-
弥漫大B细胞淋巴瘤(diffuse large B-cell lymphoma, DLBCL)是非霍奇金淋巴瘤(non-Hodgkin lymphoma,NHL)中最常见的病理类型,约占成人NHL的1/3[1]。18F-FDG PET/CT作为一种新型分子影像技术,在淋巴瘤尤其是DLBCL疗效评价及预后判断中具有重要的价值[2]。但是既往大部分研究中包含了Ⅰ~Ⅳ期的DLBCL患者,而Ⅰ期患者的预后本身就好于Ⅳ期的患者。因此,本研究回顾性分析85例Ⅱ~Ⅲ期DLBCL患者的临床资料和18F-FDG PET/CT显像资料,旨在探讨治疗前18F-FDG PET/CT显像中PET代谢参数的预后价值。
-
经过中位随访46(24~102)个月,85例患者中,26例患者复发或进展,中位PFS为31.0(2.5~99.7)个月,1年、3年无进展生存率分别为83.5%、40%;15例患者死亡,中位OS为34.8(2.0~99.7)个月,1年、3年总生存率分别为91.8%、48.2%。
-
85例患者的SUVmax、MTV、TLG的中位数分别为23.9(16.3)、25.1(95.3)cm3、424.4(1404.6)。以复发或进展作为阳性事件,ROC曲线分析结果显示:SUVmax的AUC=0.610(95%CI=0.483~0.736,P=0.109),MTV 的AUC=0.729(95%CI=0.621~0.838,P=0.001),TLG 的AUC=0.726(95%CI=0.621~0.831,P=0.001)。由于SUVmax的AUC较小,不能通过ROC曲线分析获得界值点,因此,以SUVmax的中位数23.9作为界值点进行分组。MTV的界值点为39.1 cm3(灵敏度为73.1%,特异度为69.5%);TLG的界值点为404.5(灵敏度为80.8%,特异度为61.0%)。
-
单因素分析结果显示:Ann Arbor分期、β-2微球蛋白、乳酸脱氢酶(lactate dehydrogenase, LDH)、大肿块、国际预后指数(international prognosis index, IPI)、SUVmax、MTV、TLG均是Ⅱ~Ⅲ期DLBCL患者PFS的影响因素(χ2=19.118、12.310、4.861、7.731、19.693、6.414、14.538、13.089,均P<0.05);Ann Arbor分期、β-2微球蛋白、大肿块、IPI、MTV、TLG 是患者 OS 的影响因素(χ2=14.550、5.473、5.643、15.943、13.877、12.677,均P<0.05)(表1,图1~3)。
因素 例数 中位无进
展生存
期(个月)χ2值,P值 中位总
生存期
(个月)χ2值,P值 因素 例数 中位无进
展生存
期(个月)χ2值,P值 中位总
生存期
(个月)χ2值,P值 性别 乳酸脱氢酶 男 35 31.0 χ2=0.734, 33.5 χ2=0.539, 正常 54 32.4 χ2=4.861, 34.8 χ2=2.150, 女 50 28.0 P=0.392 39.4 P=0.463 升高 31 26.0 P=0.027 33.2 P=0.147 年龄 大肿块 ≤60岁 53 32.8 χ2=1.656, 35.0 χ2=1.759, 无 64 33.8 χ2=7.731, 37.0 χ2=5.643, >60岁 32 25.5 P=0.198 30.5 P=0.185 有 21 25.0 P=0.013 26.0 P=0.018 B症状 国际预后指数 无 67 31.0 χ2=2.557, 34.8 χ2=0.074, 低危组(0~1分) 46 36.4 χ2=19.693, 43.2 χ2=15.943, 有 18 28.0 P=0.108 33.2 P=0.786 中高危组(2~4分) 39 24.1 P=0.000 28.0 P=0.000 ECOG评分 SUVmax 0~1分 76 31.0 χ2=0.899, 35.0 χ2=0.383, ≤23.9 43 36.4 χ2=6.414, 39.4 χ2=2.158, 2~3分 9 24.1 P=0.343 28.0 P=0.536 >23.9 42 26.0 P=0.011 29.0 P=0.142 Ann Arbor分期 MTV(cm3) Ⅱ期 45 34.8 χ2=19.118, 39.4 χ2=14.550, ≤39.1 48 34.7 χ2=14.538 43.7 χ2=13.877, Ⅲ期 40 25.0 P=0.000 28.0 P=0.000 >39.1 37 25.0 P=0.000 28.0 P=0.000 病理亚型 TLG 生发中心 39 33.8 χ2=1.650, 37.0 χ2=0.099, ≤404.5 41 36.4 χ2=13.089, 45.7 χ2=12.677, 非生发中心 46 30.5 P=0.199 31.0 P=0.735 >404.5 44 25.9 P=0.000 29.0 P=0.000 β-2微球蛋白 正常 68 33.8 χ2=12.310, 36.4 χ2=5.473, 升高 17 22.0 P=0.000 25.0 P=0.019 注:表中,DLBCL:弥漫大B细胞淋巴瘤;ECOG:美国东部肿瘤协作组;SUVmax:最大标准化摄取值;MTV:代谢体积;TLG:病灶糖酵解总量。 表 1 85例Ⅱ~Ⅲ期DLBCL患者的预后单因素分析
