EDITION Review Standardization Guest Article

Transcription

1 EDITION n Review The History of Hughes Syndrome n Standardization The Problem of Standardization of Anticardiolipin Antibodies Assays n Guest Article Multiple Autoantibodies Associated with Autoimmune Reproductive Failure

2 CONTENTS Editorial The study of antiphospholipid antibodies began when a serological test for syphilis was introduced in In 1941, the antigenic component in the test was found to be a phospholipid, which was subsequently termed cardiolipin. The antiphospholipid syndrome (APS) was described first in 1982 by Graham R. V. Hughes, who was so kind to write an article for this journal on the history and clinical background of APS. In 1990, three research groups independently demonstrated the requirement of β2-glycoprotein I (β2gpi) in the ELISA test system for differentiating APS-associated anticardiolipin antibodies from those occurring in syphilis and other infectious diseases. Today most researchers agree that β2-gpi is the main antigen of cardiolipin antibodies in APS. The crucial epitopes develop when β2gpi interacts with a lipid membrane composed of negatively charged phospholipids or when it is adsorbed on a polyoxygenated polystyrene plate. The ELISA for cardiolipin antibodies is one of the most frequently used test in autoimmune serological testing. Still cardiolipin antibody assays are known to have pitfalls, among other things in standardization, which is elucidated in the article of Angela Tincani, who has worked in the field of antiphospholipid antibody testing since many years. Measurement of cardiolipin antibodies is still highly recommended. This is due to the finding, that anticardiolipin antibodies remain the most sensitive markers while β2gpi antibodies are regarded as more specific markers for APS. This high specificity is always put down to the fact that they are rarely found in infectious diseases. But the more the ELISA for β2gpi antibodies is in use, the more they are detected in infectious diseases, too. Newer studies show, that infectious agents may cause the production of anti-β2gpi. Recently, a mechanism of molecular mimicry in experimental APS was described, demonstrating that bacterial peptides homologous with β2gpi amino acid sequences induce pathogenic anti-β2gpi antibodies along with APS manifestations. Co-author of this article is Yehuda Shoenfeld who is also co-author of the article on autoimmune reproductive failure in this journal. A lot of questions remain to be clarified what is the connection between β2gpi antibodies and infections in detail? What is the real benefit of measuring different antibodies in APS? What is the best marker and what does it really indicate? How can the standardization of cardiolipin and β2gpi antibody testing be improved to an acceptable level? Maybe new markers will be found, having a higher specificity for APS than cardiolipin or β2gpi antibodies. International meetings like the 10 th International Congress on Antiphospholipid Antibodies will help us to clarify open questions step by step. Enjoy reading! 3 Review The History of Hughes Syndrome 6 Standardization The Problem of Standardization of Anticardiolipin Antibodies Assays 7 Clinical Study Clinical Manifestations in Patients with Highly Elevated Anticardiolipin Antibodies 8 The Assay Evaluation of a Renewed Cardiolipin Antibodies Assay 9 Guest Article Multiple Autoantibodies Associated with Autoimmune Reproductive Failure EliA Journal is the Journal of Phadia GmbH Munzinger Straße 7, D Freiburg Editor: Contributors: Layout: Reprint 2008: 1,000 Nina Olschowka G. R.V. Hughes A. Tincani W. Papisch W. Miesbach, I. Scharrer Y. Sherer, S. Tartakover-Matalon, M. Blank, Y. Shoenfeld Melanie Tritschler, Tom Bernhard 2

3 REVIEW The History of Hughes Syndrome Graham R.V. Hughes St Thomas Hospital, The Rayne Institute, Lupus Research Unit, London, Great Britain Nearly 20 years have gone by since our first publications detailing the antiphospholipid syndrome (APS) (1,2). Those years have confirmed the early impression that here was a major, hitherto unrecognised disease, branching into all aspects of medicine, and linking thrombosis and autoimmunity. The editors have asked me for a personal view of the history and development of the discovery. While doing so, let me acknowledge at the outset, the work and input of the many colleagues who have contributed so much to the subject. In 1982, I presented my observations on the syndrome to the British Rheumato logy Society ( Heberden round ) and at the British Dermatology meeting (the Prosser-White oration) in 1983 (2). The following passages are taken from that presentation: Although many of these patients fall under the general heading of lupus, or lupus-like disease, I believe that the group is suffi ciently homogeneous and in some ways (such as the frequently negative ANA serology) suffi ciently different from typical systemic lupus erythematosus (SLE) to warrant separate consideration. The manifesta tions of this syndrome are thrombosis (often multiple) and, frequently, spontaneous abortions (often multiple), neurological disease, thrombocytopenia and livedo reticularis. The livedo reticularis is often most fl orid on the knees. This may or may not be associ ated with mild to moderate Raynaud s phenomenon. These patients blood pressures often fl uctuate, apparently correlating with the severity of the livedo, suggesting a possible reno-vascular aetiology. However, this group of patients rarely has primary renal disease. The cerebral features are prominent and of 3 varieties: headaches often migrainous and intractable. Epilepsy (or abnormal EEG s) often going back to early teenage. Fortunately, severe or diffi cult-to-control epilepsy is infrequent. Some patients have chorea. Cerebro-vascular accidents sometimes transient and seemingly attributable to migraine, are frequently progressive. The patients may develop transient ischa emic attacks or, more signifi cantly, progressively cerebral ischaemia. Two other features of the syndrome are a tendency to multiple spontaneous abortions, and peripheral thrombosis, often with multiple leg and arm vein thrombosis. We have also seen Budd-Chiari syndrome and renal vein thrombosis. We have, of course, tended to group these patients under the diagnostic umbrella of systemic lupus, though an alternative label of primary Sjögren s syndrome covers other patients, and characteristic dry Schirmer s tests and lymphocytic infi ltration of the minor salivary glands have been found in a number (though not all) of this group of patients. Table 1: The Antiphospholipid Syndrome References Stroke 1,2,8 Livedo 1,2 Headache and migraine 2 Myelopathy 