Search

Thirty cases of newly diagnosed pediatric acute myeloblastic leukemia (AML) with French-American-British (FAB) M2 morphology were analyzed with cytogenetics and a comprehensive panel of monoclonal antibodies reactive with lymphoid-, natural killer (NK)-cell-, and myeloid-associated antigens. The t(8;21)(q22;q22), or t(8;21;V)(q22;q22;V), translocation was identified in 16 of the 30 cases. Cases with the t(8;21) did not differ significantly from the remaining M2 cases with respect to expression of CD11b, CD13, CD14, CD15, CD33, CD34, CD36, CD41a, CD42b, CDw65, TdT, or HLA-DR. Expression of the B-cell antigen CD19 was detected in 13 of the 16 t(8;21) cases (81%), but in only 1 of the 14 (7%) other M2 cases (P = .00006). Expression of the CD56 NK-cell antigen was also significantly more frequent among t(8;21) cases (63% v 14%; P = .01). Coexpression of CD19 and CD56 was found only in the t(8;21) group (9 of 16 cases, P = .0009). Furthermore, this phenotype was not found in 48 evaluable cases of de novo AML of the FAB M1, M3, M4, M5, or M7 subtypes. The 14 M2 AML cases lacking the t(8;21) commonly expressed CD2 (n = 5) or CD7 (n = 8). However, no case with the t(8;21) expressed either antigen (P = .01 and .0005, respectively). Thus, the t(8;21) biologic subgroup of pediatric M2 AML has distinct immunophenotypic characteristics that distinguish it from other types of de novo AML.

Presenting features of 120 consecutive children with T-cell acute lymphoblastic leukemia (ALL), representing 15% of all patients diagnosed as having ALL during the study period, were analyzed to determine relationships with treatment outcome. Patients' ages ranged from 1.7 to 18.8 years (median, 10.3 years) and their leukocyte counts from 1.7 to 1,070 x 10(9)/L (median, 100 x 10(9)/L). Central nervous system (CNS) leukemia was present in 12.5% of the cases, a mediastinal mass in 61%, and L2 lymphoblast morphology in 32%. A relatively high proportion of cases, 26%, had normal karyotypes at presentation. Of the cases tested, membrane CD1 expression was found in 38% of cases, CD3 in 33%, CD4 in 50%, CD5 in 94%, CD8 in 55%, and CD10 in 35%. Four presenting features were found to confer an increased risk of treatment failure: age greater than or equal to 15 years, L2 lymphoblast morphology, abnormal karyotype, and membrane CD3 expression. This study illustrates the heterogeneity of presentations of childhood T-cell ALL and suggests that the relative importance of risk factors in ALL differs according to immunophenotype and treatment strategy.

Leukemic cells from 89 (24%) of 369 children with newly diagnosed acute lymphoblastic leukemia (ALL) were found to have a pre-B immunophenotype. By comparison with blasts having the common ALL phenotype, the pre-B cells were more likely to have a DNA index less than 1.16 (P = 0.02), a pseudodiploid karyotype (P less than 0.001), and a chromosomal translocation (P = 0.001). Increased serum lactic dehydrogenase levels (P = 0.001) were also characteristic of pre-B ALL; otherwise, the clinical and laboratory features of the two groups were similar. A nonrandom chromosomal translocation, t(1;19)(q23;p13.3), was identified in blast cells from 16 (23%) of the 70 patients with pre-B ALL and adequate chromosome banding studies; different translocations were found in 11 of the remaining patients. The presence of any chromosomal translocation in the pre-B group was significantly related to a higher leukocyte count, an increased level of serum lactic dehydrogenase, an increased percentage of S-phase cells, black race, and a blast cell DNA index less than 1.16. Four presenting features were found to confer an increased risk of treatment failure among pre-B patients: pseudodiploidy, chromosomal translocation, black race, and higher serum lactic dehydrogenase level. In a multivariate analysis, pseudodiploidy emerged as the strongest factor for predicting relapse in pre-B ALL. The frequent association of chromosomal abnormalities of known adverse prognostic significance and high serum lactic dehydrogenase levels with pre-B-cell ALL explains, at least in part, the poor treatment outcome reported for children with this subtype of leukemia.

