Search

The PML (for 'ProMyelocytic Leukemia') gene product is a nuclear zinc finger protein, identified when the chromosomal translocation fusing this gene to the retinoic acid receptor was found in acute promyelocytic leukemia. Recently, a frequent occurrence of autoantibodies against the PML protein was detected in primary biliary cirrhosis (PBC) sera, suggesting that this protein could represent an autoantigenic trigger in PBC. Chronic GVHD features are close to those of PBC and in addition, antinuclear and antinucleolar antibodies are frequently detected in patients' sera. In order to determine if an abnormal expression of PML, followed by the development of anti-PML antibodies, can be implicated in chronic GVHD pathogenesis, we studied the expression of PML in the skin of seven patients with chronic GVHD as well as the presence of circulating anti-PML antibodies. PML was highly expressed by the lesional skin keratinocytes, but circulating antibodies were never detected. PML is induced by interferon (IFN) gamma. The expression of PML by GVHD epidermis is likely secondary to the IFN gamma produced by infiltrating lymphocytes. Since PML display growth suppressor properties, the role of this protein in tissue lesions is discussed.

Aplastic anemia is a rare disease in children that is most commonly idiopathic and less often a hereditary disorder. Hereditary bone marrow failure (BMF) syndromes, however, should be considered both in children and in adults before any attempt at treatment. Precise diagnosis is important because it will modify prognostic treatment options and the results of bone marrow transplantation. In this review, we will report recent results of treatment of Fanconi anemia and other hereditary BMF syndromes.

Over the past decade, allogeneic cord blood transplantation (CBT) has been widely used for treating patients with malignant disorders. However, the reported low incidence of GVHD observed after allogeneic CBT might be a major drawback in leukemic recipients and at present it is not clear whether CBT really predisposes patients to an increased risk of leukemia relapse. In order to further elucidate the role of CBT in children with hematological malignancies, 54 patients with ALL or AML given either a related (31 cases) or an unrelated (23 cases) CBT in 25 centers participating in the Eurocord Registry were analyzed. Overall survival of related and unrelated CBT recipients was substantially similar, the most important factor influencing patients' outcome being disease state at time of transplantation. In fact, due to a markedly increased relapse rate, poor-risk children (ie patients transplanted in more advanced disease) experienced a significantly worse EFS than those given CBT in a more favorable disease phase (ie CR1 or CR2). These data confirm that allogeneic CBT from either a related or an unrelated donor is a feasible procedure able to cure a significant proportion of children with acute leukemia, especially if transplanted in a favorable phase of disease.

Umbilical cord blood T cells are less functional. Different explanations have been proposed. In this study we analyze the Vbeta T cell cord blood repertoire. All the Vbeta families are expressed. We found only the overexpression of three Vbeta: Vbeta 5-1, Vbeta 5-2, Vbeta 21-2.

Eurocord Transplant has established a registry for studying results of cord blood transplant. We have analyzed 78 patients who have received a related CBT between October 1988 and December 1996. The median follow-up time was 29 months (1-99). The median age was 5 years (0.2-20), median weight 19 kg (5-50). Forty-six patients had a malignant disease: 32 acute leukemia (AL), six chronic myeloid leukemia (CML), four myelodysplastic syndrome, two neuroblastoma and two non-Hodgkin lymphoma. Thirty-two patients were transplanted for non-malignant diseases including 17 bone marrow failure syndromes (BMFS), three sickle cell anemia, five thalassemia and seven inborn errors. The donor was an HLA-identical sibling in 60 cases and an HLA-mismatched donor in 18 cases. As conditioning, 36 patients received irradiation and 40 patients received associated busulfan-containing regimens. GVHD prophylaxis consisted of CsA alone in 36 cases, CsA associated with prednisone in eight cases, CsA, methotrexate (Mtx) with or without prednisone in 28 cases and CsA with monoclonal antibody or ATG in four cases. The median number of nucleated cells (NC) infused/kg was 3.9 x 10(7) (0.7-15). One-year survival was 63 +/- 6%. Age, weight, HLA identity and negative cytomegalovirus (CMV) serology in the recipient were significant favorable prognostic factors. Among these 78 patients, the incidence of gradeor = II GVHD was 9% in HLA-matched CBT and 50% in mismatched CBT (P0.001). Neutrophil engraftment was associated with age6 years (P = 0.02) and weight20 kg (P = 0.02). It was 73% in patients receiving3.7 x 10(7) nucleated cells (NC) infused/kg and 85% in patients receiving more (P = 0.06). Favorable factors for platelets engraftment were age6 years (P = 0.03), weight20 kg (P = 0.002) and HLA identity (P0.0001). Related cord blood transplantation offers a good alternative to BMT. Theses results are in favor of freezing cord blood in families in whom a transplant might be indicated.