2, 10 Miscarriage 1, 2, 9 Guillain-Barré 1, 2, 5 Thrombocytopenia 1, 2, 11 Reno-vascular hypertension 2 Chorea 2 Pulmonary hypertension 2, 6, 7 Deep vein thrombosis 1 Epilepsy 2 Visual field defects 2 Budd-Chiari Syndrome 2, 7 Immunoassays 4 Primary APS 2, 3 To my mind, however, the most striking, and often the most serious feature of the disease, is the tendency to thrombosis, particularly cerebral thrombosis. So prominent has this feature been that we have some patients in their 40 s and 50 s who had been diagnosed as primary cerebrovascular disease or when the labile hypertension has been observed as hypertensive cerebrovascular disease. The fi nding that many of these patients may have high titres of circulating anticardiolipin antibodies leads us to believe that a new line of investigation may be possible in such patients. Over the next 3 years, we published original papers highlighting a number of now widely accepted features including stroke, livedo reticularis, chorea, thrombocytopenia, myelopathy, renal and liver thrombosis (including Budd-Chiari syndrome), pulmonary hyper tension, peripheral arterial disease, recurrent pregnancy loss and myocardial infarction (Table 1). Our other contributions during those 3 years was the introduction of immunoassays for antiphospholipid antibodies (apl), the organisation of the first international APS conferences in 1984 and 1986 in London (and the setting up of serum standardisation groups and wet workshops ) and the highlighting of the fact that the antiphospholipid syndrome (APS) could be distinct from lupus the primary APS. For a distinct syndrome it most certainly is. occurs in ANA-negative LE patients, atypical lupus patients and, as expected, individuals with no lupus at all (3). Background For me, there were two main influences leading to the description of APS. Firstly, my long-term interest in central nervous system (CNS) lupus and the recognition that many features of CNS lupus could not be adequately explained by straightfor- 3

4 REVIEW ward inflammatory processes. Secondly, the unparalleled clinical experience afforded by the large clinics built up in our health care system. I believe that a bonus from seeing these large numbers (there are currently 2500 lupus patients and 800 Hughes syndrome patients on our register at St Thomas Hospital) is that it allows the recognition of clinical subsets. In 1975, I spent a year in Jamaica on loan from the Hammersmith Hospital to set up a rheumatology clinic. One of the research interests of my colleague Wendell Wilson and I that year was Jamaican neuropathy, a transverse myelopathy, now known to be due to HTLV-1 infection. We hypothesised that some of the features ANA positivity and false positive VDRL bore similari ties to the previously described lupoid sclerosis. These thoughts led to the hypothesis that antibodies to for example cardiolipin, might cross-react with neuronal phospholipids such as cephalin and sphingomyelin, and, conceivably result in direct neuronal damage. This theory didn t quite work out. However, over the next 5 years, during the late 1970 s, on wardrounds, clinics and research meetings, it became clear to all of us that we were seeing a syndrome a very distinct syndrome associated with antiphospholipid antibodies (apl) (3). Personnel Table 2: International antiphospholipid conferences 1984 London, U.K London, U.K Kingston, Jamaica 1990 Sirmione, Italy 1992 San Antonio, Texas, U.S.A Louvain, Belgium 1996 New Orleans, U.S.A Sapporo, Japan 2000 Tours, France 2002 Taormina, Italy The early clinical studies included Drs Bernie Calaco, Charles Mackworth-Young, Margaret Byron, Sozos Loizou. Visiting research fellows included Ron Asherson, Mee Ling Boey, John Chan, Helene Englert, Genevieve Derue, Angela Tincani, and later Ricard Cervera and Robert Roubey. In the early 1980 s, two colleagues in particular became central to the defining of the syndrome Drs Nigel Harris and Aziz Gharavi lifelong friends who developed the assays for apl (4) and who have both gone on to become leaders in their field. After our move to St Thomas Hospital in 1985, I was joined by Dr Munther Khamashta who has become not only my close colleague and friend, but an internationally renowned contributor to this subject, in particular to the investi gation and management of recurrent pregnancy loss. International Collaboration The subject has captured the imagination of clinicians and researchers worldwide. The 2 yearly international APS congress (Table 2) now attracts large numbers of contributors and abstracts. In recent years, the number of publications on APS has outstripped that on lupus. There are now 2 major textbooks on the subject, as well as patients guides and websites (www. hughes-syndrome.org). At the 8th international meeting, arranged by Dr Takao Koike in Japan, an outstanding workshop produced a set of preliminary classification criteria. International study groups, led by colleagues including Marie Claire Boffa, Ricard Cervera, Yehuda Shoenfeld, Donato Alarcon-Segovia, Ware Branch, Robert Roubey and Munther Khamashta, amongst others, are setting the pace in clinical, serological and therapeutic aspects of the disease. Hughes Syndrome Two major advances came in the early 1990 s the finding by Outi Vaarala and her colleagues of cross-reactivity of apl with oxidised LDL (12) and the discovery of the importance of co-factor. The recognition that the syndrome was not strictly antiphospholipid but more correctly anti- 2/phospholipid or anti-prothrombin/phospholipid or even anti-cofactor/ phospholipid led a number of colleagues at the 6th international congress in Louvain, Belgium, to suggest the alternative title Hughes syndrome (13). I am very proud to have been honoured in this way. I know some dislike the use of eponyms. Others have cited case reports of thrombosis and abortion in lupus dating back to Osler and even to Heberden. I would strongly defend the eponym the clinical description and large series of patients with livedo, strokes, migraines, thrombosis, myelopathy, miscarriage, thrombopenia, arterial disease and the association with antiphospholipid antibodies (for which we described immunoassays) is, I would argue, a far stronger claim than the single case or handful of case reports so often associated with eponyms. I am grateful to those colleagues and friends who have espoused the title Hughes Syndrome. Clinical Features Brain: For me, APS is still very much a neurological disease. I am still impatient, 20 years on, to find that APS has yet to gain a prominent place in neurology textbooks as one of the most important neurological diseases. It will. Perhaps one reason is that most of our