Introduction
The report on the autoreactive potential of the broadly neutralizing recombinant human HIV-1 antibodies 4E10, 2F5 and Igh1b12 by Haynes et al. [1] provided important information for HIV-1 vaccine development and passive immunotherapy with these antibodies. The human monoclonal antibodies 2G12, IgG1b12, 2F5 and 4E10 belong to the very few broadly neutralizing HIV antibodies identified to date [2-6]. 2F5 and 4E10 bind to adjacent linear epitopes on the gp41 ectodomain; 2G12 binds to a mannose-dependent conformational epitope on gp120, and Igh1b12 binds to the gp120 CD4 binding site [2-6]. The induction of antibodies with cross-clade neutralization capacity and binding specificities of 2G12, IgG1b12, 2F5 and 4E10 is a primary goal for vaccine development [7,8]. Passive immunization studies in rhesus macaques have indicated the high potential to prevent sexual, intravenous and mother-to-child HIV transmission [9-13].
Recently, 2F5, 4E10 and IgG1b12 were reported to be autoreactive antibodies [1]. 2F5 was found to bind to cardiolipin, a negatively charged phospholipid mainly found in mitochondrial membranes, to histones and to centromere B autoantigen. 4E10 reacted with cardiolipin and the systemic lupus erythematosus autoantigen SS-A/Ro, phosphatidylserine, phosphatidylcholine, phosphatidylethanolamine, sphingomyelin and prothrombin. While IgG1b12 did not bind cardiolipin, it reacted with ribonucleoprotein, double-stranded DNA, centromere B autoantigen and histones. 2G12 was negative for any of these autoreactivites and, therefore, not subject to safety concerns. The implications of these findings, particularly the reactivity of 2F5 and 4E10 to cardiolipin, questioned the further testing of these monoclonal antibodies in humans, as persistent high titers of anti-phospholipid antibodies, especially anti-cardiolipin antibodies, have been associated with increased risk of thrombogenic conditions. Increased prevalence of antibodies to cardiolipin and lupus anticoagulant, an inhibitor of coagulation, have been described in autoimmune diseases, especially in patients with recurrent thrombosis, spontaneous abortions or thrombocytopenia, and in viral infections, including HIV-1 [14-17]. However, anti-cardiolipins found in viral infections were reported as usually being transient and rarely being pathogenic, in contrast to 'classical' anti-cardiolipin found in autoimmune diseases such as systemic lupus erythematosus [5,15,18]. Consistently, the presence of anti-cardiolipin in HIV-1-infected patients was not associated with detectable lupus anticoagulant activity and did not increase the risk for thrombosis or thrombocytopenia compared with HIV-1-infected patients harboring no anti-cardiolipin [14,15,18]. In contrast, other determinants such as increased procoagulant factors, microparticles and decreased anticoagulant factors have been implicated as mechanisms for thrombotic diseases in HIV-1-infected patients [19].
To date, only 2F5, 2G12 and 4E10 have undergone clinical testing in humans following positive findings during acute infection studies in macaques, safety during repeat-dose toxicity studies and lack of tissue cross-reactivity.
In the course of four clinical phase I and phase II trials to evaluate the safety and efficacy of passive immunotherapy, 39 HIV-1-infected human subjects received a total of 418 intravenous antibody infusions [20-24]. Subjects received between 4 and 16 infusions of each monoclonal antibody in weekly intervals in doses ranging from 1 to 5 g antibody per infusion. Over the entire study period, subjects received between 8.5 and 80 g antibodies. Antibody infusions have been very well tolerated in all clinical trials and no severe treatment-emergent adverse events have been noted.
The recent findings of Haynes et al. [1] prompted further studies of the potential risks for patients to experience thrombogenic complications when receiving 2F5 and 4E10 infusions. To investigate the binding properties described by Haynes et al. under more physiological conditions, antibody spiking experiments were conducted with pooled adult and neonate plasma, since the safety of the antibodies in neonates is critical for further clinical investigation evaluating the prevention of vertical HIV-1 transmission. Patient plasma samples obtained in three clinical trials were analyzed retrospectively for anti-cardiolipin and anti-phosphatidylserine titers using clinically validated assays. Coagulation profiles were analyzed during the most recent clinical study. Investigations were performed pre- and postinfusion of subjects treated with high doses of the monoclonal antibodies 2F5, 4E10, and 2G12. In addition safety data of the four clinical studies conducted with the antibodies have been reviewed with respect to symptoms that have been associated with the presence of anti-phospholipid antibodies.
