University of California, San Diego researchers say they have shown one way deadly brain tumors called gliomas evade drugs aimed at blocking the epidermal growth factor receptor (EGFR), a cell signaling protein that is crucial for tumor growth. They also say that a particular EGFR mutation is important not only to initiate the tumor, but for its continued growth as well. The findings appear this week in PNAS.

In aggressive gliomas, extra copies of the EGFR gene are produced, and half of such tumors also carry an EGFR mutation, which ramps up tumor growth and portends a poor prognosis. Clinical trials of anti-EGFR agents have been disappointing; brain tumors may respond initially, but later become resistant to the drugs. To better understand why, the researchers wanted to find out if the mutant EGFR was needed by tumors for their continued growth.

The team created a genetic system in mice in which they could control the expression of mutated EGFR, turning it off and on with the drug tetracycline. They found that the tumors' growth would stop for a period of time when tetracycline blocked EGFR, much like what is seen in patients who respond to EGFR inhibitors. But the tumors would start to grow again, even without EGFR, meaning something else was driving tumor growth.

The researchers examined individual tumors that had sidestepped or "escaped" the need for mutant EGFR to sustain their growth. In some cases, tumors that would normally have killed mice in 20 days were stable for months with the blocked expression of mutant EGFR. The scientists used microarray technology to test for genes that had not been previously expressed in the tumors but were now overexpressed in tumors that no longer required EGFR. They finally found one, KLHDC8 which, when inhibited, halted tumor growth.

"That finding makes us think that this gene would be a reasonable target," Cavenee said. "About half of the individual tumors that didn't need mutant EGFR to grow expressed that gene and, if we silenced the gene, those tumors did not grow."

Cavenee thinks this could be a model for the behavior of other tumors. "If the tumors use the same strategy to get around receptor inhibitors, then targeting that alternate pathway plus the receptor up front should give a longer response because it's hitting the primary event plus the escape route," he said.

Now the research team is searching for other genes expressed in tumors that can escape EGFR dependence, and looking for biological pathways that might be involved.