Search

Top of pageAbstract We developed oncolytic adenoviruses that selectively target cancer cells with active |[beta]|-catenin signaling. To achieve this, Tcf promoters were inserted into early virus promoters E1A, E1B and E4. The virus (Ad5Tcf) demonstrated 1000-fold selectivity to tumor cells in vitro, but only a limited efficacy in vivo upon intravenous administration. We then used a range of approaches to increase in vivo efficacy of this virus. First, RGD motif was inserted into the fiber HI loop to improve infection of cancer cells. This modification increased overall efficacy of the virus and allowed infection of cell lines that did not express the natural CAR receptor. Next, we used combination therapy with m-TOR inhibitor, RAD001, which mainly acts as anti-angiogenic drug, to slow down tumor growth in a way that would not adversely affect the virus replication. This combined treatment with 10E11 particles of intravenous Ad5Tcf or Ad5Tcf-RGD virus and 5 mg/kg/day RAD001 was significantly better than any of the agents alone and could almost completely stop growth even of the most aggressive tumor xenografts. In addition, RAD001 is now undergoing phase II clinical trials as an immunosuppressant, and thus has a capability of limiting immune response to oncolytic adenoviruses. In our experiment, 5 mg/kg/day of RAD001 could completely abolish primary antibody response to adenovirus in immunocompetent mice. To further increase efficacy of the Tcf-retargeted virus, we expressed the Escherichia coli nitroreductase nfsB gene from an internal ribosome entry site (IRES) in the adenovirus L5 major late transcript. This enzyme converts prodrug CB1954 into a toxic derivative. The virus (Ad5Tcf-NTR) expressed nitroductase late with high specificity to cancer cells. This modification has markedly increased virus efficacy in vitro. In vivo, however, the gain from the enzyme-prodrug scheme was only marginal when used alone or in combination with RAD001. As we demonstrated, virus distribution in mouse xenografts was very uneven at late terms, and massively infected and dying tumor regions were separated by necrotic masses and connective tissue streaks that are impassable to the virus. Even minor isolated uninfected tumor fragments then quickly overgrew the killed part. As a conclusion, successful virus therapy of tumors would require further effort to improve virus delivery to achieve even infection and enhanced virus spread within the tumor.

In virtually all colon cancer cell lines |[beta]|-catenin is activated. The target genes of |[beta]|-catenin are activated through the interaction of |[beta]|-catenin with the transcription factor TCF. Previously in our laboratory conditionally replicating adenoviruses with TCF sites in the early viral promoters were constructed. We inserted TCF sites into the E1A, E1B and E4 promoters. This virus, vKH1, shows 1000 fold selectivity for colon cancer cell lines compared to wild type Ad5.

In virtually all colon cancer cell lines b-catenin is activated. The target genes of b-catenin are activated through interaction of b-catenin with the transcription factor TCF. Previously in our laboratory conditionally replicating adenoviruses with TCF binding sites in the early viral promoters were constructed. We inserted TCF sites into the E1A, E1B and E4 early promoters of the wild type adenovirus. This virus, vKH1, shows 1000-fold selectivity for colon cancer cells compared to wild type Ad5.