One of the most relevant concerns in long-term survivors of paediatric acute lymphoblastic leukaemia (ALL) is the development of neuropsychological sequelae. The majority of the published studies report on patients treated with chemotherapy and prophylactic central nervous system (CNS) irradiation, little is known about the outcome of patients treated with chemotherapy-only regimens. Using the standardised clinical and neuropsychological instruments of the SPOG Late Effects Study, the intellectual performance of 132 paediatric ALL patients treated with chemotherapy only was compared to that of 100 control patients surviving from diverse non-CNS solid tumours. As a group, ALL and solid tumour survivors showed normal and comparable intellectual performances (mean global IQ 104.6 in both groups). The percentage of patients in the borderline range (global IQ between 70 and 85) was comparable and not higher as expected (10% cases and 13% controls, expected 16%). Only 2 (2%) of the former ALL and 1 (1%) of the solid tumour patients were in the range of mental retardation (global IQ<70). Former known risk factors described in children treated with prophylactic CNS irradiation, like a younger age at diagnosis of ALL and female gender, remained valid in chemotherapy-only treated patients. The abandonment of prophylactic CNS irradiation and its replacement by a more intensive systemic and intrathecal chemotherapy led to a reduction, but not the disappearance of late neuropsychological sequelae.

Background. Although the demonstration of leptomeningeal dissemination is the most important predictor of poor outcome in children with medulloblastoma, there is lack of consensus on the prognostic value of a positive cerebrospinal fluid (CSF) cytology (i.e., stage M1).Patients and methods. Eighty-six pediatric medulloblastoma patients treated in Switzerland between 1972-1991 were retrospectively studied regarding the influence of M-stage on prognosis. 39 were M0, 13 M1, 15 Mx, 17 M2, and 2 M3.Results: Five- and 10-year overall survival rates were 76% and 54% for MO, 68% and 50% for Mr, 36% and 25% for M1, and 22% and 22% for M2-3 (P < 0.001), respectively. No significant survival differences were observed between M1 and M2-3 patients. Among 26 patients with only postoperative CSF cytologies, seven were positive. Their outcome was similar to that of six preoperatively staged M1 and significantly different from that of MO patients (P = 0.001). In 14 patients both pre and postoperative CSF cytology was performed. Total agreement was observed between the pre- and postoperative results (six positive and eight negative). Among the 19 M2-3 patients CSF cytology was positive in eight, negative in five, and unknown in six.Conclusions: A positive CSF cytology either pre- or postoperatively predicts for a poor outcome, similar to that observed in stage M2-3 patients. A postoperative cytology is likely to be concordant with cytologic results obtained preoperatively, and seems to have the same prognostic significance. A negative cytology however, does not exclude a more advanced stage.

Purpose In order to assess the potential correlation between clinical features and outcome 86 patients treated in Switzerland between 1972 and 1991 were retrospectively evaluated. Patients and Methods There were 66 males and 20 females with a median age at diagnosis of 8.2 years. All patients completed surgery and craniospinal axis irradiation, 62 patients with and 24 patients without additional chemotherapy. Resection was complete in 30 patients, subtotal in 42, and in 14 cases only a biopsy or partial resection was possible. Patients were staged following the Chang staging system (Chang et al. (1969). Radiology, 93, 1351-1359). T-stage: 24 T-2, 46 T-3, 13 T-4, and 3 T-x. M-stage: 39 M-0, 13 M-1, 17 M-2, 2 M-3, and 15 M-x. The following features were analyzed for correlation with survival: sex, age, performance status,T-stage, M-stage, resection, radiotherapy technical factors, and chemotherapy. Results M-stage at diagnosis and gender were the only clinical features independently correlated with survival.

In honour of Professor Rossi's 80th birthday we review the development of our understanding of the immune and auto-immune nature of the pathogenesis of immune thrombocytopenic purpura (ITP). The immune aspects have been documented by postviral alterations of the cellular and humoral immune system, by new methods of specific auto-antibody detection against platelet glycoproteins and by the therapeutic effect of administering immunoglobulin concentrate from healthy blood donors. The various possible mechanisms of action of immunoglobulin treatment have led to use of this treatment as an alternative for other immune-related disorders. The treatment of severe chronic ITP in children, however, remains unsatisfactory. With a new international clinical and laboratory study of children and adolescents with early chronic ITP we are continuing the investigation of the pathogenesis and treatment of ITP.

