Search

We recently described an immunocompetent Syrian hamster model for oncolytic adenoviruses (Ads) that permits virus replication in tumor cells as well as some normal tissues. This model allows exploration of interactions between the virus, tumor, normal organs, and host immune system that could not be examined in the immunodeficient or nonpermissive animal models previously used in the oncolytic Ad field. Here we asked whether the immune response to oncolytic Ad enhances or limits antitumor efficacy. We first determined that cyclophosphamide (CP) is a potent immunosuppressive agent in the Syrian hamster and that CP alone had no effect on tumor growth. Importantly, we found that the antitumor efficacy of oncolytic Ads was significantly enhanced in immunosuppressed animals. In animals that received virus therapy plus immunosuppression, significant differences were observed in tumor histology, and in many cases little viable tumor remained. Notably, we also determined that immunosuppression allowed intratumoral virus levels to remain elevated for prolonged periods. Although favorable tumor responses can be achieved in immunocompetent animals, the rate of virus clearance from the tumor may lead to varied antitumor efficacy. Immunosuppression, therefore, allows sustained Ad replication and oncolysis, which leads to substantially improved suppression of tumor growth.

Oncolytic vectors based on human species C adenovirus type 5 (Ad5) are being developed for cancer gene therapy. Studies on wild-type Ad5 in animal models are likely to provide important insights into the behavior of these vectors in patients. The most frequently used permissive immunocompetent animal model for Ad5 infection is the Syrian hamster. Ad5 (and Ad6) replicates in these animals and causes pathology that is similar to that seen with humans. Here, we report findings with a new Syrian hamster strain in which the STAT2 gene was functionally knocked out (KO) by site-specific CRISPR/Cas9-mediated gene targeting. STAT2 is a critical element of the Type I and Type III interferon signal transduction pathways. STAT2 KO hamsters infected intravenously with Ad5 demonstrated an accentuated pathology compared to the wild-type control animals, and the virus load in the organs (liver, lung, kidney) of STAT2 KO animals was 100- to 1000-fold higher than that in wild-type hamsters. We show that the Type I interferon pathway is disrupted in these hamsters, inasmuch as the interferon response genes PKR, OAS, and Mx2 were induced by Ad5 in the liver of wild-type but not STAT2 KO hamsters, revealing a critical role of interferon-stimulated genes in controlling Ad5 infection. Notably, the adaptive immune response to Ad5 is not adversely affected in STAT2 KO hamsters, and surviving hamsters cleared the infection by 7 to 10 days post challenge. In fact, anti-Ad5 neutralizing antibodies were 10-fold higher at 7 days postinfection in the STAT2 KO hamsters than in wild-type hamsters. T cell infiltration into the liver was similar at 3 and 7 days in the STAT2 KO and wild-type hamsters. Treatment of Ad5-infected STAT2 KO hamsters with high dose cyclophosphamide resulted in markedly increased mortality, pathogenesis, and virus replication in the liver (~1011 TCID50/g liver) at 10 days postinfection, likely because the adaptive immune response as well as other aspects of the innate response were abolished. This is the first study to report findings with a genetically modified Syrian hamster infected with human adenovirus. Further, this is the first study to show that the Type I interferon pathway plays a role in inhibiting Ad5 replication in a permissive animal model. This conclusion is in accord with a previous study in which we showed using a custom microarray platform that there is a robust up-regulation of genes involved in the innate immune response (e.g. OAS, PKR, IFN-inducible protein 10) in the liver at 18 h post-intravenous infection of wild-type Syrian hamsters with Ad5 (Ying, B. et al., Virology 485, 305, 2015). Besides providing an insight into adenovirus infection in humans, our results underscore the usefulness of Syrian hamsters as a permissive model for the study of species C human adenovirus and adenovirus-based vectors. In future studies, STAT2 KO hamsters bearing tumors could be used to explore the role of Type I and Type III interferon signaling on the efficacy of oncolytic adenovirus vectors.

Human adenovirus vectors hold great promise as "drugs" to treat cancer. They are relatively easy to grow and manipulate, are very well understood, infect dividing and non-dividing cells from a variety of tissues, and do not cause serious disease in adults. Thousands of papers have been published by many academic and industrial groups on the potential use of adenovirus vectors in cancer gene therapy. These vectors fall into two general categories: replication-defective and replication-competent (oncolytic) 1. In this issue of Molecular Therapy, Bauzon et al. 2 describe a potential step forward in the design of replication-competent adenovirus vectors.

We have developed VRX-007, a fully replication-competent (RC) oncolytic adenovirus (Ad) vector that has a faster replication cycle and is more cytolytic than wild-type Ad5 due to overexpression of the E3 11.6K Adenovirus Death Protein (ADP). VRX-007 is very efficacious against various human cancer cell lines and against tumors in the nude mouse, cotton rat, and Syrian hamster. Tissues of the last two animals are permissive for human Ad replication, and these animals have an intact immune system.

