TALOS-AMI试验中年龄与心肌梗死后抗血小板降阶治疗。
全文（开放获取） · 中文译文
摘要 背景：急性心肌梗死后双联抗血小板治疗降阶的临床效果可能因年龄而异。我们评估了经皮冠状动脉介入治疗（PCI）后病情稳定的患者中，从替格瑞洛降阶为氯吡格雷的有效性和安全性是否因年龄不同而存在差异。 患者与方法：这是一项对TALOS-AMI随机试验的预设二次分析。PCI术后1个月无事件发生且接受阿司匹林-替格瑞洛治疗的患者被随机分配至降阶组（阿司匹林-氯吡格雷）或继续治疗组（阿司匹林-替格瑞洛）。主要净临床终点为1年时主要不良心血管事件（心血管死亡、心肌梗死、卒中）与出血学术研究联合会（BARC）2型、3型或5型出血的复合终点。 结果：我们纳入2697名参与者（平均年龄60.0 ± 11.4岁；16.8%为女性）。在年龄<75岁的患者（n = 2376）中，降阶治疗降低了主要净临床终点（4.1%对比7.2%；校正风险比（aHR），0.54 [95% CI，0.38-0.77]）和出血（2.8%对比4.9%；aHR，0.54 [0.35-0.82]）。在年龄≥75岁的患者中，主要终点（6.4%对比11.6%；aHR 0.54 [0.25-1.17]）或出血（3.2%对比7.9%；aHR，0.41 [0.15-1.15]）未观察到显著差异。交互作用检验显示，年龄对主要终点（交互作用p = 0.978）、主要不良心血管事件（MACE）（p = 0.585）或BARC出血（p = 0.597）均无治疗效应修饰作用。在年龄≥75岁的患者中，14/18例出血事件发生在180天内。 结论：在PCI术后1个月病情稳定且无事件的患者中，未观察到显著的年龄与治疗之间的交互作用；因此，在老年人（≥75岁）中降阶治疗的有效性和安全性仍不确定。 无标注：试验注册：ClinicalTrials.gov（NCT02018055）。 在TALOS-AMI试验这项预设的次要分析中，在年龄<75岁的患者中，基于氯吡格雷的降阶治疗与主要净临床终点及BARC 2、3或5型出血风险降低相关，而主要不良心血管事件（MACE）在两种治疗策略之间无显著差异。在年龄≥75岁的患者中，主要净临床终点、MACE或BARC 2、3或5型出血均未观察到统计学显著的组间差异。使用预设的75岁截点、alternative 65岁和70岁截点，或年龄作为连续变量，均未检测到统计学显著的年龄-治疗交互作用；降阶治疗在老年人中的有效性和安全性仍不确定。 引言 双联抗血小板治疗（DAPT），即阿司匹林联合强效P2Y12抑制剂持续至少12个月，是急性心肌梗死（AMI）后的标准治疗方案，可降低缺血风险[]。强效P2Y12抑制的缺血获益在早期高危阶段最为显著，而在长期维持治疗期间，出血则成为主要关注问题[]。缺血与出血风险随时间的这种变化，为降阶治疗策略提供了理论依据，即仅在初始阶段使用强化DAPT，之后转为强度较低的治疗方案。这一避免出血的理念体现在当代急性冠脉综合征指南中。2025年美国心脏病学会/美国心脏协会指南指出，经皮冠状动脉介入治疗（PCI）术后1个月，对于高出血风险患者，将替格瑞洛或普拉格雷降阶为氯吡格雷并转为单药抗血小板治疗可能是合理的（均为2b类推荐，证据级别B-R）[]，而2023年欧洲心脏病学会指南指出，降阶为氯吡格雷可以考虑（2b类推荐，证据级别A），且对于高出血风险患者，1个月后可使用阿司匹林或P2Y12抑制剂单药治疗（2b类推荐，证据级别B）[]。然而，在老年人中，出血与缺血风险可能同时增加，当前标准治疗在这一人群中是否能提供相当的临床获益仍不明确。 在急性心肌梗死稳定期患者中比较替格瑞洛与氯吡格雷的研究（TALOS-AMI）中，对AMI后病情稳定并接受PCI的患者，1个月后从阿司匹林–替格瑞洛转换为阿司匹林–氯吡格雷，与不转换相比，临床结局更优，主要表现为缺血事件（心血管死亡、AMI或卒中）或临床相关出血的复合终点减少[]。结局改善主要由出血事件减少驱动，而缺血事件未见显著增加。在本项按年龄分层的TALOS-AMI研究二次分析中，患者根据预设的75岁年龄截断值进行分层，以评估DAPT降阶梯治疗的临床效果是否因年龄而异。我们假设降阶梯治疗通过减少出血而不增加缺血事件，在各年龄组中均可带来净临床获益，尽管获益的幅度和时间模式在老年患者中可能有所不同。 患者与方法 研究设计 TALOS-AMI试验的设计、方法学和主要结果细节已有报道[，]。简言之，该试验是一项在韩国32家医院开展的多中心、随机、开放标签、非劣效性研究，纳入接受经皮冠状动脉介入治疗（PCI）的急性心肌梗死（AMI）患者。所有参与者在PCI后第一个月内接受阿司匹林联合替格瑞洛治疗。1个月时，临床稳定且无复发性缺血或出血的患者被随机分配至降阶梯策略组（阿司匹林联合氯吡格雷）或继续阿司匹林联合替格瑞洛组。患者通过计划的门诊访视和电话评估接受12个月随访。数据于2014年2月14日至2018年12月31日期间获取，随访于2021年1月21日结束。研究方案已在ClinicalTrials.gov注册（NCT02018055）。研究方案已获得所有参与中心机构审查委员会的批准，包括韩国天主教大学首尔圣玛丽医院机构审查委员会（批准号KC13MIMV0691），且研究遵循《赫尔辛基宣言》进行。所有参与者均提供了书面知情同意。 在这项按年龄分层的次要分析中，我们评估了DAPT降阶治疗的临床影响是否随年龄而变化。原始试验预先规定以75岁作为界定老年组的阈值，并在额外分析中将年龄作为连续变量进行建模。采用样条曲线展示降阶治疗组与对照组在整个年龄范围内的事件发生率和风险比（HRs）。 结局 与原始TALOS-AMI研究一致，本分析的主要净临床终点为指标PCI后12个月时主要不良心血管事件（MACEs）与出血学术研究联合会（BARC）2、3或5型出血的复合终点。MACE定义为心血管死亡、心肌梗死和卒中的复合终点。关键次要结局为MACEs；BARC 2、3或5型出血的复合终点；各类型BARC出血；以及MACE与BARC 3或5型出血的复合终点。其他次要结局包括主要净临床终点的各组成部分以及全因死亡、缺血驱动的血运重建和支架内血栓形成。 统计分析 连续变量以均值±标准差表示，分类变量以计数和百分比表示。连续变量的组间比较根据数据分布情况采用独立样本t检验或Wilcoxon秩和检验。分布假设通过直方图和分位数-分位数图的目视检查进行评估。分类变量采用卡方检验进行比较。事件发生时间数据采用Cox比例风险模型进行评估，以评估治疗策略与临床结局（包括主要不良心血管事件（MACEs）和出血事件）之间的关系，并调整年龄和性别。在Cox模型中检验交互项，以评估治疗效果是否随年龄变化。我们进行了事后敏感性分析，使用65岁和70岁作为替代年龄截点。对于每个截点，在较年轻和较年长亚组内估计治疗效果，并在Cox比例风险模型中检验治疗与年龄组间的交互项。这些替代截点分析被视为探索性分析。为进一步评估年龄作为连续变量的作用，使用Cox回归模型检验了整个年龄范围内的治疗效果。为探索事件的时间模式，我们在预设时间点总结了事件发生的累积分布。此外，我们进行了180天界标分析，以评估治疗效果在早期（≤180天）和晚期（>180天）之间是否存在差异。 在晚期分析中，仅纳入在地标时间点仍无事件的患者，并从该时间点重新开始随访。Cox比例风险模型分别在各时期内应用，并使用交互项评估治疗效果随时间的差异。由于原始TALOS-AMI试验的把握度是针对总体随机队列而非年龄特异性亚组比较设计的，因此针对年龄≥75岁患者的主要净临床终点和BARC 2、3或5型出血进行了事后把握度分析，使用双侧α为0.05和观察到的亚组事件率。分析使用IBM SPSS Statistics（版本30；IBM Corp.，Armonk，NY，USA）和R软件（版本4.2.2；R Foundation for Statistical Computing，Vienna，Austria）进行。所有统计检验均采用双侧显著性阈值0.05。 结果 研究人群 研究样本的平均（标准差）年龄为60.0（11.4）岁；其中452例（16.8%）为女性。在2697名参与者中，1349人被分配至阿司匹林加氯吡格雷（降阶梯治疗）组，1348人被分配至继续阿司匹林加替格瑞洛（活性对照）组。