TB Modelling Group, TB Centre and Centre for the Mathematical Modelling of Infectious Diseases, Faculty of Epidemiology and Population Health, London School of Hygiene & Tropical Medicine, Keppel Street, London WC1E 7HT, UK.
Chinese Center for Disease Control and Prevention, Beijing 102206, China.
Sci Transl Med. 2020 Oct 7;12(564). doi: 10.1126/scitranslmed.aax4607.
More effective tuberculosis vaccines are needed to help reach World Health Organization tuberculosis elimination goals. Insufficient evidence exists on the potential impact of future tuberculosis vaccines with varying characteristics and in different epidemiological settings. To inform vaccine development decision making, we modeled the impact of hypothetical tuberculosis vaccines in three high-burden countries. We calibrated () transmission models to age-stratified demographic and epidemiological data from China, South Africa, and India. We varied vaccine efficacy to prevent infection or disease, effective in persons uninfected or infected, and duration of protection. We modeled routine early-adolescent vaccination and 10-yearly mass campaigns from 2025. We estimated median percentage population-level tuberculosis incidence rate reduction (IRR) in 2050 compared to a no new vaccine scenario. In all settings, results suggested vaccines preventing disease in -infected populations would have greatest impact by 2050 (10-year, 70% efficacy against disease, IRR 51%, 52%, and 54% in China, South Africa, and India, respectively). Vaccines preventing reinfection delivered lower potential impact (IRR 1, 12, and 17%). Intermediate impact was predicted for vaccines effective only in uninfected populations, if preventing infection (IRR 21, 37, and 50%) or disease (IRR 19, 36, and 51%), with greater impact in higher-transmission settings. Tuberculosis vaccines have the potential to deliver substantial population-level impact. For prioritizing impact by 2050, vaccine development should focus on preventing disease in -infected populations. Preventing infection or disease in uninfected populations may be useful in higher transmission settings. As vaccine impact depended on epidemiology, different development strategies may be required.
需要更有效的结核病疫苗来帮助实现世界卫生组织的结核病消除目标。对于具有不同特征和在不同流行病学环境下的未来结核病疫苗的潜在影响，现有证据不足。为了为疫苗开发决策提供信息，我们在三个高负担国家中模拟了假设的结核病疫苗的影响。我们根据中国、南非和印度的年龄分层人口统计学和流行病学数据对 () 传播模型进行了校准。我们改变了预防感染或疾病的疫苗效力、对未感染者或感染者有效的疫苗效力以及保护持续时间。我们模拟了从 2025 年开始的常规青少年早期接种和 10 年一次的大规模接种活动。我们估计了与没有新疫苗的情况相比，2050 年人群结核病发病率降低的中位数百分比（IRR）。在所有环境中，结果表明，到 2050 年，预防感染人群疾病的疫苗将产生最大影响（10 年，疾病预防效力为 70%，IRR 分别为 51%、52%和 54%，中国、南非和印度）。预防再感染的疫苗的潜在影响较低（IRR 分别为 1、12 和 17%）。如果预防感染（IRR 分别为 21%、37%和 50%）或疾病（IRR 分别为 19%、36%和 51%），仅对未感染者有效的疫苗的预测影响则处于中等水平，在高传播环境中，影响更大。结核病疫苗具有实现人群水平重大影响的潜力。为了在 2050 年前优先考虑影响，疫苗开发应侧重于预防感染人群的疾病。在未感染者中预防感染或疾病可能在高传播环境中有用。由于疫苗的影响取决于流行病学，因此可能需要不同的发展策略。
