基于术前人B-淋巴细胞趋化因子1检测的肝细胞癌患者肝切除术后复发预测模型的构建及分析

李明蔚; 高见; 翟相威; 钱相君; 文夏杰; 姚明解; 段昭君; 赵二江; 张玲; 鲁凤民

doi:10.3969/j.issn.1001-5256.2021.04.020

基于术前人B-淋巴细胞趋化因子1检测的肝细胞癌患者肝切除术后复发预测模型的构建及分析

DOI: 10.3969/j.issn.1001-5256.2021.04.020

李明蔚¹,
高见²,
翟相威¹,
钱相君^3a,
文夏杰^3b,
姚明解^3b,
段昭君⁴,
赵二江^5a,
张玲^5b, ,,
鲁凤民^{1, 3a, ,}

1.
郑州大学公共卫生学院流行病学教研室，郑州 450001
2.
上海交通大学医学院附属仁济医院，上海市肿瘤研究所，癌基因及相关基因国家重点实验室，上海 200032
3a.
北京大学基础医学院病原生物学系暨感染病中心，北京 100191
3b.
北京大学基础医学院人体解剖学与组织胚胎学系，北京 100191
4.
中国医学科学院基础医学研究所，北京协和医学院基础学院免疫学系，北京 100005
5a.
郑州大学附属肿瘤医院病案室，郑州 450008
5b.
郑州大学附属肿瘤医院肝胆胰外科，郑州 450008

基金项目:

国家自然科学基金 (81902115);

国家重点研发计划 (2017YFC0908100);

河南省基础与前沿技术研究计划项目 (162300410108)

利益冲突声明：本研究不存在研究者、伦理委员会成员、受试者监护人以及与公开研究成果有关的利益冲突。

作者贡献声明：李明蔚负责资料分析，撰写论文；高见、姚明解、段昭君、赵二江参与收集及分析数据，修改论文；翟相威、钱相君、文夏杰参与资料分析与解释；张玲、鲁凤民指导撰写文章并最后定稿。

详细信息

作者简介:
李明蔚(1995—)，女，主要从事分子流行病学研究

通信作者:
张玲，zhanglily561@126.com

鲁凤民，lu.fengmin@hsc.pku.edu.cn

中图分类号: R735.7
计量
- 文章访问数: 601
- HTML全文浏览量: 199
- PDF下载量: 45
- 被引次数: 0
出版历程
- 收稿日期: 2020-09-24
- 录用日期: 2020-12-16
- 出版日期: 2021-04-20

Construction and analysis of a predictive model for posthepatectomy recurrence in patients with hepatocellular carcinoma based on preoperative CXCL13 measurement

