Epidemiology of malignant neoplasms in soft tissue during childhood.

Employing the data of the Hokkaido Children's Cancer Registry, incidences of malignant neoplasms in soft tissue during childhood were estimated. During 25 years (1975--1999), a total of 110 such cases (0-14 years of age) were registered. The average incidence was 0.40 per 100,000 population aged 0-14 years. The incidence of rhabdomyosarcoma (RMS) was 0.24, accounting for 60% of all malignant neoplasms in soft tissue. 30% of the malignant neoplasms in soft tissue were found in the head/face/neck. The share of RMS in the upper limbs was 76%, while that in the chest was 25%. Mortality due to malignant neoplasms in soft tissue was estimated using the data of vital statistics of all Japan. The average mortality of all malignant neoplasms in soft tissue was about 0.15 per 100,000 population aged 0-14 years, which was almost constant throughout the 25 years. Both the mortality at 5-9 years of age and that at age 10-14 years increased, while that at 0-4 years had a tendency to decrease. The prolongation of the survival period in the 0-4-year age group due to the progress in therapy may have brought about a shift of the age at death to older age groups, leading to a decrease in the mortality at 0-4 years of age.

Is the mass screening for neuroblastoma ineffective?

Though a recent study (Schilling et al. 2002) concluded that the mass screening for neuroblastoma targeting children age 12 months was ineffective, we pointed out several serious problems and reestimated its effectiveness using their data. They employed the subjects in the "control area" as controls, not the "non-participants" whose biases are fewer because their area is the same as that of the participants. The incidence of neuroblastoma among the subjects in the "control area" was about 25% smaller than that of the "non-participants". This leads to underestimation of the effectiveness of the mass screening. They combined false negatives with true positives to calculate the incidence of the "screened group". But since many spontaneous regression cases are included in the true positives, this method inflates the incidence of the "screened group", leading to underestimation of the effectiveness of the mass screening. When the false negatives are compared with the non-participants, the incidence of the cases in stage 4 among the latter is about 40% of that of the former, and the mortality is less than two-thirds. The percentage of spontaneous regression cases among the true positives is estimated to be about 40%. These results are better than those of the Japanese screening programs (targeting infants age 6 months), supporting the effectiveness of mass screening for neuroblastoma.

Contribution of HPLC mass screening for neuroblastoma to a decrease in mortality.

The aim of this study is the estimation of the contribution of HPLC mass screening for neuroblastoma to the decrease in deaths due to this disease. The mortality rates of malignant neoplasms of the adrenal glands (ICD 9, 1940; ICD 10, C74; virtually all the cases of these codes are neuroblastoma during childhood) at 1-4 years of age in cohorts born in 1979-1984, 1985-1988, and 1989-1992 in the whole of Japan were calculated, using data obtained from the Ministry of Health and Welfare. The numbers of infants screened by HPLC in the cohorts were estimated through the reports of the Ministry of Health and Welfare and the database of the Japanese Society for Mass-screening. The mortality of the cohort born in 1989-1992, in which 77.8% of the live births were screened by HPLC, was 1.73 per 100,000 live births. This is about half of that (3.26) of the cohort born in 1979-1984, in which few infants were screened. On the assumption that cases of the 1985-1988 and 1989-1992 cohorts died according to the mortality rate of the 1979-1984 cohort, the expected numbers of deaths were estimated; that for the 1985-1988 cohort was 178.51 (of them, that for the infants screened by HPLC was 39.65), and that for the 1989-1992 cohort was 159.78 (of them, that for the infants screened by HPLC was 124.33). The observed numbers of deaths were 145 and 85, respectively. Assuming that non-HPLC methods have no effects and using 2 unknown quantities x (contribution of HPLC) and y (other factors), simultaneous equations (1) 178.51 - 39.65x - 178.51y = 145 and (2) 159.78 - 124.33x - 159.78y = 85 were made. Solving them, x = 0.5041 and y = 0.0757 were obtained. In conclusion HPLC screening targeting infants aged 6 months reduces death of adrenal neuroblastomas at 1-4 years of age by about 50%.

