| Purpose:Primary liver carcinoma (PLC) is a major malignant disease in China.The main treatment of early PLC is surgical resection. Unfortunately, the majority of PLC cases were technically unresectable or medically inoperable when they were diagnosed.Conformal radiotherapy (CRT) technique has increasingly became the preferred paradigm in treating medically inoperable and unresectable PLC since it has a higher response rate.Radiation-induced liver disease (RILD) has been considered the most severe complication in liver irradiation. That severity of hepatic cirrhosis (Child-Pugh score) was demonstrated to be the most important predictor for RILD. The purpose of the current study was to identify a dosimetric predictor for RILD from dose-volume histograms (DVHs) in PLC patients with Child-Pugh Grade A cirrhosis when treated with CRT.Materials and Methods:Between August 2000 and December 2008,118 PLC patients with Child-Pugh Grade A cirrhosis were enrolled and treated.The median age was 46 years (range,23-79), and males predominated, with a ratio of 108 men to 10 women. Among the patients, there were 91 cases of T3 and 27 cases of T4,with neither lymph node nor distant metastase(UICC/AJCC TNM staging system,1997).Hepatitis B virus (HBV) infection was very common,100 patients were positive for HBV. Portal vein thrombosis (PVT) was detected in 25 patients by abdominal computed tomography (CT) and B ultrasonography. All patients received CRT, and 41 of them underwent transcatheter arterial chemoembolization (TACE) before irradiation. The mean value of gross tumor volume (GTV) was 386.9±316.7cm3. A median dose of 52Gy (range,40-68) was delivered to the PLC by hypofractionated CRT (three fractions per week) with a median fraction size of 4.6Gy (range,4-6). The median irradiation fraction was 12 (range,7-17 fractions) in a median total irradiation time of 4 weeks (range, 3-6 weeks).Results:10(10/118,8.47%) patients were diagnosed as RILD. No difference was found by univariate analyses in clinical factor gender, age, TACE and PVT between patients with and without RILD (p= 0.599,0.454,0.306 and 0.128), but revealed that GTV was related to the risk of developing RILD (p=0.000). Univariate analyses demonstrated dosimetric factor V5, V10, V15, V2oand V25 were statistically significant factors associated with the development of RILD(p= 0.03,0.01,0.01,0.00 and 0.04).Multivariate analyses demonstrated GTV and V2o were independent predictors(p= 0.007 and 0.019, respectively). With V20 of 48.5% as the tolerance, the prediction of RILD achieved high accuracy 0.754 (89/118), with a sensitivity of 0.9 (9/10) and a specificity of 0.741 (80/108). Conclusions:The risk factors for occurrence of RILD were the GTV and V2o for PLC patients with Child-Pugh grade A cirrhosis treated with hypofractionated CRT. V20 may be a useful DVHs parameter in predict the risk of RILD for our individual patient treatment. Purpose:Radiation-induced liver disease (RILD) was the most severe radiation-induced complication for primary liver carcinoma (PLC) patients received conformal radiotherapy (CRT). The purpose of the current study was to describe the probability of radiation-induced liver disease (RILD) using Lyman normal tissue complication probability (NTCP) model, we also estimate a/βfor normal liver tissue based on our clinical data of hypofractionated irradiation by the normalization of linear quadratic (LQ) model.Materials and Methods:Between August 2000 and December 2008,118 PLC patients with Child-Pugh Grade A cirrhosis were enrolled and treated.The median age was 46 years (range,23-79), and males predominated, with a ratio of 108 men to 10 women. Among the patients,there were 91 cases of T3 and 27 cases of T4,with neither lymph node nor distant metastase(UICC/AJCC TNM staging system,1997).Hepatitis B virus (HBV) infection was very common,100 patients were positive for HBV. Portal vein thrombosis (PVT) was detected in 25 patients by abdominal computed tomography (CT) and B ultrasonography. All patients received CRT, and 41 of them underwent transcatheter arterial chemoembolization (TACE) befor irradiation. The mean value of gross tumor volume (GTV) was 386.9±316.7 cm3. A median dose of 52Gy (range, 40-68) was delivered to the PLC by hypofractionated CRT (three fractions per week) with a median fraction size of 4.6Gy (range,4-6). The median irradiation fractions were 12 (range, 7-17 fractions) in a median total irradiation time of 4 weeks (range, 3-6 weeks). A maximal likelihood analysis yielded best estimates for the Lyman-Kutcher-Burman (LKB) NTCP model parameters. The following expression based on the linear quadratic (L-Q) model was used to normalize the dose distribution of a dose per fraction d to a reference dose per fraction d ref:(?), where assumingα/βwas 0.5,1.0,1.5,2.0, 3.0,5.0,10.0 Gy for normal tissue, individually.Results:Of 118 patients treated with CRT,10 developed RILD. The unnormalized LKB model parameters were n= 1.0, m=0.20, tolerance dose for 50% complication risk for whole organ irradiated uniformly TD50(1)=33.6 Gy. The LKB NTCP model fit the complication data for the entire group. By combining the previous authoritative studies result of TD50 is about 40～50Gy, we estimatedα/βwas about 1.5～3 Gy. Conclusion:The Lyman NTCP model for RILD is recommended to predict RILD in PLC patients treated by hypofractionated CRT.It is the first study for us to estimateα/βvalue of human liver from clinical data. The fitting data may be used to design different dose fractionation schemes for liver irradiation. Future work still needs to evaluate the available toxicity data and determine theα/βfor liver. |
PDF Full Text Request