Stephen Rosenfeld

Paroxysmal Nocturnal Hemoglobinuria Cells in Patients with Bone Marrow Failure Syndromes

Aplastic anemia and paroxysmal nocturnal hemoglobinuria (PNH) are rare hematologic diseases that often appear in the same patient. Patients with aplastic anemia have severe thrombocytopenia, neutropenia, and anemia accompanied by absent hematopoietic precursors in an empty bone marrow (1). In contrast, the classic evidence of PNH is the intermittent appearance of dark urine due to excretion of hemoglobin, the result of intravascular hemolysis (2). The knowledge that this peculiar form of erythrocyte destruction resulted from increased susceptibility of the PNH erythrocyte to complement led to the development of laboratory assays, such as the Ham and sugar hemolysis tests. Modern clinical studies have shown that patients with PNH experience serious morbidity and mortality, mainly from venous thromboses and, especially in younger patients, pancytopenia (3). On the basis of results of the Ham test in several patients, Lewis and Dacie (4) formalized the overlap between the two diseases as the aplastic anemia-paroxysmal nocturnal hemoglobinuria syndrome. With improved survival in aplastic anemia, many patients show laboratory and clinical evidence of PNH, often months or years after completion of successful immunosuppressive therapy (5, 6). In aplastic anemia, hematopoietic cells appear to be destroyed by the patients own immune system (7). In contrast, the basis of PNH is a somatically acquired mutation in a hematopoietic stem cell. Almost all patients with PNH have molecular lesions in the PIG-A gene, which is located on the X chromosome (8). The PIG-A gene product is required at an early step in the synthesis of a glycosylphosphatidylinositol (GPI) structure, which serves as an anchor for a group of proteins that are linked to the cell surface by this greasy foot rather than by the more typical transmembrane configuration. As a result, affected cells are globally deficient in GPI-anchored proteins. Deficiency in one of these proteins, CD59, which inhibits late-acting complement component activity on the erythrocyte surface, accounts for the hemolytic component of the disease (9). The pathophysiologic basis of the clinical relation between PNH and aplastic anemia is unknown. Lewis and Dacie (4) postulated the development of an abnormal clone of haemopoietic cells in a regenerating, previously aplastic marrow as a cause, and Rotoli and Luzzatto (10, 11) speculated that an insult leading to a hypocellular environment or some feature of marrow failure might lead to either aplasia or PNH. Observations of patients with lymphoma who received monoclonal antibody treatment directed against a GPI-anchored protein have demonstrated the appearance of PNH-like lymphocytes as an inadvertent result of therapy (12). These results confirmed the in vivo plausibility of a third hypothetical mechanism: that preexisting PNH clones might be selected as a result of their relative insusceptibility to autoimmune attack (6, 10). To test these hypotheses, we used a sensitive flow cytometric method to analyze blood cells from patients in various states of marrow failure for evidence of PNH clones. Methods Patients We developed a rigorous three-color flow cytometric protocol for identification of GPI-anchored protein-deficient granulocyte populations and applied this method over 1.5 years to patients presenting to our clinic at the National Heart, Lung, and Blood Institute with newly diagnosed or previously treated bone marrow failure syndromes. Most of the patients with previously treated bone marrow syndromes had received therapy at the National Institutes of Health up to 11 years before sampling for this study. Samples from 254 participants were analyzed from April 1997 to October 1998. Of these participants, 115 had aplastic anemia, 39 had myelodysplasia, 28 had recently undergone bone marrow transplantation, 20 were controls who had undergone renal transplantation and had received antithymocyte globulin as treatment for graft rejection, 13 had large granular lymphocytosis, 18 had undergone multiple cycles of chemotherapy for cancer, and 21 were healthy controls. All clinical research samples were obtained after informed consent was given under protocols approved by the institutional review board of the National Heart, Lung, and Blood Institute. Other specimens were collected during routine phlebotomy procedures and were studied without unique patient identifiers. Aplastic anemia was defined as bone marrow cellularity of 30% or less and two of the following three laboratory abnormalities: absolute neutrophil count less than 0.5 109 cells/L, platelet count less than 20 109 cells/L, and reticulocyte count less than 60 109 cells/L. Myelodysplasia was defined by the standard morphologic criteria of either dysplastic myeloid or megakaryocytic (but not solely erythroid) bone marrow elements in patients who required 2 or more units of red blood cell transfusions per month for 2 or more