Mutations/deletions of the WT1 gene, loss of heterozygosity on chromosome arms 11p and 11q, chromosome ploidy and histology in Wilms' tumors in Japan.

Incidence rates of Wilms' tumor (WT) markedly differ in East Asian and Caucasian children. In the present study, we examined WT1 deletions/mutations and loss of heterozygosity (LOH) on 11p and 11q in a large number of WTs and compared our findings with those from 4 series of Caucasian WTs. Incidence rates of the subtle WT1 mutation in 3 of the 5 series of sporadic and unilateral WTs including ours were 4.3-6.2% and similar. However, gross homozygous WT1 deletion was more frequent in our series than in some others. In addition, our series tended to show a higher incidence of LOH limited to 11p13 and a lower incidence of LOH including 11p15 than the Caucasian one. These findings indicate some genetic differences in WT between the 2 regions. One of the 4 Caucasian series reported a correlation of germinal WT1 mutation with the predominantly stromal histology. The present study not only confirms the correlation of germinal WT1 deletion/mutation with predominant stromal histology but also establishes a correlation with somatic WT1 deletion/mutations with predominant stromal histology. While WTs with WT1 abnormalities usually showed pseudodiploidy and predominant stromal histology, those without WT1 abnormalities showed various chromosome numbers and histologic subtypes.

Prognostic impact of CD45 antigen expression in high-risk, childhood B-cell precursor acute lymphoblastic leukemia.

To evaluate the clinical implications of CD45 expression in acute childhood lymphoblastic leukemia (ALL), we measured the CD45 expression of blast cells from 133 untreated patients with childhood B-precursor ALL (n = 118) or T-ALL (n = 15). CD45 expression (> or = 20%) was detected in all 15 cases (100%) of T-ALL, and 101 cases (86%) of B-precursor ALL. In 122 cases, the fluorescence intensity of the CD45 expression was measured as a relative value; the ratio of average linear values (RALV) of CD45 on the blasts to that on CD3-positive T-lymphocytes from the same specimen. The expression was more intense in the T-ALL cases than in the B-precursor ALL cases (RALV, mean +/- SE: T-ALL 0.230 +/- 0.04 vs. pro-B ALL 0.150 +/- 0.012/pre-B ALL 0.153 +/- 0.019, p < 0.05). However, the intensity of the CD10, CD19, CD20 and CD34 antigen immunoreactivity did not correlate with the CD45 expression. Patients with hyperdiploidy (chromosome number > 50) showed significantly lower levels of CD45 expression than patients with t(1;19) or normal karyotypes (RALV, mean +/- SE: 0.081 +/- 0.022 vs. 0.133 +/- 0.03/0.143 +/- 0.019, p < 0.05). Other clinical features such as age, gender and WBC count did not correlate with CD45 expression. The prognostic implications of CD45 expression were studied in non-high-risk (low-risk + intermediate-risk) (n = 60) and high-risk patients (n = 52) with B-precursor ALL who had been treated with the risk-directed protocol of ALL-941 trial. Although CD45 expression did not correlate with the event-free survival (EFS) of the non-high-risk patients, there was a significant correlation between the expression levels and the EFS of the high-risk patients: the 3-year EFS rate of the CD45low group (n = 26, RALV = 0.017-0.132) was 88 +/- 7% versus the CD45high group (n = 26, RALV = 0.133-0.450) at 34 +/- 24% (p < 0.05). These results show that the levels of expression of the CD45 antigen on leukemic lymphoblasts are significantly correlated with the clinical features and prognosis of childhood ALL.

Development of acute myeloid leukemia after chemotherapy for osteosarcoma.

The current study describes two patients with osteosarcoma who had acute myeloid leukemia develop after treatment with multiagent chemotherapy. The incidence density for the chemotherapy protocol was 129.8 per 10,000 person-year of followup. Karyotype analysis of 16 reported patients (including the current two patients) indicated that most leukemias after treatment of osteosarcoma correlated with the use of topoisomerase II inhibitors, such as doxorubicin. The deoxyribonucleic acid-damaging activity of doxorubicin reinforced by the use of alkylating agents is highly suspected as a causative event in the development of leukemia after treatment of osteosarcoma. As the next step in the development of treatment for patients with osteosarcoma, the type and intensity of treatment must be evaluated to minimize possible leukemogenic effects without compromising the potential for cure.