Material and methods
Antibodies
The generation and characterization of 4E10, 2F5 and 2G12 have been described previously [5,25]. All three antibodies were produced by recombinant expression in CHO cells as IgG1(κ) [26]. 4E10, 2F5 and 2G12 contain identical constant regions and differ only in their variable parts, which were derived from the original hybridoma clones [25]. Clinical grade material was produced by Polymun Scientific GmbH, Vienna, Austria.
Plasma concentrations of the antibodies 4E10, 2F5 and 2G12 were determined by antibody specific enzyme-linked immunosorbent assays (ELISA) as described previously [20-22]. Briefly, plates were coated with epitope representing peptides GGGLELDKWASL for 2F5, KKWNWFDITNWGGG for 4E10, or gp160 for 2G12. Plates were washed and incubated with serial dilutions of plasma samples. Bound 2F5 and 4E10 were detected with goat anti-human IgG conjugated with horseradish peroxidase. 2G12 was detected with an anti2G12-idiotypic antibody. Plasma concentrations were determined from standard curves generated from known levels of 4E10, 2F5 and 2G12.
Mixing studies with pooled adult and neonate plasma
Pooled adult or neonate plasma was spiked with 4E10, 2F5 and 2G12 to achieve a final concentration of 40, 200 or 500 μg/ml, respectively, or with phosphate-buffered saline in controls.
Measurement of clotting activity
Activated partial thromboplastin time (aPTT; Pathromtin SL, Dade Behring Marburg Germany) and prothrombin time (PT; Thromborel S, Dade Behring Marburg Germany) were measured using a Schnittger-Gross Coagulometer (normal values for aPTT: neonates, ≤39 s; adults, ≤37 s).
The diluted Russell's viper venom time (DRVVT; Life Diagnostics, French Forest, Australia) was measured as a ratio, as was the aPTT, both using a Schnittger-Gross Coagulometer (normal aPTT ratio, < 1.1; normal DRVVT ratio, < 1.3).
Serum autoantibody-binding activity
The binding activity of serum autoantibodies against cardiolipin and phosphatidylserine as well as antibodies against beta-2-glycoprotein 1 (ß2GP1) and prothrombin were measured using commercially available ELISA kits (Orgentec Diagnostika, Mainz, Germany) according to the manufacturer's recommendations. Antibodies against the nuclear antigens nuclear ribonucleoprotein, Sm, RO/SS-A, LA/SS-B, SCL-70, JO-1, and against histones, double-stranded DNA, single-stranded DNA and centromers were measured by ELISA kits (INOVA Diagnostics, San Diego, California, USA). Antinuclear antibodies were assessed by indirect immunofluorescence using Hep-2 cells in a clinically validated antinuclear antibody assay.
Plasma samples from trial subjects
Plasma samples from subjects participating in three clinical trials were analyzed. Two studies were conducted at the Otto Wagner Spital Vienna and were published previously [20-22]. Both studies were approved by the ethics committee and all subjects gave their informed consent for the use of stored plasma samples for scientific purposes. Both studies were phase I studies investigating either the combination of 2F5 and 2G12 (study 1) or the single application of 4E10 followed by the triple combination of 4E10/2F5/2G12 (study 2) over a treatment period of 4 weeks in HIV-infected subjects not receiving any other antiretroviral therapy.
The third study was performed at the Aaron Diamond AIDS Research Center New York [24]. In brief, 10 HIV-1-infected individuals, identified and treated during acute and early infection with HAART for 15 to 60 months (mean, 29.1), were studied. Viral loads were < 50 copies/ml for at least 6 months on HAART. Patients received weekly infusions of the three monoclonal antibodies 4E10/2F5/2G12 for 16 weeks; HAART was discontinued at week 4. The first six patients (group I) were given 1 g of each of the three monoclonal antibodies. In the last four patients (group II), the doses of 2F5 and 4E10 were increased to 2 g based on their short half-lives and absence of antibody-related adverse drug reactions in group I.
