Search

This study was a second multicenter trial on childhood ALL by the Japan Childhood Leukemia Study Group (JACLS) to improve outcomes in non-T ALL. Between April 2002 and March 2008, 1138 children with non-T ALL were enrolled in the JACLS ALL-02 trial. Patients were stratified into three groups using age, white blood cell count, unfavorable genetic abnormalities, and treatment response: standard risk (SR), high risk (HR), and extremely high risk (ER). Prophylactic cranial radiation therapy (PCRT) was abolished except for CNS leukemia. Four-year event-free survival (4yr-EFS) and 4-year overall survival (4yr-OS) rates for all patients were 85.4% ± 1.1% and 91.2% ± 0.9%, respectively. Risk-adjusted therapy resulted in 4yr-EFS rates of 90.4% ± 1.4% for SR, 84.9% ± 1.6% for HR, and 66.5% ± 4.0% for ER. Based on NCI risk classification, 4yr-EFS rates were 88.2% in NCI-SR and 76.4% in NCI-HR patients, respectively. Compared to previous trial ALL-97, 4yr-EFS of NCI-SR patients was significantly improved (88.2% vs 81.2%, log rank p = 0.0004). The 4-year cumulative incidence of isolated (0.9%) and total (1.5%) CNS relapse were significantly lower than those reported previously. In conclusion, improved EFS in NCI-SR patients and abolish of PCRT was achieved in ALL-02.

L-Asparaginase has significantly improved outcome for children with acute lymphoblastic leukemia and has become an essential component of multiagent chemotherapy. However, there are many adverse events due to L-asparaginase, including acute pancreatitis. The pathology of L-asparaginase-associated pancreatitis (AAP) remains unclear. We compared patients who developed AAP (n=29) and random matched controls (n=36) who had been enrolled in the Japan Association of Childhood Leukemia Study of the ALL-02 protocol. AAP and control patients were matched for age, sex, treatment, and protocol risk. We examined correlations between AAP development and clinical symptoms, laboratory data, and concomitant medication. Abdominal pain and nausea were common presenting symptoms for AAP. There was an increased risk of AAP in patients using gastric acid-suppressing agents and antithrombin (AT) supplementation. Mean fibrinogen and AT levels before the onset of AAP were lower in AAP patients than in controls. Decreased AT and fibrinogen levels resulting from the strong suppression of protein synthesis by L-asparaginase were predictive signs for AAP. Our epidemiological approach should prove clinically useful for the diagnosis the AAP as early as possible.

Purpose This study was conducted as the first clinical trial by Japan Association of Childhood Leukemia Study to improve the outcome of B-cell acute lymphoblastic leukemia and explore a less toxic reinduction block. Patients and methods From 1997 to 2002, 563 patients with B-cell acute lymphoblastic leukemia aged 1 to 15 years were enrolled. The patients were assigned into 4 risk groups (standard, intermediate, high, or extremely high risk) and treated with regimens intensified according to the risk. Two randomized trials were conducted to compare 2 regimens with and without a 3-week reinduction therapy in the standard-risk group, and to compare the efficacy of pirarubicin with daunorubicin in the intermediate-risk and high-risk groups. Prophylactic cranial irradiation was restricted in patients with high or extremely high risk. Results The event-free survival (EFS) rate at 10 years for all patients was 77.0%. Those in the standard-risk to extremely high-risk groups were 79.3%, 72.5%, 71.7%, and 66.3%, respectively. The 15-week induction/consolidation not followed by reinduction produced 76.4% of the EFS at 10 years comparable with the regimen with reinduction therapy. Pirarubicin at 25 mg/m administered 11 times throughout the treatment produced the EFS comparable with daunorubicin at 30 mg/m. Conclusion The trial produced high survival rates in NCI-HR patients, although the outcomes in NCI-SR patients were not satisfactory possibly due to less intensive central nervous system-directed therapy.

Anthracyclines are used to treat childhood acute lymphoblastic leukemia (ALL). Even when administered at low doses, these agents are reported to cause progressive cardiac dysfunction. We conducted a clinical trial comparing the toxicities of two anthracyclines, pirarubicin (THP) and daunorubicin (DNR), in the treatment of childhood ALL. The results from our study that relate to acute and late toxicities are reported here. 276 children with B-ALL were enrolled in the trial from April 1997 to March 2002 and were randomly assigned to receive a regimen including either THP (25 mg/m2 × 11) or DNR (30 mg/m2 × 11). Acute toxicity was prospectively assessed based on the National Cancer Institute Common Toxicity Criteria. Acute hematological toxicity was also examined via some parameters. Patients with event-free survival of >5 years were retrospectively surveyed for cardiac function at 5 and 10 years and at the most recent assessment more than 10 years from the onset of ALL. Acute hematological toxicity in the early phase was more significant in the THP arm. Based on ultrasound cardiography, cardiac function was impaired in both groups during the follow-up period, but there was no significant difference between the groups except for a greater decline in fractional shortening on ultrasound cardiography in the DNR arm. While acute hematological toxicity was more significant in the THP arm, THP also appeared to be less cardiotoxic. However, the evaluation of late cardiotoxicity was limited because only a few subjects were followed beyond 10 years after ALL onset. Considering that the THP regimen produced an EFS rate comparable with that of the DNR regimen, the efficacy and toxicity of THP at reduced doses should be studied in order to identify potentially safer regimens.

ETV6-RUNX1-positive B precursor acute lymphoblastic leukemia (B-ALL) is a common subtype of pediatric B-ALL that has shown excellent outcomes in contemporary clinical trials for pediatric B-ALL. Examinations of the possibility of reducing therapeutic intensity may thus be explored. This prospective study examined outcomes in 205 pediatric patients with ETV6-RUNX1-positive B-ALL uniformly treated following the Japan Association of Childhood Leukemia Study Group (JACLS) ALL-02 protocol. The JACLS ALL-02 protocol does not employ minimal residual disease detected by polymerase chain reaction (PCR-MRD)-based risk stratification; however, 4-year event-free survival (EFS) and overall survival (OS) were 94.4 ± 1.6 and 97.5 ± 1.1%, respectively. In particular, 92 of 205 (44.9%) patients were successfully treated with a less intensive regimen involving only two cycles of high dose methotrexate and one course of re-induction therapy comprising vincristine, l-asparaginase (L-asp), pirarubicin, and prednisolone. Multivariate analysis revealed that discontinuation of L-asp and poor response to prednisolone was, respectively, associated with poor EFS (HR 6.3; 95% CI 1.3–27.0) and OS (HR 17.5; 95% CI 2.3–130), suggesting that the majority of ETV6-RUNX1-positive B-ALL cases may be cured by a less-intensive chemotherapy regimen if the risk stratification system including PCR-MRD monitoring and insufficient use of L-asp is avoided.

