia induced by inhalation of 2% isoflurane. After the studies, all animals were sacrificed by CO₂ asphyxiation.

An incision 3 to 5 mm in length was made over the fourth to fifth intercostal space of the right chest. The underlying inflating and deflating lung was thereby easily visualized through the thin fascia. Slowly, 100 μL of MSTO-211H malignant mesothelioma cellular suspension (5 × 10⁶ cells) were injected. After the injection, the skin was closed with surgical staples and mice were returned to their cages.

Intrapleural treatment with virus was done in a similar fashion as described above 10 d after tumor cell instillation into the pleural cavity. GLV-1h99 or GLV-1h68 (1 × 10⁷ plaque-forming units) was administered in 100 μL PBS and animals were gently rotated from side to side to help distribute the virus throughout the pleural cavity. Control animals (no virus) received only 100 μL PBS.

MIBG synthesis. Clinical-grade [¹²³I]MIBG was obtained from MDS Nordion. The average radiochemical purity was in excess of 97% (determined by MDS Nordion using the Sep Pak cartridge method), and the specific activity was ∼320 MBq/μmol (8.7 mCi/μmol) according to the vendor.

[¹²⁴I]MIBG was prepared using minor modifications to the reported nucleophilic isotopic exchange method (16), following a procedure previously reported by Moroz and coworkers (17). The radiochemical purity of the final product was >95% with an overall yield of >80% and the specific activity was 18.5 ± 5.2 MBq/μmol (0.5 ± 0.14 mCi/μmol). The maximum specific activities (no carrier-added synthesis) for the ¹²³I- and ¹²⁴I-labeled compounds were 8.9 and 1.2 TBq/μmol (241 and 33 Ci/μmol), respectively, due to the 7.4-fold difference in the decay rate of the two isotopes.

Clinical-grade [¹⁸F]fluorodeoxyglucose (FDG) was obtained from IBA Molecular with a specific activity of >41 MBq/μmol (>11 mCi/μmol) and a radiochemical purity of >98%.

[¹²³I]MIBG γ-camera in vivo imaging. Each animal was injected i.v. with ∼18.5 MBq (500 μCi) of [¹²³I]MIBG 48 h after intrapleural GLV-1h99 injection and imaged on a X-SPECT dedicated small-animal γ-camera SPECT-CT scanner (Gamma Medica). A photopeak energy window of 143 to 175 keV and a low-energy high-resolution parallel-hole collimator was used to acquire 10-min ¹²³I images at 2 h after [¹²³I]MIBG administration.

The X-SPECT γ-camera system was calibrated by imaging a mouse-size (30 mL) cylinder filled with an independently measured concentration (MBq/mL) of ^99mTc using a photopeak energy window of 126 to 154 keV and low-energy high-resolution collimation. The resulting ^99mTc images were exported to Intefile and then imported into the ASIPro (Siemens Pre-clinical Solutions) image processing software environment. By region of interest analysis, a system calibration factor (in cpm/pixel per MBq/mL) was derived. Animal images were likewise exported to Intefile and then imported into ASIPro and parameterized in terms of the decay-corrected percentage injected dose per gram (%ID/g) based on the foregoing calibration factor, the administered activity, the time after administration of imaging, and the image duration. Implicit in the foregoing analysis is the reasonable assumption that the sensitivities of the X-SPECT γ-camera system for ¹²³I and ^99mTc are comparable.

[¹²⁴I]MIBG micro-PET in vivo imaging. In a group of five animals (10 d after MSTO-211H tumor cell instillation into the pleural cavity), each animal was injected via the tail vein with 9.25 MBq (250 μCi) of [¹⁸F]FDG. [¹⁸F]FDG PET scanning was done 1 h after tracer administration using a 10-min list-mode acquisition. Animals were fasted 12 h before tracer administration and kept under anesthesia between FDG injection and imaging.

