Autoimmunity Reviews 11 (2012) 581–584

Contents lists available at SciVerse ScienceDirect

Autoimmunity Reviews

j ourna l homepage: www.e lsev ie r .com/ locate /aut rev

Review

Does seronegative antiphospholipid syndrome really exist?

Ricard Cervera a, Fabrizio Conti b, Andrea Doria c,⁎, Luca Iaccarino c, Guido Valesini b

a Department of Autoimmune Diseases, Hospital Clínic, Barcelona, Catalonia, Spainb Department of Internal Medicine and Medical Speciality, Rheumatology, University “La Sapienza” of Rome, Italyc Division of Rheumatology, Department of Clinical and Experimental Medicine, University of Padova, Italy

⁎ Corresponding author at: Division of RheumatoloGiustiniani, 2, 35128, Padova, Italy. Tel.: +39 049 82122+39 335 1425825 (Mobile); fax: +39 049 8212191.

E-mail address: adoria@unipd.it (A. Doria).

1568-9972/$ – see front matter © 2011 Elsevier B.V. Alldoi:10.1016/j.autrev.2011.10.017

a b s t r a c t

a r t i c l e i n f o

Available online 22 October 2011

Keywords:Seronegative antiphospholipid syndromeAntiphospholipid antibodiesAnti-phospholipid antibodies assays

The diagnosis of seronegative (SN-) antiphospholipid syndrome (APS) has been suggested for patients withclinical manifestations indicative of APS but with persistently negative results in the commonly used assaysto detect anti-cardiolipin (aCL) antibodies, anti-β2 Glycoprotein I antibodies (aβ2GPI), and lupus anticoagu-lant (LA). To date the best management of these patients is still unclear.Newemerging anti-phospholipid (aPL) assays could improve our ability in diagnosing APS. However, the availabil-ity of aPL assays in routine laboratory practice is limited. In fact, even aβ2GPI is routinely tested in only a smallnumber of laboratories, and other aPL, such as anti-prothrombin or anti-annexin antibodies, in only a few researchlaboratories.On the other hand transient or false negative aPL assay and other genetic or acquired pro-thrombotic conditionscan further complicate this issue.This paper is focused on the arguments for and against the diagnosis of SN-APS and is aimed to help the clinicianwhen approaching a patient with clinical manifestations consistent with APS diagnosis but with negative aPLusing the commonly available tests.

© 2011 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5812. Arguments for . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5823. Arguments against . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5824. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 583Take-home messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584

1. Introduction

Antiphospholipid syndrome (APS) is a new autoimmune diseasecharacterized by the association of clinical features such as arterialor venous thrombosis and pregnancy morbidity associated to persis-tent positive antiphospholipid antibodies (aPL) tests [1–3]. Antipho-spholipid antibodies represent a heterogeneous family of antibodiesthat react with serum phospholipid-binding plasma proteins (mainlyβ2GPI, prothrombin, protein C, protein S, annexin V, annexin II, oxi-dized low-density lipoprotein), phospholipid–protein complexes,and anionic phospholipids [4–6]. The aPL which are recommended tobe tested in routine laboratory practice are IgG and IgM anticardiolipin

gy, University of Padova, Via02, +39 338 8072644(Mobile),

rights reserved.

antibodies (aCL), IgG and IgM anti-β2 glycoprotein-I antibodies(aβ2GPI), detected by enzyme linked immunosorbent assay (ELISA),and lupus anticoagulant (LA) detected by clotting assays accordingto the guidelines of the International Society on Thrombosis andHaemostasis.

Classification of APS requires the combination of at least one clinicaland one laboratory criterion. A persistently positive aPL test is requiredfor the diagnosis, in fact aPL tests should be positive in two or moreoccasions at least 12 weeks apart [7].

It is not unusual to find patients with clinical manifestations charac-teristic of APS but with persistently negative aPL tests including aCL,aβ2GPI, and LA in daily clinical practice. A diagnosis of seronegativeAPS (SN-APS) has been suggested for these patients [8,9].

However, the classification of these patients is controversial; in fact,some authors are skeptical about the diagnosis of SN-APS [10–12].

One of the most critical issues is that most laboratories test onlyIgG and IgM aCL and LA, and only a few of them test aβ2GPI

582 R. Cervera et al. / Autoimmunity Reviews 11 (2012) 581–584

(primarily IgG isotype). Furthermore, the currently available assayscan not detect all potential aPL, and the new specificities, recentlyidentified, are tested in only a few research laboratories.

