SU‐E‐P‐52: Dose‐Volume Toxicity Analysis of Stereotactic Body Radiation Therapy for Hepatocellular Carcinoma

Purpose:To investigate the relationship between physical and biological effective dose (BED) to liver toxicity for SBRT fractionations.Methods:A total of 16 patients (13–10×3Gy, 2–5×10Gy and 1–3×15Gy were selected. Physical dose distributions were converted to voxel based BED values using the linear‐quadratic (LQ) and linear‐quadratic linear (LQ‐L) models for doses per fraction larger than 6Gy. Patients were graded for effective toxicity (post‐treatment minus pre‐treatment grades) using the RTOG Late Radiation Morbidity Scoring Schema associated with Radiation Induced Liver Disease (RILD). Evaluated physical dose‐volume levels consisted of V10Gy, V15Gy, V20Gy, V25Gy and V30Gy which were then converted to BED values corresponding to V16.7Gy3, V30Gy3, V46.7Gy3, V66.7Gy3 and V90Gy3, respectively. All levels were normalized to their respective patient normal liver volumes (NLV) and evaluated for correlation to RILD. Results were measured quantitatively using R‐squared regression analysis.Results:Mean Dose Tolerable to Normal Liver (MDTNL) against RILD grade resulted in an R‐squared value of 0.0104. NLV‐normalized physical dose linear regression fit of V10Gy, V15Gy, V20Gy, V25Gy and V30Gy against RILD yielded R‐squared values of 0.1041, 0.0895, 0.0698, 0.0398 and 0.0009 while BED levels of V16.7Gy3, V30Gy3, V46.7Gy3, V66.7Gy3 and V90Gy3 resulted in values of 0.0002, 0.0153, 0.0533, 0.0427 and 0.0072, respectively. The average NLV‐normalized V10Gy, V15Gy, V20Gy, V25Gy and V30Gy per grade plotted against RILD grade yielded R‐squared correlations of 0.8092, 0.6362, 0.5899, 0.5846 and 0.0224 while the BED levels of V16.7Gy3, V30Gy3, V46.7Gy3, V66.7Gy3 and V90Gy3 resulted in R‐squared correlations of 0.0003, 0.3831, 0.8476, 0.678 and 0.076, respectively.Conclusion:Regression analysis between physical dose, BED and RILD showed strong correlation for the V30Gy and V46.7Gy3 dose‐levels. Average BED and physical dose per grade both exhibit strong correlations to RILD however, a lack of statistical significance exists due to small patient sample size.