所有参与者均纳入此项按年龄分层的分析（图S1）。 在较年轻队列和较年长队列中，两组之间的大多数基线人口学特征和操作特征均达到平衡（和）。然而，较年轻队列中两组之间的估算肾小球滤过率和梗死相关动脉存在差异，较年长队列中两组之间的经皮冠状动脉介入治疗史存在差异。 按年龄组划分的患者基线特征。 数据以n（%）或均值±标准差表示。关于吸烟，较年轻活性对照组中有1例患者数据缺失。关于eGFR（<60 mL/min），较年轻降阶梯治疗组有25例、较年轻活性对照组有21例、较年长降阶梯治疗组有3例、较年长活性对照组有2例患者数据缺失。关于LVEF <40%，较年轻降阶梯治疗组有21例、较年轻活性对照组有33例、较年长降阶梯治疗组有3例、较年长活性对照组有11例患者数据缺失。 CABG：冠状动脉旁路移植术；CVA：脑血管意外；CVD：心血管疾病；eGFR：估算肾小球滤过率；LVEF：左心室射血分数；NSTEMI：非ST段抬高型心肌梗死；PCI：经皮冠状动脉介入治疗；STEMI：ST段抬高型心肌梗死。 降阶梯治疗组接受阿司匹林加氯吡格雷治疗。 活性对照组接受阿司匹林加替格瑞洛治疗。 肾功能受损定义为就诊时eGFR <60 mL/min/1.73 m²体表面积。 按年龄分组的操作特征。 数据以n（%）或均值±标准差表示。对于穿刺部位，较年轻活性对照组中有两名患者数据缺失。对于梗死相关动脉，较年轻降阶梯治疗组中有一名患者数据缺失，较年轻活性对照组中有两名患者数据缺失。 GP：糖蛋白；IVUS：血管内超声；LAD：左前降支；LCX：左旋支；LM：左主干；No.：数量；OCT：光学相干断层扫描；RCA：右冠状动脉；No. of stents：梗死相关动脉支架数量。 降阶梯治疗组接受阿司匹林加氯吡格雷治疗。 活性对照组接受阿司匹林加替格瑞洛治疗。 预设年龄亚组中的结局。 主要和关键次要结局的结果见 和 ，其他次要结局的结果详见 表 S1。在年龄 <75 岁的患者中，降阶梯治疗组的主要净临床终点发生率低于主动对照组（49 例 [4.1%] 对 85 例 [7.2%]；校正后风险比 [aHR]，0.54；95% 置信区间 [CI]，0.38–0.77；p < 0.001）。BARC 2、3 或 5 型出血（一项关键次要结局）的发生率在降阶梯治疗组中也低于主动对照组（33 例 [2.8%] 对 58 例 [4.9%] 事件；aHR，0.54；95% CI，0.35–0.82；p = 0.004）。相比之下，在较年轻队列中，主要不良心血管事件（MACE）（一项关键次要结局）的发生率在两种策略之间无差异（1.7% 对 2.5%；aHR，0.63；95% CI，0.36–1.12；p = 0.11）。在年龄 <75 岁患者的其他次要结局方面，降阶梯治疗组与主动对照组在全因死亡（7 例 [0.6%] 对 6 例 [0.5%]；aHR，1.10；95% CI，0.37–3.25；p = 0.876）、任何心肌梗死（10 例 [0.8%] 对 17 例 [1.4%]；aHR，0.56；95% CI，0.26–1.22；p = 0.145）、靶血管血运重建（16 例 [1.3%] 对 15 例 [1.3%]；aHR，1.00；95% CI，0.50–2.02；p = 0.998）或支架内血栓形成（3 例 [0.3%] 对 2 例 [0.2%]；aHR，1.43；95% CI，0.24–8.55；p = 0.697）方面均无差异。 根据年龄分层的Kaplan–Meier曲线。（A–C）代表年龄<75岁的患者，（D–F）代表年龄≥75岁的患者。主要净临床终点为PCI后1至12个月内心血管死亡、心肌梗死、卒中和BARC 2型、3型或5型出血的复合终点。MACE定义为心血管死亡、卒中和心肌梗死的复合终点。BARC出血包括根据BARC标准定义的2型、3型或5型事件。BARC：出血学术研究联盟；d，天；MACE：主要不良心血管事件；PCI：经皮冠状动脉介入治疗。 主要和关键次要结局。 校正因素为年龄和性别。 adj：校正后；BARC：出血学术研究联盟；HR：风险比；int：交互作用；MACE：主要不良心血管事件。 在年龄≥75岁的患者中，主要净临床终点（10 [6.4%] vs. 19 [11.6%]；校正后HR，0.54；95% CI，0.25–1.17；p = 0.120）、主要不良心血管事件（7 [4.5%] vs. 8 [4.9%]；校正后HR，0.88；95% CI，0.32–2.43；p = 0.801）或BARC 2型、3型或5型出血（5 [3.2%] vs. 13 [7.9%]；校正后HR，0.41；95% CI，0.15–1.15；p = 0.089）均未观察到统计学显著的组间差异。正式交互作用检验显示，在主要净临床终点（交互作用p = 0.978）、主要不良心血管事件（交互作用p = 0.585）或BARC 2型、3型或5型出血（交互作用p = 0.597）方面，无证据表明治疗效应在年龄<75岁与≥75岁患者之间存在差异。对于年龄≥75岁患者的其他次要结局，全因死亡（4 [2.5%] vs. 4 [2.4%]；校正后HR，1.01；95% CI，0.25–4.05；p = 0.991）、任何心肌梗死（2 [1.3%] vs. 3 [1.8%]；校正后HR，0.65；95% CI，0.11–3.94；p = 0.642）和靶血管血运重建（1 [0.6%] vs. 2 [1.2%]；校正后HR，0.51；95% CI，0.05–5.70；p = 0.588）在两组间均未观察到差异。 事件的时间分布及180天里程碑分析 对于BARC 2、3或5型出血，年龄≥75岁的患者中，14/18例事件（77.8%）发生在分组后180天内，而年龄<75岁的患者中为54/91例事件（59.3%）（表S2）。在180天界标分析中（表S3，及），在总体队列中（aHR，0.40；95% CI，0.24–0.67）和年龄<75岁的患者中（aHR，0.39；95% CI，0.22–0.70），降阶治疗与最初180天内BARC出血风险较低相关。各时间段之间未观察到异质性（交互作用p>0.05）。在年龄≥75岁的患者中，两个时间段内降阶治疗组与主动对照组之间均未观察到差异（≤180天：aHR，0.43；95% CI，0.14–1.39；>180天：aHR，0.32；95% CI，0.03–3.09；交互作用p=0.840）。 年龄<75岁患者180天界标分析的Kaplan–Meier曲线。(A)主要净临床终点，(B)主要不良心血管事件（MACE），以及(C)BARC 2、3或5型出血。对于每种结局，左侧图显示从随机化至第180天的累积事件发生率。右侧图显示在第180天界标时无事件且仍在观察中的参与者中，第180天之后的条件累积事件发生率。事件概率在第180天重置为零，而x轴保留自随机化以来经过的时间。虚线垂直线标记180天界标。显示的p值来自经年龄和性别校正的Cox比例风险模型。主要净临床终点是心血管死亡、心肌梗死、卒中和BARC 2、3或5型出血的复合终点。MACE定义为心血管死亡、心肌梗死和卒中的复合终点。BARC出血包括根据BARC标准定义的2、3或5型出血事件。BARC：出血学术研究联合会；MACE：主要不良心血管事件；PC：经皮冠状动脉介入治疗。 年龄≥75岁患者180天界标分析的Kaplan–Meier曲线。(A) 主要净临床终点，(B) MACE，以及(C) BARC 2、3或5型出血。对于每个结局，左侧图显示从随机化至第180天的累积事件发生率。右侧图显示在界标时无事件且仍在观察中的参与者第180天之后的条件累积事件发生率。事件概率在第180天重置为零，而x轴保留自随机化以来经过的时间。虚线垂直标记表示180天界标。显示的p值来自经年龄和性别校正的Cox比例风险模型。主要净临床终点是心血管死亡、心肌梗死、卒中和BARC 2、3或5型出血的复合终点。MACE定义为心血管死亡、心肌梗死和卒中的复合终点。BARC出血包括根据BARC标准定义的2、3或5型出血事件。BARC：出血学术研究联合会；MACE：主要不良心血管事件；PCI：经皮冠状动脉介入治疗。 使用替代年龄截点的敏感性分析 在事后敏感性分析中，高龄的替代定义并未改变研究结果。在年龄≥65岁的患者中，降阶梯治疗组473例患者中有30例（6.3%）发生主要净临床终点事件，主动对照组475例患者中有44例（9.3%）发生（校正HR，0.66；95% CI，0.41–1.04；p = 0.074）。在年龄≥70岁的患者中，相应的事件发生率分别为7.2%和9.8%（校正HR，0.72；95% CI，0.41–1.24；p = 0.235）。使用65岁截点（交互p = 0.275）或70岁截点（交互p = 0.239）均未检测到统计学显著的治疗-年龄交互作用。同样，使用任一截点，MACE或BARC 2、3或5型出血均未观察到显著交互作用（所有校正交互p ≥ 0.094；表S4）。 以年龄作为连续变量分析年龄依赖性效应 当年龄作为连续变量进行建模时，预测的累积事件发生率通常随年龄增长而增加。然而，对于主要净临床终点、MACE或BARC 2、3或5型出血，均未观察到统计学显著的治疗-年龄交互作用（总体RCS交互p分别为0.571、0.945和0.329）。未发现显著的非线性交互作用。