IKZF1 deletion is enriched in pediatric B-cell precursor acute lymphoblastic leukemia patients showing prednisolone resistance

Purpose Osteonecrosis (ON) is a serious complication of the treatment of childhood acute lymphoblastic leukemia (ALL); however, data relating to ON in Asian pediatric patients with ALL are scarce. Therefore, we performed a retrospective analysis of cohorts of Japanese patients with ALL to clarify the incidence, clinical characteristics, and risk factors of ON. Patients and Methods The incidence and characteristics of ON were determined in patients with ALL (n = 1,662) enrolled in two studies from the Japan Association of Childhood Leukemia Study (JACLS) group (n = 635 and n = 1,027 patients treated with the ALL-97 and ALL-02 protocols, respectively). Results In total, 24 of 1,662 patients suffered from ON, of which 12 of 635 and 12 of 1,027 patients were treated with the ALL-97 and the ALL-02 protocol, respectively. Of the 24 patients, 23 were older than 10 years. In multivariate analysis, age (≥ 10 years) was the sole significant risk factor for ON ( P < .001). Separate evaluation of patients ≥ 10 years of age indicated a 5-year cumulative incidence of ON of 7.2% (95% CI, 4.0% to 12.6%) and 5.9% (95% CI, 3.3% to 10.4%) in the ALL-97 and the ALL-02 protocol, respectively, which was lower than reported previously, despite an administration of dexamethasone (DEX) similar to that in comparable studies; however, concomitant administration of DEX and l-asparaginase was reduced in the JACLS protocols. Conclusion We identified a low frequency of ON in the JACLS ALL-97 and ALL-02 studies. Although the sole risk factor for ON was age (≥ 10 years), even among high-risk patients, ON incidence was significantly lower than that reported in previous studies. These results suggest that, not only the total amount of DEX, but also how DEX and l-asparaginase are administered, which affects the clearance of DEX, may be associated with ON incidence in patients with ALL.

This study reviewed the clinical characteristics of 112 pediatric B-cell precursor acute lymphoblastic leukemia (BCP-ALL) patients with TCF3-PBX1 fusion treated according to the Japan Association of Childhood Leukemia Study (JACLS) ALL02 protocol (n = 82) and Children's Cancer and Leukemia Study Group (CCLSG) ALL 2004 protocol (n = 30). The 3-year event-free survival (EFS) and overall survival (OS) rates were 85.4 ± 3.9% and 89.0 ± 3.5% in JACLS cohort, and the 5-year EFS and OS were 82.8 ± 7.0% and 86.3 ± 6.4% in CCLSG cohort, respectively, which are comparable to those reported in western countries. Conventional prognostic factors such as age at onset, initial white blood cell count, and National Cancer Institute risk have also no impact on OS in both cohorts. Surprisingly, the pattern of relapse in JACLS cohort, 9 of 82 patients, was unique: eight of nine patients relapsed during the maintenance phase and one patient had primary induction failure. However, bone marrow status and assessment of minimal residual disease on days 15 and 33 did not identify those patients. Interestingly, the two patients with IKZF1 deletion eventually relapsed in JACLS cohort, as did one patient in CCLSG cohort. International collaborative study of larger cohort is warranted to clarify the impact of the IKZF1 deletion on the poor outcome of TCF3-PBX1 positive BCP-ALL.

Background Children with acute lymphoblastic leukemia (ALL) who fail to achieve complete remission (CR) after induction therapy (induction failure: IF) have a poor prognosis; however, there have been few prospective studies in patients with IF. Patients and Methods Between April 1997 and March 2005, 27 of 1,237 leukemic patients (2.2%) failed to achieve CR after four- or five-drug induction therapy. Twenty-three of these patients entered the F-protocol study, which mainly consisted of acute-myeloid-leukemia-oriented chemotherapy followed by scheduled hematopoietic cell transplantation (HCT). Results Seventeen (73.9%) of the 23 patients responded to re-induction chemotherapy with CR. Of note, 15 (93.8%) of 16 patients with Philadelphia-chromosome-negative (non-Ph+) ALL achieved CR; in contrast, only 2 (28.6%) of 7 Ph+ patients achieved CR. Fourteen (82.4%) of 17 patients remained in CR (CCR) until their scheduled HCT, 12 of the 14 with CCR underwent HCT as scheduled, and 6 patients remain in first CR after a median of 78 months (range, 49–107 months). The 5-year overall survival (OS) rates of 16 patients with non-Ph+ and 7 patients with Ph+ were 43.8 ± 12.4% and 14.3 ± 13.2%, respectively (P = 0.012). The 5-year OS rate of the 17 patients who obtained CR by re-induction therapy and the 6 who did not were 47.1 ± 12.1% and 0%, respectively (P

In a retrospective review of the survival outcome of children with isolated central nervous system (CNS) relapse of acute lymphoblastic leukemia (ALL), we identified 79 patients with CNS relapse among the eligible patients enrolled in ALL trials of 3 Japanese pediatric oncology study groups (Japanese Children's Cancer and Leukemia Study Group [JCCLSG], Tokyo Children's Cancer Study Group [TCCSG], and Japan Association of Childhood Leukemia Study [JACLS]) between 1989 and 1999. CNS relapses were diagnosed as the first adverse event between 1991 and 1999. The median age at the time of CNS relapse was 5.0 years (range, 0.7-15.1 years). The duration of the first remission ranged from 1.4 to 54 months (median, 12.4 months), and the observation period after CNS relapse ranged from 1 to 131 months (median, 27 months). Overall, 75 of the 79 patients achieved a second complete remission, 44 of whom had second relapses in the following sites: CNS, 18 patients; bone marrow, 15 patients; combined sites, 8 patients; and testis, 2 patients. Rates of overall survival and event-free survival at 4 years were 43.7% +/- 5.8% (mean +/- SE) and 32.9% +/- 5.5%, respectively. The probability of remaining in second remission was significantly correlated with the leukocyte count (P= .005) and age (P= .02) at the initial diagnosis.

