Lupus Anticoagulants/Phospholipid-Dependent-Antibodies

August 6, 2005
16.00-19.30
Pyrmont room
Sydney, Australia

Chairman: Ph. G. De Groot, The Netherlands
Co-chairs: M. Galli, Italy;; S. Machin, UK; J. V. Pengo (Italy); H. Rand, USA; G. Reber, Switzerland, R. Roubey, USA (bold = present at the meeting)

The number of attendees at this meeting was > 200 at 16.00 h and about 50 at 19.30 h

Dr. Steven Krilis gave the first presentation. He reported on the results of a consensus meeting on new criteria for the definition of APS. This meeting was held in Sydney , Australia during the XIth International Congress on Antiphospholipid Antibodies, November 2004. The members of the workshop included rheumatologists, immunologists, obstetricians, biochemists, hematologists, and neurologists. This multidisciplinary company thoroughly discussed all the evidence available in the literature on possible criteria that define the antiphospholipid syndrome. The literature on possible criteria is far from complete and often contradictory. The agreement finally reached was not only based on evidence from the literature but also on eminence opinion and will be published shortly.

Dr. Flip de Groot gave the second presentation. He reported on a SSC-mediated muticentre study on the predictive value of a b2GPI specific LAC assay for the detection of a risk for thrombosis. Seven laboratories originally promised to participate, five laboratories, Veronique Regnault & Thomas LeCompte INSERM, Nancy, France, Guidido Reber. University Hospital, Geneva, Switzerland, Jacek Musial, School of Medicine, Cracow, Poland, Bas de Laat & Flip de Groot, UMC, Utrecht, the Netherlands Ricardo Forastiero, Universidad Favaloro, Buenos Aires, Argentinia, actually participate.

Barry Woodhams & Patrick van Dreden, Stago, (Paris, France) prepared ready-to use kits and together with the instructions how to perform the assay, the kits were send to the participants. The results were disappointing. Large variations between the results of different laboratories were observed. The final outcome, a small but significant improvement in the correlation of the new test with thrombosis over the original LAC assay was completely based on the results of two laboratories. In a attempt to understand why there were such large differences between the different laboratories, additional experiments were performed, focused on the possible differences in the performance of the assays in the different laboratories. In Northern Europe , blood is collected in 0.109 M citrate, while in Southern Europe blood is collected in 0.129 M citrate. It turns out that the b2GPI-dependent LAC assay is only functional in 0.109 M citrate and not in 0.129 citrate. These differences were not explained by the small differences in Ca2+ concentration but by differences in Zn2+ concentration. Addition of 50 m M ZnCl2 to the patient plasma not only restored the effect of cardiolipin on the b2GPI dependent LAC in 0.129 M citrate anticoagulated patient samples, it also makes the assay more sensitive in blood anticoagulated in 0.109M citrate. To be continued.

Dr Ian Jenning, also on behalf of drs Elaine Gray and Jef Arnout discussed the development of a standard for LAC testing. The goal is to develop two standards for LAC testing, one based on normal plasma spiked with monoclonal antibodies against b2GPI and prothrombin and one based on a mixture of collected patient plasmas. The major progress made the last year was obtaining ethical approvement for collectin plasmas from controls and patients. The ethical approvement will only be valid for the UK . Hopefully the standard will be there next year.

Dr Guido Reber discussed factors that influence the results of an anti-b2GPI antibody ELISA. When anti-b2GPI ELISAs from different commercial sources or different home made ELISAs were compared, large differences in results are found. To better understand the cause of these differences. the influence of coating buffer (pH), antigen source, microplate brand, washing steps, blocking buffer, ionic strength of the dilution buffer, the role of Ca2+ and the secondary antibody used was tested. The most important factors that influences the results of this ELISA were the brand of the microtitre plate, blocking buffer (only for a selected patient samples), ionic strength of the dilution buffer and the presence of Ca2+ in the dilution buffers.. There is at the moment not a strategy to overcome these problems. The only solution that might help at short run is the introduction of a standard to all commercial assays. Dr. Koike ( Saporro , Japan ) has made available human monoclonal antibodies (IgG and IgM)for this purpose. Negotiations with the commercial companies will be continued.

