Pediatric desmoid tumor: retrospective analysis of 63 cases.

PURPOSE: This study was conducted to evaluate clinical prognostic factors predictive of the probability of recurrence of desmoid tumor (DT). PATIENTS AND METHODS: Sixty-three patients with histologically confirmed diagnosis of DT were retrospectively studied. Median age at diagnosis was 13 years. Patient distribution by site was as follows: 61% extremities, 18% head and neck, 13% trunk (including 5% intraabdominal), and 8% multicentric lesions. All patients had partial or complete resections; 20 patients also received radiotherapy and/or chemotherapy. RESULTS: At a median follow-up time of 6 years since first treatment, the overall actuarial probability of having one or more recurrences was 75%. Age, sex, site, size, or number of previous recurrences had no significant impact on the likelihood of recurrence. The only factor associated with an increased proportion of recurrence-free patients was a negative margin of resection (70% v 15% with positive margins; P = .006). Of the four patients with more than 3 years follow-up since chemotherapy, two recurred, and of the 11 patients with the same follow-up after radiotherapy, four recurred, including two of five patients who received a dose of 50 Gy or more. No deaths directly related to tumor invasion were observed. CONCLUSION: A surgical approach aiming at clear margins is presently the best treatment option. When this cannot be accomplished without severe disfigurement or function impairment, partial resection is an acceptable alternative, but one associated with a high risk of regrowth. Whether adjuvant strategies should be used in this situation remains to be addressed.

In vivo cytoreduction studies and cell sorting--enhanced tumor-cell detection in high-risk neuroblastoma patients: implications for leukapheresis strategies.

PURPOSE: To improve autologous leukapheresis strategies in high-risk neuroblastoma (NB) patients with extensive bone marrow involvement at diagnosis. PATIENTS AND METHODS: Anti-G(D2) immunocytochemistry (sensitivity, 1 in 10(5) to 10(6) leukocytes) was used to evaluate blood and bone marrow disease at diagnosis and during the recovery phase of the first six chemotherapy cycles in 57 patients with stage 4 NB and bone marrow disease at diagnosis. A total of 42 leukapheresis samples from the same patients were evaluated with immunocytology, and in 24 of these patients, an anti-G(D2) immunomagnetic enrichment step was used to enhance tumor-cell detection. RESULTS: Tumor cytoreduction was much faster in blood compared with bone marrow (3.2 logs after the first cycle and 2.1 logs after the first two cycles, respectively). Bone marrow disease was often detectable throughout induction, with a trend to plateau after the fourth cycle. By direct anti-G(D2) immunocytology, a positive leukapheresis sample was obtained in 7% of patients after either the fifth or sixth cycle; when NB cell immunomagnetic enrichment was applied, 25% of patients had a positive leukapheresis sample (sensitivity, 1 in 10(7) to 10(8) leukocytes). CONCLUSION: Standard chemotherapy seems to deliver most of its in vivo purging effect within the first four cycles. In patients with overt marrow disease at diagnosis, postponing hematopoietic stem-cell collection beyond this point may not be justified. Tumor-cell clearance in blood seems to be quite rapid, and earlier collections via peripheral-blood leukapheresis might be feasible. Immunomagnetically enhanced NB cell detection can be highly sensitive and can indicate whether ex vivo purging should be considered.

High-dose chemotherapy with autologous marrow rescue for malignant brain tumors: analysis of the impact of prior chemotherapy and cranio-spinal irradiation on hematopoietic recovery.

The relative impact of age, sex, nucleated cell dose, prior chemotherapy, prior cranio-spinal irradiation (CSI) and bone marrow harvest (BMH) site on hematological recovery after ABMT were analyzed in a multivariate model. The study population comprised 100 patients with a median age of 9 years who underwent ABMT for malignant brain tumors. Two engraftment parameters were evaluated: number of days post ABMT before (1) an absolute neutrophil count (ANC) > or = 0.5 x 10(9)/l and (2) a platelet count > or = 50 x 10(9)/I were achieved for the third consecutive day without transfusions. Increasing cell dose correlated significantly with a more prompt recovery of platelet counts and ANC. Previous chemotherapy significantly delayed both neutrophil and platelet engraftment. The group of patients who also received CSI had a very delayed platelet recovery with a median time to engraftment of 72 days. Neutrophil engraftment was also significantly delayed and occurred at a median of 23 days. This effect of CSI was independent of cell dose or prior chemotherapy. In 20 of these patients, marrow was harvested at least partially from the posterior iliac crests, which might have received significant doses of irradiation. We conclude that engraftment is significantly faster if bone marrow is harvested prior to any chemotherapy administration, and that patients who receive prior CSI may have significant engraftment delay, particularly of the platelet lineage. In this latter group of patients, marrow should not be harvested from the posterior iliac crests. Strategies that might enhance both neutrophil and platelet count recovery should be considered in patients with irradiation damage to a substantial proportion of the total hematopoietic tissue.

G-CSF serum pharmacokinetics during peripheral blood progenitor cell mobilization: neutrophil count-adjusted dosage might potentially improve mobilization and be more cost-effective.

The optimal dosing schedule of G-CSF for peripheral blood progenitor cell (PBPC) mobilization is still under investigation although many centers use 10 microg/kg/day in a single subcutaneous dose. However, G-CSF clearance increases with increasing absolute neutrophil count (ANC). Hence a G-CSF dosage adjusted to ANC might be a reasonable approach. We measured G-CSF trough serum levels by sandwich ELISA assay at different ANCs in eight patients undergoing treatment with filgrastim at 10 microg/kg/day in a single subcutaneous dose. A total of 26 samples were analyzed, and a strong correlation between increasing ANC and decreasing G-CSF levels was found by linear regression analysis (P < 0.0003, r2 = 0.4199). For ANC values above 5000/microl the trough serum levels, ie 24 h after administration, were consistently below the level that provides maximal clonogenic precursor stimulation in vitro (10 ng/ml). Serial serum G-CSF measurements performed in three patients at 0, 3, 6, 9 and 24 h after G-CSF administration, showed a reduction of the area under the curve (AUC) with increasing ANC. For an ANC of 20000/microl or greater, the G-CSF serum level fell under the maximal in vitro stimulation threshold of 10 ng/ml within 12 h. This preliminary pharmacokinetic data seems to suggest that an ANC-adjusted G-CSF dosing schedule might improve the design of PBPC mobilization regimens.

[The identification of minimal residual disease in the bone marrow and peripheral blood in neuroblastoma. The prognostic and therapeutic implications]. / Identificazione della malattia residua minima nel midollo e nel sangue periferico nel neuroblastoma. Implicazioni prognostiche e terapeutiche.

A highly sensitive and specific methodology to detect neuroblastoma cells in the bone marrow and peripheral blood of children with neuroblastoma is of critical importance for proper staging and treatment of these patients. In addition, patients with bone marrow infiltration at diagnosis need to undergo regular investigation to measure the effectiveness of chemotherapy (so called "in vivo" purging). Finally, the evaluation of autologous stem cells taken from bone marrow or peripheral blood is necessary to rule out or minimise the possibility of reinfusing tumor cells to the patient following myeloablative therapy. The authors provide a "state of the art" data on this complicated issue and give their preliminary results of their own experience, mainly concerning the immunocytological methods.

Cerebellar involvement in hypomelanosis of Ito.

Hypomelanosis of Ito (HI) (incontinentia pigmenti achromians) with cerebellar atrophy and dysmorphic features is reported in a child. The association of cerebellar anomalies and HI has been previously reported in only four cases. However, since neuroimaging studies are not routinely obtained in these patients the frequency of this association might have been underestimated. In cases of otherwise unexplained cerebellar ataxia an examination of the skin with Wood's lamp should be performed since the typical hypopigmented lesions can be hardly visible in natural light.

Pediatric peripheral blood progenitor cell collection: haemonetics MCS 3P versus COBE Spectra versus Fresenius AS104.

BACKGROUND: An increasing number of apheresis machines are becoming available for peripheral blood progenitor cell (PBPC) collection in children. STUDY DESIGN AND METHODS: At the Children's Hospital of Florence (Italy), three apheresis machines were evaluated: MCS 3P (Haemonetics) (10 procedures in 4 patients, aged 10-12 years, weight 23.5-64 kg), Spectra, (COBE) (8 procedures in 3 patients, aged 4-17 years, weight 19-59 kg), and AS104 (Fresenius) (24 procedures in 9 patients, aged 2-16 years, weight 13.6-60 kg). For PBPC quantitative analysis, CD34 cytofluorimetry was employed. Relevant variables analyzed included efficiency of CD34+ cell extraction and enrichment, mononuclear cell purity and red cell contamination of the apheresis components, and platelet count decreases after leukapheresis. RESULTS: No significant differences in CD34+ cell-extraction abilities were found. However, the AS104 provided consistently purer leukapheresis components in terms of mononuclear cell and CD34+ cell enrichment (441 +/- 59%, vs. 240 +/- 35% and 290 +/- 42% for MCS 3P and Spectra, respectively). Postapheresis platelet counts dropped the least with the AS104. The smallest patient who underwent apheresis with MCS 3P (the only machine working on discontinuous flow and hence with greater volume shifts) weighed 23.5 kg and tolerated the procedure well, with no signs of hemodynamic instability. No significant complications were observed. CONCLUSION: All machines seem to have comparable PBPC extraction efficiency, but the AS104 seems to give the component with the greatest PBPC enrichment. This feature might be relevant for further ex vivo cell processing (CD34+ cell selection, expansion, and so on).

High-sensitivity immunocytologic analysis of neuroblastoma cells in paired blood and marrow samples.

