High leucovorin doses during high-dose methotrexate treatment may reduce the cure rate in childhood acute lymphoblastic leukemia.

We explored the relationship between time to relapse and different exposure variables (serum methotrexate (S-MTX) 23, 36 and 42 h after start of administration, MTX elimination time and leucovorin (LV) dose) during high-dose MTX (HDM) treatment of 445 children with acute lymphoblastic leukemia. MTX was infused at 5 g/m2 (non-high risk) or 8 g/m2 (high risk) over 24 h, 2-9 times per patient. LV rescue dose was adjusted according to the S-MTX concentration. Time from end of the last HDM to relapse was analyzed by Cox regression analysis with the logarithms of S-MTX and LV dose as exposures. The combined results from all risk groups suggest that high LV dose is related to higher risk for relapse. Doubling of the LV dose increased the relapse risk by 22% (95% confidence interval 1-49%, P = 0.037). High LV doses correlated with high MTX levels at 23, 36 and 42 h and longer elimination time. The results suggest that high doses of LV increase the risk for relapse despite the fact that they were correlated with high MTX levels and longer MTX elimination time. The choice of MTX and LV doses may be regarded as an intricate balance between effect and counter-effect.

Cellular drug sensitivity in MLL-rearranged childhood acute leukaemia is correlated to partner genes and cell lineage.

Rearrangements in the 11q23 region, the site of the mixed lineage leukaemia (MLL) gene, are found in both childhood acute myeloid (AML) and lymphoblastic (ALL) leukaemia. We studied the in vitro drug resistance by the fluorometric microculture cytotoxicity assay (FMCA) in 132 children with AML and 178 children with ALL (aged 0-17 years). In AML, children with t(9;11) (n = 10) were significantly more sensitive to cytarabine (P < 0.001) and doxorubicin (P = 0.005) than non-11q23 rearranged patients (n = 108). Children with other 11q23 rearrangements (n = 14) differed less from non-rearranged children. The 'AML-profile' common to all three groups included relative resistance to glucocorticoids and vincristine. In ALL, children with 11q23 rearrangement (n = 22) were significantly more sensitive to cytarabine (P = 0.026) than children without 11q23 rearrangement (n = 156), also after stratification for white blood cell count. In conclusion, the findings indicate that the cellular drug resistance is correlated to both the cell lineage and the type of 11q23 rearrangement. High cellular sensitivity to cytarabine and doxorubicin might explain the excellent treatment results in children with AML and t(9;11). The present study supports the strategy of contemporary protocols to include high-dose cytarabine in the treatment of 11q23-positive patients both in AML and ALL.

Loss at 12p detected by comparative genomic hybridization (CGH): association with TEL-AML1 fusion and favorable prognostic features in childhood acute lymphoblastic leukemia (ALL). A multi-institutional study.

BACKGROUND: Genetic aberrations provide prognostic information in childhood ALL. The proportion of patients with detectable aberrations can be increased by combining G-banding with comparative genomic hybridization (CGH). PROCEDURE: We studied 79 children with ALL by CGH and G-banding, and explored the relationship of these findings to clinical features and outcome. RESULTS: CGH revealed DNA copy number changes in 57 patients (72%), 9 of whom had normal karyotype by G-banding. Gains were more frequent than losses, and changes of whole chromosomes more frequent than partial aberrations. Two frequent partial losses were found; at 9p and 12p. The 9 patients with loss at 12p were studied for the deletion of TEL (ETV6) gene and the fusion of TEL and AML1 genes by fluorescent in situ hybridization (FISH). Eight out of the 9 children with loss at 12p harbored the TEL-AML1 translocation and all 9 had the deletion of a nontranslocated TEL allele. All 9 had precursor-B phenotype and L1 morphology, and 8/9 had WBC below 50 x 10(9)/liter. All children were treated according to Nordic ALL protocols, had a good response to treatment based on day 15 bone marrow morphology, and 7 out of the 9 survived in continuous complete remission (median follow-up 74 months). CONCLUSIONS: CGH is a valuable tool in screening for genetic aberrations in childhood ALL. DNA copy number losses detected at 12p associate with TEL-AML1 fusion as well as with favorable prognostic features.

Initial P-glycoprotein expression in childhood acute lymphoblastic leukemia: no evidence of prognostic impact in follow-up.

Treatment results in childhood acute lymphoblastic leukemia (ALL) have improved remarkably during the past 20 years, but still 25% of children cannot be permanently cured. Drug resistance is a major cause of poor outcome. One of the most investigated resistance mechanisms is the P-glycoprotein (P-gp)-mediated multiple-drug resistance (MDR). The authors prospectively analyzed P-gp using flow cytometry with monoclonal antibody JSB1 in a population-based series of 103 children with ALL treated according to intensive Nordic ALL protocols. Increased P-gp expression was detected in 55 patients (53%). With a cutoff value of 1% P-gp-positive blasts in bone marrow, no difference was found in event-free survival (EFS) or overall survival between children with low vs. increased P-gp expression. The 4-year EFS in the whole series was 77%. Patients with T-ALL had higher P-gp levels than the others, 3.6% vs. 1.0% (p = .002). P-gp expression did not correlate with the white blood cell count, age, sex, or cytogenetics. The authors conclude that the level of P-gp expression cannot be used as a tool for treatment stratification in childhood ALL.

