ã
The Hematology Journal (2000) 1, 250 ± 253 2000 The European Haematology Association All rights reserved 1466 ± 4680/00 $15.00 www.nature.com/thj
Aplastic anemia and paroxysmal nocturnal hemoglobinuria: a follow-up study of the glycosylphosphatidylinositol-anchored proteins defect Maria-Elena Noguera1, Vincent Leymarie1, Enrice Bittencourt2, Eliane Gluckman2, FrancËois Sigaux1 and GeÂrard SocieÂ*,2 1
Laboratoire Central d'HeÂmatologie, HoÃpital Saint-Louis, Paris, France; 2Service d'HeÂmatologie Gree de MoeÈlle, HoÃpital Saint-Louis, Paris, France
Introduction: Flow cytometry analysis of peripheral blood cells is a simple and reliable method for establishing the diagnosis of paroxysmal nocturnal hemoglobinuria. The behavior of the clone may vary; increasing or diminishing over time but prospective study of such variations have not been reported so far. Materials and methods: We report herein the results of a prospective follow-up study of 25 patients. Our aims were twofold: ®rst, to evaluate the behavior of the clone (using ¯ow cytometry) over the time; and second, to evaluate if such variations could predict the occurrence of complications or could be used as a tool for monitoring the residual disease after bone marrow transplantation. Results: It was found that ¯ow cytometry can be used to speci®cally follow the residual disease post allogeneic marrow transplantation in four patients, and that even without transplantation the defective clone can signi®cantly decrease or even disappear (three patients). Conclusion: We found that most of the patients did have signi®cant change in the amount of aected cells during more than three years, and that an increased size of the clone poorly predicted the occurrence of complications. The Hematology Journal (2000) 1, 250 ± 253 Keywords:
¯ow cytometry; GPI; PIG-A; aplastic anemia; PNH
Introduction Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired clonal disorder of hematopoietic stem cells caused by somatic mutation in the X-linked PIG-A gene, which encodes proteins involved in the biosynthesis of the glycosylphosphatidylinositol (GPI) anchor, by which many proteins are attached to the membrane. Aplastic anemia (AA) and PNH are clearly clinically related, and are sometimes called the aplastic anemia ± PNH syndrome. Two patterns of evolution have been described. First, cases of PNH that present a progressive pancytopenia with marrow failure, and second, primary aplastic anemia in which signs of PNH clone is detected later. Today, ¯ow cytometry analysis of peripheral blood cells is a simple and reliable method for establishing the diagnosis of PNH. Furthermore, the analysis of GPI-linked proteins is more sensitive than classical PNH tests and can be *Correspondence: G SocieÂ, Service D'HeÂmatologie Gree de MoeÈlle, HoÃpital Saint-Louis, 1 Avenue Claude Vellefaux, 75475 Paris Cedex 10, France; Tel: +33 1 42 49 98 24; Fax: +33 1 42 49 98 24; E-mail:
[email protected] Received 28 February 2000; accepted 3 April 2000
performed even in multi-transfused patients. Granulocytes and monocytes appear to be the ®rst cells aected, at a time when the erythrocytes appear normal and the Ham test is negative (reviewed in1,2). The behavior of a PNH clone may vary; increasing or diminishing over time but prospective studies of such variations have not been reported so far. In addition, spontaneous clinical long-term remissions have even been observed in some patients with PNH.3 We thus undertook a prospective follow-up study of patients with a GPI-anchored protein (GPI-AP) defect with two aims: ®rst to evaluate the behavior of the PNH clone over time; and second, to evaluate if such variations could predict the occurrence of complications (such as thrombosis) or could be used as a tool for monitoring the residual disease after bone marrow transplantation.