Two biodistribution studies, one in Syrian golden hamsters (Mesocricetus auratus) and one in C57BL/6 mice, were performed to examine the dissemination and pharmacokinetics of VRX-007, a replication-competent (oncolytic) adenovirus (Ad) vector, following intravenous administration. The study in Syrian hamsters allowed us to examine the biodistribution of VRX-007 in a model that is permissive for VRX-007 (and Ad5) replication. In contrast, mice do not support replication of human Ads to an appreciable extent. Controls in the Syrian hamster study included wild-type Ad5 (replication-competent virus), AdCMVpA (a replication-defective vector), and vehicle only. Two quantitative real-time polymerase chain reaction (QPCR) assays were developed, one that specifically and exclusively detects VRX-007 vector genomic DNA (gDNA) and another that specifically detects both Ad5 and AdCMVpA gDNA. These assays were used to detect viral gDNA present in blood and various organs collected from Syrian hamsters (VRX-007 and Ad5/AdCMVpA assays) or C57BL/6 mice (VRX-007 assay only). In addition, the level of infectious virus/vector present in select Syrian hamster organs was titrated using a tissue culture infectious dose-50 (TCID50) assay. Both studies demonstrate that early after administration the virus/vector is distributed widely. In the Syrian hamster study, gDNA from the replication-competent vector/virus VRX-007 and Ad5 was more abundant than that of the replication-defective vector AdCMVpA on Days 2 and 7. On Days 29 and 92, the levels of VRX-007, Ad5, and AdCMVpA Gdna were roughly the same. In addition, there was an increase in VRX-007 and Ad5 copy number from Day 2 to Day 7 in heart, kidney, and adrenal glands (VRX-007 only) that was not seen with AdCMVpA. Infectious vector/virus was recovered from the livers, lungs, and testes of VRX-007- and Ad5-injected hamsters (Days 2 and 7 only), but not from any organs of AdCMVpA-injected hamsters. There was no detectable infectious virus on Day 29 or Day 92. The data in Syrian hamsters contrasts with that obtained in mice, where VRX-007 copy number was generally low on Day 2 (except in the liver), and in many organs decreased to undetectable levels over time. In addition, no infectious VRX-007 was recovered from mouse livers. From these studies we: 1) confirm and extend observations that Syrian hamsters are permissive for Ad replication, which occurred in the liver, lungs, and possibly other organs, 2) confirm that Ad does not replicate to an appreciable extent in C57BL/6 mice, 3) show that the liver is the primary organ of infection in Syrian hamsters and mice when virus/vector is delivered intravenously. The results of these studies will be discussed in light of data obtained from parallel toxicology studies in these two animal species.

We have previously reported on a tumor-specific conditionally replicating adenovirus vector named KD3 that overexpresses the E3-11.6K |[ldquo]|Adenovirus Death Protein|[rdquo]| (ADP). Overexpression of ADP increases vector spread within the tumors, enhancing oncolytic activity. The replication of this vector is restricted to cancer as opposed to normal cells due to mutations in the E1A protein that eliminate binding of E1A to pRB and p300. We have now constructed an improved version of KD3 that is armed with human interferon alpha (hIFN-|[alpha]|). KD3-IFN carries the hIFN-|[alpha]| cDNA in the E3 region downstream of the ADP gene. The vector retains the featureof ADP overexpression and expresses secreted and functionally active hIFN-|[alpha]| strictly at late stages of vector replication. Vectors KD3-DsRed and KD3-SEAP (encoding human secreted placental alkaline phosphatase) carrying reporter genes were also constructed and used to visualize (DsRed) or quantify (SEAP) expression levels in cell culture and in animal models. In cell culture, KD3-IFN had oncolytic activity similar to KD3, exceeding the oncolytic activity of dl1101/1107 (an adenovirus mutant with the same E1A mutation as KD3-IFN but with less ADP production). Therefore, the inclusion of the transgene did not compromise the functional manifestation of ADP overexpression, and high levels of hIFN-|[alpha]| did not adversely affect vector growth and spread.

