Potential for cardiac protection with dipeptidyl peptidase-4 inhibitors: The stromal cell-derived factor-1α hypothesis (二肽基肽酶-4抑制剂潜在的心脏保护作用:基质细胞衍生因子-1α假说)


The prevention of cardiovascular disease is among our highest priorities in managing patients with diabetes. The concerns that arose from the association between rosiglitazone use and the incidence of myocardial infarction have led drug regulatory authorities to mandate evidence of cardiovascular safety before a new antihyperglycemic agent can be approved. For the dipeptidyl peptidase (DPP)-4 inhibitors, the availability and analysis of these data have, to date, not only shown apparent safety but have also raised the tantalizing possibility of cardiovascular protection. There are, however, numerous caveats to such analyses and their interpretation. Foremost, the trials are primarily designed to determine the glucose-lowering properties of these agents and not their effects on cardiovascular outcomes, which are instead captured as adverse events. In addition, most of these studies are of short duration, have relatively few patients and mostly do not focus on subjects with pre-existing heart disease or at high risk of developing it. These limitations notwithstanding, some interesting findings are evident.

A series of meta-analyses examining the cardiovascular effects of DPP-4 inhibitors have been conducted over recent years. A very recent one of these included all published and unpublished studies in humans, examining a total of 70 trials with 41 959 patients and a mean follow-up of 44.1 weeks.[1] Consistent with previous, smaller studies, this meta-analysis reported Mantel–Haenszel odds ratios of 0.71 (95% confidence intervals [CI] 0.59, 0.86), 0.64 (95% CI 0.44, 0.94) and 0.60 (95% CI 0.41, 0.88) for major adverse cardiovascular events (MACE), myocardial infarction and mortality, respectively. Of note, the effect appeared consistent among individual drugs within the DPP-4 inhibitor class. In addition, the analysis estimated the effects that may be attributable to risk factor modification by examining data on HbA1c, blood pressure and lipids, when available. Here, the authors concluded that the reduction in myocardial infarction was greater than may be predicted on the basis of changes in conventional risk factors, leading them to suggest a role for other mechanisms.[1]

For endocrinologists, familiar with the incretin effect, it is tempting to speculate that the modest increases in glucagon-like peptide-1 may account for the observed cardiovascular effects of DPP-4 inhibitors. However, as described below, other studies (mostly in the cardiac literature) suggest that other mechanisms may be operating, focusing instead on the role of stromal cell-derived factor (SDF)-1α. This chemokine is constitutively expressed by the bone marrow, where it assists in the maintenance and differentiation of hematopoietic and vascular progenitors. However, its expression may also be induced by hypoxia-inducible factor-1-dependent mechanisms in peripheral tissues in response to ischemia.[2] For example, following myocardial infarction, SDF-1α expression in the heart is markedly increased.[3] This upregulation attenuates injury and assists in cardiac repair by protecting cardiac myocytes from apoptosis and enhancing angiogenesis by both bone marrow-derived and local endothelial-based mechanisms (Fig. 1).[2]

Figure 1.

Potential role of the stromal cell-derived factor (SDF)-1α–CXC chemokine receptor 4 (CXCR4) system in myocardial infarction. At the site of tissue ischemia, such as the ischemic myocardium, hypoxia-inducible factor (HIF)-1α is induced in response to reduced oxygen tension, which then stimulates SDF-1α expression. Binding of SDF-1α to its receptor CXCR4 protects cardiomyocytes from apoptotic cell death, recruits bone marrow-derived stem/progenitor cells and stimulates the migration of endothelial cells to enhance angiogenesis. Reproduced with permission from Takahashi.[2]

Given the newfound importance of the microvasculature in the setting of ischemic heart disease,[4, 5] the potential role of SDF-1α as a therapeutic agent has recently been explored. In a study of mice with experimental myocardial infarction, intracardiac injection of recombinant SDF-1α led to improved cardiac function and reduced scar formation compared with controls.[6] Similar benefits have also been reported using a protease-resistant SDF-1α, with evidence of enhanced progenitor cell recruitment, increased capillary density, and improved cardiac function after myocardial infarction.[7] Indeed, based on the positive findings of such reports, Juventas, a biotechnology company, has developed a non-viral DNA plasmid that expresses SDF-1α to be used for the treatment of ischemic disorders. Their lead product, JVS-100, will be used in STOP-HF, a study to evaluate the safety and efficacy of JVS-100 administered to adults with ischemic heart failure that is currently recruiting participants (see http://clinicaltrials.gov [ClinicalTrials.gov identifier NCT01643590], accessed 14 March 2013).