Dr. Eiji Matsuura discussed the presence of b2GPI-oxLDL complexes in plasmas of patients with the antiphospholipid syndrome and the presence of autoantibodies directed against these complexes. Oxidized low-density lipoprotein (oxLDL), not native LDL, binds in vitro to b 2 GPI via specific oxLDL-derived ligands to form oxLDL/ b 2 GPI complexes. Elevated serum levels of oxLDL/ b 2 GPI complexes are frequently found in patients with autoimmune diseases and antiphospholipid syndrome (APS). The presence of these complexes along with oxLDL/ b 2 GPI autoantibodies strongly suggests an important atherogenic role autoimmune-mediated atherosclerosis. IgG and IgM antibodies to oxLDL/ b 2 GPI complexes were measured by ELISA in 45 systemic lupus erythematosus (SLE) without APS, 29 with secondary APS, 53 systemic sclerosis (SSc) and 81 rheumatoid arthritis (RA) patients. Healthy blood donors (n=100) served as controls. Mean ODs of IgG antibodies were: SLE = 0.215 with 40% reacting above the cut-off (mean OD + 3SD of controls), secondary APS = 0.521 with 76% positives, SSc = 0.177 with 20% positives, and RA = 0.185 with 18% positives. SLE, secondary APS and SSc were statistically higher (p<0.001) compared to controls (0.135). Mean ODs of IgM antibodies were: SLE = 0.265 with 20% positives, secondary APS = 0.684 with 41% positives, SSc = 0.279 with 19% positives, and RA = 0.516 with 51% positives. All groups were statistically higher (p<0.007) compared to controls (0.178). IgM results did not correlate with RF activity. Secondary APS had high IgG and IgM antibody levels. However, RA and SSc, to a lesser degree, only had increased IgM antibodies. This suggests, that IgG and IgM anti-oxLDL/ b 2 GPI antibodies may play an atherogenic role in APS. Whether IgM anti-oxLDL/ b 2 GPI antibodies in RA are anti-atherogenic remains to be determined.

IgG anti-oxLDL/ b 2 GPI antibodies were also measured in 93 secondary APS patients: 37 had venous thrombosis, 42 arterial thrombosis and 14 pregnancy morbidity. Mean OD for arterial thrombosis group was 0.802 with 38% of the patients reacting above the cut-off (mean OD + 3SD of controls); 0.544 for venous thrombosis with 19% positives, and 0.137 for pregnancy morbidity with 0% positives. Mean OD for arterial and venous thrombosis groups were not statistically different (p=0.262), but both (arterial and venous) were statistically higher compared to the pregnancy morbidity group (p<0.02). Positive predictive value (PPV) for total thrombosis (arterial plus venous) of IgG anti-oxLDL/ b 2 GPI antibodies was 92% (p=0.018), for arterial thrombosis 89% (p=0.004), for venous thrombosis 77% (p=0.161) and for pregnancy morbidity 0%. Mean antibody level was higher in APS patients with arterial compared to those with venous thrombosis, but this difference was not statistical significance. However, PPV for arterial thrombosis (89%) was statistically stronger than that for venous thrombosis (77%). These results provide additional support for a pathogenic role of oxLDL/ b 2 GPI antibodies in the development of autoimmune atherosclerotic complications.

Dr Monica Galli participated in the WAPS ( W arfarin in the Anti- Phospholipid Syndrome) study. Antiphospholipid antibodies include, among others, anticardiolipin (aCL), anti- b 2-glycoprotein I (a b 2GPI), anti-prothrombin (aPT), anti-annexin V (aAV) and anti-protein S (aPS) antibodies. Since their clinical significance in the antiphospholipid syndrome has not been clearly defined, we assessed 112 patients (24 males and 88 females, aged 23-81, median 42 years) enrolled in the WAPS Study. Ninety-one subjects (81.3%) were lupus anticoagulants (LA)-positive according to the SSC criteria established in 1995. Thirty-two (28.6%) suffered from autoimmune diseases, and 87 (77.7%) from antiphospholipid syndrome. Eighty-one (72.3%) had a history of arterial (n=35) and/or venous thrombosis (n=51), and 17 (19.3%) women had suffered from one or more abortions. During a median follow-up time of 4 years, 15 (13.4%) patients experienced arterial or venous thrombosis. Commercially available ELISAs were used to measure IgG and IgM aCL (Asserachrom APA), a b 2GPI (Asserachrom Anti- b 2GPI) and aPT (Asserachrom Anti-Prothrombin) antibodies. Prototype ELISAs were used for the detection of IgG and IgM aAV and aPS antibodies. All ELISAs were kindly provided by Diagnostica Stago. Values of IgG and IgM aCL, a2GPI and aPT antibodies were expressed in units. Valeus of IgG and IgM aAV and aPS antibodies were expressed in mOD. Values were grouped by tertiles. Odds Ratio and p values were calculated by logistic regression.

The following statistically significant associations were found:

No significant association was observed between tested variables and the M isotype, irrespective of the antibody. No significant association with thrombosis registered during follow-up was found, possibly because of the small number of events.

In conclusion, these data suggest to measure IgG a b 2GPI in patients suspected of suffering from APS and raise the possibility that they may replace aCL in the diagnosis of antiphospholipid syndrome. IgG aPT measurement also seems to properly establish the syndrome. The role of aAV and aPS antibodies and, in general, of the IgM isotype remains to be elucidated.

Dr. Jacob Rand, from Montefiore Medical Center in Bronx, NY, USA provided an update on the current status of annexin A5 resistance testing in APS. Annexin A5 is a potent anticoagulant protein that crystallizes over phospholipids bilayers shielding them from availability for coagulation reactions. Antiphospholipid antibodies create defects in this shield, and thereby reduce its effectiveness as an anticoagulant. This has been translated to devise a clinical assay for detecting resistance to annexin A5 anticoagulant activity. Thus far several small blinded studies with well characterized patients and control groups have demonstrated the presence of annexin A5 resistance in APS. Additional studies with larger numbers of patients from different centers are needed to test the validity of this assay.