Material and methods Patients As in previous studies patients were considered to have de novo PNH if they had a positive Ham's test at
GPI negative cells in PNH and AA ME Noguera et al
251
diagnosis. Twenty-®ve consecutive patients were included in this study. Among these 25 patients, 23 had a positive Ham's test during follow-up, but the detection of a GPI-AP-de®cient clone by ¯ow cytometry was taken as the only evidence for PNH in this study. The prospective follow-up study began in January 1995 and data were analyzed as of 1 October 1999. The mean duration of follow-up of these patients using ¯ow cytometry was 38 months. Among these 25 patients, 14 consecutive patients with acquired AA were included in this study. These 14 patients had acquired severe AA and were treated initially with immunosuppressive therapy according to an ongoing protocol mainly associating antithymocyte globulin and cyclosporin with or without granulocyte colony stimulating factor.6 The remaining 11 patients were diagnosed as having de novo PNH. These patients were treated initially by transfusion support alone or in association with cyclosporin or danatrole. Four of these patients (three with de novo PNH and one with an AA-PNH syndrome) underwent allogeneic marrow stem cell transplantation from an unrelated donor because of a worsening of the aplasia or an occurrence of thrombosis. The median age of these 25 patients was 34 years (range 19 ± 55), and included 12 males and 13 females. In patients with AA, the interval between the diagnosis of AA and the ®rst detection of a GPI-AP was 33 months. However, it should be underlined that six patients with AA were monitored initially only using the Ham's test and thus may have already had a GPIAP for a longer period. Roughly one third of the patients with otherwise typical AA (with a negative Ham's test) are found to have a GPI-AP defect at diagnosis (reviewed in1). During the period of followup ®ve patients developed thrombosis and two patients became pregnant (with one developing thrombosis after delivery). 4,5
Methods The expression of GPI-anchored molecules was studied on monocytes and on polymorphonuclear cells (PMN). Technical aspects of monocyte, PMN and labeling, and monoclonal antibodies used have been previously described.5 Since our ®rst study in 1995 only slight modi®cations of ¯ow cytometry have been introduced: we now use the CD66b antigen as a GPI-AP expressed on PMN. The following antigens were studied: CD66b and CD16 on PMN and CD14 on monocytes. The CD13 was used as a non GPI-AP molecule in double ¯uorescence staining. All antibodies were purchased by Immunotech1 (Dardilly, France) and data were analyzed with the Coulter-XL1 ¯ow cytometer (Paris, France). Each of the 25 patients were followed regularly in the outpatients clinic. Number of analyses per patient depended on the clinic status and evolution of the disease. As a mean, six analyses (range 3 ± 12) were conducted per patient (but some patients, such as those who became pregnant, had more frequent analyses).
Results Coherence of the dierent molecules to assess the variation of the GPI-AP defective clone We ®rst checked if results, for all patients included in this series, were coherent among the dierent antibodies used (CD16, CD66b, or CD14) and among the cell populations under study (ie PMN or monocytes). In fact comparison of the results in a given patient, and in the overall population, showed that there was a good correlation using either CD16 or CD66b for PMN and that these results were consistent with those obtained using the CD14 antibody in the study of monocytes. However slight variation in the percentage of GPI negative cells led us to consider as signi®cant any variation that was more than 10% between two time points.
In most patients the follow-up of the GPI-AP defective clone is not predictive of the clinical outcome Among the 25 patients, 15 showed no signi®cant variation in the amount of GPI negative cells within the 10% range as de®ned above. Three patients had a signi®cant increase in the amount of GPI negative cells. In two patients, one with AA and one with PNH, this was not correlated with any signi®cant clinical event nor with therapeutic modi®cation. The third patient with AA was in clinical remission and became pregnant (see below). Moreover, among these 15 patients, four had thrombosis that could not be predicted by a variation in the amount of GPI negative cells (tested one to three months (median two months) before the occurrence of thrombosis). However, it should be noted that a second patient who became pregnant and subsequently developed a venous thrombosis within the central nervous system had more than 90% of GPI negative cells and thus could not be evaluated for signi®cant increase in the size of the pool within the 10% con®dence interval range.
Evolution of the GPI-AP defective clone during pregnancy Two patients became pregnant and had repeated evaluation of the GPI-AP defective clone during pregnancy. One patient with PNH had prior to, and during pregnancy, more the 90% of GPI negative cells. During pregnancy there was no signi®cant modi®cation of the whole blood cell counts and delivery was uneventful. However, one month following delivery this patient developed a fatal venous thrombosis within the central nervous system. The other patient had been successfully treated with the association of antithymocyte globulin and cyclosporin for a severe AA. At diagnosis the percentage of GPI negative cells was in the 10% range. Thereafter, while she was in hematological complete remission, the size of the clone increased gradually with the 40% range. During The Hematology Journal
GPI negative cells in PNH and AA ME Noguera et al
252
pregnancy the clone remained stable, with no signi®cant modi®cation of the whole blood cell counts and delivery was uneventful. Three months post delivery the whole blood cell counts remained normal and the GPI-AP became undetectable (Figure 1, patient `L').
