There is yet no straightforward way to determine the optimal dose level and treatment schedules for high-dose radiation therapies such as stereotactic radiation therapy, used to treat brain and lung cancer, or for high-dose brachytherapy for prostate and other cancers.
Radiation oncologists at the Ohio State University Comprehensive Cancer Center-Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (OSUCCC-James) may have solved the problem by developing a new mathematical model that encompasses all dose levels.
Typically, radiation therapy for cancer is given in daily, low doses spread over many weeks. Oncologists often calculate the schedules for these fractionated, low-dose treatment courses using a mathematical model called the linear-quadratic (LQ) Model. The same calculation model is used to evaluate radiation response, interpret clinical data and guide clinical trials.
"Unfortunately the LQ Model doesn't work well for high-dose radiation therapy," says co-author Dr. Nina Mayr, professor of radiation oncology at the OSUCCC-James. "Our study resolves this problem by modifying the current method to develop the Generalized LQ (gLQ) Model that covers all dose levels and schedules."
If verified clinically, the Generalized gLQ Model could guide the planning of dose and schedules needed for the newer radiosurgery and stereotactic radiation therapy and high-dose brachytherapy procedures that are increasingly used for cancer patients, she says.
"Developing proper radiation dose schedules for these promising high-dose treatments is very challenging," Mayr says. "Typically, it involves phase I dose-finding studies and a long, cumbersome process that allows only gradual progression from the pre-clinical and clinical trial stages to broader clinical practice."
The new gLQ Model could allow oncologists to design radiation dose schedules more efficiently, help researchers conduct clinical trials for specific cancers more quickly and make these high-dose therapies available to cancer patients much sooner, Mayr says.