Frequent increase of DNA copy number in the 2q24 chromosomal region and its association with a poor clinical outcome in hepatoblastoma: cytogenetic and comparative genomic hybridization analysis.

In a cytogenetic and comparative genomic hybridization (CGH) study of 38 hepatoblastomas, we found gain of 1q in 17 tumors (44.7%), that of 2 / 2q in 14 (36.8%), that of 20 / 20q in 9 (23.7%) and that of 8 / 8q in 8 (21.0%), loss of 4q in 4 (10.5%) and no DNA copy changes with normal karyotype or no mitotic cells in 11 (28.9%). Eleven tumors with 2 / 2q gain detected by CGH had a total chromosome 2 gain, a partial 2q gain, or a total chromosome 2 gain with an augmented partial 2q region; the common region for DNA copy gain was 2q24. Two-color fluorescence in situ hybridization (FISH) analyses using probes covering the centromere of chromosome 2 or HOXD13 (2q31) confirmed the CGH findings, and showed that the common region for gain in 2q was centromeric to HOXD13. Event-free survival (EFS) +/- standard error (SE) at 5 years was lowest in patients with 2q gain [37 +/- 15%], highest in those with no DNA copy changes [82 +/- 12%], and intermediate in those with DNA copy changes other than 2q gain [74 +/- 13%] (P = 0.0549). Multivariate analysis showed that 2q gain was an independent factor predicting a poor outcome. These findings suggest the presence of a growth-promoting gene or an oncogene in the 2q24 chromosome band, and a tumor suppressor gene in terminal 4q, which have important roles in the development and progression of hepatoblastoma.

Monitoring of minimal residual disease in children with acute promyelocytic leukemia by RT-PCR detecting PML/RARalpha chimeric gene: a retrospective study of clinical feasibility.

We studied retrospectively the clinical feasibility of minimal residual disease (MRD) monitoring by reverse transcription-polymerase chain reaction (RT-PCR) detecting the PML/retinoic acid receptor alpha (RARalpha) chimeric gene in children with acute promyelocytic leukemia (APL). MRD monitoring of APL was performed with standard and nested RT-PCR for PML/RARalpha gene, the sensitivity of which was 1 leukemic cell in 10(3)-10(4) and 1 in 10(4)-10(5) cells, respectively. Patients were nine children with APL (average age: 8.3 year; average period of follow-up: 69.2 months) who, after achieving remission with all-trans retinoic acid (ATRA), received treatment either with multidrug chemotherapy or with a combination of chemotherapy and ATRA. Out of six patients treated with multidrug-combined chemotherapy, two patients exhibited PCR positivity after six months of post- remission therapy, which shifted from the detectable range of the nested PCR to that of the standard PCR. These two patients subsequently relapsed and, together with two of the other patients receiving multidrug-combined chemotherapy, underwent allogeneic bone marrow transplantation. No MRD was detected in these patients after transplantation. In the remaining three patients who underwent cyclic treatment with alternative chemotherapy and ATRA, two showed positive RT-PCR at the nested or standard level, respectively, after six months of combined therapy, and one of them relapsed. Overall, three of four patients with MRD detected in post-remission period ultimately relapsed, while all of five patients without detectable MRD had a good prognosis. These findings suggest that impending relapse may be predicted by the detection of preceding PCR positivity with an increasing quantity of the PML/RARalpha mRNA that appears beyond six months of post-remission chemotherapy, with or without combined ATRA therapy.

Congenital malformations and childhood cancer.