Table 1. Univariate analysis for survivals of 85 patients with diffuse large B cell lymphoma of stage Ⅱ~Ⅲ disease
图 1 85例Ⅱ~Ⅲ期DLBCL患者SUVmax的PFS(A)和OS(B)生存曲线
Figure 1. Kaplan-Meier curves for progression-free survival(A) and overall survival(B) of patients 85 patients with diffuse large B cell lymphoma of stage Ⅱ~Ⅲ disease by SUVmax
-
由于MTV与TLG呈高度相关,参照文献[6]并结合本研究,多因素分析时,二者中仅纳入了TLG,结果显示:LDH、TLG是影响Ⅱ~Ⅲ期DLBCL患者PFS的独立危险因素(RR=4.891, 95%CI=1.332~11.955,P=0.017;RR=0.195, 95%CI=0.058~0.660,P=0.009)(表2),IPI、TLG是影响患者OS的独立危险因素(RR=0.508, 95%CI=0.270~0.956,P=0.036;RR=0.433, 95%CI=0.227~0.826,P=0.011)(表3)。
因素 β Wald值 RR 95%CI P值 Ann Arbor分期 −1.230 3.247 0.292 0.077~1.114 0.072 β-2微球蛋白 −0.696 2.816 0.499 0.221~1.124 0.093 LDH 1.587 5.723 4.891 1.332~11.955 0.017 大肿块 −0.818 3.288 0.441 0.182~1.068 0.070 IPI −1.232 2.370 0.292 0.061~1.400 0.124 SUVmax −0.746 2.168 0.474 0.176~1.280 0.141 TLG −1.635 6.898 0.195 0.058~0.660 0.009 注:表中,DLBCL:弥漫大B细胞淋巴瘤;PFS:无进展生存期;LDH:乳酸脱氢酶;IPI:国际预后指数;SUVmax:最大标准化摄取值;TLG:病灶糖酵解总量。 表 2 85例Ⅱ~Ⅲ期DLBCL患者PFS的多因素分析
Table 2. Multivariate analysis for progression-free survival of 85 patients with diffuse large B cell lymphoma of stage Ⅱ~Ⅲ disease
因素 β Wald值 RR 95%CI P值 Ann Arbor分期 0.073 0.063 1.076 0.610~1.897 0.801 β-2微球蛋白 −0.115 0.143 0.892 0.492~1.616 0.706 大肿块 0.295 0.868 1.343 0.722~2.499 0.351 IPI −0.678 4.406 0.508 0.270~0.956 0.036 TLG −0.837 6.443 0.433 0.227~0.826 0.011 注:表中,DLBCL:弥漫大B细胞淋巴瘤;OS:总生存期;IPI:国际预后指数;TLG:病灶糖酵解总量。 表 3 85例Ⅱ~Ⅲ期DLBCL患者OS的多因素分析
Table 3. Multivariate analysis for overall survival of 85 patients with diffuse large B cell lymphoma of stage Ⅱ~Ⅲ disease
18F-FDG PET/CT代谢参数判断Ⅱ~Ⅲ期弥漫大B细胞淋巴瘤的预后价值
Prognostic value of 18F-FDG PET/CT metabolic parameters in patients with diffuse large B-cell lymphoma of stage Ⅱ~Ⅲ disease
-
摘要:
目的 探讨Ⅱ~Ⅲ期弥漫大B细胞淋巴瘤(DLBCL)治疗前18F-FDG PET/CT显像中最大标准化摄取值(SUVmax)、代谢体积(MTV)和病灶糖酵解总量(TLG)的预后价值。 方法 回顾性分析2009年6月至2015年12月诊治的85例初诊Ⅱ~Ⅲ期DLBCL患者的临床资料及治疗前18F-FDG PET/CT显像资料。以SUVmax的41%作为阈值,得出病灶的SUVmax、MTV和TLG。由受试者工作特征(ROC)曲线判断SUVmax、MTV及TLG预测无进展生存期(PFS)的最佳界值点。采用Kaplan-Meier法进行生存分析,单因素分析采用Log-rank 检验,多因素生存分析采用Cox比例风险模型。 结果 85例DLBCL患者的SUVmax、MTV及TLG的中位数分别为23.9(16.3)、25.1(95.3)cm3、424.4(1404.6)。ROC曲线分析结果显示:SUVmax的曲线下面积(AUC)=0.610(95%CI=0.483~0.736,P=0.109),MTV的AUC=0.729(95%CI=0.621~0.838,P=0.001),TLG的AUC=0.726(95%CI=0.621~0.831,P=0.001)。