Another important issue is transient aPL seronegativity whichmay be due to different causes, such as nephrotic syndrome, treat-ment with corticosteroids [13], immunosuppressants, or antibodyconsumption during thrombotic event.

The debate on this matter is still open and in this paper the argu-ments for and against the diagnosis of SN-APS are considered.

2. Arguments for

New antigenic targets for aPL in the APS have been recently inves-tigated. In particular, it has been shown that antibodies directed tolyso(bis)phosphatidic acid (aLBPA) may represent a marker of APSshowing similar sensitivity and specificity compared to aβ2GPI [14].In addition, aLBPA are strongly associated with the presence of LA[14,15]. Moreover, antiprothrombin (aPT) has been reported as theonly antibody detected in few patients with systemic lupus erythe-matosus and history of thrombosis but persistently negative for aCLor LA (at least three times, 6 weeks apart) [16,17]. Furthermore,anti-phosphatidylethanolamine antibodies (aPE) were detected in15% of a cohort of 270 patients with thrombosis and mainly foundin the absence of the other laboratory criteria for APS, but the retro-spective design of the study did not allow the evaluation of the per-sistence of aPE positivity [18]. Thus, testing for aPT and aPE may behelpful in some selected cases of so-called SN-APS.

Recently, with the aim of identifying new potential antigenic tar-get(s) of aPL, sera from patients with so-called SN-APS were testedby a proteomic approach, analyzing endothelial cell-surface mem-brane proteins. By this analysis, vimentin was identified as a stronglyimmunoreactive autoantigen [19]. Vimentin is a ubiquitous cyto-skeleton intermediate filament protein. Surface expression of formsof vimentin has been demonstrated in several cell settings, includingapoptotic neutrophils and T-cells; activated macrophages, platelets,and vascular and microvascular endothelial cells. Antivimentin anti-bodies were first described in patients with systemic lupus erythe-matosus (SLE), and they exhibit significant correlation with thepresence of aCL [20]. It has been demonstrated that vimentin isable to bind CL in vitro, possibly as a result of electrostatic interactionbetween its positively charged amino acids and the negativelycharged amino acids of CL. Serum IgG anti-vimentin/cardiolipin anti-bodies detected by ELISAwere found not only in a large proportion ofSN-APS patients (55%), but also in almost all APS patients (92%).Anti-vimentin/cardiolipin antibodies were also detected in 43% ofpatients with SLE, in 17% of patients with rheumatoid arthritis andin no healthy subjects. The analysis of IgM anti-vimentin/cardiolipinantibodies revealed similar results. The test performedwith a secondsample obtained at least 12 weeks from the previous one, confirmedthe same result in all SN-APS patients [19].

Recently, it has been demonstrated that aPL may exert their path-ogenic role by triggering a signal transduction pathway involvingToll-like receptor 4 (TLR-4), IRAK phosphorylation, NF-kB activationwith consequent release of proinflammatory and procoagulantfactors by endothelial and monocytic cells [21–23]. In support ofthe potential pathogenic role of anti-vimentin/cardiolipin autoanti-bodies, it has been recently demonstrated that IgG specific for thevimentin/cardiolipin complex, obtained from sera of SN-APS pa-tients, induced IRAK phosphorylation and NF-kB activation in endo-thelial cells. These results suggest that vimentin may be a “new”

antigenic cofactor for aPL and the detection of these antibodiescould represent a useful tool in those patients with clinical featuressuggestive of APS in which the classical tests for aPL detection resultpersistently negative.

A new diagnostic tool, immunostaining on thin layer chromatogra-phy plates, has been proposed to evaluate aPL reactivity [24]. This

non-quantitative technique identifies the reactivity of serum aPL withpurified phospholipid molecules in the absence of phospholipid-binding proteins, but with a different exposure as compared to ELISAmethods. Nevertheless, this technique is still not suitable for screeningpurposes in SN-APS, and prospective studies need to be carried out toassess its clinical relevance as a reference test in patientswith suspectedAPS but persistently negative for conventional aPL tests.

Deep vein thrombosis,myocardial infarction and stroke are themaincauses of morbidity andmortality among APS patients due to their highrisk of recurrence; thus it is mandatory to identify among patients withsuspected APS and repeatedly negative conventional aPL tests, thosewith a true APS to offer them long-term anticoagulation as widelyrecommended for secondary thromboprophylaxis in this disease.

3. Arguments against

Patients with typical clinical manifestations of APS can present, insome determinations, seronegativity for LA as well as for IgG and IgMaCL, and aβ2GPI. However, several considerationsmust be kept inmind:

1. Weak LA effect may go undetected if the plasma sample is notcompletely platelet-free [25].

2. A small proportion of patients that are negative for IgG and IgMaCL or aβ2GPI may be positive for the IgA isotype [26].

