Radiation therapy (RT)--a standard component in the treatment of cancer--is typically administered in a uniform "one-size-fits-all" in which all patients receive the same uniform dose of RT.1 However, Javier F. Torres-Roca, MD, Senior Member in the Department of Radiation Oncology at Moffitt Cancer Center, and his team are in the process of changing this paradigm and establish a true genomic approach to personalize individual radiation dose.
Dr. Torres-Roca explains that “the current clinical approach is to treat all patients with a given cancer with the same dose of RT. For example a standard dose utilized in breast cancer is 50 Gy delivered in 25 sessions or fractions. RT doses in radiation oncology have been developed empirically, through trial and error, using a populational-based approach."
In 2017, Moffitt's Radiation Oncology Program began utilizing Cvergenx's Genomic-Adjusted Radiation Dose (GARD) model. According to Dr. Torres-Roca, "GARD is the first opportunity for a genomically-driven personalized approach in radiation oncology, and according to the Lancet Oncology Commission, is a research priority for the field.” 1-2
GARD, co-developed by Dr. Torres-Roca and Jacob G. Scott, MD, Associate Staff in the Departments of Translational Hematology and Oncology Research and Radiation Oncology at the Cleveland Clinic, is the first validated model that can predict the therapeutic effects of radiotherapy and guide radiation dose based on an individual patient's radiosensitivity. It works by combining two algorithms--the linear-quadratic (LQ) model, which helps identify dosing strategies, and the radiosensitivity index (RSI), which evaluates expression levels of 10 different genes that have been found to predict tumor radiosensitivity. The RSI was developed by Dr. Torres-Roca and Steven A. Eschrich, MD, Senior Member in the Department of Bioinformatics at Moffitt. Together, the LQ model and RSI data result in GARD scoring.
"Our research has found that GARD values are lower for those tumors that are resistant to radiation and higher for those tumors that are sensitive to radiation treatment." Therefore, patients with high GARD values have the likelihood of a positive therapeutic response to radiation therapy. Lower GARD values indicate resistance to radiation. Importantly, GARD scores have been determined in over 8,000 treatment patients and independently clinically validated in over 600 patients including breast, lung, glioma, and pancreas cancer patients.1
Utilizing GARD to personalize adjuvant radiotherapy in triple negative breast cancer management3
This year, Moffitt researchers validated the use of the GARD model in a recent clinical study [PDF] that examined two separate groups of triple-negative (TN) breast cancer patients treated with radiation therapy. The primary outcome of the analysis revealed that GARD was a significant predictor of local control in TNBC patients in both cohorts: Cohort 1 from Europe (N=58) and Cohort 2 from the prospective observational protocol at Moffitt Cancer Center (N=55). Additionally, in Cohort 2, researchers demonstrated that GARD could be utilized to personalize radiation doses based on individual patient tumor biology--suggesting the current "one-size-fits all" treatment approach is sub-optimal for a significant proportion of patients.3
"Our analyses provide the first proposed range for optimal radiotherapy dose at an individual patient level in triple-negative breast cancer and proposes a significant number of patients can be treated with lower doses of radiotherapy while still maintaining high levels of local control," said Kamran Ahmed, MD, Assistant Member of the Department of Radiation Oncology at Moffitt and lead author of the study.
As the study findings suggest, GARD can identify both patients who can be treated with lower radiotherapy doses while still maintaining the same local control, as well as patients who would benefit from dose escalation to improve local control. The authors conclude that as oncologists move towards an era of oncologic personalization, radiation oncologists can consider models such as GARD to tailor RT doses.1
Developing a new paradigm in radiation oncology
GARD is emerging as a novel radiation dose prescription model, providing the first opportunity to optimize and personalize individual radiation doses to match an individual tumor's radiosensitivity. Researchers from the Radiation Oncology Program at Moffitt are currently planning a new clinical trial where the radiation dose for breast cancer patients will be selected based on the GARD model. This trial, which will take place at Moffitt Cancer Center, will be the first time that the genomics of each individual patient will be utilized to make a RT dose decision.
"Utilizing radiosensitivity and genomic-adjusted dosage will reduce unnecessary radiation treatment, as well as fine tune more effective treatment for those cancer patients who stand to benefit," said Louis B. Harrison, MD, FASTRO, chair of the Department of Radiation Oncology at Moffitt.
"Genomics is an essential ingredient of our multi-faceted strategy to personalize radiation therapy. By taking a more tailored approach to the treatment of each individual patient, we will be able to improve oncologic outcomes and reduce health care costs."
Scott JG, Berglund A, Schell MJ, Mihaylov I, Fulp WJ, Yue B, Welsh E, Caudell JJ, Ahmed K, Strom TS, Mellon E, Venkat P, Johnstone P, Foekens J, Lee J, Moros E, Dalton WS, Eschrich SA, McLeod H, Harrison LB, Torres-Roca JF. A genome-based model for adjusting radiotherapy dose (GARD): a retrospective, cohort-based study Lancet Oncol. 2017 Feb;18(2):202-211. doi: 10.1016/S1470-2045(16)30648-9. Epub 2016 Dec 18. PMID: 27993569. IF: 33.9
Jaffee EM, Van Dang C, Agus DB, Alexander BM etal. Future cancer research priorities in the USA: a Lancet Oncology Commission. Lancet Oncol, 2017 Nov;18(11):e653-706
Ahmed KA, Liveringhouse CL, Mills MN, Figura NB, et al, Utilizing the genomically adjusted radiation dose (GARD) to personalize adjuvant radiotherapy in triple negative breast cancer management. EBioMedicine 2019;47:163-169. DOI: 10.1016/j.ebiom.2019.08.019.