Fighting Fire with Fire:
Cancer-Killing Viruses

Oncolytic herpes simplex viruses are genetically engineered to replicate and spread highly selectively in tumor cells leading to their destruction. This novel type of therapy is emerging as an effective and powerful therapeutic approach for cancer.
Although treatment modalities for cancer such as chemotherapy and radiotherapy have improved over the past decades, most malignancies remain incurable. Furthermore, these approaches are often characterized by a narrow therapeutic index between cancerous and normal cells. As a consequence, major efforts in oncology are dedicated to treating their adverse events. Novel therapeutics are clearly needed that have not only greater potency and greater selectivity for malignant tissue, but also have novel mechanisms of action that will not lead to cross-resistance with existing therapies. Oncolytic viruses have the potential to fulfill these criteria. Viral therapy of cancer is based on direct cell killing due to the lytic cycle of the virus. In addition, an immune response to tumor antigens may be initiated in patients leading to eradication of even widely spread tumor cells. Initial attempts to utilize wild-type viruses for cancer treatment in the beginning and middle part of the 20th century resulted in side effects and large variability in anti-tumor response. However, significant advances in our understanding of the molecular basis of cancer and the availability of technologies to genetically engineer viruses have led to the development of replication-selective viruses that can be specifically targeted to the tumor while sparing surrounding normal tissue.

Oncolytic viruses
There are striking similarities between tumor cells and cells infected by viruses. In both cases signal transduction pathways governing the cell cycle are impacted resulting in uncontrolled cell cycle progression. It is now evident that tumorigenesis is a multistep process in which mutations in tumor suppressors and oncogenes accumulate over time creating an imbalance between growth and growth control. The expression or function of the affected genes is altered and allows bypassing important cell cycle checkpoints, e.g. enzymes involved in cellular DNA replication are expressed at high levels. Interestingly, viruses have evolved gene products that mimick cell cycle activation in quiescent cells by physiological signals or tumor cells. These virus induced processes trigger the expression of cellular enzymes required for replication of the viral genome, which finally results in complete lysis and death of the infected host cell. Alternatively, viral proteins counteract with cellular defense mechanisms. While many wild-type viruses are able to infect, replicate in, and kill a large variety of normal cells in the body, recent advances in genetic engineering have provided opportunities to generate attenuated versions of wild-type viruses. These attenuated virus mutants carry specific deletions in their genomes restricting their replicative potential to tumor cells that already provide all the enzymes required for viral replication at the time of infection (Fig. 1). Their ability to replicate in tumor tissue allows for amplification of the input dose at the tumor site, while their lack of replication in normal tissues results in efficient clearance associated with little or no toxicity. This feature makes oncolytic viruses very unique, and can in theory increase the therapeutic index of such therapy dramatically over standard therapeutic approaches. In addition to direct cell killing, viruses can induce inflammatory cytokine responses and T-cell mediated immunity.

Oncolytic herpes simplex viruses
One of the viruses being actively pursued for cancer therapy is herpes simplex virus (HSV). HSV is a human neurotropic virus of the a-herpesvirus subfamily and consists of two serotypes, type 1 (HSV-1) and type 2 (HSV-2). It is ubiquitous in the human population and only very rarley causes serious diseases. Generally the symptoms of HSV infections are mild, such as the development of cold sores. In order to maximize the therapeutic effectiveness and safety of oncolytic HSV, the virus has been genetically modified, so that it elicits minimal toxicity in normal cells while retaining its full capability to replicate in malignant cells. HSV is suited for cancer therapy due to several features of its basic biology: It infects most tumor cell types in vivo when administered via multiple routes, and it has a well studied life cycle with identified functions of the majority of its genes. Furthermore, only a relatively low multiplicity of infection is required for effective cell killing, which usually occurs within only a few hours. Finally, anti-viral drugs to terminate viral replication are available and already approved for use in man.

Treatment of malignant brain tumors
G207 is a multi-gene mutant HSV vector that was designed for the treatment of malignant brain tumors. It carries deletions in both copies of the neurovirulence gene g34.5, and has a disruption of the viral UL39 gene locus. The g34.5 gene is a major determinant of HSV pathogenicity. Virus infection usually causes activation of a signalling cascade in the host cell, which results in shut-off of the cellular protein synthesis leading to abortive virus infection in normal cells, a mechanism that is considered as anti-viral protection. HSV has evolved the g34.5 gene product to counteract this cellular virus defense mechanism allowing efficient viral replication in the infected host cell. In contrast, viruses deficient in the g34.5 gene will not be capable to replicate in normal cells. Interestingly, in many tumor cells, in which the Ras oncogene is mutated – a very frequent event in a variety of malignancies – the signalling pathways normally blocking uncontrolled viral replication are disrupted. Although the exact mechanism of functional complementation through mutant Ras is not fully understood, the g34.5 gene product becomes dispensible for efficient replication in those tumors. The second disrupted gene in G207, UL39, encodes for the viral ribonucleotide reductase, an enzyme that is also essential for viral replication. Host gene expression in dividing cells is presumably able to complement the missing enzyme activity, rendering G207 conditionally replicating in dividing cells. These multiple and large gene disruptions make reversion to wild-type highly unlikely.
Malignant gliomas are the most common primary malignant brain tumors. Despite aggressive therapies including surgery, chemotherapy and radiotherapy, these tumors are fatal, and prognosis continues to be grim with a median survival of approximately six to nine months for recurrent tumors. Screening of human glioma cell lines for G207 susceptibility revealed that G207 can completely lyse these cells shortly after infection. In contrast, G207 at the same dose caused no cytopathic effect in primary cultures of astrocytes or neurons. Furthermore, the anti-tumor efficacy and safety has been demonstrated in a wide variety of animal tumor models, including subcutaneous and intracranial pre-established tumors. In light of the encouraging preclinical data concerning anti-tumor efficacy and safety of G207, a phase I clinical trial of G207 in recurrent malignant glioma patients has been performed, recently. A total of 21 patients was treated with escalated doses of G207. In summary, treatment with G207 was well tolerated and no serious adverse events related to G207 therapy were observed. Although the primary objective of the study was to evaluate safety, it is interesting to note that two glioma patients remain alive more than four years post G207 administration.

Outlook
The use of replication-selective herpes simplex viruses for the therapy of brain tumors has gone from an interesting concept to clinical evaluation in less than 10 years. The therapeutic index, and ultimately the clinical outcome, will depend on a complex balance between host and viral factors. A number of issues will need to be addressed in more detail, such as the role of the immune system against viral and tumor antigens. There is increasing evidence that a mechanism of synergy exists between therapy with oncolytic HSV and radiotherapy. This is of particular interest, as both treatments are based on different mechansims of cell killing, and cross-resistance is therefore unlikely to occur.
In summary, the promising results of recent clinical trials have fueled the hope that using oncolytic viruses may lead to an effective cancer therapy in the near future.

References
[1] T. Mineta, S.D. Rabkin, T. Yazaki, W.D. Hunter, R.L. Martuza, Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas, Nat Med. 1 (1995) 938-43.
[2] S. Varghese, S.D. Rabkin, Oncolytic herpes simplex virus vectors for cancer virotherapy, Cancer Gene. Ther. 9 (2002) 967-78.
[3] F. Farassati F, A.D. Yang, P.W. Lee, Oncogenes in Ras signalling pathway dictate host-cell permissiveness to herpes simplex virus 1, Nat. Cell. Biol. 3 (2001) 745-50.
[4] C. Biederer, S. Ries, C.H. Brandts, F. McCormick, Replication-selective viruses for cancer therapy, J. Mol. Med. 80 (2002) 163-175.