Share story

A gene-therapy experiment at Fred Hutchinson Cancer Research Center only involved a handful of brain-tumor patients, and on average, extended their lives by months, not years.

Even so, it was the first real progress in 30 years for patients with glioblastoma, the most common and most aggressive type of primary brain tumor — the type that killed U.S. Sen. Edward Kennedy within 15 months of diagnosis.

“I think this is actually one of those proof-of-concept milestones,” said Dr. Stanton Gerson, director of the Case Comprehensive Cancer Center at Case Western Reserve University in Cleveland, who was not involved in the study. “This is the very first clinical validation that all that science made sense.”

The new approach, led by Dr. Hans-Peter Kiem and Dr. Jennifer Adair at Fred Hutch in Seattle, was published Friday in The Journal of Clinical Investigation.

It began with the usual therapy for such tumors — powerful chemotherapy combined with a drug that disables a protein that makes some of these tumors particularly resistant to chemotherapy. More than half the patients with glioblastomas, including all seven patients enrolled in the study, have such a protein, Kiem said.

The protein-disabling drug, benzylguanine, is critically important because it allows chemotherapy to attack the tumor. But the drug also damages bone marrow, killing blood cells so people are left vulnerable to infection and bleeding, he said. For that reason, patients typically can receive only one or two cycles of chemotherapy.

The gene-therapy approach involved taking the patients’ stem cells and engineering them to become resistant to benzylguanine, so their blood cells weren’t damaged by the drug. When the stem cells were returned to the patients, their blood was protected but their tumors were left vulnerable to the chemotherapy.

Better protected against infection and bleeding, the seven patients in the study were able to receive more cycles of chemotherapy.

“We can sensitize the tumor, while the blood cells are resistant,” Kiem said. “That is the trick.”

Typical median survival for glioblastoma patients with the tumor-protecting protein is less than 13 months. The patients in this study, on average, survived 20 months, and all survived beyond one year. “This is quite remarkable,” he said.

One patient underwent nine cycles of chemotherapy, and is disease-free after five years, Kiem said, and three patients survived two years.

“There is really no other treatment that even has any promise,” Kiem added.

Usually, glioblastoma patients get a couple cycles of the most effective chemotherapy, temozolomide, and then have to switch to a different chemotherapy that doesn’t work as well.

Glioblastomas, which affect from 12,000 to 14,000 new patients each year in the U.S., have been a particularly vexing problem for cancer specialists.

“We’re better at biopsies, imaging and surgery, but we haven’t affected (patient) survival,” Gerson said.

“This is the first application where they showed the ability to gene transfer, the ability of the stem cells to be selected, the ability of the bone marrow to be protected, and the benefit to patients. That’s a lot,” said Gerson, who has done early work with the same technique, but without testing on patients.

“It is an unvalidated proof-of-concept with a remarkable early hint of perhaps a significant benefit.”

The researchers said they added a step to the treatment by giving patients a drug that helped their bodies accept and use the gene-modified blood cells as well as treat residual brain tumor.

Both Kiem and Gerson were cautiously optimistic that the results of this study could be replicated in a larger group, and that this approach might eventually help patients with other types of brain tumors — perhaps even some other types of solid tumors.

For the next step, Kiem’s group will be opening another trial in about a month, planning to look at whether they can safely increase the chemotherapy dose. “We need a larger study to prove this is better,” he said. Ideally, the protocol would eventually be tested in a randomized group of patients.

Despite this study’s small size, Kiem said his team is encouraged by the results.

“I think if we study more patients with this approach, I’m quite optimistic we’ll be able to get more long-term survivors.”

Carol M. Ostrom: 206-464-2249 or On Twitter @costrom