Subcutaneous Administration of a Replication-Competent Adenovirus Expressing HSV-tk to Cotton Rats: Dissemination, Persistence, Shedding, and Pathogenicity

Oncolytic (replication-competent) adenoviruses as anticancer agents

IMPORTANCE OF THE FIELD

Whilst therapies for neoplasies have advanced tremendously in the last few decades, there is still a need for new anti-cancer treatments. One option is genetically-engineered oncolytic adenovirus (Ad) 'vectors'. These kill cancer cells via the viral replication cycle, and amplify the anti-tumor effect by producing progeny virions able to infect neighboring tumor cells.

AREAS COVERED IN THIS REVIEW

We provide a description of basic Ad biology and summarize the literature for oncolytic Ads from 1996 to the present.

WHAT THE READER WILL GAIN

An overall view of oncolytic Ads, the merits and drawbacks of the various features of these vectors, and obstacles to further development and future directions for research.

TAKE HOME MESSAGE

Ads are attractive for gene therapy because they are relatively innocuous, easy to produce in large quantities, genetically stable, and easy to manipulate. A variety of have been constructed and tested, in pre-clinical and clinical experiments. Oncolytic Ads proved to be remarkably safe; no dose-limiting toxicity was observed in any clinical trial, and the maximum tolerated dose was not reached. At present, the major challenge for researchers is to increase the efficacy of the vectors, and to incorporate oncolytic virotherapy into existing treatment protocols.

INGN 007, an oncolytic adenovirus vector, replicates in Syrian hamsters but not mice: comparison of biodistribution studies

Preclinical biodistribution studies with INGN 007, an oncolytic adenovirus (Ad) vector, supporting an early stage clinical trial were conducted in Syrian hamsters, which are permissive for Ad replication, and mice, which are a standard model for assessing toxicity and biodistribution of replication-defective (RD) Ad vectors. Vector dissemination and pharmacokinetics following intravenous administration were examined by real-time PCR in nine tissues and blood at five time points spanning 1 year. Select organs were also examined for the presence of infectious vector/virus. INGN 007 (VRX-007), wild-type Ad5 and AdCMVpA (an RD vector) were compared in the hamster model, whereas only INGN 007 was examined in mice. DNA of all vectors was widely disseminated early after injection, but decayed rapidly in most organs. In the hamster model, DNA of INGN 007 and Ad5 was more abundant than that of the RD vector AdCMVpA at early times after injection, but similar levels were seen later. An increased level of INGN 007 and Ad5 DNA but not AdCMVpA DNA in certain organs early after injection, and the presence of infectious INGN 007 and Ad5 in lung and liver samples at early times after injection, strongly suggests that replication of INGN 007 and Ad5 occurred in several Syrian hamster organs. There was no evidence of INGN 007 replication in mice. In addition to providing important information about INGN 007, the results underscore the utility of the Syrian hamster as a permissive immunocompetent model for Ad5 pathogenesis and oncolytic Ad vectors.

Human adenovirus replicates in immunocompetent models of pancreatic cancer in Syrian hamsters

The preclinical evaluation of toxicity and antitumor effect of conditionally replicative (oncolytic) adenoviruses is hampered by the inability of human adenoviruses to replicate efficiently in murine cells. The Syrian golden hamster (Mesocricetus auratus) has been suggested as a permissive animal for adenoviral replication, and cancer cell lines derived from various hamster tumors are available. We provide evidence that wild-type adenovirus type 5 is able to infect and replicate in the pancreatic cancer cell lines HaP-T1 and H2T both in vitro and in vivo. Determination of cytopathic effect, viral spread, progeny production, and the expression of late viral proteins indicates that the complete viral cycle of adenovirus takes place, albeit less efficiently than in highly permissive human cancer cell lines A549 and HuH7. Intrahepatic inoculation of HaP-T1 and H2T cells gave rise to tumors in the liver of hamsters that resemble metastases of pancreatic cancer. The growth of HaP-T1-induced nodules was faster compared with those derived from H2T, but both caused progressive liver infiltration and peritoneal dissemination. When adenovirus was inoculated in these lesions, productive replication took place and newly formed infective virions could be recovered 4 days after administration. In conclusion, the Syrian hamster models described here offer the opportunity to evaluate the effect of oncolytic adenoviruses in an immunocompetent animal and may be a valuable tool in the preclinical evaluation of these agents.

