Oncolytic Viruses: Cytolytic agents, Replicating Immunotherapeutics Or both?

Keywords: :

• immune checkpoint inhibitor
• immunotherapy
• oncolytic virus

Cancer cells have distinct hallmarks that promote growth and metastasis, but at the same time enhance their susceptibility to viral infection [1]. As a result, oncolytic viruses (OVs) possess an intrinsic ability or can be engineered to selectively target, replicate in and kill cancer cells, while leaving healthy cells unharmed. Oncolysis allows for amplification of the initial viral load due to viral replication and subsequent spread of infectious particles to neighboring cells upon cell lysis. In addition to this direct cytotoxic effect, OVs can be armed with therapeutic transgenes to take advantage of antigen release and presentation, as well as to potentiate immune responses and infiltration within the tumor microenvironment (TME) [2,3].

Selection of the right viral backbone

A significant challenge in OV design involves maximizing OV-mediated cancer cell death by increasing viral potency, while sparing nonmalignant cells. Though OVs are naturally selective to some extent toward cancer cells, additional viral attenuation is often necessary to engineer OVs that are both safe and effective. Common methods of attenuation involve the deletion of viral genes that are essential for infection of healthy cells [4] or genes that antagonize the host antiviral immune response [5]. Selecting an OV backbone with maximal oncolytic potency is not necessarily the optimal or only way of improving tumor clearance. Mounting evidence suggests that OV-mediated oncolysis, although an important first step, must be complemented by stimulation of a host antitumor immune response in order to maximize therapeutic benefit. As one example, targeted T-cell disruption before intratumor injection of oncolytic adenovirus (AdV) eliminated OV-mediated anti-tumor activity, even though OV replication and persistence was significantly increased [6].

Following viral mediated cell lysis, assorted tumor antigens, cytokines and chemokines are released into the extracellular space [2,3]. Certain OVs, such as AdV and coxsackievirus B3, initiate the release of danger-associated molecular patterns often resulting in immunogenic cell death (ICD) and a potent antitumor immune response [3]. These inherently immunogenic viral backbones are often associated with small genome size and a limited number of proteins to antagonize host antiviral immune responses. Larger DNA based viruses that are more genetically complex, such as Herpes simplex virus type-1 (HSV-1) and vaccinia virus (VV), have evolved to encode proteins involved in host immune evasion in order to facilitate the spread of infection [7,8]. However, these viral backbones can be engineered to initiate an inflammatory response simultaneously with the release of danger-associated molecular patterns initiating ICD programs [3,8]. Despite a robust viral infection, prevention of a therapeutic antitumor immune response may still occur due to the immunosuppressive nature of the TME [9]. Thus, selecting and/or engineering a therapeutic viral backbone that can initiate an immune response, while important, is likely not enough to break the immunological tolerance found in the TME in the majority of patients.