Patients with beta-thalassemia (Hbeta th) and sickle cell anemia (SCA) can be treated with bone marrow transplantation. Stem cells from cord blood have several theoretical advantages, however, the place of cord blood transplant for hemoglobinopathies has not yet been established. We report here the EUROCORD experience of 10 patients (Hbeta th = 7, SCA = 3) transplanted with related cord blood.

This report summarizes the results of unrelated cord blood transplants. Sixty-five patients were reported to Eurocord. One year survival was 29%. Factors associated with survival were diagnosis, inborn errors had a better survival than leukemia or aplastic anemia; higher number of cells infused and negative pre-transplant CMV serology were also favorable factors. Granulocyte and platelet engraftment were associated with a higher number of nucleated cells infused, GVH incidence was low especially in CMV negative patients, it was not associated with the degree of HLA disparity. These results were similar to other series reported in the USA and showed the benefit of using mismatched unrelated cord blood in clearly defined situations.

Fanconi anaemia is a hereditary disorder characterised by chromosomal breaks increased by cross-linking agents. Bone marrow transplantation is the treatment of choice when a HLA identical sibling donor has been identified. The use of low-dose cyclophosphamide with thoraco-abdominal irradiation for the conditioning regimen of FA patients has lead to a dramatic improvement of survival, with a long-term survival of 75% at our institution. However, if most patients are completely cured of their haematological disease, there is concern about an increased frequency of secondary tumours, mostly head and neck squamous cell carcinomas of poor prognosis. Results of BMT using alternative donors (HLA mismatched related and unrelated donors) have also improved during the last decade. A better selection of the donor via high-resolution techniques for class-II HLA matching, and more recently the use of T cell depleted grafts are probably the main explanations. Despite a short follow-up and the small number of patients analysed, transplants using HLA matched family cord blood give some promising results. On the other hand, first results with unrelated cord blood remind that this approach is clearly an experimental one that has to be evaluated through international registries and prospective studies. New approaches including autologous stem cell transplantations and gene therapy are currently explored.

One hundred and ninety-one patients with acute leukaemia who received bone marrow from HLA-A, -B and -DR identical unrelated donors and were reported to EBMT and/or IMUST, were matched with 382 patients receiving autologous bone marrow for diagnosis, age, stage of disease and year of transplantation. Transplant-related mortality (TRM) was significantly higher in recipients of unrelated marrow compared to autograft recipients, 44 +/- 4% (+/- 95% confidence interval) and 15 +/- 3% at 2 years in the two groups, respectively (P10(-4)). In contrast, relapse probability was lower in recipients of unrelated marrow, being 32 +/- 5% at 2 years compared to 55 +/- 3% in recipients of autografts (P10(-4)). Two-year leukaemia-free survival (LFS) in patients with acute lymphoblastic leukaemia was 39 +/- 5% and 32 +/- 3% in the two groups, respectively. Among patients with acute myeloid leukaemia (AML), the corresponding figures were 36 +/- 6% and 46 +/- 5% in the two groups, respectively (P = NS). In AML in first remission (CR-1), the 2-year survival was 42 +/- 10% in recipients of unrelated bone marrow, compared to 69 +/- 8% in autograft recipients (P = 0.008). When all patients with acute leukaemia were included, the 2-year LFS was 38% in recipients of unrelated marrow, compared to 37% in autograft recipients (NS). In conclusion, this retrospective analysis supports the design of a prospective randomized study in patients with high-risk/advanced acute leukaemia who lack a suitable related bone marrow donor, to ascertain which of the two strategies, if any, should be favoured.

