By Keridwen Cornelius
3D printing and augmented reality help Mayo Clinic doctors treat brain tumors
A giant arteriovenous malformation floats in the room like a tentacled monster from 20,000 Leagues Under the Sea. But unlike a squid, this holographic image doesn’t squirm as I circle around it, examining it closely from every side. With a pinching motion, I can even shrink it, magnify it, or flip it over.
Near this monstrosity are wiggly silicone aneurysms, plastic vascular systems that resemble coral, and cancerous sandstone brains. They’re all part of Mayo Clinic’s Precision Neurotherapeutics Innovation (PNT) Program, which is using 3D printing and augmented reality to help surgeons practice on models of each patient’s anatomy, create personalized treatment plans, and save lives.
Typically, surgeons rely on a two-dimensional MRI, CT or PET scan. From this, they imagine what the three-dimensional structure they’ll encounter in the operating room looks like. 3D printing lets surgeons hold an exact replica of each patient’s aneurysm, arteriovenous malformation or tumor and rehearse surgical strategies to determine which might be most successful.
“From the surgical perspective, they love it because they love being able to know beforehand what they’re going to deal with and have practiced it, and that’s impactful to their navigation of care,” says Dr. Kristin Swanson, co-director of the PNT program. This technology can also help ease the patient’s fears of the unknown, because surgeons can show them their own 3D-printed aneurysm and have them simulate the operation.
With augmented reality, the surgeon wears a visor called a hololens that allows them to see everything as normal, plus a holographic image. This also helps doctors strategize before surgery. But the eventual goal is for surgeons to use this technology while operating, so they won’t have to look back and forth between the patient and an MRI image on a monitor. All that information would be within their visual field.
To improve outcomes with brain tumors, Swanson and her team in the PNT’s Mathematical Neuro-Oncology Laboratory are marrying math with oncology. Swanson holds up a 3D-printed brain of a patient with glioblastoma, the cancer that killed John McCain. A rainbow of colors illustrates the distribution of cancer cells. Dark red represents the most concentrated part of the tumor, while orange, yellow, green and blue indicate progressively more diffuse cancer cells.
These cells are not connected. Brain tumors do not resemble tulip bulbs, with a central mass sprouting tendrils. Rather, Swanson explains, “imagine you have a handful of grains of sand. You pour them on the table, so you see some of them roll away. That’s the gradient in tumor cell density. Now imagine those grains of sand are cockroaches, and they can move.”
Brain cancer cells migrate, typically preferring areas with more oxygen and blood. Mathematicians can use equations to model migration in everything from wolves in Yellowstone to tumor cells. They can also make mathematical models of tumor growth. In addition, they use two types of MRI; one creates a picture of the tumor’s dark red region, and the other gives an approximation of the yellow region. They fill in the density gradient between those areas using math. From that, they create a 3D picture of each brain tumor.
In the one Swanson is holding, the rainbow spreads almost entirely across the brain. “Patients that look like this are not good candidates for big surgeries,” she says. “You would still be leaving a ton of [cancer] behind. But they are great candidates for chemotherapy.” If, however, the math indicated this diffuse tumor was rapidly growing, radiation would be most effective. On the other hand, if the tumor is very concentrated, with tight rainbow bands and the blue area close to the red, surgery can be highly successful, she says. “We’ve proven that if the blue is close to the red and you do a surgical approach using this information, you can basically double the patient’s survival.”
The lab is also developing ways of using artificial intelligence and math to more accurately predict which genes are broken in different parts of a tumor, as well as which patients might respond best to various immunotherapies, which harness the power of the immune system. The possibilities even amaze Swanson. “The stuff that we’ve got going on here,” she says, “is just literally blowing my mind.”