Hematopoietic stem cell transplantation can cure high-risk acute leukemia (AL), but the occurrence of non-leukemic death is still high. The AIEOP conducted a prospective study in order to assess incidence and relationships of early toxicity and transplant-related mortality (TRM) in a pediatric population. Between 1990 and 1997 toxicities reported in eight organs (central nervous system, heart, lungs, liver, gut, kidneys, bladder, mucosa) were classified into three grades (mild, moderate, severe) and prospectively registered for 636 consecutive children who underwent autologous (216) or allogeneic (420) transplantation, either from an HLA compatible related (294), or alternative (126) donor in 13 AIEOP transplant centers. Overall, 47% of the patients are alive in CR (3-year EFS: 45.2%, s.e.: 2.1), 19% died in CR at a median of 60 days (90-day TRM: 14.3%, s.e.: 1.4), 34% relapsed. Toxicity of any organ, but mucosa and gut, was positively correlated with early death; moderate and severe toxicity to heart, lungs, liver and kidneys significantly increased early TRM, with estimated relative risks of 9.1, 5.5, 2.7 and 2.8, respectively, as compared to absent or mild toxicity. Patients with grade III-IV aGVHD experienced more than double (56% vs 19%) TRM than patients with grade 0-II aGVHD. A higher cumulative toxicity score, estimating the impact of toxicity on TRM, was significantly associated with transplantation from an alternative donor. Quantitative assessment allowed us to describe the extent to which 'grade' of toxicity and 'type' of involved organs were related to mortality and pre-transplant characteristics and yielded a prognostic score potentially useful to compare different conditioning regimens and predict probability of early death.

Human CMV infection is a frequent complication after HSC in children with remarkable morbidity and mortality. Antiviral drugs are relatively efficient but have numerous side effects. They are used as prophylactic, pre-emptive or therapeutic medicines. It is still a matter of debate which option is the best strategy. No uniform procedure has emerged regarding these three options, and new immunologic tools have raised more questions for physicians. To assess the current practice in the management of CMV infection, we sent a questionnaire to the EBMT centers performing hematopoietic SCT (HSCT) in children. Fifty-six out of 196 responded to the questionnaire (28.5%). Quantitative PCR was the most common monitoring tool (44/56). Only 4/56 centers use the pp65 antigenemia alone. All centers used pre-emptive strategy (56/56). 21/56 centers also used prophylactic measures, 13/21 after analysis of donor/receptor serologic status. Ganciclovir was the most common first-line agent for CMV disease (55/56). The most common dose and duration for induction treatment were 5mg/kg bid (47/55) for 14 days (20/55). There is no uniform procedure for researching resistance strain, antiviral second-line therapy or cell therapy. A harmonization process should enable sound prospective trials to improve prevention, control and cure of CMV disease in children and adolescents. =20

Autologous transplantation is a treatment option for relapsed childhood acute lymphoblastic leukemia (ALL) in second complete remission (CR2) when a suitable donor is not available. In an attempt to prevent relapses originating from graft leukemic contamination, the experimental protocol of in vitro purification of leukapheretic products with monoclonal antibodies (MoAbs), previously reported for adults, was adopted in 11 of 12 consecutive patients (median age, 9 years) with B cell precursor ALL in CR2 after late relapse (median, 37; range, 31-51 months after the onset) enrolled between July 1997 and July 1999 at a single pediatric center. At a median of 12 days after the mobilizing chemotherapy followed by G-CSF, a median of 13.9 (range, 5.9-18.7)x 10(6) CD34(+) cells/kg were collected from each patient and a median of 7.5 (range, 4.1-12.6) x 10(6) CD34(+) cells/kg underwent the purification procedure. The first step of immuno-rosetting allowed a one-log reduction of the total cell count, by eliminating more than 90% of the CD11b(+) cells; the second step, performed after incubation with anti-CD19 MoAbs, allowed the depletion of 99% (range, 93-100) of the CD19(+) cells, kept within the magnetic field of the immunodepletion column, with a median recovery of 73% (range, 55-87) of the collected CD34(+) cells. Molecular analysis assessed the in vitro eradication of detectable leukemic cells. A median reinfusion of 5.2 (range, 3.2-9.1) x 10(6) CD34(+) cells/kg for each patient (median viability, 90%), after conditioning with the 'TBI-VP16-CY' regimen, allowed prompt engraftment and immunological reconstitution; no patients experienced severe transplant-related toxicity or major infections. One patient relapsed 7 months after transplantation, while 10 patients are alive in clinical and molecular remission, at a median follow-up of 29 months (range, 15-40) (2-year EFS, 89%, s,e, 9), In conclusion, the procedure proved to be reproducible for pediatric purified autografting, highly efficient concerning stem cell recovery acid depletion of leukemia-lineage specific cells, and promising in terms of final outcome.

A constant improvement in the performance of blood cell separators has been observed in recent years, allowing better yields in peripheral blood stem cell collection (PBSC) either from healthy donors or for autologous purposes. Nevertheless, to our knowledge, no reports on the efficiency of mononuclear cell (MNC) collection in patients undergoing extra-corporeal photochemotherapy (ECP) for graft-vs.-host-disease (GvHD) have been published. We retrospectively investigated the efficiency of 167 MNC collections performed consecutively in 12 patients between January 1999 and June 2001 by means of the COBE Spectra version 4.7 (V 4.7) or version 6.0 (V 6.0), for 109 and 58 procedures, respectively. MNC fractional extraction (FE) was higher in the V 6.0 group compared to the V 4.7 group : 0.59 +/- 0.21 vs. 0.51 +/- 0.22 (P <0.05). However, platelet contamination was lower in the products obtained with V.6.0 compared to those obtained with V.4.7: 740 (630 x 10(3) /(L vs. 2,073 ( 1,429 x 10(3) /(L (P < 0.05). Only two patients with acute GvHD, both from V 4.7 group, required post-ECP platelet transfusion. The recently released version 6.0 allowed a satisfactory MNC yield with minimal platelet contamination in patients scheduled to undergo ECP for acute or chronic GvHD. (C) 2002 Wiley-Liss, Inc.