NEW YORK – Researchers at several South Korean cancer institutions are planning a 20,000-sample plasma proteomic study aimed at identifying biomarkers that could aid in early detection of cancer, particularly in young adults.

纽约——韩国几家癌症研究机构的研究人员正计划开展一项涉及20,000份血浆样本的蛋白质组学研究，旨在识别有助于癌症早期检测的生物标志物，特别是针对年轻人。

The project, which will use Thermo Fisher Scientific's Orbitrap Astral instrument and Seer's recently launched Proteograph One sample enrichment system, indicates mass spectrometry's growing suitability as a tool for large-scale proteomic analyses, an area that has largely been dominated by affinity-based platforms from companies like Olink (now a part of Thermo Fisher) and Standard BioTools..

该项目将使用赛默飞世尔科技的Orbitrap Astral仪器和Seer最近推出的Proteograph One样品富集系统，表明质谱技术作为大规模蛋白质组学分析工具的适用性日益增强，这一领域长期以来主要由Olink（现为赛默飞世尔的一部分）和Standard BioTools等公司基于亲和力的平台所主导。

Sang-Won Lee, a professor of chemistry at the Center of Proteogenome Research (CPGR) of Korea University and one of the leaders of the effort, said that with the combination of the Astral instrument, Seer's Proteograph One, and the CPGR's

韩国大学蛋白质基因组研究中心（CPGR）化学教授、该项目的负责人之一李相元表示，结合Astral仪器、Seer的Proteograph One以及CPGR的资源，

dual online liquid chromatography technology

双在线液相色谱技术

, which boosts assay throughput by eliminating the time needed for LC column equilibration between runs, the researchers 'are in a good position to analyze population-scale sample [cohorts].'

，由于无需在运行之间进行液相色谱柱平衡，从而提高了分析通量，研究人员“处于有利位置来分析大规模样本群体。”

Lee said that with his lab's LC-MS setup, he and his colleagues are able to analyze around 60 samples per day with depth of coverage of around 7,000 proteins per sample.

李表示，通过其实验室的 LC-MS 设置，他和同事们每天能够分析大约 60 个样本，每个样本的覆盖深度约为 7,000 种蛋白质。

Such levels of depth and throughput in a plasma proteomics experiment were until recently unobtainable using mass spectrometry. New advances in instrumentation and sample preparation are helping mass spec-based workflows narrow the gap with affinity methods. In addition to the South Korean effort launched this week, researchers at the Max Planck Institute of Biochemistry are using mass spec to analyze the plasma proteomes of 50,000 individuals to identify biomarkers linked to maternal and fetal health.

直到最近，使用质谱技术还无法获得如此深度和通量的血浆蛋白质组学实验数据。仪器和样品制备方面的新进展正在帮助基于质谱的工作流程缩小与亲和方法之间的差距。除了本周启动的韩国研究项目外，马克斯·普朗克生物化学研究所的研究人员也在利用质谱技术分析50,000个人的血浆蛋白质组，以识别与母婴健康相关的生物标志物。

.

。

That project

那个项目

is using a Thermo Fisher Orbitrap Astral and a perchloric acid-based sample enrichment workflow called PCA-N to analyze samples at a depth of around 2,000 proteins and will take approximately one year.

使用赛默飞Orbitrap Astral质谱仪和一种名为PCA-N的基于高氯酸的样品富集工作流程，以大约2000种蛋白质的深度分析样品，预计耗时约一年。

While both projects are still significantly smaller than the largest ongoing affinity-based plasma proteome study — UK Biobank's Pharma Proteomics Project (PPP) effort using Olink's Explore HT platform

尽管这两个项目仍然比目前规模最大的基于亲和力的血浆蛋白质组研究——英国生物银行的药物蛋白组学项目（PPP）使用Olink的Explore HT平台——要小得多。

to measure

测量

up to 5,400 proteins in 600,000 blood samples — they are comparable to a recently completed

高达5400种蛋白质在60万个血液样本中——它们与最近完成的项目相当。

PPP effort

PPP 努力

using Olink to analyze samples from roughly 54,000 UKB participants and demonstrate the feasibility of using mass spec for cohorts of this size.