3. Some patients that are seronegative for the “criteria” aPL (aCL andaβ2GPI) may occasionally have antibodies against other mem-brane phospholipids, such as phosphatidylserine, phosphatidicacid and phosphatidylinositol, which are not determined in rou-tine blood tests [27].

4. Most papers support the specificity of aPL for protein-phospholipidcomplexes rather than for phospholipids alone and some studies[28] also suggest that aPL can recognize proteins (i.e., β2GPI) inthe absence of phospholipids. A variety of other proteins havebeen implicated as targets for aPL, including prothrombin, proteinC, protein S and placental anticoagulant protein.

5. A decrease in aPL levels has been observed after the development ofnephrotic syndrome, that has been attributed to urinary loss of IgGaPL, a decrease in their synthesis, and/or increase in their catabolism[29].

6. The aPL titles may markedly decrease during treatment with corti-costeroids (i.e., for the treatment of inflammatory manifestationsof associated systemic lupus erythematosus), although the corre-lation between prednisone treatment and aPL titles has not beendefinitively documented [30].

7. Temporal disappearance of aPL during the course of a thromboticevent, probably due to their consumption, has been described [31]and the results of tests carried out immediately after the throm-botic episode cannot be considered as definitive. Several cases inwhich serial measurements of aPL during follow-up revealed apositive seroconversion have been described and the concept of“transient” SN-APS – instead of simply SN-APS – has been proposed[10].

Taking into consideration these facts, the following work-up shouldbe recommended before labeling a patient as having a SN-APS:

1. Determination of LA using a completely platelet-free plasma sam-ple (ultracentrifugation and plasma filtration are needed).

2. Determination of IgA aCL.3. Determination of aPE, phosphatidic acid and phosphatidylinositol

antibodies.4. Determination of aPT, anti-protein C, protein S and placental anti-

coagulant protein antibodies.5. Rule out the effects of nephrotic syndrome.6. Rule out the effects of steroids.7. Repeat the tests for aPL at some reasonable time point after the

thrombotic event (3–6 months).

583R. Cervera et al. / Autoimmunity Reviews 11 (2012) 581–584

4. Conclusions

When assessing a patient with clinical manifestations consistentwith APS but aPL negative using the commonly available assays, theclinician has to consider the possibility that the patient is affectedwith SN-APS.

It is a critical issue because, as suggested by Favaloro andWong [32], clinicians as well as the laboratory personnel have todecide whether to spend time and money to organize additionalaPL tests or make a diagnosis of SN-APS on the basis of the avail-able aPL assays.

As summarized in Fig. 1, in face of a patient with clinical manifes-tations related to APS we recommend that all aPL assays available inthe local laboratory be performed. If these routine assays are nega-tive a complete evaluation of genetic and acquired pro-thrombotic

Are routine aPL antibtests positive?

APS

Are there any sign and scharacteristic for A

NYES

YES N

Thrombosis not related to aPL

APS: antiphospholipid syndrome; a

Are new aPL atests pos

NO

YES NO

So called seronegative

APS

Are there any genetic or acquired pro-thrombotic conditions?

Fig. 1. Flow chart for the

conditions has to be performed and false aPL negative result, i.e.due to nephrotic syndrome, corticosteroid or immunosuppressantuse, and antibody consumption during a thrombotic event, has tobe ruled out.

In case both these conditions are excluded we suggest the testingof IgA aCL, and IgA aβ2GPI, antibodies against other membrane phos-pholipids or other proteins complexed or not with phospholipidssuch as aPT, anti-aPE, anti-vimentin/cardiolipin, anti-annexin (II andV) antibodies [3,6,17–19,32,33].

It is possible that in the near future, new aPL assays will be addedto the laboratories' repertoire and considered in the classificationcriteria.

If even these new tests are negative and in the presence of highprobability for APS diagnosis, the possibilities of a SN-APS should beconsidered and patients consequently managed [34–36].

ody

ymptoms PS?

O

O

PL: antiphospholipid antibodies

Are there any causes of transient seronegativity?

utoantigen itive?

APS

YES

NO YES

Repeat aPL test

diagnosis of SN-APS.

584 R. Cervera et al. / Autoimmunity Reviews 11 (2012) 581–584

Take-home messages

• Patients with APS related clinical manifestations but negative forroutine aPL assays are not uncommon.

• In these patients the diagnosis of seronegative anti-phospholipidsyndrome may be considered.

• Major causes of negative aPL due to laboratory reasons should becarefully considered and excluded.