earliest reports were published in the rheumatology literature rather than in the neurology world. Also, much of our work was published in European rather than American journals (our paper on apl and stroke was published in Clinical and Experimental Rheumatology a new rheumatology journal published out of Pisa by my good friend Stefano Bombardieri (8). Although strokes, myelopathy, seizures and headache are now well recognised, I believe that clinically, we are just scratching at the surface. My clinical antennae tell me that the frequency of memory loss, headache, febrile convulsions, atypical encephalitis, multiple sclerosis in apl positive individuals is still hugely underestimated. And undertreated. Other organs: Thrombosis of both external (skin) and internal organs is now known to result in a wide clinical spectrum of features. Some clinical observations are listed here: 1. Frequency of valvular disease (14) a distinguishing feature from other coagulopathies (and the interesting rarity of atrial fibrillation). 2. The importance of livedo reticularis: in my view an almost additional independent risk factor for thrombosis. 4

5 REVIEW 3. The association with accelerated arterial disease, including angina and syndrome X (15,16). 4. Thrombocytopenia, often borderline again a distinguishing feature from other coagulopathies. 5. The finding of renal artery stenosis in up to one quarter of APS patients with hypertension an important clinical observation with therapeutic implications (17). 6. Despite the autoimmune features (frequency of Sjögren s syndrome, thyroid antibodies, C4 deficiency etc.) the rarity of the transition from primary APS to lupus. Pregnancy: The description of APS has clarified one aspect of lupus pregnancy loss is almost exclusively a feature of those lupus patients with apl an extremely important point in counselling and in management. The strong association of apl with recurrent pregnancy loss (recurrent late pregnancy loss is almost uniquely due to APS) has led to one of the most clinically rewarding aspects of APS: apl positive women. Indeed, APS has been described as the major advance in obstetrics in the late 20th century (Anthony Kenney, personal communi cation). The Future At the 9th international conference on APS in 2000, Jean Charles Piette set me the task of making 10 predictions for the future of APS. Here are some guesses. 1. APS will come to be recognised as the commonest autoimmune disease. 2. APS will be recognised as a (treatable) differential diagnosis of some cases of Alzheimer s and of M.S. 3. Recognition of apl as the major risk factor for strokes in the under 45 age group will have enormous therapeutic and health care economic implications. 4. The management of apl positive pregnancies will be honed and improved. 5. The association of apl with infertility will be clarified. 6. APL will come to be recognised as an important player in the development of some cases of accelerated atheroma. 7. The interaction of infective agents, apl and some cases of thrombosis will become clearer. 8. Proper epidemiological studies which assess racial and geographical markers will be defined. 9. The disease is genetically determined. More precise genetic markers will be defined. 10. Improvements in treatment will include: better anticoagulation control (including INR self-testing), newer anticoagulants, better-defined use of I.V.I.G., selective plasmapheresis and perhaps, one day, selective suppression of the clones of B cells producing pro-thrombotic apl. Perhaps some of these predictions will be clarified at Yehuda Shoenfeld s 10th APS meeting in September References 1. HUGHES GRV (1983) Thrombosis, abortion, cerebral disease, and the lupus anticoagulant. Br Med J, 287, HUGHES GRV (1984) The Prosser-White oration 1983: Connective tissue disease and the skin. Clin Exp Dermatol 9, HUGHES GRV (1985) The anticardiolipin syndrome. Clin Exp Rheumatol 3, HARRIS EN, GHARVAI AE, BOEY ML, ET AL. (1983) Anticardiolipin antibodies: detection by radioimmunoassay and association with thrombosis in systemic lupus erythematosus. Lancet 2, HARRIS EN, ENGLERT H, DERUE G, ET AL. (1983) Antiphospholipid antibodies in acute Guillain-Barré syndrome. Lancet 2, ASHERSON RA, MACKWORTH-YOUNG CG, BOEY ML, ET AL. (1983) Pulmonary hypertension in systemic lupus erythematosus. Br Med J Clin Res Ed 287, MACKWORTH-YOUNG CG, GHARAVI AE, BOEY ML, HUGHES GRV (1984) Portal and pulmonary hypertension in a case of systemic lupus erythematosus: possible relationship with a clotting abnormality. Eur J Rheumatol Inflamm 7, HARRIS EN, GHARAVI AE, ASHERSON RA, ET AL. (1984) Cerebral infarction in systemic lupus: Association with anticardiolipin antibodies. Clin Exp Rheumatol 2, DERUE GJ, ENGLERT HJ, HARRIS EN, ET AL. (1985) Fetal loss in systemic lupus: association with anticardiolipin antibodies. J Obstet Gynaecol 5, HARRIS EN, GHARAVI AE, MACKWORTH- YOUNG CG, ET AL. (1985) Lupoid sclerosis: a possible pathogenetic role for antiphospholipid antibodies. Ann Rheum Dis 44, HARRIS EN, GHARAVI AE, HEGDE U, ET AL. (1985) Anticardiolipin antibodies in autoimmune thrombocytopenic purpura. Br J Haematol 59, VAARALA O, ALFTHAN G, JAUHIAINEN M, ET AL. (1993) Crossreaction between antibodies to oxidised low-density lipoprotein and to cardiolipin in systemic lupus erythematosus. Lancet 341, GHARAVI AE, WILSON WA (1996) The syndrome of thrombosis, thrombocytopenia, and recurrent spontaneous abortions associated with antiphospholipid antibodies: Hughes syndrome. Lupus 5, FONT J, CERVERA R (2000) Cardiac manifestations in the antiphospholipid syndrome. In: Hughes Syndrome. Antiphospholipid Syndrome. Khamastha MA (ed) Springer, London 15. NAIR S, KHAMASHTA MA, HUGHES GRV (2002) Syndrome X and Hughes Syndrome. Lupus 11, GEORGE J, SHOENFELD Y (1997) The anti-phospholipid (Hughes) syndrome: a crossroads of autoimmunity and atherosclerosis. Lupus 6, GODFREY T, KHAMASHTA MA, HUGHES GRV (2000) Antiphospholipid syndrome and renal artery stenosis. Q J Med 93,

