双稳健半参模型法评价CAF和TAC化疗方案治疗乳腺癌疗效的回顾性分析
|
摘要
目的探索TAC化疗方案和CAF化疗方案治疗乳腺癌的真实疗效。方法采用回顾性队列研究的方法选择乳腺癌患者800例,探索不同的化疗方案对乳腺癌患者的生存率的影响。结果 PS匹配后经过log-rank检验两组生存曲线之间差别有统计学意义(P<0.05),得出TAC化疗方案治疗乳腺癌的效果优于CAF化疗方案。PS匹配前利用Cox回归模型和logistic回归模型,分别采用PS回归调整法、PS逆概率处理加权法(IPTW)、PS标化死亡比加权法(SMRW)和双稳健半参模型法(PS回归调整与加权法的结合,DRW)进行两种化疗方案的生存分析,加权之后基线协变量得到均衡,并经过调整的log-rank检验两组生存曲线之间差别均有统计学意义(P<0.05),且结果均显示TAC化疗方案治疗乳腺癌的效果优于CAF化疗方案。通过对不同模型之间的比较,利用DRIPTW模型和DRSMRW模型得出的处理效应的偏倚显著减小,且减小偏倚的效果优于与之相对的IPTW和SMRW法,同时也优于匹配法和回归调整法,其中DRSMRW模型法最优,相对偏倚为0.037。结论在回顾性队列研究中,TAC化疗方案治疗乳腺癌的效果优于CAF化疗方案,为乳腺癌患者指定了最佳的治疗方案。
Objective This study was aimed to explore the real efficacy of TAC and CAF chemotherapy in the treatment of breast cancer.Methods A retrospective cohort study was adopted to investigate 800 breast cancer patients,then explore the impact of different chemotherapy regimens on the survival rate of patients with breast cancer.Results The age,marital status and tumor nature of the two groups of breast cancer patients matched by PS matching were not balanced before matching,and after the matching,the survival curves of the two groups had significant difference after Log-rank test(P<0.05).Concluded that TAC chemotherapy regimen in the treatment of breast cancer better than the CAF chemotherapy.Before matching,Cox regression model and logistic regression model were used.And PS integrative regression adjustment,PS inverse probability weighting method(IPTW),PS standardized mortality weighting method(SMRW) were applied to the analysis on the two chemotherapy regimens.After adjusted Log-rank test between the two survival curves were statistically significant(P<0.05),and the results showed that TAC Chemotherapy regimen for breast cancer is superior to CAF regimen.By comparing the different models,the bias of the treatment effect obtained by the SMRW model and the DRSMRW model was significantly reduced,and the effect of reducing the bias was superior to the corresponding IPTW and SMRW methods and also better than the matching method and regression adjustment method.The DRSMRW model method was the best,the relative bias was 0.037.Conclusion In a retrospective cohort study,the effect of TAC chemotherapy regimen on breast cancerwas superior to that of CAF chemotherapy regimen.Designated the best treatment for breast cancer patients.
Objective This study was aimed to explore the real efficacy of TAC and CAF chemotherapy in the treatment of breast cancer.Methods A retrospective cohort study was adopted to investigate 800 breast cancer patients,then explore the impact of different chemotherapy regimens on the survival rate of patients with breast cancer.Results The age,marital status and tumor nature of the two groups of breast cancer patients matched by PS matching were not balanced before matching,and after the matching,the survival curves of the two groups had significant difference after Log-rank test(P<0.05).Concluded that TAC chemotherapy regimen in the treatment of breast cancer better than the CAF chemotherapy.Before matching,Cox regression model and logistic regression model were used.And PS integrative regression adjustment,PS inverse probability weighting method(IPTW),PS standardized mortality weighting method(SMRW) were applied to the analysis on the two chemotherapy regimens.After adjusted Log-rank test between the two survival curves were statistically significant(P<0.05),and the results showed that TAC Chemotherapy regimen for breast cancer is superior to CAF regimen.By comparing the different models,the bias of the treatment effect obtained by the SMRW model and the DRSMRW model was significantly reduced,and the effect of reducing the bias was superior to the corresponding IPTW and SMRW methods and also better than the matching method and regression adjustment method.The DRSMRW model method was the best,the relative bias was 0.037.Conclusion In a retrospective cohort study,the effect of TAC chemotherapy regimen on breast cancerwas superior to that of CAF chemotherapy regimen.Designated the best treatment for breast cancer patients.
引文
[1] Choi AH,Barnholtz-Sloan JS,Kim JA.Effect of radiation therapy on survival in surgically resected retroperitoneal sarcoma:a propensity score-adjusted SEER analysis.Ann Oncol,2012,23(9):2449-2457.
[2] Xie J,Liu C.Adjusted kaplan-meier estimator and log-rank test with inverse pProbability of treatment weighting for survival data.Statist Med,2005,24(20):3089-3110.
