福建省牙龈癌患者预后风险预测模型的构建

陈敏敏; 包晓丹; 邓青榕; 张泽云; 何保昌; 刘凤琼; 汪靖; 陈法

doi:10.16462/j.cnki.zhjbkz.2020.10.013

福建省牙龈癌患者预后风险预测模型的构建

doi: 10.16462/j.cnki.zhjbkz.2020.10.013

350108福建, 福建医科大学公共卫生学院流行病与卫生统计学系

基金项目:

福建医科大学公共卫生学院大学生科研创新项目 201803

详细信息

通讯作者:
陈法, E-mail:chenfa@fjmu.edu.cn

中图分类号: R780; R730.7
计量
- 文章访问数: 361
- HTML全文浏览量: 212
- PDF下载量: 25
- 被引次数: 0
出版历程
- 收稿日期: 2019-10-10
- 修回日期: 2020-01-18
- 刊出日期: 2020-10-10

A novel prognostic index for gingival cancer in Fujian Province

Department of Epidemiology and Health Statistic, School of Public Health, Fujian Medical University, Fuzhou 350108, China

Funds:

Fujian Medical University, School of Public Health, college students'science and research innovation project 201803

More Information

Corresponding author: CHEN Fa,E-mail:chenfa@fjmu.edu.cn

摘要

摘要: 目的探讨福建地区牙龈癌患者预后影响因素并构建预后风险预测模型。方法收集福建医科大学附属第一医院口腔颌面外科2005年1月―2017年6月经病理确诊的牙龈癌患者共198例，对其进行随访。利用多因素COX回归分析模型分析计算死亡风险比值（hazard rate，HR）及其95%置信区间（95% confidence interval，95% CI），筛出牙龈癌患者预后的影响因素，并基于β系数构建每个个体的预后指数（prognostic index，PI）。运用X-tile软件确定PI的最佳截断点，并根据最佳截断点把研究对象分成低、中、高危组。结果在牙龈癌患者中，年龄≥60岁（HR=1.668，95% CI：1.002~2.777，P=0.049），有复发转移（HR=3.996，95% CI：2.295~6.959，P < 0.001）的患者预后更差；临床Ⅳ期与临床Ⅰ期相比预后更差（HR=3.002，95% CI：1.134~7.947，P=0.027）；经手术治疗的患者与未经手术治疗的相比预后更好（HR=0.246，95% CI：0.118~0.511，P < 0.001）。此外，构建的PI预后预测模型预测性能良好（AIC=611.605，C-index=0.747，AUC=0.765）。结论本研究发现年龄（≥60岁）、临床分期高、未进行手术治疗及存在复发转移均是影响牙龈癌患者预后的危险因素，且构建的PI预后预测模型能够较好的预测患者预后。
- 牙龈癌 /
- COX回归分析模型分析 /
- 预后指数
Abstract: Objective To evaluate the prognostic factors and to further develop an applicable prognostic index (PI) for patients with gingival cancer in Fujian Province. Methods From January 2005 to June 2017, a total of 198 patients with pathologically confirmed gingival cancer were recruited from the First Affiliated Hospital of Fujian Medical University. The clinical and follow-up data of all the patients were collected by medical records and telephone interviews, respectively. COX regression analysis model was used to calculate the hazard ratio (HR) as well as 95% confidence interval (95% CI) and to assess independent factors for the prognosis of gingival cancer. Subsequently a prognostic index was developed based on β value of each significant variable obtained from the multivariate COX hazards regression analysis model. PI was divided into low-, medium- and high- risk groups according to the optimal cutoff points obtained by the X-tile software. The Akaike information criterion (AIC) and Harrell's C-statistic (C-index) were used to evaluate the predictability of PI. Results Among the patients with gingival cancer, the patients aged ≥ 60 years (HR=1.668, 95% CI:1.002-2.777, P=0.049), and patients with recurrence and metastasis (HR=3.996, 95%CI:2.295-6.959, P < 0.001) had poorer prognosis. The prognosis of patients with clinical stage Ⅳ was worse than that of those with clinical stage Ⅰ (HR=3.002, 95%CI:1.134-7.947, P=0.027); Patients with surgical treatment had better prognosis than those without surgical treatment (HR=0.246, 95%CI:0.118-0.511, P < 0.001). In addition, the PI prognosis prediction model has a good predictability (AIC=611.605, C-index=0.747, AUC=0.765). Conclusions This study suggests that age (≥ 60 years), high clinical stage, no surgery treatment as well as recurrence and metastasis were the risk factors for the prognosis of gingival cancer patients and the PI developed had a relatively high predictive ability.
- Gingival cancer /
- COX regression analysis model /
- Prognostic index