months, with or without thrombocytopenia or neutropenia. Monoclonal Antibodies We used CD55-PE (clone 143-30, mouse IgG1 [Research Diagnosis, Inc., Flanders, New Jersey]) and CD59-PE (clone MEM 43, mouse IgG2a [Research Diagnosis, Inc.]) for analysis of GPI-anchored protein expression on erythrocytes. Glycophorin-A-FITC (clone D2.10, mouse IgG1 [Immunotech, Westbrook, Maine]) was used as a nonanchored marker to positively identify erythrocytes. The isotypic control for erythrocyte phenotyping consisted of mouse IgG1-PE (clone X40 [Becton Dickinson, San Diego, California]) and mouse IgG2a-PE (clone X39 [Becton Dickinson]). For analysis of granulocyte GPI-anchored protein expression, we used CD66b-FITC (clone 80H3, mouse IgG1 [Immunotech]) and CD16-PECy5 (clone 3G8, mouse IgG1 [Caltag]), with CD15-PE (clone 80H5, mouse IgM [Immunotech]) as a non-GPI-anchored marker to positively identify granulocytes. The isotypic controls for granulocyte staining consisted of mouse IgG1-FITC (clone X40 [Becton Dickinson]) and mouse IgG1-PECy5 (clone MOPC-21 [Caltag]). Nonspecific Fc receptor-mediated binding of conjugated antibodies to granulocytes or erythrocytes was blocked by pre-incubating 1 mL of blood with 30 L of mouse IgG (Caltag). Antibody Staining and Flow Cytometry Analysis Blood was drawn by venipuncture into tubes containing EDTA. Samples were stained within 48 hours of collection; staining was usually done within 8 hours. (Pilot studies revealed no significant change in cytofluorometric results in samples stored for up to 48 hours.) Erythrocyte GPI-anchored protein expression was evaluated by incubating for 30 minutes at room temperature 50 L of a 1:20 dilution of whole blood with 20 L of Gly-A-FITC and either 10 L each of CD55-PE plus CD59-PE or the appropriate isotype controls. Samples were washed and resuspended in 1 mL of phosphate-buffered saline before flow cytometry. Granulocyte GPI-anchored protein expression was evaluated by incubating for 60 minutes at room temperature 100 L of whole blood with 10 L of CD15-PE and either 20 L of CD66b-FITC plus 5 L of CD16-PECy5 or the appropriate isotype controls. Erythrocytes in these samples were lysed by using a Q-Prep apparatus (Coulter, Fullerton, California) and were then fixed with paraformaldehyde. All of the samples were analyzed by using a Coulter XL flow cytometry machine equipped with a 488-nm argon laser and XL or XLII software. Strict criteria to distinguish cells lacking GPI-anchored proteins were applied for two reasons. First, like many cell surface transmembrane proteins, GPI-anchored protein can vary in expression according to the stage of cellular differentiationfor example, as a result of myeloid or erythroid maturation. In addition, in some hematologic diseases, differentiation may be abnormal, as reflected in abnormal structure and aberrant expression of specific proteins. Therefore, polymorphonuclear cells were identified on the basis of light-scatter properties that correlate with cell size and granularity and by staining with a specific antigranulocyte antibody (anti-CD15, conjugated to the fluorescent dye phycoerythrin). Two antibodies directed against distinct GPI-anchored protein (anti-CD16, conjugated to the fluorochrome PE-Cy5, and anti-CD66b, conjugated to fluorescein isothiocyanate) were used to determine the PNH phenotype. Erythrocytes were similarly analyzed by using appropriate antibody combinations (anti-glycophorin for identification of erythrocytes and anti-CD59 and anti-CD55 as labels for GPI-anchored protein). Interpretation of Flow Cytometry Data The gates used to define CD16 /CD66b granulocyte or CD55 /CD59 erythrocyte populations were set on the basis of the isotypic control analyses performed on the same day. The mean SD of double negative cells for 21 healthy participants was 0.129% 0.101% for granulocytes and 0.195% 0.116% for erythrocytes. Because many patients required red blood cell transfusion at the time of initial sampling, the classification of patients was based on granulocyte analysis; a population of GPI-anchored protein-negative granulocytes of 1% or more that was clearly separate from the wild-type CD16+/CD66b+ cells on a two-dimensional histogram (Figure 1) was required for categorization of a patient as harboring PNH cells. Figure 1. Protocol for identification of paroxysmal nocturnal hemoglobinuria ( PNH) phenotype granulocytes. A. B. C. D. E. Response Criteria Response to treatment in aplastic anemia was defined as failure to fulfill criteria for severity for at least 3 months (see above); this was previously shown to correlate with transfusion independence and absence of infections. Criteria for response to therapy in myelodysplasia were no need for transfusion for 8 or more weeks, hemoglobin values of 80 g/L or more, platelet counts greater than 20 109 cells/L, and absolute neutrophil counts of 0.5 109 cells/L or more. Statistical Analysis A Fisher exact test was used to determine two-tailed P values in 2 2 contingency tables. The chi-square test was used

