YINGLONG: A Multicenter, Prospective, Non-Interventional Study Evaluating the Safety and Tolerability of Ticagrelor in Chinese Patients with Acute Coronary Syndrome

In China, an estimated 290 million people have cardiovascular (CV) disease [1]. Acute coronary syndrome [ACS; which includes unstable angina (UA), non–ST-segment elevation myocardial infarction (NSTEMI), and ST-segment elevation myocardial infarction (STEMI)] has high morbidity, mortality, and economic burden [2, 3]. In China, the reported mortality rate of patients hospitalized with ACS is 5% [4], with poorer post-discharge outcomes for patients with NSTEMI compared to those with STEMI [5]. Improved adherence to ACS guidelines may help enhance patient outcomes in China [6].

Ticagrelor is an oral, direct- and rapid-acting reversible P2Y₁₂ receptor antagonist [7], approved for use in patients with ACS in the United States [8] and many other countries. Ticagrelor inhibits platelet aggregation by antagonizing the binding of adenosine diphosphate at the P2Y₁₂ receptor [9, 10]. Ticagrelor has an additional adenosine-mediated mechanism of action via therapeutically relevant inhibition of equilibrative nucleoside transporter 1 (ENT-1) to inhibit cellular adenosine uptake [11]. In China, co-administration of ticagrelor with aspirin (ASA) is indicated to reduce the rate of CV death, myocardial infarction (MI), and stroke in patients with ACS or a history of MI and with at least 1 high-risk factor of developing an atherothrombotic event [12]. European and Chinese guidelines recommend ticagrelor as first-line treatment for patients with ACS [13‐15].

The global PLATelet inhibition and patient Outcomes (PLATO) trial, which included a broad spectrum of 18,624 patients with ACS, showed ticagrelor to be superior to clopidogrel in reducing fatal and non-fatal CV events, without an increase in the rate of overall major bleeding [16]. Efficacy and safety results in the Asian subgroup were consistent with those of the overall population [17]; however, only a minority (n = 587, 3.1%) of patients in the PLATO trial were Chinese, 416 of whom were randomized in mainland China (AstraZeneca data on file). Ticagrelor was approved in 2012 by the China Food and Drug Administration (CFDA). However, based on CFDA drug intensive monitoring (DIM) regulations that came into force in 2011, further phase IV safety data on ticagrelor were required for patients with ACS in China. In accordance with this mandate, and to obtain a real-world assessment of the safety profile and CV event rates in a large ethnically Chinese population, 2 phase IV studies (DAYU and YINGLONG), comprising approximately 3000 Chinese patients with ACS, were conducted. The interventional DAYU study comprised 2004 Chinese patients with ACS [18].

The present post-authorization, non-interventional YINGLONG study assessed the safety and tolerability of ticagrelor treatment over 1 year in Chinese patients with ACS.

YINGLONG was a 1-year, non-interventional study of ticagrelor treatment in Chinese patients with ACS, meant to reflect actual clinical practice, and of a population consistent with the Chinese prescribing information for ticagrelor. The YINGLONG study demonstrated that, under real-world conditions, 90 mg of ticagrelor twice daily for up to 1 year is associated with a low incidence of major bleeding events, major CV events, and SAEs in Chinese patients with ACS. Concomitant use of ASA was reported in approximately 60% of patients, which was lower than expected. For example, data from the nationwide Clinical Pathways for Acute Coronary Syndromes in China (CPACS) study described ASA being used in > 90% of Chinese patients with ACS after discharge [4] and in 87% even after 12 months [19]. The low percentage of concomitant ASA use might be because of the investigators’ concern about bleeding, or low-risk patients being diagnosed with ACS, and may have contributed to the lower than expected rate of bleeding side effects. Under-reporting of prior and concomitant medications, such as ASA and statins, in this observational study also cannot be ruled out. The incidence of PLATO-defined major bleeding events was 1.1%, of which fatal/life-threatening bleeding had an incidence of 0.6%; SAEs excluding bleeding were reported in 9.8% of patients during treatment. The majority of the bleeding was minimal (n = 76, 7.3%), followed by minor (n = 25, 2.4%). Most of the bleeding events occurred within the first 6 months of the treatment. The cumulative incidence of the composite of CV death/MI/stroke was 3.6%. One death (cerebral hemorrhage) judged by the investigators to be related to ticagrelor use was reported.

Some of these results are comparable to other ticagrelor studies in patients with ACS, such as the DAYU [18], PLATO [16], and retrospective Asian subgroup analysis of PLATO [17] studies, although there are some important differences. The DAYU study [18] was also a phase IV study of ticagrelor use in Chinese patients with ACS. However, in addition to DAYU involving more patients, there were some important differences between these studies, both in design and outcomes. In the interventional DAYU and PLATO studies, the initiation of ticagrelor was part of the study protocol and the drug was provided to patients during the study, while in the non-interventional YINGLONG study, the decision to initiate ticagrelor was based on clinical practice independently from study participation and the product was not supplied by the sponsor. This can better reflect actual clinical practice and minimize potential impacts (whether positive or negative) on treatment compliance. Other differences between YINGLONG and DAYU were that YINGLONG had no collection of laboratory safety data included in the protocol and CV events were also reported as AEs, whereas the DAYU protocol included collection of safety samples and CV events were reported separately from AEs. Not including any protocol-mandated laboratory sampling in YINGLONG reduced the intervention of local clinical routines to a minimum.