The protocol was amended to include assessment of blood coagulation following the publication of the findings of Haynes et al. [1]. Pre- and postinfusion data are available for four subjects receiving 1-5 (total 2) infusions.
Reevaluation of safety data in humans
The reevaluation of monoclonal antibody safety was based on the safety data that have been collected in four clinical trials conducted with the antibodies. The following parameters were investigated for the evaluation of safety. Targeted physical examinations including diagnoses, vital signs, symptoms, and review of concomitant medications were carried out at every visit. Subjects were monitored for adverse events during and between the antibody infusions. Blood was collected on regular basis throughout the study periods for the investigation of hematology (hemoglobin, hematocrit, red blood cell count, white blood cell count, differential white blood cell count, absolute neutrophil count, platelets), liver and kidney function tests (total bilirubin, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, creatinine) and blood chemistries (glucose, triglycerides, cholesterol, sodium, potassium, chloride, lactate dehydrogenase, creatinine kinase, total protein).
Special emphasis was given to immunological investigations (CD4/CD8 subsets and percentages). Further safety evaluations were urine analysis and the investigation of immune response against the infused antibodies.
All clinical trials included follow up for prolonged periods of time (between 7 weeks and 1 year). Monoclonal antibodies 4E10 and 2F5 were detectable for 2-4 weeks after the last infusion, 2G12 for at least 10 weeks after the last infusion.
Results
In order to examine the in-vivo relevance of the reported in-vitro polyspecific reactivity of 2F5 and 4E10, the effect of monoclonal antibody infusions on clinically relevant coagulation parameters were investigated. An ongoing clinical study was amended for investigation of PT/aPTT. In this phase II study, subjects received 16 weekly doses of a combination of 4E10, 2F5 and 2G12 (5 g total per infusion) (study 4; Table 1). Four subjects were still scheduled to receive the last one to five antibody infusions at the amendment date, which allowed for a total of 12 PT/aPTT measurements directly pre- and postinfusions and seven measurements during the follow up period: At baseline (prestudy), the four subjects were negative for 4E10, 2F5 and 2G12. Trough levels of 2F5, 4E10, and 2G12 1 week after the last infusion (directly before the next infusion) were 114.9 ± 46.9, 223.8 ± 56.7, and 654.4 ± 82.9 μg/ml, respectively. Infusion of 2 g 2F5, 2 g 4E10, and 1 g 2G12 resulted in peak levels of 870.3 ± 95.8, 1051.2 ± 71.2 and 966.8 ± 79.6 μg/ml, respectively, 30 min after the end of infusion (Fig. 1a).
Monoclonal antibody infusions were associated with increases in serum anti-cardiolipin and anti-phosphatidylserine antibody titers (Pearson correlation coefficients 0.90 and 0.94, respectively; P < 0.001). The four subjects were seronegative for anti-cardiolipin antibodies and had a mean anti-phosphatidylserine titer of 5 ± 2.1 U/ml at baseline. One week after the previous infusion, mean anti-cardiolipin and anti-phosphatidylserine antibody titers of 52.1 ± 12.3 and 34.6 ± 15.7 U/ml were determined. Anti-cardiolipin and anti-phosphatidylserine titers at peak concentrations of 2F5, 4E10 and 2G12 were 105.7 ± 7.9 and 118.1 ± 29.8 U/ml, respectively. (Fig. 1b). Three weeks after the last infusion, aCL titers were undetectable and aPS titers returned to prestudy 5.0 ± 2.1 μg/ml values.
Subjects had normal PT/aPTT at study entry. Infusion of the triple antibody combination led to a mean increase in the aPTT of 9.6 ± 3.7 s at 30 min postinfusion. Trough antibody serum levels pre-infusion did not significantly influence PT/aPTT. Peak serum antibody concentrations 30 min postinfusion resulted in mild prolongations of aPTT in the four patients, either within the normal range or was grade 1 toxicity [> 1.1-1.25× upper limit of normal (ULN)] according the National Institutes of Health Division of AIDS adverse event grading table [27]. Prolongations were transient and returned to normal in all patients with wash out of 2F5 and 4E10. PT values were only minimally affected by monoclonal antibody infusions and stayed within the normal range at all time points (Fig. 1c).