摘要

摘要: 目的探究与肝细胞癌患者肝切除术后复发相关的血清学指标，构建复发预后模型以评估患者是否适合采用姑息性肝切除手术。方法共纳入了2009年2月—2013年7月于郑州大学附属肿瘤医院接受肝切除术并进行随访的肝细胞癌患者111例。收集患者的基本资料，根据患者在随访期间内是否复发分为复发组和未复发组。正态分布的计量资料两组间比较采用t检验；非正态分布的计量资料两组间比较采用Wilcoxon秩和检验。计数资料组间比较采用χ²检验。Kaplan-Meier法绘制生存曲线，log- rank检验比较差异。单因素与多因素分析采用Cox回归分析，采用受试者工作特征曲线(ROC曲线)判断预测效能。结果 Kaplan-Meier生存曲线显示AFP、ALP、GGT、FIB低值组及人B-淋巴细胞趋化因子1(CXCL13) 高值组患者的中位复发时间更长(P值均＜0.05)。AFP[HR(95%CI)=1.69(1.03~2.79)，P=0.039]、GGT[HR(95%CI)=1.89(1.14~3.14)，P=0.014]和CXCL13[HR(95%CI)=0.54(0.33~0.89)，P=0.015]为与肝切除术后复发相关的独立因素。基于其构成的预后指数：PI=0.526×AFP+0.637×GGT-0.616×CXCL13，预测姑息性切除术后0~3个月内复发的ROC曲线下面积(AUC)为0.87，灵敏度为93.75%，特异度为63.64%，但对0~6个月(AUC=0.68)及更长时间的预测效能明显下降，对姑息性切除的复发预测效能显著高于根治性切除。结论基于CXCL13、AFP和GGT构建的预后模型可用于评估患者采用姑息性肝切除术后的早期复发风险，有助于临床医生依据患者获益情况作出诊疗决策。
- 癌, 肝细胞 /
- 肝切除术 /
- 复发 /
- 预测 /
- 模型, 统计学
Abstract: Objective To investigate the serological markers associated with posthepatectomy recurrence in patients with hepatocellular carcinoma, and to establish a prognostic model to evaluate whether palliative hepatectomy is suitable for such patients. Methods A total of 111 patients with hepatocellular carcinoma who underwent hepatectomy in the Affiliated Cancer Hospital of Zhengzhou University from February 2009 to July 2013 and received follow-up were enrolled. Basic clinical data were collected and the patients were divided into recurrence group and non-recurrence group according to whether recurrence was observed during follow-up. The t-test was used for comparison of normally distributed continuous data between two groups and the Wilcoxon rank sum test was used for comparison of non-normally distributed continuous data between two groups; the chi-square test was used for comparison of categorical data between two groups. Survival curves were plotted using the Kaplan-Meier method, and survival differences were analyzed using the log-rank test. A Cox regression analysis was used to perform univariate and multivariate analyses, and the area under the ROC curve (AUC) was used to evaluate prediction efficiency. Results The Kaplan-Meier survival curves showed that the patients with low alpha-fetoprotein (AFP), alkaline phosphatase, gamma-glutamyl transpeptidase (GGT), and fibrinogen and high CXCL13 had a longer median time to recurrence (P < 0.05). AFP (hazard ratio [HR][95%CI]=1.69(1.03~2.79), P=0.039), GGT (HR[95%CI]=1.89(1.14~3.14), P=0.014), and CXCL13 (HR[95%CI]=0.54(0.33~0.89), P=0.015) were independent factors associated with posthepatectomy recurrence. The prognostic index PI=0.526×AFP+0.637×GGT-0.616×CXCL13 established based on these factors had an AUC of 0.87, a sensitivity of 93.75%, and a specificity of 63.64% in predicting recurrence within 0-3 months after palliative hepatectomy, with a significant reduction in prediction efficiency for recurrence within 0-6 months (AUC=0.68) or a longer period of time. The recurrence prediction efficiency of this model for palliative hepatectomy was significantly higher than that for radical resection. Conclusion The prognostic model established based on CXCL13, AFP, and GGT can be used to evaluate the risk of early recurrence after palliative hepatectomy and thus helps clinicians to make diagnosis and treatment decisions based on patients' benefits.
- Carcinoma, Hepatocellular /
- Hepatectomy /
- Recurrence /
- Forecasting /
- Models, Statistical