Prognostic factors of juvenile myelomonocytic leukemia (JMML) have not been clarified because of its very low incidence and inaccuracy in the diagnosis. The purpose of this study was to evaluate children with JMML given an allogeneic hematopoietic stem cell transplantation (SCT) and the role of different variables potentially influencing outcome in a nationwide survey in Japan based on the newly proposed criteria by the International JMML Working Group. The study patients were 27 children who underwent SCT among 55 JMML patients retrospectively collected in the survey. The source of grafts was HLA-identical siblings in 12 cases, HLA-matched unrelated individuals in 10 and others in five. Total body irradiation was used in 18 cases. Event-free and overall survival (OS) at 4 years after SCT were 54.2 +/- 11.2% (s.e.) and 57.9 +/- 11.0% (s.e.), respectively. Six patients died of relapse and three of complications. Patients with abnormal karyotypes showed a significantly lower OS than those with normal karyotypes (P0.001). Patients below 1 year of age showed a significantly higher OS than those of 1 year of age or more (P = 0.02). Patients with grade 0-1 acute graft-versus-host disease (GVHD) or chronic GVHD had a more favorable OS than those without them, although they were not statistically significant (P0.05). Other variables studied were not associated with OS. A multivariate analysis of these factors yielded the abnormal karyotype as the only significant risk factor for lower OS (risk ratio: 11.0; 95% CI: 2.7-45.1).

Between April 1994 and March 1997, 143 children (age range, 1-15 years) with newly diagnosed acute lymphoblastic leukemia (ALL), except for those patients with t(9;22), were treated according to protocol-94 of the Osaka Childhood Leukemia Study Group. In this trial, the intensity of chemotherapy was enforced in the consolidation and reinduction phases by introducing AML-type block therapies consisting of concentrated administration of 4 to 6 drugs during 5 or 6 days. For patients in the higher risk groups, rotational combination chemotherapy was introduced following the early phase. A total of 124 children with B-cell precursor ALL (B-pre ALL) were classified into 3 groups, the ultrahigh-risk group (UHRG) (15 patients), the high-risk group (HRG) (61 patients), or the standard-risk group (SRG) (48 patients), based on age. leukocyte count, immunophenotype, central nervous system leukemia, response to treatment, and selected chromosomal abnormalities. The complete remission rate was 93%, and the 6-year event-free survival (EFS) rate was 79%+/-4%. EFS rates for the UHRG, HRG, and SRG groups were 67%+/-12%, 80%+/-6%, and 81%+/-6%, respectively. Nineteen patients with T-cell ALL were treated with the protocol for the UHRG. Thirteen patients (68%) attained complete remission, and the 6-year EFS rate was 55%+/-12%. Thus, intensification of chemotherapy improved the EFS rate and AML-type block therapies appeared to be effective as the consolidation and reinduction therapies for B-pre ALL.

To assess the clinical significance of monitoring minimal residual disease in t(8;21)(q22;q22) AML, RT-PCR assay was conducted during the clinical course of 12 patients who had undergone BMT or conventional chemotherapy. Two cases relapsed after BMT and chimeric RNA was detected soon after BMT in their bone marrow cells. The other three cases, in whom chimeric RNA was not detected after BMT, are in CR at 21 to 33 months following BMT. Similarly, four out of 7 cases who showed negative chimeric RNA after completion of chemotherapy have been in CR at 11 to 34 monthsfollowing completion of chemotherapy.The present findings appear different from other studies which reported the detection of AMLI-ETO chimeric RNA in long-term CR patients.

Genetic alterations of Ikaros family zinc finger protein 1 (IKZF1), point mutations in Janus kinase 2 (JAK2), and overexpression of cytokine receptor-like factor 2 (CRLF2) were recently reported to be associated with poor outcomes in pediatric B-cell precursor (BCP)-ALL. Herein, we conducted genetic analyses of IKZF1 deletion, point mutation of JAK2 exon 16, 17, and 21, CRLF2 expression, the presence of P2RY8-CRLF2 fusion and F232C mutation in CRLF2 in 202 pediatric BCP-ALL patients newly diagnosed and registered in Japan Childhood Leukemia Study ALL02 protocol to find out if alterations in these genes are determinants of poor outcome. All patients showed good response to initial prednisolone (PSL) treatment. Ph⁺, infantile, and Down syndrome-associated ALL were excluded. Deletion of IKZF1 occurred in 19/202 patients (9.4%) and CRLF2 overexpression occurred in 16/107 (15.0%), which are similar to previous reports. Patients with IKZF1 deletion had reduced event-free survival (EFS) and overall survival (OS) compared to those in patients without IKZF1 deletion (5-year EFS, 62.7% vs. 88.8%, 5-year OS, 71.8% vs. 90.2%). Our data also showed significantly inferior 5-year EFS (48.6% vs. 84.7%, log rank P = 0.0003) and 5-year OS (62.3% vs. 85.4%, log rank P = 0.009) in NCI-HR patients (n = 97). JAK2 mutations and P2RY8-CRLF2 fusion were rarely detected. IKZF1 deletion was identified as adverse prognostic factor even in pediatric BCP-ALL in NCI-HR showing good response to PSL.

