Paroxysmal nocturnal haemoglobinuria: the current scenario.

AV Pathare, D Mohanty 
 Dept of Hematology, KEM Hospital, Mumbai. , USA

Correspondence Address:
A V Pathare
Dept of Hematology, KEM Hospital, Mumbai.
USA

Full Text

Paroxysmal nocturnal haemoglobinuria (PNH) is an acquired clonal chronic haemolytic anaemia in which intravascular haemolysis results from an intrinsic defect in the membrane of red cells which makes the red cells highly susceptible to complement. In the 1930’s Ham in the USA[1] and Dacie in the UK[2] developed the acidified serum test, which became the defining diagnostic test for PNH.

In contrast to all other haemolytic anemias due to an intrinsic red cell abnormality, PNH is an acquired rather than an inherited disorder. This fact, together with the finding that normal cells co-exist in the patient’ blood with those that are hypersensitive to complement, led some 35 years ago to the working hypothesis that PNH arises through a somatic mutation in a haemopoietic cell[3].

The first evidence in support of this model was provided by showing that in women with PNH who were heterozygous at the X-linked locus for the enzyme glucose 6-phosphate dehydrogenase (G6PD), the patients’total red cell population showed expression of both G6PD alleles, but the fraction of red cells susceptible to lysis - which we call the PNH red cell population - showed expression of only one, suggesting that it was monoclonal in origin[4]. Subsequently, it was shown that individual erythroid colonies from PNH patients had either the normal or the PNH phenotype, but they were never mixed[5].

Quite recently a gene called PIG-A has been isolated by expression cloning, shown to correct the membrane abnormality of PNH cells and mapped to the X chromosome on Xp21.3[6],[7]. Sequence analysis of this gene has revealed mutations, usually point mutations or small insertion-deletions in the PNH cell population but not in the normal cell population of all PNH patients[8],[9],[10],[11].

Over the past five years, flow cytometry has been employed extensively to demonstrate the deficiency in PNH red cells of GPI-linked proteins such as CD59, CD48 and CD55[12],[13],[14]. This technique lends itself to accurate quantitation, as well as to testing for the PNH abnormality even in the peripheral blood granulocytes with anti-CD14 antibody and in the lymphocytes using both the B cell and T monoclonal antibodies. Granulocytes are always affected, and usually the proportion of PNH granulocytes is larger than that of PNH red cells, with values of 90% or more, thereby providing a more accurate measure of the size of the PNH clone. For this reason, if the clone is small, it may be undetectable by testing red cells; and the Ham test may have been reported as negative, but it may be detectable by testing granulocytes. Thus, flow cytometry has a higher sensitivity than the Ham test in detecting a PNH clone.

The natural history of PNH is that of a very chronic disorder, which may afflict the patient continuously for decades. Without treatment the median survival is estimated to be about eight years; the most common causes of death being thrombosis or haemorrhage associated with severe thrombocytopenia[15]. This latter complication signals a very important feature of PNH, namely its two-way relationship to AA.

On one hand, a patient who originally presented with the classical clinical picture of PNH, which we refer to as florid PNH - usually develops pancytopenia and is found to have a hypoplastic and eventually an aplastic bone marrow. This situation is usually termed as spent PNH. On the other hand, it happens that a patient with AA, sometimes years since the original diagnosis, develops clinical and laboratory features of PNH. This development is estimated to take place in up to 30% of patients treated with antilymphocyte globulin (ALG) (16), but this does not at all mean that it is caused by ALG, since it may occur in patients who were treated otherwise (say with androgens) or who had ‘spontaneous remission’of AA.

In fact, it appears that any patient who has survived AA without BMT is at risk of developing PNH. Thus, in terms of a temporal sequence PNH may be followed by AA or AA may be followed by PNH. But on closer inspection we find, not surprisingly, that features of the two may actually co-exist: for instance, a patient with florid PNH may have at the same time severe thrombocytopenia with very few megakaryocytes in the bone marrow. It is convenient to refer to these patients as having the PNH-AA syndrome[17].

::   Pathophysiology

Ever since the introduction of the Ham test, the study of PNH has been centered on two main issues namely, the nature of the red cell lesion which makes it susceptible to complement lysis, and secondly the basis for the long-term balance between the PNH blood cell population and an apparently normal cell population.

In PNH patients, a number of proteins that are membrane linked are deficient. Indeed, they all belong to the class of proteins that are anchored to the membrane through a glycosyl phosphatidyl inositol (GPI) anchor. One of these proteins, the membrane inhibitor of reactive lysis (MIRL; also known as CD59), is of major importance in preventing attack to the membrane by the C5-C9 complement complex. Inherited absence of this protein produces a phenotype very similar to PNH[18], and therefore acquired CD59 deficiency offers a good explanation for intravascular hemolysis in PNH.

It is now clear that the deficiency of GPI-linked proteins is the direct consequence of somatic mutations in the PIG-A gene[19]. Although the biochemical function of the PIG-A protein is yet to be completely elucidated, it is almost certainly involved in the transfer of acetylglucosamine onto phosphatidylinositol, the first step in the biosynthesis of the GPI anchor. Since GPI-linked proteins are deficient on the membrane of cells belonging to the erythroid, myeloid, megakaryocytic and lymphoid lineages, one can infer that the somatic mutation responsible for the PNH phenotype has taken place in a totipotent cell, thus qualifying PNH as a stem cell disorder. The thrombotic features of PNH are probably the result of an intrinsic abnormality of the membrane of platelets, which causes them to become inappropriately activated within the circulating blood.

