Virus that Attacks Brain Cancer Headed for Clinical Trials


by Jim Schnabel

June 21, 2010

Malignant glioma has a five-year survival rate of less than 10 percent, so researchers have little to lose by seeking bold treatments for it. One of these involves the use of “oncolytic” viruses, which infect and destroy tumor cells. Such treatments have been held back in part because of the risk that such viruses could infect non-cancerous cells and even spread to healthy people. But a team of European researchers has now reported animal tests of an oncolytic virus that infects and rapidly destroys brain cancer cells while failing to replicate within non-cancerous cells.

“We’re planning to start a clinical trial by the end of the year,” says Karsten Geletneky, a neurologist at the German Cancer Research Center in Heidelberg and first author on the report, which will appear in a forthcoming issue of the journal Neuro-Oncology but was recently released online.

Geletneky and his colleagues used an oncolytic virus known as a type H-1 parvovirus (H-1PV), a small single-stranded DNA virus that normally infects rats, and can also infect humans, but does not appear to cause disease in any host. Initially, in the brains of lab rats, the researchers implanted rat-derived brain tumors known as malignant gliomas, and waited until they were established. Then they injected H-1PV, either intravenously or directly into the tumor.

Both modes of treatment greatly improved average survival times in the treated rats. “Some of the animals had large tumors and were on the brink of death,” says Geletneky. “Within a few days of receiving the virus the animals got a lot better, and in some of them the large tumors simply disappeared”—as confirmed by magnetic resonance imaging. By contrast, all the control rats who had received no treatment died within three weeks.

For the direct injection mode, eight of twelve rats showed strong improvements in overall health and in terms of tumor shrinkage. Four of these responders were soon sacrificed for analysis, but four others were followed for a year:  These not only lost their tumors but also stayed cancer-free without further treatment.

The intravenous mode was designed to test the ability of the virus to cross from the bloodstream into the brain. Six of the nine rats in this group (who received either 8 or 12 high-dose intravenous injections of virus over several days) lost their tumors and stayed cancer-free for the six-month followup period.

Then the researchers implanted human-derived gliomas into the brains of six rats whose immune systems had been weakened so that they would not attack the human-like cells. Geletneky and his colleagues later treated the rats with an aggressive combination of intravenous and direct injections of H-1PV. All of the rats showed strong responses compared with untreated controls. Those that had begun with small tumors appeared to become cancer-free. Those that had begun with larger tumors showed improved health, while their tumors stopped growing.

In all cases, the researchers found evidence that H-1PV infected healthy cells in the rats— but did not replicate in these healthy cells, did not cause apparent damage, and survived only briefly. The H-1PV replicated for a second generation only in the tumor cells, which it killed.

While the main anti-cancer effect came from the virus, there was evidence for a secondary effect from the animals’ immune systems. The rats with weakened immune systems experienced less dramatic tumor-regression from the virus therapy. By contrast, most of the rats with healthy immune systems not only were cured of glioma, but also apparently became immune to it.

“When we tried to rechallenge those animals with a large dose of these normally-aggressive tumor cells, they were completely protected,” said Geletneky. “And we attribute that more to the immune effect than the reactivation of the virus, because that happened about a year after the treatment when we weren’t able to find any traces of remaining virus.”

How and when the viral infection stimulated such a strong immunity to the glioma is something that Geletneky and his colleagues would like to know more about, and plan to study in the clinical trials. Other candidate oncolytic virus therapies have failed in part because they provoked a swift immune reaction against themselves, thus weakening or negating their cancer-killing effect. So far it appears that H-1PV doesn’t provoke such a self-negating response.

“The balance between anti- and pro-oncolytic effects of the immune response is a subject of intense study in several labs,” says Balveen Kaur, a cancer neurologist at Ohio State University who has investigated other oncolytic virus therapies. She adds that one oncolytic virus now in late-stage clinical trials against melanoma is designed specifically to produce immunity-stimulating proteins, thus acting somewhat as a tumor vaccine in addition to being a tumor killer.

Kaur calls the H-1PV study by Geletneky and colleagues promising, but notes that the key question is “whether it will be able to persist long enough in humans to be effective.”

The forthcoming clinical trial of H-1PV, sponsored by Oryx, a Munich-based biotechnology company, should be able to answer that question. But in principle, since almost all viruses have some natural capacity to evolve, researchers should be able to engineer agents such as H-1PV to maximize their infectivity and killing effect against the specific tumor cells found in a given individual. This engineering could be done, for example, by repeatedly selecting the better-performing strains of the virus as it grows in cell cultures.

“We have some early hints that this might be an interesting strategy to follow,” says Geletneky. “If we take some glioma cells that are not so susceptible to the virus, and passage the virus on those cells, then after around ten passages we tend to get a virus that kills these glioma cells a lot better.”