Certain types of chromosomal abnormalities have been shown to exert strong independent influence on treatment outcome in acute lymphoblastic leukemia (ALL). To identify the changes most closely associated with prognosis, we analyzed the completely banded blast cell karyotypes of 161 children with this disease. One hundred twenty-five cases had one or more chromosomal abnormalities, with 45 showing translocations. The frequency of translocations was highest (58%) among patients with pseudodiploid karyotypes and lowest (0%) in the hyperdiploid group defined by 51 or more chromosomes. During the maximum 6-year follow-up period, 30 of the 45 patients with a translocation failed therapy, compared with only 27 of the 116 who lacked this feature. Life-table estimates of event-free survival indicate that only 14% of the translocation group will be in complete remission at 3 years. The percentages of failures associated with random and nonrandom translocations were virtually identical (68% v 65%). When entered in a Cox proportional hazards model with seven other types of chromosomal abnormalities, and then with 11 clinical and laboratory variables of known prognostic value in ALL, translocation emerged as the strongest single predictor of treatment outcome (P less than 0.0001). The model indicated that translocation increases the risk of treatment failure six times by comparison with the absence of this feature. These findings offer an explanation for the majority of early treatment failures in childhood ALL, including those previously attributed to ploidy classification.

Factors contributing to the development of central nervous system (CNS) leukemia, and the impact of leukemic involvement of this site on subsequent remission length, were determined in 184 children with acute nonlymphoblastic leukemia who had been treated in two successive clinical trials. Preventive CNS therapy in both studies consisted of intrathecal methotrexate (12 mg/m2) given monthly during the first six months of therapy and then every three months until all treatment was stopped. Children with CNS leukemia at diagnosis or relapse were given intrathecal chemotherapy weekly for four weeks and then monthly throughout the remainder of the treatment course. Those continuing in complete remission received 2,400 rad cranial irradiation plus five doses of intrathecal methotrexate before cessation of therapy. The 38 children (20.7%) with CNS leukemia at diagnosis were more likely to have an initial leukocyte count greater than or equal to 25 X 10(9)/L (P = .01) and age less than 2 years (P = .03). The presence of CNS leukemia at diagnosis did not adversely affect the remission induction rate (P = .13) or the length of complete remissions (P = .73). CNS relapse ended initial remissions in 11 patients only and did not preclude subsequent long-term survival, as four of these children are off therapy and in second complete remission for 33+ to 78+ months. Three features at diagnosis were predictive of CNS relapse: monocytic or myelomonocytic leukemia (P = .002); age less than 2 years (P = .0001); and leukocyte count greater than or equal to 25 X 10(9)/L (P = .012). By stepwise Cox regression analysis, each factor was found to have independent predictive value. Despite the apparent effectiveness of intrathecal methotrexate as preventive CNS treatment, our findings indicate that more effective prophylaxis is needed for patients with features predisposing to CNS relapse.

Leukemia cell karyotypes were determined at diagnosis for 136 of 159 consecutive patients with acute lymphoblastic leukemia (ALL) who were followed for up to 35 mo. Ninety patients (67%) had abnormal karyotypes. Five chromosome categories were designated, based on the distribution of modal numbers: hyperdiploid greater than 50 (n = 41), hyperdiploid 47-50 (n = 18), pseudodiploid (n = 28), normal (n = 46), and hypodiploid (n = 3). Treatment response was assessed for the categories in terms of time to failure (induction failure, first relapse, or death). Children in the hyperdiploid greater than 50 category had the best responses to treatment, with only 2 failures, and those in the pseudodiploid category had the poorest (p less than 0.001). The remaining 3 chromosome categories had intermediate responses and formed a third prognostic group. This same influence of chromosome number on time to failure was evident within the 2 clinical prognostic groups: high risk, signified by a leukocyte count greater than 100 X 10(9)/liter, meningeal leukemia, mediastinal mass, or the presence of blasts that formed rosettes with sheep erythrocytes at 37 degrees C, and standard risk, indicated by the absence of these features. The influence of chromosome number on time to failure was also the same within the historically favorable prognostic group that had common ALL. Results of a multivariate analysis indicated that chromosome number was the strongest single predictor of outcome (p less than 0.001) and was the only variable that added significant prognostic information to leukocyte count (p less than 0.001). The combination of chromosome number and leukocyte count should more clearly distinguish patients with ALL at low or high risk of relapse.