Allogeneic BMT for patients with acute myeloid leukemia (AML) is presently a reference therapy. The indications for this therapy mainly rely upon prognostic factors, and their importance is constantly reassessed. To examine the impact of time from diagnosis to transplant on survival and leukemia-free survival (LFS), we analyzed 109 patients from the database of the SFGM comprising patients who had all received an HLA-identical allogeneic BMT for a diagnosis of AML in first complete remission (CR1) between January 1987 and December 1992. All patients were conditioned with cyclophosphamide (CY) and total body irradiation (TBI) (CYTBI), and methotrexate (MTX) + cyclosporin A (CsA) were used as graft-versus-host disease (GVHD) prophylaxis. Patient characteristics were: age = 33 +/- 9, M/F = 64/45, white blood cell count (WBC) at diagnosis = 27 +/- 42 x 10(9)/l, FAB distribution: M1 and M2 = 55; M3 = 15, M4 and M5 = 33, M0, M6 and M7 = 6. Karyotyping was carried out for 64 patients: 32 had a normal karyotype, 16 had good prognosis abnormalities (t(8;21), t(15;17), inv 16) and 16 patients had other abnormalities. Eleven patients needed two courses of induction to achieve CR. Time between diagnosis and BMT was 120 (64-287) days. Forty-nine patients developed gradeor = 2 acute GVHD (actuarial probability = 46%). With a median follow-up of 50 months (27-100), the 5-year probabilities for transplant-related mortality (TRM), relapse, overall survival and LFS are respectively 25%, 26%, 59% and 55%. A multivariate analysis showed that survival is adversely influenced by three independent factors: time to transplant (120 days vsor = 120 days), acute GVHD (grade 2-4 vs grade 0-1) and age (33 vsor = 33). LFS is only influenced by the first two of these factors. The favorable impact of a shorter time from diagnosis to transplant should lead to performing the transplant as early as possible. Practically speaking, this means that when such therapy is chosen for a patient with CR1 AML, the search for an allogeneic donor should begin immediately and transplant be performed as soon as possible.

Recipients of HLA-matched, related or unrelated allogeneic BMT who were CMV seropositive or those receiving unmanipulated marrow from a seropositive donor were randomised to receive one of three treatment regimens, i.v. acyclovir 500 mg/m2 three times a day from 5 days before transplant to 30 days after transplant followed by oral acyclovir 800 mg four times a day for a further 6 months, i.v. acyclovir followed by placebo, or 400 mg oral acyclovir four times a day followed by placebo (control). This paper reports the 1 year data on the same cohort of patients which was previously reported. Intravenous acyclovir (i.v./PCB) significantly reduced the risk of CMV infection when compared to the control group. The frequency of adverse events reported was comparable among the three groups. The mortality rate was significantly reduced by the sequential use of i.v. acyclovir followed by oral acyclovir, resulting in a 19% survival advantage at 1 year from transplant.

A total of 1634 recipients of HLA-identical sibling bone marrow with acute leukemia were treated with the combination of cyclosporin A (CsA) and methotrexate as prophylaxis against graft-versus-host disease (GVHD). The probability of relapse decreased with an increasing grade of acute GVHD, especially in patients grafted in first remission (CR-1): P0.01 and P0.001 for acute lymphoblastic leukemia and acute myeloid leukemia, respectively. Among patients surviving at least 3 months without a relapse, chronic GVHD was associated with a decreased incidence of relapse in CR-1 (P0.0001 for both diagnoses), and a better LFS. Among patients in CR-1, the probability of relapse was the same in those with limited or with extensive chronic GVHD. However, in patients with intermediate stage of the disease (or = 2nd remission or 1st relapse) with previous acute GVHD, those with extensive chronic GVHD had a reduced probability of relapse (P = 0.05). The graft-versus-leukemia effect of acute and especially chronic GVHD was confirmed by multivariate analyses. Overall, the highest LFS was seen in patients with chronic GVHD and no or grades I-II acute GVHD. The lowest LFS was seen in patients having acute GVHD grades III-IV without chronic GVHD.

We investigated in 15 consecutive patients a possible correlation between expression of CMV or EBV in labial salivary gland (LSG) biopsies performed 100 days after allogeneic BMT and subsequent development of chronic GVHD. Three techniques were performed for the detection of each virus: immunohistochemistry, in situ hybridization and PCR. Eleven patients developed chronic GVHD. Histologic examination detected a moderate lymphoid infiltrate (grade 1 according to Sale's score) in LSG biopsy in only one patient. CMV genes or proteins could not be detected in any patients. Likewise, EBV genome or proteins were not detected by in situ hybridization and immunohistochemistry. However, in three of the 15 patients, EBV DNA was detected by PCR in LSG biopsies. Only one of these three patients developed chronic GVHD. Therefore, at the present time, the presence of a lymphoid infiltrate on lip biopsies performed at day 100 post-BMT does not appear to be sensitive enough for the diagnosis or the prediction of the subsequent development of chronic GVHD. Moreover, the absence of EBV and CMV expression in a day-100 LSG biopsy does not preclude the development of chronic GVHD.