Because SDF-1α is rapidly inactivated by DPP-4-mediated cleavage, inhibition of this enzyme presents an alternative strategy of augmenting local SDF-1α bioactivity.[8] Consistent with this notion, Zaruba et al. have shown that DPP-4 inhibition reduces cardiac infarct size in mice, an effect that is enhanced by the coadministration of granulocyte colony-stimulating factor (G-CSF) to mobilize bone marrow-derived endothelial progenitor cells.[9] That study not only provides the non-clinical data for a patent asserting the benefit of DPP-4 inhibition in ischaemic heart disease,[10] but also serves as the scientific basis for the Sitagrami Study examining the effects of sitagliptin with G-CSF in patients with acute myocardial infarction (see http://clinicaltrials.gov [ClinicalTrials.gov identifier NCT00650143], accessed 14 March 2013).

In individuals with diabetes, both soluble and cell surface (CD26) DPP-4 activities are increased.[11, 12] These findings provide a plausible explanation, at least in part, for the comparatively poor prognosis for patients with diabetes and ischaemic heart disease even after adjustment for clinical and angiographic variables.[13, 14] In addition, it offers a rational basis for hypothesizing that DPP-4 inhibition may be particularly cardioprotective in the diabetes setting. Indeed, in studies by us and others using diabetic animal models of cardiomyopathy and acute myocardial infarction, DPP-4 inhibition attenuated many of the structural and functional attributes of disease.[4, 12] The effects of DPP-4 inhibition have also been examined in humans with type 2 diabetes, whereby 4 weeks sitagliptin administration increased both plasma SDF-1α concentrations and the number of circulating endothelial progenitor cells.[15]

In conclusion, the SDF-1α hypothesis provides a plausible basis to explain the relatively fewer MACE in subjects treated with DPP-4 inhibitors reported in meta-analyses. However, caution is still needed to avoid overinterpreting the clinical trial data, which, to date, are largely based on adverse event reporting. Fortunately, definitive answers will hopefully soon be forthcoming, with a series of large prospective randomized controlled cardiovascular studies using a range of DPP-4 inhibitors currently in progress that are expected to be completed in 2013–2018 (see http://clinicaltrials.gov [ClinicalTrials.gov identifiers 01107886, 00790205, 01243424 and 00968708], accessed 14 March 2013).


RG is the Canadian Research Chair in Diabetes Complications and his research was supported, in part, by the Canada Research Chair Program, the Canadian Institutes of Health Research and the Heart and Stroke Foundation of Canada.


RG has received grant support and consulting fees from, and has given lectures for, Bristol-Myers Squibb, AstraZeneca and Merck.


最近几年已经有一系列的Meta分析调查了二肽基肽酶-4抑制剂对心血管的影响。其中最近有一个分析收录了所有发表的以及未发表的临床研究,总共调查了70个试验,患者总数为41959名,平均随访期为44.1周[1]。与既往更小一些的研究结果相类似,这个Mata分析报告的主要心血管不良事件(MACE)、心肌梗死以及死亡率的Mantel–Haenszel优势比分别为0.71(95%CI 0.59, 0.86)、0.64(95% CI 0.44, 0.94)以及0.60(95%CI 0.41, 0.88)。值得注意的是,在DPP-4抑制剂这个药物分类中每个药物的影响都相似。另外,通过可得到的HbA1c、血压以及血脂数据,这个分析还评估了校正危险因素后所造成的影响。在这里,作者的结论是心肌梗死的风险下降程度比可以预测的基于常规危险因素变化所导致的风险下降程度更大,由此推测还有其他作用机制的影响[1]。


Figure 1.