Clonogenic cells from 41 children with newly diagnosed acute myeloid leukemia (AML) were tested in vitro for their sensitivity to cytarabine (Ara-C) and daunorubicin (DNR). The findings were then compared with the patients' responses to induction chemotherapy that uniformly included Ara-C and DNR. Light-density marrow cells were incubated with either or both drugs for one hour and cultured over leukocyte feeder layers; clusters and colonies were scored on days 7, 10, and 14. Only the percentage of cell kill in the presence of 1.8 mumol/L DNR was significantly associated with responses to induction therapy: median of 45% (range, 0% to 98%) for patients achieving complete remission v 16% (range, 4% to 23%) for nonresponders (P = .007). The relationship between clonogenic cell kill less than or equal to 23% and clinical responses was striking. Of the 11 evaluable patients with in vitro findings in this category, ten either failed induction therapy or relapsed within 1 year after attaining remission. Kaplan-Meier analysis of relapse-free survival times indicated longer durations of remission for patients whose blast cells showed increased sensitivity in vitro to Ara-C alone, DNR alone, or a combination of the two agents. Seven of 11 patients with cell kills of greater than or equal to 49% in the presence of 1.25 mumol/L Ara-C remain free of leukemia, compared with only one of 12 whose cells were less sensitive to the drug (P = .006). We conclude that the in vitro sensitivity of clonogenic leukemic progenitors to DNR and Ara-C correlates with treatment outcome in children with newly diagnosed AML.

Transient leukemoid reactions that resemble acute leukemia have been well described for infants with trisomy 21 (Down syndrome). We report a phenotypically normal 3-day-old boy with hepatosplenomegaly, leukocytosis, and circulating myeloblasts. On chromosome analysis, trisomy 21 was found in all blood and bone marrow cells. However, only 4% of cultured skin fibroblasts were trisomic and the other 96% were normal, thus indicating mosaicism. Without treatment, the leukocyte count gradually returned to normal and the organomegaly diminished. Subsequently, chromosome analysis of blood and bone marrow disclosed a predominance of cells with a normal karyotype. These findings suggest that mosaicism could be responsible for the transient leukemoid reactions in some newborns--i.e., the trisomic cells may temporarily gain a proliferative advantage over the normal cells, perhaps by inhibiting their growth. Serial cytogenetic studies, as well as chromosome analysis of more than one tissue, may help to distinguish transient leukemoid reactions from acute leukemia in infants.

We studied the clinical and biologic features of 10 cases of acute leukemia that met standard French-American-British (FAB) criteria for acute myeloid leukemia (AML) but in which the blast cells also expressed the T-cell-associated CD2 surface antigen. All cases had greater than 3% myeloperoxidase and Sudan black B-positive leukemic blasts, and blasts from seven cases contained Auer rods. Reactivity of the cells with a panel of monoclonal antibodies (MAbs) indicated that leukemic cells in all cases expressed myeloid-associated (CD11b, CD13) surface antigens, further supporting the diagnosis of AML. However, blasts from every patient coexpressed the T-cell-associated surface CD2 and CD7 as well as cytoplasmic CD3 antigens. Blasts from five patients expressed surface CD25, whereas blasts from only one expressed surface CD3. Five patients had rearranged T-cell receptor beta-chain genes, whereas only three had rearranged T-cell receptor gamma-chain genes. This pattern of lineage-related gene expression appears to define a distinct subtype of AML with T-lymphoid features (CD2+ AML) and could reflect either aberrant gene expression in leukemic blasts or transformation of a pluripotent stem cell having a flexible pattern of gene expression. Clinically, these 10 patients presented at an older age with a higher leukocyte count and a higher frequency of lymphadenopathy than did children whose blast cells were characteristic of myeloid leukemia. Patients with CD2+ AML also had poorer responses to remission induction therapy (50% v 80% entered complete remission, P = .05). However, each of the five children who failed induction chemotherapy on AML protocols had a striking response to drug combinations usually reserved for lymphoid leukemia. We conclude that this leukemia with mixed lymphoid and myeloid characteristics is a distinct biologic and clinical entity.

Sequential studies in children with acute lymphoblastic leukemia have demonstrated that at diagnosis or relapse there is defective utilization of complement by the alternative pathway. Thus, the sera of 17/18 patients fail to completely consume C3 to C9 when incubated with zymosan or cobra venom factor (CoF). This underutilization is due to a specific inhibitor of C3 activation which has been partially isolated. By remission, the inhibotor disappears and the CoF and zymosan assays return to normal. In addition, serum levels of C3 and factor B are elevated at the time of diagnosis or relapse but fall to below 3 S.D. from the mean in nearly 60 per cent of the cases during induction therapy. Similarly, serum C4 levels which are normal at diagnosis fall to less that 3 S.D. from the mean in 7/12 cases during treatment. Low C3 levels correlate well with factor B values, suggesting that if C3 to C9 are utilized after the inhibitor has been eliminated, such utilization occurs primarily through the alternative pathway. Presumably, as illustrated by the low C4 levels, this activity is mediated by the ampliciation loop of the alternative pathway involving classical pathway generation of C3b.

