Oral anticoagulants in thrombotic antiphospholipid syndrome: Leave the old road for a new trail?

I read with interest the meta-analysis by Pau Cerdà et al. on direct oral anticoagulants compared to vitamin K antagonists in antiphospholipid syndrome (APS) [ ]. Overall, I agree with the Authors’ conclusions that the use of rivaroxaban in APS patients is associated with an increase rate of recurrences, at least in those with an arterial index event or triple positivity for antiphospholipid (aPL) antibodies. However, the statement that rivaroxaban and VKAs are equally effective in preventing venous thromboembolism must be taken with caution. Patients with venous index events may present arterial events thereafter more frequently but not exclusively when they have a high-risk aPL antibody profile. In any patient with a venous index event, risk factors for arterial events should be carefully considered [ ]. A critical issue underlined by the Authors in the meta-analysis refers to the important clinical heterogeneity in the included studies as participants’ characteristics are quite different. The risk of recurrence in APS largely depends on the aPL antibody profile. Triple-positive patients (positive Lupus Anticoagulant (LA), anticardiolipin (aCL) and anti β2-glycoprotein I (aβ2GPI) antibodies, same isotype) are at much higher risk than patients with double or single positivity [ ]. In the RAPS trial [ ], only 28% of the randomized patients were triple positive and therefore it is not surprising that no thromboembolic events occurred during the six months of follow up. In the Ordi-Ros study [ ], the triple positive patients were 61.1% in the rivaroxaban arm and 60% in the Vitamin K Antagonists (VKA) arm. Most of the remaining patients had isolated LA. Actually, LA identifies two distinct groups of patients with different antibody patterns and different association with thromboembolic events [ ]. During a follow up of 10 years, individuals with isolated LA without a previous thromboembolic event (carriers) showed an incidence rate of thromboembolic events of 1.3% person/year [ ]. On the other hand, carriers of triple-positivity during a mean follow-up of 4.5 years had an incidence rate of thromboembolic events of 5.3% person/year [ ]. Thus, around one third of patients in the Ordi-Ros trial may have been at low risk of thromboembolic events. These data underline the concept that studies in this setting should only consider patients or carriers with homogeneous aPL antibody profiles. Indeed, the sensitivity analysis performed in the study by Pau Cerdà et al. [ ] by antibody profiles (triple or non-triple aPL positivity) showed that rivaroxaban was associated with a higher rate of thromboembolic recurrences in triple aPL antibody profiles and not in non-triple aPL antibody profiles. Now, we have two unsolved issues around the use of Direct Oral Anticoagulants (DOACs) in the setting of APS: i. the use of DOACs in patients with an incomplete aPL profile (non-triple positive patients); ii. the diagnosis of LA to assess triple positivity in patients on anticoagulant treatment to avoid the use of DOACs.