由于SUVmax的AUC较小,不能通过ROC曲线分析获得界值点,因此以SUVmax的中位数23.9作为界值点进行分组。MTV的界值点为39.1 cm3(灵敏度为73.1%,特异度为69.5%);TLG的界值点为404.5(灵敏度为80.8%,特异度为61.0%)。单因素分析结果显示:Ann Arbor分期、β-2微球蛋白、乳酸脱氢酶、大肿块、国际预后指数(IPI)、SUVmax、MTV、TLG是Ⅱ~Ⅲ期DLBCL患者PFS的影响因素(χ2=19.118,P=0.000;χ2=12.310,P=0.000;χ2=4.861,P=0.027;χ2=7.731, P=0.013;χ2=19.693, P=0.000;χ2=6.414, P=0.011;χ2=14.538,P=0.000;χ2=13.089, P=0.000)。Ann Arbor分期、β-2微球蛋白、大肿块、IPI、MTV、TLG是患者总体生存期(OS)的影响因素(χ2=14.550, P=0.000;χ2=5.473, P=0.019;χ2=5.643, P=0.018;χ2=15.943, P=0.000;χ2=13.877, P=0.000;χ2=12.677, P=0.000)。由于MTV与TLG呈高度相关,多因素分析时,二者中仅纳入了TLG,结果显示乳酸脱氢酶、TLG是影响患者PFS的独立危险因素(RR=4.891, 95%CI=1.332~11.955,P=0.017;RR=0.195, 95%CI=0.058~0.660,P=0.009),IPI、TLG是影响患者OS的独立危险因素(RR=0.508, 95%CI=0.270~0.956,P=0.036;RR=0.433, 95%CI=0.227~0.826,P=0.011)。 结论 18F-FDG PET/CT所测得的TLG是影响Ⅱ~Ⅲ期DLBCL患者PFS及OS的独立预后因素,TLG比SUVmax具有更好的预测价值,对DLBCL患者预后判断具有一定的参考价值。 -
关键词:
- 淋巴瘤,B细胞 /
- 正电子发射断层显像计算机体层摄影术 /
- 氟脱氧葡萄糖F18 /
- 预后
Abstract:Objective To investigate the prognostic value of the maximum standardized uptake value(SUVmax), metabolic tumor volume(MTV) and total lesion glycolysis(TLG) calculated from pretreatment 18F-FDG PET/CT results in patients with diffuse large B-cell lymphoma of stage Ⅱ~Ⅲ disease. Methods A total of 85 patients with DLBCL of stageⅡ~Ⅲ disease were enrolled from June 2009 to December 2015, and the clinic data and pretreatment 18F-FDG PET/CT data were retrospectively analyzed. MTV and TLG of tumor tissue were calculated from PET/CT images with the threshold value of 41% of the SUVmax. The optimal cutoff point of progression-free survival(PFS) of SUVmax, MTV and TLG were investigated by using receiver operating characteristic(ROC) curve analysis. The Kaplan-Meier method and Log-rank test were respectively used for survival analysis and univariate analysis, and COX proportional hazards model for multivariate analysis. Results The SUVmax, MTV and TLG of 85 patients were 23.9(16.3), 25.1(95.3) cm3, 424.4(1404.6), respectively. ROC curve showed that the area under the curce(AUC) of SUVmax, MTV and TLG were 0.610, 0.729 and 0.726(95% CI: 0.483–0.736, P=0.109; 0.621–-0.838, P=0.001; 0.621–0.831, P=0.001), respectively. The median SUVmax(23.9) was used as the cutoff points due to smaller AUC of SUVmax. The cutoff point of MTV was 