HTML全文

图 1 两组的Kaplan-Meier曲线图

下载: 全尺寸图片幻灯片

图 2 Cox模型预测3、6、12、24个月复发情况的ROC曲线

下载: 全尺寸图片幻灯片

表 1 肝切除术后未复发组和复发组患者术前指标比较

指标	未复发组(n=40)	复发组(n=71)	统计值	P值
男/女(例)	33/7	57/14	χ²=0.82	0.775
年龄(岁)	53.83±11.05	54.85±10.11	t=-0.36	0.717
肝硬化(无/有，例)¹⁾	6/34	7/62	χ²=0.57	0.655
肿瘤数目(1个/≥2个，例)	33/7	57/14	χ²=0.98	0.775
肿瘤大小(＞5 cm/≤5 cm，例)¹⁾	20/16	42/27	χ²=0.28	0.599
AFP(ng/μl)	14.82(2.43~1000.00)	329.60(15.37~1210.00)	Z=-2.74	0.006
TBA(μmol/L)	12.20(7.03~17.00)	12.20(8.40~20.70)	Z=-0.31	0.760
TBil(μmol/L)	13.65(8.60~17.25)	15.00(10.80~19.80)	Z=-2.18	0.029
ALT(U/L)	40.50(24.25~62.50)	31.00(24.00~49.00)	Z=-1.27	0.204
AST(U/L)	35.50(25.25~60.75)	42.00(30.00~63.00)	Z=-1.59	0.111
PA(mg/L)	160.50(125.65~197.73)	160.50(129.00~185.00)	Z=-1.10	0.269
Alb(g/L)	40.10(35.50~44.10)	40.50(36.80~44.20)	Z=-0.35	0.729
Glo(g/L)	31.08±7.18	32.34±5.13	t=-1.07	0.287
A/G	1.35(1.08~1.54)	1.25(1.06~1.48)	Z=-0.83	0.407
FIB(g/L)	2.80(2.25~3.20)	2.85(2.45~3.62)	Z=-1.15	0.250
ALP(U/L)	94.50(68.00~136.25)	108.00(80.00~140.00)	Z=-1.04	0.298
GGT(U/L)	46.00(30.20~81.25)	90.00(39.00~149.00)	Z=-2.69	0.007
NLR	2.37(1.58~3.16)	2.46(1.75~3.16)	Z=-0.08	0.936
LMR	4.55±1.94	4.37±1.68	t=0.50	0.617
RBC(×10¹²/L)	4.40(4.11~4.76)	4.40(3.99~4.72)	Z=-1.35	0.892
CXCL13(pg/ml)	114.99(55.86~149.29)	65.91(22.14~142.82)	Z=-1.43	0.153
PT(s)	12.90(11.85~13.63)	12.90(12.20~14.10)	Z=-1.21	0.226
INR	1.09(1.01~1.14)	1.09(1.03~1.20)	Z=-0.88	0.380
PLT(×10⁹/L)	147.00(117.00~182.75)	147.00(85.00~204.00)	Z=-0.40	0.692
Child分级A/B(例)¹⁾	34/2	59/3	χ²=0.202	0.904
BCLC分期A/B/C(例)¹⁾	30/6/1	34/14/21	χ²=13.17	0.001
注：1)部分病例的部分数据缺失，亚组间的数目存在差异。

