| Objective: The purpose of the study was to use CBCT scans to evaluatesetup errors in radiotherapy of thoracic esophageal cancer and evaluate theeffect of intrafraction mobility against the physical dosimeter at target andnormal tissues by a simulation of three dimensional (3D) treatment planningsystem. To use three-dimensional semiconductor detection matrix Delta4tomeasure the dose output of the accelerator and monitor the accelerator’smechanical and dose error and performance, so as to guide clinic to adjusttreatment plan.Methods and materials: Twenty-three patients with thoracicesophageal cancer of the first course radiation treatment were enrolled intothis study from December2011to September2012. All patients were fastenedwith a subpressure vacuum pad and underwent CT simulation in the supineposition. Transfering the CT image to the treatment planning system anddefining the GTV, CTV, PTV and the organs at risk (OAR). The physicistfirstly designed the treatment planing. Thereafter, the planning CT scans weretransmitted to the XVI workshop. CBCT scans were performed before andafter radiotherapy and automatically registered to planning CT scansaccording to gray value registration. Two doctors confirmed the variance ofthe osteal marks, and output a3D direction (left to right, superior to inferior,anterior to posterior) of the setup errors through the XVI software. The setuperror of pre-treatment is the date about placement before treatment. Becausethe setup error is corrected before treatment, the date about placement aftertreatment is called the intrafraction mobility. The difference of setup errorabout different parts and different treatment week were discussed. The effectsof intrafraction mobility were simulated in the treatment planning system by moving the isocenter and then the dose distribution recalculated withoutchanging the field angle, shape and weighing. PLAN1was the originaltreatment plan, CBCT scans were performed after each radiotherapy andsimulated the actual treatment process of patients if a≥3mm difference wasfound in any direction of intrafractions. two new plans was obtained. PLAN2:In this scenario, the fraction which a≥3mm difference was found in anydirection would be simulated, the prescription dose of each fraction was notchanged. On the contrary, that was should not be simulated and use thePLAN1. PLAN3: In this scenario, the actual fraction which a≥3mmdifference was found in any direction would be simulated, and the simulatedfractions were averagely distributed to the total fractions and prescription dosewere averaged, it assumed that a≥3mm difference was found in any directionof all fractions. PLAN2, PLAN3were compared with PLAN1and evaluatedthe effect of intrafraction mobility against the physical dosimeter at target andnormal tissues.19treatment plannings of IMRT were verified by usingthree-dimensional semiconductor detection matrix Delta4. The measured dosedistribution was compared with planned dose distribution by the Gamma (3mm/3%) criteria.Result:(1) The medium setup errors (cm) of pre-treatment of upper,middle and lower segment esophageal carcinoma in X directions were0.39±0.32,0.28±0.25and0.42±0.38(F=8.357, P=0.000), respectively. Thosein Y directions were0.63±0.51,0.65±0.58and0.61±0.43(F=0.147,P=0.863). Those in Z directions were0.26±0.19,0.32±0.24and0.30±0.17(F=3.699, P=0.026). The medium setup errors of pre-treatment of uppersegment and lower segment esophageal carcinoma in X directions was largerthan that of middle segment. The medium setup errors of pre-treatment ofmiddle segment esophageal carcinoma in Z directions was larger than that ofupper segment. The systematic setup errors (cm) and random setup errors (cm)of pre-treatment of upper segment esophageal carcinoma in X, Y, Z directionswere0.07,0.47,-0.15and0.50,0.66,0.28, respectively. That of middlesegment were-0.04,0.45,-0.11and0.37,0.75,0.38. That of lower segment were0.29,0.23,0.04and0.48,0.72,0.34.(2) The medium intrafractionmobility of upper, middle and lower segment esophageal carcinoma in Xdirections were0.12±0.15,0.10±0.10,0.10±0.10(F=1.995，P=0.138),respectively. Those in Y directions were0.12±0.10,0.11±0.10,0.12±0.10(F=0.423，P=0.656). Those in Z directions were0.13±0.11,0.12±0.10,0.08±0.07(F=6.089，P=0.003). The systematic errors (cm) and random errors (cm)of intrafraction of upper segment esophageal carcinoma in X, Y, Z directionswere0.03,0.03,-0.10and0.19,0.15,0.13, respectively. That of middlesegment were-0.03,0.01,-0.10and0.13,0.15,0.11. That of lower segmentwere-0.02,0.05,-0.05and0.14,0.15,0.09.