Haber and his colleagues analyzed blood samples from 27 patients with non-small cell lung cancer, 23 who had EGFR gene mutations and four who did not. CTCs were identified in all samples and in genetic analyses from mutations 92 percent of the time.
Mutations in EGFR, a protein, can help predict whether these tumors will respond to a family of drugs called tyrosine kinase inhibitors.
"Even in the three to four months that we followed patients, the genetic make-up of the tumors changed. Resistant mutations appear and other mutations appear, obviously because we're doing things [with drug therapy] to the cancer," Haber said. "But the way we practice oncology we don't typically test for that. We do one biopsy which takes a tiny, tiny amount and assume that for the rest of the course, the tumor is the same."
"It's important to know in real time what you're treating," he continued. "We need to be able to follow the patient without needing to re-biopsy the tumor every time."
The technology is in its infancy, however. "This is still in a very, very early stage where it takes a long time to handle every sample, to flow the blood through the chip," Haber said. "This is a proof of principle that we can do this. We need a much more automated system for larger clinical trials."
Dr. Len Horovitz, a pulmonary specialist at Lenox Hill Hospital in New York City, said that "you have to have some circulating cells to do this test, but it's very exciting because they're getting a genetic fingerprint of a tumor which will tell an oncologist what therapy the tumor might respond to or not respond to.
"It's expensive, but it may well be that if we can identify patients who can have a personalized regimen that works, we will be saving the cost of treating all those patients with regimens that don't work," he added.
All rights reserved