Advancing the implementation of variant-level functional data into clinical databases and clinical practice

摘要 限制基因组医学变革临床治疗潜力的主要挑战之一是 目前对大多数人类基因变异的功能缺乏了解。因此，大多数人 临床上遇到的遗传变异被归类为不确定意义的变异(Untify SignalfifiCance，VUS)，这种变异不能 鉴于它们与疾病的未知关系，可用于临床决策。VUS的问题是 对于来自历史上被排除在研究之外的人群的个人来说，尤其严重，使现有的 实施基因组医学时的医疗保健不平等。 不同水平的功能数据有可能通过提供致病性来克服其中的许多挑战 来自不同人群的变异信息。例如，我们和其他人已经证明了 变异效应的多重分析(MAVES)可以在临床重要基因中分辨出很大一部分vus (例如，49%的BRCA1 VU、69%的TP53 VU和93%的DDX3X VU)，以及多个美国和国际 联盟目前正在使用MAVE为所有可能的编码和非编码变体产生功能数据 与所有与人类疾病相关的基因有关。考虑到功能信息的潜力 为了加强基因组医学的实施，最近更新了临床指南以 在解释变异致病时，建议使用变异水平的功能数据。然而，尽管 大规模功能数据集的潜在临床效用，如何最好地实施它们，以便临床医生 能否将不同的功能数据适当地结合到临床实践中仍是未知的。 在这项提案中，我们的目标是使用以下方法解决基因组医学中这一未得到满足的需求： 首先，我们将生成一个框架，用于标准化和传播经过策划的大型功能 将数据转化为面向临床医生的资源，并将此框架实施到ClinVar中(目标1) 其次，我们将在ClinVar中执行概念验证集成，将不同级别的函数数据集成到两个 评估不同水平功能数据的临床吸收和影响的大型临床实践(目标2)。 最后，我们将建立和传播培训临床医生的资源，使他们了解整合 将功能数据转化为临床实践(目标3)。 总体而言，这项建议有可能显著推进基因组数据在临床上的实施 通过使临床医生能够适当地利用新出现的不同级别的功能数据来解决VU， 从而使基因组医学更加公平和有效。

SUMMARY One of the major challenges limiting the potential of genomic medicine to revolutionize clinical care is the current lack of knowledge about the function of most human genetic variants. Consequently, the majority of clinically encountered genetic variants are classified as variants of uncertain significance (VUS), which cannot be used for clinical decision making given their unknown relationship to disease. The VUS problem is particularly acute for individuals from populations historically excluded from research, compounding existing healthcare inequities when implementing genomic medicine. Variant-level functional data has the potential to overcome many of these challenges by providing pathogenicity information for variants from diverse populations. For example, we and others have demonstrated that multiplexed assays of variant effect (MAVEs) can resolve a large fraction of VUS in clinically important genes (e.g., 49% of BRCA1 VUS, 69% of TP53 VUS, and 93% of DDX3X VUS), and several U.S. and international consortia are currently using MAVEs to produce functional data for all possible coding and non-coding variants associated with all genes that have been linked to human disease. Given the potential of functional information to augment the implementation of genomic medicine, clinical guidelines have recently been updated to recommend the use of variant-level functional data when interpreting variant pathogenicity. However, despite the potential clinical utility of large-scale functional datasets, how best to implement them such that clinicians can appropriately incorporate variant functional data into clinical practice remains unknown. In this proposal we aim to address this unmet need in genomic medicine using the following approach: ● First, we will generate a framework for standardizing and disseminating curated large-scale functional data into clinician-facing resources, and implement this framework into ClinVar (Aim 1) ● Second, we will perform a proof-of-concept integration of variant level functional data in ClinVar into two large clinical practices to evaluate the clinical uptake and impact of variant-level functional data (Aim 2). ● Finally, we will build and disseminate resources for training clinicians on best practices for integrating functional data into clinical practice (Aim 3). Overall, this proposal has the potential to significantly advance the implementation of genomic data into clinical practice by enabling clinicians to appropriately leverage emerging variant-level functional data to resolve VUS, thereby making genomic medicine more equitable and impactful.