by ecancer reporter Will Davies
In June, I wrote a brief introduction to the microbiome; the teeming world of bacteria, fungi and obligate flora that inhabit every nook and cranny of a human body.
While a healthy body can maintain symbioses of numerous species in a secure niche, the dysbioses resulting from any number of causes can tip the balance towards tumourigenesis.
While the main focus of the review was on bacterial pathogens that can, through their presence or absence, result in cancer, I only briefly touched on viruses.
Viruses are somewhat more widely known to induce cancer growth, with reports ranging from as long ago as a century to as recently as last week, linking their infection of hosts with a number of cancers.
However, while a viral infection may sound unfortunate, it might also be just what the doctor ordered1.
With viruses only being so-named in 18982, a virus (influenza, in this case) was reported as inducing remission in a leukaemia patient as early as 19043.
This curious co-infection followed the first use of X rays to cure cancer by only a few months4, but the development of oncolytic virotherapy moved slowly5, with attempts to reproduce the clinical benefits seen in 1904 usually resulting in sometimes prolonged remission6 followed by swift and terminal relapse7.
A major limiting factor to the specificity available with previous viral therapies has been the technology available at their time of development.
Over the years, there have been several attempts to reinvigorate the field, which have only recently gained any traction.
While attempts had been made to find the most tolerable wild-type virus8,9, the field languished under poor public perception. With a century of theory behind virotherapy, it has only been in the last decade, with the advent of genomic editing, that progress has been made10.
Today, we have Oncorine; a genetically engineered adenovirus to target p53 mutations in tumour cells11. Administered via direct intratumoural injections to treat head and neck cancers, Oncorine was developed through American trials12 but remains licensed only in China13,14, and the only internationally licensed oncolytic virus (OV).
That may all be about to change, but we’ll come back to that later. First, it’s worth taking a moment considering what makes for a potential therapy.
Tumour biology is, by its nature, a hard target for therapy; the mutations that drive tumourigenisis can come down to single nucleotide polymorphisms, effectively atom-level difference between cancer cells and healthy tissues.
Seeing as unaltered cells are the natural habitat for viruses, this can be taken as twice as tricky, or a solution to both problems – what better to infiltrate tissues, disrupt cell cycles and attract immune activity than virus?
The features set out in Pol et al 15 sum up what can be broadly agreed as the hallmarks for desirable oncolytic virotherapy;
1. a refined oncotropism, based on the targeting of tumour-associated antigens (TAAs) exposed on the surface of malignant cells
2. an optimized selectivity of replication, based on various systems that allow for the expression of essential viral proteins only in cells of a predetermined tissue, transformed cells, cells exhibiting specific molecular defects, or cells exposed to precise microenvironmental conditions (naturally or artificially)
3. an exacerbated cytotoxicity, based on the expression of potentially lethal enzymes or other tumor-targeting molecules;
4. an enhanced capacity to boost tumour-targeting immune responses, based on the expression of TAAs (in the context of so-called “oncolytic vaccination”), co-stimulatory molecules, immunostimulatory cytokines, or chemokines;
5. a limited standalone immunogenicity, based on coating/encapsulation strategies or changes of the viral surface that reduce the recognition of circulating viruses by the immune system and reticular phagocytes.
More simply put, the essential traits of an effective virotherapy are selective targeting of tumour cells, replication within these cells only, resulting in the damage of these cells only, and attracting immune attention to these cells only; only after the previous tasks have been accomplished.
Straightforward stuff, right?
The trouble begins with even administering the therapy. The tumour microenvironment itself can hinder therapy, as the pressure between cells makes for slow diffusion of virion particles16.
However, digestive enzymes can be grafted into the engineered viruses genome alongside the cytotoxic treatments17, softening surrounding cells and helping to promulgate the infection18.
Once finally delivered, the recombinant nature of viral RNA makes for an effective means of introducing cell-cycle factors, or modulating endogenous signals.
For example human granulocyte macrophage colony stimulating factor (GM-CSF) upregulation in the case of CG007019(currently in phase III trials against bladder cancer) to inhibit tumour growth20.
PANVAC, a pox-based vaccine is designed to attract a local immune response with transgenes resulting in MUC-1, ICAD1 and LFA3 expression, and is currently recruiting patients21. Interleukin expression is another target of special interest for immune modulation, especially of IL-1222,23 and IL-1524,25.
Alternatively, a more direct route of directing cell death could be reawakening subdued p53 signalling. Healthy expression of p53 can direct tumour suppression, as regulated by ubiquitin degradation to keep it in check, but mutated p53, as found in over half of human solid cancers, allows for tumour cell proliferation.
Using adenovirus as a vector to reintroduce wild-type p53 to mutant strains has been found as effective for HNSCC as methotrexate26, and functions well to treat premalignant oral dysplasia when administered as a mouthwash27. Another means of sorting p53 as a target is employed by Onyx-015, an adenovirus which selectively replicates in p53-lacking cells to the point of lysis28.
The lytic cycle of viruses, once within the tumour, is well understood, and is itself a target of research to treat viral and retroviral pathogens29.
Not only does it persist where conventional chemotherapy might fail30 the development of resistance within tumour cells can actually increase its effectiveness; the upregulation of cell growth and proliferation pathways, such as RAS31 and NF-kB32, in advanced cancer results in the incorporation of more of the infectious viral genome.
The hijacking of cellular machinery to replicate viral proteins also directs energy and resources away from tumour cell proliferation. The lysis of infected tumour cells ultimately results in regression of the tumour mass and associated vasculature33.