6 STANDARDIZATION The Problem of Standardization of Anticardiolipin Antibody Assay Angela Tincani Rheumatology, Allergy and Clinical Immunology, Brescia Hospital, Italy Anticardiolipin (acl) assays were first described 19 years ago (1), and about 2 years later they were introduced in clinical diagnostic laboratories (2) where they were able to shed new light on autoimmune diseases (3). Despite this successful career, the acl assay remains a difficult test and up to now the reproduci bility in different laboratories is still problematic. A number of different kits are now available and, despite the great efforts made to standardize the methodology (4, 5), each company tries to improve the test by technical variations. This attitude is certainly understandable, in fact the acl assay has some basic problems like, for example, the nature of the antigen itself. Despite the name of the test, we know that the majority of antibodies detected in patients suffering from the Antiphospholipid Syndrome bind beta 2 glycoprotein I (β2gpi). In this respect, since we are speaking of autoantibodies, we have to assume that they will react better with human β2gpi. In the classical acl ELISA the source of β2gpi is adult or foetal bovine serum, which is recognized because of the high homology existing among the β2gpi molecules derived from different animal species. However this is not an absolute rule and some patients samples can bind only human β2gpi; in addition the amount of β2gpi in bovine serum preparations may vary at least to a certain extent. In figure 1, different reaction patterns of 3 sera with acl antibodies are shown. Despite these considerations, the acl assay should be performed according to some general guidelines in order to give compar able results. For example it is probably not justified to modify the original protocol of the classical acl ELISA introducing human, instead of bovine, β2gpi. In fact, when we mention the acl assay we mean the classical acl ELISA that led to the definition of the syndrome and that was quoted in the Classi fication Criteria (6). On the other hand, if we want to explore the reaction to human β2gpi, which is theoretically the one we are looking for, it is probably worthwhile to perform anti-β2gpi instead of acl assay. However, a recent collaborative work (7) done within the European APL Forum, led us to the conclusion that the acl assay protocol still includes many critical variables besides quality and quantity of β2gpi in the system. The identification of the cut off point is not standardized. Certain improvement could probably be achieved by simply recommen ding each Centre to calculate its own cut off point using the same procedure that should be percentile, according to the original observation that normal samples do not have a normal distribution in this test. In addition, calibrators included in the test are often of different origin and can display different reactivity patterns. In this respect if human or humanized monoclonal antibodies (8) will be available as a standard, they will provide a theoretically stable calibration system. Finally it should be underlined that the acl assay, 19 years after its first publication, remains a semi-quantitative test. Therefore, the results expressed in terms of range (negative, low positive, medium positive and high positive) instead of number of units can provide a great improvement in the con sistence of the answers obtained in different laboratories. This politic should also help in many clinical decisions because the diagnosis and the consequent therapeutical intervention is linked to the detection of a medium/high titre of antibodies (6) while titres considered low from the technical point of view are not clinically relevant. In fact, given the high variability of acl assays, the low positive values can often belong to the false positive category, as was recently observed by the Italian Autoantibodies Standardization Workshop FIRMA (manuscript in prepara tion). References 1. HARRIS EN, GHARAVI AE, BOEY ML ET AL. (1983) Anticardiolipin antibodies: detection by radioimmunoassay and association with thrombosis in systemic lupus erythematosus. Lancet 2, HARRIS EN, GHARAVI AE, PATEL SP, HUGHES GRV (1987) Evaluation of the anticardiolipin antibody test: Report of a standardization workshop held in April Clin Exp Immunol 68, LEVINE JS, BRANCH DW, RAUCH J (2002) The antiphospholipid syndrome. N Engl J Med 346, HARRIS EN (1990) THE SECOND International Anti-Cardiolipin Standardization Workshop / The Kingstone Anti-Phospholipid Antibody Study (KAPS) Group. Am J Clin Pathol 94, HARRIS EN, PIERANGELI S, BIRCH D (1994) Anticardiolipin Wet Workshop Report. Fifth International Symposium on Antiphospholipid antibodies. Am J Clin Pathol 101, WILSON WA, GHARAVI AE, KOIKE T ET AL. (1999) International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome. Arthritis Rheum 42, TINCANI A, ALLEGRI F, SANMARCO M ET AL. (2001) Anticardiolipin antibody assay: a methodological analysis for a better consensus in routine determinations a cooperative project of the European Antiphospholipid Forum. Thromb Haemost 86, Fig 1: acl IgG assay. Results of 3 different samples from patients with APS. At 1:50 dilution (A) the β2gpi in the system is partly derived from patients serum itself. When serum samples are diluted 1:100 (B), the addition of an excess of bovine β2gpi (B) can improve the performance also if these antibodies were mostly directed to human β2gpi (D). 8. ICHIKAWA K, TSUTSUMI A, ATSUMI T ET AL. (1999) A chimeric antibody with the human γ1 constant region as a putative standard for assays to detect IgG β-2-glycoprotein I antibodies. Arthritis Rheum 42,

7 CLINICAL STUDY Clinical Manifestations in Patients with Highly Elevated Anticardiolipin Antibodies Wolfgang Miesbach, Inge Scharrer Medical Department of Internal Medicine, Haemostaseology, University Hospital, Frankfurt Introduction Antiphospholipid (apl) antibodies are generally associated with thrombotic events of the antiphospholipid syndrome (APS), which appears to be the most common acquired thrombophilic defect. The clinical mani festations of APS are very heterogeneous, probably reflecting the heterogenity of apl-antibodies, and include arterial and/or venous thrombotic events and recurrent fetal loss [1]. The laboratory assay of apl antibodies can be performed by either using coagulation assays to detect lupus anticoagulants (LA) or by solid phase ELISA-assays to detect antibodies against proteins or phospholipids, such as cardiolipin or β 2 -glycoprotein-i [2,3]. Anticardiolipin (acl) antibodies are considered to be specific markers of the antiphospholipid syndrome. The acl-elisa was found to be the most frequently performed test in patients with suspected APS. However, it still is unclear whether IgG-and IgMisotypes of acl antibodies are distinct markers of certain clinical manifestations. To study the relationship between clinical presentations of APS and the IgG and IgM isotypes of acl-antibodies, we included only patients with highly elevated titres (over 8 times higher than normal) because the higher the anticardiolipin level the greater the likelihood of manifestation of thrombotic disease [4,5]. Patients Between 1994 and 2002 over 1,000 patients were tested positive with elevated titres of IgG- and IgM-aCL by a β 2 -glycoprotein I-dependent enzyme-linked immunosorbent assay (Pharmacia ELISA). We describe the clinical features of 34 patients with acl-igg-titres >100 GPL (normal range: GPL) and 27 patients with acl-igm > 80 MPL (normal range: MPL). Results The group of 34 patients with highly elevated acl-igg-titres >100 GPL consisted of 18 female patients and 16 male patients (female :male ratio 1.125:1). The median age at the date of the first examination was 45 years. 