Immunocompetent syngeneic cotton rat tumor models for the assessment of replication-competent oncolytic adenovirus

Oncolytic adenoviruses as a treatment for cancer have demonstrated limited clinical activity. Contributing to this may be the relevance of preclinical animal models used to study these agents. Syngeneic mouse tumor models are generally non-permissive for adenoviral replication, whereas human tumor xenograft models exhibit attenuated immune responses to the vector. The cotton rat (Sigmodon hispidus) is susceptible to human adenovirus infection, permissive for viral replication and exhibits similar inflammatory pathology to humans with adenovirus replicating in the lungs, respiratory passages and cornea. We evaluated three transplantable tumorigenic cotton rat cell lines, CCRT, LCRT and VCRT as models for the study of oncolytic adenoviruses. All three cells lines were readily infected with adenovirus type-5-based vectors and exhibited high levels of transgene expression. The cell lines supported viral replication demonstrated by the induction of cytopathogenic effect (CPE) in tissue culture, increase in virus particle numbers and assembly of virions seen on transmission electron microscopy. In vivo, LCRT and VCRT tumors demonstrated delayed growth after injection with replicating adenovirus. No in vivo antitumor activity was seen in CCRT tumors despite in vitro oncolysis. Adenovirus was also rapidly cleared from the CCRT tumors compared to LCRT and VCRT tumors. The effect observed with the different cotton rat tumor cell lines mimics the variable results of human clinical trials highlighting the potential relevance of this model for assessing the activity and toxicity of oncolytic adenoviruses.

Effect of adenoviral mediated overexpression of fibromodulin on human dermal fibroblasts and scar formation in full-thickness incisional wounds

Fibromodulin, a member of the small leucine-rich proteoglycan family, has been recently suggested as a biologically significant mediator of fetal scarless repair. To assess the role of fibromodulin in the tissue remodeling, we constructed an adenoviral vector expressing human fibromodulin cDNA. We evaluated the effect of adenovirus-mediated overexpression of fibromodulin in vitro on transforming growth factors and metalloproteinases in fibroblasts and in vivo on full-thickness incisional wounds in a rabbit model. In vitro, we found that Ad-Fibromodulin induced a decrease of expression of TGF-beta(1) and TGF-beta(2) precursor proteins, but an increase in expression of TGF-beta(3) precursor protein and TGF-beta type II receptor. In addition, fibromodulin overexpression resulted in decreased MMP-1 and MMP-3 protein secretion but increased MMP-2, TIMP-1, and TIMP-2 secretion, whereas MMP-9 and MMP-13 were not influenced by fibromodulin overexpression. In vivo evaluation by histopathology and tensile strength demonstrated that Ad-Fibromodulin administration could ameliorate wound healing in incisional wounds. In conclusion, although the mechanism of scar formation in adult wounds remains incompletely understood, we found that fibromodulin overexpression improves wound healing in vivo, suggesting that fibromodulin may be a key mediator in reduced scarring.

Preclinical safety and biodistribution of adenovirus-based cancer vaccines after intradermal delivery

The recombinant adenoviral (Ad) vector is being considered as a cancer vaccine platform because it efficiently induces immune responses to tumor antigens by intradermal immunization. The aims of this study were to evaluate the potential toxicities and biodistribution after a single dose or six weekly intradermal doses of Ad2/gp100v2 and Ad2/MART-1v2, which encode tumor-associated antigens gp100 and MelanA/MART-1, respectively. The only dose-related toxicities associated with intradermal administration of these Ad vectors were inflammatory cell infiltrates in the draining lymph nodes and injection sites that persisted 83 days after administration. The biodistribution of Ad DNA as detected by real-time polymerase chain reaction was largely confined to the injection sites and draining lymph nodes of mice treated with either a single dose or multiple doses of Ad vector and in the spleens of mice treated with multiple doses of Ad vector. Adenoviral DNA was transiently detected in the bone marrow, lung, or blood of only one animal for each site and was below the limit of assay quantification (<10 copies/microg DNA). The vector persisted in the skin and lymph nodes as long as 92 days after the last dose. We conclude that Ad vectors delivered by intradermal administration provide a safe, genetic vaccine delivery platform that induces desirable immune responses at the immunization sites and the lymph nodes that, ultimately, result in immune responses specific to the tumor antigens.