Leukemic cell chromosomal findings in 27 infants were analyzed. Among the 18 cases of acute nonlymphoblastic leukemia (ANLL), all but two were classified as monocytic or myelomonocytic. The remaining nine cases were acute lymphoblastic leukemia (ALL), seven lacking the common ALL antigen and two having cytoplasmic immunoglobulin (pre-B phenotype). Twenty-five cases (93%) had an abnormal karyotype, 21 (84%) being pseudodiploid. Chromosomal translocations were detected in 67% of the ANLL cases and in 78% of the ALL cases. Nonrandom chromosomal abnormalities included the t(9;11)(p21-22;q23) in three cases of monocytic leukemia, inversion of chromosome 16 in three cases of myelomonocytic leukemia with bone marrow eosinophilia, and t(4;11)(q21;q23) in one case of ALL. Chromosomal regions preferentially involved in infant leukemia included 11q23-25 (13 cases), 9p21-22 and 10p11-13. All but one of the 24 cases with chromosomal breakage or rearrangement had breakpoints that corresponded to known fragile sites, half of which were at 11q23-25, a finding that may have pathogenetic importance. The CALLA- or pre-B phenotype and the presence of chromosomal translocations in most infants with ALL provide a biological explanation for their poor prognosis.

Cellular DNA content distributions of propidium-iodide-stained bone marrow blasts were determined by flow cytometry (FCM) for 225 untreated children with acute leukemia and were correlated with leukemia cell phenotype and karyotype. Aneuploidy of the primary malignant stem line was detected in 54 cases (24%): 51 hyperdiploid and 3 hypodiploid. A second stem line with approximately twice the DNA content of the primary stem line was recognized by FCM in 28 cases (23 ALL, 5 ANLL) and may be an important source of leukemia cell heterogeneity. The degree of DNA content abnormality detected by FCM was highly correlated (r = 0.98) with the number of whole chromosome gains or losses in the leukemia karyotype. Aneuploidy detectable by FCM was more frequent in acute lymphoblastic leukemia (ALL) (52 of 173, 30.1%) than in acute nonlymphoblastic leukemia (2 of 52, 3.8%) (p less than 0.001). In the ALL group, aneuploidy was significantly correlated with the cell surface expression of common ALL antigen: 46 of 127 antigen-positive cases were aneuploid compared to 6 of 46 antigen-negative cases (p less than 0.003). Only 2 of 21 cases of T-cell ALL without common ALL antigen had detectable aneuploidy, which was significantly less than in the common ALL group (p = 0.02). The median percentage of cells in S-phase was significantly greater for B-cell and erythrocyte rosette-positive T-cell ALL, than for the other phenotypic subgroups. We conclude that aneuploidy and S-phase cell percentage are correlated with the state of leukemia cell differentiation. The biologic basis for the correlation is not established, but may be linked to the process of malignant transformation.

The clinical significance of a low percentage of myeloperoxidase-positive blast cells in childhood acute nonlymphoblastic leukemia was determined. Of 155 consecutive cases studied by cytochemical staining methods, 14 were characterized by 4% to 15% (median 6%) myeloperoxidase-positive blasts. All 14 cases showed reactivity to Sudan black B stain, and 7 had Auer rods. The morphological subtypes of leukemia were M1 (8 cases), M2 (3), M4 (1), and M5 (2). Immunological marker studies disclosed the lymphoid-associated T11 antigen on cells from 8 of the 11 cases tested. Other lymphoid-related findings in these 8 cases included the T3 antigen and E rosette formation in 1 case each. Among cases that were prospectively studied for the expression of lymphoid-associated markers, 6 of 8 with low levels of myeloperoxidase positivity compared with only 1 of 44 with higher levels (greater than 15%) possessed such features (P less than 0.001). We conclude that low levels of myeloperoxidase reactivity distinguish cases of acute leukemia in which the blast cells coexpress lymphoid (T11 antigen) and myeloid markers.

