Outcomes in AYA survivors of pediatric medulloblastoma.

摘要 这项临床研究确定了对认知结果有贡献的稳健因素 髓母细胞瘤(MB)脑瘤的儿童、青少年和年轻成人幸存者(PAYA) 位于小脑。儿童脑瘤幸存者认知风险增加 损害(CI)，与以下人群相比，健康状况较差和残疾的可能性很大 没有癌症病史，相对于没有接受救生的幸存者 神经毒性放化疗。然而，认知结果的个体差异 在甲基溴的付款中，即使是最有可能的临床贡献者，也是多样和广泛的 相同(即，治疗时的年龄、放化疗风险水平、MB肿瘤亚型)。这导致了我们 研究遗传疾病和其他因素的团队可能解释广泛的 结果，从轻度到重度CI。除了遗传学之外，还需要填补重要的空白 到目前为止的研究，以确定临床风险因素和环境资源，以确定 脑梗塞风险最大的人群。这个在美国国家癌症研究所指定的癌症中心的多地点项目位于佐治亚州， 和OH将研究多因素的环境资源(例如，社区社会经济 逆境、物质困难、照顾能力和学校质量)、临床因素(例如，年龄 在治疗时，化疗风险水平、MB肿瘤类型)和个人遗传素质(即， 候选单核苷酸多态变异(SNPs))。首先，在目标1中，研究将审查 三个领域中的每一个都独立使用最先进的方法和创新的分析 找出CI的最佳预测因子。在其他人群中，多基因风险分数(PR)具有 提供了比单一SNPs更强的结果预测。因此，在基因范围内 素质领域，我们将研究靶向SNPs和邻近的交互突变来创建 一种利用机器学习工具对调控RNA表达的SNP进行加权的稳健粗糙集 与CI相关，并包含已知的表观遗传学、转录和 蛋白质。其次，在目标2中，我们将创建多域风险算法，使用最敏感的 三个领域(即临床风险、环境资源和遗传因素)的CI预测因素 素质)。这一经验性的风险算法将通过识别 适应风险的放化疗的条件和预防干预措施， 缓解和管理CI。与2018年STAR法案和Precision Medicine保持一致 这些发现将有助于及早识别有CI风险的个人，并提供 治疗目标，以改善帕亚斯的整体生活质量和适应功能 儿童癌症的病症。

Abstract This clinical research identifies the robust factors that contribute to cognitive outcomes in pediatric, adolescent, and young adult survivors (PAYAS) of medulloblastoma (MB) brain tumors located in the cerebellum. Pediatric brain tumor survivors are at increased risk of cognitive impairment (CI) and have substantial likelihood of poor health and disability relative to those who do not have a cancer history, and relative to survivors who did not have to undergo lifesaving neurotoxic chemoradiation treatment. However, individual differences in cognitive outcomes among PAYAS of MB are diverse and wide ranging, even when the most likely clinical contributors are the same (i.e., age at treatment, chemoradiation level of risk, MB tumor subtype). This led our team to examine genetic diatheses and other contributors that may explain the wide range of outcomes, from mild to severe CI. In addition to genetics, there is a need to fill important gaps in the research to date to identify both clinical risk factors and environmental resources that identify those at greatest risk of CI. This multisite project at NCI-designated cancer centers in GA, AL, and OH will examine multifactorial environmental resources (e.g., neighborhood socioeconomic adversity, material hardships, caregiving capacity, and school quality), clinical factors (e.g., age at treatment, chemoradiation level of risk, MB tumor type), and individual genetic diathesis (i.e., candidate single nucleotide polymorphism variants (SNPs)). First, in Aim 1, the study will examine each of the three domains independently using state of the art methods and innovative analyses to identify the best predictors of CI. In other populations, polygenetic risk scores (PRS) have provided a stronger prediction of outcomes than single SNPs alone. Therefore, within the genetic diathesis domain, we will examine targeted SNPs and neighboring interactive mutations to create a robust PRS utilizing machine learning tools that weights SNPs regulating RNA expression associated with CI, and incorporates known functional impact of epigenetics, transcriptions and proteins. Second, in Aim 2, we will create a multi-domain risk algorithm, using the most sensitive predictors of CI across the three domains (i.e., clinical risks, environmental resources, and genetic diathesis). This empirically derived risk algorithm will inform precision medicine, by identifying conditions for risk-adapted chemoradiation treatment and prophylactic interventions to prevent, mitigate and manage CI. Consistent with the STAR Act of 2018 and the Precision Medicine Initiative, these findings will allow for early identification of individuals at risk for CI and provide targets for treatments in order to improve overall quality of life and adaptive functioning in PAYAS of childhood cancer.