基于多叶准直器日志文件的鼻咽癌调强放疗的累积剂量验证

The IMRT accumulative dosimetric verification of NPC using MLC log files

目的 利用多叶准直器（MLC）日志文件验证鼻咽癌患者动态调强放疗的累积剂量,分析MLC叶片位置偏差与剂量学关系.方法 跟踪记录5例鼻咽癌患者的28个分次动态调强放疗,共计1 400个日志文件.通过Argus软件解析和提取MLC叶片的实际位置信息,并与MLC叶片的计划位置信息比较,计算每个MLC叶片的位置偏差和所有患者的整体平均叶片位置偏差.将MLC文件导入治疗计划系统（TPS）进行剂量重建计算,分别比较和分析每个病例重建计划和原计划的28次累积剂量差异及单次相对剂量差异.结果 单个叶片位置偏差＜0.50 mm约占90％,0.50～1.00mm之间的占4％ ～ 11％,1.00～ 1.50 mm约占8‰,整体平均叶片位置偏差为0.16 mm.与原始计划累积剂量相比,重建计划的叶片位置偏差导致靶区累积相对剂量差异＜±0.1％,危及器官的累积相对剂量差异＜±0.6％,差异均无统计学意义（P＞0.05）;而与原始计划的单次剂量相比,重建计划的靶区和脑干、脊髓和右眼晶状体单次剂量差异有统计学意义（z=-2.02 ～4.61和4.46、-4.51、2.07,P＜0.05）.结论 基于动态调强日志文件的分次累积剂量验证,一定程度上减小了单次MLC叶片位置偏差对剂量的影响,验证结果可更加精确地反映患者的实际受照剂量.

Objective To verify the accumulative dose of intensity-modulated radiotherapy （IMRT） of nasopharyngeal carcinoma （NPC） using linac logfiles,and analyze the relationship between the leaf position errors of multileaf collimator （MLC） and dose distribution changes.Methods In total,1 400 dynalog files of 28 fractions of 5 NPC patients were tracked and recorded.MLC leaf position information was compared with the planned MLC leaf position information to calculate the MLC leaf position errors of each leaf and the overall mean leaf position error of all patients,which were analyzed and extracted through Argus program.The MLC files were imported into treatment planning system（TPS） to recalculate the dose.Twenty-eight accumulative dose differences between original plans and reconstruction plans,and single relative dose differences were analyzed.Results About 90% of the leaf position errors were less than 0.50 mm,4%-11% of the leaf position errors were between 0.50-1.00 mm,while there were about 8‰ leaf position errors within 1.00-1.50 mm and the average leaf position error was 0.16 mm.Compared with accumulated dose of original plans and that of reconstruction plans,the average relative dose discrepancies in targets and organs at risk （OARs） induced by leaf position errors were less than ± 0.1% and ± 0.6%,and the differences were not statistically significant （P 〉 0.05）.Compared with single dose of original plans,single dose differences in targets,brainstem,spinal cord and right lens of reconstruction plans were statistically significant （z =-2.02-4.61,4.46,-4.51,2.07,P 〈 0.05）.Conclusions The accumulative dose verification based on multi-fraction dynamic intensity-modulated MLC log files,to some extent,lessen the dose influence of the single-fraction leaf position errors and the verification results could manifest the practical irradiation dose more accurately.