下载: 导出CSV

表 2 各指标高低两组的中位复发时间

指标	例数	中位复发时间(月)	χ²值	P值	指标	例数	中位复发时间(月)	χ²值	P值
性别(例)			0.03	0.874	A/G(例)			3.08	0.079
男	90	6.00			＞1.30	56	10.00
女	21	5.00			≤1.30	55	5.00
年龄(例)			0.39	0.532	FIB(例)			5.34	0.021
＞56.00岁	48	6.00			＞2.85 g/L	48	4.50
≤56.00岁	63	5.00			≤2.85 g/L	63	10.00
肝硬化(例)			0.05	0.816	ALP(例)			4.08	0.043
有	96	6.00			＞106.00 U/L	54	4.50
无	13	6.00			≤106.00 U/L	57	10.00
肿瘤数目(例)			0.39	0.532	GGT(例)			10.20	0.001
＞1个	90	5.00			＞73.50 U/L	55	4.00
1个	21	5.00			≤73.50 U/L	56	12.00
肿瘤大小(例)			2.21	0.137	NLR(例)			0.03	0.856
＞5 cm	62	5.00			＞2.44	55	6.00
≤5 cm	43	8.00			≤2.44	56	6.00
AFP(例)			10.04	0.002	LMR(例)			0.12	0.727
＞67.65 ng/μl	54	4.00			＞4.43	55	6.00
≤67.65 ng/μl	57	14.00			≤4.43	56	6.00
TBA(例)			0.27	0.601	RBC(例)			0.89	0.345
＞12.20 μmol/L	43	6.00			＞4.40×10¹²/L	49	5.00
≤12.20 μmol/L	68	6.00			≤4.40×10¹²/L	62	6.00
TBil(例)			1.73	0.188	CXCL13(例)			7.90	0.005
＞14.35 μmol/L	55	6.00			＞81.23 pg/ml	55	10.00
≤14.35 μmol/L	56	6.00			≤81.23 pg/ml	56	4.50
ALT(例)			0.91	0.339	PT(例)			2.28	0.131
＞35.50 U/L	55	7.00			＞12.90 s	50	8.00
≤35.50 U/L	56	6.00			≤12.90 s	61	5.00
AST(例)			3.66	0.056	INR(例)			1.41	0.235
＞39.50 U/L	55	5.00			＞1.09	45	5.00
≤39.50 U/L	56	12.00			≤1.09	66	6.00
PA(例)			0.00	0.962	PLT(例)			0.28	0.596
＞160.50 mg/L	47	6.00			＞147.00×10⁹/L	52	6.00
≤160.50 mg/L	64	6.00			≤147.00×10⁹/L	59	6.00
Alb(例)			1.04	0.307	Child分级(例)			0.66	0.719
＞40.50 g/L	55	5.00			A级	93	6.00
≤40.50 g/L	56	8.00			B级	5	3.00
Glo(例)			0.98	0.320	BCLC分期(例)			12.36	＜0.001
＞31.50 g/L	52	5.00			A	64	15.00
≤31.50 g/L	59	7.00			B	20	6.00
≤31.50 g/L	59	7.00			C	22	4.00

下载: 导出CSV

表 3 与肝癌复发相关的单因素分析

指标	β	HR (95%CI)	P值
AFP(ng/μl)	0.734	2.08(1.28~3.39)	0.003
TBA(μmol/L)	0.123	1.13(0.70~1.83)	0.618
TBil(μmol/L)	0.302	1.35(0.84~2.17)	0.211
ALT(U/L)	-0.220	0.80(0.50~1.29)	0.803
AST(U/L)	0.436	1.55(0.96~2.48)	0.070
PA(mg/L)	-0.011	0.99(0.61~1.61)	0.964
Alb(g/L)	-0.235	0.79(0.49~1.27)	0.332
Glo(g/L)	0.231	1.26(0.78~2.04)	0.347
A/G	-0.404	0.67(0.42~1.08)	0.096
FIB(g/L)	0.530	1.17(1.05~2.74)	0.030
ALP(U/L)	0.461	1.59(0.99~2.54)	0.056
GGT(U/L)	0.749	2.12(1.29~3.47)	0.003
NLR	0.168	1.18(0.68~2.06)	0.550
LMR	0.100	1.10(0.68~1.78)	0.683
RBC(×10¹²/L)	0.216	1.24(0.78~1.99)	0.369
CXCL13(pg/ml)	-0.666	0.51(0.31~0.84)	0.009
PT(s)	0.345	1.41(0.88~2.26)	0.152
INR	0.274	1.32(0.82~2.12)	0.259
PLT(×10⁹/L)	0.123	1.13(0.70~1.82)	0.614

下载: 导出CSV

表 4 Cox模型预测不同时间段复发情况的AUC、灵敏度及特异度

时间	复发率	AUC(95%CI)	灵敏度	特异度
0~3个月	30.63%	0.82(0.74~0.89)	91.18%	63.64%
0~6个月	50.45%	0.68(0.59~0.77)	71.43%	63.64%
0~12个月	56.76%	0.65(0.55~0.74)	68.25%	64.58%
0~24个月	60.36%	0.61(0.51~0.70)	64.18%	61.36%