Infectivity enhancement for adenoviral transduction of canine osteosarcoma cells

The full realization of conditionally replicative adenoviruses (CRAds) for cancer therapy has been hampered by the limited knowledge of CRAd function in vivo and particularly in an immunocompetent host. To address this issue, we previously proposed a canine adenovirus type 2 (CAV2)-based CRAd for clinical evaluation in canine patients with osteosarcoma (OS). In this study, we evaluated infectivity-enhancement strategies to establish the foundation for designing a potent CAV2 CRAd with effective transduction capacity in dog osteosarcoma cells. The results indicate that the native CAV2 fiber-knob can mediate increased binding, and consequently gene transfer, in both canine osteosarcoma immortalized and primary cell lines relative to previously reported Ad5 infectivity-enhancement strategies. Gene delivery was further enhanced by incorporating a polylysine polypeptide onto the carboxy terminus of the CAV2 knob. This vector demonstrated improved gene delivery in osteosarcoma xenograft tumors. These data provide the rationale for generation of infectivity-enhanced syngeneic CAV2 CRAds for clinical evaluation in a dog osteosarcoma model.

Cotton Rat Tumor Model for the Evaluation of Oncolytic Adenoviruses

Oncolytic human adenovirus (Ad) vectors exert their antitumor effect by replicating in and lysing tumor cells. These vectors are commonly evaluated in immunodeficient mice bearing human tumor xenografts. However, this model suffers because the mice are immunodeficient and are not permissive for human Ads. We have developed a cotton rat model to test the selectivity, immunogenicity, and efficacy of oncolytic Ad vectors. The cotton rat is a rodent species that is semipermissive for human Ads. We show that the cotton cancer rat cell line LCRT supports the replication of human Ad in tissue culture and that the cells are destroyed on virus replication. When injected subcutaneously, LCRT cells formed tumors in immunocompetent cotton rats, and the growth of these tumors was delayed by the injection of an oncolytic Ad vector. Replication of the Ad vector in the tumor was demonstrated by sampling tumor tissue and isolating infectious virus particles at various times after intratumoral injection of the virus. We propose that the cotton rat can be used as an animal model to evaluate oncolytic Ad vectors.

Regulatory Aspects in the Development of Gene Therapies

Preclinical therapeutics development research is directed toward fulfilling two overlapping sets of goals. A set of scientific goals includes defining the best molecule or biologic construct for the task at hand, and proving the case for its development. The second set of goals addresses regulatory requirements necessary to introduce the agent into human subjects. In the case of “small molecule” drugs, in most cases the identity of the molecule and appropriate safety studies are straightforward. In contrast, the development of biologic agents, including gene therapies discussed here, presents distinct challenges. The nature of the “drug” may be an organism subject to mutation or selection of variants through recombination. Its properties may vary depending on the scale and method of its preparation, purification, and storage. How to test adequately for its safety prior to first introduction in humans may not be straightforward owing to intrinsic differences in response to the agent expected in humans as compared to animals.

Anex vivo human model system to evaluate specificity of replicating and non-replicating gene therapy agents

BACKGROUND

Inefficiency, aspecificity and toxicity of gene transfer vectors hamper gene therapy from showing its full potential. On this basis significant research currently focuses on developing vectors with improved infection and/or expression profiles. Screening assays with validity to the clinical context to determine improved characteristics of such agents are not readily available since this requires a close relationship to the human situation. We present a clinically relevant tissue slice technology to preclinically test improved vector characteristics.

