Abstract：
BACKGROUND: Clinical trials in older adults are increasingly focused on functional outcomes, and the composite outcome of dementia, disability, and death is gaining pivotal importance. Genetic variation, particularly the APOE epsilon(ε) variants, may modify responses to new treatments. Although APOE ε4 is known to influence these outcomes separately, the magnitude of its effect on this composite outcome remains unknown. We tested the hypothesis that APOE ε4 increases, whereas APOE ε2 decreases, the risk of a composite outcome of dementia, disability, and death.
METHODS: We evaluated clinical and genomic data from the Health and Retirement Study collected from 1992 to 2020. We used variants rs429358 and rs7412 to determine APOE genotypes, modeled dominantly (carriers/noncarriers). We conducted survival analysis, using multivariable Cox proportional hazards models with a composite endpoint of dementia, disability, and death. Our primary analysis evaluated participants with genetic data and no previous dementia or disability. In secondary analyses, we focused on persons aged > = 75 years without heart disease or stroke, a subpopulation increasingly important in clinical trials of older adults.
RESULTS: We included 14,527 participants in the primary analysis. Over a median of 18 (Interquartile Range [IQR] 12-24) years, 6711 (46%) participants developed the composite outcome. In Cox analyses, APOE ε4 associated with higher risk (HR:1.15, 95%CI:1.09-1.22) of the composite outcome, whereas APOE ε2 associated with lower risk (HR:0.92, 95%CI:0.86-0.99). In the secondary analysis, we included 3174 participants. Over a median of 7 (IQR 4-11) years, 1326 participants (42%) developed the composite outcome. In Cox analyses, APOE ε4 associated with higher risk (HR:1.25, 95%CI:1.10-1.41) of the composite outcome, whereas APOE ε2 associated with lower risk (HR:0.84, 95%CI:0.71-0.98).
CONCLUSIONS: APOE ε variants are linked to the risk of dementia, disability, and death in older adults. By examining these variants in clinical trials, we can better elucidate how they might alter the effectiveness of tested interventions. Importantly, this genetic information could help identify participants who may have greater absolute benefit from such interventions.
METHODS: We evaluated clinical and genomic data from the Health and Retirement Study collected from 1992 to 2020. We used variants rs429358 and rs7412 to determine APOE genotypes, modeled dominantly (carriers/noncarriers). We conducted survival analysis, using multivariable Cox proportional hazards models with a composite endpoint of dementia, disability, and death. Our primary analysis evaluated participants with genetic data and no previous dementia or disability. In secondary analyses, we focused on persons aged > = 75 years without heart disease or stroke, a subpopulation increasingly important in clinical trials of older adults.
RESULTS: We included 14,527 participants in the primary analysis. Over a median of 18 (Interquartile Range [IQR] 12-24) years, 6711 (46%) participants developed the composite outcome. In Cox analyses, APOE ε4 associated with higher risk (HR:1.15, 95%CI:1.09-1.22) of the composite outcome, whereas APOE ε2 associated with lower risk (HR:0.92, 95%CI:0.86-0.99). In the secondary analysis, we included 3174 participants. Over a median of 7 (IQR 4-11) years, 1326 participants (42%) developed the composite outcome. In Cox analyses, APOE ε4 associated with higher risk (HR:1.25, 95%CI:1.10-1.41) of the composite outcome, whereas APOE ε2 associated with lower risk (HR:0.84, 95%CI:0.71-0.98).
CONCLUSIONS: APOE ε variants are linked to the risk of dementia, disability, and death in older adults. By examining these variants in clinical trials, we can better elucidate how they might alter the effectiveness of tested interventions. Importantly, this genetic information could help identify participants who may have greater absolute benefit from such interventions.
摘要：
背景：老年人的临床试验越来越关注功能结局，和痴呆的复合结果，残疾,死亡越来越重要。遗传变异,特别是APOEε(ε)变体，可能会改变对新疗法的反应。尽管已知APOEε4会分别影响这些结果，其对这一复合结局的影响程度尚不清楚.我们检验了APOEε4增加的假设，而APOEε2减少，痴呆症复合结局的风险，残疾,和死亡。
方法：我们评估了从1992年到2020年收集的健康与退休研究的临床和基因组数据。我们使用变异rs429358和rs7412来确定APOE基因型，主要建模(运营商/非运营商)。我们进行了生存分析，使用具有痴呆复合终点的多变量Cox比例风险模型，残疾,和死亡。我们的主要分析评估了具有遗传数据且以前没有痴呆或残疾的参与者。在二级分析中，我们关注年龄>=75岁无心脏病或中风的人，在老年人的临床试验中越来越重要的亚群。
结果：我们在主要分析中纳入了14,527名参与者。超过18年（四分位数范围[IQR]12-24年）的中位数，6711名(46%)参与者形成了复合结果。在考克斯分析中，APOEε4与复合结局的较高风险（HR：1.15，95CI：1.09-1.22）相关，而APOEε2与较低的风险相关(HR：0.92，95CI：0.86-0.99)。在次要分析中，我们包括3174名参与者。超过7年（IQR4-11年）的中位数，1326名参与者(42%)形成了复合结果。在考克斯分析中，APOEε4与复合结局的较高风险（HR：1.25，95CI：1.10-1.41）相关，而APOEε2与较低的风险相关(HR：0.84，95CI：0.71-0.98)。
结论：APOEε变异与痴呆的风险有关，残疾,和老年人的死亡。通过在临床试验中检查这些变异，我们可以更好地阐明它们可能如何改变经过测试的干预措施的有效性.重要的是,这些遗传信息可以帮助识别可能从此类干预措施中获得更大绝对益处的参与者.
方法：我们评估了从1992年到2020年收集的健康与退休研究的临床和基因组数据。我们使用变异rs429358和rs7412来确定APOE基因型，主要建模(运营商/非运营商)。我们进行了生存分析，使用具有痴呆复合终点的多变量Cox比例风险模型，残疾,和死亡。我们的主要分析评估了具有遗传数据且以前没有痴呆或残疾的参与者。在二级分析中，我们关注年龄>=75岁无心脏病或中风的人，在老年人的临床试验中越来越重要的亚群。
结果：我们在主要分析中纳入了14,527名参与者。超过18年（四分位数范围[IQR]12-24年）的中位数，6711名(46%)参与者形成了复合结果。在考克斯分析中，APOEε4与复合结局的较高风险（HR：1.15，95CI：1.09-1.22）相关，而APOEε2与较低的风险相关(HR：0.92，95CI：0.86-0.99)。在次要分析中，我们包括3174名参与者。超过7年（IQR4-11年）的中位数，1326名参与者(42%)形成了复合结果。在考克斯分析中，APOEε4与复合结局的较高风险（HR：1.25，95CI：1.10-1.41）相关，而APOEε2与较低的风险相关(HR：0.84，95CI：0.71-0.98)。
结论：APOEε变异与痴呆的风险有关，残疾,和老年人的死亡。通过在临床试验中检查这些变异，我们可以更好地阐明它们可能如何改变经过测试的干预措施的有效性.重要的是,这些遗传信息可以帮助识别可能从此类干预措施中获得更大绝对益处的参与者.
