Abstract Several oral glycoprotein (GP) IIb/IIIa antagonists, Sibrafiban, Orbofiban and Lotrafiban, have been studied in large phase III trials; each has failed to provide efficacy and has been associated with increased mortality. Roxifiban has pharmacokinetic and pharmacodynamic properties believed to be more favorable than the earlier oral agents. Here, we revisit the controversial hypothesis of platelet activation liabilities of GP IIb/IIIa antagonists. The effects of site occupancy by four fibans (Roxifiban, Sibrafiban, Orbofiban and Lotrafiban) on platelet activation was assessed using P-selectin expression, fibrinogen binding and microaggregate formation. All four fibans inhibited ADP and TRAP-stimulated fibrinogen binding and microaggregate formation in a concentration-dependent manner, whereas P-selectin expression was relatively unaltered. To more vigorously test for activation liabilities, the effects of transition from peak to trough receptor occupancy upon platelet stimulation was analyzed. The high affinity of Roxifiban for resting platelets precluded reduction of site occupancy by dialysis or gel filtration. A method was developed that takes advantage of the rapid equilibrium of Roxifiban between platelets and soluble GPIIb/IIIa. The platelet occupancy is controlled by the ratio of platelet GPIIb/IIIa to soluble GPIIb/IIIa. This method allows in vitro investigation of peak/trough transitions on platelet activation. A decrease in site occupancy from peak to trough of Roxifiban or Sibrafiban did not result in increased activation of platelets. The loss of platelet-bound antagonist upon incubation with purified soluble GPIIb/IIIa returned fibrinogen binding/microaggregate formation to no drug levels. In conclusion, these studies do not provide evidence for an activation liability of GPIIb/IIIa antagonists in vitro.