METHODS

Slices were prepared from rat, mouse and human liver samples and from tumor tissue. Specificity of gene expression and replication was determined by infecting target and non-target tissue slices with transcriptionally retargeted adenoviruses and oncolytic viruses.

RESULTS

Using rat liver slices, we demonstrate efficient knob-mediated adenoviral infectivity. A favorable tumor-on/liver-off profile, resembling in vitro and mouse in vivo data, was shown for a tumor-specific transcriptionally retargeted adenovirus by infecting slices prepared from tumor or liver tissue. Similar liver-off data were found for mouse, rat and human samples (over 3-log lower activity of the tumor-specific promoter compared to cytomegalovirus (CMV)). More importantly, we show that this technology when applied to human livers is a powerful tool to determine aspecific replication of oncolytic viruses in liver tissue. A 2- to 6-log reduction in viral replication was observed for a tumor-specific oncolytic virus compared to the wild-type adenovirus.

CONCLUSIONS

The precision-cut tissue slice technology is a powerful method to test specificity and efficiency of gene transfer as well as of viral replication using human tissue.

Vaccination by genetically modified dendritic cells expressing a truncated neu oncogene prevents development of breast cancer in transgenic mice

Dendritic cells (DCs) are powerful antigen-presenting cells that process antigens and present peptide epitopes in the context of the major histocompatibility complex molecules to generate immune responses. DCs are being studied as potential anticancer vaccines because of their ability to present antigens to naive T cells and to stimulate the expansion of antigen-specific T-cell populations. We investigated an antitumor vaccination using DCs modified by transfer of a nonsignaling neu oncogene, a homologue of human HER-2/neu, in a transgenic model of breast cancer. BALB-neuT mice develop breast cancers as a consequence of mammary gland-specific expression of an activated neu oncogene. We vaccinated BALB-neuT mice with bone marrow-derived DCs transduced with Ad.Neu, a recombinant adenovirus expressing a truncated neu oncoprotein. The vaccine stimulated the production of specific anti-neu antibodies, enhanced interferon-gamma expression by T cells, and prevented or delayed the onset of mammary carcinomas in the mice. Over 65% of vaccinated mice remained tumor free at 28 weeks of age, whereas all of the mice in the control groups developed tumors. When challenged with a neu-expressing breast cancer cell line, vaccinated tumor-free animals had delayed tumor growth compared with controls. The antitumor effect of the vaccine was specific for expression of neu. Studies showed that CD4+ T cells were required in order to generate antitumor immunity. Importantly, the effectiveness of the vaccine was not diminished by preexisting immunity to adenovirus, whereas the protection afforded by vaccination that used direct injection of Ad.Neu was markedly reduced in mice with anti-adenovirus antibody titers. DCs modified by recombinant adenoviruses expressing tumor-associated antigens may provide an effective antitumor vaccination strategy.

Gene transfer vector biodistribution: pivotal safety studies in clinical gene therapy development

Techniques allowing for gene transfer vectors biodistribution investigation, in the frame of preclinical gene therapy development, are exposed. Emphasis is given on validation and test performance assessment. In the second part, specific gene vector distribution properties are reviewed (adenovirus, AAV, plasmid, retroviruses, herpes-derived vectors, germline transmission risks). The rationale for biodistribution by quantitative PCR, animal study and result interpretation is discussed. The importance and pivotal role of biodistribution study in gene transfer medicine development is shown through the determination of target organs for toxicity, germline transmission assessment and determination of risks of shedding and spreading of vectors in the gene transfer recipient and the environment.