Background: It is important to establish proper intensity of chemotherapeutic regimen for low risk group of childhood B-precursor ALL (BCP-ALL) to avoid late effects. For this purpose, in Japan Association of Childhood Leukemia study (JACLS) ALL-02 clinical trial, we defined standard risk (SR) group as the patients ranged from 1 to 10 years showing white blood cell (WBC) count of less than 10,000/μL at the diagnosis without predonisolone poor responder, t(4;11), t(1;19) and CNS3 to treat with less intensive chemotherapy (JACLS ALL02 SR protocol). Herein, we report the outcome of JACLS ALL02 SR protocol and analyzed prognostic factors to identify super low risk group which is curable by less intensive chemotherapy. Patientss and treatment: JACLS ALL02 SR protocol consisted of early phase therapies for total 4 months including an induction therapy using vincristine (VCR), steroids, pirarubicin (THP-ADR), and L-asparaginase (L-asp) for 4 weeks, a consolidation therapy using cyclophosphamide and a combined administration of cytarabine and dexamethasone (DEX) or 6-mercaptopurine (6-MP), a sanctuary therapy with only two cycles of high-dose methotrexate (MTX) of 3g/m2, one course of 4 drugs re-induction therapy and maintenance phase with 6-MP and oral MTX combined with VCR and PSL every 5 weeks by the end of the second year. In addition, total of 12 times of triple intrathecal therapy were administered by the end of the first year. Estimation of event free survival (EFS) and overall survival (OS) was performed using the Kaplan-Meier method and the differences were compared using the log-rank test. Multivariate analysis was performed using a Cox regression model. Results: 1252 newly diagnosed ALL patients were enrolled from April 2002 to May 2008 in JACLS ALL02 clinical trial, which include 388 patients (192 males and 196 females) in SR group (31.0%). The median age at diagnosis was 4.2 years (range, 1.0 to 9.9 years) and the median WBC count was 3,700/μL (range, 370 to 9,984 /μL). The median follow-up time was 6 year and 5 months. 4y / 8y EFS and OS (±SE) of the SR group are 91.5±1.4% / 89.4±1.7% and 97.9±0.7% / 95.2±1.2%, respectively. 4y / 8y EFS of the patients of 1 to 5 years old (n=300) and 6 to 9 years old (n=88) are 93.5±1.4% / 93.0±1.5% and 84.9±3.9% / 75.8±5.8%, respectively (log rank p=0.0003). 4y / 8y EFS of the patients with day 15 M1 marrow (n=287 ) was 92.8±1.6% / 91.6±1.8%, while those of the patients with day 15 M2 marrow (n=101) was 88.0±3.3% / 83.7±3.9% (log rank p=0.0446). When 388 patients were further classified into 4 subgroups based on genetic abnormalities, such as ETV6-RUNX1 (n=89, 22.9%), high hyperdiploid (HHD) (n=97, 25%), normal karyotype (n=147, 37.9%), and others (n=55, 14.2%), 4y-EFS/OS of each group was 97.6±5.9 /100% (ETV6-RUNX1), 91.5±2.9 /100% (HHD), 89.6±2.5 /95.1±1.8% (normal karyotype), and 86.9±4.6 /98.2±1.8% (others), suggesting poor EFS of B-others group. When HHD subgroup was divided into two groups, such as triple trisomy (TT; n=35) (trisomy of chromosome 4,10,17) and non-TT (n=62), 4y-EFS of each sub-group was 100% and 84.5±4.5%, respectively (log rank p=0.0161). Interestingly, day 15 M2 marrow was associated with poor 4y/8y EFS only in the normal karyotype subgroup (day 15 M1 vs M2 92.0±2.6% / 90.2±3.1% vs, 81.3±6.9% / 77.2±7.7%, log rank p=0.0497). The rates for NCI-CTC grade 4 infection, allergy, increase of transaminase were 1.80%, 0.26%, and 22.9%, respectively. Multivariate analysis identifies the patients of 6 to 9 years old (HR=3.178, p= Discussion: Although JACLS ALL-02 SR protocol was less intensive chemotherapeutic regimen compared to the contemporary protocols used for NCI-SR patients, good outcome of the 4y EFS and OS (91.5±1.4 % and 97.9±0.7 %) was achieved with good feasibility. Because the specific genetic subgroups such as B-others and HHD without TT were associated with relatively poor outcome, those patients should be treated with more intensive chemotherapy, especially for the patients showing initial poor response (day 15 M2). On the other hand, ETV6-RNUX1 and TT subgroups showed excellent outcome. Thus, appropriate reduction of therapy should be explored for these subgroups. Disclosures No relevant conflicts of interest to declare.

SUMMARY The age-related changes in proportion of various subsets within lymphocytes were investigated in cord blood and peripheral blood from healthy children and adults. The percentages of T and B cells did not show age-related changes, whereas natural killer (NK) cells increased significantly with age. Within lymphocytes or the CD3+ T cell population the proportion of CD45RAbright+ lymphocytes decreased and that of CD45RO+ cells increased, while that of CD45RAdim+ cells showed no age-related change. Within lymphocytes, the percentage of CD45RAbright+ CD4+ cells decreased, together with a decline of that of CD4+ cells. The proportions of CD45RAbright+ CD8+ cells and S6F1bright+ CD8+ cells increased with age, and the age-dependent increase of the proportion of CD8+ cells seems to be mainly attributable to the increases in these subsets. The CD45RAdim+ CD4+ and CD45RAdim+ CD8+ cells co-expressing CD45RO at a low level nevertheless showed no age-related changes. In γδ T cells, both δTCS1+ and δTCS1- T cells increased with age, but the δTCS1-γδ T cells increased more than the δCS1+ subset. Among lymphocytes, the percentages of CD20+, CD21+ and CD22+ cells remained similar, with no age-related changes, but the proportion of CD5+ cells within lymphocytes or B cells decreased. The proportions of CD16+ NK cells among lymphocytes increased with age, and this change was attributable to the increase of CD56+ cells.

By immunophenotyping and ultrastructural cytochemistry, the disorders involving megakaryocytic lineage cells have been clarified. These disorders are termed acute megakaryocytic leukemia (AMKL) and transient abnormal myelopoiesis (TAM). The characteristics of blasts in these disorders have been extensively investigated from various standpoints including cytochemistry, cytogenetics, ultrastructure and in vitro-colony differentiation. The target cells of AMKL and TAM are immature cells close to stem cells which are capable of differentiating into lineage cells such as megakaryocytes, erythrocytes and myeloid cells. Phenotypically, these blasts frequently express antigens appearing at an early stage in the hematopoietic differentiation pathway. They thus often emerge as mixed phenotypes as seen in mixed lineage leukemia of immature cell origin.

We recently encountered a patient with acute lymphoblastic leukemia (ALL) who showed temporal monocytosis of an unusually high cell count (5,000-30,000 monocytoid cells/μ1) five times after treatment

Ten leukemia cases with mixed phenotype were investigated in terms of clinical characteristics and cellular origin. Three patients were infants and six patients were older children. Six of them had a high leukocyte count and a mediastinal mass was found in three cases. All but one showed hepatosplenomegaly and/or lymphoadenopathy. In spite of intensive chemotherapy, most of them responded poorly. Cytochemical analysis of their leukemic cells revealed a low percentage of positivity for myeloperoxidase reactivity (< 25%) in two cases and electron microscopic platelet peroxidase reactivity was found in one of three analyzed cases. Phenotypically, these cells all expressed CD7, and other T-lineage-associated, B-lineage-associated, and/or myeloid-associated antigens were also detected to some extent. In addition, three cases expressed CD41 and one case expressed CD56. The T-cell receptor (TCR) genes and immunoglobulin gene were in the germline configuration in seven cases. In three rearranged cases two showed only the TCR-δ gene rearrangement, and one had both TCR-γ and δ gene rearrangements. Cell culture studies with 12–0-tetradecanoyl-phorbol-13-acetate (TPA) revealed differentiation to the T-lineage in two cases and to a myeloid lineage in one case. Megakaryocytic differentiation was detected in two cases in culture without TPA. These results suggest that the cells from these cases arose from stem cells capable of both lymphoid and nonlymphoid differentiation. Although the cells were heterogeneous with regard to their potency of differentiation, they have similar clinical characteristics. Because of poor prognosis, it is important to identify this type of leukemia, and allogenic or autologous bone marrow transplantation should be considered. Cancer 68:2273–2280, 1991.

We analyzed multidrug-resistance gene (mdr1 gene) expression in a patient with undifferentiated sarcoma of the liver using the cloned cDNA for the mdr1 gene. Tissue samples were available at the time of initial diagnosis and of two intracranial relapses after chemotherapy with a regimen including doxorubicin and teniposide. The level of mdr1 gene expression was increased sevenfold in the intracranial tumor at the time of first relapse and 11-fold at the second relapse. This case may be an example of acquired multidrug resistance associated with overexpression of the mdr1 gene.

A 10-month-old male infant underwent splenectomy because of life-threatening hypersplenism due to malignant histiocytosis. The spleen tissue positive for herpes simplex virus (HSV)-DNA, histologically and immunologically revealed a marked B cell depletion replaced by proliferation of activated cytotoxic T cells. S100-positive histiomonocytoid cells and lysozyme-positive hemophagocytes were also significantly increased. No metaphases were obtained by cytogenetic studies and Southern blot analysis of spleen cell DNA demonstrated only germ bands of immunoglobulin and both T gamma and beta genes, providing no evidence of monoclonality in this case of so-called malignant histiocytosis.

A 5-year-old boy received CD34-positive HLA haplo-identical bone marrow transplantation from his father as treatment for refractory advanced neuroblastoma. He had residual disease in the para-aortic lymph nodes and multiple bones after the transplant. However, all of his residual disease had disappeared completely 3 years later. He developed grade I acute graft-versus-host disease (GVHD) but had no symptoms of chronic GVHD or any other complications. This case demonstrates the possibility of a graft-versus-tumor effect against neuroblastoma by HLA-mismatched allogeneic hematopoietic stem cell transplantation.

The plasma levels of a panel of cytokines and cytokine-associated molecules (IL-1α, IL-2, IL-4, IL-6, IL-10, IL-12, IL-15, macrophage colony-stimulating factor (M-CSF), interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α), soluble IL-2 receptor (sIL-2R), soluble tumor necrosis factor receptor I or II (sTNFRI or II)) were assessed in 56 plasma samples of 13 pediatric patients undergoing hematopoietic stem cell transplantation (HSCT, bone marrow in 12 and cord blood in one) from unrelated donors. Eight patients developed severe (grade III–IV) acute GVHD (aGVHD). The plasma IL-6, IL-10, M-CSF, sTNFRI and II levels were significantly high in the severe aGVHD group compared to the mild aGVHD group (grade 0–II). The plasma IL-15 level increased transiently in the early period following HSCT and remained high in the severe aGVHD group even after 4 weeks following HSCT. Based on analysis of the correlations between the kinetics of the plasma cytokine levels after HSCT and the clinical manifestations of aGVHD, IL-15 and/or M-CSF were involved in the development of aGVHD, following elevation of the plasma IL-10 and sTNFRI or II levels. These kinetics suggest that IL-10 and sTNFRs worked as suppressor cytokines and seemed to suppress clinical manifestations of aGVHD. Furthermore, it seemed that the plasma ratio of IL-10/sTNFRII from 5 to 12 weeks following HSCT was linked to the poor outcome in the patients with severe aGVHD, suggesting that IL-10 plays an important role in protecting hosts from transplantation-related complications, including GVHD. Bone Marrow Transplantation (2001) 27, 1153–1161.

Abstract 1506 BACKGROUND: A wide range of neuropsychological sequelae, noted in long-term survivors of ALL, have been in part ascribed to cranial irradiation (CRT), especially in children who are more vulnerable than adults. We had attempted to abandon prophylactic CRT by intensifying systemic chemotherapy and protracted triple IT (TIT) in ALL-02 trial; whereas one third of patients received CRT for CNS directed therapy in the former study (ALL-97). We present here the results of JACLS ALL-02 trials especially about the effect of the CNS-directed therapy. PATIENTS AND METHODS: Between 2002 and 2008, 1139 children with newly diagnosed non-T ALL with 1–18 years of age were enrolled to JACLS ALL-02 trial. Patients with Ph+ALL and T-ALL were registered to independent protocol. The criteria for diagnosis of CNS disease of the JACLS ALL-02 trial were defined as a CSF pleocytosis of > 5 cells /microL and the presence of recognizable blast cells on a well-stained cytospin preparation, or the presence of cranial-nerve palsies at a diagnosis of leukemia. Treatment group was divided into 3 groups according to the BFM-style initial prednisolone (PSL) response and the modified National Cancer Institute (NCI) workshop criteria. PSL good responders were divided into standard risk (SR) and high risk (HR) by WBC 10,000 and age 10. BCP-ALL with PSL poor responders and acute mixed lineage leukemia/ acute unclassified leukemias were treated in extremely high risk (ER). The SR patients with unequivocal CNS involvement (CNS3) at diagnosis were treated as an HR group. The patients in M1 marrow at day 33 with M2/M3 at the day 15 bone marrow (BM) were assigned to shift higher risk after induction therapy. Treatment of ALL-02 consists of early phase and maintenance phase. Early phase includes induction, consolidation, sanctuary (2 courses of high-dose MTX at a dose of 3g /m2), and re-induction therapy for 15 to 26 weeks depending on risk group. Prophylactic CRT was replaced by protracted TIT (MTX, CA, HDC) in all except the patients with CNS3 at diagnosis in ALL-02 trial. The patients with CNS3 at diagnosis received CRT at a dose of 12 Gy. Depending on the risk group, protracted TIT was given during induction, intensification and maintenance in ALL-02 (12 doses in SR and 15 in HR/ER). Accumulative doses of DEX were 120 mg/m2 in SR, whereas 170 mg/m2 in HR and 670 mg/m2 in ER. Treatment duration is 24 months for any risk. The patients assigned ER were candidate for allogeneic stem cell transplantation (SCT) by the end of early phase, provided HLA-matched siblings were available. The probability of event-free survival (EFS) and overall survival were constructed using the Kaplan-Meier method. Events in the analysis of EFS included induction failure, death, relapse and secondary malignant neoplasm. All statistical analyses were done according to intent-to treat methods. RESULTS: The numbers of patients at each risk in ALL-02 were 457 (40%) for SR, 543 (48%) for HR, and 139 (12%) for ER. The number of patients with CNS3 at diagnosis was 31 (2.7%). Of 1139 patients, 16 patients (1.4%) had CNS relapses of the leukemia (an isolated CNS relapse: 10 pts, a combined CNS and BM relapse: 6 pts). The 5-year EFS rate ALL-02 was 83.7% (SE=1.1), slightly better than for ALL-97 (79.3%, SE=1.7, p =0.054). The 5-year cumulative risk of an isolated and total CNS relapse was lower in ALL-02 than in ALL-97 (isolated CNS relapse: 0.9% vs 2.7%, p=0.017, total CNS relapse: 4.5% vs 1.4%, p =0.01). The proportion of CRT and SCT were 2%, 8% in ALL-02 respectively, whereas 31%, 11% in ALL-97 respectively. CONCLUSIONS: The ALL-02 trial found that CRT was abolished successfully, in all except those with CNS3 at diagnosis, with substitution of more intensive systemic chemotherapy and protracted TIT. Disclosures: No relevant conflicts of interest to declare.

