A precision medicine approach to improve prediction of severe toxicity in fluorouracil chemotherapy

提高氟尿嘧啶化疗严重毒性预测的精准医学方法

• 批准号: 10436992
• 负责人: Steven Offer
• 金额: $ 36.37万
• 依托单位: MAYO CLINIC ROCHESTER
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-01 至 2026-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10436992
• 关键词: Address; African American population; Alleles; American; Applications Grants; Biological Markers; Breast; Cancer Burden; Cancer Patient; Clinical; Clinical Research; Clinical Trials; Collection; Colorectal Cancer; Coupled; DPYD gene; Data; Data Collection; Dihydropyrimidine Dehydrogenase; Dose; Enzymes; Ethnic Origin; European; Financial Hardship; Fluorouracil; Genetic; Goals; Guidelines; Haplotypes; Impairment; Incidence; Individual; Life; Machine Learning; Malignant Neoplasms; Masks; Measurement; Medicine; Metabolism; Pathway interactions; Patients; Pharmaceutical Preparations; Pharmacogenetics; Plasma; Population; Public Health; Publishing; Quality of life; Race; Reaction; Research Personnel; Risk; Safety; Severities; Signal Transduction; Structure; Systemic Therapy; Testing; Therapeutic; Therapeutic Uses; Toxic effect; Treatment Efficacy; Underrepresented Populations; Variant; Western Europe; Work; adverse drug reaction; base; candidate marker; chemotherapy; clinically relevant; dose individualization; expectation; experience; genetic variant; high risk; improved; medication safety; multidimensional data; novel; precision medicine; predictive modeling; predictive test; risk variant; sample collection; side effect; volunteer

PROJECT SUMMARY/ABSTRACT My long-term goal is to improve the toxicity profiles for cancer therapeutics. One-third of cancer patients treated with the commonly prescribed chemotherapeutic 5-fluorouracil (5-FU) experience severe and life- threatening toxicity to standard doses of the drug. An appreciable fraction of those patients die—not due to cancer, but because of side-effects related to treatment. Clinical studies indicate that the majority of patients who experience severe toxicity to 5-FU are deficient for an enzyme called dihydropyrimidine dehydrogenase (DPD, DPYD gene); however, only four genetic variants in DPYD have been adequately characterized to be considered predictive of 5-FU toxicity in clinical studies. My preliminary studies demonstrate that these four variants explain only a small fraction of severe 5-FU toxicities and have exceedingly limited clinical value outside of individuals with European ancestry. The primary objective of the studies proposed in this grant application is to identify additional biomarkers of 5-FU toxicity risk that can be used to individualize 5-FU dosing with the goal of improving the safety profile for the drug. My overall hypothesis is that expanded biomarker-based pre-treatment tests will more accurately identify patients with DPD deficiency, as well as the relative degree to which the DPD function is impaired, enabling more accurate dose optimization. My rationale is that improved biomarker-based approaches to dose individualization have strong potential to improve the safety profile for this commonly used therapeutic. Aim #1 will identify risk alleles for severe 5-FU–related toxicity in understudied populations. Aim #2 will characterize multi-marker haplotype contributions to 5-FU toxicity. In Aim #3, I will develop an integrated predictive model of 5-FU toxicity using deep machine learning. It is my expectation that the proposed studies, which will leverage multiple large patient and volunteer data and specimen collections to address various aspects of my primary hypothesis, will answer key questions that have vexed pharmacogenetics researchers for decades. In doing so, the proposed studies are expected to identify clinically relevant biomarkers that can be used to improve the safety profile of 5-FU through dose optimization.

项目摘要/摘要 我的长期目标是改善癌症治疗药物的毒性。三分之一的癌症患者 用常用的化疗药物5-氟尿嘧啶(5-FU)治疗后，会出现严重的生命危险- 对标准剂量的药物有毒性威胁。这些患者中有相当一部分死亡--不是由于 癌症，但因为治疗相关的副作用。临床研究表明，大多数患者 对5-FU有严重毒性的人缺乏一种名为二氢嘧啶脱氢酶的酶 (DPD，DPYD基因)；然而，DPYD中只有四个遗传变异被充分描述为 在临床研究中被认为是5-FU毒性的预测。我的初步研究表明，这四人 变种只能解释一小部分严重的5-FU毒性，并且临床价值非常有限 在欧洲血统的个人之外。这项拨款建议进行的研究的主要目的 应用是确定可用于个体化的5-FU毒性风险的其他生物标记物 给药的目的是改善药物的安全性。我的总体假设是扩展后的 基于生物标记物的治疗前测试将更准确地识别DPD缺乏症患者，以及 DPD功能受损的相对程度，可实现更准确的剂量优化。我的理论基础 改进的基于生物标记物的剂量个体化方法有很大的潜力来改善 这种常用的治疗药物的安全性简介。目标1将确定严重5-FU相关的风险等位基因 在研究不足的人群中的毒性。目的#2将描述多标记单倍型对5-FU的贡献 毒性。在目标3中，我将使用深度机器学习开发一个5-FU毒性的集成预测模型。它 我的期望是，拟议的研究将利用多个大型患者和志愿者数据，并 样本收集，以解决我的主要假设的各个方面，将回答以下关键问题 几十年来一直困扰着药物遗传学研究人员。在这样做的过程中，拟议的研究有望确定 临床相关的生物标志物，可用于通过剂量优化改善5-FU的安全性。