It is thought that the exposure of tumour and viral cell fragments to circulating immune cells can also attract and educate an innate anti-tumour immunity34, especially in combination with checkpoint immunotherapy35.
More on that later, but for now we have a cell that is replicating its viral load and then dying with an eruption that spreads the infection to nearby cells.
All seems to be going well, until it isn’t:
This process by which viruses infiltrate, infect and ultimately lyse out of their host cells is at once their best weapon, and our worst weakness, a problem shared by many other cancer therapies – how can we selectively kill cancer cells without harming healthy host tissue?
Going back to the earliest studies of infections concurrent with cancer, Russian Far East encephalitis virus was noted as completely ablating sarcoma cells in animal models, shortly before the host passed due to infection related disease36. Similar risks were reported from early observations of hepatitis and Hodgkin’s disease37, and the innate patient immunity to adenovirus38 or mumps39 limited many further attempts.
Another issue within the host beyond tissue specificity is our circulating sentinels, the immune system that has been trained through millions of years of evolution to recognise and lay waste to infection. How can we get the viruses to tumour sites without being cleared by host immunity, and then have them last long enough to attract immune reactions to the tumour sites?
While intratumoural delivery has been the standard so far, systemic therapies would be preferred, especially to target metastatic tissues. To avoid antibody development, or recognition when using a common virus40, immunoapheresis of patient plasma has been trialled, though requires ongoing refinement. Similarly, immunosuppressed patients have hosted above-expected tumour response to OV therapy41. Careful modulation of a patient’s immune system, then, may strike the balance of oncolytic penetration and limiting anti-tumour immunity42.
If only there were some way in which immunity and oncology could be joined, or a virus somehow repurposed to suppress host immunity. What a future that might be…
The most significant development came in 1991, with the deletion of the tk gene from a Herpes simplex virus, resulting in viral replication being confined to dividing cells only43. With viral genomes edited to limit virulence or remove pathogenicity towards non-tumour presenting cells entirely44, oncolytic therapy sounds like an ideal cytotoxic therapy, culminating in the 2015 approval of a first-in-class treatment for advanced melanoma, called talimogene laherparepvec45, or T-VEC for short.
Much like the 1991 leap, T-VEC (formerly Oncovex46) is a modified herpes simplex 1 virus that is in administered intralesionally. In its original announcement, 16.3% of patients receiving T-VEC to treat unresectable metastatic melanoma achieved durable response – tumour shrinkage >6 months – compared to only 2.1% of the control group. In later analysis of these responders "T-VEC resulted in a decrease in size by ≥50 % in 64 % of injected lesions (N = 2116), 34 % of uninjected non-visceral lesions (N = 981), and 15 % of visceral lesions (N = 177). Complete resolution of lesions occurred in 47 % of injected lesions, 22 % of uninjected non-visceral lesions, and 9 % of visceral lesions. Of 48 patients with durable responses, 23 (48 %) experienced PPR, including 14 who developed new lesions only. No difference in overall survival was observed, and median duration of response was not reached in patients with PPR versus those without PPR.”47.
T-VEC has now been approved by the FDA to treat advanced melanoma , and is being investigated in other organ sites49.
Alongside the cell damage mediated by viral lysis, it also has a secondary mode of action that has enkindled new interest in oncolytics; while it can gain entry to healthy and cancerous cells, it’s only replication-competent in cancer cells as a result of gene deletions for Infected Cell Proteins 34.550,51 and 4752. In their place, GM-CSF, a pro-inflammatory cytokine, has been inserted to recruit a host immune response53.
The tumour microenvironment is notoriously immunosuppressive. However, as the infected tumour cells go through the process of cell death54, they express an upregulated amount of tumour-associated antigens to circulating immune cells attracted by the GM-CSF.
When these are recognised in turn by antigen-presenting cells and activated T cells, the hosts adaptive immune system swarms and devours the now-noticeable cancer cells. By painting cancer as an actionable target, like any other infection, the patient’s own immune symptom becomes empowered to heal lesions, and perhaps even develop an immune complement that recognises similar antigens, should relapse occur, and take the initiative55.
The immune activation of T-VEC and other OVs is raising many eyebrows as immunotherapy continues to gain pace in clinical application, most potently as a combined therapy. Checkpoint immunotherapies have become a landmark development in the fight against cancer by their own merits56, and are noted as working synergistically with cytotoxic chemotherapy57 and radiotherapy58. Removing the brakes on T cells and giving them a readily lysing target seems like a straight forward win59, but some researchers are taking this a step further.
Engineered adenovirus and measles virus strains which encode anti-CTLA-460 and PD-L1 antibodies61 directly into their host tumours have reported significant tumour shrinkage and remission. Another exciting angle is rewriting tumour cells to secrete Bispecific T cell Engagers (BiTE) which directly recruits T cell activity62.
Be it as an adjuvant to checkpoint therapy63, or using the transcriptive machinery of the virus to encode a tumour cells own doom into its DNA, the combination of OV and immunotherapy promises to be an exciting avenue to future research, with reports of local tumour shrinkage from injected lesions being complemented by memory-T cells infiltrating distant tumours64,65.
Finally having the technology to edit viral genome and transcription has sparked a surge of interest in oncolytic therapy, over a century since its first recorded description.
With seemingly as many ongoing trials are there are viruses15,66, and many more that have needed to get this far67, effective combinations are already beginning to emerge68 extending survival and delaying disease.
After a long wait, it seems oncolytic therapy is almost certain to take a place alongside checkpoint inhibitors as a pivotal step forwards in the treatment of cancer.
Image Credit: Dr Thomas Splettstoesser. Shared under Creative Commons licence CC BY-SA 4.0
References: