Brief Report: Induction of sustained remission in recurrent catastrophic antiphospholipid syndrome via inhibition of terminal complement with eculizumab

Authors


Abstract

Objective

Catastrophic antiphospholipid syndrome (CAPS) is characterized by histopathologic evidence of small vessel thrombosis, dysfunction of multiple organs occurring over a short period of time, and laboratory confirmation of the presence of antiphospholipid antibodies (aPL). Treatment of CAPS focuses on anticoagulation therapy and on removal of aPL that promote thrombosis by activating endothelial cells, monocytes, and platelets. Studies in animal models support the hypothesis that a more targeted intervention, such as complement inhibition, may be an effective means to prevent aPL-induced thrombosis. Herein we describe use of an inhibitor of complement activation to treat CAPS that was refractory to conventional therapy.

Methods

Our patient was a young man who had recurrent CAPS characterized by multiple arterial thromboses in large and small vessels despite maximal anticoagulation, immunosuppression, and plasma exchange therapy. We treated him with eculizumab, an anti-C5 monoclonal antibody that blocks activation of terminal complement.

Results

Administration of eculizumab, at doses that blocked complement activity, aborted acute progressive thrombotic events, reversed thrombocytopenia, and was associated with no further clinical episodes of thrombosis during >3 years of therapy.

Conclusion

This first report of the use and clinical efficacy of eculizumab, an inhibitor of complement activation, in the treatment of CAPS demonstrates both the importance of complement (specifically, terminal complement components) in the pathogenesis of CAPS and the therapeutic benefit of complement inactivation.

Catastrophic antiphospholipid syndrome (CAPS) is a rapidly developing, life-threatening, multisystem disease with high mortality rates. Patients with this syndrome have clinical evidence of multiorgan involvement that evolves over a short period of time, histopathologic evidence of multiple small-vessel occlusions in the lung, brain, heart, kidneys, skin, and/or gastrointestinal tract, and laboratory confirmation of the presence of antiphospholipid antibodies (aPL) (1). Although <1% of patients with APS develop CAPS, the very high morbidity and mortality in this subgroup pose a serious clinical challenge that can be best addressed by advancing our currently limited understanding of disease pathogenesis (2).

More than half of patients who develop CAPS have an identifiable triggering event, most commonly infection, a finding that is also supported by studies in experimental models (3). Our group and others have suggested that uncontrolled activation of complement may initiate and amplify cellular features characteristic of CAPS, such as endothelial cell activation, monocyte tissue factor expression, and platelet aggregation, as well as the histologic hallmark of the disease, thrombotic microangiopathy (TMA) (4–6). In mouse models, C5a–C5a receptor interactions produce lesions like those seen in TMA, and blockade of terminal complement activation at C5, or interruption of complement C5a–C5a receptor interactions, prevents complications of APS (4). The combination of aPL and endothelial cell activation leads to micro- and macrovascular occlusion, inhibition of natural anticoagulants, and release of proinflammatory factors (1, 7). The disease process is amplified and disseminated by excessive cytokine release from affected and necrotic tissues and has been likened to the events in the “systemic inflammatory response syndrome” (1).

Although inflammation clearly contributes to disease pathogenesis, treatment of CAPS has focused on anticoagulation therapy, removal of aPL, and suppression of aPL production (3). A better understanding of disease pathogenesis is needed in order to improve the prognosis of patients with CAPS. Herein we describe a patient with disease that was refractory to traditional aggressive management of CAPS, in whom long-term remission was achieved following therapy with the terminal complement inhibitor, eculizumab. These results underscore the potential importance of complement activation.

PATIENT AND METHODS

The patient, a 28-year-old man, presented with pulmonary emboli at age 12 years and was found to have systemic lupus erythematosus with arthritis, malar rash, and thrombocytopenia. He was subsequently maintained on a regimen of hydroxychloroquine and anticoagulation therapy with warfarin. The patient was first admitted for treatment of his current illness in January 2007, when his right foot became cold and painful. Angiography revealed a right tibial arterial thrombosis. Results of the initial laboratory evaluation showed the following values: hemoglobin 9.0 gm/dl, platelet count 183,000/μl, international normalized ratio (INR) 3.0, lactate dehydrogenase (LDH) 272 units/liter, and positive IgG and IgM anticardiolipin (46 units/ml and 70 units/ml, respectively [normal <20]), as well as IgG and IgM anti–β2-glycoprotein I (anti-β2GPI) (149 units/ml and 70 units/ml, respectively [normal <12]). He was treated with intravenous heparin and intraarterial urokinase, but showed no improvement in vascular flow. His course was complicated by worsening thrombocytopenia (platelet count 53,000/μl) and microangiopathic hemolytic anemia (3–4 schistocytes per high-power field [hpf], LDH 752 units/liter, haptoglobin <3 mg/dl). The thrombocytopenia was thought to be related to heparin therapy. Argatroban (2 μg/kg/minute), a direct thrombin inhibitor, was substituted for heparin. Arterial thrombectomy was unsuccessful.

Over the next 2 weeks, the patient developed fever, livedo reticularis, gangrene of the right foot that extended to the calf, severe abdominal pain, abnormal results on liver function tests (aspartate aminotransferase [AST] 99 IU/liter [normal 10–40], alanine aminotransferase [ALT] 167 IU/liter [normal 10–45]), elevated D-dimer levels (991 ng/ml [normal <500]), and progressive thrombocytopenia (platelet count from 150,000/μl to 35,000/μl) and anemia (hemoglobin 7.8 gm/ dl, 4–5 schistocytes/hpf, LDH 1,282 units/liter, haptoglobin <3 mg/dl). CAPS was diagnosed. The patient was treated with pulse steroids (methylprednisolone 500 mg/day for 4 days) and intravenous (IV) immunoglobulin (400 mg/kg/day for 5 days). Despite these measures, a right below-the-knee amputation was required. Anticoagulation therapy was switched from argatroban to subcutaneous fondaparinux (10 mg/day), and the patient was discharged (March 2007).

Twenty-four hours later, the patient was readmitted with left tibial arterial thrombosis that was confirmed by arteriography and evolved into a syndrome characterized by fever, abdominal pain, pulmonary infiltrates (small ground-glass opacities), thrombocytopenia (platelet count 65,000/μl), TMA (4–5 schistocytes/hpf, LDH 1,271 units/liter, haptoglobin <3 mg/dl), abnormal liver function test results (AST 78 IU/ liter, ALT 106 IU/liter) and IgG and IgM anti-β2GPI antibodies (58 units/ml and 16 units/ml, respectively). Recurrent CAPS was diagnosed, and he was treated with plasma exchange (total of 54 sessions), IV immunoglobulin (400 mg/kg/day for 5 days), IV argatroban (2 μg/kg/minute), oral clopidogrel (75 mg/day), and aspirin (325 mg/day). In addition, he received immunosuppressive therapy with high-dose pulse steroids (methylprednisolone 500 mg/day for 5 days) followed by prednisone (60 mg/day), IV cyclophosphamide (500 mg followed by 50 mg/day orally for 14 days), and rituximab (375 mg/m2/week for 3 weeks). Despite these aggressive measures, thrombocytopenia persisted (platelet count 32,000/μl), D-dimer levels remained elevated (1,910 ng/ml), and a left below-the-knee amputation was required. Histopathologic studies of the amputation specimen confirmed recent and chronic small- and large-vessel thromboses in both arterial and venous beds (Figure 1).

Figure 1.

Recent and chronic vascular thromboses in left leg amputation specimen. A, Artery occluded by recent thrombus, showing central blood clot and marginal organization. Original magnification × 5. B, Small vein showing recent blood clot and peripheral capillary recanalization. Original magnification × 25. C, Vein showing chronic thrombosis with complete occlusion by fibrous tissue, focal chronic inflammation (white arrow), and focal capillary recanalization (black arrow). Original magnification × 10.

The patient was discharged after a 2-month hospitalization during which he was treated with fondaparinux, clopidogrel, aspirin, prednisone, and hydroxychloroquine. His IgG anticardiolipin titer had fallen to 5 units/ml.