图1 间质细胞衍生因子(SDF)-1α–CXC趋化因子受体4(CXCR4)系统在心肌梗死中的潜在作用。在缺血组织部位,例如缺血心肌层,氧分压降低可诱导产生缺氧诱导因子(HIF)-1α,接着它可刺激SDF-1α的表达。SDF-1α与它的受体CXCR4结合后可防止心肌细胞凋亡,补充骨髓来源的干/祖细胞,刺激内皮细胞的迁移以增强血管生成作用。转载得到了Takahashi的许可[2]。

由于新近发现了微血管在缺血性心脏病环境中的重要性[4, 5],有人将SDF-1a作为潜在的治疗药物进行了研究。在一项实验心肌梗死小鼠的研究中,与对照组相比,心内注射重组的SDF-1α可改善心脏功能并减少瘢痕形成[6]。心肌梗死后使用抗蛋白酶的SDF-1α进行治疗的研究也报告了相似的获益结果,显示祖细胞恢复增强、毛细血管密度增加以及心肌梗死后心脏功能改善[7]。的确,根据这些报告中的阳性结果,Juventas,一家生物技术公司,已经研制出了一种非病毒的DNA质粒,它可以表达SDF-1α,用来治疗缺血性疾病。他们的主导产品,JVS-100,将被用于STOP-HF,一项用来评估缺血性心脏衰竭的成人使用JVS-100治疗的安全性与有效性的研究中,目前正在招募受试者(参见http://clinicaltrials.gov [ClinicalTrials.gov identifier NCT01643590],2013年3月14日访问)。

因为SDF-1α在DPP-4的介导下会被迅速地分解灭活,所以一种替代的治疗策略就是要抑制这种酶以增加局部SDF-1α的生物活性[8]。Zaruba等的研究结果与这个观念一致,他们发现抑制小鼠的DPP-4之后可以减少其心肌梗死面积,而且通过同时使用粒细胞集落刺激因子(G-CSF)动员骨髓来源的血管内皮祖细胞后可以增强这种效应[9]。这项研究不仅第一次为我们提供了DPP-4抑制作用有益于缺血性心脏病的非临床数据[10],而且它还是Sitagrami研究(调查西格列汀联合G-CSF对急性心肌梗死患者的影响)的科学依据(参见http://clinicaltrials.gov [ClinicalTrials.gov identifier NCT00650143],2013年3月14日访问)。

在糖尿病患者中,可溶解的以及细胞表面(CD26)的DPP-4活性都增加了[11, 12]。这些发现为我们提供了一个至少部分合理的解释,关于为什么合并糖尿病与缺血性心脏病的患者即使经过临床与血管造影变量的校正后其预后还是相对较差[13, 14]。另外,它还为抑制糖尿病患者的DPP-4可能具有特别的心脏保护作用这个假说提供了一个合理的依据。的确,在我们以及其他人使用糖尿病动物模型对心肌病与急性心肌梗死进行的研究中,DPP-4抑制剂可以缓解许多由疾病所导致的心脏结构与功能改变[4, 12]。在2型糖尿病患者中也调查了DPP-4抑制剂的作用,在这项研究中给予西格列汀治疗4周以后血浆SDF-1α浓度以及循环中的内皮祖细胞数量都有增加[15]。

总之,SDF-1α假说为我们提供了一个合理的依据,它可以解释在Meta分析中为什么使用DPP-4抑制剂治疗的受试者MACE的发生率相对较低。然而,我们仍然需要小心谨慎地避免过度解读这些临床试验数据,迄今为止的临床试验数据主要都是基于不良事件报告。幸运的是,不久之后就有希望看到明确的答案,有一系列使用各种DPP-4抑制剂的大型前瞻性随机对照心血管研究目前正在进行当中,它们预期在2013–2018年完成(参见http://clinicaltrials.gov [ClinicalTrials.gov identifiers 01107886、00790205、01243424与00968708],2013年3月14日访问)。




RG曾经接受过Bristol-Myers Squibb, Astra-Zeneca and Merck.等公司提供的资助与顾问费,并且为它们讲学过。