Genetic testing can identify patients with HPV-positive throat cancer who may benefit from lower radiation doses, according to Cleveland Clinic research. The study, published in the Journal of Clinical Investigation , builds on a growing body of evidence that radiation treatment can be personalized using tumor genomics, potentially shifting treatment approaches from the norm, where radiation is prescribed at a uniform dose, to one called Genomic Adjusted Radiation Dose (GARD), where radiation is prescribed to a desired effect.
The current standard radiation dose for HPV-positive throat cancer is 70 Grays (Gy), which offers cure rates between 80% and 95% but can cause long lasting side effects like problems with swallowing and breathing. So far, attempts to lower radiation doses (for example, to 60Gy) have failed in clinical trials. have been no proven strategies for de-escalating the dose,
"There's been a feeling in the field that we're stuck," says study lead author Jacob Scott, MD, DPhil, a radiation oncologist at Cleveland Clinic. "All the evidence tells us lower doses should work, but clinical trials haven't been able to prove it. We wanted to ask: could the missing piece be genetics?"
They turned to the genomic-adjusted radiation dose (GARD) model, which Dr. Scott had developed in collaboration with Moffitt Cancer Center radiation oncologist Javier Torres-Roca, MD .
GARD uses tumor gene expression to calculate the minimum radiation dose required to control cancer. Unlike models that rely only on clinical features such as tumor size or smoking history, GARD integrates genomic data from ten radiosensitivity genes to predict patient-specific response.
Dr. Scott and Dr. Torres-Roca previously validated GARD across multiple cancer types . To see whether GARD could be used in HPV+ head and neck cancer, Drs. Scott and Torres-Roca teamed up with internationally renowned head and neck oncologist Lisa Licitra, MD, from the Fondazione IRCCS Istituto Nazionale dei Tumori in Milan, Italy. Dr. Licitra and her team were the driving forces behind the Big Data to Decide Project , the world's largest head and neck cancer patient database.
Analyzing data from 191 patients in the database confirmed that higher GARD scores were associated with improved survival outcomes, even when patients received the same radiation dose. The researchers then applied GARD retrospectively to participants from an unsuccessful 2024 clinical trial that tested 60 Gy instead of 70 Gy. While overall survival was slightly lower at 60 Gy (96–98% vs. 99%), GARD analysis revealed that about 22% of patients would likely have maintained excellent outcomes at a lower personalized dose.
"This is critical context for planning the next wave of clinical trials," Dr. Licitra says. "It shows that genetics can help us select the right patients for reduced doses, which is something we couldn't do before."
"This work builds directly on nearly two decades of research into radiosensitivity and genomics," adds Dr. Torres-Roca. "We've shown that integrating genomics into radiation oncology is not only feasible, but essential if we are to move past the limitations of one-size-fits-all dosing."
The team hopes these findings will pave the way for new clinical studies that integrate GARD into decision-making before treatment begins.
"We already have open trials using GARD in other cancers," Dr. Scott says. "To our knowledge, this is one of only two approaches that have successfully lowered radiation dose for patients. We're close to reaching the holy grail in radiation oncology, moving to truly personalized treatment."
