Translating proof-of-concept for platelet slip into improved antithrombotic therapeutic regimens

Abstract

Platelets are central to thrombosis. Research at the intersection of biological and physical sciences provides proof-of-concept for shear rate-dependent platelet slip at vascular stenosis and near device surfaces. Platelet slip extends the observed biological “slip-bonds” to the boundary of functional gliding without contact. As a result, there is diminished engagement of the coagulation cascade by platelets at these surfaces. Comprehending platelet slip would more precisely direct antithrombotic regimens for different shear environments, including for percutaneous coronary intervention (PCI). In this brief report we promote translation of the proof-of-concept for platelet slip into improved antithrombotic regimens by: (1) reviewing new supporting basic biological science and clinical research for platelet slip; (2) hypothesizing the principal variables that affect platelet slip; (3) applying the consequent construct model in support of-and in some cases to challenge-relevant contemporary guidelines and their foundations (including for urgent, higher-risk PCI); and (4) suggesting future research pathways (both basic and clinical). Should future research demonstrate, explain and control platelet slip, then a paradigm shift for choosing and recommending antithrombotic regimens based on predicted shear rate should follow. Improved clinical outcomes with decreased complications accompanying this paradigm shift for higher-risk PCI would also result in substantive cost savings.

Keywords:

• Cardiovascular
• platelet adhesiveness
• shear strength
• slip-bond
• thrombosis

The contemporary evolution of prescribed antithrombotic regimens is driven by the evolution of our understanding of the mechanism of thrombosis, with subsequent influence by associated clinical research.¹ Our understanding of the mechanism of thrombosis is based primarily on basic biological science, which increasingly intersects with the physical sciences.² Effective antithrombotic regimens are then established through randomized controlled trials. These regimens form the principal foundation for practice guidelines. However, practices change when new information becomes available that downgrades or dispatches a regimen (e.g., bivalirudin³ and low molecular weight dextran,⁴ respectively, for percutaneous coronary intervention [PCI]).

Platelets are central to thrombosis.⁵ High shear rates activate the multimeric protein von Willebrand factor (vWF) through a conformation change in its A1 domain.⁶ That conformational change promotes binding to the platelet glycoprotein (GP)Ibα component of the GPIb-IX-V receptor, resulting in platelet adhesion (with subsequent activation, secretion, and aggregation). This high-shear-dependent conformational change and accompanying adhesion is an example of the transition of the A1 domain from a “slip-bond” to a “catch-bond.”^7,8 However, further increases in shear rate appear to reverse this transition from catch-bond back to slip-bond.⁹

Biological slip in the bond context differs from fluid dynamic slip. In fluid dynamics, slip at a fluid-solid interface indicates that the fluid has a non-zero velocity at the surface of the solid (Figure 1).⁵ Nevertheless, fluid dynamics provides the opportunity to model biological slip out to the boundary of functional gliding. In parallel, a perfect biological catch-bond at optimal shear would correlate with fluid dynamic no-slip, and a perfect slip-bond would correlate with perfect slip.

Figure 1. Velocity profile of fluid flow through a cylinder.⁵ Lower left panel: no-slip boundary condition. Middle left panel: partial slip. Upper left panel: perfect slip. True slip (upper right panel) indicates the slip layer at the boundary is of molecular dimension, and effective (or apparent) slip (lower right panel) indicates the local velocity varies over a finite, albeit small, mesoscopic distance. Fluid flow is in the z-direction, steady, laminar and incompressible. R = cylinder inside radius; u_R = fluid velocity at cylinder inside surface (“slip velocity”); u_max = maximum flow velocity.

A recently published research article in Platelets described a validated model of platelet slip at vascular stenosis and near PCI device surfaces (Figure 2).⁵ In this model the measured slip of microspheres-albeit potentially effective-provides proof-of-concept for platelet slip, and shows that their slip is shear-dependent.

Figure 2. Representative illustrations from the validated model of platelet slip at vascular stenosis and near PCI device surfaces.⁵ Platelets were modeled using 3.12 μm diameter polystyrene (non-adhesive) microspheres. The model test section was cylindrical, flow was continuous, and the view presented is longitudinal of the horizontal middle plane. The velocities of the microspheres were determined by laser-Doppler velocimetry (blue rings and discs: ○;●). These velocities were compared with COMSOL Multiphysics computer-simulations of plasma velocity profiles (black solid lines). The model had the capacity to insert a central solid cylinder to model a PCI guide wire (a “cylinder-within-a-cylinder”). The was 20–90% by cross sectional area. The Reynolds number of 300 is within the range of coronary flow.²⁵ All velocities were normalized to the mean velocity u₀, calculated from direct volumetric measurement of bulk fluid flow. Computer simulations applied the no-slip boundary condition at each surface, noting wall shear stresses in the unobstructed coronary circulation promote minimal plasma slip.⁵⁶ The blue discs (●) highlight partial or effective slip (i.e.: non-zero velocity) of microspheres immediately adjacent to the surfaces of the stenoses and central cylinder surface. PCI = percutaneous coronary intervention.

