Oncolytic immunotherapy
appears unparalleled in its ability to increase anti-tumor immunity while
decreasing local immunosuppression. Laboratory data is now being confirmed in
clinical trials. Accordingly, Big Pharma has lost some of its skepticism and
acquired a number of oncolytic viruses companies. However, there is much room
for improvement with regard to the early generation viruses that have been
studied in humans thus far.
It is
increasingly accepted that human cancers come in two immunological varieties:
hot and cold. The former are characterized by infiltration of CD8+ lymphocytes,
a high mutational burden resulting in neoantigens, and signs of
immunosuppression (eg PDL-1 expression). The latter lack these characteristics.
One reason is that the mutations present in cold tumors are not very
immunogenic, or perhaps they have been selected against through immunoediting
(Hemminki A). A subclass of a cold tumor is an immune excluded tumor where the
presence of stromal molecules such as beta-catenin prevent T-cells from
reaching the tumor interior.
What is the
utility of making cold tumors hot?
Hot tumors
can be treated quite effectively with checkpoint inhibitors (eg. ipilimumab,
pembrolizumab, nivolumab, atezolizumab, avelumab, durvalumab), which are
already approved but which only work in a small proportion of patients – those
with hot tumors (Hemminki A). Oncolytic viruses have single agent efficacy as
proven by the recent approval of T-Vec in the US and EU, but their best use may
well be in combination regimens. Of note, oncolytic viruses can make cold
tumors hot, but they do not seem to add to the toxicity of checkpoint
inhibitors (Andtbacka et al, Curti et al, Puzanov et al, Ribas et al). In fact,
clinical data has shown that oncolytic viruses are among the safest oncological
approaches.
What then
is the best way to make cold tumors hot?
In this
regard, nothing is theoretically as potent and multifunctional as oncolytic
viruses. They can release tumor neoantigens, they can convert local
immunosuppression into a pro-inflammatory environment conducive to anti-tumor
responses, they can stimulate both pathogen and damage associated pattern
recognition receptors (PAMP and DAMP) and cause immunogenic cell death.
Moreover, anti-viral immunity results in anti-tumor immunity through eg. CD4+
mediated mechanisms such as epitope spreading. Virus seems very dangerous to
the immune system and having a lot of it in the tumor can break tolerance and
regain immunological recognition of the tumor.
Importantly,
not all viruses are alike. Some viruses are quite stealthy, which might appear
advantageous for systemic delivery, but if they can’t induce immunity at the tumor,
perhaps they shouldn’t be called immunotherapy at all? Some viruses such as
vaccinia are immunogenic, but result in antibody responses instead of cellular
responses. This would promote rapid neutralization instead of the sought-after effects
in the tumor microenvironment. Based on human data, adenoviruses are
unparalleled in their ability to stimulate the adaptive arm of the immune
system (Hemminki A). Specifically, they result in induction of CD8+ and CD4+ T-cell
responses, which are key to conversion of cold tumors into hot tumors.
Accordingly, single-agent clinical efficacy is most prominent in cold tumors
(Taipale et al 2016a), but the potential in the context of anti-PD1/PDL-1 is
tremendously exciting (Cervera-Carrascon et al).
Moreover,
5/3 chimeric adenoviruses combine the best of both worlds by being able to
transduce distant metastases in humans (Koski et al). Thus, even after local
delivery, systemic efficacy occurs through two mechanisms: the virus transduces
metastases through blood and the immune response generated can also be
body-wide. 5/3 chimeric adenoviruses utilize blood cells such as red blood
cells and lymphocytes to avoid neutralizing antibodies. Again, not all viruses
share these appealing characteristics.
Arming with
transgenes combines the best features of conventional gene therapy with the
self-amplifying oncolytic platform. Demonstrating how easy it might be to
improve on current results, many oncolytic viruses currently being tested
clinically lack an arming device, while most of the rest feature the prototypic
GMCSF. A multitude of more sophisticated arming devices have been tested in the
laboratory and are entering clinical trials (Cervera-Carrascon et al).
The problem
with oncolytic viruses is their complexity. Most people in drug development,
biotech and pharmaceutical companies have received training in molecular, cell
or other applied biology, biochemistry, or maybe a pharmacy. In these molecular
worlds, viruses have far too many diverse effects on cells, the immune system,
the tumor microenvironment and other organs of the body, to be easily
comprehensible. Decision-makers in Big Pharma often have a business or legal
training meaning that virology and immunology can be challenging.
A skeptic
might claim that the optimal drug development molecule is simple, with a
specific target and uncomplicated mechanism of action. Oncolytic viruses are
certainly none of these things. In contrast, simpler approaches such as STING
agonism, TLR agonists, checkpoint inhibitors and even gene modified adoptive
T-cell therapy have thus far been more attractive to Pharma (Tang et al).
Moreover, there is a lingering fear of neutralizing antibodies being a problem
for oncolytic viruses, even if the clinical data proves otherwise. In reality, anti-viral
immune responses probably enhance the effects of oncolytic viruses (Ricca et al,
Gujar et al, Taipale et al 2016b).
While STING
agonists and TLR agonists might work, many oncolytic viruses, including
adenoviruses, induce the same biological effects. However, oncolytic viruses
have multiple further anti-tumor mechanisms over mere stimulation of a single
molecule. When going to war (against cancer), which is better? To target an
anti-aircraft missile battery or maybe an enemy destroyer, or to employ a
comprehensive attack plan targeting enemy ground, air and naval forces, their
information infrastructure, energy resources and supply lines. Not only do
oncolytic viruses achieve this locally but they seek out other enemy bases both
near and far, and attack those automatically. Each dying cancer cell releases
thousands of vicious guerillas to penetrate deeper into and beyond enemy lines.