Both Chinese ticagrelor phase IV studies—DAYU and YINGLONG—showed similar safety results and incidences of CV events. In DAYU, during the 1-year follow-up treatment with ticagrelor, the incidences of fatal/life-threatening and PLATO-defined major bleeding events were 0.8% and 1.3%, respectively; the incidence of the composite of CV death/MI/stroke was 4.2%, and SAEs other than bleeding were reported in 5.8% of the patients during treatment [18]. These rates were comparable with those found in YINGLONG.

However, there are important differences in the rates of major bleeding and dyspnea between PLATO [16] and YINGLONG that deserve mention. Such differences should be interpreted with respect to the different study designs, PLATO being a randomized, controlled, phase III trial and YINGLONG being a non-interventional, phase IV study. The higher rate of major bleeding reported in PLATO (11.6%) compared with YINGLONG (1.1%) could be accounted for by several reasons. For example, the PLATO trial required patients to not only have ACS but also additional risk factors, which may have impacted the bleeding risk. Patients planned for urgent coronary artery bypass grafting (CABG) are less likely to be put on ticagrelor because of label restrictions, and hence were not eligible for YINGLONG, whereas in the PLATO trial such patients were allowed. A large part of the major bleeds in the PLATO trial were procedure- and CABG-related [16]. There were more patients on heparin/low–molecular weight heparin in PLATO [16] (51.6%) than in YINGLONG (where 23.5% and 10.0% of patients reported taking enoxaparin sodium and heparin sodium, respectively, subsequent to the first dose of ticagrelor). The YINGLONG study included patients already initiated on ticagrelor, and physicians may have selected patients with an anticipated low bleeding risk for ticagrelor treatment; i.e., a selection bias because the study was non-interventional and the treatment decision was made independently of any aspect of the study. By contrast, the non-interventional design would be expected to reduce risk for other types of bias seen in interventional, highly controlled studies with frequent visits and a free drug supply.

The lower rate of dyspnea (3.6%) in the YINGLONG study than that reported in the PLATO trial (13.8%) [16] or in the subgroup of East Asian patients (13.4%, 37/276) [17] in the PLATO trial could be accounted for by several factors. Because PLATO was conducted between 2006 and 2008 prior to ticagrelor’s approval, and YINGLONG between 2015 and 2017 after ticagrelor’s approval, patients with dyspnea—a known AE of ticagrelor—may have been less likely to accept participation in the YINGLONG study. Furthermore, cardiologists may have been less likely to initiate ticagrelor in patients with conditions associated with dyspnea. In PLATO, 15.1% of patients treated with ticagrelor had a history of dyspnea, whereas this information was not recorded in YINGLONG. Despite standardized training of investigators prior to study commencement in YINGLONG, under-reporting of events cannot be ruled out and may have also contributed to the lower rates of dyspnea observed in YINGLONG compared with PLATO. Intrinsic patient factors might also account for these differences, although there were proportionately similar numbers of females and patients aged greater than 75 years in both studies. However, the low number of Asian patients in PLATO prohibits any speculation. Dyspnea rates in YINGLONG were similar to those found in the PHILO study of patients with ACS from Japan, Taiwan, and South Korea, where the dyspnea rate was 5.7% (22/401) [20], and in the DAYU study [18], where the dyspnea rate was 3.4% (68/2001).

YINGLONG has several strengths and limitations. Consistent with the inherent nature of an observational study design and the lack of a control group, causality cannot be inferred. Events may have been under-reported, and, because of the low event rates, it is possible that a preponderance of low-risk patients was included. However, patient characteristics were mostly comparable with those of the PLATO trial, and similar to those seen in clinical practice, as well as other observational studies of patients with ACS in China [5, 20]. Although patients had to pay for ticagrelor in the YINGLONG study, the patients’ possible differential ability to pay is unlikely to have influenced the results because in the acute stages the choice of treatment depends more on the severity of the disease. However, the impact of patients’ ability to pay on treatment selection over the duration of the study cannot be excluded. A strength of the YINGLONG study is its prospective, non-interventional design, where the choice of treatment followed standard practice, which could increase external validity compared to a randomized controlled trial. Nonetheless, because the treatment was not randomized, unknown confounders could have influenced the results and causality between the intervention and the outcomes cannot be determined. In China, tertiary care hospitals represent the apex of medical care; thus, given the generalizability of our findings based on the study design, the results reported here can influence medical care at lower levels. However, the results may also not be applicable to all of China given the inclusion of only a limited number of tier 3 hospitals. It is therefore possible that the patients at these sites could differ from those in other hospitals throughout China. Moreover, because clopidogrel is commonly prescribed in China for patients with ACS, our results are particularly important given the high rates of cytochrome P450 family 2 subfamily C member 19 (CYP2C19) loss-of-function alleles present in Chinese patients [21‐23], which increases the risk of adverse CV outcomes with clopidogrel [23, 24]. Chinese patients with ACS and CYP2C19 loss-of-function alleles thus require alternative dual antiplatelet therapy, such as with ticagrelor. This underscores the value of having more data on ticagrelor in Chinese patients with ACS.

In conclusion, this real-world study provides further safety data on ticagrelor use for up to 1 year in Chinese patients with ACS, finding low rates of dyspnea and PLATO-defined bleeding events. The results are consistent with the known safety profile of ticagrelor, with no new safety findings.