A good biomarker is one whose levels go up or down as a patient’s disease worsens or wanes. A great biomarker also gives key insights into disease development. A really great biomarker does both of these things and also serves as a treatment target.
With a protein called netrin-1, Edward Smith, MD, and Michael Klagsbrun, PhD, seem to have hit the trifecta. In a recent paper in Cancer Research, they report a clear relationship between urine netrin levels and medulloblastoma, the most common malignant brain tumor of children.
And show that netrin fuels the tumor’s invasion into healthy brain tissue.
And that blocking netrin may, at least in the laboratory, check the tumor’s spread.
From brain to bladder
For years, Smith—a brain surgeon and Boston Children’s Hospital’s director of pediatric cerebrovascular surgery—has been searching for ways to use urine to watch for relapses in children with brain tumors.
The search started when Smith attended a seminar by Marsha Moses, MD, PhD, director of the hospital’s Vascular Biology Program (VBP), who studies urine biomarkers for ovarian and breast cancer. Inspired, he started learning from her how to isolate proteins from urine and assess them as potential biomarkers.
In 2008, the pair showed that a panel of urine biomarkers correlated with the presence of a variety of brain tumors. But just knowing a tumor is there wasn’t enough for Smith. As a surgeon, he wanted to know what the tumor is doing.
“The biggest problem I face as a surgeon is invasion,” he says. “Invasive tumors put fingers out into the surrounding tissue, making them harder to operate on. They’re also harder to treat with chemotherapy and radiation and more prone to metastasis.”
While gene signatures can tell us something about a medulloblastoma’s behavior, reading those signatures requires having tumor tissue on hand. Gene signatures also can’t help track a brain tumor’s response to treatment—at least not without taking additional biopsies—or raise an alarm if a tumor returns. Those are things that are best detected using markers pulled from an easily collected fluid. And urine is probably the easiest, least invasive fluid of all.
To home in on urine markers that could provide intel on tumor presence and behavior, Smith teamed up with Klagsbrun. For more than 40 years, Klagsbrun has been working in the VBP to tease apart the intricate web of chemical signals influencing blood vessel growth (angiogenesis) and cancer development. He is particularly interested in semaphorins, a family of axon guidance factors that influences nerve growth during development and also seem to play a role in angiogenesis.
Netrin is also an axon guidance factor. During fetal development, netrin and semaphorins work in opposition. Netrin creates a trail of chemical breadcrumbs that young nerve fibers follow as they grow. Semaphorins block or deflect nerve growth.
“There’s a balance between them,” Klagsbrun says, “between pro-growth and anti-growth.” Or, as Smith puts it, “Netrin is like the gas, semaphorins the brakes.”
Looking at cell lines, the pair found that netrin holds a great deal of pro-growth sway over medulloblastomas. Adding netrin to medulloblastoma cell lines caused a jump in tumor cell invasion while also activating a signaling pathway known to be involved in cancer development (thus making netrin a potential therapeutic target.) At the same time, blocking netrin using a variety of techniques (e.g., receptor blockade, siRNA, neutralizing antibodies and small molecules) reduced invasion.
Nimbler tumor screening
Turning to patient samples, Smith and Klagsbrun found that the tumor cells of one patient produced significantly more netrin than adjacent, normal brain cells. They also found that urine from 16 medulloblastoma patients contained between five and 10 times as much netrin as urine from 12 healthy controls, and that urine netrin levels drop after a patient undergoes surgery.
As a bonus, urine levels alone could single out patients with highly invasive tumors, the ones that had spread more aggressively.
“You can’t always see on an MRI if a child will have an invasive tumor,” Smith notes. “But here we saw a gradient of netrin levels that increased as we looked from healthy controls across to high-risk patients.”
For the moment, Smith envisions cheaper, measurable and more frequent screening of medulloblastoma patients after treatment.
“Measuring urine netrin could help guide decisions about when a patient might need a scan and additional treatment,” he says. “The ‘pie-in-the-sky’ scenario would be to use urine biomarkers like netrin to tailor a patient’s treatment regimen. Because urine proteins change more quickly than scans, we could be more nimble with our treatment approach.”