Venous thrombosis is a major problem in paroxysmal nocturnal hemoglobinuria (PNH). The studies of Hillmen and colleagues in this issue of Blood show that an antibody to complement C5, eculizumab (Soliris), markedly inhibits thrombosis in affected patients.

Although some studies show that anticoagulation can prevent thrombosis in some cases of PNH, most experts agree that once thrombosis has occurred, it is likely to progress or to recur despite adequate anticoagulation. A better prophylaxis and treatment is needed.

The study by Hillmen and colleagues published in this issue of Blood may provide the answer.

The tendency of patients with PNH to suffer thrombosis was first noted in the 1930s and was emphasized by Crosby in the early 1950s.1  It has since been recognized as a major part of the syndrome and has been marked in several epidemiological studies as a very bad prognostic sign and the most common cause of death in PNH. About 30% to 40% of patients of European origin have serious thrombosis at some time; for unexplained reasons, only 5% to 10% of patients of East Asian (Chinese, Japanese, and Thai) or Mexican origin develop this complication.2,3 

The reason for this propensity is not entirely clear. Intravascular hemolysis may provide altered membrane surfaces upon which coagulation may be initiated. More likely, it is the effect of the activation of complement on platelets and perhaps endothelial cells. In platelets, the deposition of C9 complexes on the surface stimulates their removal by vesiculation; these vesicles are very thrombogenic. Because PNH platelets lack the mechanism for downregulating C9 deposition (ie, CD59), even a minimal stimulus from activated complement results in a greatly increased production of these vesicles.

The drug eculizumab, a humanized monoclonal antibody against C5 inhibiting its acti-vation, has been shown in both double-blind and safety studies to dramatically reduce the hemolysis of PNH. Although thrombosis was not a primary end point of these studies, Hillmen et al have analyzed the rate of thrombosis in patients who were taking the drug and compared it with the rate seen while patients were not taking the drug. There is a dramatic difference.

The study on thrombosis was not done by the method that is often considered the gold standard—the blinded placebo versus the drug trial. Rather, much of the data for the “placebo” arm was retrospective, covering a time period before the drug was given equal to the time during which the patient was taking the drug. Nevertheless, the difference is so great that some systemic bias would be needed to explain it, which is very unlikely.

Eculizumab will be extensively used in the treatment of PNH. The careful studies that are planned will tell whether the incidence of thrombosis is decreased (as we would expect) and whether previous or acquired thromboses can be controlled. From the current data, it would seem appropriate to use eculizumab in the treatment and prevention of thrombosis in selected patients with PNH.

Conflict-of-interest disclosure: The author declares no competing financial interests. ■

REFERENCES

1
Crosby
 
WH
PNH: relation of the clinical manifestations to underlying pathogenic mechanisms.
Blood
1953
, vol. 
8
 (pg. 
769
-
778
)
2
Socie
 
G
Mary
 
J-Y
De Gramont
 
A
, et al. 
Paroxysmal nocturnal haemoglobinuria: long term follow-up and prognostic factors.
Lancet
1996
, vol. 
348
 (pg. 
573
-
577
)
3
Nishimura
 
J
Kanakura
 
Y
Ware
 
RE
, et al. 
Clinical course and flow cytometric analysis of paroxysmal nocturnal hemoglobinuria in the United States and Japan.
Medicine (Baltimore)
2004
, vol. 
83
 (pg. 
193
-
207
)