by ecancer reporter Clare Sansom

Immunotherapeutic agents that act by stimulating the patient’s immune systems to destroy cancer cells – so-called active immunotherapies – are among the most promising of new types of drug in oncology.

They are expected to be particularly beneficial in treating metastatic cancer, although there are still questions about their long-term toxicity.

Oncolytic viruses are viruses that have been engineered to selectively infect, replicate in and lyse tumour cells while stimulating the patient’s immune system to induce tumour specific immunity.

The biotherapeutics company Jennerex Inc., based in San Francisco, California, USA has engineered a vaccinia virus to induce both tumour cell lysis and tumour-specific immunity.

This virus, named JX-594 or Pexa-Vec, has a disrupted thymidine kinase gene that enables it to selectively infect tumour cells, and two inserted transgenes – human granulocyte-macrophage colony-stimulating factor (hGM-CSF) and beta-galactosidase –that allow replication assessment and immune stimulation respectively.

Phase I trials of this virus have already shown that it is able to induce complete responses in solid tumours.

David Kirn of Jennerex and Tony Reid of the University of California, San Diego, USA, working with a large research team including clinicians based throughout North America and in South Korea, have now completed a Phase II dose-finding trial of this virus in advanced hepatocellular carcinoma (HCC).

This deadly disease is the most common type of liver cancer worldwide and is particularly prevalent in East Asia, with about half the over 600,000 yearly deaths from HCC occurring in China.

Thirty patients with advanced HCC were randomized to receive three treatments with either a high dose (10⁹ PFU) or a low dose (10⁸ PFU) of JX-594 by infusion; approximately half the patients in each arm of the trial had antibodies to the vaccinia virus at the start of the trial.

Only one patient was unable to receive all three doses of the virus, and this was due to an adverse event not related to this treatment.

The virus was generally well tolerated at both doses; the only severe adverse event, nausea and vomiting, occurred in the high dose group and the types and levels of mild and moderate toxicity were similar in both arms.

All patients experienced mild or moderate flu-like symptoms during the first day after receiving the virus.

The abdomen and liver of each patient was examined using serial dynamic MRI scans throughout the trial, and the results examined by independent experts who had been blinded to the trial arm.

Liver responses were assessed using the Modified Response Evaluation Criteria in Solid Tumors (mRECIST) for liver cancer, which are based on tumour shrinkage.

Patients in both arms showed similar responses based on these criteria, with measured control rates of 47% in the high dose arm and 46% in the low dose arm at the primary radiographic endpoint eight weeks into the study.

Tumour blood flow and necrosis were measured using tumour contrast enhancement and assessed using the modified Choi criteria; these results also showed similar responses between the doses and between tumours that had been directly injected with the vaccine and distant, tumours in the same patients.

The researchers assessed replication of the JX-594 virus and expression of the transgenes hGM-CSF and beta-galactosidase using several methods.

Expression of hGM-CSF that is delayed five or more days after viral infusion is an indicator of viral replication; this protein could be detected in the plasma of 69% of the high-dose and 46% of the low-dose patients on day 5 of the trial, and it was followed by measurable anti-tumour immunity.

Despite similar tumour responses, the overall survival of patients receiving the high dose of the vaccine was significantly higher than that of those receiving the low dose.

The median overall survival of patients receiving the high dose was 14.6 months for patients in the high dose group and 6.7 months for those in the low dose group (hazard ratio 0.39, p = 0.020) and multivariate analysis confirmed the viral dose received as the best predictor of overall survival.

The results of this trial confirm that advanced liver tumours respond to the oncolytic virus JX-594 and strongly suggest that this virus has a dose-dependent effect on overall survival in this hard-to-treat cancer.

Several larger randomized trials of this virus including a placebo arm are planned or in progress, and this promising therapy is also being tested in other solid tumours.

Reference

Heo, J., Reid, T., Ruo, L. and 23 others (2013). Randomized dose-finding clinical trial of oncolytic immunotherapeutic vaccinia JX-594 in liver cancer. Nature Medicine, published online ahead of print 10 February 2013. doi:10.1038/nm.3089