全文（开放获取） · 英文原文
【Abstract】 BACKGROUND: The clinical effects of dual antiplatelet therapy de-escalation after acute myocardial infarction may differ by age. We evaluated whether the efficacy and safety of de-escalation from ticagrelor to clopidogrel differed by age in stabilized patients after percutaneous coronary intervention (PCI). PATIENTS AND METHODS: This was a prespecified secondary analysis of the TALOS-AMI randomized trial. Patients event-free 1 month after PCI receiving aspirin-ticagrelor were randomly allocated to de-escalation (aspirin-clopidogrel) or continuation (aspirin-ticagrelor). The primary net clinical endpoint was a composite of major adverse cardiovascular events (cardiovascular death, myocardial infarction, stroke) and Bleeding Academic Research Consortium types 2, 3, or 5 bleeding at 1 year. RESULTS: We included 2697 participants (mean age 60.0 ± 11.4 years; 16.8% women). Among patients aged <75 years (n = 2376), de-escalation reduced the primary net clinical endpoint (4.1% vs. 7.2%; adjusted hazard ratio (aHR), 0.54 [95% CI, 0.38-0.77]) and bleeding (2.8% vs. 4.9%; aHR, 0.54 [0.35-0.82]). Among patients aged ≥75 years, no significant differences were observed for the primary endpoint (6.4% vs. 11.6%; aHR0.54 [0.25-1.17]) or bleeding (3.2% vs. 7.9%; aHR, 0.41 [0.15-1.15]). Interaction testing showed no treatment effect modification by age for the primary endpoint (p for interaction = 0.978), MACE (p = 0.585), or BARC bleeding (p = 0.597). Among patients aged ≥75 years, 14/18 bleeding events occurred within 180 days. CONCLUSIONS: Among stabilized, event-free patients 1 month after PCI, no significant age-treatment interaction was observed; therefore, the efficacy and safety of de-escalation in older adults (≥75 years) remain uncertain. UNLABELLED: Trial registration: ClinicalTrials.gov (NCT02018055). In this prespecified secondary analysis of the TALOS-AMI trial, clopidogrel-based de-escalation was associated with lower risks of the primary net clinical endpoint and BARC type 2, 3, or 5 bleeding among patients aged <75 years, while MACE did not differ significantly between treatment strategies. Among patients aged ≥75 years, no statistically significant between-group differences were observed for the primary net clinical endpoint, MACE, or BARC type 2, 3, or 5 bleeding. No statistically significant treatment-by-age interaction was detected using the prespecified 75 years cutoff, alternative 65 and 70 years cutoffs, or age as a continuous variable; the efficacy and safety of de-escalation specifically in older adults remain uncertain. 