使用Olink分析了来自大约54,000名UKB参与者的样本，并证明了在如此规模的队列中使用质谱的可行性。

Lee highlighted the depth of coverage provided by the Seer Proteograph One assay, noting that measuring 7,000 proteins in plasma 'is quite remarkable.'

李强调了Seer Proteograph One检测方法提供的覆盖深度，指出在血浆中测量7000种蛋白质“非常了不起”。

Seer's technology uses nanoparticles to boost proteomic coverage and is based on the observation that when incubated in a biological sample, nanoparticles collect proteins, which form a 'corona' that researchers can then isolate and analyze. While past Proteograph assays have required that samples be incubated with multiple nanoparticles in multiple wells, which required multiple mass spec experiments per sample, the Proteograph One, which the company launched this week at the American Society for Mass Spectrometry annual meeting in Baltimore, lets this incubation take place in a single well, allowing for one mass spec run per sample.

Seer的技术利用纳米颗粒提升蛋白质组覆盖范围，其基础是观察到当纳米颗粒在生物样本中孵育时，会收集蛋白质并形成一个“冠状层”，研究人员可以将其分离并分析。过去，Proteograph检测需要将样本在多个孔中与多种纳米颗粒孵育，每个样本需要进行多次质谱实验，而该公司本周在巴尔的摩举行的美国质谱学会年会上推出的Proteograph One系统，则可以在单个孔中完成孵育，每个样本只需进行一次质谱运行。

Using this more streamlined system, researchers can prepare up to 80 samples for mass spec analysis in five hours..

使用这种更精简的系统，研究人员可以在五小时内为质谱分析准备多达80个样本。

Seer Chair and CEO Omid Farokhzad said that he believes mass spec experiments analyzing hundreds of thousands of plasma samples are now feasible. He also noted that as the scale of experiments continues to rise, the per-sample price of Seer's technology — which has been a sticking point for some users — will go down..

Seer公司主席兼首席执行官Omid Farokhzad表示，他认为现在进行分析数十万份血浆样本的质谱实验是可行的。他还指出，随着实验规模的不断扩大，Seer技术的单样本成本（这一直是部分用户的一个痛点）将会下降。

'It's becoming very competitive pricing-wise,' he said. 'Early on when people said Seer is expensive, that was because we had a lot of customers doing tens of samples, maybe some doing hundreds of samples. But now I have customers … coming in who want to study 10,000 samples. That [customer] is going to get a very significantly discounted price.'.

“从价格角度来看，竞争变得非常激烈，”他说道。“早期当人们说Seer很贵时，那是因为我们有许多客户进行几十个样本的测试，可能有些人做几百个样本。但现在我有客户……进来想要研究一万个样本。这样的客户将获得非常大的折扣。”

Farokhzad declined to give a price range for the Proteograph One but said it is competitive with pricing for Olink's and Standard BioTools' affinity-based platforms. Seer said in a statement that per-sample cost has fallen by roughly 60 percent since 2021.

法罗赫扎德拒绝透露 Proteograph One 的价格范围，但表示其定价与 Olink 和 Standard BioTools 的基于亲和力的平台具有竞争力。Seer 在一份声明中表示，自 2021 年以来，单样本成本已下降了大约 60%。

Lee said that he and his colleagues are running an initial discovery experiment consisting of 5,000 samples taken from individuals spanning 10 different cancer types and 500 healthy controls. This portion of the project will take around three months, he said. The researchers then plan to follow up their findings in that cohort with a 5,000-sample verification experiment and a 10,000-sample validation experiment.

李表示，他和他的同事正在进行一项初步发现实验，该实验包括从涵盖10种不同癌症类型的个体中提取的5000个样本和500个健康对照。他说，该项目的这一部分将耗时约三个月。研究人员随后计划在该队列中通过一项5000个样本的验证实验和一项10000个样本的确认实验来跟进他们的发现。

Lee said they hope to complete the data collection in around two years..

李表示，他们希望在大约两年内完成数据收集工作。

The samples are being sourced from the Seoul National University Hospital, the National Cancer Center, and Samsung Medical Center, and the study is being supported by funding from the South Korean government's K-Health MIRAE initiative, which Lee described as analogous to the Advanced Research Projects Agency for Health (ARPA-H) in the US..

