Dyspnoea and antiplatelet agents: revisited

To the Editor: We have read with interest the review by Serebruany et al. (1) in which the authors attempt to explain the aetiology of the adverse event of dyspnoea (a subjective sensation of breathlessness or difficulty breathing) that has been reported by some patients treated with the reversible oral P2Y12 receptor antagonist AZD6140. As the authors noted, 10.5% of patients (35 of 334) treated with AZD6140 90 mg twice daily and 15.8% of patients (51 of 323) treated with AZD6140 180 mg twice daily reported dyspnoea in the DISPERSE2 (Dose-finding Investigative Study to assess the Pharmacodynamic Effects of AZD6140 in atherosclerotic disease) trial, which was a randomised, double-blind, three-group, phase II study in patients with acute coronary syndromes (2). These rates were similar to those observed in the dose-ranging DISPERSE trial conducted in patients with stable atherosclerosis (3). However, the authors neglected to mention that 6.4% of the patients (21 of 327) in the group treated with clopidogrel (300-mg load followed by 75 mg once daily) in the DISPERSE2 trial also reported dyspnoea (2). As previously reported (3, 4), nearly all of these instances of dyspnoea were considered mild or moderate in seriousness, were sporadic, self-limited and rarely led to study discontinuation (< 1% of patients). In DISPERSE2, the number of permanent discontinuations related to dyspnoea in each AZD6140 arm (none in the 90-mg twice daily arm; two in the 180-mg twice daily arm) was the same as or lower than the number in the clopidogrel arm (AstraZeneca, data on file, R&D Mölndal, Sweden). Although we are interested in new hypotheses to explain the aetiology of dyspnoea reported by these patients, we find that the speculations by Serebruany et al. (1) are inconsistent with the available data. In their review, the authors state that AZD6140 is an ATP analogue that is metabolised to adenosine. In fact, AZD6140 (a cyclo-pentyl-triazolopyrimidine) lacks the triphosphate moiety of ATP and is not an ATP analogue, as key changes have been introduced into both the core purine and sugar moieties. Hence, none of its measured metabolites are adenosine, nor do they possess adenosine-like physicochemical properties. While an indirect mechanism related to endogenous adenosine cannot be excluded, currently available information does not support a direct adenosine-like effect of AZD6140 or its metabolites. Furthermore, the authors suggest that because AZD6140 is a reversible antiplatelet agent, it ‘…may have damaged already-activated platelets during the unbinding from the cell surface in patients with vascular disease…’. However, the reversible binding of AZD6140 to cell receptors is similar in nature to the reversible binding of agonists such as ADP (adenosine diphosphate) or PGI2 (prostaglandin I2), and hence dependent on its plasma concentration and dissociation constant (Koff). Neither reversible antagonists nor agonists, whether endogenous or synthetic, are known to ‘damage’ their respective receptors or platelets during their dissociation. The authors (1) attempt to associate dyspnoea with thrombotic thrombocytopenic purpura (TPP), and cite observations of TPP with ticlopidine, clopidogrel, and abciximab. These agents are all irreversible inhibitors of P2Y12 or Gp IIb/IIIa receptors that cause permanent structural changes to these receptors (in contrast with the reversible inhibition by AZD6140); such changes may increase the risk of triggering an immune response, as Serebruany et al. further describe. As noted in a previous letter in the European Heart Journal (4), the clinical evidence to date does not support TPP as the cause of the dyspnoea reported by patients treated with AZD6140. Investigators have not reported thrombocytopenia, fluid retention or any objective signs of heart failure, or bronchospasm to suggest a specific, physiological cause in those patients who reported dyspnoea with AZD6140. To date, we have phase II data from over 1000 patients which, together with the final data from the ongoing phase III study (A study of Platelet Inhibition and Patient Outcomes (PLATO)), will constitute a considerable patient safety database that will provide definitive information on the overall profile of AZD6140. The PLATO study compares the 90-mg twice daily AZD6140 maintenance dose with standard treatment with clopidogrel in 18,000 patients. The size of the study ensures that sufficient data will be obtained to allow physicians to conduct an informed benefit-to-risk analysis. Meanwhile, AstraZeneca is committed to further research on potential mechanisms of dyspnoea associated with AZD6140. In summary, the speculations of Serebruany et al. (1) are partly based upon an erroneous belief that AZD6140 metabolises to adenosine and are not supported by any of the available preclinical or clinical data. Careful safety monitoring will continue in the ongoing primary, large-scale phase III outcomes study, and additional studies will investigate both the aetiology and clinical impact of dyspnoea associated with AZD6140. We look forward to a better understanding of this issue based on the results of these studies.