创伤后应激障碍(PTSD)和帕金森病(PD)有越来越多的全身性炎症的证据,然而,推定的生物标记的发现的不一致和缺乏可复制性延迟了这一领域的进展。平台之间性能的差异可能导致生物标志物文献中缺乏共识,正如许多精神疾病所见,包括PTSD.因此,需要高性能,可扩展,以及用于发现和开发用于药物开发和临床诊断的炎症生物标志物的验证平台。为了确定用于未来生物标志物发现工作的最佳平台,我们对5种领先的平台技术进行了全面的跨平台和交叉分析评估.最初的评估集中在四种与PTSD有关的细胞因子-白细胞介素(IL)-1β,IL-6,肿瘤坏死因子(TNF)-α,和干扰素(IFN)-γ。评估平台性能并了解脑部疾病患者的可能测量结果,血清和血浆样本来自PTSD患者(n=13)或帕金森病患者(n=14)以及健康对照(n=5)。我们比较了许多常见分析参数的平台性能,包括测定精度,灵敏度,内源性分析物检测频率(FEAD),平台之间的相关性,使用连续稀释系列测量细胞因子的并行性。单分子阵列(Simoa™)超敏感平台(Quanterix),MESOV-Plex(中尺度发现),和LuminexxMAP®(Myriad)由各自的供应商进行,而Luminex®和Quantikine®高灵敏度ELISA测定法由R&DSystem的生物标志物测试服务进行评估。在所有分析物和临床群体中检测内源性分析物的灵敏度最高的测定(即最高的FEAD),是Simoa™平台。相比之下,MESOV-plex的性能变化更大,研发Luminex®和Quantikine®,而Myriad的LuminexxMAP®在所有分析物和样品中表现出低FEAD。Simoa™还在检测内源性细胞因子方面表现出高精度,如<20%的变异系数(%CV)在重复运行中来自健康对照样品,PTSD患者,PD患者。相比之下,MESOV-Plex,R&DLuminex®和Quantikine®在细胞因子的精度方面具有不同的性能。MyriadLuminexxMAP®无法包含在精度估计中,因为供应商未重复运行样品。对于跨平台性能比较,观察到IL-6的跨平台相关性最高,因此除Myriad的LuminexxMAP®外,所有平台在IL-6测量中都具有强相关性(r范围=0.59-0.86).对于其他细胞因子,跨平台的相关性很低甚至没有,这样报告了IL-1β的测量结果,TNF-α,和IFN-γ在不同的试验中变化。一起来看,这些发现提供了新的证据,表明免疫测定的选择可以极大地影响所报告的细胞因子的发现.当前的研究提供了有关平台之间和免疫分析之间性能变化的关键信息,这些信息可能有助于在未来的研究中选择分析。Further,结果强调,随着新技术的出现,需要对免疫测定进行比较评估,特别是考虑到缺乏细胞因子定量评估的参考标准。
There is mounting evidence of systemic inflammation in post-traumatic stress disorder (PTSD) and Parkinson\'s disease (PD), yet inconsistency and a lack of replicability in findings of putative biological markers have delayed progress in this space. Variability in performance between platforms may contribute to the lack of consensus in the biomarker literature, as has been seen for a number of psychiatric disorders, including PTSD. Thus, there is a need for high-performance, scalable, and validated platforms for the discovery and development of biomarkers of inflammation for use in drug development and as clinical diagnostics. To identify the best platform for use in future biomarker discovery efforts, we conducted a comprehensive cross-platform and cross-assay evaluation across five leading platform technologies. This initial assessment focused on four cytokines that have been implicated PTSD - interleukin (IL)-1β, IL-6, tumor necrosis factor (TNF)-α, and interferon (IFN)-γ. To assess platform performance and understand likely measurements in individuals with brain disorders, serum and plasma samples were obtained from individuals with PTSD (n = 13) or Parkinson\'s Disease (n = 14) as well as healthy controls (n = 5). We compared platform performance across a number of common analytic parameters, including assay precision, sensitivity, frequency of endogenous analyte detection (FEAD), correlation between platforms, and parallelism in measurement of cytokines using a serial dilution series. The single molecule array (Simoa™) ultra-sensitive platform (Quanterix), MESO V-Plex (Mesoscale Discovery), and Luminex xMAP® (Myriad) were conducted by their respective vendors, while Luminex® and Quantikine® high-sensitivity ELISA assays were evaluated by R&D System\'s Biomarker Testing Services. The assay with the highest sensitivity in detecting endogenous analytes across all analytes and clinical populations (i.e. the highest FEAD), was the Simoa™ platform. In contrast, more variable performance was observed for MESO V-plex, R&D Luminex® and Quantikine®, while Myriad\'s Luminex xMAP® exhibited low FEAD across all analytes and samples. Simoa™ also demonstrated high precision in detecting endogenous cytokines, as reflected in < 20 percent coefficient of variance (%CV) across replicate runs for samples from the healthy controls, PTSD patients, and PD patients. In contrast, MESO V-Plex, R&D Luminex® and Quantikine® had variable performance in terms of precision across cytokines. Myriad Luminex xMAP® could not be included in precision estimates because the vendor did not run samples in duplicate. For cross-platform performance comparisons, the highest cross-platform correlations were observed for IL-6 such that all platforms - except for Myriad\'s Luminex xMAP® - had strong correlations with one another in measurements of IL-6 (r range = 0.59 - 0.86). For the other cytokines, there was low to no correlation across platforms, such that reported measurements of IL-1β, TNF-α, and IFN-γ varied across assays. Taken together, these findings provide novel evidence that the choice of immunoassay could greatly impact reported cytokine findings. The current study provides crucial information on the variability in performance between platforms and across immunoassays that may help inform the selection of assay in future research studies. Further, the results emphasize the need for performing comparative evaluations of immunoassays as new technologies emerge over time, particularly given the lack of reference standards for the quantitative assessments of cytokines.