下载: 导出CSV

表 5 Cox模型预测姑息性切除和根治性切除不同时间段复发情况的AUC、灵敏度及特异度

方法	例数	时间	AUC(95%CI)	灵敏度	特异度
姑息性切除	72	0~3个月	0.87(0.77~0.94)	93.75%	70.00%
		0~6个月	0.72(0.60~0.82)	72.92%	66.67%
		0~12个月	0.68(0.56~0.79)	69.81%	68.42%
		0~24个月	0.63(0.51~0.74)	92.86%	37.50%
根治性切除	39	0~3个月	0.50(0.34~0.66)	100.00%	32.43%
		0~6个月	0.57(0.40~0.72)	62.50%	61.29%
		0~12个月	0.54(0.31~0.70)	50.00%	72.41%
		0~24个月	0.51(0.35~0.67)	0	82.14%

下载: 导出CSV

参考文献(18)

[1]	BRAY F, FERLAY J, SOERJOMATARAM I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2018, 68(6): 394-424. DOI: 10.3322/caac.21492.
[2]	ZHOU Z, LIN X, LIN N, et al. Establishment and predictive value of clinical prediction model for the recurrence after resection of hepatocellular carcinoma[J/CD]. Chin J Hepat Surg(Electronic Edition), 2017, 6(3): 192-196. DOI: 10.3877/cma.j.issn.2095-3232.2017.03.010. 周正, 林晓, 林楠, 等. 肝细胞癌切除术后复发的临床预测模型的建立及其预测价值[J/CD]. 中华肝脏外科手术学电子杂志, 2017, 6(3): 192-196. DOI: 10.3877/cma.j.issn.2095-3232.2017.03.010.
[3]	ZLOTNIK A, YOSHIE O. The chemokine superfamily revisited[J]. Immunity, 2012, 36(5): 705-716. DOI: 10.1016/j.immuni.2012.05.008.
[4]	LI B, SU H, CAO J, et al. CXCL13 rather than IL-31 is a potential indicator in patients with hepatocellular carcinoma[J]. Cytokine, 2017, 89: 91-97. DOI: 10.1016/j.cyto.2016.08.016.
[5]	DUAN Z, GAO J, ZHANG L, et al. Phenotype and function of CXCR5⁺CD45RA^-CD4⁺ T cells were altered in HBV-related hepatocellular carcinoma and elevated serum CXCL13 predicted better prognosis[J]. Oncotarget, 2015, 6(42): 44239-44253. DOI: 10.18632/oncotarget.6235.
[6]	BINDEA G, MLECNIK B, TOSOLINI M, et al. Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer[J]. Immunity, 2013, 39(4): 782-795. DOI: 10.1016/j.immuni.2013.10.003.
[7]	Ministry of Health of the People's Republic of China. Diagnosis, management, and treatment of hepatocellular carcinoma (V2011)[J]. J Clin Hepatol, 2011, 27(11): 1141-1159. DOI: 10.3969/j.issn.1001-5256.2011.11.004 中华人民共和国卫生部. 原发性肝癌诊疗规范(2011年版)[J]. 临床肝胆病杂志, 2011, 27(11): 1141-1159. DOI: 10.3969/j.issn.1001-5256.2011.11.004
[8]	LI CY. Expression and effect of CXCL 13 chemokine in hepatocellullar carcinoma[D]. Changchun: Jilin University, 2017. 李春艳. 趋化因子CXCL13在肝细胞癌中的表达及作用机制研究[D]. 长春: 吉林大学, 2017.
[9]	LI C, KANG D, SUN X, et al. The effect of C-X-C motif chemokine 13 on hepatocellular carcinoma associates with Wnt signaling[J]. Biomed Res Int, 2015, 2015: 345413. DOI: 10.1155/2015/345413.
[10]	FÖRSTER R, MATTIS AE, KREMMER E, et al. A putative chemokine receptor, BLR1, directs B cell migration to defined lymphoid organs and specific anatomic compartments of the spleen[J]. Cell, 1996, 87(6): 1037-1047. DOI: 10.1016/s0092-8674(00)81798-5.