Arming viruses with immune modulators

Along with danger signals, cytokine release from infected cells is an important characteristic of OV-mediated ICD. This can lead to a localized inflammatory response and subsequent recruitment of lymphoid cells to the site of infection. As one might predict, the constellation of cytokines released during tumor infection varies depending on the OV platform being used. For instance, vesicular stomatitis virus and VV infections are known to promote the release and transcriptional activation of proinflammatory chemokines, CXCL1 and CXCL5, as well as cytokines like GM-CSF, promoting neutrophil infiltration [2]. Oncolytic reovirus infections locally increase the production of proinflammatory cytokines such as IL-6 and GM-CSF, allowing for increased lymphoid cell recruitment and MHC Class I molecule expression [3]. To augment the natural induction of cytokines, several groups have adopted the approach of encoding particular cytokines within OV backbones. Attractive candidates for this strategy include IL-12 [10] and TNF-α [11] which have promising therapeutic potential but unacceptable toxicities when administered systemically. In principle, OV-driven expression of this class of cytokine ensures localized, high dose treatment at the site of infection without systemic exposure. Many cytokines and co-stimulatory molecules have been encoded as transgenes into OVs, including CCL3 and CCL5 in AdV, IL-2 in HSV-1 and Newcastle Disease Virus, IL-12 in AdV and vesicular stomatitis virus, among others [2,3]. T-Vec, commercially known as Imlygic, is the first US FDA and EMEA approved OV built upon an HSV backbone that has been engineered to encode GM-CSF to stimulate the production/recruitment of granulocytes and monocytes within the TME and stimulate T-cell dependent immune responses against patient tumor antigens [12]. More recently, to stimulate and link innate and adaptive antitumor immune responses, assorted virally encoded immune modulators have been investigated. For instance OX40/OX40L, associated with increasing the effector to regulatory T-cell ratios, and the production and signaling of cytokines by T cells, antigen presenting cells, natural killer (NK) cells, and NK T-cells [13,14]; the Glucocorticoid-induced TNF receptor (GITR)/GITRL, with regulatory T-cell inhibitory effects and activating effects on CD8⁺ effector T-cells [15]; the inducible co-stimulator (ICOS)/ICOSL, which plays a key role in stimulating the proliferation, differentiation and function of various effector cells [16]; and CD137/CD137L, which inhibits apoptosis, and enhances the proliferation and effector functions of T- and NK cells [17].

A complementary strategy to initiating a natural T-cell mediated response against tumor antigens is to encode within the virus, Bi-specific T-cell Engagers or BiTEs [18]. These molecules consist of linked variable chain antibody fragments directed against the T-cell antigen CD3 and a specific tumor-associated antigen. BiTEs secreted from an infected cell will spread throughout the TME and lead to the formation of artificial immunological synapses between malignant cells and any T cell in the vicinity [18]. The net result of this strategy is to kill both infected and uninfected tumor cells with the goal of overall improved therapeutic outcomes.

When a robust T-cell response is initiated against a tumor, immunological resistance can emerge by cancer cell expression of immune checkpoints that lead to paralysis of invading tumor-specific-T-cells [19]. To overcome this problem so-called ‘immune checkpoint inhibitor’ or ICI antibodies have been developed. For example, the ICI antibody directed against the CTLA-4, is effective in some melanoma patients, but often associated with severe immune-related toxicity and has generally lower efficacy in nonmelanoma tumors [19]. A second class of ICI antibodies targeting programmed PD-1 and PD-L1 have demonstrated clinical activity across a wider range of cancer types and in general have reduced toxicity when compared with anti-CTLA-4 [19]. Although these are both exciting therapeutics in their own right, they remain effective only in a minority of cancer patients and have off-target toxicities associated with systemic delivery. There is now considerable effort aimed at combining ICIs with OVs either as two stand-alone therapeutics or as a virus expressing an ICI transgene. In both preclinical and clinical settings combination of ICI antibodies with OVs have proven effective in a range of malignancies including refractory triple-negative breast cancer [20], melanoma [21] and gliomas [22]. Combining ICIs with OVs as a single therapeutic has been accomplished with a number of OV platforms including (but not limited to) VV [23], Influenza A virus [24], and Myxoma virus [25]. Often, these virally-encoded checkpoint inhibitors were equally or more effective than either therapy alone or in combination, and improved safety profiles [23–25].

The co-incident development of OVs and anticancer immune modulators has led to the convergence of the two classes of therapeutics. OVs seem destined to be not only cytolytic agents and immune adjuvants, but perhaps equally important, replicating cancer gene therapy vectors that selectively and locally express anticancer immune modulators.

Financial & competing interests disclosure

JC Bell is a founder and sits on the Board of Directors for Turnstone Biologics. JCB is supported by the Canadian Institutes of Health Research (CIHR), the Canadian Cancer Society Research Institute, the Ontario Institute of Cancer Research, and the Terry Fox Foundation. ATC is supported by a CIHR Master’s Award. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.