Although this model was very fundamental, it nonetheless extends the slip-bond concept to the boundary of shear-induced functional gliding of platelets along a surface. This model reconciles theslip-/catch-bond concept with classic theory for early experimental observations of increasing-then-decreasing platelet adhesion with monotonically increasing shear.⁹ Moreover, diminished platelet adhesion results in diminished engagement of the coagulation cascade by platelets at both vascular stenosis and device surfaces. Thus, the perceived necessity of anticoagulation is demoted. Correspondingly, early observational studies indicate disposal of anticoagulation altogether indeed decreases bleeding events without an increase in ischemic events.^10,11

In this research,⁵ for a mean velocity of blood flow down a normal proximal left anterior descending coronary artery at 30 cm×sec⁻¹, the critical threshold fluid shear rate to promote slip was calculated to be 318.3 sec⁻¹. However, this threshold shear rate does not account for confounding biological phenomena.

Here we present new biological science and clinical research that support the concept and presence of platelet slip. We also hypothesize the principal variables that affect platelet slip and apply the consequent construct model to relevant guidelines, including those for urgent, higher-risk PCI. Finally, we suggest future research pathways.

(1) Although unintentional, the new supporting basic biological science is compelling. Arce, et al., reported that the discontinuous autoinhibitory module flanking the A1 domain of vWF resists tensile force (shear), but cooperatively unfolds above a certain threshold to expose A1.¹² They also reported that, at very high shear rates (10000 sec⁻¹) and compared with whole blood, the antagonist VHH81 partially suppresses and ARC1172 achieves near-complete suppression of vWF-mediated platelet adhesion.

Arce, et al., confirmed for whole blood the anticipated “catch-bond” increase in platelet adhesion with an increase in shear rate from 1000–2500 sec⁻¹ (Figure 3 left panel¹²). However, they then document a significant-and disproportionate¹³-decrease in platelet adhesion with the further increase in shear rate from 2500–10 000 sec⁻¹. Importantly, these shear rates are above the range for shear-induced platelet detachment.¹⁴ Additionally, their parallel-plate flow chamber did not include any ADAMTS13-type protease. Applying our model, the decrease in platelet adhesion indicates a transition from catch-bond to slip-bond toward the boundary of gliding at a critical threshold plasma shear rate between 2500–10000 sec⁻¹. Conveniently, this shear rate range spans the rate identified in early experimental observations (2600 sec⁻¹)⁹. The addition of VHH81 and ARC1172 caused, respectively, shear-induced amplification of and physiologically near perfect platelet slip compared with whole blood alone, indicating the slip-promoting properties of these antagonists.

Figure 3. Comparison of perfused whole blood platelet adhesion to collagen (% Area covered) at shear rates 1000 –10 000 sec⁻¹ with no inhibitors, VHH81 (14 μg/mL) or ARC1172 (1 μM) mixed with recalcified blood before perfusion.¹² Comparison between platelet coverage [adhesion] was analyzed with a two-way ANOVA with mixed effects and Tukey’s multiple comparison correction. p values after multiple comparison corrections are displayed for relevant groups. Data are means±standard deviation with n = number of fields analyzed per condition. Reproduced with permission.

The new clinical research supporting platelet slip is two-fold: First, a retrospective study of antiplatelet therapy only, without anticoagulation, for urgent, higher-risk PCI;¹⁵ and second, a randomized trial comparing antiplatelet therapies without anticoagulation for the treatment of acute ischemic stroke (AIS).¹⁶

The PCI study¹⁵ consisted of 481 consecutive patients receiving an antithrombotic regimen of antiplatelet therapy only (oral aspirin, P2Y₁₂ inhibitor; intravenous GPIIb-IIIa inhibitor), without anticoagulation. This antithrombotic regimen diverges from PCI guidelines which have consistently supported a combination of antiplatelet and anticoagulant therapies.¹⁷ Most patients (88.4%) presented with non-ST elevation acute coronary syndrome (NSTE-ACS). Additional high-risk features were prevalent.¹⁸ For example, a majority (68.0%) had active New York Heart Association class III-IV congestive heart failure (CHFIII-IV). Also, nearly one-half (48.2%) had diabetes mellitus (DM; insulin dependent-23.3%). All procedures were performed via femoral artery access.

