An expanded meaning for "early detection" emerged at the 101st annual meeting for the American Association for Cancer Research in April. Namely, detecting cancer earlier no longer means simply diagnosing patients before they reach late-stage disease — rather, it means catching oncogenic changes years before the first symptom appears.
"What we're looking at is evolving over decades, and yet we're not doing too good a job at understanding what's going on over those several decades," Sanjiv Sam Gambhir, director of the Canary Center for Early Detection at Stanford University, told AACR attendees during a discussion moderated by National Cancer Institute Deputy Director Anna Barker. "Instead, we're facing the problem much too late."
To combat the issue, Gambhir and his colleagues at the Canary Center are bringing two disparate diagnostic teams together — those who traditionally perform in vitro diagnostics examining blood-based biomarkers and those who screen for markers in vivo with molecular imaging techniques. "They're really two sides of the same coin," Gambhir said. "If these two groups start to really talk together, we think the biomarker problem — both in imaging and in blood-based biomarkers — becomes much more solvable."
When screening for blood-based biomarkers — such as proteins, microRNAs, DNA, or circulating cells, among others — researchers can only ascertain a downstream perspective. "It's like being on a freeway, and seeing which cars are driving by," Gambhir said. While researchers are at an advantage to screen for multiple markers using this approach, he said, "you're not able to get to the site of the tumor. ... You're at the mercy of what's going down the freeway."
Conversely, with molecular imaging tactics, clinicians can determine the means through which specific molecules make their way to the tumor site. The downside is that the method is less amenable to multiplexing — it's only possible to monitor a few targets at once, and "you have to be very careful about what few things you interrogate," Gambhir said.
Working with several collaborators within the Canary network, he said that the amalgamation of in vitro and in vivo teams "is proving to be a useful marriage that's leading to new solutions." One such solution hinges on the novel use of a standby clinical tool: ultrasound. "The concept's actually pretty simple," Gambhir said. "If you tune sound the right way — not my voice, certainly, but ultrasound — and apply it to a very diffuse area or a focal area, you will cause a release of biomarkers."
Once sound-targeted biomarkers are liberated from the tumor cells, they transiently decrease within the bloodstream as they're degraded, metabolized, or filtered, he said, enabling researchers to track their movements. In a proof-of-principle study published in PNAS last October, Gambhir and his colleagues demonstrated the efficacy of their technique by measuring carcinoembryonic antigen levels before and after the application of ultrasound in a mouse tumor xenograft model. "As it turns out, it's quite reproducible across several biomarker types," he said.
Gambhir and co-principal investigator Charles Drescher at the Fred Hutchinson Cancer Research Center are being funded by NCI to move a related technique into clinical trials. They plan to image the ovarian cancer vascular network for VEGFR-2-targeted microbubbles — gaseous particles that change their shape in response to isonation — using contrast-enhanced ultrasound, according to the project's description. "You can actually imagine a person getting a blood test, [they] themselves serving as their own control, getting a zap of sound, and then getting another blood test within a few minutes and seeing a spike-up in their blood-biomarker levels," Gambhir said at AACR.
In the realm of early cancer detection, Gambhir said that this technique could be a game-changer. "It's one way we can fundamentally change things because we don't have to [wait to] let the tumor grow over time," he said.