ANAHEIM, CA—A herpes virus that has been modified to kill tumor cells while stimulating the immune system shows promise in Phase I/II trials, according to data presented here at the American Association for Cancer Research Annual Meeting.
Previous research has shown that herpes strains that lack a functional neurovirulence factor, ICP34.5, cause tumor-specific cell lysis. Intratumoral injection of the modified viruses in Phase I trials in glioblastoma or melanoma patients induced tumor-specific cell lysis and showed that the oncolytic viruses were safe.
The viral strains used in previous trials, however, were derived from laboratory strains that were only moderately effective at inducing cell lysis.
To improve on those initial observations, Robert S. Coffin, PhD, Chief Scientific Officer and Founder of BioVex Ltd. and colleagues generated a new oncolytic herpes viral strain. They started with recent clinical isolates of virus, which were more effective at inducing cell lysis than the laboratory-derived strains were. In addition, the team engineered the virus to express human granulocyte macrophage colony-stimulating factor (GM-CSF) protein, which is a strong immune stimulator.
In theory, the new GM-CSF-expressing virus could attack the tumor in two ways: It could induce tumor cell lysis directly; and it could act as a tumor-specific vaccination, using the GM-CSF to boost an immune response to the tumor antigens that were released into the bloodstream as the cancer cells ruptured.
In the new trial, which is the first patient trial of the novel virus, the team treated 26 late-stage cancer patients who had subcutaneous malignant lesions that could be easily injected with the virus and monitored for changes. The patients had a variety of tumor types including melanoma and breast, head and neck, and colon cancers.
In the first part of the trial, patients received a single intratumoral injection of virus at one of three doses, 106, 107, or 108 plaque-forming units (pfu)/ml, depending on the volume of the tumor.
The drug appeared to be well tolerated, Dr. Coffin said, with mild fevers and flu-like symptoms the most common side effects.
In the second phase of the trial, patients received a series of three injections. Because side effects were more common and more severe in patients who were initially seronegative for herpes simplex virus, the researchers designed a tolerizing regimen in which patients first received a low-dose intratumoral injection of 106 pfu/ml, followed by two injections of 108 pfu/ml. The three injections were spaced two to three weeks apart.
Viral DNA was apparent in the serum of 10 patients and virus near the site of injection in three patients after the injections. GM-CSF was detected in both the tumor and in the patients' blood.
“We saw that the virus replicated in the tumor, and that GM-CSF was expressed from the virus,” Dr. Coffin said. “Necrosis was present in the tumor samples taken after injection,” indicating that the virus was causing tumor cell death.
Pre-injection biopsy samples were not available for comparison, however. More than half of the tumors injected showed some response.
The patients were not assessed for overall clinical response because they were so far advanced in their disease progression, he said. “We did see local effects.”
Significantly, researchers saw that some of the injected tumors flattened, as did some noncontiguous neighboring tumors. Signs of inflammation were also present in some tumors, although this response had been seen more frequently in the animal models during preclinical testing than in the Phase I/II human trials, Dr. Coffin noted.
Despite this cautionary note, John Mendelsohn, MD, President of the University of Texas M. D. Anderson Cancer Center, who moderated a news conference in which the work was presented, stressed the importance of this observation.
“I think it is worth emphasizing that some of the tumors that were not injected also showed inflammation, which suggests that immune response was being developed,” he said. “That is a main focus of the study.”
“This is a very exciting approach. Many groups are studying ways to use viruses to kill tumors, and using viruses to carry a stimulus to the immune response is a clever combination.”
Dr. Coffin concluded, “We believe that oncolytic viruses have shown considerable promise clinically. Based on that we developed an improved version and have now taken that to the clinic. We hope to combine oncolysis with a tumor vaccine approach.”
The team plans to initiate several Phase II trials in different cancers, including head and neck cancer, breast cancer, and melanoma.
Oncolytic Measles Virus Attacks Liver Cancer In Preclinical Model
In another study presented at the meeting and discussed at the same news conference, an attenuated vaccine strain of the measles virus was shown to kill liver cancer cells and slow tumor growth in a mouse model of hepatocellular carcinoma.
The attenuated vaccine strain of measles virus, MV-Edm, is known to differentially kill a variety of tumor cell types but has not been previously tested in liver cancer. The main cell surface receptor for the vaccine strain is CD46, which is overexpressed in a variety of tumors. This overexpression is thought to contribute to the tumor-specific killing induced by the virus.
Now, Boris Blechacz, MD, a Postdoctoral Research Fellow in the Molecular Medicine Program at the Mayo Clinic, and colleagues showed that CD46 receptor is overexpressed in hepatocellular carcinoma cells, leading them to test whether the virus could be used as a therapeutic vaccine for liver cancer.
When the team added the virus to cell cultures of liver cancer cells, they saw that the virus induced formation of cell syncytia. That means that if one cell is infected, it can induce the cell death of numerous neighboring cells in the so-called bystander effect. Evidence was also found of apoptosis in the infected cells.
To test whether a modified MV-Edm virus that expresses a marker protein could induce cell death in liver cancer, the team injected the virus into xenograft hepatocellular tumors in mice. Each animal received five injections over the course of ten days.
Dr. Blechacz found that animals treated with the virus survived significantly longer than control animals injected with killed virus, suggesting that the virus did have antitumor activity. The virus induced complete regression in one third of the animals treated with active virus.
In a second set of experiments, the team repeatedly tested the possibility of using intravenous injections of the virus. Additionally, they modified the virus so that it expressed the human sodium-iodide symporter protein, which is normally expressed in the thyroid gland and facilitates iodine uptake.
Again the animals treated with active virus had a significantly longer median survival than those treated with killed virus.
Furthermore, radioactive iodine was injected several days after the animals were injected with symporter-expressing virus, and it was found that the radioactivity concentrated in the tumors.
“We have a high level of transgene expression, which gives us an opportunity for an enhanced therapeutic effect with combined viral-radiotherapy,” Dr. Blechacz said.
‘Demonstrates Utility of Animal Models’
Dr. Mendelsohn noted that although it is popular within some parts of the cancer research community to “bash” animal models, this study demonstrates the utility and importance of the approach. He said he expects to see this oncolytic viral approach tested in clinical trials in the future.
“These results clearly demonstrate the high potential of MV-Edm as a novel vector for cancer gene therapy of hepatocellular carcinoma,” Dr. Blechacz concluded.