• All aPL tests routinely available in the local laboratory should beperformed.

• New emerging aPL assays should be tested in a larger number oflaboratories.

References

[1] Saponjski J, Stojanovich L, Djokovic A, et al. Systemic vascular diseases in the anti-phospholipid syndrome. What is the best diagnostic choice? Autoimmun Rev2011;10(4):235–7.

[2] Erkan D, Espinosa G, Cervera R. Catastrophic antiphospholipid syndrome: updateddiagnostic algorithms. Autoimmun Rev 2010;10(2):74–9.

[3] Biggioggero M, Meroni PL. The geoepidemiology of the antiphospholipid antibodysyndrome. Autoimmun Rev 2010;9(5):A299–304.

[4] Martin E,Winn R, Nugent K. Catastrophic antiphospholipid syndrome in a community-acquiredmethicillin-resistant Staphylococcus aureus infection: a review of pathogenesiswith a case for molecular mimicry. Autoimmun Rev 2011;10(4):181–8.

[5] Alessandri C, Conti F, Pendolino M, et al. New autoantigens in the antiphospholipidsyndrome. Autoimmun Rev 2011;10:609–16.

[6] Iaccarino L, Ghirardello A, Canova M, et al. Anti-annexins autoantibodies: theirrole as biomarkers of autoimmune diseases. Autoimmun Rev 2011;10(9):553–8.

[7] Miyakis S, Lockshin MD, Atsumi T, et al. International consensus statement on anupdate of the classification criteria for definite antiphospholipid syndrome (APS).J Thromb Haemost 2006;4:295–306.

[8] Hughes GRV, Khamashta MA. Seronegative antiphospholipid syndrome. AnnRheum Dis 2003;62:1127.

[9] Alessandri C, Conti F, Conigliaro P, et al. Seronegative autoimmune diseases. AnnN Y Acad Sci 2009;1173:52–9.

[10] Miret C, Cervera R, Reverter JC, et al. Antiphospholipid syndromewithout antiphospho-lipid antibodies at the time of the thrombotic event: transient ‘seronegative’ antipho-spholipid syndrome? Clin Exp Rheumatol 1997;15:541–4.

[11] Criado PR, Rivitti EA, Sotto MN, et al. Livedoid vasculopathy as a coagulation dis-order. Autoimmun Rev 2011;10(6):353–60.

[12] Jawad AS. Seronegative antiphospholipid syndrome. Ann Rheum Dis 2004;63(5):608 author reply 608.

[13] Zen M, Canova M, Campana C, et al. The kaleidoscope of glucorticoid effects onimmune system. Autoimmun Rev 2011;10(6):305–10.

[14] Alessandri C, Bombardieri M, Di Prospero L, et al. Anti-lysobisphosphatidic acidantibodies in patients with antiphospholipid syndrome and systemic lupuserythematosus. Clin Exp Immunol 2005;140:173–80.

[15] Sorice M, Ferro D, Misasi R, et al. Evidence for anticoagulant activity and beta2-GPIaccumulation in late endosomes of endothelial cells induced by anti-LBPA anti-bodies. Thromb Haemost 2002;87:735–41.

[16] Bertolaccini ML, Gomez S, Pareja JF, et al. Antiphospholipid antibody tests:spreading the net. Ann Rheum Dis 2005;64:1639–43.

Patients with systemic lupus erythematosus (SLE) are at higher risk of infogens is usually recommended; otherwise, patient’s response to vaccinaticontroversial point, because of the small number of patients studied, almoRheum 2011;63:2396-406) have studied the efficacy of influenza vaccinaSLE patients, clinically and serologically monitored at baseline and at diffevaccination program. Based on cumulative score of influenza-specific anvaccine responders. Overall humoral response to the vaccine was lower inresponse was associated with European ancestry, hemolytic anemia and cdisease flaring and autoantibody titer increasing after vaccination. Also, seexperienced a disease flare than in patients who did not have a flare folloAnna Ghirardello

Clinical, demographic and serological factors predicting efficacy and saf

[17] Bizzaro N, Ghirardello A, Zampieri S, et al. Anti-prothrombin antibodies predictthrombosis in patients with systemic lupus erythematosus: a 15-year longitudi-nal study. J Thromb Haemost 2007;5(6):1158–64.

[18] Sanmarco M, Gayet S, Alessi MC, et al. Antiphosphatidylethanolamine antibodiesare associated with an increased odds ratio for thrombosis. A multicenter studywith the participation of the European Forum on antiphospholipid antibodies.Thromb Haemost 2007;97:949–54.