BACKGROUND: We investigated the epidemiology of congenital malformations and childhood cancer. PROCEDURE: By employing the cases of the Registry of Childhood Malignancies in Hokkaido Prefecture, Japan, from 1969 to 1996, the numbers of malignancies in cases (diagnosis at 0-14 years of age) with Down syndrome (DS), mental retardation (MR) excluding DS, luxatio coxae congenita (LCC), congenital heart disease (CHD) excluding DS, undescended testicle (UT), and cleft palate-lip (CPL) were calculated. Using the percentages of malignancies in the 2,349 cases without malformation, expected numbers of malignancies in the cases with the malformations were calculated. The observed numbers were statistically compared to expected ones. RESULTS: In the DS cases, leukemia developed with a significantly high frequency, but no UT cases developed leukemia. No brain tumor was preceded by DS. This could not be explained only by early death from coexisting diseases such as CHD, insofar as the CHD cases without DS developed brain tumors more frequently than expected. The ratio of acute lymphoblastic leukemia (ALL) to acute nonlymphoblastic leukemia (ANLL) was different among the malformations. The MR cases developed ANLL more frequently than expected, whereas the CPL cases developed ALL more frequently. The distribution of the age at diagnosis for Wilms' tumor was different according to the underlying malformation. CONCLUSIONS: Malformations might have some factors that inhibit or delay as well as promote the development of certain malignancies.

Epidemiology of childhood brain tumors in Japan.

Using the vital statistics issued annually by the Japanese Government from 1960 to 1994, the mortality of brain tumors during childhood (0-14 years of age) in Japan was estimated. Since there are few nationwide registries of childhood cancers with a sufficiently high registration rate, the incidence of brain tumors was calculated using the Registry of Childhood Malignancies in Hokkaido Prefecture from 1969 to 1996. Though the mortality due to malignant diseases as a whole during childhood has been decreasing, and though there should be progress in therapeutic methods, the mortality due to brain tumors in children has been increasing. The incidence of brain tumors in Hokkaido Prefecture has been increasing as well. There was no tendency for the incidence of medulloblastoma to decrease. The incidence of childhood brain tumors has been increasing in Japan, though the cause is unknown. The period of the present study corresponded to a period of high economic growth. Thus, a study on environmental factors would be interesting.

Correlation of chromosome abnormalities with presence or absence of WT1 deletions/mutations in Wilms tumor.

Of 40 Wilms tumors with chromosome abnormalities, 6 were hypodiploid, 10 were pseudodiploid, 7 were hyperdiploid with 47 to 49 chromosomes, and 17 were hyperdiploid with 50 or more chromosomes, mostly including +12. WT1 deletions/mutations were found in one hypodiploid, eight pseudodiploid, and one hyperdiploid (47-49 chromosomes) tumor, but in none of the hyperdiploid (> or =50 chromosomes) tumors. Of the 10 tumors with WT1 abnormalities, 6 had a homozygous WT1 deletion, 1 had a nonsense WT1 mutation and loss of heterozygosity at 11p, 1 had an intragenic hemizygous WT1 deletion without detectable WT1 mutation, and 2, which occurred in Wilms tumor-aniridia-genitourinary abnormalities-mental retardation syndrome patients, had a hemizygous deletion and a missense or frameshift mutation of WT1. Six of the nine tumors with homozygous or hemizygous WT1 deletions had chromosome aberrations involving chromosome band 11p13 in one of the two chromosomes 11. While one hypodiploid and one pseudodiploid patient died of the disease, and one hyperdiploid (47-49 chromosomes) patient was alive in nonremission, all hyperdiploid (> or =50 chromosomes) patients had no evidence of disease at the last follow-up. Our data show that chromosome aberrations are closely correlated to WT1 abnormalities and suggest that hyperdiploid (> or =50 chromosomes) Wilms tumors may be characterized by the absence of WT1 abnormalities and possibly also by a favorable prognosis.

Improvement in CNS protective treatment in non-high-risk childhood acute lymphoblastic leukemia: report from the Japanese Children's Cancer and Leukemia Study Group.