Acute megakaryoblastic leukemia (AMkL), defined by the presence of the platelet-associated glycoprotein IIb/IIIa complex on malignant cells, was diagnosed in 4 (4%) of 103 consecutive children with untreated acute leukemia or 4 (21%) of 19 children with acute nonlymphoblastic leukemia (ANLL). Particular features in the four children with AMkL were an age below 12 months at diagnosis (two patients), the absence of a significant hepatosplenomegaly (three patients), a leukocyte count below 20 x 10(9)/L with only a few blast cells in the peripheral blood (four patients), a technically difficult bone marrow aspiration (three patients), the presence of many megakaryocytes in marrow particles (two patients), and an inconclusive cytochemistry (four patients). The four children with AMkL were treated according to protocols for ANLL and a complete remission was obtained in all patients. One patient died from relapse after 3 months, one patient is a long-term survivor (38+ months), and two patients still on chemotherapy are disease-free for 11+ and 13+ months.

In children with acute lymphoblastic leukemia (ALL), megakaryocytopoiesis was investigated in vitro by the semisolid agar culture technique. In untreated ALL the median number of committed megakaryocyte progenitor cells (CFU-Mk) in the bone marrow was 2 (range less than 0.1-8) per 10(5) bone marrow cells instead of 30 (range 14-93) in controls, the impairment being dependent on the degree of leukemic bone marrow infiltration. However, if the number of CFU-Mk was related to residual nonleukemic bone marrow cells only, two-thirds of the children investigated had a frequency of CFU-Mk within the normal range. After 2 weeks of induction therapy the majority of the children had a low number of CFU-Mk in the bone marrow (median 6, range 0.5-40), a fact that could no longer be explained by a dilution of CFU-Mk by leukemic cells. After 4 weeks of chemotherapy (day 29) the frequency of CFU-Mk had risen to 18 (range 3-67) per 10(5) bone marrow cells, a value still significantly (p less than 0.01) below the normal range. In contrast to the changes in the number of CFU-Mk the median cell number per megakaryocyte colony remained constant during induction of remission. After cessation of long-term chemotherapy all children investigated had a normal number of CFU-Mk, suggesting that no permanent chemotherapy-related damage to committed megakaryocyte progenitor cells was induced.

Elimination of neoplastic B cell populations from autologous bone marrow grafts also removes normal B lymphocytes. This is potentially hazardous for the reconstitution of the immune system in patients undergoing high-dose chemotherapy and total body irradiation followed by autologous marrow rescue. Five pediatric patients with B cell non- Hodgkin's lymphoma in first remission undergoing such a regimen were studied. They received bone marrow pretreated with anti-Y 29/55 monoclonal antibody and complement. B and T lymphocyte subpopulations reached normal levels within 6 months after autologous bone marrow transplantation (ABMT), and serum immunoglobulin levels became normal within 4 to 9 months. Vaccination with diphtheria and tetanus toxoid, trivalent poliomyelitis vaccine of the Salk type, and pneumococcal capsular antigens (38 to 54 months after transplantation) gave rise to specific antibody production. ABO isoagglutinins could be demonstrated in all patients. The response pattern was similar to that of patients who received unmanipulated autologous bone marrow. It is concluded that ex vivo anti-Y 29/55 depletion of the marrow graft does not induce relevant disturbances of humoral immune functions.

In the human bone marrow the nuclear enzyme terminal deoxynucleotidyl transferase (TdT) is expressed by cells during early stages of lymphocyte differentiation. In order to investigate a possible regulation of lymphopoiesis at this level of differentiation, the relative frequency and the in vitro 3H-thymidine labeling index (3HdT-LI) of TdT-positive bone marrow cells were assessed in patients with different functional activities of the immune system. TdT-positive lymphoid precursor cells could be detected in the bone marrow of all children investigated, including six patients with various forms of immunodeficiency. Neither a transient hyperfunction of the immune system during the immunological rebound after cessation of long-term chemotherapy for acute lymphoblastic leukemia, nor a congenital or acquired hypofunction of the immune system had any detectable influence on the invariably high in vitro 3HdT-LI of TdT-positive bone marrow cells, a phenomenon possibly related to an autonomous and high turnover of this precursor cell compartment in the human bone marrow.

Autologous bone marrow transplantation was performed in 28 pediatric and adult patients with various neoplasias. Long-term remissions were obtained in one patient with yolk sac tumor and in 9 patients with B-cell non-Hodgkin's lymphoma. The relapse rate was decreased in patients receiving in-vitro decontaminated marrow (anti-Y 29/55 and complement).

In our i.v. IgG (Sandoglobulin ®) multicenter study for children with ITP unresponsive to previous treatment we observed some patients with an unsatisfactory long-term response to i.v. IgG [1]. They required more than one IgG infusion per month in order not to fall below 20 × 109/l platelets and/or not to show signs of hemorrhagic diathesis. Therefore, we proposed a trial combining i.v. IgG treatment and ciclosporin therapy for these children with such a severe form of chronic ITP.