HTML全文

图 1 使用X-tile软件确定最佳截断点

Figure 1. Using X-tile software to determine the optimal cut-off point

下载: 全尺寸图片幻灯片

表 1 牙龈癌预后影响因素的单因素分析[n (%)]

Table 1. Univariate analysis of prognostic factors for gingival cancer patients [n (%)]

变量	删失	死亡	HR(95% CI)值	logrank P值
年龄(岁)				0.009
＜60	64(50.79)	27(37.50)	1.000
≥60	62(49.21)	45(62.50)	1.870(1.158~3.018)
性别				0.391
男	72(57.14)	47(65.28)	1.000
女	54(42.86)	25(34.72)	0.808(0.497~1.314)
BMI^a(kg/m²)				0.218
18.5~	77(61.11)	48(66.67)	1.000
＜18.5	20(15.87)	7(9.72)	0.523(0.235~1.161)
>24	29(23.02)	17(23.61)	1.101(0.633~1.916)
吸烟				0.024
无	89(70.63)	41(56.94)	1.000
有	37(29.37)	31(43.06)	1.703(1.066~2.718)
饮酒				0.312
无	100(79.37)	54(75.00)	1.000
有	26(20.63)	18(25.00)	1.317(0.771~2.249)
文化程度				0.467
小学及以下	43(34.13)	33(45.83)	1.000
中学	73(57.94)	34(47.23)	0.774(0.478~1.252)
大专及以上	10(7.93)	5(6.94)	1.183(0.458~3.051)
合并症				0.019
无	47(37.30)	17(23.61)	1.000
有	79(62.70)	55(76.39)	1.903(1.103~3.282)
临床分期				0.001
Ⅰ期	23(18.25)	5(6.94)	1.000
Ⅱ期	25(19.84)	10(13.89)	1.151(0.393~3.370)
Ⅲ期	19(15.08)	11(15.28)	1.590(0.552~4.580)
Ⅳ期	59(46.83)	46(63.89)	3.285(1.301~8.291)
组织学分级				0.006
高分化	53(42.06)	32(44.44)	1.000
中分化	44(34.92)	18(25.00)	0.773(0.432~1.383)
低分化	20(15.87)	17(23.61)	2.365(1.297~4.313)
其他	9(7.15)	5(6.95)	1.124(0.437~2.890)
口腔部位				0.915
口裂以上	45(35.71)	25(34.72)	1.000
口裂以下	74(58.73)	44(61.11)	0.906(0.553~1.484)
其他	7(5.56)	3(4.17)	0.859(0.259~2.852)
口腔癌复发转移				＜0.001
无	101(80.80)	19(27.14)	1.000
有	24(19.20)	51(72.86)	4.241(2.502~7.189)
病理类型				0.497
鳞癌	107(84.92)	64(88.89)	1.000
腺癌及其他	19(15.08)	8(11.11)	0.775(0.372~1.618)
NLR^b				0.203
≤1.94	42(33.33)	13(18.06)	1.000
1.95~	51(40.48)	25(34.72)	1.535(0.782~3.012)
≥3.67	17(13.49)	19(26.39)	2.107(1.039~4.276)
其他	16(12.70)	15(20.83)	1.350(0.638~2.857)
血小板(10⁹ /L)				0.654
≤350	61(48.42)	23(31.94)	1.000
>350	50(39.68)	33(45.83)	1.286(0.752~2.198)
其他	15(11.90)	16(22.23)	1.151(0.598~2.215)
PNI^c				0.783
≤48.25	57(45.24)	28(38.89)	1.000
>48.25	54(42.86)	28(38.89)	1.206(0.713~2.038)
其他	15(11.90)	16(22.22)	1.085(0.582~2.023)
PLR^d				0.979
≤146.57	58(46.03)	25(34.72)	1.000
>146.57	52(41.27)	31(43.06)	1.003(0.590~1.703)
其他	16(12.70)	16(22.22)	0.944(0.496~1.797)
手术				＜0.001
无	7(5.60)	16(22.22)	1.000
有	118(94.40)	56(77.78)	0.208(0.116~0.373)
放化疗				0.792
无治疗	53(42.74)	25(34.72)	1.000
化疗	18(14.52)	15(20.83)	0.907(0.472~1.745)
放疗	29(23.39)	9(12.50)	0.658(0.307~1.411)
放疗+化疗	24(19.35)	23(31.94)	1.155(0.653~2.041)
注：BMI^a：体重指数，body mass index；NLR^b：中性粒细胞/淋巴细胞比值，neutrophil-to-lymphocyte ratio；PNI^c：预后营养指数，prognostic nutritional index；PLR^d：血小板/淋巴细胞比值，platelet-to-lymphocyte ratio。