High-sensitivity immunocytochemistry was used to evaluate the relative frequency of neuroblastoma cells in bone marrow and peripheral blood in patients with neuroblastoma (NB). A total of 51 concomitant paired blood and marrow samples (102 total) from 35 patients with NB (age 4 months-31 years; stage 29 stage IV, 4 stage III, 2 stage IVS; 14 at diagnosis, 18 in relapse, 12 during treatment, and 7 off-therapy) were analyzed. Cytospins containing up to 10(6) cells each were prepared using the mononuclear cell (MNC) fraction. For immunocytologic staining, a primary mouse monoclonal anti-GD2 antibody (3F8), a secondary antimouse biotinylated antibody, and a streptavidin-alkaline phosphatase complex were used. A minimum of two cytospins containing a mean of 1.4 x 10(6) total MNCs was analyzed in addition to a negative and a positive control. No circulating tumor cells were detected when the concomitant marrow samples were negative or had <10 positive cells per 106 MNC (23 of 51 samples). Of the 18 marrow samples positive at 10-10,000 cells per 106 MNC, 6 had detectable NB cells in the corresponding blood sample, whereas for marrow samples with >10,000 NB cells per 10(6) MNC (1%), the concomitant blood sample was positive for 9 of the 10. When both marrow and blood samples were positive (15 BM-PB pairs), NB cell frequency was significantly lower in blood, with a mean difference of 2.14 logs (median 2.22, range -0.16-4.8, standard error 0.38). In patients with NB, circulating tumor cell frequencies seem to be substantially lower than in concomitant marrow samples, with a mean difference of >2 logs.

Mutations in the thrombopoietin receptor, Mpl, in children with congenital amegakaryocytic thrombocytopenia.

Congenital amegakaryocytic thrombocytopenia (CAMT) is a rare disorder of undefined aetiology. The disease presents with severe thrombocytopenia and absence of megakaryocytes in the bone marrow. Furthermore, CAMT patients may develop bone marrow aplasia. To obtain more insight into the mechanism underlying CAMT, five children were analysed. All patients had increased plasma thrombopoietin (Tpo) levels, indicating a platelet production defect. Bone marrow-derived CD34+ stem cells from three patients were cultured in an in vitro liquid culture system to study megakaryocytopoiesis. CD34+ cells from two of the three patients failed to differentiate into megakaryocytes. The lack of megakaryocyte formation could imply that a defect in the c-mpl gene, encoding the Tpo receptor, exists. Sequencing of c-mpl revealed mutations in four of five patients. Three patients had point mutations and/or a deletion in the coding regions of c-mpl. All point mutations led to an amino acid substitution or to a premature stop codon. In one patient, a homozygous mutation in the last base of intron 10 was found that resulted in loss of a splice site. This study showed that mutations in c-mpl could be the cause of thrombocytopenia in CAMT in the majority of patients. Furthermore, Tpo has been shown to have an anti-apoptotic effect on stem cells. Therefore, mutations in c-mpl might not only affect megakaryocyte formation but may also impair stem cell survival, which could explain the occurrence of bone marrow failure as final outcome in patients with CAMT.

Treatment of Diamond-Blackfan anemia with recombinant human interleukin-3.

This report describes the response of eighteen Diamond-Blackfan anemia (DBA) patients to recombinant human interleukin-3 (rhIL-3). rhIL-3 was administered subcutaneously once daily on an escalating dose schedule (0.5 to 10 micrograms/kg/d). The rhIL-3 dose was escalated every 21 days until erythroid response was attained, grade III or IV nonhematologic toxicity was observed, or the maximum rhIL-3 dose was reached. Four patients experienced clinically significant erythroid responses. Two of the responders were steroid-dependent and transfusion-independent, while two were steroid-independent and transfusion-dependent. Baseline clinical or laboratory parameters, in particular in vitro bone marrow erythroid progenitor assays, were not useful in predicting rhIL-3 response. rhIL-3 administered at 5 to 10 micrograms/kg/d was associated with an increase in total white blood cell count, secondary to increases in neutrophils, eosinophils, and lymphocytes. Patients experienced a dose-dependent elevation in absolute eosinophils across the entire dose range. Two of the responding patients remain on maintenance rhIL-3, without diminution of effect at 244 and 370 + days. rhIL-3 was discontinued in the other two responders, because of the development of deep venous thrombi.

Successful treatment of Diamond-Blackfan anemia with interleukin 3.

This report describes the response of 18 Diamond-Blackfan anemia (DBA) patients to recombinant human interleukin 3 (rhIL-3). rhIL-3 was administered s.c. once daily on an escalating dose schedule (0.5-10 micrograms/kg/day). The rhIL-3 dose was escalated every 21 days until erythroid response was attained, grade III or IV nonhematologic toxicity was observed, or the maximal rhIL-3 dose was reached. Four patients experienced clinically significant erythroid responses. Two of the responders were steroid-dependent and transfusion-independent, while two were steroid-independent and transfusion-dependent. Baseline clinical or laboratory parameters, in particular in vitro bone marrow erythroid progenitor assays, were not useful in predicting rhIL-3 response. Two of the responding patients remain on maintenance rhIL-3 without diminution of effect at 490 and 855+ days. rhIL-3 was discontinued in the other two responders because of the development of deep venous thrombi.