Antiplatelet therapy: the clot thickens

In this issue of Blood, Kaiser and colleagues show that platelets participate in venous thrombosis through their activation to a procoagulant phenotype.¹ They demonstrate pharmacologic inhibition of procoagulant platelet formation as a viable potential therapy for the prevention and treatment of venous thromboembolism (VTE) with decreased hemorrhagic risk relative to traditional antiplatelet agents.

Strong platelet stimulation using soluble or mechanical agonists results in the formation of activated platelet subpopulations with distinct phenotypes. Aggregatory platelets have the familiar activated phenotype. Features of this subpopulations include an aggregatory phenotype with activation of integrin α_IIbβ₃, the release of α and dense granules to the platelet surface as most frequently measured by surface expression of CD62P or P-selectin, and filipodial and lamellipodial spreading. The second subpopulation consists of procoagulant platelets. Distinguishing phenotypic characteristics of the procoagulant platelet are the externalization of negatively charged phospholipids to facilitate coagulation complex assembly, often measured by annexin V binding; rounding to a “balloon” with procoagulant platelet spreading or vesiculation; and decreased aggregatory potential associated with secondary inactivation of integrin α_IIbβ₃.² 

Query of the activation status of circulating platelets in venous blood can provide insight into peripheral mechanisms. In patients with suspected deep venous thrombosis or pulmonary embolism, Kaiser et al show that the amount of circulating procoagulant platelets was associated with VTE while other classical markers of platelet activation such as integrin α_IIbβ₃ activation of CD26P surface expression were not associated. Furthermore, mechanically retrieved thrombi from patients contained a substantial population of phosphatidylserine-positive platelets. In 2 murine models of venous stasis and thrombosis, thrombus weight was correlated with the amount of circulating platelet procoagulant platelets, but other canonical markers of platelet activation were not. Using confocal live imaging, balloon-shaped procoagulant platelets were observed as an early feature within the growing thrombus.

The authors tested the impact of inhibition of pathways specifically implicated in the procoagulant transition on venous thrombosis. Mitochondrial permeability transition pore (mPTP) formation is implicated as an initiating event. Cyclophilin D null platelets have impaired mPTP formation and a substantially decreased potential to become procoagulant.³ Resting platelets maintain an asymmetric cytoplasmic membrane with externalized cationic phospholipids and internalized anionic phospholipids including phosphatidylserine and phosphatidylethanolamine. TMEM16F has been implicated as the scramblase responsible for the rapid disruption of this membrane asymmetry and anionic phospholipid externalization to facilitate tenase and prothrombinase complex assembly and procoagulant activity in platelets.⁴ In mice with absence of either platelet cyclophilin D or platelet TMEM16F, both thrombus incidence and thrombus weight were significantly decreased.

To test the viability of a pharmacologic approach, Kaiser and colleagues used the clinically approved carbonic anhydrase inhibitor methazolamide, which blocks procoagulant platelet spreading and phosphatidylserine externalization.⁵ Carbonic anhydrase inhibitors are classically used to facilitate diuresis and are used in conjunction with furosemide in the treatment of decompensated heart failure.⁶ As with the genetic knockouts of cyclophilin D or TMEM16F, methazolamide treatment specifically abrogated platelet transition to a procoagulant phenotype and inhibited the incidence of venous thrombosis and decreased thrombus mass.

Procoagulant platelet formation has been implicated in the pathogenesis of ischemia-reperfusion injury in the central nervous system⁷ and the gut⁸ and in antibody-mediated platelet activation in COVID-19.⁹ This study adds to the growing body of evidence that indicates the clinical potential of inhibition of specific activated platelet phenotypes and functional pathways. Prior to their clinical application, it will be important to better define the impact of inhibition in other models. In a model of arterial thrombosis induced by photochemical injury, the absence of platelet cyclophilin D results in accelerated thrombosis.¹⁰ A better understanding of these in vivo models will provide insight into the mechanisms driving these differences in thrombotic outcome.

Perhaps in the next decade, the clinical armamentarium for treatment of VTE will move beyond the traditional antiplatelet agents like aspirin and clopidogrel, and we will contemplate therapy with antiplatelet aggregatory, antiplatelet release, or antiprocoagulant platelet agents. The plot thickens as we gain new insights into how platelets contribute to VTE.

Conflict-of-interest disclosure: S.M.J. declares no competing financial interests.