21 % of the patients were suffering from secondary APS. The group of 27 patients with highly elevated acl-igm-titers > 80 MPL consisted of 15 female patients and 12 male patients (female : male ratio 1.25:1). The median age at the date of the first examination was 49 years and 23 % of the patients were suffering from secondary APS. The main clinical features of these patients are summarized in table 1. The most common clinical manifestations of the patients with high acl-igg-titers > 100 GPL were arterial thrombosis (50 %), e.g. transient ischemic attacks and strokes (26 %), myocardial infarctions (12 %) and peripheral arterial thromboses (12 %). 44 % of the patients were presented with venous thromboses, 15 % with thrombo cyto penia and 12 % with fetal loss. In nearly 18 % of the patients no thromboem bolic disease could be found. In comparison to that result, in over 42 % of the patients with highly elevated acl-igm titres no thromboembolic disease could be found. Arterial thrombosis occurred in 44 % of the patients in this group. It was the most common clinical manifestation, e.g. transient ischemic attack and stroke (33 %). Myocardial infarction (7 %) and peripheral arterial thrombosis (4 %) were less frequent. 23 % of the patients were presented with venous thrombosis and 4 % with thrombocytopenia. Discussion By including only patients with high elevated acl- IgG or acl-igm-titres the prevalence of the clinical features of patients with APS and different apl antibodies can be distinguished more clearly. Both groups showed nearly the same female : male ratio and the proportion of patients with SLE-associated APS. In patients with highly elevated acl-igg and acl-igm-titres arterial thrombosis occurred more frequently than venous thrombosis. Cerebral ischemic attacks were the most frequent arterial events. acl-igg seemed to be the strongest risk factor for thromboembolic disease, especially arterial thrombosis. In contrast, a recently published study of a large cohort of 1,000 patients with definite APS showed that venous thrombosis is the most frequent event in patients with definite APS [6]. The risk of thromboembolic disorder is higher in patients with high acl-igg-titres than in patients with high acl-igm-titres. In over 40 % of the patients with high acl-igm-titres > 80 MPL no thromboembolic disease could be found. Precise identification of apl antibodies may improve the clinical predictability of APS. The manifestation of thrombotic disease, especially the risk of arterial thrombosis is higher, the higher the acltitre is. Further investigations with a large cohort of patients are necessary to demonstrate the characteristics of different groups of patients with APS. References 1. CUADRADO MJ, HUGHES GR (2001) Hughes (antiphospholipid) syndrome. Clinical features. Rheum Dis Clin North Am 27, HARRIS EN, PIERANGELI SS, GHAVARI AE (1998) Diagnosis of the antiphospholipid syndrome: A proposal for use of laboratory tests. Lupus 7 (Suppl), S144-S TSUTSUMI A, ICHIKAWA K, MATSUURA E, KOIKE T (1998) Anti-β 2 -Glycoprotein-I antibodies. Lupus 7 (Suppl 2), S98-S TANNE D, D OLHABERRIAGUE L, TRIVEDI AM, ET AL. (2002) Anticardiolipin antibodies and mortality in patients with ischemic stroke: a prospective follow-up study. Neuroepidemiology 21, Table 1. Clinical manifestations Clinical manifestations No. (%) of patients IgG-aCL>100 GPL (n=34) IgM-aCL>80 MPL (n=27) Arterial thrombosis 18 (53) 12 (44) Venous thrombosis 15 (44) 6 (22) Thrombocytopenia 5 (15) 1 (3,7) (< platelets µl) Fetal loss 4 (12) 0 No thromboembolic event 6 (18) 11 (41) SLE-associated APS 7 (21) 6 (22) 5. ANDREASSI C, ZOLI A, RICCIO A, ET AL. (2001) Anticardiolipin antibodies in patients with primary antiphospholipid syndrome: a correlation between IgG titre and antibody-induced cell dysfunctions in neuronal cell cultures. Clin Rheumatol 20, CERVERA R, PIETTE JC, FONT J, ET AL. (2002) Antiphospholipid Syndrome - Clinical and immunologic manifestations and patterns of disease expression in a cohort of 1,000 patients. Arthritis Rheum 46,

8 THE ASSAY Evaluation of a Renewed Cardiolipin Antibodies Assay Requirements of GMP Conform Production GMP in Product Development (Design Control) Market Need initial testing secure raw material supply in constant quality Wolfgang Papisch Pharmacia Diagnostics, Pharmacia Deutschland GmbH, Feasibility expectes properties It is just 20 years ago that cardiolipin was used for the first time as an antigen for the determination of anti phospho lipid antibodies [1]. Today, the testing for cardiolipin antibodies is one of the most frequently used tests in autoimmune serological testing. A large number of commercially available assays from various manufacturers can be used in the laboratory. Conferences focussing exclusively on phospholipid antibodies are organized regularly and gain increasing interest. Still, cardiolipin antibodies assays are known to have pitfalls in terms of reproducibility, robustness and standardi zation. The assay design of the Varelisa Cardiolipin Abs Assays has been reworked in order to improve the perfor mance of the assay in these respects. The clinical performance is evaluated with a defined set of samples from antiphospho lipid syndrome (APS) patients and disease controls (CMV, EBV, HCV). High sensitivities and specificities were found (Table 1). In this context, it should be pointed out that the diseases in the control group are known to be critical with respect to potential positivity for cardiolipin antibodies. In agreement with other findings [2], the titers found in the majority of these patients were low (<25 U/ml). (Figure 1) Figure 2 Verification Validation Registration / CE-labelling GMP in Production Raw Material set detailed specifications develop production and test methods secure production equipment test / validate product properties data for package insert and registration secure production documentation secure market availability by regulatory compliance qualified suppliers raw material of constant quality n = 141 (64 APS; 77 Controls) Sensitivity in % Specificity in % Table 1: Sensitivity, specifi city, positive predictive value, negative predictive value, efficiency and positive/negative likelihood ratio of Varelisa Cardiolipin IgG, IgM, IgA Antibodies and Screen PPV in % NPV in % Efficiency in % Positive Likelihood Ratio Negative Likelihood Ratio Varelisa Cardiolipin IgG Abs 75,0 75,3 71,6 78,4 75,2 3,04 0,33 Varelisa Cardiolipin IgM Abs 39,1 85,7 69,4 62,9 64,5 2,73 0,71 Varelisa Cardiolipin IgA Abs 17,2 93,5 68,8 57,6 58,9 2,65 0,89 Varelisa Cardiolipin Screen (IgG,M,A): at Cut Off Ratio >1.0 (lower limit of eq. area) 90,6 57,1 63,7 88,0 72,3 2,11 0,16 at Cut Off Ratio >1.2 (upper limit of eq. area) 81,3 80,5 77,6 83,8 80,9 4,17 0,23 Raw Material Testing Production Step 1 In Process Controls Production Step x Batch Release Testing production and testing according to well-defined written procedures all steps are documented (Batch Record) deviations are reported and process improvements are initiated changes (process / product) only under controlled conditions Final Product secure constant quality of the products Figure 3 Figure 1: APS patients and controls in Varelisa Cardiolipin IgG, IgM, IgA Antibodies 8

