单剂次与两剂次HPV疫苗接种的成本效果：基于肯尼亚-印度中低收入国家复合模型的马尔可夫队列建模分析

摘要(中文译文)

宫颈癌在低收入和中等收入国家（LMICs）造成了不成比例的疾病负担，其中第二剂次脱种显著限制了两剂次HPV疫苗接种规划在真实世界中的有效性。单剂次接种方案可消除脱种问题，但提供的保护持续时间较短。单剂次接种在经济上是否优于两剂次接种，以及在何种条件下两种策略会因所采用的健康结局指标不同而产生不同的推荐意见，在已发表的成本效果分析中受到的关注有限，尤其是针对直接头对头方案比较的研究。我们构建了一个静态Markov队列模型，并对照观察到的发病率数据进行了表面效度评估，在由肯尼亚（低收入）和印度（中低收入）构建的LMIC复合模型中，追踪100,000名9岁女孩的终身时间跨度（至99岁）。比较了三种策略：不接种疫苗（参照）、1剂次HPV疫苗接种（对HPV 16/18相关CIN2+的效力为87%；基础病例保护持续时间为20年）和2剂次疫苗接种（效力93%；终身保护）。两种策略的规划覆盖率均为49%；15%的第二剂次脱种率使2剂次有效覆盖率降至41.7%。从医疗保健支付方角度估算成本（2024年美元）、避免的伤残调整生命年（DALYs）和增量成本效果比（ICERs），采用3%年贴现率，在静态直接效应框架内进行，该框架不捕捉群体免疫。采用《疾病控制优先事项》第三版中的支付意愿阈值：200美元/DALY（低收入）和500美元/DALY（中低收入）。通过10,000次模拟的概率敏感性分析（PSA）和预先设定的情景分析来表征不确定性，考察第二剂次脱种率从5%至45%以及1剂次保护持续时间为10年、20年（基础病例）和终身的情况。与不接种疫苗相比，两种策略均占优势（节省成本且更有效）。每100,000名女孩中，1剂次预防了64例浸润性宫颈癌（ICC）病例和77例死亡，避免了5,545个DALYs，净成本节省1,233,242美元（ICER：-222美元/DALY）。2剂次预防了更多病例（87例）和死亡（95例），但在基础病例20年保护假设下，避免的DALYs更少（5,149个），节省的成本也更少（980,690美元）。这种差异反映了1剂次保护在15-29岁女性中的年龄集中性（每避免一例病例为86.4个DALYs，而2剂次为59.2个），并且在结构上依赖于20年保护持续时间的假设；在终身保护下，这种DALY效率差异会缩小。在PSA中，1剂次在99.3%的模拟中占优势（平均ICER：-195美元/DALY；95%可信区间[CrI]：-315至-64美元），并在100%的模拟中在两个阈值下均具有成本效果。2剂次在91.8%的模拟中占优势（平均ICER：-133美元/DALY；95% CrI：-284至+136美元）。1剂次ICER对第二剂次脱种率不变；2剂次ICER从5%脱种率时的-141美元/DALY恶化至45%脱种率时的-105美元/DALY。单剂次策略实现更高每美元DALY效率的发现，在结构上依赖于20年单剂次保护持续时间的基础病例假设，不应被解释为普遍的经验性发现。在这一特定假设下，消除第二剂次脱种使有效覆盖率从41.7%提高到49.0%，并将保护集中在15-29岁女性中，每避免一例病例贡献更多的寿命损失年。敏感性分析表明，如果单剂次保护时间更短（10年）或延长至终身，这种DALY效率优势会缩小或逆转。在所有考察的情景中，两剂次策略预防的绝对癌症病例数和死亡数均多于单剂次。这些是在肯尼亚-印度复合模型、静态直接效应框架内生成的情景特异性、探索性发现；国家规划决策应纳入国家特异性脱种率、关于单剂次保护持续时间的新兴长期证据、当地适用的结局指标以及经国家校准的建模，然后再得出政策结论。