In a group of 16 animals, 4 subgroups of 3 to 5 animals each were studied (5 animals in subgroups 1 and 2; 3 animals in subgroups 3 and 4). Each animal was injected via the tail vein with 9.25 MBq (250 μCi) of [¹²⁴I]MIBG. Animals in subgroups 1 and 2 were injected with GLV-1h99 48 and 72 h before [¹²⁴I]MIBG administration. Subgroup 3 animals received GLV-1h68 48 h before radiotracer administration; subgroup 4 animals were not injected with virus, receiving only 100 μL PBS. Potassium iodide was used to block the uptake of radioactive iodine by the thyroid. [¹²⁴I]MIBG PET was done for 10 min, 1 h, 2 h, and 4 h after tracer administration, for 15 min at 12 h, for 30 min at 24 h, and for 60 min at 48 h. After tracer administration and between imaging time points, the animals were allowed to wake up and maintain normal husbandry.

Imaging was done using a Focus 120 micro-PET dedicated small-animal PET scanner (Concorde Microsystems, Inc.). Three-dimensional list-mode data were acquired using an energy window of 350 to 700 keV for ¹⁸F and 410 to 580 keV for ¹²⁴I, respectively, and a coincidence timing window of 6 ns. These data were then sorted into two-dimensional histograms by Fourier rebinning. The image data were corrected for (a) nonuniformity of scanner response using a uniform cylinder source-based normalization, (b) dead time count losses using a single-count rate-based global correction, (c) physical decay to the time of injection, and (d) the ¹²⁴I branching ratio. The count rates in the reconstructed images were converted to activity concentration (%ID/g) using a system calibration factor (MBq/mL per cps/voxel) derived from imaging of a mouse-size phantom filled with a uniform aqueous solution of ¹⁸F.

Image analysis was done using ASIPro. At all acquired scans (including [¹⁸F]FDG PET, serial [¹²⁴I]MIBG PET, and [¹²³I]MIBG γ-camera), regions of interest were manually drawn over tumor, lung, liver, and skeletal muscle. For each tissue and time point after injection, the measured radioactivity was expressed as %ID/g. The maximum %ID/g value was recorded for each tissue, and from these, tumor-to-organ ratios for lung, liver, and skeletal muscle were then calculated.

Immunohistochemistry. After the final image, the animals were sacrificed and the tumors were harvested and frozen in Tissue-Tek OCT compound (Sakura Finetek USA, Inc.). Tissues were cut into 5-μm-thick sections and mounted on glass slides. Cryosections were fixed and stained with H&E and 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (1 mg/mL) in an iron solution of 5 mmol/L K₄Fe(CN)₆, 5 mmol/L K₃Fe(CN)₆, and 2 mmol/L MgCl₂, as previously described (18), to identify virally mediated lacZ expression.

Statistics. A two-tailed unpaired t test was applied to determine the significance of differences between values using the MS Office 2003 Excel 11.0 statistical package (Microsoft).

Cytotoxicity assays in vitro showed dose-dependent lytic activity. Four mesothelioma and four pancreatic cancer cell lines showed lytic cytotoxicity following exposure to GLV-1h99 (hNET-expressing virus) and to GLV-1h68 (non–hNET-containing virus). Similar cytotoxicity was observed with GLV-1h99 and GLV-1h68 at a MOI of 1.0 (Fig. 1A) and a dose-dependent lytic effect was also shown (Fig. 1B). At a MOI of 0.1, all MSTO-211H and H2052 mesothelioma cells as well as 80% of the PANC1 pancreatic cancer cells were dead at day 7. Oncolysis seemed to be more gradual over time in PANC1 cells compared with the more sigmoidal lytic time profile in MSTO-211H cells (Fig. 1). The mesothelioma cell line JMN and the pancreatic cancer cell line HS766T were more resistant and showed only 80% cell death by day 7 at a MOI of 1.0 (Fig. 1C). MiaPaCa2 and BxPC3 (pancreatic cancer cell lines) and H2373 (mesothelioma cell line) were sensitive to the virus only at a higher MOI of 10 (data not shown).