样本来源于首尔国立大学医院、国家癌症中心和三星医疗中心，该研究得到了韩国政府的K-Health MIRAE计划的资助，李博士将其比作美国的健康高级研究计划局（ARPA-H）。

The project is focused on identifying markers for detecting cancers in young adults, which Lee said comprise a relatively large proportion of cancer cases in South Korea.

该项目专注于识别检测年轻人癌症的标志物，李表示，在韩国，年轻人癌症病例占相当大的比例。

'It is a real national health issue,' he said. 'We want to have some kind of diagnostic method to detect cancers early on for young adults.'

“这是一个真正的国家健康问题，”他说。“我们希望有一种诊断方法可以早期检测年轻人的癌症。”

The effort is entirely proteomics-focused, Lee said, adding that one factor driving the decision to use mass spec-based methods for the study is the depth of information provided by mass spec beyond just the protein identification and relative quantification data generated by affinity platforms. He noted that while using the Seer-Astral setup, he and his colleagues are able to measure around 7,000 proteins per sample.

李表示，这项工作完全专注于蛋白质组学，并补充说，决定在研究中使用基于质谱的方法的一个推动因素是质谱提供的信息深度，远超亲和平台生成的蛋白质鉴定和相对定量数据。他指出，在使用Seer-Astral设置时，他和同事们能够测量每个样本中大约7,000种蛋白质。

The system collects large amounts of information that do not inform protein identifications but could nonetheless prove useful..

该系统收集大量信息，这些信息虽然不能确定蛋白质身份，但可能仍然有用。

The 7,000 proteins measured per sample represent the roughly 20 percent of MS/MS spectra that could be confidently linked back to an MS spectra, enabling a protein identification. That, Lee noted, means roughly 80 percent of the MS/MS data generated by the mass spec are not being used.

每个样本测量的7000种蛋白质代表了大约20%的MS/MS谱图，这些谱图可以被可靠地链接回MS谱图，从而实现蛋白质鉴定。李指出，这意味着质谱仪生成的MS/MS数据中大约80%未被使用。

New methods of data analysis may allow the researchers to make better use of that data in the future, improving the experiments' depth of coverage, Lee said. In the present, he and his colleagues are using AI to look for patterns in these unidentified MS/MS spectra that can distinguish between cancer cases and controls..

李表示，未来数据分析的新方法可能会让研究人员更好地利用这些数据，提高实验的覆盖深度。目前，他和他的同事正在使用人工智能寻找这些未识别的MS/MS谱图中的模式，以区分癌症病例和对照组。

'Instead of trying to identify those unidentified MS/MS spectra by using a conventional mass spec proteomics approach, we decided to use an ID-free AI approach where we find the hidden patterns in those MS/MS spectra and use those patterns to diagnose cancer,' Lee said.

“我们没有选择使用传统的质谱蛋白质组学方法来鉴定那些未识别的MS/MS光谱，而是决定采用一种无需鉴定的AI方法，即在这些MS/MS光谱中找到隐藏的模式，并利用这些模式来诊断癌症，”李说。

The approach is similar to one employed by South Korean

这种方法与韩国采用的方法类似

proteomics firm Bertis

蛋白质组学公司Bertis

, which Lee said is also participating in the initiative. Bertis likewise uses AI to analyze unidentified MS/MS spectra to look for patterns that could be used for early detection of cancer. The company, which in 2023 launched a San Diego-based US subsidiary, Bertis Bioscience, is developing tests for pancreatic and ovarian cancer.

，李表示也参与了这项计划。Bertis同样使用人工智能分析未识别的MS/MS光谱，寻找可用于早期癌症检测的模式。这家在2023年成立了位于圣地亚哥的美国子公司Bertis Bioscience的公司，正在开发胰腺癌和卵巢癌的检测方法。

It offers a proteomic breast cancer test, Mastocheck, which is intended as a complement to mammography in women with dense breasts, in Asian markets including South Korea and Singapore and in the Middle East..

它提供了一种蛋白质组学乳腺癌检测方法Mastocheck，旨在作为乳腺致密的女性在亚洲市场（包括韩国和新加坡）以及中东地区乳腺X光检查的补充。

Lee said Bertis is a likely candidate for commercializing discoveries that emerge from the project.

李表示，伯蒂斯是将该项目中出现的发现商业化的可能候选人。