The PCI procedural success rate was high (99.2%). Additionally, there was a low rate of 72-hour and 30-day major adverse cardiovascular events (MACE; combined all-cause death/myocardial infarction/urgent target vessel revascularization: 2.6% [death-0%] and 4.4% [death-0.8%], respectively). Finally, adverse bleeding event rates were low (Bleeding Academic Research Consortium [BARC] 5/death¹⁹: 0% and 0.2%, respectively; BARC 3: 1.2% and 2.1%, respectively).

Moreover, the subgroup of biomarker positive NSTE-ACS patients (N = 228) similarly demonstrated a significant proportion of high-risk clinical features including active CHFIII-IV in 70.6% and DM in 43.4% (insulin dependent-19.7%).²⁰ As anticipated, their 72-hour and 30-day MACE rates were somewhat higher: 4.3% [death-0%] and 7.5% [death-1.8%/cardiac-0.5%], respectively. However, bleeding events remained low (BARC 5/death: 0% and 0.4%, respectively; BARC 3: 1.8% and 2.6%, respectively). Also, urgent target vessel revascularization was low (0% and 1.3%, respectively). These results compare favorably with contemporary-but-lower-risk biomarker positive NSTE-ACS PCI patients receiving combined antiplatelet and anticoagulation therapies (at 30 days: MACE 9.6% [death-1.7%/cardiac-1.1%]; “serious bleeding” [comparable to the summation of BARC 3+5]: 4.3%).²¹

Shear-dependent platelet slip can account for these positive results, acknowledging the influence of concurrent antiplatelet therapy (Figure 4 upper panel). The antiplatelet therapy was mandatory in the peri-stenosis high-to-extremely high-shear environment (4,000–40,000 sec⁻¹)²² to limit adhesion, activation and aggregation of platelets by activated plasma vWF. Additionally, the GPIIb-IIIa inhibitor was particularly useful for patients poorly responsive to aspirin and/or P2Y₁₂ inhibitor²³ and to disperse preexisting platelet aggregates.²⁴ Also, the GP inhibitor served as a safety net in eliminating any platelet adhesion to surface vWF and other proteins (e.g., adsorbed Factor-1) for the intermediate-to-high shear rate ranges spanning normal coronary arterial flow (200–800 sec^{−1 22,25}). The Achilles’ Heel to the concurrent antiplatelet therapy is that any ADAMTS13 cleaving of unfolded vWF would similarly promote bleeding events. This acquired von Willebrand syndrome would potentially resolve following PCI-induced restoration of normal shear rates.²⁶

Figure 4. Effect of shear rate on platelets at a PCI device surface (upper panel)⁵ and at a site of endothelial disruption (lower panel). Left-low shear; Right-high shear.

The trial¹⁶ randomized 1177 patients with AIS without occlusion of large or medium-sized vessels to either the intravenous GPIIb-IIIa inhibitor tirofiban or oral low-dose aspirin (but no anticoagulation). The principal finding was that tirofiban resulted in a greater likelihood of a positive outcome compared with aspirin at 90 days (modified Rankin score 0 or 1: tirofiban-29.1% versus aspirin-22.2%; p = .02). Moreover, incidences of intracranial hemorrhage were very low (but higher with tirofiban: 1.0% versus 0%, respectively; p = .03).

AIS in the absence of large or medium-sized vessel occlusion generally involves the deep penetrating arteries.¹⁶ In this high-shear environment,²² despite the disrupted endothelium and exposed sub-endothelial proteins that characterizes atherothrombosis,²⁷ platelet slip explains the lack of engagement of the coagulation cascade (Figure 4 lower panel).

(2) The results of the above new supporting research contribute to hypothesizing three principal variables that affect platelet slip. These three variables are physical and biological:

(1) Local shear rate.

(2) Quantitative area/volume/density, composition and catch-/slip-bond propensity of proteins within plasma, on device and exposed sub-endothelial surfaces.

(3) Adhesiveness of platelets (both intrinsic and drug-induced [e.g., downstream effect of P2Y₁₂ inhibitors on platelet P-selectin, decreasing adhesiveness²⁸]).

These three variables provide a construct model that is simply a more explicit version of Virchow’s triad.²⁹ This triad consists of stasis, vessel damage and hypercoagulability in describing the etiology and risk of deep vein thrombosis (DVT).

(3) This construct model of platelet slip generally supports current guidelines regarding antithrombotic regimens, which are based most securely on randomized trials (and occasionally meta-analyses). For example, the management of DVT and most atrial fibrillation (low-shear) focus on anticoagulation.^30,31 Conversely, management of AIS (high-shear) focuses on antiplatelet therapies.³²

NSTE-ACS and PCI both involve high-shear environments.²² Nonetheless, both sets of those guidelines advocate a combination of antiplatelet and anticoagulation therapies.^33,34 Notably, retrospective studies of pre-PCI NSTE-ACS patients indicate that, in the presence of oral dual antiplatelet therapy (aspirin and P2Y₁₂ inhibitor), a reduced dose¹¹ or even zero anticoagulation³⁵ result in no significant increase in ischemic but a decrease in bleeding events.

