口腔医学研究

口腔医学研究 ›› 2020, Vol. 36 ›› Issue (10): 915-920.DOI: 10.13701/j.cnki.kqyxyj.2020.10.007

• 口腔肿瘤学研究 • 上一篇    下一篇

基于SEER数据库的唾液腺粘液表皮样癌预后因素分析

张曙光1, 王玉龙2, 徐文光3,4, 尹西腾3,4, 韩伟3,4, 邹会会3,4*   

  1. 1.淮安市肿瘤医院(淮安市淮安医院)口腔颌面外科 江苏 淮安 223200;
    2.复旦大学附属肿瘤医院 上海 200032;
    3.南京大学医学院附属口腔医院口腔颌面外科 江苏 南京 210008;
    4.南京大学医学院附属口腔医院中心实验室 江苏 南京 210093
  • 收稿日期:2020-03-04 出版日期:2020-10-28 发布日期:2020-10-19
  • 通讯作者: *邹会会,E-mail:weiwoshi@foxmail.com
  • 作者简介:张曙光(1972~ ),男,江苏人,本科,副主任医师,研究方向:口腔颌面肿瘤。
  • 基金资助:
    江苏省卫生青年医学重点人才(编号:QNRC2016121)

Prognostic Factors of Mucoepidermoid Carcinoma of Salivary Glands: a SEER Database-based Study

ZHANG Shuguang1, WANG Yulong2, XU Wenguang3,4, YIN Xiteng3,4, HAN Wei3,4, ZOU Huihui3,4*   

  1. 1. Department of Stomatology, Huai' an Tumour Hospital (Huai'an Hospital), Huai'an 223200, China;
    2. Department of Oncology, Shanghai Medical College, Fudan University, Shanghai 200032, China;
    3. Department of Oral and Maxillofacial Surgery, Nanjing Stomatological Hospital, Medical School of Nanjing University, Nanjing 210008, China;
    4. Central Laboratory of Stomatology, Nanjing Stomatological Hospital, Medical School of Nanjing University, Nanjing 210093, China
  • Received:2020-03-04 Online:2020-10-28 Published:2020-10-19

PDF (PC)

207

摘要： 目的: 本研究利用SRER(Surveillance, Epidemiology, and End Results)数据库对唾液腺粘液表皮样癌(Mucoepidermoid carcinoma,MEC)的临床病理资料进行统计分析并筛选其预后风险因素。方法: 下载2004~2015年间美国SEER数据库中唾液腺MEC患者临床病理数据和随访资料。首先将SEER数据库纳入的MEC患者分为建模组和验证组,接着将国内数据库中的MEC数据作为独立的外部验证组。用单因素分析初步筛选了影响患者预后的临床病理因素;随后采用COX模型多因素分析获取影响患者预后的独立因素,并构建列线图(nomogram)并验证模型区分能力与一致性。结果: 共纳入2209例唾液腺MEC患者,其中建模组1657例,验证组552例。从南京大学附属口腔医院和复旦大学附属肿瘤医院的肿瘤数据库中纳入234例唾液腺MEC患者进行外部验证。单因素分析结果显示年龄、性别、临床分期、T分期、N分期、M分期、组织学分级、是否手术、是否行颈部淋巴结清扫是影响患者预后的重要临床病理因素;多因素分析表明年龄、T分期、N分期、M分期、组织学分级、是否手术、是否行颈淋巴结清扫是影响患者预后的独立因素。成功构建了个体化预测唾液腺MEC患者预后的列线图,区分度检验显示一致性指数(concordance index C-index)为0.794(95%CI 0.744-0.844),表明该模型达到一个良好的区分度。建模组和内外部验证组的一致性检验的结果表明列线图所预测的生存率与实际生存率具有较好的一致性。结论: 年龄、性别、临床分期、T分期、N分期、M分期、组织学分级、是否行颈部淋巴结清扫是影响唾液腺MEC患者的临床病理因素。本文构建的列线图可以较好地预测唾液腺MEC患者的预后。

关键词: 唾液腺, 粘液表皮样癌, SEER数据库, 预后, 列线图

Abstract: Objective: To analyze and screen prognostic risk factors of salivary gland MEC. Methods: The clinicopathological data and follow-up data of salivary gland MEC patients were downloaded from the SEER (surveillance, epidemiology, and end results) database in the United States between 2004 and 2015. Patients were randomly divided into the training and validation cohorts. The MEC patients in the domestic database were selected as an independent external verification group. The univariate analysis was initially employed to screen the clinicopathological factors that affected the prognosis of MEC patients. The multivariate analysis was used to obtain independent factors affecting the prognosis of patients. The nomogram was constructed and the distinguishing ability and consistency of the model were tested. Results: A total of 2209 patients with MEC of salivary glands were included in this study. Patients were assigned as training set (n=1657), and the rest were selected as SEER validation set (n=552). An external validation was performed by a set of independent 234 MEC patients from Nanjing Stomatological Hospital and Fu Dan Hospital in China .The univariate analysis showed that age, gender, clinical stage, T stage, N stage, M stage, and histological grade neck dissection were important clinicopathological factors affecting the prognosis of patients. Multivariate analysis showed that age, T stage, N stage, M stage, histological grade, and neck dissection were independent factors affecting the prognosis of MEC patients of salivary glands. An individualized nomogram was successfully constructed to predict the prognosis of patients with MEC of salivary glands, and the discriminability test showed that the C-index was 0.794 (95%CI 0.744-0.844), indicating that the model reached a good discriminability. The consistency test showed that the predicted survival rate by the nomogram was consistent with the actual survival rate. Conclusion: Age, gender, clinical staging, T staging, N staging, M staging, histological grading, and neck dissection were the prognostic factors of MEC patients in salivary glands. The constructed nomogram could effectively predict the prognosis of patients with salivary gland MEC.

Key words: salivary gland, mucoepidermoid carcinoma, SEER database, prognosis, nomogram