As these clinical results in patients who had received a combination of three monoclonal antibodies did not allow for dissecting effects contributed by each single antibody, a further analysis reexamined the binding properties of 2G12, 4E10, and 2F5 to different autoantigens separately and under different conditions in vitro. Using clinical routine assays, binding was first investigated in the absence of human serum (i.e., diluted in infusion buffer) to the following: cardiolipin; phosphatidylserine; ß2GP1; the nuclear antigens nuclear ribonucleoprotein, Sm, RO/SS-A, LA/SS-B, SCL-70, JO-1; histones; double-stranded DNA; single-stranded DNA; centromers; antinuclear antibodies; and prothrombin. 2G12 tested negative for any cross reactivity while 4E10 exhibited significant and dose-dependent reactivity with cardiolipin and phosphatidylserine and showed low binding to ß2GP1 and prothrombin (data not shown). Antibody 2F5 showed only low binding to phosphatidylserine, which became detectable at 1000 μg/ml. In contrast to the data published by Haynes et al. [1], antibody 2F5 did not bind to cardiolipin (data not shown).
Subsequently, the binding properties of the antibodies were investigated in the presence of human plasma, which also allowed the effects on the coagulation profiles to be studied in vitro. Pooled adult and neonate plasmas were spiked with increasing antibody concentrations relevant for therapeutic and prophylactic clinical application (40, 200 and 500 μg/ml). Results found in the buffer system were confirmed (Fig. 2). No cross-reactivities were detected for 2G12. 4E10 exhibited significant and dose-dependent reactivity with cardiolipin and phosphatidylserine and showed low-level binding to ß2GP1 and prothrombin that was undetectable at 40 and 200 μg/ml but became detectable at 500 μg/ml (Fig. 2). Antibody 2F5 did not bind to cardiolipin or any other antigen tested with the exception of low binding to phosphatidylserine, which was detectable only in neonate plasma at 500 μg/ml. Binding to cardiolipin was not detected at any concentration, neither in adult nor neonate plasma.
Consistent with the observed cross-reactive binding properties, only 4E10 exerted influence on the coagulation profile of adult and neonate plasma. Dose-dependent prolongations of the PT and aPTT were detected in adult and neonate plasma spiked with 4E10. Additionally, increase in the DRVVT ratio was found in adult but not neonate plasma. 4E10-induced prolongations of clotting times were mild and did not exceed grade I toxicities even at a concentration of 500 μg/ml. At 40 μg/ml, all test results remained within the normal range. In contrast, plasma spiked with 2G12 or 2F5 did not exhibit significant variations in the coagulation profile even at the highest test concentration (Fig. 2).
To further confirm that only 4E10, but not 2F5 or 2G12, was responsible for the observed effect on coagulation profiles in vivo, cryopreserved samples from earlier phase I studies were investigated retrospectively using routine assays for anti-cardiolipin and anti-phosphatidylserine antibodies.
The first set of samples originated from a study in which patients had been treated with the combination of 2F5 and 2G12. Infusions consisted of 1 g 2F5 and 1 g 2G12 and resulted in median peak plasma concentrations of 374 μg/ml and 605 μg/ml for 2F5 and 2G12, respectively [20]. Six of seven patients had undetectable anti-cardiolipin antibody titers over the entire study period. One patient showed low anti-cardiolipin antibody titer at study entry prior to antibody infusion, which did not increase during the study period. Measurements of anti-phospholipid showed detectable concentrations at base line and a minimal increase from 20.9 ± 6.0 to 28 ± 5.4U/ml when comparing pre- and postinfusion values (Fig. 3).
The second set of samples originated from a study in which a single infusion of antibody 4E10 had been administered on study day 1 followed by three infusions at 1 week intervals of the triple combination of 4E10, 2F5 and 2G12 [21]. Plasma of six subjects was available for investigating the effect of the single infusion of 1 g 4E10 on 4E10, anti-cardiolipin antibodies and anti-phosphatidylserine antibodies plasma titers. At baseline, subjects had undetectable titers of 4E10 and anti-cardiolipin and a mean anti-phosphatidylserine titer of 13.3 ± 9.9U/ml. At 30 min postinfusion, plasma titers of 4E10, aCL and aPS were 345.5 ± 52.6, 74.5 ± 21.1 and 58.9 ± 14.2 U/ml, respectively (Fig. 4a).
