Ticagrelor was associated with lower fracture risk than clopidogrel in the dual anti-platelet regimen among patients with acute coronary syndrome treated with percutaneous coronary intervention

Osteoporosis is a global pandemic. It is clinically silent until complications occur, manifesting as fragility fractures [1]. Hip and vertebral fractures are the more devastating types of fragility fractures associated with significant morbidities and mortality [2]. Patients with coronary artery disease (CAD) may be at a higher risk of osteoporosis and fragility fractures, potentially mediated through the common pathophysiological mechanism of atherosclerosis [3, 4]. Furthermore, patients with acute coronary syndrome (ACS) tend to be older with multiple comorbidities, increasing the risk of major osteoporotic fractures (MOF). Impairment in activities of daily living after fractures can lead to significant morbidity, creating substantial functional limitations and reducing patients’ quality of life.

Dual anti-platelet therapy (DAPT), consisting of aspirin and a P2Y12 inhibitor, is the standard of care in patients with ACS following percutaneous coronary intervention (PCI) [5]. Inhibition of platelet aggregation by DAPT is deemed essential in reducing the risk of stent thrombosis and ischemic events such as stroke and peripheral artery disease [6]. Recently, novel agents, namely ticagrelor and prasugrel, have emerged as potent P2Y12 inhibitors with better antithrombotic effects than clopidogrel [7, 8]. Observational studies have demonstrated differences in fracture risks associated with different anti-platelet agents [9]. A cohort study by Jorgensen et al. showed that the risk of fracture was increased by approximately 50% among clopidogrel users compared to non-users [10]. Mechanistically, P2Y12 receptors are expressed not only in platelets but also in osteoblasts. Inhibition of P2Y12 receptors by clopidogrel could lower osteoblastic activity and differentiation, reducing new bone formation and thereby increasing fracture risk [11]. On the other hand, ticagrelor, in addition to P2Y12 inhibition, prevents cellular adenosine uptake by the equilibrative nucleoside transporter (ENT)-1, thereby increasing extracellular adenosine levels [12]. This may promote osteoblastic differentiation and bone formation. There are yet observational studies reporting fracture risks associated with ticagrelor use.

Given the higher fracture risks among patients with CAD [13], it is clinically relevant to evaluate whether clopidogrel or ticagrelor is associated with a better fracture risk profile. Based on the observations in the pre-clinical studies, we hypothesize that ticagrelor use is associated with a lower fracture risk than clopidogrel use. Here, we performed a retrospective cohort study comparing the risk of MOF among Chinese patients who underwent PCI and were subsequently treated with clopidogrel versus ticagrelor in the DAPT regimen.

A total of 21284 patients were considered for inclusion: 1496 (7.0%) were excluded—age younger than 18 years, died before hospital discharge, not prescribed with P2Y12 inhibitor on hospital discharge, prescribed with prasugrel (instead of ticagrelor or clopidogrel) on hospital discharge. Of the remaining 19788 patients analyzed, 1134 (5.7%) were excluded from the complete case analysis due to missing values in any of the variables used in the propensity score model (Fig. 1). A total of 3018 pairs of patients prescribed either ticagrelor or clopidogrel upon hospital discharge were generated by 1:1 propensity score matching. Baseline characteristics of all patients before and after propensity score matching are shown in Supplementary Table 1 and Table 1, respectively. Before matching, the two groups had significant differences in sex, age, diabetes, hypertension, cerebrovascular disease, previous MI, previous heart failure, CKD, anemia, obesity, and baseline medications. After matching, the two groups were well balanced in the baseline characteristics with standardized differences < 0.1 for all variables (Table 1). In the matched cohort, the mean age was 61.4 ± 11.4 years, with a male predominance (only 15.9% were women). The median follow-up was 6.5 years (IQR: 5.0–7.8).

This is the first study in the literature addressing the impact of the choice of P2Y12 inhibitor in the DAPT regimen on the risk of MOF. Among patients undergoing PCI for ACS, ticagrelor use was associated with reduced risks of MOF compared with clopidogrel. From the perspective of bone health, ticagrelor may be the preferred agent in the DAPT regimen.

Ticagrelor, a potent P2Y12 inhibitor, has superior efficacy over clopidogrel in treating CAD [18, 19]. In line with the results of the randomized controlled trials, our group has previously reported the superiority of potent P2Y12 inhibitors (predominantly ticagrelor) over clopidogrel in our current post-PCI cohort with a lower hazard of MI (including but not exclusively reinfarction) after PCI in the potent P2Y12 inhibitor group (adjusted HR 0.71; 95% CI 0.59–0.87; P < 0.001) [14]. Therefore, further evaluation of the fracture risk among patients receiving clopidogrel versus ticagrelor in the post-PCI cohort is essential. Patients with ACS undergoing PCI are at an increased risk of fracture per se. Previous observational studies have demonstrated an increased risk of osteoporosis and vertebral fracture following coronary events [3, 18]. Furthermore, given the role of aspirin in DAPT, proton pump inhibitors (PPI) are often co-prescribed with aspirin to reduce the risk of peptic ulcer disease [20]. PPI is a known risk factor for osteoporosis, increasing the risk of osteoporotic fractures [21, 22]. In our cohort, over 80% of the individuals were put on PPI. Overall, we observed an annualized fracture risk of 0.3–0.6%. Given the male preponderance and the relatively young age of the cohort (mean age of around 60 years), this level of fracture risk required attention.