【Introduction】 Dual antiplatelet therapy (DAPT) with aspirin and a potent P2Y12 inhibitor for at least 12 months is the standard of care after acute myocardial infarction (AMI), reducing the ischemic risk []. The ischemic effect of potent P2Y12 inhibition is most prominent in the early, high-risk period, whereas bleeding becomes the dominant concern during long-term maintenance therapy []. These time-dependent shifts in the ischemic and bleeding risks provide a rationale for de-escalation strategies that limit intensive DAPT to the initial phase and transition to less intensive therapy thereafter. This bleeding-avoidance rationale is reflected in contemporary acute coronary syndrome guidelines. The 2025 American College of Cardiology/American Heart Association guideline states that, 1 month after percutaneous coronary intervention (PCI), de-escalation from ticagrelor or prasugrel to clopidogrel and transition to single antiplatelet therapy for patients at a high bleeding risk may be reasonable (both are Class of Recommendation 2b and Level of Evidence B-R) [], whereas the 2023 European Society of Cardiology guideline states that de-escalation to clopidogrel may be considered (Class 2b, Level A) and that aspirin or P2Y12 inhibitor monotherapy after 1 month may be considered for patients at a high bleeding risk (Class 2b, Level B) []. However, in older adults, bleeding and ischemic risks can increase concurrently, and whether the current standard therapy provides a comparable clinical benefit in this subset of the population remains unclear. In the Ticagrelor vs. Clopidogrel in Stabilized Patients with Acute Myocardial Infarction (TALOS-AMI) study on patients stabilized after AMI and who underwent PCI, switching from aspirin–ticagrelor to aspirin–clopidogrel after 1 month resulted in superior clinical outcomes, compared with not switching, primarily as reductions in a composite of ischemic events (cardiovascular death, AMI, or stroke) or clinically relevant bleeding []. The improvement in outcomes was mainly driven by a reduction in bleeding events, without a significant increase in ischemic events. In this age-stratified secondary analysis of the TALOS-AMI study, patients were stratified according to a prespecified age cutoff of 75 years to assess whether the clinical effects of DAPT de-escalation differed by age. We hypothesized that de-escalation would confer a net clinical benefit across age groups by reducing bleeding without increasing ischemic events, although the magnitude and temporal pattern of benefit might differ in older patients. 【Patients and methods】 【Study design】 Details of the TALOS-AMI trial design, methodology, and main findings have been reported [,]. In brief, it was a multicenter, randomized, open-label, noninferiority study conducted at 32 hospitals in South Korea, enrolling patients with AMI who underwent PCI. All participants received aspirin plus ticagrelor for the first month after PCI. At 1 month, patients who were clinically stabilized without recurrent ischemia or bleeding were randomly assigned to either a de-escalation strategy with aspirin and clopidogrel or continuation of aspirin and ticagrelor. Patients were followed up for 12 months through scheduled outpatient visits and telephonic assessments. Data were obtained between February 14, 2014 and December 31, 2018, and follow-up concluded on January 21, 2021. The study protocol was registered at ClinicalTrials.gov (NCT02018055). The study protocol was approved by the institutional review boards of all participating centers, including the Institutional Review Board of Seoul St. Mary’s Hospital, The Catholic University of Korea (approval No. KC13MIMV0691), and the study was performed in accordance with the Declaration of Helsinki. All participants provided written informed consent. In this age-stratified secondary analysis, we evaluated whether the clinical impact of DAPT de-escalation differs with age. The original trial prespecified 75 years as the threshold for defining the older age group, and age was modeled as a continuous variable in additional analyses. Spline curves were used to show the event rates and hazard ratios (HRs) for the de-escalation group versus the control group across the entire age range. 