XVir-N-31-based glioma-oncovirotherapy and its combination with an immune checkpoint inhibition that targets PD-1/PD-L1: Assessment of immune responses

Abstract

Glioblastoma (GBM) is the most common malignant primary brain tumor in adults. Even with the current standard of care, including optimal surgical resection, radio- and chemotherapy, the median survival is limited to less than 20 months. Therefore, new therapeutic strategies are urgently needed. Reasons for the bad outcome are, belong others, the highly invasive growth and the strong immunosuppression of this tumor. A promising approach to treat GBM is oncolytic virotherapy. It has been shown that oncolytic viruses (OV) provide, aside oncolysis, immunostimulatory effects. By the release of cytokines and damage associated molecular pattern (DAMP) proteins from OV infected tumor cells, OVs are able to initiate immunogenic cell death (ICD). Subsequently, tumor specific immune cells may be activated and attracted, leading to an anti-tumor response even towards those GBM cells, that have invaded in the healthy brain and that are located far away from the original tumor. In preclinical trials the oncolytic adenovirus (OAV) XVir-N-31 was already found to be effective in the therapy of experimental GBM as it significantly prolonged the survival of GBM bearing mice. The aim of this project was to examine the impact of XVir-N-31 on its capacity to induce ICD and to determine its immunostimulatory, anti-tumoral effects both in vitro and in vivo using an immuno-humanized GBM mouse model. Besides others, one of the typical immunosuppressive features of GBM is the strong surface exposure of programmed cell death ligand 1 (PD-L1) on GBM cells, leading, by interaction with programmed cell death 1 (PD-1) expressed on immune cells, to the exhaustion of these cells. Therefore, the impact of an additional immune checkpoint inhibitory therapy was examined. In this regard, an XVir-N-31 based OVT in combination with a blockade of the PD-1/PD-L1 axis was conducted, either by the systemic application of Nivolumab in combination with an intra-tumoral injection of XVir-N-31, or by the local expression of an anti-PD-L1 neutralizing antibody that is coded by the XVir-N-31 derivate XVir-N-31-anti-PD-L1. For this, the functionality of XVir-N-31-anti-PD-L1 was confirmed in vitro. In contrast to the wild type-like adenovirus dl309, which possesses higher cytotoxicity, XVir-N-31 and XVir-N-31-anti-PD-L1 induce ICD in vitro and in vivo as determined by the release of DAMPs and proinflammatory proteins. In vivo, a single intratumoral injection of XVir-N-31 increased the amount of tumor infiltrating T lymphocytes and natural killer cells even more than dl309. Furthermore, this effect was not only restricted to virus-injected tumors but was also visible in untreated tumors located in the contralateral hemisphere. The additional blockade of the PD-1/PD-L1 interaction by either multiple systemic applications of Nivolumab or by XVir-N-31-anti-PD-L1 further enhanced the DAMP concentration in the tumor, but also increased the number of tumor infiltrating lymphocytes (TILs). This was true for both, virus-injected as well as for contralateral located, untreated GBMs. Whereas a single, intratumoral injection of dl309 or XVir-N-31 led to massive tumor volume reduction of injected tumors, only the combination of an XVir-N-31 based OVT in combination with the blockade of the PD-1/PD-L1 interaction showed a significant growth reduction of contralateral tumors. Overall, the obtained data provide strong evidence that XVir-N-31 is a promising therapeutic agent for a successful treatment of GBM and that its immune activating properties and the induction of ICD is of greater importance than its cell killing capacity. For the improvement of the therapy and an induction of strong abscopal effects on tumor cells that are located far away from the site of virus injection, an additional blockade of the PD-1/PD-L1 interaction, ideally by XVir-N-31-anti-PD-L1 expressed anti-PD-L1, seems to be highly beneficial.