Leukemic blast cells are thought to arise from clonal expansion of a single transformed hematopoietic cell. This generality is supported by the rarity of convincing reports on acute myeloblastic leukemia (AML) with two karyotypically independent clones. Relying on sequential cytogenetic analyses, we identified such clones in two children with relapsed AML. The first case, classified as M2 leukemia in the French-American-British (FAB) classification system, had a t(8;21) (q22;q22) at diagnosis; 16 months later, at relapse, the leukemic cells had uniform morphologic features similar to those observed at diagnosis, except that two independent clones were present: one with the original t(8;21) and the other with t(11;22)(q23;q13) [corrected]). The second case was initially classified as FAB M1 leukemia with a t(8;21) (q22;q22). At relapse, 16 months later, the blast cells appeared morphologically uniform and similar to the diagnostic specimen; however, in addition to the original t(8;21) clone, there was a t(1;11) (p32;q23) [corrected]. These findings suggest that separate leukemogenic events affecting different progenitor cells can occur in rare cases of AML. The presence of two karyotypically independent clones could also be explained by multistep leukemogenesis; that is, more than one cell from a common pool of preleukemic cells could be affected by the transforming event, resulting in two independent clones. Alternatively, in light of recent reports of therapy-related leukemias with an 11q23 translocation, the new independent clone in these two patients could represent a therapy-related secondary malignancy. Thus, 11q23 translocations may occur preferentially in stem cells that are more susceptible to treatment-induced malignant transformation.

A four-drug regimen, based on cell kinetic principles, induced complete remissions in 68 of 95 children (72%) with acute nonlymphocytic leukemia (ANLL). Patients entered remission after 2-5 weekly cycles of vincristine-daunorubicin (day 1) followed by sequential cytosine arabinoside and 6-azauridine (days 4-7). With continuation therapy of monthly vincristine-doxorubicin-cyclophosphamide, weekly cytosine arabinoside, and daily 6-mercaptopurine, the median duration of complete remission was 10 mo and the median survival time 21 mo. Portal triaditis, evident in 11 of 23 patients with liver biopsies, was associated with long remissions. A larger spleen size (greater than 5 cm) and a higher myeloblast labeling index (greater than 10%) at diagnosis were clearly related to shorter durations of remission. Splenectomy within 1 mo of remission had no statistically significant effect on the frequency of relapse or length of remission. Patients without central nervous system (CNS) leukemia at diagnosis, all treated prophylactically with intrathecal methotrexate, had a low frequency of initial CNS relapse (3/56, 5%). The 2-yr disease-free survival rate is 29% (20 of 68 patients attaining complete remission). fifteen patients have completed 2.5 yr of therapy, and each remains in continuous complete remission, off treatment, for 1+ -36+ mo. This induction chemotherapy was as effective as more intensive regimens, with the advantage of less toxicity and shorter periods of hospitalization.

We treated 24 children and adolescents with stage III or IV lymphoblastic non-Hodgkin's lymphoma, using a protocol designed for patients with poor-prognosis acute lymphoblastic leukemia (ALL). Early therapy consisted of teniposide plus cytarabine administered before and immediately after prednisone, vincristine, and asparaginase. The two-drug combination was also given intermittently with continuous 6-mercaptopurine and methotrexate during the first year of continuation chemotherapy. Periodic intrathecal methotrexate and delayed cranial irradiation were used to prevent central nervous system involvement. Anthracycline compounds, alkylating agents, high-dose methotrexate, and involved-field irradiation were not used in any phase of treatment. Twenty-two (96%) of the 23 evaluable patients achieved complete remission. With a median follow-up of 2 1/2 years, only four patients have relapsed; the remainder have been disease-free for eight months to more than five years. The projected four-year continuous complete remission rate is 73% for all patients and 79% for the 19 with mediastinal involvement at diagnosis. These results demonstrate that use of teniposide plus cytarabine with an otherwise conventional plan of ALL therapy is an effective approach to the treatment of childhood lymphoblastic lymphoma.

We describe two adolescent girls with acute lymphoblastic leukemia (ALL) whose leukemia cells were near-haploid. Their lymphoblasts stained in a block pattern with periodic acid Schiff and had "common ALL" surface markers confirmed by indirect immunofluorescence. Each patient had two populations of blasts, one near-haploid and one hyperdiploid, which was an exact doubling of the near-haploid karyotype. The first patient had a predominant population of cells with 26 chromosomes and a few with 52, while the second had a predominance of cells with 56 and a minority with 28. Flow cytometric analysis of DNA content initially detected the minor near-haploid population in the second patient, which was confirmed later by cytogenetic review of the marrow sample. In addition to our two patients, only four patients have been reported with near-haploid ALL. Of these six, five were girls, five were adolescents, and five had short survivals (median, 10 mo). All six had disomy of chromosome 21 with or without disomy for chromosomes 10, 14, 18, or X (four patients each). Thus, near-haploid ALL may represent a unique subgroup of ALL with a poor prognosis. To detect these and other possible subgroups, we have included cytogenetic analysis and flow cytometric analysis of DNA content in our initial evaluation of patients with ALL.