The accurate determination of adenovirus titers is crucial because as little as a three-fold difference in the amount of virus administered can dramatically affect the outcome of an experiment, especially when high doses are used in animal models. The plaque assay is the standard method for determination of infectious titers, while physical titers (infectious plus non-infectious virus) are typically determined by optical absorbance at 260 nm. Although the plaque assay is a well-established method, there can be considerable variability in titers depending on the cell line used, the passage of the cells, the researcher performing the assay, and the laboratory protocol. For example, in our hands, the plaque assay titers determined in A549 cells are typically several-fold higher than those obtained on HEK 293 cells. In addition, the plaque assay requires as many as four weeks to complete, particularly for vectors which replicate slowly, such as those in which the adp gene is deleted. We have developed a unique quantitative real-time PCR assay for rapid adenovirus quantification. Our method includes a DNase treatment prior to capsid disruption to eliminate non-encapsidated DNA. Also, the real-time PCR primers are located in the E4 region at the right end of the genome, distant from the packaging signals at the left end of the genome. With this design, only full-length encapsidated adenovirus DNA should be measured. Our assay has been used to titer a wide variety of adenovirus stocks, including viruses with various point mutations, insertions, and deletions. In addition, we have assayed both cesium chloride-banded preparations and crude lysates, as well as stocks which have been stored in different buffers, at different temperatures, and for different lengths of time (ranging from recently prepared stocks to those prepared 20 years ago). Regardless of these differences, we have found this assay to produce rapid and consistent titers. Interestingly, the titers determined by real-time PCR correlated with infectious titers determined by the plaque assay and the endpoint dilution assay. This method can be utilized to verify the previously determined titer of virus stocks as well as to rapidly predict the infectious titer of new virus stocks.

Replication-competent adenovirus vectors for cancer gene therapy are typically evaluated in immunodeficient nude or SCID mice bearing human xenograft tumors. However, the lack of an intact immune system in these animals lessens the relevance of these models to a clinical scenario. We have developed a novel Syrian hamster animal model for the study of oncolytic adenovirus vectors. Initially, we observed the expression of early and late adenovirus proteins in infected hamster cells at levels comparable to that seen in human A549 cells. Single step growth curves established that human adenovirus can replicate within and produce high virus yields in Syrian hamster cells. The yield per cell (burst size) was approximately 500 plaque forming units per cell, only one order of magnitude less than that obtained with A549 cells, one of the most permissive human cell lines known. We then confirmed that human adenovirus type 5 (Ad5) can replicate in the lungs of Syrian hamsters following endotracheal instillation. Three different Syrian hamster cancer cell lines were examined as subcutaneous tumor models in Syrian hamsters, and all three of these were found to form tumors following subcutaneous inoculation. Growth of these tumors was markedly suppressed following intratumoral injection with VRX-007, a replication-competent adenovirus vector developed in our laboratory, which overexpresses ADP (E3-11.6K). Furthermore, we found that VRX-007 replicates within tumors. Subcutaneous tumors formed by one of these hamster cell lines metastasized to the lungs and other organs. Intratumoral injection of VRX-007 suppressed the degree of metastasis. Histopathological analysis revealed areas of necrosis and degenerating tumor cells within tumors treated with VRX-007. These necrotic areas were located within viable areas of tumors, not in the central necrotic area, which most tumors contained. The necrotic regions were characterized as having a central core of debris which was primarily acellular, immediately surrounded by neutrophil infiltration and more peripherally surrounded by macrophages and fibroblasts. No such areas were observed within vehicle treated tumors. Our findings that Syrian hamster cells support productive adenovirus replication in vitro and that vectors developed in our laboratory show efficacy and replication in vivo demonstrate that the Syrian hamster can serve as a useful animal model to evaluate the safety and efficacy of oncolytic adenovirus vectors.

Top of pageAbstract The need for new anti-cancer treatments is evident from the fact that malignant neoplasms are the second leading cause of death in the United States and third leading cause of death world-wide. Oncolytic adenovirus (Ad) vectors, which kill cancer cells by harnessing the cytolytic nature of the virus life cycle, are a new class of anti-cancer agents that have been demonstrated to be safe in humans but that have thus far yielded disappointing clinical efficacy results, despite demonstrating efficacy in pre-clinical models. We have constructed several new oncolytic Ad vectors that contain features designed to enhance vector efficacy while maintaining safety. All of our vectors over-express ADP, the viral protein required late in infection for efficient cell lysis and release of progeny virions at the culmination of the infectious cycle. Because our vectors over-express ADP, infected cells are lysed rapidly and progeny virions are released efficiently, which translates into improved cell-to-cell dissemination of the vector. Our vectors are also engineered to express the cancer therapeutic protein tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL), a member of the TNF family of cytokines that shows specificity for neoplastic cells. TRAIL expression and secretion provides a means of enhancing vector efficacy by providing a way to kill uninfected tumor cells. The TRAIL cDNA was inserted into our vectors such that it would be expressed similarly to ADP, that is, abundant expression late in the infectious cycle. We have shown in cultured cancer cells that our TRAIL-expressing vectors synthesize a large quantity of TRAIL late in the infection and are highly cytolytic to both infected and uninfected cells. TRAIL expression in infected cells leads to apoptosis in nearby uninfected cells and ADP expression results in efficient cytolysis of infected cells. As a safety feature, all of the vectors contain a deletion that eliminates expression of nearly all of the E3 proteins; these proteins function to protect infected cells from the immune system. As an additional safety feature, some vectors also contain deletions within the E1A gene that result in selective replication of the vector in cancer cells because the mutant E1A proteins are no longer able to bind to pRB family members or p300/CBP. The vectors were shown to reduce the growth of Hep3B liver cell tumor xenografts grown in nude mice.

