Comparison And Selection Of Dosimetric Parameters Associated With Acute Radiation-induced Pulmonary Injury

1.Comparison of the predictive abilities of dosimetric parameters associated with acute symptomatic radiation-induced pulmonary injuryPurpose:To assess the utility of dosimetric parameters as predictors of acute symptomatic radiation-induced pulmonary injury and to identify the optimal metrics.Methods:Between November 2002 and August 2005,91 patients were enrolled onto this study.There were 81 males and 10 females with a median age of 56 years(ranged 19-75 years).Patients received neoadjuvant platinum-based chemotherapy first.The medium pre-RT cycle was one.Chemo-regimen induced NP,GP,IEP and so on. Radiation dose ranged from 30-71 Gy.Cad-plan TM(version 6.08) and Pinnacle3 TM (version 7.0-7.4) treatment planning system were applied to generate Maximum lung dose(Dmax),mean lung dose(Dmean),V5 to V50 which is the percent of pulmonary volume received at least 10 to 55 Gy.Equivalent uniform dose(EUD) and normal tissue complication probability(NTCP) was calculated by Matlab(version R2007a). Radiation-induced pulmonary injury was graded according to Common Terminology Criteria for Adverse events 3.0(CTCAE3.0).Student's t test,Wilcoxon rank sum test, chi-square test and the logistic regression test were used to make univariate and multivariate analysis.The predictive abilities of models(sensitivity and specificity) were calculated and compared based on the area beneath receiver operating characteristic(ROC) curves.Results:Thirty-one of 91(34%) patients developed CTCAE3.0 grade≥2 acute pulmonary injury.Ten developed grade 2 pneumonitis,20 grade 3,1 grade 4.There was no grade 5 toxicity.On univariate analysis,except for Dmax and V5,dosimetric parameters were significantly relative to grade≥2 acute pulmonary injury.The single best predictor was V20 with the area under ROC curve 0.69.The cut-off was 23%with sensitivity of 80.65%and specificity of 58.33%.NTCP and EUD,with the largest area under ROC curve 0.67,were not superior to V20.On logistic regression analysis,only V20 remained significantly correlated with grade≥2 acute pulmonary injury.However,V20 was marginally associated with grade≥3 acute pulmonary injury in univariate analysis(p=0.056).But patients with grade≥3 acute pulmonary injury were much older and had worse pre-RT pulmonary function tests(PFTs) than those who without severe comlications.Conclusion:Dosimetric parameters were significant correlated with acute symptomatic radiation-induced pulmonary injury(CTCAE3.0≥2) and had good negative predictive abilities.The single best predictor for CTCAE3.0 grade≥2 acute pneumonitis was V20.Dosimetirc parameters was also associated with severe acute pulmonary injury(CTCAE3.0≥3) and had negative predictive abilities.Elderly patients(≥65 years) and patients with poor pre-RT PFTs(FEV1＜1.9L) were more likely to have severe acute pulmonary injury. 2.Analysis of the efficacy and safety of three-dimensional conformal radiotherapy for the treatment of non-small cell lung cancer after pneumonectomyPurpose:To analyze the efficacy and safety of three-dimensional conformal radiotherapy(3DCRT) for the treatment of non-small cell lung cancer(NSCLC) after pneumonectomy,and to provide the reference standard for the clinical best treatment plan.To assess the predictive ability of dosimetric paramters in those patients.Methods:We were reviewed retrospectively 17 patients undergoing 3DCRT after pneumonectomy for lung cancer from February 2002 to November 2008.There were 16 males and 1 females with a median age of 55 years(ranged 40-70 years).Five patients had stageⅢa disease,eleven had stageⅢb and one had stageⅣ.Thirteen patients received platinum-based chemotherapy prior or posterior to radiotherapy. Chemo-regimen induced NP,EP and TP.Fourteen patients received conventional radiotherapy with 30-66Gy/15-33Fx/21-49d,other three patients with 45Gy/15Fx/22d, 55Gy/22Fx/33d and 64Gy/24Fx/31d.Complications were graded according to Common Terminology Criteria for Adverse events 3.0(CTCAE3.0).The predictive abilities of dosimetric parameters were assessed referring to the cut-off point mentioned in the first part of this paper.The statistical analysis was performed with Stata 10 for windows software system and estimated the survival rate of patients using Kaplan-Meier survival anlysis method.Results:Median follow-up was 20.6 months(ranged 1.6-74).1-,2-,3- year survival rate was 74.3%,65.0%and 52.0%.Five patients had treatment-related complications: one had a wound crack and died of it(CTCAE3.0 grade 5),one had fistula after radiotherapy(CTCAE3.0 grade 3),two had acute radiation-induced pulmonary injury(CTCAE3.0 grade 1-2) and the other had a dyspnea on exertion(CTCAE3.0 grade 1).Because of lung dose was limited to a low scope (under the cut-off point),only one patient developed acute symptomatic radiation-induced pulmonary injury and no severe acute pulmonary injury occurred.Conclusion:When treating patients after pneumonectomy,a reasonable plan can avoid severe pulmonary injury.But there were still some severe treatment-related complications.So we should make a comprehensive assessment of patients,including pulmonary function tests,when treating patients after pneumonectomy.There should be a certain time interval between surgery and radiotherapy.