Oncolytic Viruses for Cancer Therapy: Overcoming the Obstacles

Abstract

Targeted therapy of cancer using oncolytic viruses has generated much interest over the past few years in the light of the limited efficacy and side effects of standard cancer therapeutics for advanced disease. In 2006, the world witnessed the first government-approved oncolytic virus for the treatment of head and neck cancer. It has been known for many years that viruses have the ability to replicate in and lyse cancer cells. Although encouraging results have been demonstrated in vitro and in animal models, most oncolytic viruses have failed to impress in the clinical setting. The explanation is multifactorial, determined by the complex interactions between the tumor and its microenvironment, the virus, and the host immune response. This review focuses on discussion of the obstacles that oncolytic virotherapy faces and recent advances made to overcome them, with particular reference to adenoviruses.

Figure 1.

Mechanisms of tumor selectivity of several oncolytic viruses. The interferon (IFN)/double-stranded RNA-activated protein kinase (PKR) pathway is a natural anti-viral defense system. IFNs produced by infected cells result in the upregulation of PKR. On binding to viral double-stranded RNA (dsRNA), PKR autophosphorylates, which in turn phosphorylates the α subunit of eIF-2. Phosphorylated eIF-2α sequesters eIF-2B, a guanine nucleotide exchange factor. Without eIF-2B, the GDP bound to eIF-2 cannot be exchanged for GTP. As a result eIF-2 is unable to bring the initiator transfer RNA (tRNA) to the 40S ribosomal subunit, and the synthesis of viral protein is inhibited. Inactivated IFN and activated Ras pathways are frequently found in cancer (the latter could inhibit PKR), and some naturally-found viruses can replicate selectively in cancer but not normal cells, including the Newcastle disease virus (NDV) [21], reovirus [22], vaccinia virus [23], and vesicular stomatitis virus (VSV) [24]. The herpes simplex virus (HSV) protein ICP34.5 interacts with cellular phosphatase 1α to dephosphorylate eIF-2α, leading to synthesis of proteins needed for virus replication. Deletion of gene that encodes for ICP34.5 (RL1) results in selective replication in tumors with a defective IFN/PKR pathway [25]. The influenza virus NS1-deleted mutant is also dependent on this defective pathway [26]. Adenoviruses normally produce virus-associated (VA) RNAs to inhibit PKR. As such, engineered VAI-deleted adenovirus (dl331) could replicate selectively in tumors with an activated Ras pathway [27]. Epstein-Barr virus (EBV) also expresses RNAs similar to VA RNAs and these can complement dl331, resulting in selectivity in EBV-associated tumors [28].

Figure 2.

Engineered replication selectivity of oncolytic adenoviruses (Ads) by deletion of the E1A, E1B 19K or E1B 55K gene. Retinoblastoma protein (pRb) is normally hypophosphorylated and binds to transcription factors of the E2F family to regulate the G1-to-S checkpoint of the cell cycle. Upon stimulation by mitogenic signals, upregulation of cyclins enables cyclin-dependent kinases (CDKs) to phosphorylate pRb, releasing E2F that leads to the expression proteins needed for DNA synthesis and thus cell cycle progression. E2F upregulates p14^ARF, which inhibits Mdm2. Mdm2 normally results in p53 degradation. p53 is a transcription factor that is upregulated and activated by stress signals such as virus infection or DNA damage. It results in the expression of proteins that induce apoptosis (Bax), cell cycle arrest (p21^CIP1/WAF via its inhibition of CDK2) or DNA repair. p16^INK4A is a tumor suppressor that inactivates CDK4/6. The adenoviral E1A proteins bind to pRb to release E2F, so that viral DNA could be replicated. E1A also promotes the acetylation of pRb by p300/CBP, causing pRb to associate with Mdm2 to inhibit p53. Because cancer cells are often in the S phase, E1A CR2-deleted Ad5 mutant (dl922-947) could selectively replicate in and destroy replicating cancer cells but not normal resting cells [29]. E1B 19K binds to and inhibits Bax. The tumor selectivity of E1B 19K-deleted Ad2 (dl250) is due to multiple defects in the apoptotic pathways, where survival of the virus in normal cells would be limited owing to rapid apoptosis induction in the presence of tumor necrosis factor-α (TNF-α) [30]. E1B 55K interacts with the adenovirus E4 open reading frame 6 (E4orf6) protein to form an E3 ubiquitin ligase complex that targets p53 for degradation. It also induces the expression of cyclin E as well as simultaneously inhibits cellular mRNA export and promotes the export of late viral mRNAs. E1B 55K-deleted Ad could replicate in tumor selectively because of non-functioning p53 [31], cyclin E overexpression [32], and E1B 55K-independent late viral RNA export in cancer but not normal cells [33].

Figure 3.

Obstacles to successful delivery of oncolytic viruses to tumor cells. After intravenous injection, viruses are neutralized by pre-existing antibodies and complement activation. Adenoviruses (Ads) also interact with blood cells. Recent work has revealed that Ad5 binds to erythrocytes via the Coxsackie and adenovirus receptor (CAR) and complement receptor 1 (CR1) in the absence and presence of anti-Ad5 antibodies, respectively [178]. Sequestration into other organs and the reticuloendothelial system is a particular problem, often with resulting toxicities. From the blood stream, viruses have to pass through a mixture of extracellular matrix, cells (including normal and immune cells) and high interstitial fluid pressure before reaching the tumor. They then have to attach to the cellular receptor (often trapped in tight junction), be internalized, translocate to the nucleus, replicate, produce structural and other proteins, lyse the cell and release their progenies – some of these steps could be inhibited by factors such as the natural host immune response, hypoxic environment, soluble factors, and genetic changes in the tumor cell.