Home free? Not after 3!

Jacob H. Rand

In this issue of Blood, Pengo and colleagues report a prospective study in which they found that patients who tested strongly positive for all 3 standard antiphospholipid (aPL) assays (“triple positives”) had a significant likelihood of experiencing a first thromboembolic event and developing the aPL syndrome (APS).1

Unfortunately, patients cannot be diagnosed for APS until after they have already had at least 1 clinical event.2 Positive laboratory tests alone are not sufficient; the consensus-based definition of the syndrome requires the presence of a clinical manifestation such as a thromboembolic event or defined pregnancy complications for the diagnosis to be made.2 The patient may already have suffered an incapacitating or even fatal event such as stroke, massive pulmonary embolism, or multiorgan failure before the diagnosis is identified. Therefore, finding ways to identify patients who are at increased risk before that first event would be a very valuable contribution because it would open the door to developing preventive treatments.

Cumulative incidence of first thromboembolic events in 104 patients who were triple positive for antiphospholipid antibody tests. See Figure 1 in Pengo et al.1

The finding of positive aPL tests in otherwise asymptomatic patients is not a rare occurrence. Typical situations in which this occurs include preoperative coagulation screening where a prolonged activated partial thromboplastin time is attributed to a lupus anticoagulant, testing for sexually transmitted disease that reveals biologic false-positive syphilis serologies, and diagnostic screens for potential autoimmune condition. However, these laboratory abnormalities are generally dismissed as false positives because they are rarely associated with clinical events. Prophylactic anticoagulation is generally not warranted in these settings because the hemorrhagic risks far outweigh any benefits and many physicians have offered these patients empiric treatment with aspirin; however, this has not been proven effective in at least 1 prospective clinical trial.3 While the current article by Pengo et al was not designed to test the efficacy of the drug, this study also did not find any reduction in thromboembolic events for the patients who were taking aspirin.1

The uncertainties about the clinical significance of positive aPL tests is attributable to the very nature of the assays themselves. As opposed to virtually all of the other clinical laboratory tests, the aPL assays were not designed to measure a disease-specific analyte or to report on a disease mechanism. Rather, they were derived through careful incremental and often fortuitous empiric observations that were made by astute clinicians over decades. Consider how confusing and irrational the aPL tests must be to the student who is learning about them for the first time! First, there is the lupus anticoagulant, a test that detects neither systemic lupus erythematosus nor a bleeding tendency. Then there are the anticardiolipin antibody IgG and IgM assays, which were developed from efforts to quantify the biologic false-positive syphilis test and are not intended to measure antibodies against cardiolipin at all, but are meant to detect antibodies that recognize β2-glycoprotein I (β2GPI) and other proteins that are bound to cardiolipin in that assay. Finally, there are the anti-β2GPI IgG and IgM assays that do measure antibodies that are specific for this protein, but whose role in the APS disease process is not yet understood.

It is therefore especially remarkable that Pengo and colleagues were able to show a predictive use for the combination of these nonmechanistic assays.1 In a well-designed, prospective observational study, they found that patients who tested positive for all 3 of the cardinal aPL tests went on to have a significantly increased risk of developing a thromboembolic event, with an estimated rate of ∼ 5% annually and a cumulative incidence of 37% over the course of the study (see figure). Interestingly, they also identified 2 clinical factors that correlated with increased risk: male sex and a prior risk factor for deep vein thrombosis.

Why should triple positivity be associated with an increased risk for thromboembolism? The simplest and most plausible explanation may be the following: Each of the assays, by itself, has a significant false-positive rate, therefore, triple positivity filters out a group of patients with the lowest likelihood of false positivity and the highest likelihood of having thrombogenic aPL antibodies. Along these lines, my collaborators, Wahezi et al, recently reported that children with autoimmune disorders who were triple positive for aPL tests were also more likely to display resistance to annexin A5 anticoagulant activity than those having a single abnormal aPL test (Wahezi D, manuscript accepted).

Perhaps the most important benefits of this paper by Pengo et al is its contribution to the design of effective primary prevention trials. By identifying these significant predictive risk factors, Pengo et al have identified a subject group that will have a sufficiently high event rate that would allow investigators to test measures that might prevent a first, and potentially devastating, thromboembolic event.1 Ideally, if there were a sufficient number of subjects, the candidate treatments for a multiarmed trial might include antithrombotic drugs, hydroxychloroquine, and statin agents. Hydroxychloroquine, a synthetic antimalarial drug with immunosuppressive properties, is a particularly attractive candidate4,5 because it has an established risk profile, having been used for several decades to treat systemic lupus erythematosus and the drug appears to be associated with reduction of thromboembolic events. More recently, our own laboratory has demonstrated in vitro that the drug interferes with the formation of aPL antibody-β2GPI immune complexes on phospholipid bilayers, with aPL antibody binding to cells, and can also restore annexin A5 crystallization on endothelial and placental trophoblast cell membranes.6,7 It is encouraging that only 2 of the 18 patients in this cohort who were treated with hydroxychloroquine had a first thromboembolic event1; however, it must be emphasized that the potential use of this drug will have to be tested in a prospective clinical trial.

The development of precise assays that measure key steps in the aPL disease processes can only come after scientists have elucidated the immunologic and molecular bases for this enigmatic autoimmune thrombophilic disorder. Nevertheless, even before all the definitive answers are known, investigators like Pengo et al will continue to use the current clinical tests to find successful ways to hit home runs for patients, to prevent and treat this disorder.


  • Conflict-of-interest disclosure: The author declares no competing financial interests. ■