Immunoblot analysis confirmed dose-dependent levels of transgene hNET expression. The two most sensitive mesothelioma cell lines (MSTO-211H and H2052) and the most sensitive pancreatic cancer cell line (PANC1), based on the cytotoxicity assays, were chosen for immunoblot analysis and compared with the endogenous hNET-expressing neuroblastoma cell line, SK-N-SH. The levels of hNET expression 24 hours after GLV-1h99 (hNET-expressing virus) viral infection at a MOI of 1.0 were investigated (Fig. 2A). In addition to the ∼80-kDa hNET band, two low-molecular weight immunoreactive bands (∼50-55 kDa and ∼37-40 kDa, respectively) are seen in the blots of the GLV-1h99–infected cells; these two low-molecular weight bands are barely visible in the blots of the SK-N-SH neuroblastoma cells. Similar to the cytotoxicity assay, there was a viral dose-dependent expression of hNET at different MOIs (0.1, 1.0, 5, and 10; shown for MSTO-211H cells in Fig. 2B). Strong hNET expression was found in the MSTO-211H and PANC1 cell lines by 12 hours after GLV-1h99 viral infection, peaking at 24 hours followed by a gradual decline over 72 hours (shown for MSTO-211H cells in Fig. 2C and D). A similar pattern of hNET expression was observed in the other cell lines, although the hNET immunoblot bands were less intense (data not shown). The neuroblastoma cell line (SK-N-SH), expressing endogenous hNET, served as a positive control for the immunoblot analysis and radiotracer uptake studies; the GLV-1h68 virus-infected (non–hNET-containing virus) and the uninfected mesothelioma and pancreatic cancer cell lines served as negative controls (Fig. 2B).

In vitro [¹²³I]MIBG uptake showed peak of radiotracer uptake 48 hours after virus infection. The time course of [¹²³I]MIBG accumulation was studied in PANC1 and MSTO-211H cells following infection with the GLV-1h99 (hNET-expressing virus) virus at a MOI of 1.0. [¹²³I]MIBG accumulation in noninfected MSTO-211H and PANC1 cells was low (Fig. 3A and B, respectively). There was no significant increase in radiotracer uptake 24 hours after infection of the cells with GLV-1h68 (non–hNET-containing virus, negative control). In contrast, there was a significant (P < 0.01) increase in [¹²³I]MIBG accumulation in both cancer cell lines at all time points (12, 24, 48, and 72 hours) after infection with GLV-1h99 (Fig. 3A and B). Peak radiotracer uptake was observed at 48 hours after virus infection in both cell lines. The natural hNET-expressing neuroblastoma cell line (SK-N-SH) served as a positive control. Total cell protein in the [¹²³I]MIBG uptake assays was unchanged over the first 24 hours following GLV-1h99 infection compared with uninfected cells. At 48 and 72 hours after viral infection, there was a decrease in measured cell protein (Fig. 3C and D).

hNET expression imaging by [¹²³I]MIBG scintigraphy and [¹²⁴I]MIBG PET. Following direct injection of hNET-expressing GLV-1h99 virus into MSTO-211H orthotopic pleural tumors, viral localization was visualized by [¹²⁴I]MIBG PET imaging of hNET expression in pleural tumors (Fig. 4A). [¹²⁴I]MIBG was i.v. administered 48 or 72 hours after intrapleural virus injection, and sequential PET imaging was done 1 to 48 hours after radiotracer administration. Tumor radioactivity values (%ID/g) were measured and tumor-to-organ ratios were calculated. The highest levels of radioactivity in the pleural tumors were found 48 hours after injection of GLV-1h99 (hNET-expressing virus) followed by tumors that were injected with GLV-1h99 72 hours before [¹²⁴I]MIBG administration. Low levels of radioactivity were observed in tumors that were injected with GLV-1h68 (non–hNET-containing virus) and in tumors that were not injected with virus (Fig. 4B). Maximum activity in both the pleural tumors and remote organs (background) was observed at the time of the initial measurement, 1 hour after radiotracer administration. Tumor and remote organ activity decreased over time (1-72 hours) in all four groups of animals. The decrease in tumor activity was more rapid over the first 12 hours after [¹²⁴I]MIBG administration in the two control groups: tumors injected with GLV-1h68 (non–hNET-containing virus) or no virus.