BACKGROUND: Prevention of central nervous system (CNS) leukemia by early introduction of therapy to this sanctuary site is an essential component of modern treatment strategy for acute lymphoblastic leukemia (ALL). However, the optimal form of preventive CNS therapy remains debatable. PROCEDURE: To address this issue, we evaluated the efficacy of CNS preventive therapy for 572 children with ALL who achieved complete remission in the Children's Cancer and Leukemia Study Group (CCLSG) ALL874 (1987-1990) and ALL911 (1991-1993) studies. They received risk-directed therapy based on age and leukocyte count. In the ALL 874 study, the non-high-risk (low-risk [LR] + intermediate risk [IR]) patients were randomly assigned to the conventional cranial irradiation (CRT) regimen (L874A and I874A) and the high-dose methotrexate (HDMTX) regimen without CRT (L874B and I874B). The former patients received 18-Gy CRT plus 3 doses of intrathecal (i.t.) MTX and the latter patients received 3 courses of HDMTX at 2 g/m2 plus 13 doses of ITMTX (L874B) or 4 courses of HDMTX at 4.5 g/m2 plus 1 dose of ITMTX (I874B). RESULTS: The 7-year probabilities (+/- SE) of CNS relapse-free survival were 97.3% +/- 2.6% (L874A, n = 41) vs. 90.3% +/- 5.3% (L874B, n = 39) (P = 0.25) in the LR patients, and 100% (I874A, n = 55) vs. 78.5% +/- 6.5% (I874B, n = 54) (P = 0.002) in the IR patients. The corresponding disease-free survival (DFS) rates were 79.4% +/- 6.5% vs. 74.4% +/- 7.3% (P = 0.62) in the LR group and 63.3% +/- 6.8% vs. 58.3% +/- 7.2% (P = 0.66) in the IR group. Thus, the HDMTX regimen could not provide better protection of CNS relapse as compared with the CRT regimen, although their overall efficacy was not significantly different. In the ALL 911 study, intensive systemic chemotherapy with extended i,t, injections of MTX plus cytarabine achieved a high CNS relapse-free survival (98% +/- 1.9% at 7 years) and a favorable DFS (85.5% +/- 5% at 7 years) in the IR patients. The patients in the high-risk (HR) group in both ALL874 and ALL911 studies received the 18-Gy or 24-Gy CRT with intensive systemic chemotherapy. Their 7-year probabilities of CNS relapse-free survival ranged from 88% to 95%, among which the T-ALL patients had a risk of CNS leukemia, which was 3-4 times higher compared with B-precursor ALL patients. CONCLUSIONS: These results indicate that long-term intrathecal CNS prophylaxis as well as appropriate systemic therapy for the non-high-risk patients can provide protection against CNS relapse equivalent to that provided by cranial irradiation.

[Studies of childhood non-Hodgkin's lymphoma--treatment results with the CCLSG NHL 960 protocol. Children's Cancer and Leukemia Study Group (CCLSG)].

We report here on the preliminary treatment findings of a CCLSG NHL 960 study that was initiated in March 1996. In this study, 37 patients with non-Hodgkin's lymphoma were assigned to 4 different treatment groups according to disease stage and histology: (1) localized disease; (2) advanced disease, lymphoblastic type; (3) advanced disease, large cell type; and (4) advanced disease, Burkitt type. The first three groups received the modified protocols of the NHL 890 study. Groups 1 and 3 received COPADM induction therapy (CPM, VCR, PRD, ADR, and MTX). After achieving remission, Group 1 received only maintenance therapy consisting of alternate administration of 7 drugs, while Group 3 received additional intensification therapy with combination chemotherapy consisting of MTX and Ara-C, followed by a maintenance phase involving the administration of 9 drugs. Group 2 received COPADL induction therapy (CPM, VCR, PRD, ADR, and LASP) and consolidation/intensification therapies followed by a maintenance phase. Group 4 received short-term intensive COPADM polychemotherapy. Twelve patients with localized with localized disease (stage I-II) and 25 patients with advanced disease (stage III-IV) were enrolled in this study. Except for 2 patients in the advanced disease stages who died earlier in the course of the study, all patients remained in remission.

Long survivors with Ki-1 lymphoma having t(2;5) (p23;q35). Does the presence or absence of t(2;5) influence the prognosis of patients with Ki-1 lymphoma?