Abstract

Cervical cancer imposes a disproportionate burden in low- and middle-income countries (LMICs), where second-dose dropout substantially limits the real-world effectiveness of two-dose HPV vaccination programmes. Single-dose schedules eliminate dropout but provide shorter-duration protection. Whether single-dose vaccination is economically preferred to two-dose vaccination, and under what conditions the two strategies yield divergent recommendations depending on the health outcome metric used, has received limited attention in published cost-effectiveness analyses, particularly for the direct head-to-head schedule comparison. We developed a static Markov cohort model, assessed for face validity against observed incidence data, tracking 100,000 girls aged 9 years through a lifetime horizon (to age 99) within an LMIC composite derived from Kenya (low-income) and India (lower-middle-income). Three strategies were compared: no vaccination (reference), 1-dose HPV vaccination (efficacy 87% against HPV 16/18-attributable CIN2+; 20-year base-case protection duration), and 2-dose vaccination (efficacy 93%; lifetime protection). Programme coverage was 49% for both strategies; a 15% second-dose dropout rate reduced 2-dose effective coverage to 41.7%. Costs (2024 USD), disability-adjusted life years (DALYs) averted, and incremental cost-effectiveness ratios (ICERs) were estimated from a healthcare payer perspective with 3% annual discounting, within a static direct-effect framework that does not capture herd immunity. Willingness-to-pay thresholds of USD 200/DALY (low-income) and USD 500/DALY (lower-middle-income) were applied from Disease Control Priorities, 3rd edition. Uncertainty was characterised through 10,000-simulation probabilistic sensitivity analysis (PSA) and pre-specified scenario analyses examining second-dose dropout rates from 5 to 45% and 1-dose protection durations of 10 years, 20 years (base-case), and lifetime. Both strategies were dominant (cost-saving and more effective) versus no vaccination. Per 100,000 girls, 1-dose prevented 64 invasive cervical cancer (ICC) cases and 77 deaths, averting 5,545 DALYs with a net cost saving of USD 1,233,242 (ICER: -USD 222/DALY). Two-dose prevented more cases (87) and deaths (95) but, under the base-case 20-year protection assumption, averted fewer DALYs (5,149) and saved less (USD 980,690). This divergence reflects the age-concentration of 1-dose protection in women aged 15-29 years (86.4 DALYs per averted case vs. 59.2 for 2-dose) and is structurally dependent on the 20-year protection duration assumption; at lifetime protection, this DALY efficiency difference diminishes. In PSA, 1-dose was dominant in 99.3% of simulations (mean ICER: -USD 195/DALY; 95% credible interval [CrI]: -USD 315 to -USD 64) and cost-effective at both thresholds in 100% of simulations. Two-dose was dominant in 91.8% of simulations (mean ICER: -USD 133/DALY; 95% CrI: -USD 284 to +USD 136). The 1-dose ICER was invariant to second-dose dropout rate; the 2-dose ICER deteriorated from -USD 141/DALY at 5% dropout to -USD 105/DALY at 45% dropout. The finding that the single-dose strategy achieves greater DALY efficiency per dollar is structurally dependent on the base-case assumption of 20-year single-dose protection duration and should not be interpreted as a general empirical finding. Under this specific assumption, elimination of second-dose dropout raised effective coverage from 41.7 to 49.0% and concentrated protection in women aged 15-29 where each averted case contributes more years of life lost. Sensitivity analyses show this DALY efficiency advantage narrows or reverses if single-dose protection is shorter (10 years) or extends to lifetime. Across all scenarios examined, the two-dose strategy prevented more absolute cancer cases and deaths than single-dose. These are scenario-specific, exploratory findings generated within a Kenya-India composite under a static direct-effect framework; national programme decisions should incorporate country-specific dropout rates, emerging long-term evidence on single-dose protection duration, locally applicable outcome metrics, and country-calibrated modelling before drawing policy conclusions.

如何引用

Frontiers in public health 的其他文献

• 医疗服务提供者与患者对中医服务价格改革的认知：来自中国东北地区的证据
• 血清持久性有机污染物与多囊卵巢综合征风险之间的关联：一项具有中介分析的病例对照研究
• 平台算法压力与中国零工员工的职业承诺：一项双刃剑效应研究
• 从接入到能力：当代中国的数字素养与健康不平等
• 广西地中海贫血现状及展望
• 老年人艾滋病毒/艾滋病负担的长期演变：七国集团国家75岁以上个人的全球疾病负担分析
• 中国农村代际支持与老年人心理健康：证据和政策启示
• 当远程压力从不睡觉时：心理/空间分离作为数字游牧民睡眠质量的中介
• 推进非洲卫生领域的人工智能和数据科学：全球卫生优先事项的教育、合作和应用
• 传染病医院手术职业暴露的流行病学：一项为期5年的回顾性研究