PNH often presents the unusual combination of pancytopenia and reticulocytosis in the peripheral blood with erythroid hyperplasia in the bone marrow. This finding, together with the close relationship between PNH and AA, is consistent with the notion that an element of bone marrow failure may be present in every patient with PNH. Indeed, the proportion of PNH neutrophils in PNH patients who have normal or reduced neutrophil counts is often 90% or more, indicating an absolute decrease in normal neutrophils. In addition, markedly reduced numbers of erythroid and myeloid progenitors have been reported in the peripheral blood of patients with PNH, even when they had no overt AA[13]. It is therefore apparent that there are two components that are absolutely essential for the pathogenesis of PNH. PNH clones arise through spontaneous somatic mutations in the PIG-A gene of haemopoietic stem cells. As long as the remaining stem cells are normal, clinical PNH will not develop. Thus, the development of PNH is conditional on a background of BMF and PNH always co-exists with BMF. BMF is clinically obvious in patients who initially present with AA and then develop PNH. In patients who initially present with PNH, BMF may not be obvious because by the time of diagnosis, the PNH clone has expanded to the point where it provides a substantial proportion of the patient’s hematopoiesis. Moreover, the PNH clone has a long but probably finite life span. If, by the time the PNH clone is exhausted, and the BMF has not recovered, the patient evolves clinically from PNH to AA. If however, by the time the PNH clone is exhausted, and the BMF has recovered, the patient evolves to ‘cured’ PNH.

The existence of a florid PNH clone while the rest of hematopoiesis is depressed suggests that the PNH clone can be spared selectively from the injury affecting the rest of the bone marrow. In order to explain this[3], one may surmise specifically that the damage to stem cells causing BMF is mediated through a GPI-linked surface molecule, therefore in this case, the PNH cells lacking these molecules will survive. The very defect of the PNH clone, may thus endow it with a relative survival or growth advantage in a patient with BMF. If the patient has such a clone, he or she will present with PNH; otherwise he or she would present with overt AA. Since the PNH clone is often very florid, one can infer that the BMF affects selectively the non-PNH stem cells. While PNH is often regarded as a further ‘complication’ of AA[16], it is important to realize that a PNH clone can support substantially the patient’s haemopoiesis[20]. Furthermore, the existence of two or more PNH clones in the same patient can be seen as a case of convergent evolution in the population genetics of hemopoietic cells.

::   Treatment

The only definitive treatment for PNH is bone marrow transplantation (BMT). In cases in which bone marrow failure has progressed to the stage of qualifying for severe aplastic anaemia, and if an HLA-identical sibling is available, BMT must be regarded therefore as the treatment of choice[21]. In florid PNH without evidence of severe bone marrow failure, BMT has been attempted only in a handful of cases in which an identical twin was available and without myeloablative and immunosuppressive treatment: this type of BMT has not been successful. In view of the potentially life-threatening nature of the thrombotic complications of PNH, and in view of the currently improved results of conventional BMT from an HLA-identical sibling with full bone marrow ablation, this treatment must be now regarded as a valid option for all patients who have such a donor available and who are willing to accept the attending risks.

For all other patients with florid PNH, and for all those who do not have a potential donor, treatment must be supportive. Once the diagnosis is firmly established it is very important to explain to patients that they can live with PNH, perhaps for many years. Without raising hopes too high, it is fair to mention that adequate support may see them through to spontaneous recovery. Blood transfusion is imperative when exacerbation of haemolysis threatens life, but it should not be used on a regular schedule. Rather, the tolerance of the individual patient to anaemia should be assessed, as blood transfusion is indicated only when the haemoglobin level falls below the tolerated level. It is imperative to use on-line white cell filters for all transfusions. The use of this precaution has made it clear that the previously reported instances of haemoglobinuria triggered by blood transfusion, results in fact from white cell reactions activating complement rather than from an increase in the haematocrit as such. A neglected cause of worsening anaemia is iron deficiency consequent to urinary iron loss.

Any patient with PNH, who has experienced venous thrombosis, whether peripheral or hepatic, should be placed on prophylactic warfarin indefinitely, because there is serious risk of recurrence and venous thrombosis is one of the main causes of death. Although it may be regarded as perfectly rational to introduce warfarin in newly diagnosed patients even before they develop thrombosis, this has not been a recommended policy. In the treatment of hepatic venous thrombosis or Budd-Chiari syndrome, there is an immediate indication for thrombolytic therapy with tissue plasminogen activation. Surprisingly, this has been effective in some patients even as late as four weeks after the onset of symptoms[22].

Although the haemolysis is PNH is complement-mediated, there is no objective evidence that corticosteroids will decrease the rate of haemolysis and hence there is no rationale for their use in PNH. However, immunosuppressive treatment with anti-lymphocytic globulin and cyclosporine has been sometimes used successfully[23],[24].

References