建立基于常规CT征象和肿瘤微环境免疫类型（TIMT）的模型来预测非小细胞肺癌（NSCLC）术后辅助化疗的持久临床获益（DCB）。总共205名NSCLC患者接受了手术术前 CT 并分为两组：DCB（无进展生存期 (PFS) ≥ 18 个月）和非 DCB（NDCB，PFS <18 个月）。对 PD-L1 和 CD8 肿瘤浸润淋巴细胞 (TIL) 的密度百分位数进行量化以估计 TIMT。收集临床特征和常规CT征象。采用多元逻辑回归来选择最具辨别力的参数，构建预测模型，并将模型可视化为列线图。采用受试者工作特征（ROC）曲线、校准曲线和决策曲线分析（DCA）来评估预测性能和临床实用性。 118 例 DCB 患者和 87 例 NDCB NSCLC 患者接受术后辅助化疗。 DCB 组和 NDCB 组之间的 TIMT 存在统计学差异 (P < 0.05)。临床特征（神经元特异性烯醇化酶、鳞状细胞癌抗原、Ki-76 和 cM 分期）和常规 CT 征象（毛刺、气泡状透明度、胸膜回缩、最大直径和静脉期 CT 值）在不同患者之间有所不同。四个 TIMT 组（P < 0.05）。此外，DCB 组和 NDCB 组之间的临床特征（淋巴细胞计数 [LYMPH] 和 cM 分期）和常规 CT 征象（气泡样透明和胸腔积液）也不同 (P < 0.05)。多变量分析显示 TIMT、cM 分期、LYMPH 和胸腔积液与 DCB 独立相关，并用于构建列线图。组合模型的曲线下面积 (AUC) 为 0.70 (95%CI: 0.64-0.76)，敏感性和特异性分别为 0.73 和 0.60。 传统 CT 征象和 TIMT 为预测临床结果提供了一种有前途的方法。接受术后辅助化疗的 NSCLC 患者。版权所有 © 2024 大学放射科医生协会。由爱思唯尔公司出版。保留所有权利。

To develop a model based on conventional CT signs and the tumor microenvironment immune types (TIMT) to predict the durable clinical benefits (DCB) of postoperative adjuvant chemotherapy in non-small cell lung cancer (NSCLC).A total of 205 patients with NSCLC underwent preoperative CT and were divided into two groups: DCB (progression-free survival (PFS) ≥ 18 months) and non-DCB (NDCB, PFS <18 months). The density percentiles of PD-L1 and CD8 + tumor-infiltrating lymphocytes (TIL) were quantified to estimate the TIMT. Clinical characteristics and conventional CT signs were collected. Multivariate logistic regression was employed to select the most discriminating parameters, construct a predictive model, and visualize the model as a nomogram. Receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA) were used to evaluate prediction performance and clinical utility.Precisely 118 patients with DCB and 87 with NDCB in NSCLC received postoperative adjuvant chemotherapy. TIMT was statistically different between the DCB and NDCB groups (P < 0.05). Clinical characteristics (neuron-specific enolase, squamous cell carcinoma antigen, Ki-76, and cM stage) and conventional CT signs (spiculation, bubble-like lucency, pleural retraction, maximum diameter, and CT value of the venous phase) varied between the four TIMT groups (P < 0.05). Furthermore, clinical characteristics (lymphocyte count [LYMPH] and cM stage) and conventional CT signs (bubble-like lucency and Pleural effusion) differed between the DCB and NDCB groups (P < 0.05). Multivariate analysis revealed that TIMT, cM stage, LYMPH, and pleural effusion were independently associated with DCB and were used to construct a nomogram. The area under the curve (AUC) of the combined model was 0.70 (95%CI: 0.64-0.76), with sensitivity and specificity of 0.73 and 0.60, respectively.Conventional CT signs and the TIMT offer a promising approach to predicting clinical outcomes for patients treated with postoperative adjuvant chemotherapy in NSCLC.Copyright © 2024 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.