Smith Ruben, Shaw Leslie, Palmqvist Sebastian, Mattsson-Carlgren Niklas, Klein Gregory, Tonietto Matteo, Quijano-Rubio Clara, Rank Christopher M, Andreadou Myrto, Burnham Samantha C, Stomrud Erik
Clinical Memory Research Unit, Department of Clinical Sciences Malmö, Lund University, Lund, Sweden.
Memory Clinic, Skåne University Hospital, Malmö, Sweden.
Neurol Ther. 2025 Jul 20. doi: 10.1007/s40120-025-00798-8.
Amyloid- and tau-positron emission tomography (PET) are promising modalities for detecting pathological changes associated with Alzheimer's disease (AD); however, their application is limited. Although the use of cerebrospinal fluid (CSF) biomarkers as alternatives to amyloid-PET and tau-PET has been explored, no in vitro diagnostic-approved or commercial CSF biomarker assays are currently available for detecting tau pathology in clinical practice.
In this study, we determined and validated the optimal cutoff value for the ratio of tau phosphorylated at a threonine residue at position 181 (pTau) to β-amyloid(1-42) (Aβ) in CSF, measured with the Elecsys Phospho-Tau (181P) CSF and β-Amyloid(1-42) CSF immunoassays (Roche Diagnostics International Ltd, Rotkreuz, Switzerland), based on its concordance with binary tau-PET status. Clinical performance was explored using CSF measurements and tau-PET scans retrospectively obtained from a subset of subjects with mild cognitive impairment and dementia due to AD in two independent cohorts, Alzheimer's Disease Neuroimaging Initiative-2/3 (ADNI-2/3; N = 133) and Swedish BioFINDER-2 (N = 62).
In the first part of this analysis (ADNI-2/3 pre-analytics), a CSF pTau/Aβ cutoff value of 0.0395 was selected as the best compromise between positive percent agreement (PPA) and negative percent agreement (NPA) for the tau-PET visual read endpoint. After adjustment to account for differences between the ADNI-specific protocol and the manufacturer's recommended pre-analytical protocol used in BioFINDER-2, the optimal CSF pTau/Aβ cutoff value was set at 0.037. The adjusted cutoff was validated in BioFINDER-2 and was associated with a PPA of 85.7% (95% confidence interval [CI] 70.6, 93.7), NPA of 70.4% (95% CI 51.5, 84.1), and overall percent agreement (OPA) of 79.0% (95% CI 67.4, 87.3); the positive and negative likelihood ratios were 2.89 and 0.203, respectively.
The Elecsys CSF pTau/Aβ ratio may be a reliable tool for identifying tau pathology in clinical practice.
淀粉样蛋白和tau蛋白正电子发射断层扫描（PET）是检测与阿尔茨海默病（AD）相关病理变化的有前景的方法；然而，它们的应用有限。尽管已经探索了使用脑脊液（CSF）生物标志物作为淀粉样蛋白PET和tau蛋白PET的替代方法，但目前尚无经体外诊断批准或商业化的CSF生物标志物检测方法可用于临床实践中检测tau蛋白病理。
在本研究中，我们使用Elecsys磷酸化tau蛋白（181P）CSF和β-淀粉样蛋白（1-42）CSF免疫分析（罗氏诊断国际有限公司，瑞士罗特kreuz），基于其与二元tau蛋白PET状态的一致性，确定并验证了CSF中苏氨酸残基181位磷酸化tau蛋白（pTau）与β-淀粉样蛋白（1-42）（Aβ）比值的最佳临界值。使用从两个独立队列中因AD导致轻度认知障碍和痴呆的受试者亚组中回顾性获得的CSF测量值和tau蛋白PET扫描，探索临床性能，这两个队列分别是阿尔茨海默病神经影像学倡议-2/3（ADNI-2/3；N = 133）和瑞典生物标志物发现者-2（N = 62）。
在本分析的第一部分（ADNI-2/3分析前），选择CSF pTau/Aβ临界值0.0395作为tau蛋白PET视觉读取终点的阳性百分比一致性（PPA）和阴性百分比一致性（NPA）之间的最佳折衷值。在调整以考虑ADNI特定方案与生物标志物发现者-2中使用的制造商推荐的分析前方案之间的差异后，最佳CSF pTau/Aβ临界值设定为0.037。调整后的临界值在生物标志物发现者-2中得到验证，其PPA为85.7%（95%置信区间[CI]70.6，93.7），NPA为70.4%（95%CI 51.5，84.1），总体百分比一致性（OPA）为79.0%（95%CI 67.4，87.3）；阳性和阴性似然比分别为2.89和0.203。
Elecsys CSF pTau/Aβ比值可能是临床实践中识别tau蛋白病理的可靠工具。