We experienced three patients with CD30+ diffuse large cell lymphoma having chromosomal abnormalities. The first patient was an 8-year-old girl with bilateral cervical lymphadenopathy. A biopsy of a cervical lymph node revealed diffuse large cell lymphoma (stage III), positive for CD30 and a chromosomal abnormality, t(2;5). She attained a remission and is now in complete remission 108 months after diagnosis, despite frequent relapses. The second patient was a 13-year-old boy with right axillar and supraclavicular lymph-node adenopathy. A biopsy of a cervical lymph node revealed diffuse large cell lymphoma (stage III), positive for CD30 and a chromosomal abnormality, t(2;5). He attained remission and was in continuous first remission 112 months after diagnosis. The third patient was an 11-year-old boy with fever and bilateral cervical lymph node revealed diffuse large cell lymphoma (stage III), positive for CD30 and chromosomal abnormality without t(2;5). He showed a very aggressive clinical course. Only the patients with Ki-1 lymphoma having t(2;5) survived over 100 months from the diagnosis, despite the advanced stage of the disease. These findings and a review of the literature showed that the presence or absence of t(2;5) may influence the outcome of Ki-1 lymphoma.

[Bone marrow relapse in high-risk pediatric patients with acute lymphoblastic leukemia: a comparison of relapse times and initial clinical features of patients on different protocols. Children's Cancer and Leukemia Study group (CCLSG)].

To clarify the efficacy of modern intensive chemotherapy for ALL patients with unfavorable features, we compared the time to failure and initial clinical features of children who relapsed in the bone marrow or combined sites, as documented by early CCLSG studies (H811 and H851; 1981-1987) and later studies (H874 and H/HH911; 1987-1993) concerning high-risk ALL patients. In the later studies patients outcomes with new intensive regimens employing early intensification and reinduction therapy were apparently better than those of patients in the early studies with conventional regimens. When we compared the number of relapsed patients based on duration of first remission, we found that the improved outcomes for patients in the later studies were due to a decrease in the number who relapsed 7-36 months after the start of treatment (intermediate relapse), and that the percentage of those who relapsed within the first 6 months of therapy (early relapse) was higher. Patients with high initial WBC counts tended to relapse much earlier than those with low initial WBC counts. However, in the later studies, patients with high WBC counts often relapsed after the termination of therapy (late relapse). These results suggest that the intensive chemotherapy regimens used in the later studies can prevent the development of drug resistant leukemic clones, except in extremely high-risk patients likely to relapse within the first 6 months of therapy.

Mass screening for neuroblastoma by HPLC and by non-HPLC: a review.

There have been a number of contradictory views concerning the efficacy of mass screening for neuroblastoma. Three Japanese and one cohort study from Quebec, and three Japanese cross-sectional studies were reviewed. The four cohort studies revealed that mass screening using high performance liquid chromatography (HPLC) reduced the incidence of this disease by about half in children aged from 1 to 4 years while little reduction was noted in screening utilizing non-HPLC methods. There is a large difference in the efficacy of mass screening depending on the method used. In the three cross-sectional studies, there was no discrimination between screened and unscreened cases, or between subjects screened by HPLC and non-HPLC methods. The percentage of children in the cross-sectional studies screened by HPLC was low. Cross-sectional studies seem inappropriate in assessing the effectiveness of current HPLC mass screening.

Mutations in the E-domain of RAR portion of the PML/RAR chimeric gene may confer clinical resistance to all-trans retinoic acid in acute promyelocytic leukemia.

The binding of all-trans retinoic acid (ATRA) to the ligand-binding region in the E-domain of retinoic acid receptor-alpha modifies the transcriptional activity of RARalpha protein. ATRA probably induces differentiation of acute promyelocytic leukemia (APL) cells by binding to the E-domain of the RARalpha portion (RARalpha /E-domain) of PML/RARalpha chimeric protein. Therefore, molecular alteration in the RARalpha /E-domain of the chimeric gene is one mechanism by which patients with APL may acquire resistance to ATRA therapy. In this study using reverse transcription-polymerase chain reaction and single-strand conformation polymorphism, DNA segments amplified from the RARalpha /E-domain in fresh APL cells of 23 APL patients (8 males and 15 females from 4 to 76 years of age) were screened for mutations. Of those patients, 3 patients (1 with de novo and 2 with relapse) had clinical resistance to ATRA therapy. We found mutations in the RARalpha /E-domain of PML/RARalpha chimeric gene exclusively in the 2 patients who exhibited ATRA-resistance at relapse, whereas the mutations were not detected at their initial onset. Interestingly, these patients received a prolonged or intermittent administration of ATRA before relapse with ATRA-resistance. The mutations lead to the change of amino acid in the ligand-binding region of RARalpha /E-domain, Arg272Gln, or Met297Leu according to the amino acid sequence of RARalpha, respectively. Further study demonstrated that the in vitro ligand-dependent transcriptional activity of the mutant PML/RARalpha protein was significantly decreased as compared with that of wild-type PML/RARalpha. These findings suggest that mutations in the RARalpha /E-domain of the PML/RARalpha chimeric gene may confer clinical resistance to ATRA therapy in patients with APL.