下载: 导出CSV

表 2 牙龈癌预后影响因素的多因素分析

Table 2. Multivariate analysis of prognostic factors for gingival cancer patients

特征	β值	HR(95% CI)值	P值	PI模型评分
年龄(岁)
＜60		1.000
≥60	0.511	1.668(1.002~2.777)	0.049	5
吸烟
无		1.000
有	0.269	1.308(0.796~2.151)	0.290
临床分期
Ⅰ期		1.000
Ⅱ期	0.394	1.483(0.494~4.457)	0.482	4
Ⅲ期	0.217	1.242(0.417~3.700)	0.697	2
Ⅳ期	1.099	3.002(1.134~7.947)	0.027	11
组织学分级
高分化		1.000
中分化	-0.683	0.505(0.272~0.939)	0.031	7
低分化	0.402	1.494(0.767~2.913)	0.238	4
其他	-0.618	0.539(0.180~1.612)	0.269	6
复发/转移
无		1.000
有	1.385	3.996(2.295~6.956)	＜0.001	14
合并症
无		1.000
有	0.055	1.056(0.582~1.916)	0.875
手术
无		1.000
有	-1.404	0.246(0.118~0.511)	＜0.001	14

下载: 导出CSV

表 3 PI模型预测性能及趋势性检验[n (%)]

Table 3. PI model prediction performance and trend test[n (%)]

预后指数分组	删失	死亡	HR(95% CI)	HR(95% CI)^a
低危组	86(68.25)	15(20.83)	1.000	1.000
中危组	25(19.84)	23(31.94)	3.996(2.084~7.662)	4.241(2.131~8.440)
高危组	15(11.90)	34(47.22)	7.383(4.008~13.602)	6.987(3.643~13.398)
P for trend			＜0.001
C-index			0.747
AUC			0.765
AIC			611.605
注：^a调整的变量：性别、BMI、学历、吸烟、饮酒、营养指数、病理类型、血小板、NLR、PLR、口腔部位、合并症、放化疗。

下载: 导出CSV

参考文献(27)