Late clonal diseases of treated aplastic anemia

Recent progress in the treatment of aplastic anemia has dramatically changed the previously grim prognosis for these patients. Improvements in bone marrow transplantation and immunosuppression have increased the number of long-term survivors so that immediate survival is no longer the sole concern. Here, we review the major clinical studies and summarize recent analyses of risk factors for developing paroxysmal nocturnal hemoglobinuria (PNH), myelodysplastic syndrome (MDS), acute leukemia, or solid tumor after treatment for aplastic anemia. We also examine biologic clues that may shed light on the interrelationship between aplastic anemia and clonal diseases.

High-dose cyclophosphamide in severe aplastic anaemia: A randomised trial

BACKGROUND High-dose cyclophosphamide has been proposed as an alternative immunosuppressive agent for treatment of severe aplastic anaemia, with a response rate similar to that with regimens containing antithymocyte globulin (ATG) but neither relapse nor clonal haematological complications. We undertook a phase III, prospective, randomised trial to compare response rates to immunosuppression with either high-dose cyclophosphamide plus cyclosporin or conventional immunosuppression with ATG plus cyclosporin in previously untreated patients. METHODS Between June, 1997, and March, 2000, 31 patients were enrolled. 15 were assigned cyclophosphamide (1 h intravenous infusion of 50 mg/kg daily for 4 days) and 16 were assigned ATG (40 mg/kg daily for 4 days); both groups received cyclosporin, initially at 12 mg/kg daily with adjustment to maintain concentrations at 200-400 microg/L, for 6 months. The primary endpoint was haematological response (no longer meeting criteria for severe aplastic anaemia). The trial was terminated prematurely after three early deaths in the cyclophosphamide group. Analyses were by intention to treat. FINDINGS Median follow-up was 21.9 months (range 1-33). There was excess morbidity in the cyclophosphamide group (invasive fungal infections, four cyclophosphamide vs no ATG patients; p=0.043) as well as excess early mortality (three deaths within the first 3 months cyclophosphamide vs no ATG patients; p=0.101). There was no significant difference at 6 months after treatment in the overall response rates among evaluable patients (six of 13 [46%] cyclophosphamide vs nine of 12 [75%] ATG). INTERPRETATION A longer period of observation will be necessary to assess the secondary endpoints of relapse and late clonal complications as well as disease-free and overall survival. However, cyclophosphamide seems a dangerous choice for treatment of this disorder, given the good results achievable with standard therapy.

Immunosuppressive treatment of aplastic anemia with antithymocyte globulin and cyclosporine.

Immunosuppression is the treatment modality for the majority of patients with aplastic anemia, most of whom are not candidates for allogeneic stem-cell transplantation. Antithymocyte globulin (ATG) or antilymphocyte globulin (ALG) have proven to be essential components of all regimens. Initial response rates can be improved by the addition of cyclosporine A (CsA), and this combination has become the standard of care for appropriate patients. Several new approaches to immunosuppression are being studied, including the optimal timing of administration of these drugs, the use of novel immunosuppressive agents, and the addition of early- and late-acting hematopoietic growth factors.

Unique region of the minor capsid protein of human parvovirus B19 is exposed on the virion surface.

