Implications of cardiovascular outcome trials with injectable antidiabetic agents
注射用抗糖尿病药物心血管转归试验的意义
Injectable therapies, including glucagon-like peptide-1 (GLP-1) receptor agonists and insulin, play an important role in the treatment of type 2 diabetes mellitus (T2DM) and consequently are prominently featured in both the American Diabetes Association (ADA) and American Association of Clinical Endocrinologists (AACE) treatment guidelines.1, 2 Following the release of the 2008 US Food and Drug Administration (FDA) Guidance for Industry, all new drugs in development for the treatment of T2DM were required to demonstrate that they did not result in an unacceptable increase in cardiovascular risk.3 Thus, GLP-1 receptor agonists, many of which were in development at the time the guidance was released, have been required to perform robust cardiovascular outcome trials as a post-marketing requirement of their approval. In addition, the FDA has recently required a post-marketing cardiovascular safety study for a basal insulin. In this paper, the results of these cardiovascular outcomes trials are reviewed and the implications for treatment are discussed.
Glucagon-like peptide-1 receptor agonists
To date, there have been five pre- and post-marketing cardiovascular outcomes trials completed with GLP-1 receptor agonists.
The Evaluation of Lixisenatide in Acute Coronary Syndrome (ELIXA) trial (NCT01147250) randomized 6068 subjects with T2DM and a recent acute coronary syndrome (ACS) event to treatment with lixisenatide, a short-acting exenatide-derived GLP-1 receptor agonist, or placebo.4 After a median follow-up of 25 months, there were 406 primary major adverse cardiac events (MACE) in the lixisenatide group and 399 in the placebo group (hazard ratio [HR] 1.02; 95% confidence interval 0.89–1.17). Thus, lixisenatide was non-inferior to placebo (P < 0.001), but did not demonstrate superiority.
In the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) study (NCT01179048), 9340 patients with T2DM and high cardiovascular risk (~80% with prior cardiovascular disease events) were randomized to liraglutide, a human GLP-1 analog, or placebo.5 After a median follow-up of 3.8 years, 608 primary MACE occurred in the group randomized to liraglutide, compared with 694 in the group randomized to placebo (HR 0.87; 95% CI 0.78–0.97; P < 0.001 for non-inferiority; P = 0.01 for superiority). In addition, fewer patients in the liraglutide group died than in the placebo group (219 vs 278; HR 0.78; 95% CI 0.66–0.93; P = 0.007).
The Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes (SUSTAIN-6; NCT01720446) was a preapproval trial comparing the effects of two doses (0.5 and 1.0 mg) of semaglutide, a human GLP-1 analog, with placebo in 3297 patients with T2DM and high cardiovascular risk, 83.0% of whom had established cardiovascular disease, chronic kidney disease, or both.6 After a median follow-up of 2.1 years, the primary MACE outcome occurred in 108 patients randomized to semaglutide and in 146 patients randomized to placebo (HR 0.74; 95% CI 0.58–0.95; P < 0.001 for non-inferiority). Although the 26% reduction in the primary endpoint observed with semaglutide was remarkable, the study was neither designed nor powered to detect superiority. In contrast with earlier studies, an increase in retinopathy events was noted in patients randomized to semaglutide compared with placebo (HR 1.76; 95% CI 1.11–2.78; P = 0.02). This adverse effect has been attributed to the rapid and marked improvement in glycemia in patients with pre-existing retinopathy and high baseline HbA1c levels, similar to that reported with insulin treatment.6
The effects of the GLP-1 receptor agonist exenatide delivered once weekly in a depot formulation were compared with placebo in the Exenatide Study of Cardiovascular Event Lowering (EXSCEL) trial (NCT01144338) in 14 752 patients with T2DM and cardiovascular risk (73% with previous cardiovascular disease).7 After a median follow-up of 3.2 years, the primary end-point occurred in 839 patients randomized to exenatide and in 905 patients randomized to placebo (HR 0.91; 95% CI 0.83–1.00; P < 0.001 for non-inferiority; P = 0.06 for superiority).
Finally, top-line results in the pre-approval FREEDOM-CVO trial (NCT01455896), which compared ITCA 650 (exenatide delivered using an implanted mini-osmotic pump; Intarcia Therapeutics, Boston, MA, USA) with placebo in over 4000 patients with T2DM and high cardiovascular risk, demonstrated the cardiovascular safety of the product, but did not apparently demonstrate a benefit.8 In addition to these completed studies, the Researching Cardiovascular Events with a Weekly Incretin in Diabetes (REWIND) study (NCT01394952) comparing dulaglutide to placebo in 9900 patients with T2DM and high cardiovascular risk completed recruitment in 2013 and is scheduled for completion in 2019.9 Other robust cardiovascular outcomes trials are in the design phase for newer GLP-1 receptor agonists.
