Increasing oncoselectivity of virotherapy: Receptor targeted viruses against CD30-positive malignancies

Classical Hodgkin lymphoma (cHL) is a hematopoietic malignancy with a characteristic cellular composition. The tumor mass is made up of infiltrated lymphocytes and other cells of hematologic origin but only very few neoplastic cells that are mainly identified by the diagnostic marker CD30. While most patients with early stage cHL can be cured by standard therapy, treatment options for relapsed or refractory cHL are still inadequate, although immunotherapy-based approaches for the treatment of cHL patients have gained ground in the last decade. As targeting molecule CD30 is in focus since it is only minimally expressed on normal cells. Oncolytic viruses (OV) combine the selective lysis of tumor cells with the induction of an antitumoral response. Due to the broad tropism, strategies to increase tumor specificity were developed. One approach is the retargeting of cell entry of viral particles to a receptor of choice through genetic modification of the viral envelope. This thesis investigates CD30-targeted oncolytic viruses as novel therapeutic concept against cHL selectively destroying CD30+ tumor cells. CD30-targeted measles virus (MV CD30) and vesicular stomatitis virus (VSV CD30) were generated in a previous work and rely on a recently described CD30-specific single chain variable fragment (scFv). For VSV CD30 the VSV glycoprotein G reading frame is replaced by those of the CD30-targeted MV glycoproteins. In the current thesis, selectivity for the infection of CD30+ cHL cells was demonstrated by infection experiments of two cHL cell lines L 428 and KM H2. In vitro killing capacity of VSV CD30 vs. MV CD30 was examined, demonstrating that VSV CD30 induced more rapid and efficient killing of both cHL cell lines. In contrast, CD30+ activated healthy T lymphocytes were protected from infection with CD30-targeted viruses. Notably, VSV CD30 yielded much higher titers than MV CD30 and could be purified out of supernatant of infected cells. For preclinical studies, two xenograft tumor mouse models of human cHL in immunodeficient NSG mice were designed. First, a local subcutaneously growing cHL mouse model was established. Subcutaneous KM H2 but not L 428 tumors responded to VSV CD30 treatment with growth inhibition. In addition, replicating virus could be recovered from tumor tissue. After stabilizing tumor growth using “Matrigel” and demonstrating that virus replication is not hindered in this extracellular based hydrogel, KM H2 was identified as a potentially appropriate tumor model. Using the established local mouse tumor model, antitumoral activities of both MV CD30 and VSV CD30 were evaluated. Remarkably, intratumorally, as well as systemically injected VSV CD30 significantly slowed down tumor growth of cHL xenografts resulting in a substantially prolonged survival of tumor-bearing mice. Second, a disseminated growing cHL mouse model was established to represent a clinically more relevant tumor burden. At least in some mice, VSV CD30 treatment induced tumor growth inhibition, again underlining the CD30 targeting capacity. In contrast to cHL, CD30+ anaplastic large cell lymphoma (ALCL) cell lines were not susceptible for the infection with VSV CD30 but rather for the infection with MV CD30. These findings demonstrate that oncolytic potential of VSV CD30 cannot be generalized and has to be separately evaluated for each type of tumor. In summary, this study is a proof of concept that VSV CD30 has oncolytic activity against xenograft cHL tumors. It will pave the way for further preclinical investigations to use VSV CD30 as a novel therapeutic agent for the treatment of cHL