Combination Therapy with Conditionally Replicating Adenovirus and Replication Defective Adenovirus

Low gene transfer rate is the most substantial hurdle in the practical application of gene therapy. One strategy to improve transfer efficiency is the use of a conditionally replicating adenovirus (CRAD) that can selectively replicate in tumor cells. We hypothesized that conventional E1-deleted adenoviruses (ad) can become replication-competent when cotransduced with a CRAD to selectively supply E1 in trans in tumors. The resulting selective production of large numbers of the E1-deleted ad within the tumor mass will increase the transduction efficiency. We used a CRAD (Delta24RGD) that produces a mutant E1 without the ability to bind retinoblastoma but retaining viral replication competence in cancer cells with a defective pRb/p16. Ad-lacZ, adenovirus-luciferase (ad-luc), and adenovirus insulin-like growth factor-1R/dominant-negative (ad-IGF-1R/dn; 482, 950) are E1-deleted replication-defective adenoviruses. The combination of CRAD and ad-lacZ increased the transduction efficiency of lacZ to 100% from 15% observed with ad-lacZ alone. Transfer of media of CRAD and ad-lacZ cotransduced cells induced the transfer of lacZ (media transferable bystander effect). Combination of CRAD and ad-IGF-1R/dn increased the production of truncated IGF-1R or soluble IGF-1R > 10 times compared with transduction with ad-IGF-1R/dn alone. Combined intratumoral injection of CRAD and ad-luc increased the luciferase expression about 70 times compared with ad-luc alone without substantial systemic spread. Combined intratumoral injection of CRAD and ad-IGF-1R/482 induced stronger growth suppression of established lung cancer xenografts than single injections. The combination of CRAD and E1-deleted ad induced tumor-specific replication of CRAD and E1-deleted ad and increased the transduction rate and therapeutic efficacy of these viruses in model tumors.

A Conditionally Replicating Adenovirus for Nasopharyngeal Carcinoma Gene Therapy

Successful attainment of tumor-specific gene expression was achieved in nasopharyngeal carcinoma (NPC) by exploiting the exclusive presence of the Epstein-Barr virus (EBV) genome in the cancer cells. In the current study, we have utilized an EBV-dependent transcriptional targeting strategy to construct a novel conditionally replicating adenovirus, adv.oriP.E1A. After treatment with adv.oriP.E1A, we observed extensive cell death in the EBV-positive NPC cell line C666-1. In contrast, no cytotoxicity was observed in a panel of other human EBV-negative cell lines, including fibroblasts from the nasopharynx. In vitro adenoviral replication was confirmed by the time-dependent increase in the expression of adenoviral capsid fiber protein and adenoviral DNA after C666-1 cells were infected with adv.oriP.E1A. Tumor formation was inhibited for more than 100 days after ex vivo infection of C666-1 cells with adv.oriP.E1A. Combination of local tumor radiation and adv.oriP.E1A caused complete disappearance of established tumors for at least 2 weeks in two distinct EBV-positive NPC xenograft models. Safety of this treatment was determined through the systemic delivery of adv.oriP.E1A in vivo, whereby minimal temporary perturbation of liver function was observed. We have successfully established a conditionally replicating adenovirus for EBV-positive NPC, which is both safe and efficacious, indicating a strategy that may be therapeutically applicable.

A novel tetracycline-controlled transactivator-transrepressor system enables external control of oncolytic adenovirus replication

The use of restricted replication-competent adenoviruses (RRCAs) inducing tumor cell-specific lysis is a promising approach in cancer gene therapy. However, the use of RRCAs in humans carries considerable risk, since after injection into the patient, further regulation or inhibition of virus replication from the outside is impossible. Therefore, we have developed a novel system allowing external pharmacological control of RRCA replication. We show here that a tumor-selective E1B-deleted RRCA can be tightly regulated by use of doxycycline (dox)-controlled adenoviral E1A gene expression, which in turn determines vector replication. RRCA replication is switched on by addition and switched off by withdrawal of dox. The system results in efficient tumor cell killing after induction by dox, whereas cells are unaffected by the uninduced system. It was also employed for efficient external control of transgene expression from cotransfected replication-deficient adenovectors. Furthermore, the use of a liver cell-specific human alpha1-antitrypsin (hAAT)-promoter driving a tetracycline-controlled transcriptional silencer allowed specific protection of cells with hAAT-promoter activity in the absence of dox in vitro and in vivo, delineating a new principle of 'tissue protective' gene therapy. The concept of external control of RRCAs may help to improve the safety of cancer gene therapy.