Antitumor immune response during glioma virotherapy

See the article by Yoo et al. in this issue, pp. 1131–1140.

Recent reports indicate that oncolytic viruses constitute a promising therapeutic approach for malignant brain tumors. Recently, Lang and colleagues reported that treatment of patients with recurrent malignancy with a single intratumoral injection of an oncolytic adenovirus resulted in a survival longer than 3 years in 20% of patients.¹ Similar results were reported in a cohort of patients treated with a replication competent poliovirus.²

Although there were anecdotal reports consistent with evidence of tumor regression or improvement during the course of a viral infection or a vaccination, the concept of cancer immunotherapy was pioneered by Coley and collaborators, who proposed the use of inactivated pathogens to treat patients with progressing tumors.³ The modern development of the oncolytic virus field was probably ignited by the seminal report by Martuza and colleagues in the early 90s, who demonstrated that infection of human glioblastoma bearing mice with a genetically modified oncolytic herpes simplex virus (oHSV) resulted in the destruction of the tumor.⁴ This study proposed that viruses can be engineered to selectively replicate in cancer cells. Since then, research in several laboratories has been focused on elucidating the fundamental molecular mechanisms behind the anticancer effect of HSV, particularly in gliomas, where oHSV has been tested in several clinical trials in adults⁵ and, more recently, in children (NCT03911388, NCT02031965, NCT02457845).

While HSV has demonstrated an antiglioma effect as a single agent in some patients, the results from the trials strongly suggest that it should be administered in combination with other therapies to induce a therapeutic effect in the majority of patients with malignant gliomas.

One of the hopes of the Cancer Genome Project was the generation of a new type of cancer therapy based on the development of drugs targeting specific gene/protein abnormalities.⁶ This rationale has triggered an enthusiasm for novel types of personalized and precise strategies to specifically inhibit molecular pathways that are aberrantly activated in cancer cells. The MEK pathway constitutes a hub for signals that originate in the cellular membrane and are then trafficked to promote mitosis and cell proliferation.⁷ Due to its significance in oncogenesis and tumor maintenance, this pathway has been the target of several inhibitors, and some of these small molecules have reached clinical settings. One of these inhibitors, trametinib, shows promise for the treatment of pediatric gliomas,⁸ particularly in combination with BRAF inhibitors, since the abnormalities in the BRAF and MEK pathways often coexist in the same tumor. Difficulty in inducing a dramatic antiglioma effect in every patient using a single agent, a criticism of virotherapy, also applies to small-molecule inhibitors.

In this issue of Neuro-Oncology, Yoo and collaborators report the antiglioma effect of combining oHSV with trametinib.⁹ Intriguingly, they showed that oHSV infection improves the efficacy of trametinib by suppressing trametinib-mediated feedback reactivation of the mitogen-activated protein kinase signaling pathway. One of the most innovative angles of the report is that the viral infection of the tumor enhanced blood–brain barrier penetration of trametinib, improving the efficiency of the chemical treatment. Importantly, the benefits of this combination were reciprocal; trametinib treatment led to a significant reduction in microglia/macrophage-derived tumor necrosis factor alpha (TNFα) secretion in response to oHSV treatment, but without suppressing the activation of CD8+ T cell–mediated immunity. This is a key effect because T-cell activation was one of the major multifaceted mechanisms encompassed by combining oHSV and trametinib, which ultimately resulted in the significant improvement of the survival of glioma-bearing immune-competent mice.

The combination of oncolytic viruses with chemotherapy has been thoroughly tested, and often resulted in an additive or synergistic anticancer effect, regardless of the chosen drug. The reasons and rationale for this combined effect are multiple and may involve the capability of the virus infection to generate targets for the drug, as well as the drug-mediated lysis of the cell, improving the imperfect release of infective virion particles by the host cells,¹⁰ in addition to the improvement of tumor drug intake resulting from the virus-mediated disruption of the blood–brain barrier. This is an original and important observation of Yoo’s study. However, the same mechanism that facilitates the entry of the drug to the brain may, in turn, lead to an increased chance of unwanted toxicity. Furthermore, blood–brain barrier disruption may modify the therapeutic index of chemotherapeutic agents, altering the ability of the clinician to accurately calculate the optimal dose of the targeted inhibitor.

The role of the antitumor immune response is thought to be a key element of virotherapy.^1,2 This immune aspect is a double-edged sword: It is probably directed against both the virus and the tumor. Currently, it is difficult to ascertain, particularly in the clinical setting, whether the response is against the tumor or the virus. In many instances, the only valid evidence would be a consistent and persistent decrease in the tumor mass without the addition of any other therapy. More importantly, it is also unknown why the antitumor immune response is successful in inducing tumor regression in approximately 20% of patients. Intense research is focused on discovering the molecular and cellular mechanisms that underlie the switch that may allow the shift from an initial antivirus immune response to an antitumor one, or, alternatively to precisely describe the tumor microenvironmental factors that permit both immune responses to coexist.¹¹ Keeping with this line of thinking, it is intriguing that data from Yoo and collaborators showed that inhibition of TNFα, a T helper cell 1 cytokine, and as such part of the canonical immune response against both viruses and tumors, did not preclude the antitumor immune response. This observation suggests that TNFα may be responsible for the concentration of the immune response against the virus, and thereby keeping the tumor-associated antigens in the “blindside” of the immune response. Further studies are required to independently test this hypothesis.

Disclosure

C.G-M. and J.F. report ownership interest (including patents) in DNAtrix, Inc. C.G-M. and J.F. are consultants for DNAtrix, Inc.

Funding

This work was supported by the National Institutes of Health/National Institute of Cancer (Brain Cancer SPORE 5P50CA127001-10), the US Department of Defense (CA160525), and the Cancer Prevention Research Institute of Texas (RP170066). The funding bodies were not involved in the study design, the data collection and analysis, the decision to publish, or the preparation of the manuscript.