“This system allows us to add another dimension to what is already personalized medicine,” says Dr. Price, who also joined Columbia in April as vice chairman for physics at Columbia’s Vagelos College of Physicians and Surgeons. “The degrees of freedom have increased significantly, because not only are we creating a unique plan for every patient, but when it’s merited, we can personalize every single treatment that’s given.”

Typically, a patient’s radiation plan is tailor-made by a collaborative team of radiation oncologists, dosimetrists, and medical physicists over the course of about a week. Before radiotherapy begins, the team takes computerized tomography (CT) images of the area of the body that needs treatment and compiles a map (known as the radiation prescription) that indicates which parts need higher doses of radiation than others. In traditional radiotherapy, this map doesn’t change, even weeks into treatment when the tumor may have shrunken or other shifts in anatomy have likely occurred.

Conversely, a small number of commercially-available “adaptive” radiotherapy systems, like Ethos™, allow clinicians to update a patient’s plan on a session-to-session basis, in real-time. So before the patient receives any radiation, Ethos™ double-checks their anatomy with a previously-acquired CT scan. And instead of having the radiation physician draw the outlines of organs and tumors, which normally take several hours to complete, Ethos™ utilizes artificial intelligence (AI) to automatically determine whether the borders of those regions have changed and create a new radiotherapy plan based on these changes.

“This new plan takes only minutes to create, and meanwhile, the patient never gets up off the treatment table,” Dr. Price says. “Then, the physician and the physicist look at this new plan and check the AI’s work. It’s hardcoded into the system that every single computer-generated region and resultant plan must be checked by a human (i.e. physician) before a treatment is delivered.”

In a recent study with Ethos™, Dr. Price and his former colleagues at Vanderbilt University School of Medicine found that the system’s updated plans improved target coverage and decreased the maximum dose to nearby organs-at-risk in patients with cervical and rectal cancers.

Other clinicians at the Herbert Irving Comprehensive Cancer Center (HICCC) look forward to harnessing this technology for their own patients and research studies, with hopes of lowering the toxicity and side effects of treatment.

While conventional radiation machines treated patients based on fixed plans from images, advancements in technology over the years have allowed for more precise targeting and a greater sparing of critical, healthy structures in the brain. The ability to deliver smaller radiation volumes have led to less adverse side effects.
With Ethos™ therapy now at his fingertips, Dr. Wang hopes to take treatment for glioblastoma a step further with the incorporation of adaptive radiation plans. He was recently awarded a three-year, multi-center research grant to investigate the use of Ethos™ therapy in conjunction with magnetic resonance imaging (MRI) for patients with glioblastoma and other brain tumors.
“Normally, we would use a static radiation plan based off the CT scan that was acquired on the day after surgery. But who is to say that the brain anatomy hasn’t changed, especially after a tumor has been resected? Sometimes the surgical cavity can collapse,” he says. “If we are able to shrink radiation volumes by adapting their plan, that may potentially help with the patient’s quality of life.”
Dr. Wang, who also serves as co-director of the Center for Radiosurgery and chair of the Quality and Patient Safety Committee in Radiation Oncology at NewYork-Presbyterian/Columbia University Irving Medical Center, plans to explore the possible benefits of adding information from advanced MRI scans prior to and during the third week of radiation therapy. The purpose of the MRI scans is to see changes in the tumor and surgical cavity and identify sites where the tumor is more (or less) likely to progress that could be targeted with a higher (or lower) radiation dose.

Pictured, clockwise from left:

Drs. Lisa Kachnic, Christine Chin, Michael Price, David Horowitz, and Michelle Savacool.

Many patients with anal cancer can be cured, especially if the disease is caught early. But individuals whose cancer has spread to lymph nodes or other areas of the body have a much higher risk of recurrence and need the highest doses of radiation for treatment.

Dr. Horowitz aims to build on the pioneering work of Lisa A. Kachnic, MD, professor and chair of radiation oncology and associate director for Cancer Network Strategy at the HICCC, who previously demonstrated that a combination of tailored radiation and chemotherapy can lead to a cure. Her innovative technique, called dose-painted intensity-modulated radiation therapy (DP-IMRT), delivers a high dose of radiation to large tumor deposits while giving a lower dose to areas at risk that have no evidence of cancer.

“The challenge [with DP-IMRT] is that we know these tumors actually shrink dramatically over the course of a patient’s treatment. So the problem that sometimes arises is that normal tissue, such as the bowel or bladder, can move from the time that we do our original planning for the radiation to when the patient is halfway through the course, for example,” he says. “That means we might be giving higher doses of radiation to normal organs than we would ideally like.”

Artificial intelligence-based planning and delivery of radiation, also known as adaptive radiation, provides a fast, simple way to make adjustments, potentially on a daily basis. Dr. Horowitz is the lead co-investigator on a multi-center research grant that Dr. Kachnic recently received to evaluate adaptive radiation for patients with anal cancer on their new Ethos™ radiation delivery machine. Dr. Horowitz believes that adaptive radiation may change the way we deliver radiation as it has the ability to spare patients from the sometimes serious side effects while also maximizing the radiation dose directly going to the tumor itself.