The best part is that these guerrillas convert the locals (the immune system)
against the corrupt regime (malignant cells) which has enslaved the population
(the body). Oncolytic viruses are the veritable neutron bomb (in the 1960s
sense) since they kill enemy personnel (cancer cells) while leaving the
buildings (normal organs) intact, which underlies their impressive clinical
safety.
Given the strength
of the theoretical background, it is not surprising that the emerging clinical
data is robust. Thus far, clinical checkpoint inhibitor combination data has
been published for two types of herpes viruses and one coxsackie virus
(Andtbacka et al, Curti et al, Puzanov et al, Ribas et al). Studies with many
other viruses are ongoing (Tang et al). Of note, none of the viruses used thus
far have been designed with T-cells in mind. Since checkpoint inhibitors work
through T-cells we believe that viruses developed with this specific goal will
provide exciting patient data. In preclinical studies we have been able to cure
every single rodent resulting in 100% survival (Cervera-Carrascon et al,
www.tiltbio.com), which is far superior over the aforementioned viruses.
In all
studies featuring checkpoint inhibitors and oncolytic viruses, safety has been
excellent and efficacy impressive. As usual in medicine we have to of course wait
for randomized phase 3 data to be sure, but already Big Pharma has taken note. At
least four deals in or above the 1 billion USD ballpark have been announced
(BioVex/Amgen, Psioxus/BMS, Benevir/J&J, Turnstone/Abbvie). The invisible
hand of the market seems to be grabbing for oncolytic viruses.
Amgen’s
T-Vec/pembrolizumab trial, which is furthest along, went directly from phase 1b
to randomized phase 3 and enrolled all patients in an incredibly short time, by
January 2018. Results might be available within the next year. The good news
for biotechs working in this area is that this is only the beginning. Looking
at the history of medicine, there is limited first-mover advantage. Rather,
efficacy and safety (and marketing) determine which drugs will provide the
biggest revenues. While the oncolytic immunotherapy space is certainly no Blue
Ocean any more, it is not quite a Red Ocean yet either (as defined by Kim &
Mauborgne), so there is room for innovative biotechs to operate. There is a
mind-boggling number of clinical trials ongoing in combination immune-oncology
(Tang et al) but the rationale is nowhere stronger than with T-cell targeting
oncolytic adenoviruses (www.tiltbio.com).
In summary,
there is now little doubt that oncolytic viruses will be an important part of
how we will treat cancer in the future. Clinical
data has given all stakeholders reasons-to-believe that laboratory data
translates well into humans. Immunotherapies such as checkpoint inhibitors,
tumor infiltrating lymphocytes (TIL) and chimeric antigen receptor (CAR) T-cell
therapies have demonstrated that even metastatic cancer can sometimes be cured.
With oncolytic viruses, we can increase the proportion of patients with this
outcome.
Akseli
Hemminki, MD, PhD
Professor
of Oncology
CEO of TILT
Biotherapeutics Ltd.
References
Andtbacka
RHI et al. Final results of a phase II multicenter trial of HF10, a
replication-competent HSV-1 oncolytic virus, and ipilimumab combination
treatment in patients with stage IIIB-IV unresectable or metastatic melanoma. J
Clin Oncol 2017 35:15_suppl, 9510-9510
Cervera-Carrascon
V et al. TNFa and IL-2 armed adenoviruses enable complete responses by
anti-PD-1 checkpoint blockade. Oncoimmunology 2018; 7(5): e1412902.
Curti BD et
al. Activity of a novel immunotherapy combination of intralesional
Coxsackievirus A21 and systemic ipilimumab in advanced melanoma patients
previously treated with anti-PD1 blockade therapy. J Clin Oncol 2017
35:15_suppl, 3014-3014.
Gujar et
al. Antitumor Benefits of Antiviral Immunity: An Underappreciated Aspect of
Oncolytic Virotherapies. Trends Immunol 2018 Mar;39:209-221.
Hemminki A.
Crossing the Valley of Death with Advanced Therapy. Published by Nomerta,
Turku, Finland, 2015. Available at http://www.nomerta.net and several e-book
stores globally
Koski A.
Biodistribution analysis of oncolytic adenoviruses in patient autopsy samples
reveals vascular transduction of non-injected tumors and tissues. Mol Ther,
2015; 23: 1641-52.
Puzanov I,
et al. Talimogene Laherparepvec in Combination With Ipilimumab in Previously
Untreated, Unresectable Stage IIIB-IV Melanoma. J Clin Oncol. 2016;34: 2619-262016.
Ribas A et
al. Oncolytic Virotherapy Promotes Intratumoral T Cell Infiltration and
Improves Anti-PD-1 Immunotherapy. Cell. 2017;170:1109-1119.
Ricca JM et
al. Pre-existing Immunity to Oncolytic Virus Potentiates Its Immunotherapeutic
Efficacy. Mol Ther. 2018 Apr 4;26:1008-1019.
Taipale K
et al. Chronic activation of innate immunity correlates with poor prognosis in cancer
patients treated with oncolytic adenovirus. Mol Ther, 2016a; 24: 175-83.
Taipale K
et al. Predictive and Prognostic Clinical Variables in Cancer Patients Treated
With Adenoviral Oncolytic Immunotherapy. Mol Ther. 2016b;24:1323-32
Tang J et
al. Comprehensive analysis of the clinical immuno-oncology landscape. Ann
Oncol. 2018;29:84-91.