基于放射生物学模型的非小细胞肺癌不同放射治疗方案对比研究  

Comparative study of different radiotherapy schemes for NSCLC based on radiobiological model

作　　者：王鹏[1] 陈杰[1] 王克强[1] 张文学[1] Wang Peng;Chen Jie;Wang Keqiang;Zhang Wenxue(Department of Radiotherapy,Tianjin Medical University General Hospital,Tianjin 300070,China)

机构地区：[1]天津医科大学总医院放射治疗科,天津300070

出　　处：《中国医学装备》2024年第12期32-38,共7页China Medical Equipment

摘　　要：目的:基于非小细胞肺癌(NSCLC)放疗评估的生物模型,对比不同放射治疗方案在肿瘤控制概率(TCP)和正常组织并发症概率(NTCP)的差异。方法:采集2021年4月至2022年7月就诊于天津医科大学总医院的15例NSCLC患者放疗资料,所有患者分别采用低分辨率泊松(TCP-Poisson-LQ)模型、Zaider-Minerbo(TCP-ZM)模型和TCP-Logit模型拟合TCP曲线,采用Lyman-Kutcher-Burman(LKB)模型和线性二次(LQ)模型拟合NTCP曲线,比较不同模型在肿瘤控制率、放射性肺炎和放射性心包炎上的适用性,并比较常规放疗方案(方案1)、最大治疗增益比方案(方案2)和平均肺剂量(MLD)<20 Gy时最大分割次数方案(方案3)在TCP和NTCP中的差异。结果:TCP-Poisson-LQ模型在60~70 Gy处的平均TCP为(87.2±11.92)%,符合临床所需剂量。全肺平均受量<26 Gy时,NTCP-LQ模型计算的放射性肺炎发生率高于NTCP-LKB模型。在不同方案的比较中,方案3的TCP均值为(81.56±11.20)%,高于其他两种方案(60.28±8.04)%和(69.46±18.09)%,差异均有统计学意义(t=-6.196、-1.969,P<0.05)。方案3的平均放射性肺炎的发生率为(19.24±0.43)%,高于方案1和方案2的(15.07±3.24)%和(15.89±4.55)%,差异有统计学意义(t=-5.878、-2.386,P<0.05)。结论:采用Poisson-LQ模型和NTCP-LQ模型分别计算NSCLC患者的TCP和放射性肺炎发生率较为合理,MLD<20 Gy时最大分割次数方案(方案3)可以在保证治疗安全的前提下有效提高TCP。Objective:To compare the difference of tumor control probability(TCP)and normal tissue complication probability(NTCP)between different radiotherapy schemes bases on biological model of non-small cell lung cancer(NSCLC)that used in assessing radiotherapy.Methods:The radiotherapy data of 15 NSCLC patients who admitted to Tianjin Medical University General Hospital from April 2021 to July 2022 were collected.The low resolution Poisson(TCP Poisson LQ)model,Zaider Minerbo(TCP-ZM)model and TCP Logit model were respectively adopted to fit TCP curve for all patients.Lyman-Kutcher-Burman(LKB)model and linear quadratic(LQ)model were adopted to fit NTCP curves for comparing applicability of several models in tumor control rate,radiation pneumonitis and radiation pericarditis,and the differences in TCP and NTCP among conventional radiotherapy regimen(scheme 1),regimen of maximum gain ratio of treatment(scheme 2),and maximum segmentation frequency regimen(scheme 3)as mean lung dose(MLD)<20 Gy.Results:The average TCP of the TCP Poisson LQ model was(87.2±11.92)%at 60-70 Gy,which met the requirement of clinical dose.The incidence rate of radiation pneumonitis,which was calculated by the NTCP-LQ model,was higher than that by the NTCP-LKB model when the average radiation dose of whole lung was less than 26 Gy.In the comparison of different schemes,the TCP mean of scheme 3 was(81.56±11.20)%,which was respectively higher than that of other two schemes(60.28±8.04)%and(69.46±18.09)%,and the differences of them were statistically significant(t=-6.196,-1.969,P<0.05).The average incidence of radiation pneumonitis in Scheme 3 was(19.24±0.43)%,which was respectively higher than that in Scheme 1 and Scheme 2[(15.07±3.24)%and(15.89±4.55)%],respectively,and the differences of them were statistically significant(t=-5.878,-2.386,P<0.05).Conclusion:It is reasonable to use Poisson-LQ model and NTCP-LQ model to calculate TCP,and incidence of radiation pneumonitis in NSCLC patients.The maximum segmentation frequency scheme(Scheme 3)can effec

关 键 词：非小细胞肺癌(NSCLC) 放射生物模型 肿瘤控制概率(TCP) 正常组织并发症概率(NTCP) 

分 类 号：R816.41[医药卫生—放射医学]