In some instances the GPI-AP defective clone signi®cantly decreased As shown in Figure 1, the GPI-AP defective clone decreased substantially in three patients. Patient `L' is described above. The two other patients had AA. One patient `W' had hematological remission but remained cyclosporin-dependent. Patient `Z' had a signi®cant decrease of the GPI-AP defective clone during relapse.
Follow-up study of the patients who underwent allogeneic marrow transplantation We recently published a series of patients who underwent allogeneic marrow transplantation from an HLA-identical sibling donor.7 In this series we reported that, following transplantation, the GPI negative cells rapidly disappeared. However none of these patients had been followed sequentially. In the present study, four additional grafted patients were included. All
Figure 1 Signi®cant decrease in the amount of GPI negative cells in three ungrafted patients.
Figure 2 GPI negative cells disappear soon after transplantation. The Hematology Journal
underwent allogeneic marrow transplantation from unrelated donors and had a prolonged history of PNH or AA before transplant. As shown in Figure 2, the rapid disappearance of the GPI negative cells following transplantation was con®rmed. This was correlated with the engraftment of donor-derived cells in chimerism study using PCR ampli®cation of variable number of tandem repeat sequences.7
Discussion Flow cytometry is the reference tool for the diagnosis of PNH.2 With widespread use of ¯ow cytometry it is now recognized that roughly one third of patients with otherwise typical AA have a GPI-AP defect.1 Since the early description of the GPI-AP defect in PNH and in the PNH-AA syndrome numerous reports have described the usefulness of ¯ow cytometry in assessing this defect.4,5,8 ± 12 However, none speci®cally looks at the dynamic behavior of the GPI-AP defective clone and the eventual predictive value of its variation in the occurrence of complications such as thrombosis. From our study the following conclusions could be drawn: (1) most of the patients showed signi®cant change in the amount of aected cells during the study period; (2) increased size of the clone poorly predicted the occurrence of complications; (3) ¯ow cytometry can be used to follow speci®cally the residual disease post allogeneic marrow transplantation; and (4) even without transplantation, the GPI-AP defective clone can signi®cantly decrease or even disappear in some patients. These results deserve some comments with regard to the conclusions drawn in other publications. First, other studies including larger numbers of patients will be needed to con®rm that the clone remains stable without signi®cant variation during an extended period of time in a given patient. However, given the rarity of the disease it is likely that only multi-institutional studies will be able to give a de®nitive response to this question. A similar approach will also be needed to know whether or not the increase in the size of the GPI-AP defective clone can predict the occurrence of complications in these patients. This is of primary clinical importance since thrombosis is the leading cause of death in these patients.13 Furthermore, we described recently that micro-particles of platelets origin and with pro-coagulant activity are highly elevated in patients with PNH.14 It is not known however, whether variations in the level of the GPI-AP defective clone can be correlated with the variations in the amount of micro-particles and subsequent risk of thrombosis. It is now clear from our own experience (this study and7) as well that of other authors15 ± 17 that ¯ow cytometry can be used to follow the residual disease after allogeneic marrow transplantation. It should be kept in mind, however, that the sensitivity limit of ¯ow cytometry in the detection of a GPI-AP clone is within the 5% range.
GPI negative cells in PNH and AA ME Noguera et al
253
Finally, the disappearance of the GPI-AP clone has been described by Hillmen et al.3 to occur in 15% of non-grafted long-term survivor patients with PNH and has been considered as evidence of spontaneous cure of the disease. While our data showed a signi®cant decrease in the amount of GPI negative cells in three out of 25 patients (12%) this decrease was associated with a complete clinical remission in only one patient. In conclusion, the use of ¯ow cytometry in following patients with PNH provides useful information
especially before and after transplantation, or in the case of signi®cant reduction of the GPI-AP defective clone. However, no clear correlation between the clinical outcome of the patients and the variations of the clone could be drawn. This would require a study including a larger number of patients during an extended period of time, ideally coming from multiinstitutional cooperation.
