Background: Before therapeutic C5 inhibition, thromboembolism accounted for 40-67% of deaths in paroxysmal nocturnal hemoglobinuria (PNH). Anticoagulation alone is ineffective in preventing thromboembolism. Further, bleeding risk is significant due to co-occurrence of marrow failure and hepatic dysfunction. C5 inhibition decreases recurrent thromboembolism, however many patients remain on anticoagulation. There is limited data whether anticoagulation in PNH patients with history of thromboembolism can be safely discontinued.
Aims: Compare the risk of recurrent thromboembolism in PNH patients with and without anticoagulation on C5 inhibition.
Methods: We reviewed the electronic medical records of patients at Johns Hopkins Hospital between 1/2005-10/2020 with documented PNH clones treated with eculizumab or ravulizumab for >6 months. Patients with history of thromboembolism by imaging or high clinical suspicion were selected. The period on C5 inhibition included thromboembolic events from treatment initiation through last follow-up or bone marrow transplant, as long as therapy was continued with <1-week interruption. Thromboembolic rates for the period pre-C5 inhibition and during C5 inhibition were calculated as the total events divided by the time in years on a per patient basis and compared using the Fisher exact test. This study was approved by the Johns Hopkins IRB.
Results: Of 21 patients with history of thromboembolism, 11 discontinued anticoagulation, 6 never received or could not tolerate anticoagulation, and 4 continued anticoagulation after initiation of C5 inhibition (Figure 1, Table 2). Thrombosis rate pre-C5 inhibition was 26.3 events/100 patient-years compared with 1.5 events/100 patient-years on anti-C5 monotherapy (P <0.001) and 5.4 events/100 patient-years on combined anticoagulation and C5 inhibition (P=0.016). Two thromboembolic events on anti-C5 monotherapy were provoked and treated with 3-6 months of anticoagulation. Thrombosis rates between the anti-C5 monotherapy and C5 inhibitor plus anticoagulation groups were not significantly different (P=0.4).
Thromboembolic events in PNH patients treated with C5 inhibition.
| Patient Group | Patients treated with anti-C5 monotherapy (n=17) | Patients treated with C5 inhibitor and indefinite anticoagulation (n=4) |
| Diagnosis Classical PNH PNH/AA |
10 7 |
4 – |
| Sex Male Female |
10 7 |
1 3 |
| Median age of diagnosis (range) | 24 years (10-59) | 41 years (36-61) |
| Median granulocyte clone (range) | 97% (73-100) | 87% (78-99) |
| Median time prior to anti-C5 treatment (range) | 5 years (1-14) | 3.5 years (1-8) |
| Median time on anti-C5 treatment (range) | 10 years (0.5-14) | 9.5 years (5-13) |
| Median time on anticoagulation (range) | 1 month (0-11 years) | 9 years (5-19) |
| Location of TE pre-C5 inhibitor DVT pulmonary embolism abdominal vein dermal small bowel cerebrovascular inferior vena cava renal vein ureter tonsillar |
1 2 9 1 3 3 1 2 1 1 |
1 1 2 – – 2 – – – – |
| Location TE on C5 inhibitor DVT pulmonary embolism |
2 – |
1 1 |
Baseline patient characteristics and thromboembolic (TE) events.
Conclusions: Discontinuation of anticoagulation for secondary prevention of thromboembolism in PNH patients well-controlled on terminal complement inhibition appears safe.
To cite this abstract in AMA style:
Gerber G, DeZern A, Chaturvedi S, Brodsky R. Recurrent Thromboembolic Risk in Paroxysmal Nocturnal Hemoglobinuria Patients not on Anticoagulation Treated with Terminal Complement Inhibition [abstract]. Res Pract Thromb Haemost. 2021; 5 (Suppl 2). https://abstracts.isth.org/abstract/recurrent-thromboembolic-risk-in-paroxysmal-nocturnal-hemoglobinuria-patients-not-on-anticoagulation-treated-with-terminal-complement-inhibition/. Accessed November 28, 2022.« Back to ISTH 2021 Congress
ISTH Congress Abstracts - https://abstracts.isth.org/abstract/recurrent-thromboembolic-risk-in-paroxysmal-nocturnal-hemoglobinuria-patients-not-on-anticoagulation-treated-with-terminal-complement-inhibition/