【Outcomes】 Consistent with the original TALOS-AMI study, the primary net clinical endpoint of this analysis was a composite of major adverse cardiovascular events (MACEs) and Bleeding Academic Research Consortium (BARC) type 2, 3, or 5, at 12 months after the index PCI. A MACE was defined as a composite of cardiovascular death, myocardial infarction, and stroke. Key secondary outcomes were MACEs; a composite of BARC type 2, 3, or 5 bleeding; each individual type of BARC bleeding; and a composite of MACE and BARC type 3 or 5 bleeding. Other secondary outcomes comprised the individual components of the primary net clinical endpoint as well as all-cause mortality, ischemia-driven revascularization, and stent thrombosis. 【Statistical analysis】 Continuous variables were reported as means ± standard deviations, whereas categorical variables were expressed as counts and percentages. Group comparisons for continuous variables were performed using the independent-samples t test or the Wilcoxon rank-sum test, as appropriate for the data distribution. Distributional assumptions were evaluated using visual inspection of histograms and quantile–quantile plots. Categorical variables were compared using the chi-square test. Time-to-event data were evaluated with Cox proportional-hazards models to assess the relationship between treatment strategy and clinical outcomes, including MACEs and bleeding events, with adjustment for age and sex. Interaction terms were tested within the Cox models to assess whether treatment effects varied according to age. We performed post hoc sensitivity analyses using alternative age cutoffs of 65 and 70 years. For each cutoff, treatment effects were estimated within the younger and older subgroups, and treatment-by-age-group interaction terms were tested in Cox proportional-hazards models. These alternative-cutoff analyses were considered exploratory. To further evaluate age as a continuous variable, treatment effects across the full age spectrum were examined using Cox regression models. To explore the temporal pattern of events, we summarized the cumulative distribution of event occurrence over time at prespecified time points. In addition, we performed a 180-day landmark analysis to evaluate whether the treatment effects differed between the early (≤180 days) and late (>180 days) periods. For the late-period analysis, only patients who remained event-free at the landmark time point were included, and follow-up was reinitiated from that time. Cox proportional-hazards models were applied separately within each period, and an interaction term was used to evaluate the differences in the treatment effects over time. As the original TALOS-AMI trial was powered for the overall randomized cohort rather than age-specific subgroup comparisons, a post hoc power analysis was performed for patients aged ≥75 years for the primary net clinical endpoint and BARC type 2, 3, or 5 bleeding by using a two-sided α of 0.05 and the observed subgroup event rates. Analyses were conducted using IBM SPSS Statistics (version 30; IBM Corp., Armonk, NY, USA) and R software (version 4.2.2; R Foundation for Statistical Computing, Vienna, Austria). A two-tailed significance threshold of 0.05 was applied for all statistical tests. 