[11]	KIM CH, ROTT LS, CLARK-LEWIS I, et al. Subspecialization of CXCR5⁺ T cells: B helper activity is focused in a germinal center-localized subset of CXCR5⁺ T cells[J]. J Exp Med, 2001, 193(12): 1373-1381. DOI: 10.1084/jem.193.12.1373.
[12]	CUCCHETTI A, PISCAGLIA F, CATURELLI E, et al. Comparison of recurrence of hepatocellular carcinoma after resection in patients with cirrhosis to its occurrence in a surveilled cirrhotic population[J]. Ann Surg Oncol, 2009, 16(2): 413-422. DOI: 10.1245/s10434-008-0232-4.
[13]	CHEN K, YANG HJ, DENG XF, et al. Risk factors of primary hepatocellular carcinoma patients early term recurrence after radical resection and prediction model establishment[J]. Chin J Cancer Prev Treat, 2018, 25(5): 344-348. https://www.cnki.com.cn/Article/CJFDTOTAL-QLZL201805009.htm 陈凯, 杨洪吉, 邓小凡, 等. 肝癌根治性切除术后早期复发危险因素分析及预测模型构建[J]. 中华肿瘤防治杂志, 2018, 25(5): 344-348. https://www.cnki.com.cn/Article/CJFDTOTAL-QLZL201805009.htm
[14]	PAN YX, CHEN JC, FANG AP, et al. A nomogram predicting the recurrence of hepatocellular carcinoma in patients after laparoscopic hepatectomy[J]. Cancer Commun (Lond), 2019, 39(1): 55. DOI: 10.1186/s40880-019-0404-6.
[15]	YE YJ, LIU B, CHEN W, et al. Prognosis and staging system analysis of hepatocellular carcinoma patients with negative serum alpha-fetoprotein[J]. J Clin Hepatol, 2019, 35(3): 535-541. DOI: 10.3969/j.issn.1001-5256.2019.03.017. 叶颖剑, 刘波, 陈伟, 等. 血清AFP阴性的肝细胞癌患者预后及分期系统分析[J]. 临床肝胆病杂志, 2019, 35(3): 535-541. DOI: 10.3969/j.issn.1001-5256.2019.03.017.
[16]	LIU AX, WANG HQ, BO WT, et al. Clinical efficacy and prognostic factors analysis of hepatectomy for hepatocellular carcinoma[J]. Chin J Dig Surg, 2019, 18(4): 368-374. DOI: 10.3760/cma.j.issn.1673-9752.2019.04.012. 刘爱祥, 王海清, 薄文滔, 等. 肝细胞癌肝切除术的临床疗效及预后因素分析[J]. 中华消化外科杂志, 2019, 18(4): 368-374. DOI: 10.3760/cma.j.issn.1673-9752.2019.04.012.
[17]	XUE Y, ZHANG J, BAI Y, et al. Establishment and validation of a prognostic model for hepatocellular carcinoma after radical liver resection[J]. J BUON, 2019, 24(4): 1420-1428. http://www.researchgate.net/publication/336824473_Establishment_and_validation_of_a_prognostic_model_for_hepatocellular_carcinoma_after_radical_liver_resection/download
[18]	Department of Medical Administration. National Health and Health Commission of the People's Republic of China. Guidelines for diagnosis and treatment of primary liver cancer in China (2019 edition)[J]. J Clin Hepatol, 2020, 36(2): 277-292. DOI: 10.3969/j.issn.1001 -5256.2020.02.007. 中华人民共和国国家卫生健康委员会医政医管局. 原发性肝癌诊疗规范(2019年版)[J]. 临床肝胆病杂志, 2020, 36(2): 277-292. DOI: 10.3969/j.issn.1001 -5256.2020.02.007.