Blood metabolomics as a next-generation cancer diagnostic platform technology

癌症的早期诊断可以改善患者的预后，挽救生命。不幸的是，目前很多癌症的诊断都很晚，处于治疗选择有限的阶段。这一问题的一个潜在原因是常规成像技术（如CT扫描和MRI）的处方不完善，这两种技术都是常用的癌症诊断方法。目前，几乎所有的癌症都是通过观察有症状的病人的影像来诊断的，但医生们的工作是在不完整的支持信息下进行的，并且需要在何时给病人成像以及成像结果表明什么（如果有的话）方面做出不完美的决定。除了患者预后较差之外，目前的方法对医疗保健系统来说既昂贵又浪费，同时也给患者及其家属带来了压力。氧组学技术通过一种简单、通用的血液测试来诊断癌症，再加上机器学习，可以识别血液中被称为代谢物的小分子的癌症特异性模式。我们的测试适合现有的患者护理途径，并提供快速的早期诊断，允许在正确的时间为正确的患者进行正确的影像学检查，从而改善患者的预后。我们已经证明，我们的技术可以用于在服务不足的非特异性症状患者中准确发现癌症，包括在通过成像发现癌症之前发现癌症的病例。这个应用程序的目标是进一步发展我们的技术，使它可以作为一种规范的临床测试，可以被医生每天使用。为此，我们将开发基础分类软件，建立特定关键癌症的临床证据，并与监管机构合作，确保我们测试的质量和可重复性。

Early diagnosis of cancer improves patient outcomes and saves lives. Unfortunately, a lot of cancers are currently diagnosed late, at stages where treatment options are limited. One underlying cause of this problem is imperfect prescription of routine imaging technologies like CT scanning and MRI, both of which are commonly used for cancer diagnosis. Currently, almost all cancers are diagnosed by observing a symptomatic patient with imaging, but doctors are working with incomplete supporting information and a requirement to make imperfect calls about when to image patients and what imaging results indicate, if anything at all. Apart from poorer patient outcomes, the current approach is both costly and wasteful for the healthcare system while also being stressful for patients and their families.Oxomics technology diagnoses cancer using a simple, universal blood test, coupled with machine learning which identifies cancer-specific patterns of small molecules in the blood called metabolites. Our test fits into existing patient care pathways and offers rapid earlier diagnoses, allowing referral of the right patients for the right imaging at the right time, improving patient outcomes. We have already shown that our technology can be used to accurately find cancer in an under-served population of patients with nonspecific symptoms, including cases where we found cancers before they were visible by imaging.The goal of this application is to develop our technology further so that it is available as a regulated clinical test that can be used every day by doctors. We will do this by developing the underlying classification software, building clinical evidence in specific key cancers and working with regulators to ensure quality and reproducibility from our tests.