Leveraging Pathogen Sequence and Contact Tracing Data to Enhance Vaccine Trials in Emerging Epidemics.
Leveraging Pathogen Sequence and Contact Tracing Data to Enhance Vaccine Trials in Emerging Epidemics.
复制标题
利用病原体序列和接触者追踪数据来加强在新出现的流行病中的疫苗试验。
DOI：
10.1097/ede.0000000000001367
复制
发表时间：
2021-09-01
期刊：
影响因子：
--
通讯作者：
Lipsitch M
中科院分区：
文献类型：
--
作者：
Kahn R;Wang R;Leavitt SV;Hanage WP;Lipsitch M
在疫情爆发期间进行疫苗试验的提前规划提高了我们快速确定疫苗效力和潜在影响的能力。疫苗抗传染性效力（VEI）是了解疫苗全面影响的重要指标，但目前在许多试验设计中无法识别，因为它需要了解感染者的疫苗接种状况。基因组学的最新进展提高了我们重建传播网络的能力。我们的目标是评估是否增加试验与病原体序列和接触者追踪数据可以允许他们估计VEI。我们开发了一个传播模型，在疫情环境中进行疫苗试验，结合病原体序列数据和接触者追踪数据，并为可能的感染者分配概率。然后，我们提出并评估VEI估计的性能。我们发现，在完美的知识感染者-被感染者对，我们能够准确地估计VEI。序列数据的使用导致传输网络的不完美重建，使VEI的估计偏向于零，使用深度序列数据的方法比使用共有序列数据的方法表现更好。纳入接触者追踪数据可减少偏差。病原体基因组学增强了VEI的可识别性，但不完善的传播网络重建使估计值偏向于零，并限制了我们检测VEI的能力。鉴于偏倚方向一致，使用这些方法从试验中获得的估计值将提供真实VEI的下限。序列和流行病学数据的结合产生了最准确的估计，强调了接触者追踪的重要性。
Advance planning of vaccine trials conducted during outbreaks increases our ability to rapidly define the efficacy and potential impact of a vaccine. Vaccine efficacy against infectiousness (VEI) is an important measure for understanding a vaccine’s full impact, yet it is currently not identifiable in many trial designs because it requires knowledge of infectors’ vaccination status. Recent advances in genomics have improved our ability to reconstruct transmission networks. We aim to assess if augmenting trials with pathogen sequence and contact tracing data can permit them to estimate VEI. We develop a transmission model with a vaccine trial in an outbreak setting, incorporate pathogen sequence data and contact tracing data, and assign probabilities to likely infectors. We then propose and evaluate the performance of an estimator of VEI. We find that under perfect knowledge of infector–infectee pairs, we are able to accurately estimate VEI. Use of sequence data results in imperfect reconstruction of transmission networks, biasing estimates of VEI towards the null, with approaches using deep sequence data performing better than approaches using consensus sequence data. Inclusion of contact tracing data reduces the bias. Pathogen genomics enhance identifiability of VEI, but imperfect transmission network reconstruction biases estimates towards the null and limits our ability to detect VEI. Given the consistent direction of the bias, estimates obtained from trials using these methods will provide lower bounds on the true VEI. A combination of sequence and epidemiologic data results in the most accurate estimates, underscoring the importance of contact tracing.