We analysed the organization of V delta genes and delta recombining element (delta Rec) in 27 children with B-cell precursor acute lymphoblastic leukaemia. Twenty-two of 54 alleles showed rearrangements of the T-cell receptor (TCR) delta locus. These rearrangements resulted either from D2D delta 3 (2 alleles) or V delta 2(Dn)D delta 3 (20 alleles) recombinations, and the other V delta and delta Rec were not rearranged. Of 23 alleles with deletion of C delta and rearrangements of J alpha, V delta 2, V delta 4 and V delta 5 appeared to rearrange to J alpha on five alleles. With regard to the relationship between the rearranged V alpha/delta and J alpha genes, gene segments 5' to V delta 2 frequently rearranged to J alpha more proximal to C alpha, whereas V delta 2 and gene segments 3' to V delta 2 showed a tendency to rearrange to J alpha distal to C alpha. Based on these findings, we suggest that the initial recombination event of the TCR-alpha/delta gene may be D2D delta 3 joining, followed by V delta 2 recombination with the D2D delta 3 complex. It was also suggested that use of V alpha/delta and J alpha/delta may depend on the distance between the involved V alpha/delta and J alpha/delta at least in B-lineage cells. These rearrangements in B-precursor cells appear to be aberrant. However, this recombinational process may be one of the normal differentiation pathways in T-lineage cells, because cells with a V delta 2(Dn)D delta 3 rearrangement were detected in 0.1-0.01% of normal peripheral mononuclear cells by the polymerase chain reaction.

Cells from 203 childhood leukemia patients were analyzed using a broad panel of markers, such as immunological marker studies and Southern blot and Northern blot analyses to define the lineage-specific classification of childhood leukemia. Phenotypically, they were divided into B lineage (62.6%), T lineage (9.8%), nonlymphoid (14.3%), and uncertain lineage (13.3%). Two of the B-lineage ALL and two of the T-lineage ALL studied did not show immunoglobulin (Ig) and T-cell receptor (TCR) gene rearrangements, respectively. Therefore those four cases were excluded from the final classification. The uncertain lineage leukemia, which includes undifferentiated leukemia and mixed lineage leukemia, could be further subclassified at the DNA and RNA levels. On the basis of these findings, we discuss the definition for B-lineage and T-lineage cells, the incidence of dual genotypes or spillover, heterogeneity of undifferentiated leukemia, and a new classification for mixed—lineage leukemia. In addition, NK—lineage leukemia associated with Epstein-Barr viral infection is also presented.

We investigated N-ras activation in childhood acute lymphoblastic leukemia (dALL) by the polymerase chain reaction (PCR) and the oligonucleotide hybridization method. The frequency of point-mutation of the N-ras gene was not high (2 of 15), and one positive case who relapsed was analyzed in detail. Although N-ras gene activation was detected at both onset and relapse, the mutation sites were different. At onset, Gly (GGT) was changed to Ser (AGT) at codon 12, and at relapse, Gly (GGT) to Asp (GAT) was observed at the same codon. In addition, the DNA at relapse showed a remarkably higher transforming activity than the DNA at onset on two independent recipient cell lines. The identical cell surface phenotype and the same rearrangement patterns of both the immunoglobulin (Ig) heavy chain and T-cell receptor (TCR) gamma chain genes indicated that the leukemic cells at onset and those at relapse were derived from the same precursor cell. Therefore, this case supports the concept that ras activation is not the event initiating leukemogenesis, but may be involved in leukemic progression.

BACKGROUND: With current intensified chemotherapy, more than 70% of children with acute lymphoblastic leukemia (ALL) are curable. However, prognosis of the patients at higher risk has been unsatisfied. More appropriate risk assignment and innovative treatment is expected to be developed. We conducted a clinical trial of JACLS ALL-97 including multi-agent block therapy followed by hematopoietic stem cell transplantation (SCT) for the selected higher risk patients. PATIENTS AND METHODS: Between April 1997 and March 2002, 674 patients aged 1 to 15 years with newly diagnosed ALL (excluding mature B-cell ALL) were enrolled on the JACLS ALL-97 protocol. Excluding 75 T-ALL, 26 mixed lineage leukemia other than B-precursor ALL with myeloid markers, 9 acute unclassified leukemia and 1 B-precursor ALL with pretreatment, 563 patients with B-precursor ALL were eligible for this analysis. Treatment group was divided 5 groups according to the modified National Cancer Institute (NCI) workshop criteria. Standard risk (SR) and IR (intermediate risk) divided by WBC10 × 103 account for NCI-SR. High risk (HR) and ER (extremely high risk) divided by WBC 100 × 103 account for NCI-HR. The patients with ALL positive for Philadelphia chromosome and for translocation with chromosome11q23 were assigned to the highest risk group F. The patients in complete remission (CR) at day 35 with M2/M3 at the day 14 marrow were assigned to shift higher risk after induction therapy. Treatment of ALL-97 consists of early phase and maintenance phase. Early phase includes induction therapy, consolidation therapy, sanctuary therapy, and re-induction therapy for 19 to 26 weeks dependent on risk group. Early phase in IR/HR protocol is identical to the SR protocol except the addition of two doses of DNR and three doses of CPM. Maintenance phase contains standard MTX and MP with monthly VCR and PSL intensification for SR, and rotational therapy of a set of MTX and MP with a set of VCR, PSL, ASP, and (DNR or CPM) for IR/HR/ER/F. The IR protocol is identical to the HR protocol without cranial irradiation. Treatment duration is 24 months for any risk. The patients assigned ER and F were candidate for allogeneic stem cell transplantation by the end of early phase. Transplant procedures depended on the institute. RESULTS: Number of patients at each risk was 204 for SR, 158 for IR, 128 for HR, 36 for ER, 27 for Ph+ and 10 for 11q23. Six of them were treated with incorrect risk protocol. Thirty-four patients received SCT in first CR. Following the induction therapy, 550 of 564 patients (97.5%) achieved CR. 23 patients(8 in SR, 9 in IR, 4 in HR and 2 in ER, 4.4% of all) were shift to higher risk due to the findings of day 14 marrow. Five-year overall survival rate (OS) and event-free survival rate (EFS) for all patients was 90.6% and 77.0%, respectively. Five-year EFS for NCI-SR and HR was 81.6% and 67.6%, respectively. According to risk group, 5-year EFS for SR, IR, HR, ER, Ph+ ALL, and ALL with 11q23 were 86.6%, 77.0%, 71.9%, 68.7%, 40.7%, and 70.0%, respectively. 5-year OS for them were 94.0%, 93.9%, 92.1%, 83.8%, 55.6%, and 68.6%, respectively. CONCLUSIONS: By the risk-adapted therapy in the JACLS ALL-97 trial, high cure rate could be achieved for children with B-precursor ALL.