39.1 cm3(sensitivity=73.1% and specificity=69.5%), and the cutoff point of TLG was 404.5(sensitivity=80.8% and specificity= 61.0%). Univariate analysis showed that Ann Arbor stage, β-2 MG, lactate dehydrogenase(LDH) level, bulky disease, international prognostic index(IPI), SUVmax, MTV and TLG were relative factors affecting PFS(χ2=19.118, P=0.000; χ2=12.310, P=0.000; χ2=4.861, P=0.027; χ2=7.731, P=0.013; χ2=19.693, P=0.000; χ2=6.414, P=0.011; χ2=14.538, P=0.000; χ2=13.089, P=0.000), and Ann Arbor stage, β-2 MG, bulky disease, IPI, MTV and TLG were relative factors affecting overall survival(OS)(χ2=14.550, P=0.000; χ2=5.473, P=0.019; χ2=5.643, P=0.018; χ2=15.943, P=0.000; χ2=13.877, P=0.000; χ2=12.677, P=0.000). As MTV and TLG measures correlated strongly, only TLG measures were used for multivariate analysis. LDH level and TLG were statistically significant predictors of PFS(RR=4.891, 95%CI=1.332–11.955, P=0.017; RR=0.195, 95%CI=0.058–0.660, P=0.009), and IPI and TLG were statistically significant predictors of OS on multivariate analysis(RR=0.508, 95%CI=0.270–0.956, P=0.036; RR=0.433, 95%CI=0.227–0.826, P=0.011). Conclusions TLG in pretreatment 18F-FDG PET/CT is an independent prognostic factor for predicting progression-free survival and overall survival time in patients with DLBCL of stageⅡ-Ⅲ disease. TLG may be more useful than SUVmax for prognosisand has certainly reference value. -
表 1 85例Ⅱ~Ⅲ期DLBCL患者的预后单因素分析
Table 1. Univariate analysis for survivals of 85 patients with diffuse large B cell lymphoma of stage Ⅱ~Ⅲ disease
因素 例数 中位无进
展生存
期(个月)χ2值,P值 中位总
生存期
(个月)χ2值,P值 因素 例数 中位无进
展生存
期(个月)χ2值,P值 中位总
生存期
(个月)χ2值,P值 性别 乳酸脱氢酶 男 35 31.0 χ2=0.734, 33.5 χ2=0.539, 正常 54 32.4 χ2=4.861, 34.8 χ2=2.150, 女 50 28.0 P=0.392 39.4 P=0.463 升高 31 26.0 P=0.027 33.2 P=0.147 年龄 大肿块 ≤60岁 53 32.8 χ2=1.656, 35.0 χ2=1.759, 无 64 33.8 χ2=7.731, 37.0 χ2=5.643, >60岁 32 25.5 P=0.198 30.5 P=0.185 有 21 25.0 P=0.013 26.0 P=0.018 B症状 国际预后指数 无 67 31.0 χ2=2.557, 34.8 χ2=0.074, 低危组(0~1分) 46 36.4 χ2=19.693, 43.2 χ2=15.943, 有 18 28.0 P=0.108 33.2 P=0.786 中高危组(2~4分) 39 24.1 P=0.000 28.0 P=0.000 ECOG评分 SUVmax 0~1分 76 31.0 χ2=0.899, 35.0 χ2=0.383, ≤23.9 43 36.4 χ2=6.414, 39.4 χ2=2.158, 2~3分 9 24.1 P=0.343 28.0 P=0.536 >23.9 42 26.0 P=0.011 29.0 P=0.142 Ann Arbor分期 MTV(cm3) Ⅱ期 45 34.8 χ2=19.118, 