[19] Ortona E, Capozzi A, Colasanti T, et al. Vimentin/cardiolipin complex as a new an-tigenic target of the antiphospholipid syndrome. Blood 2010;116:2960–7.

[20] Blaschek MA, Boehme M, Jouquan J, et al. Relation of antivimentin antibodiesto anticardiolipin antibodies in systemic lupus erythematosus. Ann Rheum Dis1988;47:708–16.

[21] Raschi E, Testoni C, Bosisio D, et al. Role of the MyD88 transduction signalingpathway in endothelial activation by antiphospholipid antibodies. Blood 2003;101:3495–500.

[22] Conti F, Sorice M, Circella A, et al. Beta-2-glycoprotein I expression on monocytesis increased in anti-phospholipid antibody syndrome and correlates with tissuefactor expression. Clin Exp Immunol 2003;132:509–16.

[23] Sorice M, Longo A, Capozzi A, et al. Anti-beta2-glycoprotein I antibodies inducemonocyte release of tumor necrosis factor alpha and tissue factor by signal trans-duction pathways involving lipid rafts. Arthritis Rheum 2007;56:2687–97.

[24] Sorice M, Griggi T, Circella A, et al. Detection of antiphospholipid antibodies byimmunostaining on thin layer chromatography plates. J Immunol Methods1994;173:49–54.

[25] Brandt JT, Triplett DA, Alving B, et al. Criteria for the diagnosis of lupus antico-agulants: an update. On behalf of the Subcommittee on Lupus Anticoagulant/Antiphospholipid Antibody of the Scientific and Standardisation Committee of theISTH. Thromb Haemost 1995;74:1185–90.

[26] Gharavi AE, Harris EN, Asherson RA, et al. Anticardiolipin antibodies: isotype dis-tribution and phospholipids specificity. Ann Rheum Dis 1987;46:1–6.

[27] López-Soto A, Cervera R, Font J, et al. Isotype distribution and clinical significanceof antibodies to cardiolipin, phosphatidic acid, phosphatidylinositol and phospha-tidylserine in systemic lupus erythematosus: prospective analysis of a series of 92patients. Clin Exp Rheumatol 1997;15:143–9.

[28] Espinosa G, Cervera R, Font J, et al. Antiphospholipid syndrome: pathogenicmechanisms. Autoimm Rev 2003;2:86–93.

[29] Pérez-Vázquez ME, Cabiedes J, Cabral AR, et al. Decrease in serum antiphospholi-pid antibody levels upon development of nephrotic syndrome in patients withsystemic lupus erythematosus: relationship to urinary loss of IgG and other fac-tors. Am J Med 1992;92:357–62.

[30] Silveira LH, Jara LJ, Espinoza LR. Transient disappearance of serum antiphospho-lipid antibodies can also be due to prednisone therapy. Clin Exp Rheumatol1996;14:217–26.

[31] Drenkard C, Sánchez-Guerrero J, Alarcón-Segovia D. Fall in antiphospholipid anti-body at time of thromboocclusive episodes in systemic lupus erythematosus. JRheumatol 1989;16:614–7.

[32] Favaloro EJ, Wong RCW. Laboratory testing for the antiphospholipid syndrome:making sense of antiphospholipid antibody assays. Clin Chem LabMed 2011;49:447–61.

[33] Doria A, Zen M, Canova M, et al. SLE diagnosis and treatment: when early is early.Autoimmun Rev 2010;10(1):55–60.

[34] Scoble T,Wijetilleka S, KhamashtaMA.Management of refractory anti-phospholipidsyndrome. Autoimmun Rev 2011;10(11):669–73.

[35] Soltész P, Kerekes G, Dér H, et al. Comparative assessment of vascular function inautoimmune rheumatic diseases: considerations of prevention and treatment.Autoimmun Rev 2011;10(7):416–25.

[36] Barabas AZ, Cole CD, Graeff RM, et al. The role of autoimmunologists in investigatingand treating autoimmune disorders. Autoimmun Rev 2011;10(3):166–70.

ections than healthy subjects, and vaccination against common path-on in terms of efficacy or impact on the disease course still remains ast with quiescent or inactive disease. Recently, Crowe et al. (Arthritistion and the impact on the disease course in a number of unselectedrent time points after vaccination, during a 3-year trivalent influenzatibody responses, SLE patients were equally divided into high or lowSLE patients than in healthy controls, mainly in low responders. Pooroncurrent prednisone treatment, as well as with higher probability ofrum interferon-α activity wasmore likely to be higher in patients whowing vaccination.

ety of influenza vaccination in systemic lupus erythematosus