With modern risk-adapted multi-agent chemotherapy, complete remission (CR) is obtained in the majority of children with acute lymphoblastic leukemia (ALL) and long-term survival free from disease can be expected in more than 70% of them. However, the prognosis of the patients without CR (induction failure; IF) has been dismal. We conducted a prospective study of F-protocol followed by a hematopoietic stem cell transplantation (SCT) in patients with IF. Purpose: To evaluate the efficacy and safety of the treatment strategy adopted in the JACLS ALL F-protocol. Patients and methods: Between April 1997 and March 2005, 1,254 patients with newly diagnosed ALL (excluding mature B-cell ALL) were enrolled on two consecutive JACLS ALL protocols (ALL-97 and ALL-02). Excluding induction death, 19 (1.6%) out of 1,207 patients with non-Ph+ ALL and 9 (30%) out of 30 patients with Ph+ ALL could not achieve CR following four- or five-drug induction chemotherapy. Among them, 25 patients in total were entered the study with guardian’s written consent. F-protocol consisted of the AML oriented re-induction chemotherapy (3 days of 8 mg/m2 MIT, 500 mg/m2 CA, and 40 mg/m2 PSL, followed by 3 days of 200 mg/m2 VP16, CA, and PSL a week later), four block consolidation therapy consisting of two alternative regimens A and B (A: HD-DEX, VP16, CA, MIT, B: HD-MTX, ASP, PSL, CA, THP-ADR) and maintenance therapy. Patients with CR were scheduled to receive SCT after the second consolidation therapy before the maintenance therapy. Transplant procedures depended on the institute. Results: Following the re-induction therapy, 15 of 16 patients with non-Ph+ ALL (93.8%) achieved CR and 13 (86.7%) of them had continued CR until 27th week of treatment (CCR), which was the expected limiting time of SCT. Eight of 11 patients receiving SCT at the first CCR are alive. Two of 13 patients with CCR did not receive SCT by a physician’s decision, and relapsed during maintenance therapy. However they are alive following SCT in second CR. Five-year overall survival rate (OS) and event free survival rate for all 16 cases was 53% and 32.8%, respectively. Estimated five-year OS for patients receiving SCT at the first CCR was 59.7%. On the other hand, only 3 of 9 patients with Ph+ ALL (33.3%) achieved CR. No patients without response to re-induction therapy could achieve CR by any subsequent chemotherapy, but one success with SCT. Among 3 patients attaining a CR with re-induction therapy, 2 could continue CR and underwent SCT. One is alive in CCR and the other deceased from TRM. The other patient, relapsed during consolidation therapy, underwent SCT with success. Five-year OS for all 9 patients with Ph+ ALL was 33.3%. Concerning about adverse events of F-protocol, there was only one non-hematological toxicity NCI-CTC grade 4, which was pancreatitis due to ASP. Conclusions: F-protocol could produce a high remission induction rate in non-Ph+ ALL, but not in Ph+ ALL. Improved outcome of non-Ph+ ALL patients with IF seemed to be obtained with F-protocol timely followed by SCT.

T-cell acute lymphoblastic leukemia (T-ALL) and T-cell lymphoblastic lymphoma (T-LBL) account for most of the childhood T-lymphoid malignancy(LM). T-ALL is usually treated by the same protocol to the B-progenitor ALL. It is obvious that biologically, T-cell behaviors are different from those of B-progenitor cell, and cell origins are same among T-ALL and T-LBL. Thus we conducted a strategy initiating T-ALL specific protocol, differing from protocols for B-progenitor ALL. Furthermore, we indicated the same protocol to advanced T-LBL. The aims of this study were to evaluate the efficacy and safety of indicating the same protocol for childhood T-ALL and advanced T-LBL, and to reveal the prognostic factors of childhood T-LM. 70 eligibleT-ALL patients enrolled in the JACLS ALL-97 study between 1997-2001, and 32 eligible patients with stage III and IV T-LBL enrolled in the JACLS NHL-T98 trial between 1998-2002 were analyzed. Median age was 9y8m (2y~15y3m) for T-ALL and 11y11m (3y~15y4m) for T-LBL. Male/female ratio was 46/24, 26/6, respectively. Mediastinal mass was found in 31/70(44.3%) for T-ALL, 19/21(90.5%) for stageIII and 6/11(54.5%) for stageIV T-LBL. The treatment for 2 years consisted of the induction therapy (VCR, HD-MTX, CA, PSL, ASP), the 5-drug consolidation therapy A and B, both including high dose of ASP, and maintenance therapy with block-rotated treatment using drugs above. Complete remission(CR) at the end of induction therapy was obtained 65/70(92.9%) for T-ALL, 15/21(71.4%) for stage III and 10/11(90.9%) for stage IV T-NHL. 5-year overall survival(OAS) rate for T-ALL, stage III and stage IV T-LBL was 81.1%, 63.9% and 81.8%, respectively. 5-year event free survival (EFS) rate was 72.9%, 47.6% and 72.3%, respectively. Relapse after CR occurred in 12/65 with T-ALL, 6/15 with stage III and 1/10 with stage IV T-LBL. Single variant analysis revealed that there were no significant difference in OAS or EFS for T-ALL patients based on WBC, NCI index, but statistical difference in OAS or EFS based on age(older than 10y worse), the existence of mediastinal mass(absence, worse) In T-LBL, there were no statistical differences based on age, existence of mediastinal mass. Multivariate analysis revealed, for T-ALL and T-LBL patients as a whole, that age > 10 years was a risk factor in both OAS and EFS, absence of mediastinal mass and stage III T-LBL were risk factors in OAS. Our data shows that indicating same T-cell specific protocol, for T-ALL and advanced T-LBL has a potential to improve the prognosis of T-LM. The older age, and stage III T-LBL appeared as prognostic factors. Moreover, mediastinal mass with bone marrow involvement was a favorable factor for childhood T-LM. Although some risk factors were documented, it is needed to clarify unknown prognostic factors and develop the more effective, stratified T-cell specific protocols.