[1]	魏温涛.常见头颈部恶性肿瘤的诊治[M].青海:青海人民出版社, 2005. Wei WT. The diagnosis and treatment of malignant tumours of head and neck[M]. Qinghai:Qinghai Peolple's Publishing House, 2005.
[2]	郭朱明, 王顺兰, 张诠, 等. 116例牙龈癌的治疗与预后分析[J].癌症, 2008, 27(3):307-310. DOI: 10.3321/j.issn:1000-467X.2008.03.016. Guo ZM, Wang SL, Zhang Q, et al. Treatment and prognosis of gingival carcinoma:a report of 116 cases[J]. Chin J Cancer, 2008, 27(3):307-310. DOI: 10.3321/j.issn:1000-467X.2008.03.016.
[3]	傅锦业, 高静, 郑家伟, 等.口腔癌相关危险因素的流行病学调查分析[J].中国口腔颌面外科杂志, 2011, 9(4):316-322. DOI: 10.3969/j.issn.1672-5085.2013.24.116. Fu JY, Gao J, Zheng JW, et al. Epidemiologic study of risk factors of oral cancer[J]. China Journal of Oral and Maxillofacial Surgery, 2011, 9(4):316-322. DOI: 10.3969/j.issn.1672-5085.2013.24.116.
[4]	Rietbergen MM, Witte BI, Velazquez ER, et al. Different prognostic models for different patient populations:validation of a new prognostic model for patients with oropharyngeal cancer in Western Europe[J]. BJC, 2015, 112(11):1733-1736. DOI: 10.1038/bjc.2015.139.
[5]	华贻军, 洪明晃, 罗东华, 等. 406例鼻咽癌患者预后多因素分析[J].中国肿瘤临床, 2005, 32(8):435-438. DOI: 10.3969/j.issn.1000-8179.2005.08.005. Hua YJ, Hong MH, Luo DH, et al. Multivariable analysis of prognostic predictors for 406 patients with nasopharyngeal carcinoma[J]. Chin J Clin Oncol, 2005, 32(8):435-438. DOI: 10.3969/j.issn.1000-8179.2005.08.005.
[6]	宋丽华, 宋现让, 张锡芹, 等.小细胞肺癌患者预后因素分析[J].中华肿瘤杂志, 2004, 26(7):413-416. DOI: 10.3760/j.issn:0253-3766.2004.07.008. Song LH, Song XR, Zhang XQ, et al. Prognostic factors in patients with small cell lung cancer[J]. Chin J Oncol, 2004, 26(7):413-416. DOI: 10.3760/j.issn:0253-3766.2004.07.008.
[7]	柳巍, 聂绍发, 陈继贵, 等. 370例结直肠癌患者预后影响因素及预后预测分析[J].中华流行病学杂志, 2007, 28(7):700-703. DOI: 10.3760/j.issn:0254-6450.2007.07.019. Liu W, Nie SF, Chen JG, et al. Study on the prognostic factors of colorectal cancer and on suggested model for prediction[J]. Chin J Epidemiol, 2007, 28(7):700-703. DOI: 10.3760/j.issn:0254-6450.2007.07.019.
[8]	Chen F, Cao Y, Huang J, et al. A novel prognostic index for oral squamous cell carcinoma patients with surgically treated[J]. Oncotarget, 2017, 8(33):55525-55533. DOI: 10.18632/oncotarget.14821.
[9]	Balkau B, Hu G, Qiao Q, et al. Prediction of the risk of cardiovascular mortality using a score that includes glucose as a risk factor. The DECODE study[J]. Diabetologia, 2005, 47(12):2118-2128. DOI: 10.1007/s00125-004-1574-5.
[10]	Chen F, Lin LS, Liu FQ, et al. Three prognostic indexes as predictors of response to adjuvant chemoradiotherapy in patients with oral squamous cell carcinoma after radical surgery:a large-scale prospective study.[J]. Head neck, 2019, 41(2):301-308. DOI: 10.1002/hed.25495.
[11]	Kevin C. Lee, Sung-Kiang Chuang, Elizabeth M, et al. Which clinicopathologic factors affect the prognosis of gingival squamous cell carcinoma:a population analysis of 4345 cases[J]. J Oral Maxillofac Surg, 2019, 77(5):986-993. DOI: 10.1016/j.joms.2019.01.007.
[12]	万艳, 尚政军.牙龈癌的流行病学危险因素研究[J].临床口腔医学杂志, 2013, 29(7):402-404. DOI: 10.3969/j.issn.1003-1634.2013.07.007. Wan Y, Shang ZJ. Epidemiologic study of risk factors of gingival carcinoma[J]. J Clin Stomatol, 2013, 29(7):402-404. DOI: 10.3969/j.issn.1003-1634.2013.07.007.
[13]	伍俊锋, 林李嵩, 陈法, 等.福建省口腔鳞状细胞癌预后指数构建[J].中华流行病学杂志, 2018, 39(6):841-846. DOI: 10.3760/cma.j.issn.0254-6450.2018.06.028. Wu JF, Lin LS, Chen F, et al. A novel prognostic index for oral cancer in Fujian Province[J]. Chin J Epidemiol, 2018, 39(6):841-846. DOI: 10.3760/cma.j.issn.0254-6450.2018.06.028.
