非小细胞肺癌 (NSCLC) 的放射治疗可能会对灌注肺造成放射损伤。灌注损失可以通过正电子断层扫描发射（PET）灌注成像来测量，但是，这种方式可能无法广泛使用。对比双能计算机断层扫描 (DECT) 可能是 PET/CT 的替代方案。这项工作的目的是研究 PET 和 DECT 在 NSCLC 中确定的剂量反应曲线 (DRC) 的等效性。来自前瞻性临床试验盲审 (BLINDED FOR REVIEW) 的 PET 和 DECT 数据集包含在本次预先计划中试验分析。患者在基线/治疗后 3 个月/12 个月的同一天接受 68Ga 大聚集白蛋白 (68Ga-MAA) PET/CT 检查和 DECT 对比检查。灌注肺是根据 PET/DECT 中最大标准化摄取值 (%SUVmax)/碘浓度 (%IoMax) 的阈值来定义的。 PET 和 DECT DRC 之间的等效性是通过比较 (1) 两个 DRC 归一化重叠的平均值 (aNO)（范围从 0（无重叠）到 1（完全重叠））和 (2) a 的斜率来确定的。应用于 DRC 的线性模型。在参加临床试验的 19 名患者中，14/10 的患者分别在中位数 = 4.5 个月/13.5 个月时进行了治疗后影像学检查。在 30%SUVmax/35%IoMax 下，aNO 最大化，并且线性模型的 PET 和 DECT 斜率之间的差异在每个时间点最小化（斜率 = 3 个月时的 0.76%/Gy / 0.75%/Gy 和 0.86%/Gy） / 0.87%/Gy（12 个月时通过 PET / DECT 确定）。在 NSCLC 患者治疗后 3 个月和 12 个月时，通过 DECT 确定的剂量反应关系与 PET 相当。版权所有 © 2023。由 Elsevier Inc. 出版。

Radiotherapy treatment for non-small cell lung cancer (NSCLC) may result in radiation damage to the perfused lung. The loss in perfusion may be measured from positron tomography emission (PET) perfusion imaging, however, this modality may not be widely available. Dual-energy computed tomography (DECT) with contrast may be an alternative to PET/CT. The purpose of this work is to investigate the equivalence of dose-response curves (DRC) determined from PET and DECT in NSCLC.PET and DECT datasets from the prospective clinical trial BLINDED FOR REVIEW (BLINDED FOR REVIEW) were included in this pre-planned trial analysis. Patients underwent 68Ga-macroaggregated albumin (68Ga-MAA) PET/CT examination and DECT with contrast on the same day at baseline/3-months/12-months post-treatment. The perfused lung was defined from a threshold based on the maximum standardized uptake value (%SUVmax)/iodine concentration (%IoMax) in PET/DECT. The equivalence between PET and DECT DRC was established by comparing (1) the average of the normalized overlap of the two DRCs (aNO), ranging from 0 (no overlap) to 1 (perfect overlap), and (2) the slope of a linear model applied to DRCs.Out of the 19 patients enrolled in the clinical trial, 14/10 patients had a post-treatment imaging session at median = 4.5 months/13.5 months, respectively. With 30%SUVmax/35%IoMax, aNO was maximized and the difference between PET and DECT slope of the linear model was minimized at each time point (slope = 0.76%/Gy / 0.75%/Gy at 3 months and 0.86%/Gy / 0.87%/Gy at 12 months determined from PET / DECT).Dose-response relationship determined from DECT was comparable to PET at 3- and 12-months post-treatment in NSCLC patients.Copyright © 2023. Published by Elsevier Inc.