References 1 Socie G. Recent advances in paroxysmal nocturnal hemoglobinuria. From the biology to the clinic. Hematology and Cell Therapy 39: 175, 1997. 2 Rosse WF. Paroxysmal nocturnal hemoglobinuria as a molecular disease. Medicine 76: 63, 1997. 3 Hillmen P, Lewis SM, Bessler M, Luzzatto L, Dacie JV. Natural history of paroxysmal nocturnal hemoglobinuria. New England Journal of Medicine 333: 1253, 1995. 4 Vu T, Griscellibennaceur A, Gluckman E, Sigaux F, Carosella ED, Menier C, Scrobohaci ML, Socie G. Aplastic anaemia and paroxysmal nocturnal haemoglobinuria: A study of the GPI-anchored proteins on human platelets. British Journal of Haematology 93: 586, 1996. 5 Griscellibennaceur A, Gluckman E, Scrobohaci ML, Jonveaux P, Vu T, Bazarbachi A, Carosella ED, Sigaux F, Socie G. Aplastic anemia and paroxysmal nocturnal hemoglobinuria: Search for a pathogenetic link. Blood 85: 1354, 1995. 6 Bacigalupo A, Broccia G, Corda G, Arcese W, Carotenuto M, Gallamini A, Locatelli F, Mori PG, Saracco P, Todeschini G, Coser P, Iacopino P, Vanlint MT, Gluckman E. Antilymphocyte globulin, cyclosporin, and granulocyte colony-stimulating factor in patients with acquired severe aplastic anemia (SAA): A pilot study of the EBMT SAA Working Party. Blood 85: 1348, 1995. 7 Bemba M, Guardiola P, Garderet L, Devergie A, Ribaud P, Esperou H, Noguera MH, Gluckman E, Socie G. Bone marrow transplantation for paroxysmal nocturnal haemoglobinuria. British Journal of Haematology 105: 366, 1999 8 Schubert J, Vogt HG, Zielinskaskowronek M, Freund M, Kaltwasser JP, Hoelzer D, Schmidt RE. Development of the glycosylphosphatidylinositol-anchoring defect characteristic for paroxysmal nocturnal hemoglobinuria in patients with aplastic anemia. Blood 83: 2323, 1994. 9 Schubert J, Stroehmann A, Scholz C, Schmidt RE. Glycosylphosphatidylinositol (GPI)-anchored surface antigens in the allogeneic activation of T cells. Clinical and Experimental Immunology 102: 199, 1995.
10 Schrezenmeier H, Hertenstein B, Wagner B, Raghavachar A, Heimpel H. A pathogenetic link between aplastic anemia and paroxysmal nocturnal hemoglobinuria is suggested by a high frequency of aplastic anemia patients with a de®ciency of phosphatidylinositol glycan anchored proteins. Experimental Hematology 23: 181, 1995. 11 Tooze J, Saso R, Marsh JCW, Papadopoulos A, Pulford K, Gordon-Smith EC. The novel monoclonal antibody By114 helps detect the early emergence of a paroxysmal nocturnal hemoglobinuria clone in aplastic anemia. Experimental Hematology 23: 1484, 1995. 12 Yamaguchi M, Nakao S, Takamatsu H, Chuhjo T, Shiobara S, Matsuda T. Quality of hematologic recovery in patients with aplastic anemia following cyclosporine therapy. Experimental Hematology 23: 341, 1995. 13 Socie G, Mary JY, Degramont A, Rio B, Leporrier M, Rose C, Heudier P, Rochant H, Cahn JY, Gluckman E. Paroxysmal nocturnal haemoglobinuria: Long-term follow-up and prognostic factors. Lancet 348: 573, 1996. 14 Hugel B, Socie G, Vu T, Toti F, Gluckman E, Freyssinet JM, Scrobohaci ML. Elevated levels of circulating procoagulant microparticles in patients with paroxysmal nocturnal hemoglobinuria and aplastic anemia. Blood 93: 3451, 1999. 15 Endo M, Beatty PG, Vreeke TM, Wittwer CT, Singh SP, Parker CJ. Syngeneic bone marrow transplantation without conditioning in a patient with paroxysmal nocturnal hemoglobinuria: In vivo evidence that the mutant stem cells have a survival advantage. Blood 88: 742, 1996. 16 Graham ML, Rosse WF, Halperin EC, Miller CR, Ware RE. Resolution of Budd-Chiari syndrome following bone marrow transplantation for paroxysmal nocturnal haemoglobinuria. British Journal of Haematology 92: 707, 1996. 17 Kawahara K, Witherspoon RP, Storb R. Marrow transplantation for paroxysmal nocturnal hemoglobinuria. American Journal of Hematology 39: 283, 1992.
The Hematology Journal