[Treatment of children with non-Hodgkin's lymphoma with CCLSG NHL 855/890 protocols long-term outcome and incidence of secondary malignancies].

We report here on treatment results of consecutive CCLSG NHL studies (NHL855, 1985-1989; NHL890, 1989-1996). The NHL855 protocol consisted of an induction phase of five drugs (VCR, PRD, CPM, DXR, and high-dose MTX) and a maintenance phase of 7 drugs. The probabilities of EFS at 7 years were 78% (SE, 10%) for the patients with localized disease, and 38% (SE, 7%) for those with advanced disease. In the NHL 890 protocol, the patients were assigned to two different treatment groups according to their histology and received different consolidation therapy; non-lymphoblastic subtype was treated almost identically to NHL855 while LASP and VP-16 were newly added for the lymphoblastic subtype. The 7-year EFS improved to 91% (SE, 6%) for localized disease, and 61% (SE, 6%) for advanced disease. A remarkable improvement was particularly evident for lymphoblastic type with mediastinal mass. Optional trial of high-dose sequential chemotherapy and peripheral blood progenitor cell auto grafting resulted in an unfavorable outcome. The 7-year EFS according to main histological subgroups were as follows: 84% (10%) for large cell type, 67% (11%) for Burkitt's-type, 58% (10%) for lymphoblastic type. Secondary cancer occurred in two of the 163 patients studied. Both patients were AML (M0/M4) and MLL rearrangement was detected in the M4 case.

[Prognostic implication of DNA contents on long-term outcome of childhood acute lymphoblastic leukemia].

Prognostic value of cellular DNA content was evaluated in 189 children with acute lymphoblastic leukemia. Treatment outcome of the three different DNA index (DI) groups (Group A, DI = 1.0 vs. Group B, DI 1.01-1.15 vs. Group C, DI > or = 1.16) was compared between the two treatment risk groups (standard-risk and high-risk groups) stratified by the initial leukocyte count and age. In the standard-risk group, these groups had 10-year event free survival (EFS) rate (SE) of 62% (6%), 40% (21%) and 87% (6%), respectively (p < 0.05). In the high risk group, they had 10-year EFS rate of 30% (5%), 33% (27%) and 60% (19%), respectively (p < 0.01). Use of the DI, leukocyte count and age may be sufficient to distinguish the patients with an extremely low risk of failing to the standard ALL therapy from the patients with a relatively high-risk of treatment failure.

[Childhood acute promyelocytic leukemia treated with all-trans retinoic acid].

We treated two children with acute promyelocytic leukemia (APL) in whom complete remission was successfully induced by oral administration of all-trans retinoic acid (ATRA). We followed these patients with conventional chemotherapy. The first patient has remained in continuous complete remission. However, the other patient relapsed during the maintenance therapy and died of progressive disease in spite of a second treatment with ATRA and chemotherapy. From a clinical point of view, the latter case had a hyperleukocytosis on admission. Also morphologically speaking, this patient had a different M3 variant than the first case. There are two major isoforms of PML/RAR alpha transcripts, so called short and long type transcripts, according to the breakpoints in the PML genes. In the first case the "long type' isoform was detected by reverse transcriptase polymerase chain reaction (RT/PCR) amplification. On the other hand the "short type' isoform was observed in the latter case. Also the second case became PCR positive at relapse, although the detectable isoform was negative during remission. The "short type' isoform may be related to the poor prognosis and RT/PCR analyses may be a powerful to detect early relapse.