Clinical and laboratory features associated with CD33 expression were analysed in 123 children with B-precursor acute lymphoblastic leukemia (ALL), including 85 at onset, 34 at relapse and four in a refractory state to induction therapy. CD33 was demonstrated in 13 patients (15.3%) at onset, and it was associated with coexpression of T-cell and multipotential hematopoietic cell-associated antigens, i.e. CD2, CD4 and CD7, which were observed in four of 11 analysed patients (P < 0.01). Patients with CD33 expression were older than those without CD33 (P < 0.01). Although CD33 was the strongest predictor of a poor outcome (event-free survival, 44% for CD33+ and 75% for CD33-patients; P = 0.0041) in univariate analysis, multivariate analysis did not demonstrate significance (P = 0.0645). Fourteen of 38 patients (36.8%) at relapse or in a refractory state showed CD33 expression. Analysis of CD33 expression had also been performed at onset in 16 of these patients and showed acquisition of CD33 in six of 13 patients who had been negative for this antigen at onset. Thus, it seems that CD33+ B-precursor ALL is derived from undifferentiated cells minimally committed to B-cell lineage and more homogeneous than so-called My+ B-precursor ALL with regard to the clinical and biological features. The frequent expression of CD33 on the cells which acquired resistance to chemotherapy may have resulted from expansion of a CD33+ original minor clone or clonal evolution.

The T-cell receptor beta-chain (T beta) gene expression was examined in 16 children with B-lineage acute lymphoblastic leukemia (ALL), including eight patients with rearrangement of the T beta gene as well as immunoglobulin (Ig) heavy chain gene rearrangement. In contrast to the 1.3 kb full-length transcripts of the T beta observed in T-lineage leukemia and lymphoma cells, no transcript of the T beta gene was detected in 10 patients, including four with T beta gene rearrangement. Low levels of T beta transcripts were found in three patients with T beta gene rearrangement and two patients without T beta gene rearrangement, but those transcripts were truncated. In contrast to those findings, a single patient with T beta gene rearrangement showed abundant 1.3 kb T beta transcripts. These data indicate that T beta gene expression is not restricted to T-lineage cells and demonstrate the heterogeneity of B-lineage ALL at the expression level of the T beta gene. Our findings also suggest that T beta gene expression is not always accompanied with T beta gene rearrangement.

Nine children with mediastinal non-Hodgkin's lymphoma (NHL) were treated according to our new regimen which is characterized by intensified therapy with high-dose cytosine arabinoside (HDCA). After induction therapy with a combination of five drugs, such as vincristine, doxorubicin, cyclophosphamide, 1-asparaginase, and prednisolone, intermediate dosages of methotrexate (MTX) (1 g/m2) and HDCA (1.5 g/m2 x 12 doses) were administered. All but one patient (88.9%) achieved complete remission and then received this intensified therapy. With a median follow-up period of 25.5 months, five patients are still in complete remission, but three patients have relapsed. From the phenotypic point of view, these relapsed patients showed only very immature T-cell differentiation antigens such as CD2 and CD7 (or CD5). These results suggest that HDCA as intensified therapy for children with mediastinal NHL seems to be effective. However, for patients with an immature phenotype of T-lineage cells, more sophisticated regimens should be prepared.

A 2-year-old boy with B-lineage non-Hodgkin's lymphoma is described. He presented with growing skin tumors on his head, and biopsy specimens showed a malignant lymphoma of diffuse lymphoblastic type. Sixty-four percent of bone marrow cells were replaced with lymphoblasts, and they expressed B-lineage markers (CD19 and HLA/DR). Southern blot analysis demonstrated immunoglobulin heavy chain gene rearrangements (two rearranged and one germline) with the germline configuration of the T-cell receptor beta chain gene. Ten months later he relapsed with blasts of M5 morphologic type and a myeloid phenotype with the germline configuration of the immunoglobulin genes. During the next 2 months, myeloid blasts with immunoglobulin gene rearrangement which was identically rearranged with one of the two rearranged bands detected at diagnosis appeared. The most likely explanation for these findings is that initially the patient seemed to have at least two different clones of blasts, and clonal selections occurred during the treatments.

A 10-year-old boy with chronic active Epstein-Barr virus (EBV) infection developed T-cell lymphoproliferation in the terminal stage of hepatic failure. The phenotypes of the proliferating lymphocytes were CD3+, CD4+ and HLA-DR+. The genomic DNAs from these cells demonstrated two rearranged T-cell receptor beta-chain genes and contained the EBV genome. These findings indicate that EBV can infect T lymphocytes and might cause clonal T-cell lymphoproliferation.

On the basis of negativity for myeloperoxidase (MPO) and absence of lineage-associated antigens on the cell surface, 11 children were diagnosed as having acute undifferentiated leukemia. To analyze the molecular events associated with hematopoietic cell differentiation, we analyzed the configuration of the immunoglobulin (Ig) and T-cell receptor (TCR) delta, alpha, gamma, and beta genes in these patients. In parallel, transcription of the genes for MPO, terminal deoxynucleotidyltransferase (TdT), CD3-gamma, Ig-mu, TCR-gamma, and beta was also examined. Six patients showed rearrangements of both the Ig heavy (H) and TCR-delta genes, frequently accompanied with Ig-kappa, TCR-alpha, gamma, and beta gene rearrangements. These findings indicated that the leukemic cells from the six patients had been committed to the lymphoid lineage. This concept was supported by the presence of TdT transcripts in three analyzed specimens from these patients. In contrast, the remaining five patients did not display rearrangements of the Ig or TCR genes, and TdT transcripts were undetectable in two patients tested. MPO transcripts were not detected in four patients analyzed, thus providing no evidence of myeloid differentiation. After hybridization with the CD3-gamma gene, three of six patients showed transcription of the CD3-gamma gene. In addition to CD3-gamma transcripts, one patient with rearrangements of the Ig-H, TCR- delta, alpha, gamma, and beta genes also had full-length TCR-beta and gamma transcripts, indicating a T-precursor-cell origin of the leukemic cells from this patient. The Ig and TCR genes were in the germline configuration in the other two patients with CD3-gamma transcripts. One of them did not express the CD7 antigen but did express the CD33 antigen on the cell surface, suggesting that CD3-gamma transcription may not always be an event restricted to cells differentiating along the T-cell lineage.