GARD is a pan-cancer predictor of clinical outcome in radiation treated patients.

BackgroundDespite advances in cancer genomics, radiation therapy (RT) is still prescribed based on an empiric one-size-fits-all paradigm. Previously, we proposed a novel algorithm using the genomic adjusted radiation dose (GARD) to personalize RT prescription dose based on the biological effect of a given physical RT dose, calculated using individual tumor genomics. We hypothesize that GARD will reveal interpatient heterogeneity associated with opportunities to improve outcomes compared to physical RT dose alone. To test this hypothesis, and the GARD-based RT dosing paradigm, we performed a pooled pan-cancer analysis in 11 separate clinical cohorts of 1,615 unique patients with 7 different cancer types that represent all available cohorts with the data required to calculate GARD, together with clinical outcome. MethodsUsing 11 previously-published datasets of cancers including breast, head and neck, non-small cell lung, pancreas, endometrium, melanoma and glioma, we defined two clinical endpoints: (i) time to first recurrence and (ii) overall survival, comprising 1,298 (982 +RT, 316 -RT) and 677 patients (424 +RT, 253 -RT), respectively. We used Cox regression stratified by cohort to test association between GARD and outcome with separate models using RT dose and sham-GARD for comparison. Interaction tests between GARD and treatment (+/- RT) were performed using the Wald statistic. ResultsPooled analysis of all available data reveal that GARD as a continuous variable is associated with recurrence (HR = 0.982, CI [0.970, 0.994], p = 0.002) and survival (HR = 0.970, CI [0.953, 0.988], p = 0.001). The interaction test revealed the effect of GARD on survival depends on whether or not that patient received RT (Wald statistic: p=0.011). Physical RT dose and sham-GARD were not significantly associated with either outcome. ConclusionsThe biologic effect of radiation therapy, as quantified by GARD, is significantly associated with recurrence and survival for those patients treated with radiation: it is predictive of RT benefit; and physical RT dose is not. We propose integration of genomics into radiation dosing decisions, using a GARD-based framework, as the new paradigm for personalizing RT prescription dose.