Tumor-to-organ (lung, liver, and muscle) ratios were calculated from the PET image data (as described in Materials and Methods) and the highest values were obtained for the group of animals that were infected with GLV-1h99 (hNET-expressing virus) 48 hours before radiotracer administration (Fig. 5). Comparing the animals that were treated with GLV-1h99 48 hours before [¹²⁴I]MIBG administration with the animals that received no virus, the ratio differences were highly significant (P < 0.01) at the 2-hour imaging time point and significant (P < 0.05) at the 1-hour imaging time point. Nearly the same low tumor-to-organ ratios were found for the two control groups of animals and the tumor-to-organ ratios decreased over time.

For localization of the tumors and for comparison with a clinically used imaging technique, [¹⁸F]FDG PET imaging was also done. [¹²⁴I]MIBG PET and [¹⁸F]FDG PET imaging were compared (Figs. 4 and 5). The pleural tumors were visualized by [¹⁸F]FDG PET imaging, but image contrast at 48 and 72 hours after GLV-1h99 virus (hNET-expressing virus) injection was greater with [¹²⁴I]MIBG PET compared with [¹⁸F]FDG PET. The [¹²⁴I]MIBG and [¹⁸F]FDG tumor-to-lung, tumor-to-liver, and tumor-to-muscle ratios in control animals were similar.

In vivo hNET expression in the pleural tumors after GLV-1h99 (hNET expressing) virus administration could also be imaged by [¹²³I]MIBG planar scintigraphy. All GLV-1h99–injected animals showed localized accumulation of [¹²³I]MIBG radioactivity in the virus-injected pleural tumors compared with the control animals that received no virus (Fig. 6). The tumor-to-background ratios for the GLV-1h99–infected animals were with 2.4 ± 0.2, significantly (P < 0.01) higher compared with the group that received no virus, 1.5 ± 0.1.

Immunohistochemistry confirmed viral presence in tumors. All animals were sacrificed and examined to confirm the presence of pleural tumors. All pleural lesions were shown to be malignant pleural mesothelioma on H&E staining (Fig. 7A and B). In addition, all tumors infected with vaccinia virus stained positive for lacZ, confirming the presence of the virus in tumors and indicating that all tumors visualized by [¹²⁴I]MIBG PET or [¹²³I]MIBG scintigraphy reflect GLV-1h99 expression of a functional hNET transporter protein (Fig. 7C and D).

Oncolytic viral therapy dates back more than a century (19) and has had a mixed and fluctuating level of acceptance in the medical community. Nevertheless, it still holds promise as a biological treatment for some standard therapy–resistant cancers (11). Several viruses (e.g., adenovirus, herpes simplex virus, Newcastle disease virus, myxoma virus, vaccinia virus, and vesicular stomatitis virus) have been shown to infect and replicate in cancer cells and to selectively kill them (oncolysis). Oncolytic viral therapy has been used in cancer treatment and has evolved from the use of wild-type viruses to genetically engineered viruses that express therapeutic transgenes. Clinical and preclinical trials involving different viral strains and constructs have shown to be safe and to have potent antitumor effects (10). Oncolytic viruses have also shown enhanced efficacy involving combination regimens with approved chemotherapeutics and radiotherapy (20). Two oncolytic viruses (G207 and H101) have entered randomized phase III clinical testing (10), and marketing approval was obtained for H101 in 2005 (6). Future studies are likely to focus on optimization of viral doses and administration routes, interaction with the immune system, and in vivo monitoring through imaging. The ability to noninvasively and repetitively identify anatomic sites of viral targeting and to measure the magnitude of viral infection could provide important safety, efficacy, and toxicity information during clinical studies of viral oncolysis.

