Abstract:

Objective: To construct a prognostic nomogram model combined with the clinical features of patients with head and neck squamous cell carcinoma (HNSCC) and to predict the disease-specific analysis, based on Surveillance, Epidemiology, and End Results (SEER)—an authoritative cancer statistical database.Methods: All patients diagnosed with HNSCC from 2004 to 2016 were selected from SEER registry database, including age, gender, race, marital status, primary tumor site, tumor stage (TNM), surgical situation, etc. The population epidemiological characteristics, prognostic factors, and treatment-related outcomes were calculated after adjustment for different variables. Cox proportional-hazards model was applied to achieve univariate and multivariate analysis. Finally, the performance evaluation model was used to determine the prediction model.Results: A total of 1 026 cases' available information was found in SEER database. Multivariate survival analysis of Cox proportional-hazards model showed that the risk of death in female patients was higher than that in male patients (P<0.001); patients aged over 60 years had higher risk of death than that in patients aged younger than 60 years (P<0.001); for the primary tumor site, taking tongue as a reference, the death risk of buccal mucosa, maxillary sinus, the floor of mouth was lower (P<0.05) and that in other locations (gingiva, palate, oropharynx, nasopharynx, larynx) was higher (P<0.05); in TNM stage, the risks of stageⅡ (P<0.001), stageⅢ (P<0.001) and stageⅣ (P<0.001) were relatively higher compared with stageⅠ patients; similarly, in terms of radiochemotherapy and immunotherapy, patients who did not receive radiochemotherapy or immunotherapy had an increased risk of death compared with patients who received radiochemotherapy or immunotherapy (P<0.001).Conclusion: Gender, HPV infection, TNM stage, radiotherapy and chemotherapy are the important factors for the survival of patients suffered from HNSCC. Patients with high risk factors should be paid more attention to after diagnosis, which is helpful for doctors to make clinical decisions in actual therapies.

Key words: head and neck squamous cell carcinoma, prognostic analysis, database of SEER, nomogram

摘要：

目的： 基于美国国家癌症研究所监测，流行病学和最终结果 (Surveillance，Epidemiology，and End Results， SEER)数据库，构建头颈部鳞状细胞癌（head and neck squamous cell carcinoma，HNSCC）患者列线图模型并进行预后分析。方法： 利用SEER肿瘤登记注册数据库，提取2004—2016年诊断为HNSCC的所有患者的信息，包括年龄、性别、种族、婚姻状况、肿瘤部位、肿瘤分期（TNM）、手术情况等分类数据，并计算人群流行病学特征、预后因素和调整不同变量后的治疗相关结果；使用Cox比例风险模型进行单变量和多变量分析；利用绩效评价模型确定预测模型。结果： SEER数据库中总共检索到1 026例HNSCC可用信息。Cox回归模型多因素分析结果显示，女性患者的死亡风险高于男性患者(P<0.001)；年龄≥60岁患者的死亡风险高于年龄<60岁患者(P<0.001)；从发病部位来看，以舌为参照，除颊黏膜、上颌窦、口底处的死亡风险低于舌以外(P<0.05)，牙龈、腭、口咽、鼻咽及喉部的死亡风险均高于舌(P<0.05)；TNM分期中，以Ⅰ期患者为参照，Ⅱ期(P<0.001)、Ⅲ期(P<0.001)和Ⅳ期(P<0.001)风险相对更高；在放疗与化疗方面，相比接受化疗或放疗患者，未行化疗或放疗患者的死亡风险增加(P<0.001)；在免疫治疗方面，未行免疫治疗的患者死亡风险增加(P<0.001)。结论： 年龄、HPV感染情况、TNM分期、放化疗等均是HNSCC患者生存的重要影响因素，对具有高危险因素的患者在诊断后应给予更多关注，这也有助于医生在实际治疗过程中进行临床决策。

关键词: 头颈部鳞状细胞癌, 预后分析, SEER数据库, 列线图

CLC Number: 

• R782

• R739.8