9 GUEST ARTICLE Technical data like assay variance, serum/ plasma correlation, performance in internal and external quality assess ment schemes and more were evaluated. The consistency of results was found to be highly reliable in the internal quality assessment scheme. The retrospective evaluation of external quality assessment sera (NEQAS) shows a high agreement with the designated targets (details not shown). GMP [3] and quality assurance related aspects are shown schematically and demonstrate the complexity of this field. GMP in Product Development (Figure 2) covers elements essential for the way of an idea to a high quality final product. GMP in Production (Figure 3) documents in a simplified scheme the way from Raw Material to Final Product. The strict use of GMP conform procedures, although being costly and time-consuming, guarantees welldocumented products and con sistency over time. In addition, the application of these guidelines may increase and ease aspects like standardi zation and confidence in the test systems used. This is an aspect which is currently being discussed for cardiolipin Abs testing. It should be pointed out that aspects of standardization in the field of autoantibodies testing are gaining more and more importance and interest. Welldocumented and controlled production of assays will facilitate standardization and comparability of results easier. Literature 1. HARRIS EN, GHARAVI, AE BOEY, ML ET AL.(1983) Anticardiolipin antibodies: detection by radioimmunoassay and association with thrombosis in systemic lupus erythematosus. Lancet 2, CACOUB P, MUSSET L, PIETTE JC ET AL. (1997) Anticardiolipin, Anti-beta2- Glycoprotein I, and Antinucleosome Antibodies in Hepatitis C Virus Infection and Mixed Cryoglobulinemia. J Rheumatol 24, SCRIPTS/CDRH/CFDOCS/CFCFR/ CFRSEARCH.CFM Link to document: 21CFR820 Multiple Autoantibodies Associated with Autoimmune Reproductive Failure Yaniv Sherer, Shelly Tartakover-Matalon, Miri Blank, Yehuda Shoenfeld Department of Medicine B and Center of Autoimmune Diseases, Sheba Medical Center, Tel-Hashomer, and Sackler Faculty of Medicine, Tel-Aviv University, Israel Abstract Some cases of reproductive failure can be attributed to autoimmune factors. Among these factors, several autoantibodies were found in association with such clinical manifestations, mainly in patients having systemic lupus erythematosus or antiphos pholipid syndrome. These autoantibodies include classical anti-phospholipid antibodies such as anti-cardiolipin, antiβ2-glycoprotein-i, anti-phosphatidylserine, and anti-phosphatidyl ethanolamine. There are also some non-classical antiphospholipid antibodies directed to prothrombin, thromboplastin or mitochondrial antibodies of the M5 type, which were also found in patients with reproductive failure. Moreover, animal models as well as some human studies suggest that other autoantibodies, including anti-thyroglobulin, antilaminin-1, anti-corpus luteum, anti-prolactin, anti-poly (ADP-ribose) and lymphocytotoxic antibodies, also play a role in these clinical manifestations. Even though currently there is not enough data to support a firm associ ation between some of these autoantibodies and reproductive failure, future studies are likely to confirm this and to expand the number of autoantibodies screened in these patients. Introduction Reproductive failure complicates many autoimmune diseases such as systemic lupus erythematosus (SLE) and is also a major manifestation of the antiphospholipid syndrome (APS). APS is characterized by various aspects of reproductive failure such as recurrent abortions at various stages, in trauterine growth retardation, pre-eclampsia and, possibly, also infertility [1]. Even though recurrent pregnancy loss and other features of reproductive failure might result from various factors, autoimmune factors provide a major etiology for unexplained recurrent pregnancy loss. Herein we briefly review the association of autoantibodies with reproductive failure with a special emphasis on autoantibodies other than anti-phospholipid antibodies (apl). This association is evident from either (and sometimes both) human studies or murine models. Classical anti-phospholipid antibodies apl are the hallmark of APS, and one of the 2 major clinical manifestations of APS include various aspects of reproductive failure. Thorough and comprehensive discussion of the role of apl in reproduction is beyond the scope of this review. Nonetheless, some remarks are provided regarding the groups of autoantibodies that are known as the classical apl: Anti-cardiolipin Anti-cardiolipin is probably the most classical apl, as it is used for the definition of APS. It is highly associated with recurrent pregnancy loss, but also with other mani festations of APS such as focal CNS involvement [2,3]. Animal models support the pathogenic role of anti-cardiolipin in pregnancy [4]. Infusion of anti-cardiolipin antibodies to pregnant mice resulted in a lower fecundity rate, an increased resorption index of embryos, lower number of embryos per pregnancy, and lower weights of embryos and placentae than in the control mice [5]. Following active immunization with a human pathogenic monoclonal IgM acl (H-3), primary APS developed in BALB/c mice: the mice had high titers of acl with clinical manifestations typical for APS, mainly ob stetric manifestations [6]. Anti-phosphatidylserine The role of anti-phosphatidylserine anti bodies in pregnancy loss is also evident from murine studies. Passive induction of APS has been reported in 2 different studies [7,8]. Moreover, active immunization with phos phatidylserine also led to obstetric clinical manifestations typical of experimental APS: lower fecundity rate, number of embryos per pregnancy, and weights of embryos and placentae, but higher fetal resorption rate than in the control mice [9]. 9