To date, all trials with GLP-1 receptor agonists have met their primary objective of demonstrating the cardiovascular safety of the product. The reason why some GLP-1 receptor agonists showed an apparent cardiovascular benefit (liraglutide, semaglutide) whereas others did not is unknown. Is it possible that patients with acute coronary syndrome (in the ELIXA trial) are too advanced for a GLP-1 receptor agonist to manifest a benefit? Or perhaps the cardiovascular benefit requires a sustained 24-h GLP-1 effect, which is not seen with lixisenatide? If that is the case, why did the well-powered EXSCEL trial with exenatide once weekly narrowly miss achieving statistical significance for superiority? Could differences in the molecules account for the differences in outcomes? It is notable that the two trials showing an apparent benefit were conducted with human GLP-1 receptor analogs,5, 6 whereas those trials not demonstrating a benefit used exenatide-based products.4, 7, 8 Ultimately, we cannot know with certainty, but it is likely that differences in the design, target patient population, size, duration of exposure to the drug and execution account for much of the heterogeneity in outcomes.
Basal insulins
In addition to studies with GLP-1 receptor agonists, there are also robust cardiovascular outcomes trials with insulin glargine and insulin degludec, two widely used basal insulin analogs. The Outcome Reduction with an Initial Glargine Intervention (ORIGIN) study (NCT00069784) compared the effects of insulin glargine with usual care (which did not use insulin) in 12 537 patients with cardiovascular risk factors and prediabetes or early diabetes.10 After a median follow-up of 6.2 years, there were no significant differences in the rates of either of the coprimary endpoints (three- or four-point MACE), but insulin glargine was associated with an increased risk of severe hypoglycemia.
In the recent Trial Comparing Cardiovascular Safety of Insulin Degludec versus Insulin Glargine in Patients with Type 2 Diabetes at High Risk of Cardiovascular Events (DEVOTE; NCT01720446), insulin degludec was compared to insulin glargine in a randomized double-blind parallel arm trial in 7637 patients with T2DM and high cardiovascular risk (85% of whom had established cardiovascular disease, chronic kidney disease, or both).11 After a median follow-up of 2 years, the primary outcome of MACE occurred in 325 patients randomized to degludec and in 356 patients randomized to glargine (HR 0.91; 95% CI 0.78–1.06; P < 0.001 for non-inferiority). Notably, severe hypoglycemia was significantly less frequent among those in the degludec than glargine group (relative risk [RR] 0.60; P < 0.001 for superiority).
What are the implications of these studies for the treatment of patients with T2DM? Based on the available evidence, it appears that GLP-1 receptor agonists can be used safely in patients with T2D and high cardiovascular risk. However, there does not seem to be a class effect with regard to to lowering the risk of cardiovascular events. Among the completed trials, only the LEADER trial demonstrated superiority, which led to an update of the product label by the FDA. Liraglutide is now indicated “to reduce the risk of major adverse cardiovascular events in adults with type 2 diabetes mellitus and established cardiovascular disease.”12 Recent updates to the ADA and AACE treatment guidelines also reflect this new indication.1, 2 The DEVOTE trial demonstrated that insulin degludec was non-inferior to insulin glargine with regard to cardiovascular outcomes, but was associated with a significantly lower risk of severe hypoglycemia. This information is also now included in the insulin degludec label. Thus, data from these cardiovascular outcomes trials not only provide information on the cardiovascular safety of injectable agents for T2DM, but they also provide robust data on which to base treatment decisions when selecting agents to decrease cardiovascular disease risk or hypoglycemia. Although beyond the scope of this review, these trials have provided valuable information on the overall safety of these drugs, including rare events, such as pancreatitis, in a broad population of patients with T2DM.
In summary, recent robust cardiovascular outcomes studies have provided reassuring data about the cardiovascular safety of GLP-1 receptor agonists and basal insulin, and suggest that some products may be associated with a significant benefit. This information is invaluable to clinicians seeking to optimize outcomes in complicated patients with T2DM and multiple comorbidities.
Acknowledgements
The author is supported, in part, by the National Institutes of Health (DP3 DK113358, U01 DK098245) and the Helmsley Trust/Juvenile Diabetes Research Foundation (3-SRA-2017-425-M-R).
Disclosure
REP has received consulting and/or speaker fees from AstraZeneca, Takeda, and Novo Nordisk; consulting fees from Boehringer Ingelheim, GlaxoSmithKline, Hanmi Pharmaceutical, Janssen Scientific Affairs, Ligand Pharmaceuticals, Eli Lilly, Merck, Pfizer, and Eisai; and research grants from Gilead Sciences, Lexicon Pharmaceuticals, Ligand Pharmaceuticals, Eli Lilly, Merck, Novo Nordisk, Sanofi US, and Takeda. All fees for REP's services were paid directly to Florida Hospital, a non-profit organization.
