摘要
目的探索中性粒细胞/淋巴细胞比值(NLR)、淋巴细胞/单核细胞比值(LMR)在外周T细胞淋巴瘤(PTCL)患者预后分析中的应用。方法选择2008年7月至2018年8月,四川省肿瘤医院研究所诊治的121例PTCL患者为研究对象。其中,男性患者为92例,女性为29例,中位年龄为55岁(15~83岁)。收集患者的一般临床资料、实验室检查结果及影像学检查结果进行回顾性分析。根据患者初诊时血常规检查结果中,中性粒细胞绝对计数、淋巴细胞绝对计数、单核细胞绝对计数计算NLR、LMR。LMR与NLR的最佳截断值通过受试者工作特征(ROC)曲线计算获得,并且根据截断值将患者分为低、高NLR组和低、高LMR组。组间不同临床特征患者构成比的比较,采用χ2检验。采用Kaplan-Meier法绘制低、高LMR组和低、高NLR组患者的无进展生存(PFS)、总体生存(OS)曲线。采用Log-rank检验对PTCL患者PFS、OS率进行单因素分析,纳入分析的影响因素包括性别、年龄、Ann Arbor分期、病理类型,美国东部肿瘤协作组(ECOG)评分、国际预后指数(IPI)评分、PTCL预后指数(PIT)评分、B症状、乳酸脱氢酶(LDH)水平、血小板计数、贫血、Ki67指数、NLR、LMR。将单因素分析结果中有统计学意义及有临床指导意义的影响因素,纳入COX比例风险回归模型进行多因素分析。本研究遵循的程序符合2013年修订版《世界医学会赫尔辛基宣言》的要求。结果①通过NLR的ROC曲线获得NLR最佳截断值为3.822,敏感度为0.577,特异度为0.623,曲线下面积(AUC)为0.591(95%CI:0.488~0.694)。通过LMR的ROC曲线获得LMR最佳截断值为2.715,敏感度为0.519,特异度为0.710,AUC为0.614(95%CI:0.512~0.716)。②按照NLR最佳截断值,57例患者纳入低NLR组(NLR<3.822),64例患者纳入高NLR组(NLR≥3.822)。这2组患者临床特征比较,差异均无统计学意义(P>0.05)。按照LMR最佳截断值,75例患者纳入低LMR组(LMR<2.715),46例纳入高LMR组(LMR≥2.715)。低LMR组年龄>60岁者为38例,年龄≤60岁者为37例,高LMR组年龄>60岁者为13例,年龄≤60岁者为33例;低LMR组Ann Arbor分期系Ⅰ~Ⅱ期者为13例,Ⅲ~Ⅳ期者为62例;高LMR组Ⅰ~Ⅱ期者为16例,Ⅲ~Ⅳ期者为30例;低LMR组PTCL非特指型(PTCL-NOS)者为24例,血管免疫母细胞淋巴瘤(AITL)者为37例,间变性淋巴瘤激酶(ALK)-间变大细胞淋巴瘤(ALCL)者为14例,高LMR组PTCL-NOS者为25例,AITL者为11例,ALK-ALCL者为10例。这2组患者的年龄、Ann Arbor分期、病理类型构成比分别比较,差异均有统计学意义(χ2=5.870、4.764、8.297,P=0.015、0.029、0.016);2组患者其他临床特征比较,差异均无统计学意义(P>0.05)。③患者中位随访期为32个月(5~101个月)。低、高NLR组患者中位PFS期分别为16、11个月,低NLR组患者3年PFS率显著高于高NLR组(28.8%比10.2%;χ2=5.537,P=0.019)。低、高NLR组患者中位OS期分别为56、19个月,低NLR组患者3年OS率亦显著高于高NLR组(61.1%比27.8%;χ2=9.341,P=0.002)。低、高LMR组患者中位PFS期分别为10、16个月,低LMR组患者3年PFS率显著低于高LMR组(11.9%比31.7%;χ2=5.391,P=0.020);低、高LMR组患者中位OS期分别为17、56个月,低LMR组患者3年OS率亦显著低于高LMR组(28.5%比62.5%;χ2=8.999,P=0.003)。④PTCL患者预后多因素分析结果显示,Ann Arbor分期为Ⅲ~Ⅳ期(HR=0.544,95%CI:0.314~0.944,P=0.030),ECOG评分>1分(HR=0.349,95%CI:0.221~0.551,P<0.001),Ki67指数≥80%(HR=0.421,95%CI:0.253~0.699,P=0.001)及NLR≥3.822(HR=0.615,95%CI:0.400~0.944,P=0.026)是影响PTCL患者PFS的独立危险因素。Ann Arbor分期为Ⅲ~Ⅳ期(HR=3.632,95%CI:1.726~7.642,P=0.001),ECOG评分>1分(HR=4.311,95%CI:2.530~7.347,P<0.001),Ki67指数≥80%(HR=2.691,95%CI:1.500~4.828,P=0.001)及LMR<2.715(HR=0.450,95%CI:0.265~0.764,P=0.003)是影响PTCL患者OS的独立危险因素。结论NLR、LMR可作为PTCL患者的预后预测指标。但是,由于本研究仅为回顾性分析,尚需大样本、前瞻性、随机对照试验,对此结论进一步研究、验证。
Objective To investigate the application of neutrophil/lymphocyte ratio(NLR)and lymphocyte/monocyte ratio(LMR)in prognostic analysis of patients with peripheral T-cell lymphoma(PTCL).Methods From July 2008 to August 2018,a total of 121 patients with PTCL diagnosed and treated in Sichuan Cancer Hospital&Institute were selected as the research subjects.Among them,there were 92 male patients and 29 females,with median age of 55 years(15-83 years).Patients′general clinical data,laboratory test results and imaging test results were collected for retrospective analysis.The NLR and LMR were calculated based on the absolute neutrophil count,absolute lymphocyte count,and absolute monocyte count in the routine blood test results of the patient at the initial diagnosis.The optimal cut-off values of LMR and NLR were calculated from the receiver operating characteristic(ROC)curve,and patients were divided into low and high NLR groups,low and high LMR groups,according to optimal cut-off values.Chi-square test was used to compare the composition ratio of different clinical features between groups.Kaplan-Meier method was used to draw progression free survival(PFS)and overall survival(OS)curves of patients in low and high LMR groups and low and high NLR groups.Log-rank test was used to conduct univariate analysis of PFS and OS rates in PTCL patients.The influencing factors included gender,age,Ann Arbor stage,pathological type,Eastern Cooperative Oncology Group(ECOG)score,International Prognostic Index(IPI)score,Prognostic Index for PTCL(PIT)score,B symptoms,lactate dehydrogenase(LDH)level,platelets count,anemia,Ki67,NLR,LMR.Influencing factors with statistical significance in the results of univariate analysis and clinical guiding significance were included in the COX proportional hazard regression model for multivariate analysis.The procedure followed in this study was in accordance with the requirements of the Declaration of the World Medical Association Helsinki revised in 2013.Results①NLR optimal cut-off value obtained by the NLR ROC curve was 3.822,the sensitivity was 0.577,the specificity was 0.623,the area under curve(AUC)was 0.591(95%CI:0.488-0.694).LMR optimal cut-off value obtained from LMR ROC curve was 2.715,sensitivity was 0.519,specificity was 0.710,AUC was 0.614(95%CI:0.512-0.716).