To evaluate whether the results of bone marrow transplantation have improved in Europe with time, we analyzed the outcome for 2195 patients with acute leukemia. 1405 had acute myeloid leukemia (AML) and 790 had acute lymphoblastic leukemia (ALL), and were allografted in first complete remission between September 1979 and December 1991 with marrow from an HLA-identical sibling donor. We found a continuing improvement more evident since 1987 for AML and since 1986 for ALL. A substantial reduction in the 3 years transplant related mortality (TRM): 26 vs 39% for AML (P = 10(-4)), and 25 vs 39% for ALL (P = 10(-4)), has resulted in an increase of the 5-year actuarial leukemia-free survival (LFS). 57 vs 45% for AML (P10(-4)) and 54 vs 45% (P = 10(-4)) for ALL. Four important changes have occurred. (1) Graft-versus-host disease (GVHD) prevention has involved an increased use of cyclosporin A (CsA) alone and subsequently its use in combination with methotrexate: this was associated with lower TRM both in AML and ALL; (2) Use of total body irradiation as pretransplant regimen has decreased; (3) a shorter interval from remission to BMT is more common; (4) an older population of patients has undergone BMT. Multivariate analyses were performed separately in AML and ALL. In AML four variables significantly influenced TRM favorably: year of BMT (P = 10(-4)), younger age at BMT (P = 10(-4)), prevention of GVHD including CsA (P = 0.008), sex match other than female donor to male recipient (P = 0.002). The relapse incidence (RI) was lower in patients with FAB M1-2-3 vs M4-5 (P = 0.0004). The LFS improved by year of BMT (P = 0.0004), younger age at BMT (P = 10(-4)), prevention of GVHD including CsA (P = 0.01), FAB M1-2-3 (P = 0.03). In ALL, three variables were associated with a lower TRM: year of BMT (P = 10(-4)), younger age at BMT (P = 10(-4)), sex combination other than female to male (P = 0.008). The LFS was better after 1986 (P = 0.0004) and in younger patients (P = 10(-4)). However a better outcome after 1986/87 was observed in patients receiving the same GVHD prophylaxis: therefore, other unidentified factors resulting in better patient care have also contributed to this. The improved results of allogeneic BMT are entirely related to a reduction in TRM without loss of the antileukemic effect since relapse incidence has not changed over the years.

From 1984 to 1991, 514 patients were treated by BMT in 1 center. 254 patients survived more than 3 months and, in 38 patients, 47 liver biopsies were performed for chronic liver dysfunction characterized by cholestasis. The aim of the present study was to evaluate the possible causes of liver disease at the time of biopsy. One clinician analyzed clinical data and was able to propose up to 3 diagnoses including GVHD, viral hepatitis, drug-related hepatitis, chronic veno-occlusive disease (VOD) or other. Two pathologists reviewed histologic sections and were also able to propose up to 3 diagnoses. Clinically, 1, 2 or 3 diagnoses were proposed in 30, 60 and 10% of cases, respectively. Pathologically, 1, 2 or 3 diagnoses were proposed in 13, 62 and 25%, respectively. Histologic changes of GVHD were present in 40 of 47 biopsies and concordance between the clinician and the pathologists on the presence of GVHD lesions was found in 77% of biopsies. Viral hepatitis was proposed 22 times by the clinician and 19 times by pathologists. Viral hepatitis, usually hepatitis C, was associated with GVHD in 16 cases. Diagnoses of chronic VOD and drug-related hepatitis were proposed less often. In summary, more than 1 diagnosis was suggested for many of the patients studied, GVHD being the most frequent. The simultaneous presence of GVHD, viral diseases, chronic VOD and drug-induced diseases could explain the high incidence of cholestasis in the long-term post-BMT.