10 GUEST ARTICLE Anti-phosphatidylethanolamine There is evidence in some studies that antiphosphatidylethanolamine is associated with the manifestations of APS such as thrombosis, recurrent fetal loss, neurological features and livedo reticularis [10,11]. Anti-β2-glycoprotein-I β2-glycoprotein-i is probably the autoantigen in APS. Autoantibodies directed to it are strongly associated with thrombosis, recurrent fetal loss, thrombocytopenia, hemolytic ane mia, and heart valve disease [12-14]. Immuni zation with β2-glycoprotein-i also resulted in the induction of the typical manifestations of obstetric APS [15]. Non-classical anti-phospholipid antibodies In addition to the above-mentioned autoantibodies, there are autoantibodies directed to antigens associated with the coagulation system, and some also serve as co-factors for apl. There are some reports that also suggest a link to reproductive failure. Anti-thromboplastin Thromboplastin is composed of a complex of phospholipids, lipoprotein and cholesterol, and it enables activation of coagulation factor VII upon its binding. The presence of these antibodies has been associated with thrombosis, thrombocytopenia, hemolytic anemia and fetal loss in SLE patients [16]. They are also correlated with lupus anticoagulant and anti-cardiolipin antibody presence. Anti-mitochondrial antibodies of M5 type These antibodies are directed towards an unknown antigen located in the inner mem branes of mitochondria (50 kda). They are associated with recurrent fetal loss, hemolytic anemia and thrombocytopenia, and have controversial association with thrombosis [17,18]. They are found in up to 31% of SLE patients, and correlate with the presence of other classical apl. Anti-prothrombin Anti-prothrombin autoantibodies are mainly associated with thrombosis in APS and outside the setting of APS. One example would be IgG anti-prothrombin and anti-β2gpi antibodies measurement at the beginning of a 5-year coronary prevention trial, taken from 106 patients who experienced non-fatal myocardial infarction or cardiac death and 106 subjects without coronary episodes during the follow-up [19]. Anti-prothrombin levels were significantly higher in patients than in controls, and the level of apt in the highest third of distribution predicted a 2.5-fold increase in the risk of cardiac 10 events. The association of these antibodies with pregnancy loss is still controversial, but a recent study emphasizes a significant association between the presence of anti-prothrombin antibodies and pregnancy loss in APS, especially early pregnancy loss (submitted for publication). Other autoantibodies In addition to apl, there are other autoanti bodies reported to be associated with repro ductive failure in general or in the setting of APS / SLE. Anti-thyroglobulin Some reports support an association between anti-thyroglobulin antibodies and pregnancy loss. Even though thyroid dysfunction can explain this association, the increase in miscarriages cannot always be explained by thyroid dysfunction alone as it can be encountered in the presence of normal thyroid function [20,21]. This suggests that the higher rate of miscarriages observed in women with autoimmune thyroid distur bances primarily reflects an autoimmune phenomenon, rather than or in addition to a consequence of an overt thyroid hormone abnormality. Thus, the presence of antibodies to the thyroid could represent a secondary marker of a predisposition for an auto immune disease rather then the actual cause of pregnancy loss. We have recently conducted a study in which active immuni zation of mice with thyroglobulin resulted in the production of anti-thyroglobulin antibodies with an increased rate of fetal resorptions compared with non-immunized mice (submitted for publication). No thyroid pathology could be identified in these mice. Anti-corpus luteum Autoantibodies directed to the corpus luteum glycoprotein were found in 22% of female SLE patients who were under 40 years of age. These antibodies were associated with early stages of ovarian dysfunction [22]. This finding might tentatively link these anti bodies to infertility, but this assumption should be further confirmed. Anti-prolactin Autoantibodies to prolactin have been reported in 5% of SLE patients, but in 41% of SLE patients having also hyperprolactin emia [23]. Their presence is associated with decreased disease activity in lupus, but no clear clinical manifestation. They are associated with increased lymphocyte counts and decreased anti-dna antibody levels. Even though not determined, the association of these antibodies with hyperprolactinemia might signify decreased conception rates in these patients, as hyperprolactinemia can lead to infertility. Lymphocytotoxic antibodies The autoantibodies reacting with lympho cytes are a heterogeneous group of antibodies which target different membranal and intracellular antigens. They are detected in 28-90% of SLE patients, but they also appear in viral diseases, malignancies and rheuma toid arthritis. Many different clinical associations have been described with these autoantibodies including cognitive dys function, lupus nephritis, and also spon taneous abortions [24]. Anti-poly (ADP-ribose) Poly (ADP-ribose) is a branched homopoly mer synthesized from the respiratory coen zyme NAD+ at the site of DNA breakage. Autoantibodies against it have been reported in 42-73% of SLE patients, and in these patients they have been associated with obstetric complications including abortion and premature delivery [25]. Anti-laminin IgG anti-laminin-1 antibodies were found in more than 30% of 177 recurrent aborters, and the levels of these antibodies were higher in recurrent aborters than in healthy pregnant and non-pregnant women. Accordingly, the live birth rate of subsequent pregnancies in IgG anti-laminin-1 positive recurrent aborters was significantly lower than in anti-laminin-1 negative recurrent aborters [26]. Animal models also support such a role for anti-laminin-1. Intravenous injections of rabbit antibodies to laminin induced a high inci dence of abortions, fetal death, retroplacental hematomas and hemorrhages in surviving litters. The antibodies were found in the parietal and visceral yolk sac [27]. We recently conducted a study in which immunization with laminin-1 resulted in the production of anti-laminin-1 antibodies and increased fetal resorption rates (submitted for publication). These findings are not sur prising as laminins critically contribute to cell differentiation, cell shape and movement, maintenance of tissue phenotypes, and promotion of tissue survival, and hence they are most important during development of early pre-implantation embryos, during the implantation process and during the organo genesis in post-implantation embryos. Conclusions Autoimmune factors provide a tool for the differential diagnosis and may be involved in the etiology of reproductive failure. Autoantibodies associated with reproductive failure include mainly apl, but the above-mentioned data suggest that many other antibodies might be at least a marker of such clinical presentation. Unfortunately, regarding most of these non-classical and other autoantibodies, there

