Precision Medicine.
Precision Medicine.
复制标题
精密医学。
DOI：
10.1161/circulationaha.118.036781
复制
发表时间：
2018-11-13
期刊：
影响因子：
37.8
通讯作者：
Sobhani K
中科院分区：
文献类型：
--
作者：
Van Eyk JE;Sobhani K
循环。 2018； 138：2172–2174。 DOI：10.1161/流通AHA。 118.036781 2018 年 11 月 13 日 2173 个针对特定分析物（包括蛋白质）的核酸国家文库。适体对其靶标具有不同程度的亲和力和特异性，在这方面，它们的表现与抗体类似。例如，Williams 等人最近部署了 SOMAscan 适体平台，用于回顾性分析来自 ILLUMINATE 心血管药物试验（一项检查 Torcetrapib/阿托伐他汀和阿托伐他汀对心脏病患者临床心血管事件影响的研究）的血液样本，使用单个适体来捕获/识别目标蛋白。拥有单一捕获试剂意味着蛋白质鉴定的特异性仅基于单一结合区域。将此与夹心免疫测定（即 ELISA）进行对比，其中 2 个抗体（通常是单克隆抗体）必须与目标蛋白的 2 个不同（但相连）表位结合。当两种抗体的含量都超过靶标时，ELISA 即可实现出色的定量和特异性，并且在结合时（以三明治形式），产生与靶标浓度成正比的线性信号。或者，质谱分析通常基于多个肽的定量，每个肽对于目标蛋白来说都是独特的，因此它们代表不同的（但不相关的）表位。通过发现或靶向质谱法对每种肽进行的鉴定和定量是相同的，并且基于肽转换（肽氨基酸序列特有的分解片段），提供正确的鉴定（即特异性）和定量。在 Ngo 等人最近发表的一篇文章中 2，作者进行了 SOMAscan 研究，随后尝试通过基于靶向质谱的分析来验证其生物标志物命中。他们证明可以滴定目标蛋白标准品、建立部分滴定曲线并设置检测下限值（表中定义）。值得注意的是，他们还表明，除了目标蛋白之外，许多其他蛋白也与适体结合。正如 Gramolini 等人所指出的，3“需要确定所使用的每种适体是否容易受到脱靶配体的影响，以及靶标定量是否会因体内变化而受到干扰”（例如，蛋白质形式）。展望未来，必须评估每个适体（或任何抗体或捕获试剂）的分析特异性及其定量靶蛋白的能力。如果没有这个，就无法知道分析物的特异性。从蛋白质组学发现到临床检测的道路充满挑战。我们建议发现方法必须朝着分析特异性、灵敏度、精确度、再现性和准确性进行评估；这样做将为生物标志物检测和伴随诊断的开发提供可靠的途径，这些检测和伴随诊断从一开始就可以被视为临床有效。我们鼓励蛋白质组学发现界遵守分析和临床实验室科学制定的严格性和标准。通过基于质谱的发现方法，通常将数百到数千种蛋白质一起定量，许多团体已经了解应用参考标准、区分定性和定量测定、制定蛋白质/肽鉴定要求、评估再现性、建立系统适用性、推进自动化样品制备（以帮助实现准确性和再现性）以及结合生物信息学和数据共享以满足……的重要性。
Circulation. 2018; 138: 2172–2174. DOI: 10.1161/CIRCULATIONAHA. 118.036781 November 13, 2018 2173 natorial libraries of nucleic acids directed against specific analytes, including proteins. Aptamers have varying degrees of affinity and specificity for their targets, and, in this regard, they perform in a manner analogous to antibodies. With the SOMAscan aptamer platform, which was, for example, recently deployed by Williams et al1 to retrospectively analyze blood samples from the ILLUMINATE cardiovascular drug trial (A Study Examining Torcetrapib/Atorvastatin and Atorvastatin Effects on Clinical CV Events in Patients With Heart Disease), a single aptamer was used to capture/identify the target protein. Having a single capture reagent means that the specificity of protein identification is based solely on a single binding region. Contrast this to the sandwich immunoassay (ie, ELISA), where 2 antibodies (typically monoclonal antibodies) must bind to 2 different (but linked) epitopes of the target protein. Excellent quantification and specificity are achieved with ELISA when both antibodies are present in excess of the target and, when bound (in a sandwich-like format), produce a linear signal that is directly proportional to the target concentration. Alternatively, mass spectrometry assays are typically based on the quantification of multiple peptides, each being unique to the target protein such that they represent different (but unlinked) epitopes. The identification and quantification of each peptide by discovery or targeted mass spectrometry are the same and are based on peptide transitions (breakdown fragments that are unique to the amino acid sequence of the peptide) providing the correct identification (ie, specificity) and quantification. In a recent publication by Ngo et al, 2 the authors performed a SOMAscan study and subsequently attempted to validate their biomarker hits by targeted mass spectrometry–based analysis. They demonstrated that it is possible to titrate target protein standards, establish partial titration curves, and set lower limit-of-detection values (defined in the Table). It is important to note that they also showed that many other proteins bound to the aptamer besides their target protein. As pointed out by Gramolini et al, 3 “there is a need to ascertain whether each aptamer utilized is susceptible to off-target ligands, and whether target quantification is confounded by alterations in vivo”(eg, proteoforms). Moving forward, it will be imperative that each aptamer (or any antibody or capture reagent) is assessed for analytic specificity and its ability to quantitate target protein (s). Without this, analyte specificity is not known. The road from proteomic discovery to a clinical assay is challenging. We propose that discovery methods must move toward being assessed for analytic specificity, sensitivity, precision, reproducibility, and accuracy; doing so will provide a reliable path to the development of biomarker assays and companion diagnostics that can both be deemed clinically valid from inception. We encourage the proteomics discovery community to adhere to the rigor and standards set by both analytic and clinical laboratory science. With mass spectrometry–based discovery methods, where typically hundreds to thousands of proteins are quantified together, many groups have understood the importance of applying reference standards, differentiating between qualitative and quantitative assay, developing protein/peptide identification requirements, assessing reproducibility, establishing system suitability, advancing automated sample preparation (to help achieve accuracy and reproducibility), and incorporating bioinformatics and data sharing to meet …