医学
一致性
组内相关
接收机工作特性
逻辑回归
置信区间
切断
回顾性队列研究
Pet成像
队列
放射科
核医学
试验预测值
认知功能衰退
外科
队列研究
卡帕
疾病
内科学
正电子发射断层摄影术
曲线下面积
吻合
代理终结点
尤登J统计
疾病严重程度
认知
定量评估
目视检查
临床试验
作者
Luqiang Jin,Xinyu Wang,Fahuan Song,Jie Hou,Yueqian Bi,Xuehua Wen,Xu Wang,Fangyuan Lai,Aiping Cheng
标识
DOI:10.1097/mnm.0000000000002184
摘要
OBJECTIVES: This study aimed to validate the clinical utility of visual and software-based quantitative AV-45 PET analyses and compare treatment effects between lecanemab and deep cervical lymphatic-venous anastomosis (dcLVA). METHODS: This retrospective cohort study included Alzheimer's disease patients who received 6-month lecanemab therapy or dcLVA surgery between July 2024 and 2025. AV-45 PET was performed 1 week before and 6 months after treatment. Standardized uptake value ratios (SUVRs) and centiloids (CLs) were obtained using visual and quantitative analyses. Agreement was assessed using Cohen's κ and intraclass correlation coefficients (ICC). Receiver operating characteristic (ROC) analysis evaluated diagnostic performance. Logistic regression used Informant Questionnaire on Cognitive Decline in the Elderly score greater than or equal to 3.3 as the outcome. RESULTS: Baseline characteristics were comparable between groups (all P > 0.05). Visual assessment showed good agreement with centiloid quantification ( κ = 0.72, ICC = 0.663). ROC analysis identified ΔCL as the optimal marker (area under the curve = 0.764, P = 0.026), with a cutoff of -11.5%. ΔCL ≥ -11.5% independently predicted cognitive decline in the lecanemab group (odds ratio = 10.281, 95% confidence interval = 1.289-82.005; P = 0.028). Cognitive decline was less frequent in the lecanemab group (23.5 vs. 81.8%; P < 0.001). ΔCL differed significantly between groups ( P = 0.006), whereas ΔSUVR did not. CONCLUSION: Visual and centiloid-based AV-45 PET analyses show good concordance for monitoring Alzheimer's disease treatment response. ΔCL is a robust marker of lecanemab efficacy, which is superior to dcLVA in reducing amyloid-β burden and delaying cognitive decline.
科研通智能强力驱动
Strongly Powered by AbleSci AI