This study aimed to establish a nomogram for the prognostic prediction of patients with early-onset lung cancer (EOLC) in both overall survival (OS) and cancer-specific survival (CSS). We retrieved EOLC patients diagnosed between 2004 and 2015 from the Surveillance, Epidemiology, and End Results (SEER) database and further divided them into training and validation sets randomly. The prognostic nomogram for predicting 3-, 5- and 10-years OS and CSS was established based on the relative clinical variables determined by the multivariate Cox analysis results. Furthermore, the predictive performance of the nomogram was assessed by concordance index (C-index), calibration curve, receiver operating characteristic (ROC) curve and decision curve analysis (DCA) curve. A total of 1,822 EOLC patients were selected and randomized into a training cohort (1,275, 70%) and a validation cohort (547, 30%). The nomograms were established based on the statistical results of Cox analysis. In training set, the C-indexes for OS and CSS prediction were 0.797 (95% confidence interval [CI]: 0.773-0.818) and 0.794 (95%CI:0.771-0.816). Significant agreement in the calibration curves was noticed in the nomogram models. The results of ROC and DCA indicated nomograms possessed better predict performance compared with TNM-stage and SEER-stage. Furthermore, the areas under the curve (AUC) of the nomogram for OS and CSS prediction in ROC analysis were 0.766 (95%CI:0.745-0.787) and 0.782 (95%CI:0.760-0.804) respectively. In conclusion, the prognostic nomogram provided an accurate prediction of 3-, 5-, and 10-year OS and CSS of EOLC patients which contributed clinicians to optimize individualized treatment plans.

中文翻译：

---

45 岁以下肺癌患者预后列线图的开发和验证。

本研究旨在建立一个列线图，用于预测早发性肺癌 (EOLC) 患者的总生存期 (OS) 和癌症特异性生存期 (CSS)。我们从监测、流行病学和最终结果 (SEER) 数据库中检索了 2004 年至 2015 年间诊断出的 EOLC 患者，并进一步将其随机分为训练集和验证集。用于预测 3、5 和 10 年 OS 和 CSS 的预后列线图是基于由多变量 Cox 分析结果确定的相关临床变量建立的。此外，列线图的预测性能通过一致性指数 (C-index)、校准曲线、受试者工作特征 (ROC) 曲线和决策曲线分析 (DCA) 曲线进行评估。总共选择了 1,822 名 EOLC 患者并将其随机分配到一个训练队列中（1,275，70%) 和验证队列 (547, 30%)。列线图是基于 Cox 分析的统计结果建立的。在训练集中，OS 和 CSS 预测的 C 指数为 0.797（95% 置信区间 [CI]：0.773-0.818）和 0.794（95%CI：0.771-0.816）。在列线图模型中注意到校准曲线的显着一致性。ROC 和 DCA 的结果表明，与 TNM 阶段和 SEER 阶段相比，列线图具有更好的预测性能。此外，ROC 分析中 OS 和 CSS 预测列线图的曲线下面积 (AUC) 分别为 0.766 (95%CI:0.745-0.787) 和 0.782 (95%CI:0.760-0.804)。总之，预后列线图提供了对 EOLC 患者 3 年、5 年和 10 年 OS 和 CSS 的准确预测，有助于临床医生优化个性化治疗计划。