注射疗法, 包括胰高血糖素样肽-1(GLP-1)受体激动剂和胰岛素, 在治疗2型糖尿病(T2DM)中发挥了重要作用, 因此被收进了在美国糖尿病协会(ADA)和美国内分泌医师协会(AACE)治疗指南中重度推荐1,2。在2008年美国食品和药品管理局(FDA)发布业界指导后, 所有用于治疗T2DM的新药都要求证明它们不会导致无法接受的心血管风险增加3。因此,GLP-1受体激动剂, 在该指导发布时, 许多还处于开发过程中, 被要求进行强有力的心血管转归试验作为通过审批的上市后要求。此外,FDA最近还要求对基础胰岛素进行上市后心血管安全性研究。在本文中回顾了这些心血管转归试验的结果, 并讨论了治疗的意义。
胰高血糖素样肽-1受体激动剂
迄今为止, 已经完成了5项上市前和上市后GLP-1受体激动剂心血管转归试验。
利西拉来治疗急性冠状动脉综合征的评估(Evaluation of Lixisenatide in Acute Coronary Syndrome,ELIXA)试验(NCT01147250)对6068名患有T2DM和最近发生急性冠状动脉综合征(acute coronary syndrome,ACS)事件的受试者随机分配至利西拉来(一种短效艾塞那肽衍生的GLP-1受体激动剂)或安慰剂治疗4。中位随访25个月后, 利西拉来组有406例原发性主要不良心脏事件(major adverse cardiac events,MACE),安慰剂组有399例(危险比[HR] 1.02;95% CI 0.89-1.17)。因此, 利西拉来不劣于安慰剂(P <0.001),但也没有表现出优越性。
在利拉鲁肽治疗糖尿病的疗效和机制:评估心血管转归结果(Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results,LEADER)研究(NCT01179048)中,9340名有较高心血管风险的T2DM患者(约80%既往发生过心血管疾病事件)被随机分配接受人GLP-1类似物利拉鲁肽或安慰剂治疗5。中位随访3.8年后, 利拉鲁肽组发生608例原发性MACE,而安慰剂组发生694例(HR 0.87;95% CI 0.78-0.97;非劣效性P = 0.001;优势P = 0.01)。此外, 利拉鲁肽组的患者死亡人数少于安慰剂组(219 vs 278;HR 0.78;95% CI 0.66-0.93;P = 0.007)。
在2型糖尿病患者中使用索马鲁肽评估心血管和其他长期转归的试验(Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes,SUSTAIN-6; NCT01720446)是一项审批前试验, 在3297名有较高心血管风险的T2DM患者(83.0%患有心血管疾病和/或慢性肾脏疾病)中, 比较两种剂量(0.5和1.0 mg)的人GLP-1类似物索马鲁肽和安慰剂的疗效6。中位随访2.1年后, 索马鲁肽组中有108例患者发生原发性MACE结果, 而安慰剂组为146例患者(HR 0.74;95%CI 0.58-0.95;非劣效性P < 0.001)。虽然在索马鲁肽组观察到的主要终点减少26%非常显著, 但此项研究既没有设计, 也没有效力来检测优越性。与之前的研究相比, 索马鲁肽组患者中的视网膜病变事件发生率高于安慰剂组(HR 1.76;95% CI 1.11-2.78;P = 0.02)。这种不良反应归因于在已有视网膜病变和高基线 HbA1c水平的患者中血糖迅速显著改善, 这与胰岛素治疗相似6。
在艾塞那肽降低心血管事件研究(Exenatide Study of Cardiovascular Event Lowering,EXSCEL)(NCT01144338)中, 在14752名有心血管风险的T2DM患者(73%既往曾有心血管疾病)中, 比较了每周一次长效GLP-1受体激动剂艾塞那肽和安慰剂的疗效7。中位随访3.2年后, 艾塞那肽组中有839名患者发生主要终点, 安慰剂组中有905名患者发生主要终点(HR 0.91;95%CI 0.83-1.00;非劣效性P < 0.001;优越性P = 0.06)。
