The role of antiphospholipid antibody pattern in the recurrence of thrombosis in patients with antiphospholipid syndrome

The antiphospholipid syndrome (APS) is a hypercoagulability condition of autoimmune origin clinically characterized by the development of arterial, venous and/or microvascular thrombosis and pregnancy complications (recurrent early miscarriages, fetal deaths after the 10th week of gestation and/or premature births) associated to the presence of antiphospholipid antibodies (aPL). It can occur either as an isolated condition (primary APS), or in the context of an underlying autoimmune disease, most commonly systemic lupus erythematosus (SLE). Three aPL are included in the current classification criteria: anti-cardiolipin IgG/IgM, anti-β2-glycoprotein I (β2GPI) IgG/IgM and lupus anticoagulant (LAC). Although several others have been identified, they are not tested in most laboratories. The standard of treatment is oral anticoagulation for thrombotic APS and prophylactic/therapeutic low molecular weight heparin plus low dose aspirin for obstetric APS. Nonetheless there is a high rate of recurrence of clinical manifestations in patients undergoing standard of care. Known risk factors for thrombotic recurrence are a previous arterial thrombosis, high risk aPL profiles such as triple aPL positivity and LA persistent positivity, association with SLE and presence of genetic and acquired risk factors for thrombosis. Moreover, aPL persistence over time seems to be associated with thrombotic recurrence, but data on the course of aPL tests over time are limited and mostly with a short follow-up. Recently a score system, named global APS score (GAPSS), that includes the three aPL plus anti-phosphatydilserine-prothrombin (aPS-PT) IgG/M, hypertension and dyslipidemia, has been proposed as a risk stratification tool for APS. Its “adjusted” (aGAPSS) does not include aPS-PT. Moreover, a score for measuring damage, named Damage Index for APS (DIAPS), has been devised. The aGAPSS is by its very nature dynamic, since both cardiovascular risk factors and aPL can change over time, and therefore needs to be monitored longitudinally at repeated time points. Furthermore, even though a disease activity index for APS is lacking, the aGAPSS, as a dynamic score that predispose to clinical manifestations, could be used a surrogate of disease activity and predictor of damage. Starting from these premises, the objective of my thesis has been to analyze the pattern of aPL persistence over time and its association with clinical manifestations and the role of longitudinal monitoring of the aGAPSS as a predictor of clinical recurrence and a surrogate of disease activity predicting damage accrual, in a cohort of patients with APS. In the first study we assessed the rate of aPL persistence in a cohort of 200 individuals with a diagnosis of APS. Median follow-up duration was 172.5 months (IQR 120-240). 56% of patients presented persistent aPL positivity, defined as positivity at medium-high titer in at least two thirds of total determinations. Higher aPL titers and multiple aPL positivity at baseline correlated with aPL persistence, while single aCL and LA positivity were associated with a transient aPL profile. Patients with persistent aPL presented a higher rate of recurrence of clinical manifestations (thrombosis and/or pregnancy morbidity) in comparison to patients with transient aPL (43.8% vs 23.9%) with an OR=2.48 (95% CI 1.34-4.58, p=0.003). The rate of thrombotic recurrence, analyzed separately, was also higher in patients with persistent aPL profile (40.2% vs 18.2%) with an OR=3.02 (95% CI 1.57-5.81, p<0.001). We therefore concluded that more than half of patients with APS maintained persistent aPL positivity over more than 14 years of follow-up. Multiple positivity at baseline increased the likelihood of aPL persistence in the follow-up. Persistent aPL increased the odds of thrombotic and thrombotic plus obstetric recurrence. In the second study we aimed to assess if the one-time aGAPSS measured at baseline, presented a different association with recurrence of thrombosis and/or pregnancy morbidity in comparison to the average aGAPSS over time, measured as a mean of up to 6 yearly measurements (in patients without recurrence) or up the event in patients with recurrence, and to the delta of the aGAPSS, computed as the difference between the aGAPSS before clinical recurrence (if present) or at the end of follow-up (in case of no clinical recurrence) and the basal aGAPSS. Only patients with at least three annual determinations of the aGAPSS were included. In more than half of patients under VKA therapy, we tested the percentage of time spent within the therapeutic range (TTR) to verify that there was no difference between patients with and without thrombotic recurrence. In our cohort of 200 APS patients, we found a higher baseline, mean and delta aGAPSS in patients with recurrence than patients without. Conversely, when comparing patients with thrombotic recurrence to patients with only obstetric recurrence, the baseline aGAPSS was not significantly different, while both the mean aGAPSS and the delta were significantly higher in the first group. The mean aGAPSS was higher in patients with arterial thrombotic recurrence than venous recurrence. TTR was similar in patients under VKA who did or did not present thrombotic recurrence. When performing a cox regression analysis to find the best cut-off value for predicting recurrence of clinical manifestations, a mean aGAPSS > 13 had the highest HR for the event. We therefore concluded that periodic, at least annual monitoring of aPL and cardiovascular risk factors gives a more realistic picture to stratify the risk of recurrence in APS patients and predispose preventive strategies. In the third study we tested if the mean aGAPSS over time, calculated in an identic fashion as the previous study, can be used as a predictor of damage accrual measured through DIAPS change during follow-up (calculated as the difference between the DIAPS at the end of follow-up and the basal DIAPS). aGAPSS showed a positive linear correlation with DIAPS change over time in the multivariate analysis. When comparing patients with high vs low damage accrual (setting a cut-off of 1 point of DIAPS increase during follow-up) we found that mean aGAPSS over time was higher in patients with high damage. Moreover, in the multivariate logistic regression analysis aGAPSS was a predictor of high damage accrual. Lastly, when analyzing the association between DIAPS and mortality, we found that baseline DIAPS correlated with increased odds of death during follow-up. We could therefore conclude that mean aGAPSS over time could serve as a predictor of damage accrual in APS patients. The overall conclusions of the present thesis are therefore that: 1. aPL persistence was associated with multiple positivity and higher aPL titers at baseline and correlated with recurrence of clinical manifestations. 2. aGAPSS longitudinal monitoring, assessed as a mean or delta, compared to one-time assessment of the score, better defined the risk for recurrence of clinical manifestations. A persistently high aGAPSS and an aGAPSS increasing over time were predictors of thrombotic recurrence. 3. Mean aGAPSS over time linearly correlated with DIAPS change and was associated with high damage. Considering its dynamic nature, it can be used as a surrogate of disease activity and a predictor of damage accrual in APS patients. 4. In light of our results results, periodic monitoring of aPL and cardiovascular risk factors is highly recommended in APS patients.