In October 2007, the patient was readmitted with severe chest and abdominal pain, thrombocytopenia (platelet count 34,000/μl), anemia (hemoglobin 8.0 gm/dl, 3–4 schistocytes/hpf), elevated liver enzyme levels (AST 408 IU/liter, ALT 501 IU/liter), elevated LDH levels (754 units/liter) and high IgG and IgM anticardiolipin antibody titers (64 units/ ml and 46 units/ml, respectively). Magnetic resonance imaging of the abdomen revealed a 14 × 10–cm hematoma in the right abdominal wall and infarction of the left lobe of the liver with thromboses in the hepatic artery and a branch of the portal vein, despite therapeutic anticoagulant activity (anti–factor Xa activity 0.9 IU/ml). Anticoagulation therapy was stopped due to the abdominal wall hematoma, and the patient was re-treated with plasma exchange and pulse methylprednisolone. After 6 days of plasma exchange, his clinical condition and laboratory results improved and anticardiolipin antibody became undetectable, but thrombocytopenia (platelet count 27,000/μl) and microangiopathic hemolytic anemia persisted.

Given the recurrent thromboses despite documented therapeutic anticoagulant activity, plasma exchange, and immunosuppression, we decided to administer eculizumab, a humanized monoclonal antibody against complement C5 that blocks its cleavage and prevents the generation of the prothrombotic and proinflammatory molecules C5a and membrane attack complex C5b–9 (8). IV eculizumab induction dosing (6 doses of 600 mg approximately every week) began in November 2007. Serum samples were collected for pharmacokinetic and pharmacodynamic studies of eculizumab and analyzed as previously described (8). Anticoagulation therapy with bivalirudin, a direct thrombin inhibitor, and therapy with prednisone (60 mg/day) and hydroxychloroquine were continued. The patient was also immunized against Neisseria meningitidis and received prophylactic ciprofloxacin.

Following the initial 6 doses of eculizumab, treatment was interrupted for 2 weeks due to Staphylococcus aureus sepsis (secondary to an infection at the site of a Hickman catheter) complicated by acute respiratory distress syndrome. Three weeks later the patient developed left hemiparesis and right frontoparietal peripheral infarcts. He had evidence of TMA with anemia (hemoglobin 7.8 gm/dl, 1–2 schistocytes/hpf) and thrombocytopenia (platelet count 78,000/μl). Pharmacokinetic and pharmacodynamic analyses during the eculizumab induction period showed complement breakthrough; therefore, when treatment with eculizumab was reintroduced, the induction dose was increased to 900 mg/week for 3 weeks. Subsequently maintenance doses of 1,200 mg every other week were given (Figure 2). Followup analyses demonstrated that the 1,200 mg maintenance doses resulted in eculizumab trough levels that were consistently in the therapeutic range (≥35 μg/ml) and serum complement activity remained blocked (≤20%) (Figure 2).

Figure 2.

Pharmacokinetic and pharmacodynamic analyses of eculizumab with corresponding platelet counts. Serum concentrations of eculizumab (pharmacokinetic data), serum complement activity (pharmacodynamic data), and platelet counts are shown. Serum eculizumab levels of >35 μg/ml (horizontal dashed line) typically result in complete complement blockade (≤20% serum complement activity in the pharmacodynamic assay).

There were no further neurologic events, thrombocytopenia and anemia resolved (platelet count 228,000/μl, hemoglobin 8.8 gm/dl), and D-dimer levels normalized (428 ng/ml) in the setting of therapeutic levels of eculizumab in the serum (Figure 2). The patient's neurologic deficits completely resolved, and he was discharged with full function on a regimen of prednisone (10 mg/day) and lepirudin (1.25 mg/kg administered subcutaneously twice daily). The patient continued to receive eculizumab (1,200 mg every other week) and remained stable and free of further thrombotic episodes for 1 year. Thereafter, the dose of eculizumab was decreased to 600 mg/month and lepirudin was replaced by warfarin (INR target 2.5–3.0). There have been no further clinical events for more than 3 years.

DISCUSSION

This report describes a patient with relapsing CAPS, an uncommon, life-threatening manifestation of APS (2, 9). Dramatic cases such as this can provide important insights into disease pathogenesis and advance understanding as to why some individuals develop catastrophic phenotypes. For example, lupus activity has been considered a poor prognostic factor in patients with CAPS. Our patient had a history of mild lupus, but his manifestations were primarily cutaneous and articular, and his lupus had remained inactive on a regimen of hydroxychloroquine. His first 2 thrombotic episodes at the onset of CAPS occurred in large arteries, more typical of classic APS than of CAPS, which is characterized by TMA. While no infections that could trigger the event were apparent, tissue necrosis due to ischemia in the legs during these episodes may have resulted in systemic inflammation.