Vaccinia virus is perhaps the most widely administered medical product in history; it is certainly the most successful biological product. Vaccinia also displays many of the qualities thought necessary for an effective antitumor agent and it is particularly well characterized in humans due to its role in the eradication of smallpox. Vaccinia has a short life cycle and spreads rapidly; it has inherent systemic tumor targeting, a high propensity to induce cell lysis, well-defined biology, and a large cloning capacity (12). The large insertional cloning capacity allows for the inclusion of several functional and therapeutic transgenes. With the insertion of reporter genes not expressed in uninfected cells, viruses can be localized and the course of viral therapy can be monitored. A noninvasive, clinically applicable method for imaging viruses in target tissue or specific organs of the body would be of considerable value during oncolytic viral therapy in patients.

In this study, we describe the use of a genetically modified vaccinia virus, GLV-1h99, which has been engineered for specific targeted treatment of cancer and for noninvasive imaging. GLV-1h99 is able to efficiently infect, replicate in, and lyse a variety of human pancreatic and mesothelioma cancer cell lines. The oncolytic potency of GLV-1h99 was shown to be similar to the non–hNET-containing parent virus, GLV-1h68, in eight pancreatic and mesothelioma cancer cell lines. GLV-1h68 has also been shown to successfully treat an orthotopic animal model of mesothelioma with pleural disease (21).

The reporter gene chosen for insertion into GLV-1h99 was based on the very favorable PET and SPECT imaging characteristics of the hNET-MIBG reporter imaging system (17) and because [¹²³I]MIBG is an approved radiopharmaceutical for clinical imaging of neuroendocrine tumors (22, 23). In contrast to a study published by McCart et al. (24) using an oncolytic vaccinia virus expressing the human somatostatin receptor SSTR2, hNET is a transporter-based reporter gene system. Receptors usually have a 1:1 binding relationship with a radiolabeled ligand; transporters provide signal amplification through transport-mediated concentrative intracellular accumulation of the radiolabeled substrate. hNET is a transmembrane protein that mediates the transport of norepinephrine, dopamine, and epinephrine across the cell membrane (25). It is one of several human reporter genes that are currently being used in preclinical studies (17, 26) and has a high potential for rapid translation into clinical reporter gene imaging studies (27, 28).

The hNET immunoblots showed protein expression in all infected cell lines and the expression was time and dose dependent. The antibody used recognizes a degraded or less glycosylated form of the protein (∼50-55 kDa and ∼37-40 kDa) as well as a more highly glycosylated hNET protein (∼80 kDa). Interestingly, the low-molecular weight bands were more intense early (12-24 hours) after GLV-1h99 infection compared with later time points (48-72 hours; Fig. 2). These bands also appear in SK-N-SH neuroblastoma cells and in hNET-transduced cell lines (17) but at much lower intensity. We suspect that the prominence of the low-molecular weight bands is the effect of viral infection, replication, and lysis and that the low-molecular weight immunoreactive protein may be nonfunctional with respect to MIBG transport and accumulation.