11 GUEST ARTICLE is currently not enough data to suggest a firm association between their presence or elevated levels and clinical manifestations of reproductive failure. Development of animal models along with clinical studies would be able to disclose which autoantibodies could indeed serve as such a marker. Nonetheless, it seems that more autoantibodies would be used in the future for screening patients having currently what is termed unexplained reproductive failure. References 1. SHOENFELD Y, SHERER Y, BLANK M (1998) Antiphospholipid syndrome in pregnancy animal models and clinical implications. Scand J Rheumatol 27 (Suppl 107), LOVE PE, SANTORO SA (1990) Antiphospholipid antibodies: anticardiolipin and lupus anticoagulant in systemic lupus erythematosus (SLE) and in non-sle disorders. Ann Intern Med 112, NINOMIYA C, TANIGUCHI O, KATO T, ET AL. (1992) Distribution and clinical significance of lupus anticoagulant and anticardiolipin antibody in 349 patients with systemic lupus erythematosus. Intern Med 31, SHOENFELD Y, BLANK M, SHERER Y (2001) Induction and treatment of the antiphospholipid syndrome lessons from animal models. Eur J Clin Invest 31, BLANK M, COHEN J, TODER V, SHOENFELD Y (1991) Induction of anti-phospholipid syndrome in naive mice with mouse lupus monoclonal and human polyclonal anticardiolipin antibodies. Proc Natl Acad Sci USA 88, BAKIMER R, FISHMAN P, BLANK M, ET AL (1992) Induction of primary antiphospholipid syndrome in mice by immunization with a human monoclonal anticardiolipin antibody (H-3). J Clin Invest 89, VOGT E, LYDEN TW, ROTE NS (1992) Monoclonal antiphosphatidylserine antibody induces intrauterine growth retardation in BALB/c mice. Clin Exp Rheumatol 10, BLANK M, TINCANI A, SHOENFELD Y (1994) Induction of antiphospholipid syndrome in naive mice with purified IgG antiphosphatidylserine antibodies. J Rheumatol 21, YODFAT O, BLANK M, KRAUSE I, SHOENFELD Y (1996) The pathogenic role of antiphosphatidylserine antibodies: active immunization with the antibodies leads to the induction of antiphospholipid syndrome. Clin Immunol Immunopathol 78, KARMOCHKINE M, BERARD M, PIETTE JC, ET AL. (1993) Antiphosphatidylethanolamine antibodies in systemic lupus erythematosus. Lupus 2, BERARD M, CHANTOME R, MARCELLI A, BOFFA MC (1996) Antiphosphatidylethanolamine antibodies as the only antiphospholipid antibodies. I. Association with thrombosis and vascular cutaneous diseases. J Rheumatol 23, CABIEDES J, CABRAL AR, ALARCON- SEGOVIA D (1995) Clinical manifestations of the antiphospholipid syndrome in patients with systemic lupus erythematosus associate more strongly with anti-β 2 - glycoprotein-i than with antiphospholipid antibodies. J Rheumatol 22, DAY HM, THIAGARAJAN P, AHN C, ET AL. (1998) Autoantibodies to β 2 -glycoprotein I in systemic lupus erythematosus and primary antiphospholipid antibody syndrome: clinical correlations in comparison with other antiphospholipid antibody tests. J Rheumatol 25, SANFILIPPO SS, KHAMASHTA MA, ATSUMI T, ET AL. (1998) Antibodies to β 2 -glycoprotein I: a potential marker for clinical features of antiphospholipid antibody syndrome in patients with systemic lupus erythematosus. J Rheumatol 25, BLANK M, FADEN D, TINCANI A, ET AL. (1994) Immunization with anticardiolipin cofactor (beta-2-glycoprotein I) induces experimental antiphospholipid syndrome in naive mice. J Autoimmun 7, FONT J, LOPEZ-SOTO A, CERVERA R, ET AL. (1997) Antibodies to thromboplastin in systemic lupus erythematosus: isotype distribution and clinical significance in a series of 92 patients. Thromb Res 86, TINCANI A, MERONI PL, BRUCATO A, ET AL. (1985) Anti-phospholipid and antimitochondrial type M5 antibodies in systemic lupus erythematosus. Clin Exp Rheumatol 3, LA ROSA L, COVINI G, GALPERIN C, ET AL. (1998) Anti-mitochondrial M5 type antibody represents one of the serological markers for anti-phospholipid syndrome distinct from anti-cardiolipin and anti-β 2 glycoprotein I antibodies. Clin Exp Immunol 112, VAARALA O, PUURUNEN M, MÄNTTÄRI M, ET AL. (1996) Antibodies to prothrombin imply a risk of myocardial infarction in middle-aged men. Thromb Haemost 75, MATALON-TARATKOVER S, BLANK M, ORNOY A, SHOENFELD Y (2001) The association between anti-thyroid antibodies and pregnancy loss. Am J Reprod Immunol 45, DENDRINOS S, PAPASTERIADES C, TARASSI K, ET AL. (2000) Thyroid autoimmunity in patients with recurrent spontaneous miscarriages. Gynecol Endocrinol 14, PASOTO SG, VIANA VST, MENDONCA BB, ET AL. (1999) Anti-corpus luteum antibody: a novel serological marker for ovarian dysfunction in systemic lupus erythematosus? J Rheumatol 26, LEAÑOS A, PASCOE D, FRAGA A, BLANCO- FAVELA F (1998) Anti-prolactin autoantibodies in systemic lupus erythematosus patients with associated hyperprolactinemia. Lupus 7, BRESHNIHAN B, GRIGOR RR, OLIVER M ET AL. (1977) Immunological mechanism for spontaneous abortion in systemic lupus erythematosus. Lancet 2, KANAI Y, ISONISHI S, TERASHIMA Y (1989) Antibody to poly (ADP-ribose) is an indicator of obstetric complications in pregnant patients with systemic lupus erythematosus. Immunol Lett 21, INAGAKI J, MATSUURA E, NOMIZU M, ET AL. (2001) IgG anti-laminin-1 autoantibody and recurrent miscarriages. Am J Reprod Immunol 45, FOIDART JM, YAAR M, FIGUEROA A, ET AL. (1983) Abortion in mice induced by intravenous injections of antibodies to type IV collagen or laminin. Am J Pathol 110,

12 AVAILABLE ISSUES No. 1, 1996 No. 2, 1996 No. 3, 1996 No. 1, 1997 No. 1, 1998 No No. 1, 1999 No. 2, 1999 No. 1, 2000 No. 1, 2001 No. 2, 2001 No. 3, 2001 No. 1, 2002 Special Ed. No.1, 2002 No. 2, 2002 Special Ed. No. 1, 2003 No. 2, 2003 No.2, 2002 No. 3, 2003 No. 1, 2004 No. 2, 2004 Special Ed. No.1, 2004 No. 1, 2005 No. 1, 2006 Printed on recycled paper Special Ed. No.1, 2007 No. 1, 2007 Phadia GmbH, Munzinger Str. 7, D Freiburg Germany, Tel: , Fax: , Head office Sweden Austria Belgium Brazil Denmark Finland France Germany Italy Japan Norway Portugal Spain Distributors Switzerland Taiwan The Netherlands United Kingdom / Ireland Other Countries Order No Freiburg 02/2008