Focusing on patients with ACS treated with first-ever PCI, our study demonstrated a higher risk of MOF with clopidogrel than ticagrelor, favouring using the latter agent in the post-PCI setting. While the association between clopidogrel use and fracture risk has been reported in prior studies [10], there is a lack of relevant studies on potent P2Y12 inhibitors such as ticagrelor. The nationwide cohort study by Jorgensen et al. showed that treatment with clopidogrel was associated with an increased risk of overall fractures and osteoporotic fractures, especially in patients with treatment duration longer than one year [10]. A dose-dependent relationship was also observed, where individuals on higher doses of clopidogrel had an increased risk of fractures [10]. Even though the latest study by Jorgensen et al. revealed no significant association between clopidogrel treatment and fracture risk in patients with stroke and transient ischemic attack [23], that might be explained by the fact that such patients were immobilized in bed for prolonged periods, which reduced the risk of falls and subsequent fractures [23].

The difference in the fracture risk between ticagrelor and clopidogrel users could be understood from the respective effects of the P2Y12 inhibitors on the bone in pre-clinical studies. Syberg et al. showed that clopidogrel delays osteoblast growth and reduces cell viability [24]. The resulting decrease in mature osteoblasts causes a decline in alkaline phosphatase activity and collagen production, inhibiting new bone formation. Furthermore, clopidogrel induces the precursor cells to preferentially differentiate into the adipocytic lineage instead of the osteoblastic lineage, creating a negative bone balance [24]. While P2Y12 receptors are expressed in both osteoblasts and osteoclasts, the effect of clopidogrel on the reduction of osteoclast activity and bone resorption is smaller in magnitude than that on the reduction of osteoblast differentiation and bone formation [24]. Hence, Syberg et al. postulated that prolonged exposure to clopidogrel could negatively impact bone health. On the other hand, the lower fracture risk observed with ticagrelor use could be ascribed to its ability to enhance bone formation via an adenosine-dependent mechanism. Mediero et al. demonstrated that in vivo, direct or indirect stimulation of adenosine A_2A receptor (A2AR) by specific agonists or by enhancing endogenous adenosine levels enhances new bone formation and regeneration [25]. Meanwhile, ticagrelor regulates osteoblast and osteoclast function by blocking adenosine uptake, increasing extracellular adenosine levels, thereby stimulating A2AR [12]. A2AR activation modulates the expression of semaphorins 4D (Sema4D) and 3A (Sema3A), which are important regulators of bone homeostasis [26]. Specifically, A2AR stimulation reduces the release of Sema4D by osteoclasts, inhibits Sema4D-mediated osteoclast activation and increases the secretion of Sema3A by osteoblasts [26]. This helps promote osteoblast differentiation and inhibit osteoclast activity, consequently increasing new bone formation.

Secondary analyses revealed consistently lower HR of fractures over vertebrae, hip and upper limbs associated with ticagrelor use. However, the HR of fractures over vertebrae and hip did not reach statistical significance, which may be explained by the lower number of events at these two sites. In the study by Jorgensen et al., fracture risk over different sites (any fractures, hip, forearm and spine) was consistently higher in the clopidogrel-treated group compared to those not treated with clopidogrel [10]. This suggests that the differential effects of clopidogrel versus ticagrelor on the bone are probably systemic. Furthermore, subgroup analyses showed that the observed protective effects of ticagrelor over clopidogrel on fracture risk were consistent across various subgroups, denoting individuals with different fracture risk profiles—age, sex, presence of diabetes and CKD at baseline, and prior history of fractures. Our results suggested the favorable effects of ticagrelor over clopidogrel on bone health across patients with different baseline fracture risks.

There are several limitations in the present study. First, although we applied propensity score matching for various baseline characteristics, including the fracture risk profile of the clopidogrel and ticagrelor groups, the issue of unmeasured confounding and bias could not be entirely excluded given the observational nature of this cohort study. Secondly, bone mineral density measurement and bone turnover markers were not available. Nonetheless, bone mineral density was not the typical determinant of eligibility for various P2Y12 inhibitors. In addition, fracture events were the more important outcomes from patients’ perspective. Thirdly, treatment adherence could not be ascertained like all large-scale pharmacoepidemiologic studies using electronic health databases. Fourthly, as vertebral fractures were not systematically assessed by quantitative vertebral morphometry, we could only identify clinical, but not morphometric, vertebral fractures. Vertebral fractures are frequently asymptomatic [27], so morphometric vertebral fractures would go undetected. This could lead to an underestimation of the number of events of vertebral fractures. Finally, this study was performed among Chinese patients. The results may not be generalizable to patients of other ethnicities.

In this cohort of Chinese patients who underwent PCI for ACS, ticagrelor was associated with a lower risk of MOF than clopidogrel use. Our results support the use of ticagrelor in the DAPT regimen from the perspective of bone health.