Capsids of the B19 parvovirus are composed of major (VP2; 58 kD) and minor (VP1; 83 kD) structural proteins. These proteins are identical except for a unique 226 amino acid region at the amino terminus of VP1. Previous immunization studies with recombinant empty capsids have demonstrated that the presence of VP1 was required to elicit virus-neutralizing antibody activity. However, to date, neutralizing epitopes have been identified only on VP2. Crystallographic studies of a related parvovirus (canine parvovirus) suggested the unique amino-terminal portion of VP1 assumed an internal position within the viral capsid. To determine the position of VP1 in both empty capsids and virions, we expressed a fusion protein containing the unique region of VP1. Antisera raised to this protein recognized recombinant empty capsids containing VP1 and VP2, but not those containing VP2 alone, in an enzyme-linked immunosorbent assay. The antisera immunoprecipitated both recombinant empty capsids and human plasma-derived virions, and agglutinated the latter as shown by immune electron microscopy. The sera contained potent neutralizing activity for virus infectivity in vitro. These data indicate that a portion of the amino terminus of VP1 is located on the virion surface, and that this region contains intrinsic neutralizing determinants. The external location of the VP1-specific tail may provide a site for engineered heterologous epitope presentation in novel recombinant vaccines.

Platelet von Willebrand factor in Hermansky-Pudlak syndrome

The Hermansky‐Pudlak Syndrome (HPS) is an autosomal recessive inherited disorder characterized by oculocutaneous albinism, tissue accumulation of ceroid pigment, and a mild to moderate bleeding diathesis attributed to storage‐pool deficient (SPD) platlets. Patients have platelet aggregation and release abnormalities. In addition, low levels of plasma von Willebrand factor (vWF) antigen in some HPS patients have been associated with a greater bleeding tendency than would be predicted from either condition alone. Other HPS patients have severe bleeding despite normal levels of plasma vWF, suggesting that at least one additional factor is responsible for their bleeding diathesis. Because platelet vWF levels have been well correlated with clinical bleeding times in patients with von Willebrands disease, we have measured the platelet vWF activity and antigen levels in 30 HPS patients and have attempted to correlate their clinical bleeding with these values. The platelet vWF activity levels in patients was significantly lower than that of normal subjects (P < 0.0001). The patients as a group also had slightly lower values of plasma vWF activity when compared with normals (P ∼ 0.03). In 11 of the HPS patients, the multimeric structure of plasma vWF showed a decrease in the high molecular weight multimers and an increase in the low molecular weight multimers. In correlating the platelet and plasma vWF values with the bleeding histories, we were not able to show a predictable relationship in the majority of the patients. Am. J. Hematol. 59:115–120, 1998.

Important statistical considerations in classifier systems

The performance of a classifier system may be limited due to the following: (1) nonmonotonic relationships between individual predictor co-factors and outcomes, (2) prevalence imbalances between development data and application environment data, and (3) failure to account for cost-gain economics. These issues are explored, and statistically-based techniques for treating them are presented. In addition, probabilistic and fuzzy interpretations of classifier outputs are discussed, a likelihood ratio transformation of classifier outputs is suggested and two new cost-gain indexes that rate classifier systems in global economic terms are introduced.

Standard Web-based clinical radiology images and reports at the point of care

Web-based standard radiology images were successfully piloted and initially deployed to over 1300 users for on-sight intranet access. User feedback from the pilot yielded positive reports for ease of use, convenience and acceptable format. Technical deployment of thin client technology was found to be complex, time consuming and replaced with an alternate vendor strategy. One month implementation utilization data suggests moderate use of standard images. Recommendations yielded interest in pursuing Web access to enhanced digital radiology images and off-sight access.

Clinical research protocol mapping: a description of a pilot

Clinical and research activities are focused around protocols. A protocol map is a descriptive visual tool that outlines all the care events that occur during multiple encounters for a patient over the life of a protocol. The map, which is similar to a clinical pathway, also defines specific interdisciplinary standards of practice and key interventions that are required to facilitate research and patient outcomes. The inability to aggregate data, retrieve information or generate reports prompted the need to computerize the mapping process. The objectives of the protocol-mapping project were: (1) to test the concept of whether research protocol information mapped to a pathway could be automated using pilot software, and (2) to determine if an automated protocol map could be used to generate information to project resource utilization volumes. The project was divided into two phases. The implementation of the latter phase of the pilot study incorporated the activities of software development, the definition and mapping of pathways for research, the collection of clinical data and the generation of reports. Evaluation of the pilot software was based on the accomplishment of the project goals and objectives. The protocol map pilot software was successful in developing a robust tool for capturing protocol information and generating resource utilization data. The findings from this pilot software be used in developing requirements for a new clinical research information system that integrates clinical and research information needs.