Profiling circulating metabolites for the diagnosis of pulmonary hypertension

肺动脉高压（PAH）是一种进行性和致死性疾病。最初，PAH发展为 无症状或具有通常未被初级医生识别的轻微症状。不过就算 非特异性症状的发作，如劳力性呼吸困难和疲劳，PAH仍未被识别， ≥ 2年。平均而言，一个病人需要去看五次全科医生，三次心脏病专家/肺病专家 在患者被转诊进行右心室（RV）导管插入术之前进行访视，以确认诊断。的 PAH诊断延迟仍然是治疗效率低下的主要原因。因此，尽管重大 该领域的进展，包括更好地了解PAH发病机制和靶向治疗 尽管有多种方法，但这种疾病的预后仍然很差，特别是对于功能III级的患者， 四.因此，非常需要新的诊断工具来缩短诊断时间间隔， 症状出现后立即开始治疗。事实上，许多研究表明， 早期开始治疗的存活率。已知PAH会改变代谢特征。我们的研究 临床前PAH模型显示，代谢的改变发生在肺动脉压变化的阶段， 血流动力学轻度，无RV功能障碍证据。我们对病人血浆的初步数据 样本表明，循环代谢物的分析可以成为追踪PAH的有效工具， 早期和发展阶段。由Metfora发现并获得专利的循环代谢物的特定面板 LLC创始人，有效地区分PAH不仅从健康的个人，而且从个人与左 心脏病（LHD）、糖尿病（DM）和其他慢性疾病。在本申请中，我们建议使用 用于PAH鉴别诊断的基于质谱（MS）的平台，即，区分PAH和 症状相似的疾病。这些疾病包括高度流行的疾病，如慢性肺、肾和 Metfora开发了一种新的基于血液的诊断方法， 多重代谢组和机器学习/深度学习（ML/DL）模型，以实现早期PAH 诊断. Metfora的靶向方法可以将健康患者与慢性疾病患者分开， 肺部状况，并进一步识别肺部疾病，准确率为90%-98%。在第一阶段，我们 将使用更大的患者队列、更广泛的疾病谱和更广泛的 技术方法，如机器学习和深度学习（ML/DL）算法。完成后 第一阶段，我们将拥有一个严格优化和验证的平台，提供PAH的鉴别诊断。 如果成功，我们革命性的诊断方法将缩短PAH的诊断时间间隔， 几年到几天，开始早期治疗的背景，提高生活质量和生存率 PAH患者的发生率。

Pulmonary arterial hypertension (PAH) is a progressive and fatal disease. Initially, PAH develops asymptomatically or with mild symptoms that are often unrecognized by primary doctors. However, even after the onset of non-specific symptoms, such as dyspnea on exertion and fatigue, PAH remains unrecognized for ≥two years. On average, it takes a patient five visits to a general practitioner, and three cardiologist/pulmonologist visits before a patient is referred for the right ventricle (RV) catheterization, which confirms the diagnosis. The delay in PAH diagnosis remains the primary reason for a low efficiency of therapy. Therefore, despite significant advances in the field, which include a better understanding of PAH pathogenesis and targeted therapeutic approaches, the disease still carries a poor prognosis, especially for the patients with Functional Class III and IV. Thus, there is a significant need for novel diagnostic tools to shorten the time-to-diagnose interval and initiate therapy immediately after symptom onset. Indeed, numerous studies have shown a significantly better survival rate when the therapy is started early. PAH is known to alter the metabolic profile. Our research on the pre-clinical PAH model shows that alterations in metabolism occur at the stage when changes in pulmonary hemodynamics are mild, and no evidence of RV dysfunction is present. Our preliminary data on patients' plasma samples showed that the profiling of circulating metabolites could become an efficient tool for tracing PAH at early and developed stages. The specific panels of circulating metabolites, discovered and patented by Metfora LLC founders, efficiently distinguish PAH not only from healthy individuals, but also from individuals with left heart disease (LHD), diabetes mellitus (DM), and other chronic conditions. In this application, we propose to use the mass spectrometry (MS)-based platform for differential diagnosis of PAH, i.e., to distinguish PAH from diseases with similar symptoms. These include highly prevalent conditions such as chronic lung, kidney, and liver diseases, cancers, etc. Metfora has developed a novel blood-based diagnostic method that utilizes multiplexed metabolic panels and a Machine Learning/Deep Learning (ML/DL) model to enable earlier PAH diagnosis. Metfora’s targeted approach allows for the separation of healthy patients from those with chronic pulmonary conditions and further identifies the lung diseases with a precision of 90%-98%. During Phase I, we will optimize and validate these panels using larger patient cohorts, a wider spectrum of conditions, and broader technical approaches, such as Machine Learning and Deep Learning (ML/DL) algorithms. Upon completion of Phase I, we will have a rigorously optimized and validated platform that provides a differential diagnosis of PAH. If successful, our revolutionary diagnostic approach will shorten the time-to-diagnosis interval for PAH from two years to a few days, initiating background for the early treatments and increasing the quality of life and survival rate for PAH patients.