One subject in this study stopped treatment after receiving one single infusion of 4E10 for personal reasons. In this patient (patient 3), the increased aCL and aPS plasma titers generated by the cross-reactivity of 4E10 disappeared over time in correlation with the wash out of 4E10 (Fig. 4b).
Investigation of samples from these two studies confirmed the in-vitro findings that only 4E10 shows cross-reactivity to cardiolipin and influences coagulation profile.
Finally, to investigate if monoclonal antibody infusions harbor a significant medical risk, all accumulated safety data were reviewed from the four clinical trials with respect to concerns arising from possible cross reactivity to cardiolipin. In summary, the antibodies were well tolerated and no serious adverse events occurred [20,21,23,24,28]. Possibly antibody-related adverse events were primarily myalgias (n = 7) and arthralgias (n = 4).
Symptoms that have been reported to be connected to persistent titers of cardiolipin antibodies, such as thrombosis, thrombocytopenia, livedo reticularis, stroke, superficial thrombophlebitis, pulmonary embolism, transient ischemic attack and hemolytic anemia, were not observed. No venous thrombosis at infusion and venipuncture sites occurred.
Laboratory parameters were reviewed with respect to changes in platelet counts and hemoglobin levels, since thrombocytopenia and hemolytic anemia have been related to persistent high titers of anti-cardiolipin antibodies. No significant changes were observed that could possibly be connected with the infusions of monoclonal antibodies. Since decreased platelet counts is a general symptom of HIV infection, a direct correlation of platelet levels was found only with viral loads, and no decrease occurred subsequent to antibody administrations (Table 1).
Laboratory safety parameters (hematology, clinical chemistry and urine analysis) showed either changes within the normal range or no change compared with pretrial levels in most patients. Transiently elevated levels of alanine aminotransferase (1.5× ULN) were found in three patients at one single time point each and were probably related to exercise. Transient microalbuminuria was detected at low levels in five patients at one time point each. Specifically, no eosinophilia was detected and patients with myalgias did not show elevated creatine kinase levels. Urine sediments were uneventful. In summary, no laboratory abnormalities were detected that could be attributed to repeated passive immunization.
Infusion-induced anti-cardiolipin titers were only observed in patients receiving 4E10 infusions and disappeared in all study subjects 2-4 weeks after the last infusion, correlating with the wash out of 4E10.
Discussion
Over recent years, it has become evident that induction of both cellular and humoral immune responses are likely to be indispensable in developing an effective HIV-1 vaccine. The most broadly neutralizing antibodies known to date, 2F5, 2G12, Igh1b12 and 4E10, are used as prime models for design of antigens for eliciting cross-clade neutralizing antibody responses. The report that three of these four antibodies (Igh1b12, 2F5 and 4E10) harbor autoreactive binding activities were daunting and challenging. Moreover, the ongoing clinical development of 2F5 and 4E10 for passive immunotherapy, including the use for prevention of vertical HIV transmission, was seriously jeopardized.
The access to clinical data and samples from patients infused with high doses of 2F5, 4E10 and 2G12 gave us the unique opportunity to examine if the reported in-vitro findings of Haynes et al. [1] were of clinical relevance and in what way they might influence future clinical investigations.
2F5, 4E10 and 2G12 have undergone clinical testing in four different studies. In these trials, a total of 39 subjects received 418 infusions consisting of 1-5 g monoclonal antibodies. Repeated high-dose monoclonal antibody infusions were tolerated well by all patients and mild adverse events only occasionally reported [20,21,23].
