We combined the metabolic features of 18F-FDG-PET/CT and hematological inflammatory indicators to establish a predictive model of the outcomes of patients with locally advanced non-small cell lung cancer (LA-NSCLC) receiving concurrent chemoradiotherapy. A predictive nomogram was developed based on sex, CEA, systemic immune-inflammation index (SII), mean SUV (SUVmean), and total lesion glycolysis (TLG). The nomogram presents nice discrimination that yielded an AUC of 0.76 (95% confidence interval: 0.66–0.86) to predict 1-year PFS, with a sensitivity of 63.6%, a specificity of 83.3%, a positive predictive value of 83.7%, and a negative predictive value of 62.9% in the training set. The calibration curves and DCA suggested that the nomogram had good calibration and fit, as well as promising clinical effectiveness in the training set. In addition, survival analysis indicated that patients in the low-risk group had a significantly longer mPFS than those in the high-risk group (16.8 months versus 8.4 months, P < 0.001). Those results were supported by the results in the internal and external test sets. The newly constructed predictive nomogram model presented promising discrimination, calibration, and clinical applicability and can be used as an individualized prognostic tool to facilitate precision treatment in clinical practice.

中文翻译：

---

基于 PET/CT 代谢特征和血液炎症指标的局部晚期 NSCLC 预后列线图模型的开发和验证

我们结合 18F-FDG-PET/CT 的代谢特征和血液学炎症指标，建立了接受同步放化疗的局部晚期非小细胞肺癌（LA-NSCLC）患者结局的预测模型。根据性别、CEA、全身免疫炎症指数 (SII)、平均 SUV (SUVmean) 和总病变糖酵解 (TLG) 开发了预测列线图。列线图具有很好的区分性，预测 1 年 PFS 的 AUC 为 0.76（95% 置信区间：0.66-0.86），敏感性为 63.6%，特异性为 83.3%，阳性预测值为 83.7%，并且训练集中的阴性预测值为 62.9%。校准曲线和 DCA 表明列线图具有良好的校准和拟合度，并且在训练集中具有良好的临床效果。此外，生存分析表明，低风险组患者的 mPFS 显着长于高风险组患者（16.8 个月 vs 8.4 个月，P < 0.001）。这些结果得到了内部和外部测试集结果的支持。新构建的预测列线图模型具有良好的区分性、校准性和临床适用性，可作为个体化预后工具，促进临床实践中的精准治疗。