AML1 gene amplification: a novel finding in childhood acute lymphoblastic leukemia.

BACKGROUND AND OBJECTIVE: We previously found a high-level amplification in chromosomal region 21q22 in two children with acute lymphoblastic leukemia (ALL) using comparative genomic hybridization. The same region harbors the AML1 gene. The aim of the present study was to investigate whether AML1 is a target gene in these amplifications. DESIGN AND METHODS: Bone marrow samples were obtained from 112 childhood ALL patients. The copy number of AML1 was studied using fluorescent in situ hybridization with a dual color DNA probe specific for the AML1 and TEL genes. RESULTS: Three of the patients had 3-to-8 fold amplification of AML1 and showed a high-level amplification of 21q22 by comparative genomic hybridization. In two of them the extra copies were shown to be located tandemly in a derivative of chromosome 21. Thirty-seven of the patients (33%) had 1-to-2 extra copies of AML1, most probably reflecting the incidence of trisomy 21 and tetrasomy 21. The TEL-AML1 fusion was less frequent in the patients with extra copies of AML1 (7/40; 18%) than in the patients with no extra copy (24/72; 33%). None of the three patients with 3-to-8 fold amplification of AML1 showed the fusion or loss of TEL. INTERPRETATION AND CONCLUSIONS: Our findings suggest that the AML1 gene is a target gene in the 21q22 amplicon in childhood ALL. To understand the role, if any, of the AML1 amplification in leukemogenesis, further studies are needed.

Reemphasis on lymphoblast L2 morphology as a poor prognostic factor in childhood acute lymphoblastic leukemia.

BACKGROUND: In childhood acute lymphoblastic leukemia (ALL), the relationship between lymphoblast L1/L2 morphology and prognosis is controversial. According to some studies L2 morphology is associated with poor prognosis, whereas in others the association disappears after adjustment for other known risk factors. PROCEDURE: We investigated the prognostic importance of lymphoblast L1/L2 morphology in childhood ALL treated with current Nordic ALL protocols in Finland. From the routine bone marrow (BM) aspirate and biopsy slides of 251 children with ALL diagnosed in 1990-1995, the blast cell morphology and early treatment responses were assessed blindly in a central review, using French-American-British (FAB) criteria with the Children's Cancer Group (CCG) modification. RESULTS: L1 morphology (>90% L1) was found in 197 (80%) children and L2 (>/=10% L2) in 49 (20%). Early treatment response was poorer in L2 than in L1: >5% blasts in the marrow on day 15 were seen in 27% of L2 as opposed to 12% of L1 (P = 0.048). The 6-year event-free survival (EFS) in the study population was 75%, 76% in L1 and 70% in L2 (P = 0.34). In the group with white blood cell count (WBC) below 50 x 10(9)/liter at diagnosis, the L2 morphology was associated with inferior survival: 6-year EFS 74% in L2 and 84% in L1 (P = 0.07), with 6-year overall survival (OS) 81% vs. 91% (P = 0.035), respectively. L2 morphology was not associated with any other adverse prognostic factor analyzed. CONCLUSIONS: With the intensive Nordic ALL protocols, lymphoblast L2 morphology is an independent poor prognostic factor, influencing both the early response to treatment and, in the low-WBC group, the ultimate outcome, and should be reemphasized in risk categorization of childhood ALL.

Comparative genomic hybridization and conventional cytogenetic analyses in childhood acute myeloid leukemia.

Comparative genomic hybridization (CGH) analysis was performed on bone marrow specimens from 19 children with acute myeloid leukemia (AML) at diagnosis. The results of CGH were compared to those of conventional cytogenetic analysis. The most common CGH aberrations were gains of whole chromosomes 6 and 8, both of which appeared three times. Two losses were seen twice; losses of whole chromosomes 7 and X. The CGH findings were concordant with the results of conventional karyotyping. CGH did not add new information to the karyotypes. Since no high-level amplification was found among the samples and standard karyotyping was highly successful, we do not advocate routine use of CGH in the diagnostic evaluation of childhood AML.

Multiple drug resistance mediated by P-glycoprotein is not a major factor in a slow response to therapy in childhood ALL.

In childhood acute lymphoblastic leukemia (ALL), early response to treatment is an important prognostic factor and drug resistance is a major cause of poor outcome. One of the most investigated resistance mechanisms is P-glycoprotein (P-gp)-mediated multiple drug resistance (MDR). We analyzed P-gp using flow cytometry with monoclonal antibody JSB1 in a series of 118 children with ALL, 103 at diagnosis and 15 at relapse. Increased P-gp expression was found in 55 (53%) patients at diagnosis and in 11 (73%) at relapse. We also analyzed the bone marrow aspirate slides for early response to treatment in a central review. No correlation was found between P-gp and early response. Patients with T-ALL had higher P-gp levels than the others, 5.3% versus 1.0% (P = .002). We conclude that P-gp-mediated multiple drug resistance is not a factor in a slow response to ALL induction therapy.