【Results】 【Study population】 The study sample had a mean (standard deviation) age of 60.0 (11.4) years; among them, 452 (16.8%) were female. Of the 2697 participants, 1349 were assigned to the aspirin plus clopidogrel (de-escalation) group and 1348 to the continued aspirin plus ticagrelor (active-control) group. All of them were included in this age-stratified analysis (Figure S1). Most baseline demographic and procedural characteristics were balanced between the treatment groups in both the younger and older cohorts ( and ). However, the estimated glomerular filtration rate and infarct-related artery differed between the groups in the younger cohort, and the history of PCI differed between the groups in the older cohort. Baseline characteristics of patients according to age group. Data are presented as n (%) or means ± standard deviations. For smoking, data were missing for one patient in the younger active-control group. For eGFR (<60 m L/min), data were missing for 25 patients in the younger de-escalation group, 21 in the younger active-control group, 3 in the older de-escalation group, and 2 in the older active-control group. For LVEF <40%, data were missing for 21 patients in the younger de-escalation group, 33 in the younger active-control group, 3 in the older de-escalation group, and 11 in the older active-control group. CABG: coronary artery bypass grafting; CVA: cerebrovascular accident; CVD: cardiovascular disease; eGFR: estimated glomerular filtration rate; LVEF: left ventricular ejection fraction; NSTEMI: non-ST-segment elevation myocardial infarction; PCI: percutaneous coronary intervention; STEMI: ST-segment elevation myocardial infarction. The de-escalation group was administered aspirin plus clopidogrel. The active-control group was administered aspirin plus ticagrelor. Impaired kidney function was defined as an eGFR <60 mL/min/1.73 m2 of body-surface area at presentation. Procedural characteristics according to age. Data are presented as n (%) or means ± standard deviations. For the access site, data were missing for two patients in the younger active-control group. For the infarct-related artery, data were missing for one patient in the younger de-escalation group and two in the younger active-control group. GP: glycoprotein; IVUS: intravascular ultrasound; LAD: left anterior descending artery; LCX: left circumflex artery; LM: left main coronary artery; No.: number; OCT: optical coherence tomography; RCA: right coronary artery; No. of stents: number of stents for infarct-related artery. The de-escalation group was administered aspirin plus clopidogrel. The active-control group was administered aspirin plus ticagrelor. 【Outcomes in the prespecified age subgroups】 Results for the primary and key secondary outcomes are shown in and , whereas those for other secondary outcomes are detailed in Table S1. Among patients aged <75 years, the incidence of the primary net clinical endpoint was lower in the de-escalation group than in the active-control group (49 [4.1%] vs. 85 [7.2%]; adjusted HR [aHR], 0.54; 95% confidence interval [CI], 0.38–0.77; p < 0.001). The incidence of BARC type 2, 3, or 5 bleeding, a key secondary outcome, was also lower in the de-escalation group versus the active-control group (33 [2.8%] vs. 58 [4.9%] events; aHR, 0.54; 95% CI, 0.35–0.82; p = 0.004). In contrast, the incidence of MACEs, a key secondary outcome, did