Six hundred and eighteen patients with acquired aplastic anaemia grafted from an HLA-identical sibling donor between 1976 and 1990 in eight European centres were reported to the Working Party for Severe Aplastic Anaemia (SAA) Registry and were evaluable for analysis of the incidence of graft failure/rejection and the outcome of second bone marrow transplants (BMT). The number of patients experiencing graft rejection declined significantly over the study period from 32% to 8% (p0.0001). This coincided with the introduction of cyclosporine to the conditioning regimen for BMT. The graft rejection rate in the post-hepatitis SAA group was significantly lower than in the group with idiopathic SAA (4% vs 20%) (p = 0.001). The use of irradiation in the conditioning regimen significantly reduced the number of patients experiencing graft rejection (7% vs 21%) (p = 0.004). Age, sex and severity of disease did not influence the rate of sustained engraftment. Of the 85 patients experiencing graft rejection, 41 received a second transplant: their survival is 33% vs 8% for patients not transplanted a second time (p = 0.003). The major factor predicting the outcome of second BMT for SAA was the interval from first BMT. Patients receiving a second BMT within 60 days from the first BMT had a significantly poorer outcome.

The International Marrow Unrelated Search and Transplant (IMUST) Study has prospectively assessed outcome of unrelated donor BMT (UD-BMT) in comparison with a matched cohort of patients treated by HLA-identical sibling BMT (ID-BMT). We report an interim analysis of the first 165 UD-BMT and 368 ID-BMT. Eighty-two percent of UD-BMT pairs were matched by serology for HLA-A, B and DR, 10% were less well matched and HLA matching data was incomplete in 8%. The Kaplan-Meier estimated probability of survival until day 400 was 0.42 (95% confidence limits 0.33-0.51) after UD-BMT and 0.62 (0.56-0.68) after ID-BMT (p =0.001). Probability of engraftment by day 100 was 0.90 (0.85-0.95) and 0.95-0.99) after UD-BMT and ID-BMT, respectively (p =0.001). Cumulative probability of acute GVHD by day 100 was 0.52 (0.45-0.60) and 0.42 (0.37-0.47) after UD-BMT and ID-BMT, respectively (p = 0.009). After UD-BMT, 52% of patients with early disease survived until day 400 (40-64%) and 27% (14-40%) with advanced disease (p =0.001). Multifactorial analysis of survival showed success was related to the centre's experience of UD-BMT and this effect was modified by conditioning protocol. Increased probability of survival after UD-BMT in centres with most experience of the procedure is novel finding. We conclude UD-BMT is a more difficult procedure than ID-BMT but results are acceptable in patients with early disease and when UD-BMT is carried out in experienced centres.

Of 322 patients undergoing allogeneic BMT, 18 developed invasive pulmonary aspergillosis at a mean of 115 days post-transplant, with a mortality rate of 82%. Pulmonary localization was common but cerebral involvement was seen in 10 of 18 patients. The diagnosis was made ante mortem in 11 patients by direct examination of pathological samples or culture and A. fumigatus was the only species isolated. Specific antibodies were not demonstrated before or at the time of clinical symptoms and Aspergillus antigen was only seen in one patient a few days before death.

Human umbilical cord blood contains enough hematopoietic stem cells for transplantation instead of marrow cells to cure children and adults with leukemia and other potentially fatal marrow disorders. An international cord blood transplantation registry is currently collecting information about the results of these transplants. As a result of relative immune deficiency at birth, it is likely that partially matched unrelated transplants can be successful; for this purpose an international cord blood banking project is described.

Human umbilical cord blood was evaluated as an alternative to bone marrow as a source of stem cells for hematopoietic transplantation. In order to define an optimal collection procedure, we have studied the parameters that influence the collection, handling and storage of cord blood. We have attempted to correlate the quality of the samples with obstetrical and neonatal parameters. Using culture techniques we have studied the long term viability of the cells. Cell separation was also investigated. Our results suggest that most of the sample collected could be suitable for transplantation, in term of progenitors. However the observed variability between samples suggest that an efficient control of the quality of the samples is important.

27 cord blood samples from healthy newborns were processed according to a "whole blood" flow cytometric analysis. The CD3-positive T cells are characterized by their variability: 44.8 +/- 13.3% of lymphocytes with a lower expression of the gamma delta T cell receptor. The majority of the CD3+ cells are CD38+. Newborn T cells have less ability than adult T cells to express IL-2 receptors as well as HLA-DR. The CD4-positive T cells are equal to 31.0 +/- 10.8% of lymphocytes with a great prevalence of the CD4+/CD45RA+ population. The CD3+/CD8+/CD11b+ cells are increased to 23.4 +/- 7.1% of lymphocytes. The CD57 antigen is not expressed. The NK population, CD16+/CD56+ is increased to 25 +/- 11% of lymphocytes. 68% of CD19+ cord blood B lymphocytes coexpressed CD5. Thus the "suppressive" and "naïve" cells are prominently represented in cord blood.