(3) The margins required foradequate coverage of the CTV of upper segment esophageal carcinoma were6mm,17mm and6mm in X, Y, Z directions for treatment fractions withoutCBCT, respectively. That of middle segment were4mm,17mm and6mm.That of lower segment were11mm,11mm and4mm. The margins required foradequate coverage of the CTV of upper segment esophageal carcinoma were3mm,2mm and4mm in X, Y, Z directions for treatment fractions with CBCT,respectively. That of middle segment were2mm,2mm and4mm. That oflower segment were2mm,3mm and2mm.(4) Compare of setup errors atdifferent cycles, no significant difference was found between each group in theX and Y axis (P>0.05). The fifth week and other groups was found significantin the Z axis (P<0.05), no significant difference was found between the restgroups. The setup error of pre-treatment was larger than those in other weekssince the fifth week.(5) PLAN1compared with PLAN2. The average valuesof D50, D2in GTV were higher than those of PLAN2(P<0.05). The averagevalues of D100, D2in CTV had significant difference between PLAN1andPLAN2(P<0.05). The average values of D50in PTV had significantdifference between PLAN1and PLAN2(P<0.05). The rest of indexs in GTV,CTV, PTV had no significant difference between PLAN1and PLAN2(P>0.05). PLAN1compared with PLAN3. The average values of V100inGTV was higher than that of PLAN3(P<0.05). The rest of indexs in GTV,CTV, PTV had no significant difference between PLAN1and PLAN3 (P>0.05).(6) The average doses of lung for PLAN1, PLAN2and PLAN3were1374.25±182.82cGy,1374.09±182.93cGy and1362.04±194.05cGy,respectively. V20were26.32±2.43%,26.32±2.42%and25.99±2.62%,V30were15.67±2.28%,15.65±2.27%and15.24±2.24%. The averagedoses of heart for PLAN1, PLAN2and PLAN3were1382.19±1269.94cGy,1382.90±1271.32cGy and1398.66±1300.80cGy, respectively. V40were9.35±12.83%,9.34±12.83%and9.36±12.94%, V50were4.71±6.79%,4.69±6.78%and4.57±6.62%. Except lung V30about PLAN1and PLAN3,the rest of the above indexes between PLAN1and PLAN2or PLAN1andPLAN3had no statistical difference (P>0.05). Furthermore, the indexes oflung V30about PLAN3were lower than that of PLAN1. In the original plan,none of patients’ spinal cord exceeded4500cGy, while2cases of PLAN2weremore than4500cGy, with maximum4517.2cGy, while8cases of PLAN3weremore than4500cGy, with maximum5045.2cGy.(7) Using three-dimensionalsemiconductor detection matrix Delta4to measure the dose output of theaccelerator, The first to twelfth week of average passing rate (%) were94.6±2.3,91.2±3.0,92.8±3.6,88.4±4.9,86.3±6.9,95.4±2.8,94.0±3.4,94.0±4.4,77.1±7.0,85.4±5.4,85.9±5.4and85.8±5.6, respectively. The passingrate of dose verification for the first week was the best, while that for the ninthweek was the worst. The gamma values of beam of300°,0°,60°always weremore than95%throughout12weeks. The gamma value of beam of210°,150°were more scattered. The average passing rate (%) of210°to150°were92.3±6.5,97.9±3.1,97.4±2.1,96.4±4.8and92.7±4.6, respectively.Conclusion: The setup error of different direction are changingaccording to different esophageal cancer site. The PTV outside border from0.5cm to1.0cm in the Y axis in our hospital showed significantly insufficientwith CBCT. Because the setup error in X and Z axises of the late treatmentperiod patient become greater, so we suggest to replace the body membraneand make treatment plan at the beginning of the fifth week. The PTV outsideborder decreased significantly no matter where the esophageal cancerdeveloped with CBCT. Consequently CBCT are recommended to be used in radiotherapy. What’s more, intrafraction mobility can impact the originaltreatment target volume after the patients used the CBCT. However, weshould note that due to spinal cord is a tandem organ who relatively sensitiveto mobility in individual treatment, which might lead to many cases exposedto exceed maximum tolerance. Accelerator mechanical error and dose errorare objective existed and impact the target irradiation dose at some extent. |
PDF Full Text Request