[3] Scharfstein DO,Rotnitzky A,Robins JM.Adjusting for nonignorable dropoutusing semiparametric nonresponse models(with discussion).Journal of the American Statistical Association,1999,94,1096-1146.
[4] DF Mccaffrey,BA Griffin,D Almirall,et al.A Tutorial on Propensity Score Estimation for Multiple Treatments Using Generalized Boosted Models.Statistics in Medicine,2013,32 (19):3388.
[5] 涂娇,陈卉,倪平,等.倾向评分逆处理概率加权在生存分析中的应用及SAS宏实现.中国临床药理学与治疗学,2015,20(9):1011-1019.
[6] Rosenbaum PR,Rubin DB.The central role of the propensity score in observational studies for causal effects.Biometrika,1983,70:41-55.
[7] D’Agostino RB.Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group.Statistics in Medicine,1998,17:2265-2281.
[8] 李智文,任爱国.倾向评分加权分析法.中国生育健康杂志,2010,21(4):251-253.
[9] Robins JM,Hernan MA,Brumback B.Marginal structural models and causal inference in epidemiology.Epidemiology,2000,11:550-560.
[10] Hernan MA,Brumback B,Robins JM.Marginal structural models toestimate the causal effect of zidovudineon the survival of HIV-positive men.Epidemiology,2000,11:561-570.
[11] Sato T,Matsuyama Y.Marginal structural models as a tool for standardization.Epidemiology,2003,14:680-686.
[12] Austin PC.An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies.Multivariate BehavRes,2011,46(3):399-424.
[13] 赵璐,张彦,李曼.568例乳腺癌根治术后患者生存分析.中华临床医师杂志,2014,(22):79-82.
[14] 王小松,欧阳取长.4720例女性乳腺癌病例生存分析.湖南师范大学学报(医学版),2014,11(3):35-40.
[15] 肖菲菲.不同化疗方案的乳腺癌术后患者的生存疗效分析.河北省:华北理工大学,2015.
[16] 孙婷,秦国友,武振宇,等.不同混杂结构条件下各倾向性评分方法的模拟比较研究.中国卫生统计,2017,34(3):415-420.
[17] Austin PC.The use of propensity score methods withsurvival or time-to-event outcomes:reporting measures of effect similar to those used in randomized experiments.Statist Med,2014,33(7):1242-1258.
[2] Xie J,Liu C.Adjusted kaplan-meier estimator and log-rank test with inverse pProbability of treatment weighting for survival data.Statist Med,2005,24(20):3089-3110.
[3] Scharfstein DO,Rotnitzky A,Robins JM.Adjusting for nonignorable dropoutusing semiparametric nonresponse models(with discussion).Journal of the American Statistical Association,1999,94,1096-1146.
[4] DF Mccaffrey,BA Griffin,D Almirall,et al.A Tutorial on Propensity Score Estimation for Multiple Treatments Using Generalized Boosted Models.Statistics in Medicine,2013,32 (19):3388.
[5] 涂娇,陈卉,倪平,等.倾向评分逆处理概率加权在生存分析中的应用及SAS宏实现.中国临床药理学与治疗学,2015,20(9):1011-1019.
[6] Rosenbaum PR,Rubin DB.The central role of the propensity score in observational studies for causal effects.Biometrika,1983,70:41-55.
[7] D’Agostino RB.Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group.Statistics in Medicine,1998,17:2265-2281.
[8] 李智文,任爱国.倾向评分加权分析法.中国生育健康杂志,2010,21(4):251-253.
[9] Robins JM,Hernan MA,Brumback B.Marginal structural models and causal inference in epidemiology.Epidemiology,2000,11:550-560.
[10] Hernan MA,Brumback B,Robins JM.Marginal structural models toestimate the causal effect of zidovudineon the survival of HIV-positive men.Epidemiology,2000,11:561-570.
[11] Sato T,Matsuyama Y.Marginal structural models as a tool for standardization.Epidemiology,2003,14:680-686.
[12] Austin PC.An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies.Multivariate BehavRes,2011,46(3):399-424.
[13] 赵璐,张彦,李曼.568例乳腺癌根治术后患者生存分析.中华临床医师杂志,2014,(22):79-82.
[14] 王小松,欧阳取长.4720例女性乳腺癌病例生存分析.湖南师范大学学报(医学版),2014,11(3):35-40.
[15] 肖菲菲.不同化疗方案的乳腺癌术后患者的生存疗效分析.河北省:华北理工大学,2015.
[16] 孙婷,秦国友,武振宇,等.不同混杂结构条件下各倾向性评分方法的模拟比较研究.中国卫生统计,2017,34(3):415-420.
[17] Austin PC.The use of propensity score methods withsurvival or time-to-event outcomes:reporting measures of effect similar to those used in randomized experiments.Statist Med,2014,33(7):1242-1258.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |