Lupus Anticoagulants/Phospholipid-Dependent-Antibodies

July 18, 2002
13:00 to 17:00
Imperial Room
Boston Park Plaza Hotel

Chairman:  J. Arnout, Belgium
Co-chairs:  M. Galli, Italy; Ph. G. De Groot, The Netherlands; S. Machin, UK;
R. Roubey, USA; P. Sie, France

Number of attendees: 125-175

I. Diagnosis of APS: classification and nomenclature?
 
Drs. Arnout, Machin, Roubey, Harris, de Groot and Pengo reviewed how APS is currently diagnosed in their institutions and countries.  Most of the investigators test for antiphospholipid antibodies in patients with clinical features suggestive of APS.  In most institutions, the diagnosis is made based on the so-called Sapporo criteria (Wilson et al. Arthritis & Rheumatism, 42; 1999: 1309-1311).  From these presentations some general conclusions could be made: A large variety of assays is on the market and the interlaboratory agreement is relatively poor.  We need to understand better the reasons for the disagreement between assays.  There is a lack of good guidelines both for the clinical aspects and the laboratory aspects of the syndrome.  There is also a lack of good reference materials.  
 
Dr. Pengo sent a questionnaire to 54 centers in Italy regarding the diagnosis of APS and obtained responses of 23.  Twelve (12) centers only use the LA and the aCL test for the diagnosis of antiphospholipid antibodies.  Eleven (11) centers use in addition the direct anti-b2GP1 immunoassay.  Dr. Pengo himself has a longstanding experience with a homemade anti-b2GP1 immunoassay.  He reported on the results that he obtained in more than 600 consecutive patients with antiphospholipid antibodies.  According to his experience, patients testing positive for LA, aCL and anti-b2GP1 have the highest risk for thrombosis.  In his proposal for a new classification of the laboratory criteria for APS, he considers this as the most important group (Type I).  He also proposed to do a retrospective multicenter analysis of data to give more power to this observation.
 
Dr. de Groot critically reviewed the so-called Sapporo criteria.  Several meta-analysis studies have now shown that a positive LA test is much more strongly associated with thrombosis than a positive aCL.  In addition, a recent meta-analysis reported by Dr. Galli also showed that a positive anti-b2GP1 is more strongly associated with thrombosis than a positive aCL.  The Sapporo criteria include a positive b2GP1-dependent aCL as a criterion for APS; however, it is unclear how to prove the b2GP1-dependency.  Therefore it would be better to replace the aCL test by a direct anti-b2GP1 test.  The SSC should make efforts via multicenter studies to obtain sufficient data to validate ßthe following proposed classification.   Measured on two occasions, at least 6 weeks apart.
Well defined normal range; IgG, IgM and IgA included
 
II. Antiphospholipid antibodies: standardization issues
 
Dr. Ian Jennings reported on a recent UK NEQAS proficiency testing exercises in which the performance of potential LA reference materials was evaluated. Five (5) plasmas were sent to 231 centres for LA screening.  Three samples comprised normal pooled plasma to which monoclonal antibodies against b2glycoprotein1 (anti-b2GP1), prothrombin (anti-II), and a mixture of both had been added.  One further sample was from a patient with a previously identified strong LA and another from a pool of normal plasma. Although the majority of the centers gave a correct interpretation for all samples, variation between methods and reagents was observed.  At identical antibody concentrations, strength of response was greater with anti-II than with anti-b2GP1 and responsiveness of APTT reagents showed greatest agreement between patient and anti-II samples.  A high proportion of negative interpretations, however, (4/8 users) were reported with one DRVVT kit with the anti-II sample, and this was associated with incomplete correction with the concentrated phospholipid reagent.  Further study of phospholipid content in DRVVT reagents indicated significant correlation (r>0.85, P<0.03) between normalized test/confirm ratios and the phospholipid concentration in confirm reagents for 2 out of 3 spiked samples and the LA positive sample, with higher ratios obtained using reagents with higher phospholipid concentrations.  Monoclonal antibodies will be a useful tool in improving standardization between LA screening methods; it is important, however, to recognize that these artificial plasmas may not always behave in the same way as plasma from all patients with LA.
     
Dr. Philip de Groot reported on a new method to differentiate prothrombin from b2glycoprotein1-dependent lupus anticoagulants. Monoclonal antibodies, affinity purified patient antibodies, and selected patient samples were used to show that in an aPTT based clotting assay (PTT-LA, Diagnostica Stago), the use of cardiolipin-vesicles in the neutralization procedure discriminates between b2GP1- or prothrombin-dependent LA activities. Addition of cardiolipin-vesicles shortened the prolonged clotting time caused by anti-b2-glycoprotein I-antibodies with LA activity, whereas this procedure further prolonged clotting times caused by anti-prothrombin-antibodies with LA activity. In contrast, addition of PS/PC-vesicles, corrected prolonged clotting times caused by both anti-b2GP1 and anti-prothrombin antibodies with LA activity. The effects of CL on b2GP1-induced LA activity were specific for contact activation mediated clotting assays. Possible explanations for these findings are the relatively high affinity of b2-GP1 for cardiolipin, as determined by surface plasmon resonance analysis, and inhibition by anti-b2GP1 antibodies of the CL-induced prolongation of the PTT-LA.
 
Dr. Jef Arnout reviewed the current practices of reporting the LA test results.  Current ways of expressing results include the Rosner index, the delta value, a percent value, the Rosove index and several forms of ratios.  The relatively complex criteria for LA diagnosis cannot, however, be translated into one simple figure. These so-called quantitative measures lack inter-laboratory and inter-method standardization and can at most be useful to help differentiating LA from other clotting defects, thereby increasing the probability of a positive LA test.  No precise recommendations on these are available in the current guidelines.  Dr. Arnout proposed that the SSC would recommend qualitative estimations of the results using a probability scale rather than a potency scale.  He also proposed a large multicenter study involving the SSC, ECAT, UK-NEQAS and CAPS to validate the possibility of using calibrator plasmas consisting of normal plasma spiked with LA-positive Mabs at different potency levels in order to quantify the LA potency in clinical samples.
 
Drs. Nigel Harris and Silvia Pierangeli reviewed the status of standardization of the anticardiolipin assay.  Many efforts to standardize the aCL ELISA have been performed. Despite this, the interlaboratory agreement is still poor, mainly due to laboratories using procedures not conforming to proposed guidelines.  The European forum on antiphospholipid antibodies has recently published guidelines for the aCL test in Thrombosis and Haemostasis .  Although homemade bench methods can provide reliable results if the test is performed according to accepted guidelines, Dr. Pierangeli made a plead for the use of well-validated commercial kits. In selecting an aCL ELISA kit, she suggested selection of a kit that utilizes a calibration curve and reports antibody levels in GPL, MPL or APL units.   Utilization of a separate positive control having a defined value and error range are also helpful.  The use of kits in which there is no calibration curve should be discouraged since other means of determining antibody levels, such as that based on the ratio of the O.D. of an unknown sample to the level, may not be reliable.  Dr. Pierangeli also made a plead for the use of assays that are more specific for the syndrome than the original aCL assay. One of the major drawbacks of the original aCL ELISA test has been false positive results.  Binding of sera from patients with a variety of diseases, other than APS is frequent to cardiolipin coated plates.  Recently, new assays that either utilize phosphatidylserine, a mixture of negatively charged phospholipids (APhL® ELISA Kit) or ß2glycoprotein1 have been proposed for more specific measurements of antibodies in APS.
 
Dr. Peter Schur reviewed the methodological aspects of the antiprothrombin immunoassay. From his review of the literature it is clear that the area is fraught with lack of standardization in that different investigator groups used different ELISA plates, different prothrombin preparations and concentrations, different buffers, different blockers, etc. In any case most authors agree that these antibodies are found in variable frequencies in patients with SLE and APS, but more frequently in those with both.  The associations were strongest with thromboembolic events rather than miscarriages, however, Dr. Schur presented his own experience using two different buffer systems, with and without calcium. The frequency of positivity in his experience is much less than reported by others; and calcium indeed does seem to make a difference with some specimens.
 
Dr. Ian Mackie reported that a manuscript is being drafted on the multicenter study to standardize the aPT ELISA.  This manuscript will be forwarded via e-mail to the particpating laboratories and will then be submitted to the Journal of Thrombosis and Haemostasis.  

Dr. Roubey presented a new initiative to register patients with APS: the Antiphospholipid Syndrome Collaborative Registry (APSCORE).  The goal of this NIH funded registry is to enroll 1500 patients with APS and 500 individuals with antiphospholipid antibodies but without clinical manifestations of the syndrome.  APSCORE will utilize a web-based data entry/data management system.  Serum, plasma, and genomic DNA will be stored at UNC.  Enrollment started in April of 2002 and yielded approximately 50 patients so far. Investigators wishing to obtain additional information can take contact with Robert A. S. Roubey, M.D.  Principal Investigator, APSCORE via the following E-mail address:  apscore@med.unc.edu

III.  Concluding remarks.
From several questions raised from the participants it became clear that people expect some more precise guidelines on how to diagnose APS and which assays are needed.  There is also need for an update of the SSC guidelines for LA testing.  A position paper will be drafted and circulated to the active members of the Subcommittee via e-mail.

A number of active members also agreed to exchange samples and patient data in order to obtain sufficient data needed to propose a new classification of APS.

Those who want to join this effort can take contact with Dr. de Groot Ph.G.deGroot@lab.azu.nl , Dr. Pengo pengo@ux1.unipd.it , or the chairman, Dr. Arnout jef.arnout@med.kuleuven.ac.be .