not differ between strategies in the younger cohort (1.7% vs. 2.5%; aHR, 0.63; 95% CI, 0.36–1.12; p = 0.11). For other secondary outcomes among patients aged <75 years, the de-escalation and active-control groups did not differ in terms of all-cause death (7 [0.6%] vs. 6 [0.5%]; aHR, 1.10; 95% CI, 0.37–3.25; p = 0.876), any myocardial infarction (10 [0.8%] vs. 17 [1.4%]; aHR, 0.56; 95% CI, 0.26–1.22; p = 0.145), target-vessel revascularization (16 [1.3%] vs. 15 [1.3%]; aHR, 1.00; 95% CI, 0.50–2.02; p = 0.998), or stent thrombosis (3 [0.3%] vs. 2 [0.2%]; aHR, 1.43; 95% CI, 0.24–8.55; p = 0.697). Kaplan–Meier curves stratified according to age. (A–C) represent patients aged <75 years, and (D–F) represent patients aged ≥75 years. The primary net clinical endpoint was a composite of cardiovascular death, myocardial infarction, stroke, and BARC type 2, 3, or 5 from 1 to 12 months after PCI. MACEs were defined as the composite of cardiovascular death, stroke, and myocardial infarction. BARC bleeding comprises type 2, 3, or 5 events according to the BARC criteria. BARC: Bleeding Academic Research Consortium; d, days; MACE: major adverse cardiovascular event; PCI: percutaneous coronary intervention. Primary and key secondary outcomes. Adjustments were for age and sex. adj: adjusted; BARC: Bleeding Academic Research Consortium; HR: hazard ratio; int: interaction; MACE: major adverse cardiovascular event. Among patients aged ≥75 years, no statistically significant between-group differences were observed for the primary net clinical endpoint (10 [6.4%] vs. 19 [11.6%]; aHR, 0.54; 95% CI, 0.25–1.17; p = 0.120), MACEs (7 [4.5%] vs. 8 [4.9%]; aHR, 0.88; 95% CI, 0.32–2.43; p = 0.801), or BARC type 2, 3, or 5 bleeding (5 [3.2%] vs. 13 [7.9%]; aHR, 0.41; 95% CI, 0.15–1.15; p = 0.089). Formal interaction testing showed no evidence that the treatment effect differed between patients aged <75 and ≥75 years for the primary net clinical endpoint (p for interaction = 0.978), MACE (p for interaction = 0.585), or BARC type 2, 3, or 5 bleeding (p for interaction = 0.597). For other secondary outcomes among patients aged ≥75 years, no differences were observed between the groups in terms of all-cause mortality (4 [2.5%] vs. 4 [2.4%]; aHR, 1.01; 95% CI, 0.25–4.05; p = 0.991), any myocardial infarction (2 [1.3%] vs. 3 [1.8%]; aHR, 0.65; 95% CI, 0.11–3.94; p = 0.642), and target-vessel revascularization (1 [0.6%] vs. 2 [1.2%]; aHR, 0.51; 95% CI, 0.05–5.70; p = 0.588). 【Temporal distribution of events and 180-day landmark analysis】 For BARC type 2, 3, or 5 bleeding, 14/18 events (77.8%) in patients aged ≥75 years occurred within 180 days of group allocation, compared with 54/91 events (59.3%) among patients aged <75 years (Table S2). In the 180-day landmark analysis (Table S3, and ), de-escalation was associated with a lower risk of BARC bleeding during the first 180 days in the overall cohort (aHR, 0.40; 95% CI, 0.24–0.67) and in patients aged <75 years (aHR, 0.39; 95% CI, 0.22–0.70). No heterogeneity was observed between the time periods (p for interaction >0.05). Among patients aged ≥75 years, no differences were observed between the de-escalation and active-control groups in either period (≤180 days: aHR, 0.43; 95% CI, 0.14–1.39; >180 days: aHR, 0.32; 95% CI, 0.03–3.09; p for interaction = 0.840). Kaplan–Meier curves for the 180-day landmark analysis in patients aged <75 years. (A) Primary net clinical endpoint, (B) MACE, and (C) BARC type 2, 3, or 5 bleeding. For