When heparin-induced thrombocytopenia was suspected and argatroban was substituted for heparin, extensive microvascular thrombotic disease ensued, evidenced by progression of gangrene of the leg, severe abdominal pain, and microangiopathic hemolytic anemia. The rapid acceleration of disease following withdrawal of heparin suggests the possibility that heparin may have had therapeutic effects beyond anticoagulation. Indeed, our work in mouse models has shown that heparin inhibits activation of complement and thereby protects against aPL-induced pregnancy complications in mice, findings that are supported by in vitro studies (10).

Over the 10-month period prior to starting eculizumab, the patient received plasma exchange for 60 sessions, cyclophosphamide, pulse steroids, and rituximab, as well as anticoagulation therapy with argatroban, fondaparinux, aspirin, and clopidogrel. Although previous reports of series of patients or individual cases describe rituximab as an effective treatment for thrombocytopenia with aPL positivity and for CAPS (11), our patient had relapsing CAPS, characterized by severe chest and upper abdominal pain, thrombocytopenia, and new thrombotic lesions in the liver, despite treatment with rituximab and despite declining anticardiolipin antibody titers.

Given the recurrent and severe clinical manifestations, the lack of response to first- and second-line therapies, and the high mortality risk associated with CAPS, we chose to initiate an unconventional treatment, blockade of terminal complement activation with eculizumab. We have shown that terminal complement is a key innate immune effector engaged by aPL to induce thrombophilia, monocyte tissue expression, and leukocyte adhesion to endothelium (4, 5). Recent clinical evidence that uncontrolled terminal complement is a key mediator of TMA in patients with atypical hemolytic uremic syndrome (9, 12), a disease with histologic lesions similar to those in CAPS, also supports the consideration of complement inhibition as a treatment for CAPS. Eculizumab treatment benefits patients with paroxysmal nocturnal hemoglobinuria, reducing intravascular hemolysis and downstream morbidities, including recurrent thrombosis, by blocking complement-mediated pathogenic effects (8, 13, 14). In addition, eculizumab has recently been approved by the Food and Drug Administration for the treatment of atypical hemolytic uremic syndrome (in patients whose disease is resistant to plasma therapy and in whom plasma therapy is required), a disorder associated with inherited defects in complement regulation (14).

Our patient initially received an eculizumab induction dose of 600 mg/week, a regimen that did not consistently block complement (Figure 2). In addition, therapy was interrupted, causing a rebound in serum complement activity. Against this background, the patient sustained worsening transient neurologic conditions. Reinduction with 3 doses of 900 mg was therefore initiated before initiation of the maintenance dose of 1,200 mg every other week, with which inhibition of complement was sustained. Long-term eculizumab treatment combined with continued anticoagulation therapy with lepirudin resulted in the normalization of platelet counts (Figure 2). The patient has been in remission with no further thromboses.

These data demonstrate the successful treatment of recurrent CAPS that had been resistant to other interventions. The dosing of eculizumab was guided by monitoring of complement inhibitory activity. Lonze et al recently reported on a patient with a history of CAPS and renal failure due to TMA in whom eculizumab was used to prevent a recurrence of CAPS, rather than to treat active CAPS, while the patient underwent renal transplantation (15). Taken together, these results provide proof of concept that terminal complement blockade, a strategy that inhibits the key pathway of aPL-induced organ damage in animal models of APS, may prove to be an effective new approach to treating this condition. Importantly, response to eculizumab was achieved in our patient in the absence of heparin (eliminating heparin as a confounding source of complement inactivation). This report underscores the importance of investigating complement inhibitors, such as eculizumab, to treat and/or prevent CAPS and other manifestations of aPL-mediated organ damage.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Salmon had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Shapira, Allen, Salmon.

Acquisition of data. Shapira, Allen, Salmon.

Analysis and interpretation of data. Shapira, Andrade, Allen, Salmon.

Acknowledgements

We are grateful to Dr. Russell Rother for thoughtful discussions, the staff at Alexion Pharmaceuticals for pharmacokinetic and pharmacodynamic measurements of eculizumab, and Dr. Edward DiCarlo (Hospital for Special Surgery) for assistance with histopathologic analysis.

Ancillary