Department of Gynecology, Guangdong Second Provincial General Hospital, Guangzhou, China.
Department of Radiation Oncology, Nanfang Hospital, Southern Medical University, Guangzhou, China.
J Gynecol Oncol. 2023 Nov;34(6):e81. doi: 10.3802/jgo.2023.34.e81. Epub 2023 Jul 5.
To elucidate clinical characteristics and build a prognostic nomogram for patients with vulvar cancer.
The study population was drawn from the Surveillance, Epidemiology, and End Results (SEER) database. Patients were randomly assigned to training and validation sets. Cox proportional hazards model and competing risk model were used to identify the prognostic parameters of overall survival (OS) and cancer-specific survival (CSS) to construct a nomogram. The nomogram was assessed by concordance index (C-index), area under the curve (AUC), calibration plot, and decision curve analysis (DCA).
A total of 20,716 patients were included in epidemiological analysis, of whom 7,025 patients were selected in survival analysis, including 4,215 and 2,810 in training and validation sets, respectively. The multivariate Cox model showed that the predictors for OS were age, marital status, histopathology, differentiation and tumor node metastasis (TNM) stages, whether to undergo surgery and chemotherapy. However, the predictors for CSS were age, race, differentiation and TNM stages, whether to undergo surgery and radiation. The C-index for OS and CSS in the training set were 0.76 and 0.80. The AUC in the training set for 1-, 3- and 5-year OS and CSS were 0.84, 0.81, 0.80 and 0.88, 0.85, 0.83, respectively, which was similar in the validation set. The calibration curves showed good agreement between prediction and actual observations. DCA revealed that the nomogram had a better discrimination than TNM stages.
The nomogram showed accurate prognostic prediction in OS and CSS for vulvar cancer, which could provide guidance to clinical practice.
阐明外阴癌患者的临床特征并建立预后列线图。
研究人群来自监测、流行病学和最终结果（SEER）数据库。患者被随机分配到训练集和验证集中。使用 Cox 比例风险模型和竞争风险模型确定总生存（OS）和癌症特异性生存（CSS）的预后参数，以构建列线图。通过一致性指数（C 指数）、曲线下面积（AUC）、校准图和决策曲线分析（DCA）评估列线图。
共纳入 20716 例患者进行流行病学分析，其中 7025 例患者进行生存分析，包括训练集和验证集中的 4215 例和 2810 例。多变量 Cox 模型显示，OS 的预测因素为年龄、婚姻状况、组织病理学、分化程度和肿瘤淋巴结转移（TNM）分期、是否接受手术和化疗。然而，CSS 的预测因素为年龄、种族、分化程度和 TNM 分期、是否接受手术和放疗。训练集 OS 和 CSS 的 C 指数分别为 0.76 和 0.80。训练集中 1、3 和 5 年 OS 和 CSS 的 AUC 分别为 0.84、0.81、0.80 和 0.88、0.85、0.83，在验证集中也相似。校准曲线显示预测与实际观察之间具有良好的一致性。DCA 表明，列线图比 TNM 分期具有更好的区分能力。
列线图在外阴癌的 OS 和 CSS 中具有准确的预后预测能力，可为临床实践提供指导。
