Impact of human neutralizing antibodies on antitumor efficacy of an oncolytic adenovirus in a murine model.

PURPOSE: The purpose of this study was to assess the impact of anti-adenovirus neutralizing antibodies (AdNAbs) on the distribution, tolerability, and efficacy of intravenously administered oncolytic adenovirus. A translational model was developed to evaluate the impact of humoral immunity on intravenous administration of oncolytic adenovirus in humans. EXPERIMENTAL DESIGN: Initially, severe combined immunodeficient (SCID)/beige mice were passively immunized with various amounts of human sera to establish a condition of preexisting humoral immunity similar to humans. A replication-deficient adenovirus encoding beta-galactosidase (rAd-betagal) was injected intravenously into these mice. An AdNAb titer that mitigated galactosidase transgene expression was determined. A xenograft tumor-bearing nude mouse model was developed to assess how a similar in vivo titer would impact the activity of 01/PEME, an oncolytic adenovirus, after intravenous administration. RESULTS: In SCID/beige mice, there was a dose dependence between AdNAbs and galactosidase transgene expression; 90% of transgene expression was inhibited when the titer was 80. A similar titer reconstituted in the nude mice with human serum, as was done in the SCID/beige mice, did not abrogate the antitumor efficacy of the replicating adenovirus after intravenous administration. Viral DNA increased in tumors over time. CONCLUSIONS: In intravenous administration, preexisting AdNAb titer of 80 significantly attenuated the activity of a 2.5 x 10(12) particles per kilogram dose of nonreplicating adenovirus; the same titer had no affect on the activity of an equivalent dose of replicating adenovirus. Our results suggest that a majority of patients with preexisting adenovirus immunity would be candidates for intravenous administration of oncolytic adenovirus.

Intravesical Ad-IFNalpha causes marked regression of human bladder cancer growing orthotopically in nude mice and overcomes resistance to IFN-alpha protein.

We have produced prolonged, high local concentrations of interferon in vivo by intravesical instillation of adenoviruses encoding interferon-alpha (Ad-IFNalpha) together with the gene transfer-enhancing agent Syn3. We found sustained interferon protein levels for days, both in normal mouse urothelium and in human bladder cancer cells growing as superficial bladder tumors in nude mice using an orthotopic bladder model developed by us. Tumor burden in the bladder was determined utilizing cancer cells containing the green fluorescent protein. Marked tumor regression was observed following two 1-h exposures of Ad-IFNalpha/Syn3 and little or no cytotoxicity was detected in normal cells. Similar intravesical instillation of clinically relevant concentrations of IFN protein alone or Ad-IFNalpha without Syn3 was ineffective. Surprisingly, in vitro, Ad-IFNalpha also caused caspase-dependent death of bladder cancer cell lines that were resistant to high concentrations of IFN-alpha protein, including the cell line used in vivo. These findings demonstrate that Ad-IFNalpha can overcome resistance to IFN-alpha protein both in vitro and in vivo and support evaluation of intravesical Ad-IFNalpha/Syn3 for the treatment of superficial bladder cancer.

Acute hepatotoxicity of oncolytic adenoviruses in mouse models is associated with expression of wild-type E1a and induction of TNF-alpha.

Replication competent adenoviruses with various E1 modifications designed to restrict their replication to tumor cells are being evaluated as oncolytic agents in clinical trials. In mouse models, we observed that such oncolytic adenoviruses showed greater hepatotoxicity than E1-deleted adenovirus vectors following intravenous administration. Additional studies in congenic BALB/c, nude, and beige/Scid mice demonstrated dose-dependent hepatotoxicity and indicated that beige/Scid was the most sensitive strain. Comparison of E1-containing viruses showed that hepatotoxicity correlated with expression of wild-type E1a in the liver. Pharmacokinetic analysis showed rapid increases in viral DNA levels in the liver with a virus containing wild-type E1a. This was correlated with rapid induction of TNF-alpha to high levels and with rapid elevation of serum ALT. Hepatotoxicity was significantly reduced for an adenovirus with deletions in the region E1a (dl01/07) or a virus lacking E1a. The results suggest a mechanism for hepatotoxicity involving virus-induced production of local TNF-alpha release and E1a-mediated sensitization of hepatocyte killing.

Pharmacologic indicators of antitumor efficacy for oncolytic virotherapy.

Central to the development of oncolytic virotherapies for cancer will be a better understanding of the parameters that influence the outcome of virotherapy to treat disseminated cancer by i.v. administration versus regional disease by local treatment. Intratumoral administration of 01/PEME, an oncolytic adenovirus, required approximately 1000-fold less dose than i.v. administration to induce similar tumor growth inhibition. Despite the short (<10 min) circulating half-life of the virus DNA, we could monitor virus distribution to the tumor site and observed virus replication by >1000-fold increase in virus DNA copies over time. There were doses of 01/PEME for which the virus DNA concentration in the tumor increased over time but did not result in antitumor efficacy. Oncolytic virus replication at a tumor site may not be a relevant indication of antitumor efficacy. Efficient distribution to the tumor site may be one of the most critical parameters for antitumor efficacy with oncolytic virotherapy.