UC San Francisco researchers may have determined how the experimental anti-cancer drug ONYX-015 succeeds in killing cancer cells, offering a possible answer to recent questions about its mechanism of action and providing insight that could contribute to a deeper understanding of cancer biology.
The finding, reported by Frank McCormick, Ph.D., F.R.S., director of the UCSF Cancer Research Institute, at the American Association for Cancer Research (AACR) annual meeting in Philadelphia, could solidify and improve efforts to use the drug, now in phase II clinical trials. It could also lead to modified approaches for killing cancer cells with tumor-specific viruses.
ONYX-015 has shown promising results in phase II clinical trials of patients with head and neck cancer (to be reported by Onyx Pharmaceuticals, Inc. at AACR), and phase II trials have opened for patients with colon, pancreatic and ovarian cancer. (Phase II trials are small-scale studies designed to determined effective dosage.)
But recent findings in several laboratories have brought into question the presumed mechanism of action of this potent agent, a genetically engineered version of the cold virus known as adenovirus. The observations have indicated that the virus, while effective in its manipulated form against one large class of tumor cells, might also be able to kill other tumor cells, apparently by an unknown mechanism.
The observations have raised concerns that the drug could harm normal cells in humans and have presented the slight possibility that the drug could be used against a broader range of tumor cells than previously thought. More fundamentally, they have raised questions about the regulation of a particular gene, known as p53, in the pathway toward cancer.
The p53 gene plays a key role in preventing cells from slipping into uncontrolled cell growth, a hallmark of cancer. And the ability of various factors to disable the gene accounts for 60 percent of malignant tumors.
The premise of ONYX-015, conceived and designed by McCormick, is that the andenovirus could be manipulated to take advantage of a certain vulnerability that tumor cells without functioning p53 genes have, while remaining benign against normal, healthy cells, which retain functioning p53 gene.
The rationale was this: Adenovirus contains a gene called E1b that disables p53, and McCormick reasoned that by disarming the adenovirus of E1b, the virus would be unable to invade normal, healthy cells, because they would contain functional copies of the p53 gene that would prevent the virus from replicating.
But for the same reason, he surmised that the altered virus would be able to wreak havoc on cancer cells that had already disabled their p53 gene, because these cells would lack the defense mechanism to thwart the viral replication process. The genetically engineered virus would be able to replicate continuously, disrupting cell behavior, and ultimately causing cell death.
This theory has borne out, first in laboratory cell cultures and now in clinical trials. But recent laboratory studies have also demonstrated that ONYX-015 is able to replicate in tumor cells in which p53 remains intact. And this, said McCormick, "has led to speculation about the whole mechanism of action, the concern being that maybe in cancer cells there really is no correlation between viral replication and p53 status."
Now, however, building on a critical finding reported by researchers last September involving the novel p14ARF gene (Nature, Sept. 10, 1998), McCormick provides data that supports his original concept involving p53.
Specifically, he adds evidence to the earlier finding that, while some tumor cells with defects in p53 function have mutations in the gene itself, others have a defect in p14ARF, and that this defect indirectly disables p53 function.
"We have some very nice images that show that when normal cells are infected with ONYX-015, p53 is induced, and the virus is shut down. But that when tumor cells with p53 genes intact are missing p14ARF, p53 induction does not actually occur."
"We suspect," he said, "that all tumor cells susceptible to ONYX-015 have mutations in either the p53 gene or p14ARF, so that either directly or indirectly their p53 system is inactivated. This would explain why ONYX-015 can replicate in tumor cells with normal p53 genes."
"We're trying to prove this is the missing link in the lab, by putting p14ARF back into these tumor cells and seeing if it really does stop the virus from growing."
If the gene's function is proven, it could also serve as a new target for cancer therapies in some tumors lacking functional p53 genes. The ongoing research is funded by Onyx Pharmaceuticals, Inc. McCormick founded the company and is now on its scientific advisory board.