Statistical Methods in HIV Vaccine Efficacy Trials

艾滋病毒/艾滋病研究的最终目标是开发一种预防艾滋病毒感染的疫苗，这种疫苗 在美国和全球消除艾滋病毒/艾滋病将是必要的。严格的随机临床试验 评估候选艾滋病毒疫苗的效力是实现这一目标的核心研究方法。这 该项目开发了艾滋病毒疫苗疗效试验和平行重复低剂量试验的统计方法 挑战SIV疫苗在非人类灵长类动物中的试验。用于药效试验。目标1和目标2开发新的和 交叉统计方法：(1)筛分分析，即对疫苗效力的评估 艾滋病毒感染因传播艾滋病毒的遗传特征而异；及(2)免疫保护相关因素 分析，即评估疫苗对艾滋病毒感染和针对特定基因型的效力 HIV感染与接种疫苗的免疫反应不同。这两个目标的首要目标是 发展免疫保护相关性(COP)，这是指导最佳选择的免疫保护 疫苗中要包括的HIV序列，用于指导基准免疫原性终点的选择 精制疫苗方案在L/11期试验中，并用于免疫架桥疗效试验结果的预测 疫苗在新环境下的效力。这些方法将侧重于(A)高效和灵活地适应时间-- 疫苗效力的差异；(B)最大限度地提高艾滋病毒基因特征的免疫学相关性；(C) 改进用于测量免疫反应的抽样设计；以及(D)制定最佳阈值或 对结合了来自被评估的大量免疫反应的信息的免疫COP进行评分。 许多新方法将通过目标最大值来提高有效性、效率和健壮性 似然统计框架。这些方法将被开发出来，并应用于13项疫苗效力试验 (艾滋病毒8例，登革热3例，带状疱疹1例，流感1例)。 目标3通过开发新的统计方法在非人类内部进行评估，从而与目标1和目标2类似 灵长类重复低剂量挑战试验预防SIV感染的疫苗效力如何随时间变化 具有暴露SIV的遗传特征和对疫苗的免疫应答。目标3研究计划 重点是问题(A)、(B)和(D)“，并将应用于三个试验中的几个试验 研究小组。 相关性(请参阅说明)： 为了在美国和全球消除艾滋病毒/艾滋病，需要一种预防艾滋病毒感染的疫苗。是这样的 疫苗可以通过一系列确定候选疫苗的疫苗效力试验来开发。 提供对艾滋病毒感染的部分保护，并确定保护如何随 艾滋病毒的遗传学和疫苗接种的免疫反应。这个项目开发了新的统计方法 为了改进和加速这一进程；这些方法也适用于疫苗的开发 其他传染病。

An ultimate goal of HIV/AIDS research is the development of a vaccine that prevents HIV infection, which will be needed for eliminating HIV/AIDS in the U.S. and globally. Randomized clinical trials that rigorously assess the efficacy of candidate HIV vaccines are a core research approach to meeting this goal. This project develops statistical methods for HIV vaccine efficacy trials and for parallel repeated low-dose challenge trials of SIV vaccines in nonhuman primates. For efficacy trials. Aims 1 and 2 develop novel and interdigitating statistical methods for: (1) sieve analysis, i.e., the assessment of how vaccine efficacy against HIV infection varies with genetic features of transmitting HIVs; and (2) immune correlates of protection analysis, i.e., the assessment of how vaccine efficacy against HIV infection and against genotype-specific HIV infection varies with immune responses to vaccination. The overarching objective of these two aims is development of immune correlates of protection (CoPs), which are needed for guiding the optimal choice of HIV sequences to include in vaccines, for guiding the choice of immunogenicity endpoints for benchmarking refined vaccine regimens in Phase l/ll trials, and for immuno-bridging of efficacy trial results to predict vaccine efficacy in new settings. The methods will focus on (A) efficiently and flexibly accommodating time- variations in vaccine efficacy; (B) maximizing the immunological relevance ofthe HIV genetic features; (C) improving the sampling design for measuring immune responses; and (D) developing the best threshold or score immune CoPs that combine information from the large set of immune responses that are assessed. Many ofthe new methods will improve validity, efficiency, and robustness via the targeted maximum likelihood statistical framework. The methods will be developed with application to 13 vaccine efficacy trials (8 for HIV, 3 for dengue, 1 for herpes zoster, 1 for influenza). Aim 3 parallels Aims 1 and 2 by developing novel statistical methods for assessing within nonhuman primate repeated low-dose challenge trials how vaccine efficacy to prevent SIV infection over time varies with genetic features of exposing SIVs and with immune responses to vaccination. The Aim 3 research plan focuses on issues (A), (B), and (D)"and will be developed with application to several trials from three research groups. RELEVANCE (See instructions): A vaccine that prevents HIV infection is needed for eliminating HIV/AIDS in the U.S. and globally. Such avaccine may be developed through a series of vaccine efficacy trials that identify candidate vaccines conferring partial protection against HIV infection, and that identify how the protection varies with the genetics of HIV and with immune responses to vaccination. This project develops novel statistical methods for improving and accelerating this process; these methods also apply to the development of vaccines for other infectious diseases.