Umbilical cord blood collected and cryopreserved at birth contains enough hematopoietic progenitor stem cells for engraftment. HLA identical sibling cord blood transplant has been performed for the first time, in a child with Fanconi anemia. Three years latter, this child is alive with a complete donor type bone marrow. Since this first attempt, several other patients with other diseases have been transplanted successfully. Cord blood banking is a safe and easy procedure. Due to the high proliferative capacity of neonatal hematopoietic progenitors and to the relative immunological functional immaturity of neonatal lymphocytes cord blood cells could be used for matched unrelated or partially mismatched transplants.

A total of 304 children under the age of 15 years with acquired severe aplastic anemia (SAA) received immunosuppressive therapy (IS) (n = 133) or a matched bone marrow transplant (BMT) (n = 171). The projected 10-year survival is 48% and 63% respectively (p = 0.002). Results following BMT have improved considerably over the years from 49% in 1970-80, to 70% in 1981-83 (p = 0.002) and to 81% between 1984-88 (p = 0.08). Other favorable prognostic factors are the use of cyclosporin A (p = 0.004), no previous therapy (p = 0.006) and early BMT (p = 0.009). In multivariate analysis only the year of treatment proved significant (p = 0.02). In contrast, results of IS are greatly dependent on the severity of pre-treatment neutropenia with survival of 56% versus 37% for neutrophils more or less than 0.2 x 10(9)/l (p = 0.003). Poor survival was associated in univariate analysis with female sex (43%), post-hepatitis SAA (37%), children not receiving androgens (38%) and patients younger than 5 years (35%), especially if associated with a low neutrophil count (11%). In multivariate analysis only the degree of neutropenia proved significant (p = 0.005). These results suggest that IS is a satisfactory alternative therapy for children with moderately SAA in the absence of an HLA-identical sibling, although BMT remains the treatment of choice. In children under 5 years with very SAA, results with IS are so poor that a search for an unrelated matched donor is justified as early as possible.

The SAA Registry of the EBMT now contains data on 171 children younger than 15 years of age with acquired SAA and undergoing BMT between 1970 and 1988. The overall actuarial survival is 63% at 10 years. In a multivariate Cox analysis, the year of transplant was the most important prognostic factor with a significant advantage for children grafted in 1984-88 (81%) vs 1981-83 (67%) and 1970-80 (41%) (p = 0.02). Cyclosporine A given for GVHD prophylaxis, no treatment before transplant and an interval less than 90 days from diagnosis to BMT were all favourable variables in univariate analysis. As regard to transplant procedures, the better results were obtained using Cyclophosphamide and Cyclosporine A (78%) followed by Cyclophosphamide plus irradiation plus Cyclosporine A (77%). Sex, etiology and the severity of the aplasia had no impact on survival in both uni and multivariate analysis.

Four patients developed legionnaires' disease after bone marrow transplantation. Two cases occurred early after transplant and were considered as part of a hospital epidemic due to contamination of water supply. The other two cases were considered to be sporadic because they occurred 3-4 weeks after hospital discharge. The outcome was good in two patients. In the third patient, recurrent disease was probably due to acquired resistance to macrolides, and complete cure was achieved after treatment with pefloxacin and rifampicin. The fourth patient died of overwhelming infection despite early treatment with erythromycin and pefloxacin. During the same period we treated 14 patients with pefloxacin for prevention of bacterial infection, of whom none developed Legionella pneumophila infection, while three of the patients reported here were in a group of 11 patients who received only oral non-absorbable antibiotics for gut decontamination. The fourth patient in this report was receiving no antibiotics. Thus pefloxacin seems to be effective as prophylaxis against L. pneumophila infection. When the hospital water supply was heated to 60 degrees C and chlorinated, the nosocomial cases in the hospital completely disappeared.

Chronic graft-versus-host disease (GVHD) is a very heterogeneous entity with variable clinical expression. The pathophysiology involves autoimmune disorders and features of fibrosis. Four patients transplanted for severe aplastic anaemia conditioned with cyclophosphamide and thoraco-abdominal irradiation developed skin scleroderma specifically localized in the irradiation fields. This syndrome was remarkable for its late onset and the absence of factors known to trigger chronic GVHD. It is postulated that irradiation used in this protocol may induce keratinocyte damage and trigger chronic GVHD.