Experiences gathered while exploring the usefulness of intravenous immunoglobulin for children with idiopathic (immune) thrombocytopenic purpura (ITP) are reviewed in view of further investigations characterizing the effects of IgG i.v. in other immune diseases without detectable antibody deficiency. The most pertinent factors to be considered are i) the heterogeneity of an immune disease investigated; ii) the criteria used to evaluate the effects of the IgG therapy; iii) the IgG preparations used and i.v. dose-fractionation. Controlled, prospective clinical trials will be required to further explore the practical usefulness of IgG i.v.

We studied the surface markers Leu 7, Leu 1, Leu 2a and Leu 3a on lymphocytes of 54 healthy blood donors and 19 patients with humoral immunodeficiencies and compared the relative numbers of positive cells with the natural killer cell (NK) activity. In controls, and in 12 out of 19 patients (group A) the percentage of the Leu 7+ cells was positively correlated with NK activity. Seven of 19 patients had increased relative numbers of Leu 7+ cells and low NK activity (group B). In controls and group A patients about 40% of the Leu 7+ cells simultaneously expressed the cell markers Leu 1 and Leu 2a. In group B patients approximately 70% of the Leu 7+ cells carried T cell antigens. Furthermore, a distinct suppressor T cell predominance was noticed and the large granular lymphocytes of these patients showed morphological abnormalities.

Malignant lymphocyte populations in peripheral blood of patients with B-cell chronic lymphocytic leukemia, leukemic variant of B-cell non-Hodgkin's lymphoma, and hairy cell leukemia can be characterized by the use of a monoclonal murine antibody (anti-Y 29/55) which is directed against a cell membrane component normally confined to the sessile nonrecirculating cells of the B-lymphocyte population in lymphoid tissues. The present report describes the reactivity of the anti-Y 29/55 antibody with bone marrow cells obtained from children with acute lymphoblastic leukemia using an indirect immunofluorescence method in combination with morphological and cytokinetic studies. In 25 patients (acute lymphoblastic leukemia subtype: 14 common; 4 pre-B-cell; 4 null; and 3 T-cell), a maximum of 2% of cells (small lymphocytes) were stained. One patient presented with blasts exhibiting cytoplasmic and surface immunoglobulin M (IgM) (pre-B-B-cell acute lymphoblastic leukemia). About 11% of this patient's blast cells showed a positive reaction with anti-Y 29/55. They could not be differentiated by morphological criteria from the anti-Y 29/55-negative blast cell population. In another patient with pre-B-B-cell acute lymphoblastic leukemia, only 1% of anti-Y 29/55-positive cells was found. In bone marrow of children with relative lymphocytosis, 1.4 to 8.7% of mononuclear cells reacted with anti-Y 29/55. Morphologically, these cells were small lymphocytes and predominantly expressed surface IgM. In two of these children, a further subdivision of bone marrow cells could be achieved by combining anti-Y 29/55 and cytoplasmic IgM reactivity with [3H]thymidine pulse labeling. These studies revealed that the actively proliferating, normal pre-B-cell population was anti-Y 29/55-nonreactive, whereas a nonproliferating population of anti-Y 29/55-reactive, cytoplasmic IgM-positive cells probably represented B-cells with surface immunoglobulin M reacting when cytoplasmic IgM was assessed. We conclude that the reactivity of the monoclonal anti-B-cell antibody (anti-Y 29/55) is restricted to surface immunoglobulin-positive bone marrow cells and that neither leukemic or normal pre-B-cells nor common, null-cell, or T-cell acute lymphoblastic leukemia blasts react.

The survival of 20 children with medulloblastoma who received adjuvant chemotherapy with procarbazine, vincristine and prednisone after resection and craniospinal irradiation is compared with the preliminary results of the Children's Cancer Study Group (CCSG) and the international Society of Pediatric Oncology (SIOP) medulloblastoma studies. Survival with the chemotherapy used in the SPOG study ws not superior to the survival of children who received craniospinal irradiation only.

A new immunoglobulin (Ig) for intravenous use was given in high doses to 4 children with refractory idiopathic thrombocytopenic purpura (ITP) and 2 children with idiopathic aplastic anemia (IAA). Within 5-10 days after initiation of Ig therapy the platelets of the children with ITP rose to 300,000-650,000/mm3 and could be maintained at normal levels with one Ig infusion every 1-3 weeks. No reaction was observed in the 2 patients with IAA.