②According to the NLR optimal cut-off value,57 patients were included in low NLR group(NLR<3.822)and 64 patients were included in high NLR group(NLR≥3.822).Clinical characteristics of patients in the two groups were not statistically significant(P>0.05).According to the LMR optimal cut-off value,75 patients were included in low LMR group(LMR<2.715)and 46 patients were included in high LMR group(LMR≥2.715).In low LMR group,there were 38 patients aged>60 years old and 37 patients aged≤60 years old;in high LMR group,there were 13 patients aged>60 years old and 33 patients aged≤60 years old.In low LMR group,there were 13 patients with Ann Arbor stageⅠ-Ⅱ,62 cases with stageⅢ-Ⅳ;in high LMR group,there were 16 patients with stageⅠ-Ⅱ,30 cases with stageⅢ-Ⅳ.There were 24 cases of peripheral T-cell lymphoma,not otherwise specified(PTCL-NOS),37 cases of angioimmunoblastic T-cell lymphoma(AITL),14 cases of anaplastic lymphoma kinase(ALK)-anaplastic large cell lymphoma(ALCL)in low LMR group;and there were 25 cases of PTCL-NOS,11 cases of AITL,and 10 cases of ALK-ALCL in high LMR group.Differences in age,clinical stage,and pathological type between the two groups were statistically significant(χ2=5.870,4.764,8.297;P=0.015,0.029,0.016),other clinical features were not statistically significant(P>0.05).③Median follow-up time of patients was 32 months(5-101 months).Median PFS time of patients in low and high NLR groups was 16 and 11 months,respectively.3-year PFS rate of patients in low NLR group was significantly higher than that in high NLR group(28.8%vs 10.2%;χ2=5.537,P=0.019).Median OS time of patients in low and high NLR groups was 56 and 19 months,respectively.3-year OS rate of patients in low NLR group was also significantly higher than that in high NLR group(61.1%vs 27.8%;χ2=9.341,P=0.002).Median PFS time of patients in low and high LMR groups was 10 and 16 months,respectively.3-year PFS rate of patients in low LMR group was significantly lower than that in high LMR group(11.9%vs 31.7%;χ2=5.391,P=0.020).Median OS time of patients in low and high LMR group was 17 and 56 months,respectivety,and 3-year OS rate of patients in low LMR group was also significantly lower than that in high LMR group(28.5%vs 62.5%;χ2=8.999,P=0.003).④Multivariate analysis of prognostic factors in PTCL patients showed that Ann Arbor stageⅢ-Ⅳ(HR=0.544,95%CI:0.314-0.944,P=0.030),ECOG>1 score(HR=0.349,95%CI:0.221-0.551,P<0.001),Ki67≥80%(HR=0.421,95%CI:0.253-0.699,P=0.001)and NLR≥3.822(HR=0.615,95%CI:0.400-0.944,P=0.026)were independent risk factors affecting PFS of patients with PTCL.Ann Arbor stageⅢ-Ⅳ(HR=3.632,95%CI:1.726-7.642,P=0.001),ECOG>1 score(HR=4.311,95%CI:2.530-7.347,P<0.001),Ki67≥80%(HR=2.691,95%CI:1.500-4.828,P=0.001)and LMR<2.715(HR=0.450,95%CI:0.265-0.764,P=0.003)were independent risk factors affecting OS of patients with PTCL.Conclusions NLR and LMR could be prognostic indicators for patients with PTCL.However,as this study is just a retrospective analysis research,large samples,prospective,and randomized controlled trials are needed for further research and verification of this conclusion.
作者
冯薇臻
欧阳楚桐
余思思
张智慧
Feng Weizhen;Ouyang Chutong;Yu Sisi;Zhang Zhihui(Department of Oncology,Affiliated Tumor Hospital,Southwestern Medical University,Luzhou 646000,Sichuan Province,China)
出处
《国际输血及血液学杂志》
CAS
2020年第3期241-250,共10页
International Journal of Blood Transfusion and Hematology
基金
四川省卫生厅科学研究项目(18PJ190)。