Thirty-eight patients with haematological malignancies were treated with bone marrow transplantation using histocompatible immunotoxin T cell-depleted marrow siblings. All patients received conventional postgraft immunosuppression (methotrexate and/or cyclosporin A). Donor bone marrow was treated ex vivo with T101 Fab fragment coupled to ricin A-chain (T101 Fab-RTA) at a concentration of 10(-8) M of A-chain in association with NH4Cl (2 x 10(-2) M) in pH adjusted (7.8) incubation medium. A median cytoreduction of 99.5% (91-99.5) was obtained. The median of follow-up was 300 days. Only three patients developed grade II acute graft-versus-host disease (GVHD) (actuarial rate of acute GVHD: 9.1%). No chronic GVHD occurred. All patients but one engrafted. Six out of the 37 patients developed a documented bone marrow rejection (actuarial rate of graft failure: 18%). Ten patients relapsed (actuarial rate of relapse: 36.9%). These findings demonstrate that treatment of donor marrow with T101 Fab-RTA in association with NH4Cl at critical pH value can achieve a high level of mature T cell depletion and greatly reduce the incidence of bone marrow rejection and relapse after T cell-depleted allogeneic bone marrow transplantation.

Systematic clinical and serological studies to evaluate the frequency of toxoplasmosis in bone marrow transplant recipients were performed in 80 consecutive patients. Antitoxoplasma antibody titres were measured in donors and recipients before transplant and subsequently post-transplant. Before bone marrow transplant, 54 recipients were seropositive and 26 were seronegative, whereas 35 donors were seropositive and 45 were seronegative. After bone marrow transplant, the frequency of clinical and serological manifestations of toxoplasmosis appeared closely related to the recipient's serological status before transplant. In the seronegative group of patients before transplant the incidence of toxoplasmosis was low: only two patients experienced seroconversion 3 months after bone marrow transplant and one developed clinical symptoms consistent with toxoplasmosis but without cerebral involvement. Clinical toxoplasmosis or secondary elevation of antibody titres was mostly observed in pre-bone marrow transplant seropositive patients; in this group, cerebral toxoplasmosis occurred in four patients and a significant secondary rise of antibody titres was observed in 16 patients. It thus appears that toxoplasmosis is most often related to a reactivation of latent cysts. Prophylactic treatment may be useful in patients presenting serological evidence of past or latent infection before bone marrow transplant.

Treatment with GM-CSF or G-CSF is becoming widely used in patients with chronic neutropenia, or who are aplastic following chemotherapy or autologous or allogeneic bone marrow transplantation. Recently, some authors have described a phenomenon analogous to cyclic agranulocytosis following treatment with G-CSF in a patient with chronic neutropenia. We wish to describe the same phenomenon in a patient with chronic granulocytic leukemia who received GM-CSF (Sandoz) after T cell depletion in order to accelerate hematological reconstitution.

Data were obtained from 46 patients with severe aplastic anemia (SAA) who received bone marrow transplants (BMT) from donors other than genotypically HLA-identical siblings. The data were collected in the SAA Registry of the European Bone Marrow Transplant Group. The donors were non-HLA-identical siblings in six cases, parents in 28 cases, a son in one case and unrelated individuals in 11 cases. Fifteen donor-recipient pairs were HLA-A, -B and -DR identical and mutually non-reactive in mixed lymphocyte culture; nine were mismatched at one locus, 17 were mismatched at two or more loci and in five cases data were not available for D/DR determinants. Actuarial survival was predicted by the degree of mismatch. It was 45% for phenotypically HLA-identical grafts, 25% for grafts mismatched at one locus and 11% for graft mismatched at more than one locus. Whether the graft was derived from a family member or an unrelated donor seemed to be less important and results were comparable. Age, patient sex and year of transplant had no significant influence on survival. The use of cyclosporine (CSA) for graft-versus-host disease (GVHD) prophylaxis (n = 21, survival 34%) appeared superior to both methotrexate (n = 9, survival 11%) and to CSA with T cell depletion of donor marrow (n = 13, survival 14%). The causes of death were rejection (n = 15), GVHD (n = 13), pneumonitis (n = 5) and infection (n = 1). Twelve patients are alive at 16-84 months post-BMT.(ABSTRACT TRUNCATED AT 250 WORDS)