39.4 χ2=14.550, ≤39.1 48 34.7 χ2=14.538 43.7 χ2=13.877, Ⅲ期 40 25.0 P=0.000 28.0 P=0.000 >39.1 37 25.0 P=0.000 28.0 P=0.000 病理亚型 TLG 生发中心 39 33.8 χ2=1.650, 37.0 χ2=0.099, ≤404.5 41 36.4 χ2=13.089, 45.7 χ2=12.677, 非生发中心 46 30.5 P=0.199 31.0 P=0.735 >404.5 44 25.9 P=0.000 29.0 P=0.000 β-2微球蛋白 正常 68 33.8 χ2=12.310, 36.4 χ2=5.473, 升高 17 22.0 P=0.000 25.0 P=0.019 注:表中,DLBCL:弥漫大B细胞淋巴瘤;ECOG:美国东部肿瘤协作组;SUVmax:最大标准化摄取值;MTV:代谢体积;TLG:病灶糖酵解总量。 表 2 85例Ⅱ~Ⅲ期DLBCL患者PFS的多因素分析
Table 2. Multivariate analysis for progression-free survival of 85 patients with diffuse large B cell lymphoma of stage Ⅱ~Ⅲ disease
因素 β Wald值 RR 95%CI P值 Ann Arbor分期 −1.230 3.247 0.292 0.077~1.114 0.072 β-2微球蛋白 −0.696 2.816 0.499 0.221~1.124 0.093 LDH 1.587 5.723 4.891 1.332~11.955 0.017 大肿块 −0.818 3.288 0.441 0.182~1.068 0.070 IPI −1.232 2.370 0.292 0.061~1.400 0.124 SUVmax −0.746 2.168 0.474 0.176~1.280 0.141 TLG −1.635 6.898 0.195 0.058~0.660 0.009 注:表中,DLBCL:弥漫大B细胞淋巴瘤;PFS:无进展生存期;LDH:乳酸脱氢酶;IPI:国际预后指数;SUVmax:最大标准化摄取值;TLG:病灶糖酵解总量。 表 3 85例Ⅱ~Ⅲ期DLBCL患者OS的多因素分析
Table 3. Multivariate analysis for overall survival of 85 patients with diffuse large B cell lymphoma of stage Ⅱ~Ⅲ disease
因素 β Wald值 RR 95%CI P值 Ann Arbor分期 0.073 0.063 1.076 0.610~1.897 0.801 β-2微球蛋白 −0.115 0.143 0.892 0.492~1.616 0.706 大肿块 0.295 0.868 1.343 0.722~2.499 0.351 IPI −0.678 4.406 0.508 0.270~0.956 0.036 TLG −0.837 6.443 0.433 0.227~0.826 0.011 注:表中,DLBCL:弥漫大B细胞淋巴瘤;OS:总生存期;IPI:国际预后指数;TLG:病灶糖酵解总量。 -
[1] Farooq U, Maurer MJ, Thompson CA, et al. Clinical heterogeneity of diffuse large B cell lymphoma following failure of front-line immunochemotherapy[J]. Br J Haematol, 2017, 179(1): 50−60. DOI: 10.1111/bjh.14813. [2] 胡娜, 吴永港, 肖立志, 等. 18F-FDG PET/CT代谢活性参数及其在淋巴瘤中的应用[J]. 国际放射医学核医学杂志, 2015, 39(4): 342−347. DOI: 10.3760/cma.j.issn.1673−4114.2015.04.015.
Hu N, Wu YG, Xiao LZ, et al. Metabolic parameters of 18F-FDG PET/CT and their application in lymphoma[J]. Int J Radiat Med Nucl Med, 2015, 39(4): 342−347. DOI: 10.3760/cma.j.issn.1673−4114.2015.04.015.[3] 李亚明, 赵晋华. 淋巴瘤PET/CT诊断学[M]. 北京: 人民卫生出版社, 2017.
Li Y M, Zhao J H. PET/CT imaging of lymphoma[M]. Beijing: People's Medical Publishing House, 2017.[4] 梁颖, 吴宁, 方艳, 等. 18F-FDG PET/CT显像SUVmax、MTV和TLG判断弥漫性大B细胞淋巴瘤的预后价值[J]. 中华核医学与分子影像杂志, 2015, 35(2): 97−101. DOI: 10.3760/cma.j.issn.2095−2848.2015.02.005.