Thus, despite >45 years of research,³⁶ the optimal antithrombotic regimen for NSTE-ACS PCI patients remains uncertain. The above report for urgent, higher-risk PCI¹⁵ supports antiplatelet therapy only, without anticoagulation. In the PCI guidelines,³⁴ the level of recommendation for anticoagulation during any PCI is Class I (benefit>>>risk; procedure/treatment should be performed/administered). However, the level of evidence is C (very limited populations evaluated; only consensus opinion of experts, case studies or standards of care). Rather unexpectedly, in those guidelines only a cursory reference is made to the sole randomized trial evaluating antiplatelet therapy only, without anticoagulation, for PCI (qualifying for level of evidence B).

That trial (CIAO³⁷) showed a significant decrease in both ischemic and bleeding events for elective PCI patients receiving oral aspirin and P2Y₁₂ inhibitor (clopidogrel) without versus with unfractionated heparin (p = .034 and 0.048, respectively). In addition to highlighting the efficacy of antiplatelet therapy, these results underscore untoward consequences of heparin. Those consequences include: heparin-based anticoagulants can paradoxically activate platelets and induce other prothrombotic activities;³⁸ increase bleeding in a dose-dependent fashion;³⁹ and cause thrombocytopenia.⁴⁰ These limitations of the heparin family and the results of CIAO contribute to the continued use of unfractionated heparin in PCI as “the use of an FDA-approved drug for a non-FDA-approved indication.”⁴¹

(4) Future research pathways to advance the proof-of-concept for platelet slip should include basic and clinical studies:

(1) Both a contemporary validation of the original model of coronary flow and expanded flow cytometry studies; and

(2) Additional quantitative clinical research.

Unfortunately, to date, there remains no technology that can directly visualize or measure platelet slip in a biological environment. A contemporary model of coronary flow would take advantage of advanced engineering technologies using a true-scale model with physiologic pulsation and platelet-like particles.⁴² Additionally, microsphere imaging could employ newer particle image velocimetry techniques.⁴³

The flow cytometry studies would build upon prior relevant studies^9,12,23,28 with a focus on determining the critical threshold shear rate for different geometries, pulsatile flow patterns and therapeutic antagonists. Additionally, these studies could clarify mechanisms accounting for prior observations of reduced fibrin formation in higher shear environments near device surfaces.⁴⁴ These additional studies would therefore contribute to the knowledge base for slip-bonds confined to coagulation factors.⁴⁵

The quantitative clinical research would logically involve a sequential approach to extend the inference of in-vivo platelet slip and its clinical impact on thrombosis for urgent, higher-risk PCI: (1) a propensity score analysis comparing the urgent PCI patients using antiplatelet therapy only¹⁵ to a matched set of PCI patients who also received anticoagulation; then (2) a registry; and then, finally, (3) a randomized controlled trial. Each successive step would naturally be predicated upon a positive previous step. Also, the registry and randomized trial would include an independent variable of femoral versus radial artery access to facilitate resolving the accompanying controversy.⁴⁶

A positive outcome to the future research may promote parallel research into larger-diameter arterial sheath procedures with known concerning adverse bleeding event rates. For example, the bleeding event rate for transcatheter aortic valve replacement is 31.9%.^47,48 Additionally-similar to PCI^19,49-there is a significant increase in subsequent short- and long-term adverse events including death. Also, a positive outcome may facilitate improving therapies for other cardiovascular structures, devices and conditions (e.g., left atrium after appendage occlusion^50,51; left ventricular assists devices^52,53).

Moreover, a positive outcome and the resulting attenuation of PCI complications would decrease 30-day MACE for biomarker positive NSTE-ACS patients²¹ by at least 9.6–7.5 = 2.1% (net ≥21.9% decrease). Also, BARC 3+5 bleeding would decrease by at least 4.3–3.0 = 1.3% (net ≥30.2% decrease). We estimate the consequent cost-savings for similar PCI patients in the United States during their index hospitalization alone at ≥$200-500 M/year.^15,54,55 The cost-savings and novel expanded knowledge of thrombosis should support both the basic and quantitative clinical research despite not involving a new drug or device. Additionally, artificial intelligence may improve and compliment this research while providing financial relief without sacrificing authenticity.¹ The positive trials should facilitate translating the proof-of-concept of platelet slip into improved antithrombotic regimens.

Thus, a paradigm shift for choosing and recommending antithrombotic regimens based upon the intersection of basic biological and physical sciences (and directly affecting PCI) may be on the horizon.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The authors reported there is no funding associated with the work featured in this article.