We have developed the Syrian hamster as a permissive immunocompetent model to evaluate the safety and efficacy of replication-competent adenovirus (Ad) vectors for cancer gene therapy. One of our vectors is named VRX-007; VRX-007 overexpresses ADP, an Ad protein that mediates virus release and cell-to-cell spread. In order to determine the maximum tolerated dose (MTD) of VRX-007 in hamsters, we administered a single injection of VRX-007 into the jugular vein. Hamsters received 1.9 |[times]| 108 PFU, 1.9 |[times]| 109 PFU, or 1.9 |[times]| 1010 PFU. No morbidity and no gross or histopathologic abnormalities were observed. In a second study, we administered VRX-007 via the jugular vein on three consecutive days. The cumulative doses were 1.9 |[times]| 1010 PFU for the low dose group, 5.7 |[times]| 1010 PFU for the intermediate dose group, and 1.1 |[times]| 1011 PFU for the high dose group. Half of each group was sacrificed on day 4, and the remainder of the animals were sacrificed on day 14. Hamsters in the low dose group did not display signs of morbidity, and no significant findings were observed grossly at either sacrifice date. In the intermediate group, one animal died on day 7; at necropsy, there was mild to moderate gross liver pathology at day 4 but only mild liver pathology at day 14. All animals in the high dose group were either found dead or became moribund and were sacrificed by three days postinjection. Liver enzymes (ALT and AST) in the serum were measured in order to evaluate the degree of hepatotoxicity. These enzymes were normal in the low dose group. For the intermediate group, the ALT and AST levels generally correlated with the degree of gross liver pathology observed at necropsy, with mild to moderate elevation at 4 days that resolved at 14 days. For the low and intermediate groups, elevation of alkaline phosphatase levels as well as a decrease in total protein and albumin levels were observed at 4 but not 14 days. For both groups, hematological analysis revealed a dose-dependent decrease in platelet counts on day four, although platelet counts were normal again at 14 days. Based on these data, we estimate the MTD of VRX-007 when adminstered intravenously on three consecutive days into Syrian hamsters to be the low dose used in this study (1.9 |[times]| 1010 PFU total or 1.7 |[times]| 1011 VP total). This corresponds to a dose of 2.1 |[times]| 1011 PFU/kg (1.9 |[times]| 1012 VP/kg).

Adenoviruses as oncolytic vectors have been investigated for almost a decade, and show great promise for the treatment of cancer. These vectors kill tumor cells as a result of the lytic replication cycle of the virus. A significant impediment to research with replicating adenovirus vectors is the lack of suitable animal models. As adenoviruses are very species-specific, researchers studying vectors based on human Adenovirus 5 were forced to use the highly artificial immunodeficient mouse-human tumor xenograft model, where the viruses could replicate within the human tumor cells. In this system, the replication of the adenovirus vector in normal tissues and the effect of the host immune system on the anti-tumor activity of the vector cannot be studied. We have developed an immunocompetent, semi-permissive animal model to study oncolytic adenovirus vectors. In this model, we induced subcutaneous tumors in immunocompetent cotton rats by injecting the animals with a suspension of LCRT cotton rat sarcoma cells. The cotton rat was shown by others to be semi-permissive to human adenovirus infection, and to exhibit pathology similar to that observed in humans with respiratory adenoviral infection. The LCRT cell line is derived from a spontaneously arisen mammary tumor. These cells support human adenovirus replication in vitro, as evidenced by expression of adenoviral late proteins and the production of progeny virus that can initiate a new infectious cycle. As a result, LCRT cells are destroyed upon infection even with low multiplicities of virus. When injected subcutaneously into cotton rats, LCRT cells form rapidly growing subcutaneous tumors that metastasize to the lung and to various lymph nodes. They show the characteristics of fibrosarcoma tumors, and usually contain a large necrotic center and rapidly dividing cells at the periphery. We injected pre-existing or nascent subcutaneous LCRT tumors with VRX-007, an oncolytic adenovirus overexpressing ADP. ADP is an adenoviral protein expressed at the culmination of infection that facilitates cell lysis and virus egress, thus speeding up the cell-to-cell dissemination of the vector. Injection of intact VRX-007 consistently caused a significant delay in tumor growth, while tumors injected with UV inactivated vector grew at a rate similar to that of vehicle injected tumors. We have also established the intravenous maximum tolerated dose of VRX-007 in cotton rats. Based upon these results, we propose subcutaneous LCRT tumor bearing cotton rats as a novel, immunocompetent, semi-permissive animal model for the testing of oncolytic adenovirus vectors.