Improving combination chemotherapy of tuberculosis: a computational approach

项目概要/摘要 结核病（TB）是一种广泛存在的细菌性传染病，每年造成近150万人死亡。而 结核病有效药物治疗已有50多年历史， 耐药的临床病例对公共卫生产生重大影响。结核病的药物治疗方案旨在限制 同时使用多种药物（联合治疗）会出现耐药性， 开发的时间和成本。虽然美国食品和药物管理局（FDA）与 最近成立的新结核病药物方案关键路径（CPTR）倡议，现在提供监管指导 开发新的药物组合作为一个单一的单位，而几个新的抗结核病方案在临床上， 在FDA的指导下进行测试，关于如何确定最佳剂量的关键问题 这些新的联合治疗方案中的每一种药物。 新的抗结核药物组合的剂量方案通常基于找到最佳剂量， 每一个单一的药物在临床前阶段，并通过第二阶段的剂量范围临床试验。在裁剪 药物组合中每种单独药物的剂量可能会产生更有效和更好的效果。 耐受的治疗方案，体外方法学的指数增加，动物有效性研究， 为确定治疗结核病所需的三种或三种以上药物组合的剂量， 昂贵得令人望而却步。为了解决结核病药物开发中的这一差距，我们提出了一种新的剂量方法 联合药物治疗的方案设计，包括（1）使用常规的临床前和 为结核病特定艾德药物组合提供数学剂量反应模型的临床测量 患者，（2）将此数学模型与生物启发的遗传算法相结合，以设计 剂量方案的方式类似于自然选择，和（3）这些优化的经验评价 实验性TB感染模型中的治疗方案。 为了建立我们的方法与临床相关的例子，我们将设计优化的剂量方案 对于新的抗结核组合pretomanid +莫西沙星+吡嗪酰胺（PaMZ）， 耐多药（MDR）结核病患者所需的治疗选择，目前正在II期临床试验中进行评估 审判这种结核病药物组合有大量高质量的临床前和临床数据， 提供了一个健全的证据基础，以发展我们的计算框架，并测试我们的结论。成功 拟议目标的完成将建立新的方法和工具，以更好地转化临床前研究 为将来的抗结核联合用药设计临床剂量方案。虽然结核病药物的需求 发展，该项目包括适用于治疗其他疾病，如癌症， 人类免疫缺陷病毒（HIV）感染和疟疾。

Project Summary/Abstract Tuberculosis (TB) is a widespread bacterial infectious disease that kills nearly 1.5 million people annually. While effective drug therapy for TB has been available for more than 50 years, there is a substantial number of drug resistant clinical cases that are significantly impacting public health. Drug regimens for TB are designed to limit the emergence of resistance by using multiple drugs concurrently (combination therapy) which greatly increases the time and cost of their development. While the U.S. Food and Drug Administration (FDA), in partnership with the recently formed Critical Path to New TB Drug Regimens (CPTR) initiative, now provides regulatory guidance for developing new drug combinations as a single unit, and while several new anti-TB regimens are in clinical testing under this FDA guidance, there are critical questions about how to establish the optimal dose of each individual drug within these new combination regimens. Dosage regimens for new anti-TB drug combinations are generally based on finding an optimal dose for every single drug in the preclinical stage, and through Phase II dose-ranging clinical trials. While tailoring the doses of each individual drug within a drug combination could potentially yield a more effective and better tolerated treatment regimen, the exponential increase in the in vitro methodologies, animal efficacy studies, and clinical testing required to identify such doses for combinations of three or more drugs needed for TB would be prohibitively expensive. To address this gap in TB drug development we propose a new approach to dosage regimen design of combination drug therapies that consists of (1) the use of conventional preclinical and clinical measurements to inform a mathematical dose-response model for a specified drug combination in TB patients, (2) the integration of this mathematical model with a biologically inspired genetic algorithm to design dosage regimens in a manner analogous to natural selection, and (3) the empirical evaluation of these optimized regimens in experimental TB-infection models. To establish our approach with a clinically relevant example, we will design optimized dosage regimens for the new anti-TB combination pretomanid + moxifloxacin + pyrazinamide (PaMZ); a promising and urgently needed treatment option for patients with multidrug resistant (MDR) TB, currently assessed in a Phase II clinical trial. There is a large amount of high quality preclinical and clinical data for this TB drug combination that will provide a sound evidence base to develop our computational framework and to test our conclusions. Successful completion of the proposed aims will establish new methods and tools to better translate preclinical studies to clinical dosage regimen design for future anti-TB combinations. While motivated by the needs of TB drug development, this project includes innovations that apply to the treatment of other diseases such as cancer, human immunodeficiency virus (HIV) infection, and malaria.