Based on the present investigation, we conclude that only infusions of monoclonal antibody 4E10, but not 2F5 or 2G12, are connected with mild prolongations in aPTT. This effect is directly correlated to the plasma concentration of 4E10 and resolves at the end of the antibody treatment phase. The aPTT prolongation is most likely causally related to the observed binding of 4E10 to cardiolipin in vitro, thus confirming previous studies by Haynes et al. [1]. Our results also confirm that 2G12 lacks cross-reactivity with the autoantigens tested; in contrast to the findings of Haynes et al. [1], 2F5 failed to show cross-reactivity to cardiolipin when tested with a validated assay that is routinely used for the detection of anti-cardiolipin antibodies (aCL) in patients. In addition Haynes et al. [1] reported no cross-reactivity of 2F5 with phosphatidylserine, while we show low binding. Although the exact reason for these partially divergent results is unclear, minimal differences in the experimental methods used would offer an explanation. The lack of cardiolipin-binding of 2F5 and 2G12 is in good agreement with the results of coagulation assays, in which no effect on coagulation was found. Consistently, 4E10 and not 2F5 or 2G12 has to be considered an autoreactive anti-phospholipid antibodies (aPL).
Anti-phospholipid antibody activity as well as lupus anticoagulant have been found in association with acquired thrombophilia in the aPL syndrome [29]. However, many persons harboring anti-phospholipid do not suffer from thromboembolic disease. The pathogenesis of aPL syndrome was critically reviewed recently [30]: While there is an accumulating body of evidence suggesting that specific subgroups of aPL may directly contribute to disease pathogenesis, mechanisms remain unclear. Some studies find a specific correlation between clinical disease and ß2GP1 antibodies, but in view of the low affinity for this protein, it seems unlikely that interactions in solution can completely explain the pathogenesis of thromboembolic disease [30-32]. A more likely explanation would be that an underling disease fosters thromboembolic disease by affecting surface-mediated phenomena, for example by exposing negatively charged phospholipid. Then aPL would function as 'second hit' [31].
While the association between aPL and thromboembolic disease is established, individual patient risk of thrombosis cannot be predicted from a particular profile of aPL and/or lupus anticoagulant [32,33]. Furthermore it was demonstrated that sera from healthy blood donors have naturally occurring aPL that are undetectable because they are bound in immune complexes. It was speculated that these antibodies serve as scavengers removing apoptotic cells and cell debris from sites of inflammation [34,35].
In our study only one monoclonal antibody, 4E10, showed activity against phospholipid and lupus anticoagulant and this was of low titer. While some were concerned about possible risk [36], others speculated that phospholipid binding might be important for the broad neutralizing capabilities [37]. Although thromboembolic risk cannot definitely be excluded, it can be speculated that the risk of thrombosis would be low with transient low-titer activity against phospholipid and lupus anticoagulant and would mostly depend on other underlying disorders.
In the light of the present findings, coagulation profile testing for aPL titers should be part of each clinical protocol investigating monoclonal antibody 4E10.
Acknowledgements
We thank the patients for their participation. We would like to thank Thomas Wittenberger for helping with the graphical presentation of data.
Sponsorship: This work was supported in part by grants from the National Institutes of Health (NIH)-sponsored Acute HIV Infection and Early Disease Research Program (AI-41534) and the General Clinical Research Center at the National Center for Research Resources at the NIH (M01-RR00102).
References
1. Haynes BF, Fleming J, St Clair WE, Katinger H, Stiegler G, Kunert R, et al. Cardiolipin polyspecific autoreactivity in two broadly neutralizing HIV-1 antibodies. Science 2005; 308:1906-1908.
2. Purtscher M, Trkola A, Gruber G, Buchacher A, Predl R, Steindl F, et al. A broadly neutralizing human monoclonal antibody against gp41 of human immunodeficiency virus type 1. AIDS Res Hum Retroviruses 1994; 10:1651-1658.
3. Trkola A, Purtscher M, Muster T, Ballaun C, Buchacher A, Sullivan N, et al. Human monoclonal antibody 2G12 defines a distinctive neutralization epitope on the gp120 glycoprotein of human immunodeficiency virus type 1. J Virol 1996; 70:1100-1108.
4. Kessler JA II, McKenna PM, Emini EA, Chan CP, Patel MD, Gupta SK, et al. Recombinant human monoclonal antibody IgG1b12 neutralizes diverse human immunodeficiency virus type 1 primary isolates. AIDS Res Hum Retroviruses 1997; 13:575-582.