In vitro [¹²³I]MIBG uptake studies also showed time-dependent radiotracer uptake, peaking at 24 to 48 hours after viral infection (∼5-fold above control) for MSTO-211H cells. The uptake levels were lower in the GLV-1h99–infected cells compared with the hNET-expressing neuroblastoma cells. In the viral-treated cultures, it is likely that not all of the cells are infected with virus and therefore not all express the reporter gene during the early phase of viral infection. In addition, the hNET protein may not have been translocated and inserted into the cell membrane to form a functional transporter during the initial 12- to 24-hour period after infection. During the late, prelytic phase of viral infection (72 hours and beyond), the hNET transporter could be impaired, and following cell lysis, the accumulated MIBG radiotracer would be lost. Thus, there seems to be a relatively narrow window, ∼24 to 48 hours after viral infection of MSTO-211H cells, during which the hNET reporter is maximally functional. These results reflect the dynamic state between viral infection, replication, and lysis of tumor cells.

This dynamic state of viral infectivity and functional hNET expression was also observed in the in vivo imaging studies. It should be noted that the whole tumor is only partially infected with virus and tumor cells are at different stages of virus infection at any given time (shown by immunohistochemistry). Timing of [¹²⁴I]MIBG PET imaging after GLV-1h99 virus injection was very important in the in vivo studies. Better imaging results were obtained at 48 hours compared with 72 hours after viral injection. MIBG uptake in GLV-1h99–infected cells, both in vitro and in vivo, is not exactly comparable with MIBG uptake in SK-N-SH neuroblastoma cells or in cells transduced with constitutive hNET expression cassettes (17) where expression levels are constant. Nevertheless, the quantitative [¹²⁴I]MIBG PET and [¹²³I]MIBG SPECT studies showed that imaging of GLV-1h99 viral infection of MSTO-211H pleural tumors is feasible after direct tumor injection.

Similarly, the timing of PET imaging after [¹²⁴I]MIBG i.v. injection was also shown to be important. Radioactivity levels (% dose/cm³) as well as tumor-to-organ ratios in GLV-1h99–infected tumors were highest during the first 4-hour period after tracer administration. This differs from the findings of Moroz et al. (17) in a xenograft model, where constitutive expression of the hNET reporter occurred in all tumor cells and optimal imaging results were obtained at late time points (48 and 72 hours after tracer administration). This time-dependent difference probably reflects the effect of increasing cell death resulting from viral oncolysis after [¹²⁴I]MIBG injection. Oncolysis will result in a loss of [¹²⁴I]MIBG from the infected tumor cells and tumor, consistent with the rapid decrease in the PET signal after 4 hours (Fig. 4B). A similar pattern was observed in the in vitro immunoblot analysis and [¹²³I]MIBG uptake studies, showing decreasing hNET expression and radiotracer uptake at later time points after GLV-1h99 infection (Fig. 3B).

Pancreatic cancer and malignant pleural mesothelioma remain largely unresponsive to standard treatments and are rapidly fatal diseases in most cases. Thus, alternate treatment options must be considered. Preclinical studies have shown the efficacy of oncolytic herpes simplex viruses in treatment of pancreatic cancer (29) and malignant mesothelioma (4). Oncolytic viral therapy has also been shown to be synergistic with radiation and chemotherapy (30, 31). Kelly et al. (21) have shown effective killing in malignant mesothelioma cell lines and xenografts using a genetically modified oncolytic vaccinia virus, GLV-1h68, the parent virus of GLV-1h99.

We have shown cytotoxic efficacy in vitro and tumor-specific imaging following GLV-1h99 infection of an orthoptic mesothelioma tumor model. GLV-1h99 expresses the hNET human reporter gene, which was imaged with a clinically approved radiopharmaceutical, [¹²³I]MIBG, and with a positron-emitting analogue, [¹²⁴I]MIBG. This imaging paradigm could be directly translated to human studies, and clinical trials of oncolytic viral therapy would benefit from this noninvasive imaging paradigm.

N. Chen, Y.A. Yu, Q. Zhang, and A.A. Szalay are employees of Genelux Corp., the producer and patent holder of the genetically modified vaccinia viruses, GLV-1h99 and GLV-1h68, used in this study.

We thank Dr. Steven Larson (Memorial Sloan-Kettering Cancer Center) for his help and support.