Liang Y, Wu N, Fang Y, et al. Prognostic significance of SUVmax, MTV and TLG on 18F-FDG PET/CT imaging in patients with diffuse large B-cell lymphoma[J]. Chin J Nucl Med Mol Imaging, 2015, 35(2): 97−101. DOI: 10.3760/cma.j.issn.2095−2848.2015.02.005.[5] Okuyucu K, Ozaydın S, Alagoz E, et al. Prognosis estimation under the light of metabolic tumor parameters on initial FDG-PET/CT in patients with primary extranodal lymphoma[J]. Radiol Oncol, 2016, 50(4): 360−369. DOI: 10.1515/raon−2016−0045. [6] Mikhaeel NG, Smith D, Dunn JT, et al. Combination of baseline metabolic tumour volume and early response on PET/CT improves progression-free survival prediction in DLBCL[J]. Eur J Nucl Med Mol Imaging, 2016, 43(7): 1209−1219. DOI: 10.1007/s00259−016−3315−7. [7] 丁重阳, 郭喆, 吴宇霖, 等. 18F-FDG PET/CT在局限期小细胞肺癌预后评估中的价值[J]. 中华核医学与分子影像杂志, 2017, 37(11): 694−699. DOI: 10.3760/cma.j.issn.2095−2848.2017.11.005.
Ding CY, Guo Z, Wu YS, et al. Prognostic value of 18F-FDG PET/CT in patients with limited-stage small cell lung cancer[J]. Chin J Nucl Med Mol Imaging, 2017, 37(11): 694−699. DOI: 10.3760/cma.j.issn.2095−2848.2017.11.005.[8] Gallicchio R, Mansueto G, Simeon V, et al. F-18 FDG PET/CT quantization parameters as predictors of outcome in patients with diffuse large B-cell lymphoma[J]. Eur J Haematol, 2014, 92(5): 382−389. DOI: 10.1111/ejh.12268. [9] 张玥伟, 范洋, 杨志, 等. 化疗早期及中期18F-FDG PET/CT显像对56例弥漫大B细胞淋巴瘤患者预后的评估价值[J]. 中华血液学杂志, 2015, 36(10): 824−829. DOI: 10.3760/cma.j.issn.0253−2727.2015.10.004.
Zhang YW, Fan Y, Yang Z, et al. The prognosis value of early and interim 18F-FDG-PET/CT scans in patients with diffuse large B-cell lymphoma[J]. Chin J Hematol, 2015, 36(10): 824−829. DOI: 10.3760/cma.j.issn.0253−2727.2015.10.004.[10] Song MK, Yang DH, Lee GW, et al. High total metabolic tumor volume in PET/CT predicts worse prognosis in diffuse large B cell lymphoma patients with bone marrow involvement in rituximab era[J]. Leuk Res, 2016, 42(1): 1−6. DOI: 10.1016/j.leukres.2016.01.010. [11] Ceriani L, Martelli M, Conconi A, et al. Prognostic models for primary mediastinal (thymic) B-cell lymphoma derived from 18-FDG PET/CT quantitative parameters in the International Extranodal Lymphoma Study Group (IELSG) 26 study[J]. Br J Haematol, 2017, 178(4): 588−591. DOI: 10.1111/bjh.14728. [12] Zhou M, Chen Y, Huang H, et al. Prognostic value of total lesion glycolysis of baseline 18F-fluorodeoxyglucose positron emission tomography/computed tomography in diffuse large B-cell lymphoma[J]. Oncotarget, 2016, 7(50): 83544−83553. DOI: 10.18632/oncotarget.13180. [13] Adams HJ, de Klerk JM, Fijnheer R, et al. Prognostic superiority of the National Comprehensive Cancer Network International Prognostic Index over pretreatment whole-body volumetric-metabolic FDG-PET/CT metrics in diffuse large B-cell lymphoma[J]. Eur J Haematol, 2015, 94(6): 532−539. DOI: 10.1111/ejh.12467. [14] Lovinfosse P, Janvary ZL, Coucke P, et al. FDG PET/CT texture analysis for predicting the outcome of lung cancer treated by stereotactic body radiation therapy[J]. Eur J Nucl Med Mol Imaging, 2016, 43(8): 1453−1460. DOI: 10.1007/s00259−016−3314−8.