5. Stiegler G, Kunert R, Purtscher M, Wolbank S, Voglauer R, Steindl F, et al. A potent cross-clade neutralizing human monoclonal antibody against a novel epitope on gp41 of human immunodeficiency virus type 1. AIDS Res Hum Retroviruses 2001; 17:1757-1765.
6. Zwick MB, Labrijn AF, Wang M, Spenlehauer C, Saphire EO, Binley JM, et al. Broadly neutralizing antibodies targeted to the membrane-proximal external region of human immunodeficiency virus type 1 glycoprotein gp41. J Virol 2001; 75:10892-10905.
7. Parren PW, Moore JP, Burton DR, Sattentau QJ. The neutralizing antibody response to HIV-1: viral evasion and escape from humoral immunity. AIDS 1999; 13(Suppl A):S137-S162.
8. Burton DR, Desrosiers RC, Doms RW, Koff WC, Kwong PD, Moore JP, et al. HIV vaccine design and the neutralizing antibody problem. Nat Immunol 2004; 5:233-236.
9. Mascola JR, Lewis MG, Stiegler G, Harris D, VanCott TC, Hayes D, et al. Protection of macaques against pathogenic simian/human immunodeficiency virus 89.6PD by passive transfer of neutralizing antibodies. J Virol 1999; 73:4009-4018.
10. Baba TW, Liska V, Hofmann-Lehmann R, Vlasak J, Xu W, Ayehunie S, et al. Human neutralizing monoclonal antibodies of the IgG1 subtype protect against mucosal simian-human immunodeficiency virus infection. Nat Med 2000; 6:200-206.
11. Mascola JR, Stiegler G, VanCott TC, Katinger H, Carpenter CB, Hanson CE, et al. Protection of macaques against vaginal transmission of a pathogenic HIV-1/SIV chimeric virus by passive infusion of neutralizing antibodies. Nat Med 2000; 6:207-210.
12. Ferrantelli F, Rasmussen RA, Buckley KA, Li PL, Wang T, Montefiori DC, et al. Complete protection of neonatal rhesus macaques against oral exposure to pathogenic simian-human immunodeficiency virus by human anti-HIV monoclonal antibodies. J Infect Dis 2004; 189:2167-2173.
13. Hofmann-Lehmann R, Vlasak J, Rasmussen RA, Smith BA, Baba TW, Liska V, et al. Postnatal passive immunization of neonatal macaques with a triple combination of human monoclonal antibodies against oral simian-human immunodeficiency virus challenge. J Virol 2001; 75:7470-7480.
14. Weiss L, You JF, Giral P, Alhenc-Gelas M, Senger D, Kazatchkine MD. Anticardiolipin antibodies are associated with antiendothelial cell antibodies but not with antibeta 2 glycoprotein I antibodies in HIV infection. Clin Immunol Immunopathol 1995; 77:69-74.
15. Abuaf N, Laperche S, Rajoely B, Carsique R, Deschamps A, Rouquette AM, et al. Autoantibodies to phospholipids and to the coagulation proteins in AIDS. Thromb Haemost 1997; 77:856-861.
16. Petrovas C, Vlachoyiannopoulos PG, Kordossis T, Moutsopoulos HM, Autoimmun J. Antiphospholipid antibodies in HIV infection and SLE with or without antiphospholipid syndrome: comparisons of phospholipid specificity, avidity and reactivity with beta2-GPI. J Autoimmun 1999; 13:347-355.
17. Loizou S, Singh S, Wypkema E, Asherson RA. Anticardiolipin, antibeta(2)-glycoprotein I and antiprothrombin antibodies in black South African patients with infectious disease. Ann Rheum Dis 2003; 62:1106-1111.
18. Palomo I, Alarcon M, Sepulveda C, Pereira J, Espinola R, Pierangeli S. Prevalence of antiphospholipid and antiplatelet antibodies in human immunodeficiency virus (HIV)-infected Chilean patients. J Clin Lab Anal 2003; 17:209-215.
19. Klein SK, Slim EJ, de Kruif MD, Keller TT, ten Cate H, van Gorp EC, et al. Is chronic HIV infection associated with venous thrombotic disease? A systematic review. Neth J Med 2005; 63:129-136.