each outcome, the left-hand plot shows the cumulative event incidence from randomization through day 180. The right-hand plot shows the conditional cumulative event incidence after day 180 among participants who were event-free and remained under observation at the landmark. Event probability was reset to zero at day 180, whereas the x-axis retains the time elapsed since randomization. The dashed vertical line marks the 180-day landmark. Displayed p values were derived from Cox proportional hazards models adjusted for age and sex. The primary net clinical endpoint was a composite of cardiovascular death, myocardial infarction, stroke, and BARC type 2, 3, or 5 bleeding. MACE was defined as a composite of cardiovascular death, myocardial infarction, and stroke. BARC bleeding comprised type 2, 3, or 5 bleeding events according to the BARC criteria. BARC: Bleeding Academic Research Consortium; MACE: major adverse cardiovascular event; PC: percutaneous coronary intervention. Kaplan–Meier curves for the 180-day landmark analysis in patients aged ≥75 years. (A) Primary net clinical endpoint, (B) MACE, and (C) BARC type 2, 3, or 5 bleeding. For each outcome, the left-hand plot shows the cumulative event incidence from randomization through day 180. The right-hand plot shows the conditional cumulative event incidence after day 180 among participants who were event-free and remained under observation at the landmark. Event probability was reset to zero at day 180, whereas the x-axis retains the time elapsed since randomization. The dashed vertical line marks the 180-day landmark. Displayed p values were derived from Cox proportional hazards models adjusted for age and sex. The primary net clinical endpoint was a composite of cardiovascular death, myocardial infarction, stroke, and BARC type 2, 3, or 5 bleeding. MACE was defined as a composite of cardiovascular death, myocardial infarction, and stroke. BARC bleeding comprised type 2, 3, or 5 bleeding events according to the BARC criteria. BARC: Bleeding Academic Research Consortium; MACE: major adverse cardiovascular event; PCI: percutaneous coronary intervention. 【Sensitivity analyses using alternative age cutoffs】 In post hoc sensitivity analyses, alternative definitions of older age did not change the findings. Among patients aged ≥65 years, the primary net clinical endpoint occurred in 30 of 473 patients (6.3%) in the de-escalation group and 44 of 475 patients (9.3%) in the active-control group (adjusted HR, 0.66; 95% CI, 0.41–1.04; p = 0.074). Among patients aged ≥70 years, the corresponding event rates were 7.2% and 9.8%, respectively (adjusted HR, 0.72; 95% CI, 0.41–1.24; p = 0.235). No statistically significant treatment-by-age interaction was detected using either the 65-year cutoff (p for interaction = 0.275) or the 70-year cutoff (p for interaction = 0.239). Similarly, no significant interactions were observed for MACE or BARC type 2, 3, or 5 bleeding using either cutoff (all adjusted p for interaction ≥ 0.094; Table S4). 【Analysis of age-dependent effect with age as continuous variable】 When age was modeled as a continuous variable, the predicted cumulative event rates generally increased with advancing age. However, no statistically significant treatment-by-age interaction was observed for the primary net clinical endpoint, MACE, or BARC type 2, 3, or 5 bleeding (p for overall RCS interaction = 0.571, 0.945, and 0.329, respectively). No significant nonlinear interaction was identifi