The EGFR pathway is common to a variety of cancers, and a project led by researchers at the University of California, Los Angeles, is working out how to find patients with glioblastomas who will respond to EGFR inhibitors, such as Tarceva, a product of OSI Pharmaceuticals, Pfizer, and Genentech; and Iressa, made by AstraZeneca.
In a sample of tumors from 26 patients who had received therapy with either Iressa or Tarceva, "it was very clear that patients whose tumors co-expressed two proteins, EGFRvIII and Pten, were 51 times more likely to respond," Paul Mischel, an associate professor in UCLA's David Geffen School of Medicine and the Jonsson Comprehensive Cancer Center, told Pharmacogenomics Reporter. "It was a dramatic enrichment for response, and [patients were] very unlikely to have treatment failure when those proteins were present," he said.
The epidermal growth-factor receptor is amplified, over-expressed, or mutated in as many as half of glioblastoma cases, according to some studies, Mischel said. However, "in small and early clinical trials, it was hard [for investigators] to fully document [drug] effect and figure out which individual patients were likely to benefit" from treatments targeting EGFR, he said.
After ruling out whether response to EGFR-inhibiting drugs correlated to either an EGFR gene-dose effect or EGFR tyrosine kinase-domain mutations, the team hit upon a more promising factor. "Much of our work had suggested that [the] mutant EGF receptor called EGFRvIII — which is not like the kinase-domain mutations in its structure, but may have some similar effect — doesn't bind [a] ligand; it sits on the cell surface sending a chronic signal," and appeared to be a promising target, Mischel said.
The other important marker, Pten, is a phosphatase that works to keep the PI3-kinase signaling pathway in check, the same pathway through which EGFRvIII sends signals. "If you block the [EGF] receptor, you can inhibit the signaling of that receptor," said Mischel. "But if Pten is gone, that becomes dissociated from the downstream signaling — so in other words, you can shut off the receptor, but if you don't have Pten, that doesn't turn off the whole pathway downstream."
In a validation experiment with samples from a study conducted at the University of California, San Francisco, the researchers tested tissue from patients who had been treated with Tarceva. "We analyzed Pten and EGFRvIII in those samples and found exactly the same thing — that there was a dramatic enrichment for Pten and EGFRvIII co-expression in those patients who responded to the drug," Mischel said.
The group is currently involved in planning prospective clinical trials, but they are hoping that an independent group performs prospective validation. Mischel declined to name which drug manufacturers are involved in planning clinical trials. Also important is the development of clinical diagnostic tests. "We're in the earliest stages — very brief discussions" with diagnostic companies and other scientists within UCLA about how to approach test development, Mischel said.
"Glioblastoma is the most common malignant primary brain tumor for adults, and it's a very challenging disease — patients have a median survival of a year or less [after diagnosis] despite surgery, radiation, and chemotherapy," said Mischel. In the US, nearly 10,000 glioblastomas are diagnosed annually, he said.
— Chris Womack ([email protected])