20. Armbruster C, Stiegler GM, Vcelar BA, Jager W, Michael NL, Vetter N, et al. A phase I trial with two human monoclonal antibodies (hMAb 2F5, 2G12) against HIV-1. AIDS 2002; 16:227-233.
21. Armbruster C, Stiegler GM, Vcelar BA, Jager W, Koller U, Jilch R, et al. Passive immunization with the anti-HIV-1 human monoclonal antibody (hMAb) 4E10 and the hMAb combination 4E10/2F5/2G12. J Antimicrob Chemother 2004; 54:915-920.
22. Stiegler G, Armbruster C, Vcelar B, Stoiber H, Kunert R, Michael NL, et al. Antiviral activity of the neutralizing antibodies 2F5 and 2G12 in asymptomatic HIV-1-infected humans: a phase I evaluation. AIDS 2002; 16:2019-2025.
23. Trkola A, Kuster H, Rusert P, Joos B, Fischer M, Leemann C, et al. Delay of HIV-1 rebound after cessation of antiretroviral therapy through passive transfer of human neutralizing antibodies. Nat Med 2005; 11:615-622.
24. Mehandru S, Vcelar B, O'Neil M, Wrin T, Stiegler G, Shet A, et al. Monoclonal antibody infusions delay HIV-1 rebound after discontinuation of ART in a cohort of HIV-1-infected individuals treated during primary HIV-1 infection, 13th Conference on Retroviruses and Opportunistic Infections. Denver, February 2006 [abstract 178].
25. Buchacher A, Predl R, Strutzenberger K, Steinfellner W, Trkola A, Purtscher M, et al. Generation of human monoclonal antibodies against HIV-1 proteins; electrofusion and Epstein-Barr virus transformation for peripheral blood lymphocyte immortalization. AIDS Res Hum Retroviruses 1994; 10:359-369.
26. Kunert R, Steinfellner W, Purtscher M, Assadian A, Katinger H. Stable recombinant expression of the anti HIV-1 monoclonal antibody 2F5 after IgG3/IgG1subclass switch in CHO cells. Biotechnol Bioeng 2000; 67:97-103.
27. National Institutes of Health Division of AIDS. Table for Grading the Severity of Adult and Pediatric Adverse Events. Bethesda, MD: National Institutes of Health; 2004.
28. Joos B, Trkola A, Kuster H, Aceto L, Fischer M, Stiegler G, et al. Long-term multiple-dose pharmacokinetics of human monoclonal antibodies (MAbs) against human immunodeficiency virus type 1 envelope gp120 (MAb 2G12) and gp41 (MAbs 4E10 and 2F5). Antimicrob Agents Chemother 2006; 50:1773-1779.
29. Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006; 4:295-306.
30. Giannakopoulos B, Passam F, Rahgozar S, Krilis SA. Current concepts on the pathogenesis of the antiphospholipid syndrome. Blood 2007; 109:422-430.
31. de Groot PG, Derksen RH. Pathophysiology of the antiphospholipid syndrome. J Thromb Haemost 2005; 3:1854-1860.
32. Arnout J, Vermylen J. Current status and implications of autoimmune antiphospholipid antibodies in relation to thrombotic disease. J Thromb Haemost 2003; 1:931-942.
33. Pabinger I, Vormittag J. Thrombophilia and pregnancy outcomes. J Thromb Haemost 2005; 3:1603-1610.
34. Alving CR. Antibodies to lipids and liposomes: immunology and safety. J Liposome Res 2006; 16:157-166.
35. Cabiedes J, Cabral AR, Alarcon-Segovia D. Hidden antiphospholipid antibodies in normal human sera circulate as immune complexes whose antigen can be removed by heat, acid, hypermolar buffers or phospholipase treatments. Eur J Immunol 1998; 28:2108-2114.
36. Nabel GJ. Close to the edge: neutralizing the HIV-1 envelope. Science 2005; 308:1878-1879.
37. Brown BK, Karasavvas N, Beck Z, Matyas GR, Birx DL, Polonis VR, et al. Monoclonal antibodies to phosphatidylinositol phosphate neutralize human immunodeficiency virus type 1: role of phosphate-binding subsites. J Virol 2007; 81:2087-2091.
© 2007 Lippincott Williams & Wilkins, Inc.