最后, 审批前的FREEDOM-CVO试验(NCT01455896),在超过4000名有较高心血管风险的T2DM患者中, 比较了ITCA 650(使用植入式微型渗透泵给予艾塞那肽;Intarcia Therapeutics,Boston,MA,USA)与安慰剂的疗效, 结果显示此产品具有心血管安全性, 但是没有明显显示出治疗益处8。除了这些已完成的研究外, 在糖尿病患者中每周一次肠促胰岛素的心血管事件(Researching Cardiovascular Events with a Weekly Incretin in Diabetes,REWIND)研究(NCT01394952)比较了度拉糖肽与安慰剂对9900名有较高心血管风险的T2DM患者的疗效, 此项研究在2013年完成招募, 计划于2019年完成9。其他针对更新的GLP-1受体激动剂的稳健的心血管转归试验正处于研究设计阶段。
迄今为止, 所有使用GLP-1受体激动剂的试验都达到了证明产品具有心血管安全性的主要目的。一些GLP-1受体激动剂显示出明显的心血管益处(利拉鲁肽、司美鲁肽)而其他产品没有显示益处的原因尚不清楚。急性冠状动脉综合征患者(在ELIXA试验中)的疾病病程是否可能过于晚期而不能通过GLP-1受体激动剂治疗显示出益处?或许心血管疾病需要持续24小时的GLP-1效应, 但是利西拉来没有这种持续效应?如果是这样的话, 为什么在效力良好的EXSCEL试验中使用每周一次艾塞那肽几乎没有显示出统计数据的优势?分子方面的差异是否可以解释转归差异?值得注意的是, 显示出明显益处的两个试验使用的是人GLP-1受体类似物5,6,而没有显示益处的试验使用了基于艾塞那肽的产品4,7,8。最终, 虽然我们无法确切知道, 但是可能是设计、目标患者人群、研究规模、药物暴露时间和执行时间的差异导致了大部分转归方面的差异。
基础胰岛素
除了使用GLP-1受体激动剂的研究外, 还有一些使用甘精胰岛素和德谷胰岛素进行稳健的心血管转归试验, 这是两种广泛使用的基础胰岛素类似物。甘精胰岛素初始干预结果(The Outcome Reduction with an Initial Glargine Intervention,ORIGIN)研究(NCT00069784)在12537名有心血管风险因素的糖尿病前期或早期糖尿病患者中比较了甘精胰岛素与常规治疗(未使用胰岛素)的效果10。中位随访6.2年, 任何主要终点(3或4点MACE)的发生率都没有显著差异, 但甘精胰岛素与发生严重低血糖的风险增加有关。
最近的发生心血管事件风险较高的2型糖尿病患者中比较德谷胰岛素和胰岛素甘精胰岛素心血管安全性的试验(Trial Comparing Cardiovascular Safety of Insulin Degludec versus Insulin Glargine in Patients with Type 2 Diabetes at High Risk of Cardiovascular Events,DEVOTE; NCT01720446)是一项随机双盲平行臂试验, 在7637名有较高心血管风险的T2DM患者(85%患有心血管疾病和/或慢性肾病)中比较了德谷胰岛素和甘精胰岛素11。中位随访2年后, 在德谷胰岛素治疗组中有325例患者发生MACE主要转归, 而在甘精胰岛素治疗组中为356例患者(HR 0.91;95%CI 0.78-1.06;非劣效性P < 0.001)。值得注意的是, 在德谷胰岛素治疗患者中, 严重低血糖的发生率显著低于甘精胰岛素组(相对风险 [RR] 0.60;优势P < 0.001)。
这些研究对T2DM患者的治疗意义何在?根据现有证据, 似乎GLP-1受体激动剂可安全用于治疗有较高心血管风险的T2DM患者。然而, 在降低心血管事件的风险方面似乎没有类效应。在已完成的试验中, 只有LEADER试验显示出优势, 这导致FDA更新了产品标签。利拉鲁肽目前显示适用于“减少明确患有心血管疾病的2型糖尿病成人患者的主要不良心血管事件的风险12”。最近ADA和AACE治疗指南的更新也反映了这个新的适应证1,2。DEVOTE试验表明, 在心血管转归方面, 德谷胰岛素不劣于甘精胰岛素, 但与发生严重低血糖的风险显著降低有关。此信息现在也包含在德谷胰岛素的说明书中。因此, 来自这些心血管转归试验的数据不仅提供了关于T2DM注射剂心血管安全性的信息, 而且它们还提供了强有力的证据, 可以在选择降低心血管疾病风险或低血糖的药物时做出治疗决策。虽然超出了本次综述的范围, 但这些试验为大量T2DM患者提供了有关这些药物总体安全性的有价值的信息, 包括罕见事件, 例如胰腺炎。总之, 最近稳健的心血管转归研究已经提供了关于GLP-1受体激动剂和基础胰岛素的心血管安全性的令人放心的数据, 并且表明一些产品可能提供显著益处。这些信息对于希望在复杂的患有多种合并症的T2DM患者中优化转归的临床医生来说是非常宝贵的。