[14]	尚伟, 郑家伟.口腔及口咽癌新版TNM分期与NCCN诊治指南部分解读[J].中国口腔颌面外科杂志, 2018, 16(6):58-71. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgkqhmwkzz201806011 Shang W, Zheng JW. Interpretation of the new TNM classification and the NCCN guidelines for cancers of the oral cavity and oropharynx[J]. China Journal of Oral and Maxillofacial Surgery, 2018, 16(6):58-71. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgkqhmwkzz201806011
[15]	Liu FQ, Chen F, Huang JF, et al. Prospective study on factors affecting the prognosis of oral cancer in a Chinese population.[J]. Oncotarget, 2017, 8(3):4352-4359. DOI: 10.18632/oncotarget.13842.
[16]	Overholt SM. Eicher SA, Wolf P, et al. Prognostic factors affecting outcome in lower gingival carcinoma[J]. Laryngoscope, 1996, 106(11):1335-1339. DOI: 10.1097/00005537-199611000-00006.
[17]	Lo WL, Kao SY, Chi LY, et al. Outcomes of oral squamous cell carcinoma in Taiwan after surgical therapy:factors affecting survival[J]. J Oral Maxillofac Surg, 2003, 61(7):751-758. DOI: 10.1016/S0278-2391(03)00149-6.
[18]	尤金强, 王军, 王平. 127例原发齿龈鳞状细胞癌预后分析[J].中国肿瘤临床, 2005, 32(8):445-447. DOI: 10.3969/j.issn.1000-8179.2005.08.008. You JQ, Wang J, Wang P, et al. Prognostic analysis on 127 patients with squamous cell carcinoma of the gingiva[J]. Chin J Clin Oncol, 2005, 32(8):445-447. DOI: 10.3969/j.issn.1000-8179.2005.08.008.
[19]	Bell RB, Kademani D, Homer L, et al. Tongue cancer:is there a difference in survival compared with other subsites in the oral cavity?[J]. J Oral Maxillofac Surg, 2007, 65(2):229-236. DOI: 10.1016/j.joms.2005.11.094.
[20]	Hasegawa T, Shibuya Y, Takeda D, et al. Prognosis of oral squamous cell carcinoma patients with level Ⅳ/Ⅴ metastasis:an observational study[J]. J Craniomaxillofac Surg, 2016, 45(1):145-149. DOI: 10.1016/j.jcms.2016.10.011.
[21]	杨洪, 宋晓萌, 陈洁, 等. 68例下颌牙龈鳞癌预后影响因素分析[J].中国口腔颌面外科杂志, 2014, 12(6):530-534. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgkqhmwkzz201406010 Yang H, Song XM, Chen J, et al. Analysis of prognostic factors in lower gingival squamous cell carcinoma[J]. China Journal of Oral and Maxillofacial Surgery, 2014, 12(6):530-534. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgkqhmwkzz201406010
[22]	Fu XY, Chen SW, Chen WC, et al. Clinical analysis of second primary gingival squamous cell carcinoma after radiotherapy.[J]. Oral oncology, 2018, 84:20-24. DOI: 10.1016/j.oraloncology.2018.06.018.
[23]	邱蔚六.口腔癌诊治的现状和趋向[J].中华口腔医学杂志, 2002, 37(2):81-83. DOI: 10.3760/j.issn:1002-0098.2002.02.001. Qiu WL. The current status and trends for management of oral cancer[J]. Chin J Stomatol, 2002, 37(2):81-83. DOI: 10.3760/j.issn:1002-0098.2002.02.001.
[24]	Liu CH, Chen HJ, Wang PC, et al. Patterns of recurrence and second primary tumors in oral squamous cell carcinoma treated with surgery alone[J]. Kaohsiung J Med Sci, 2013, 29(10):554-559. DOI: 10.1016/j.kjms.2013.03.001.
[25]	王蕊, 孙慧玲.牙龈癌的MSCT诊断及外科治疗[J].医学影像学杂志, 2015, 25(10):1753-1755. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yxyxxzz201510015 Wang R, Sun Hl. MSCT diagnosis and surgical treatment of gingival carcinoma[J]. J Med Imaging, 2015, 25(10):1753-1755. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yxyxxzz201510015
[26]	宿伟鹏, 赵化荣.头颈部肿瘤治疗进展[J].新疆医科大学学报, 2013, 36(2):259-265. DOI: 10.3969/j.issn.1009-5551.2013.02.025. Su WP, Zhao HR. Advances in therapy of tumours of head and neck[J]. J Xinjiang Med Univ, 2013, 36(2):259-265. DOI: 10.3969/j.issn.1009-5551.2013.02.025.
[27]	Kim KY, Li SJ, Cha JD, et al. Significance of molecular markers in survival prediction of oral squamous cell carcinoma[J]. Head Neck, 2012, 34(7):929-936. DOI: 10.1002/hed.21856.