Glioblastoma represents fourty to fifty percent of all malignant brain tumors and has a poor prognosis with a median survival of twelve months and a five year survival rate of fifteen percent. Surgical removal of all tumor tissue and drug delivery through the blood brain barrier pose unique challenges to effective treatment. Recently, Phase III immunotherapy trials failed at improving survival of glioblastoma patients (Lim et al. 2018). This is likely due to a highly immunosuppressive microenvironment dominated by pro-tumorigenic macrophages that drastically limits immunosurveillance by lymphocytes. Dr. Eric Holland (Human Biology division) and Dr. McGarry Houghton (Clinical Research Division and Human Biology Division) combined their respective expertise on glioblastoma and solid tumor immunology to evaluate a new approach to eradicate this tumor, a work that was recently published in JCI Insight.
The Holland lab previously developed a genetically engineered glioblastoma mouse model where Pten is deleted homozygously in the brain upon local injection of viruses encoding for the Cre recombinase. This model faithfully recapitulates the main traits of the human disease, including the high proportion of immunosuppressive macrophages and the rare fraction of lymphocytes. In addition, as demonstrated in this study, these tumors fail to respond to the blockade of the immune checkpoint inhibitors PD-1 and CTLA-4, in agreement with the failure observed in clinical trials for this disease.
As a new strategy, Dr. Hans-Georg Wirsching (Holland lab) and colleagues developed an oncolytic herpes simplex virus (oHSV) which includes a payload cassette encoding the human UL16-binding protein 3 (oHSV-ULBP3). ULBP3 is a class 1 major histocompatibility complex-like (MHC-like) molecule that has been shown to drive the differentiation of macrophages into a proinflammatory, antitumorigenic state. Although oHSV did not improve survival of glioblastoma bearing mice compared to saline-treated animals, oHSV-ULBP3 administration almost doubled their survival. This was associated with accumulation of activated macrophages, CD8+ T cells, and a gene expression profile correlating with antigen presentation and T cell activation.
As tumor relapse is almost automatic after surgical removal of glioblastoma due to disseminated tumor cells, it was important for the authors to understand whether this localized immunogenic effect could extend to non-targeted secondary tumor sites (abscopal immunity). To this end, in addition to injecting the Cre encoding virus in the first brain hemisphere to locally deplete PTEN while activating the expression of the cell tracer luciferase, they also administered a virus encoding for a shRNA targeting PTEN, thus simulating PTEN loss without activating the cell tracer. This ingenious strategy allowed them to follow the two tumors independently. In this model, the researchers treated the glioblastoma originating from the shRNA downregulation of PTEN with either a blocking antibody against PD-1, oHSV-ULBP3 or a combination of both. Thanks to the luciferase expression in the untreated tumor, they determined by imaging that the distant tumor growth was drastically delayed in response to oHSV-ULBP3 and further inhibited in combination with anti-PD-1. The survival benefit observed with oHSV-ULBP3 alone, compared to saline, was aslo further increased when combined with anti-PD-1. Strikingly, in both oHSV-ULBP3 and oHSV-ULBP3 + anti-PD-1 conditions, an immunogenic reaction was observed in both the treated and distant tumors, demonstrating that the immune response mounted against the first tumor was strong enough to “spread” to the secondary tumor. Even more surprisingly, oHSV-ULBP3 led to an increased proportion of macrophages expressing PD-1 and additional anti-PD1 treatment further activated macrophages.
These results bring a new wave of optimism in the field of glioblastoma. The oncolytic viruses engineered to counteract the most prominent immunosuppressive population in glioblastoma open new windows of opportunity where both macrophages and T cells are activated, express PD-1, and positively respond to anti anti-PD-1 strategies. This study shows that transforming a cold tumor into a hot tumor is essential for checkpoint inhibitor blockade strategies to be effective.
This work was supported by the National Institute of Health and Oncorus.
Fred Hutch/UW Cancer Consortium member Drs. Holland, Houghton and Pierce contributed to this research.
Lim M, Xia Y, Bettegowda C, Weller M. Current state of immunotherapy for glioblastoma. Nat Rev Clin Oncol. 2018;15(7):422–442.
Wirsching H-G, Zhang H., Szulzewsky F, Arora S, Grandi P, Cimino PJ, Amankulor N, Campbell JS, McFerrin L, Pattwell SS, Ene C, Hicks A, Ball M, Yan J, Zhang J, Kumasaka D, Pierce RH, Weller M, Finer M, Quéva C, Glorioso JC, Houghton AM, Holland E. Arming oHSV with ULBP3 drives abscopal immunity in lymphocyte-depleted glioblastoma. JCI Insight. 2019;4(13):e128217. https://doi.org/10.1172/jci.insight.128217.