Heterogeneity of TdT+, HLA-DR+ acute leukaemia: immunological, immunocytochemical and clinical evidence of lymphoid and myeloid origin

The role of immunophenotype in acute lymphoblastic leukemia of infant age

The prognosis of acute lymphoblastic leukemia in infants is still significantly worse than that of older children. This is thought to be due to both clinical and biological factors, such as high white blood cell (WBC) counts at diagnosis, irregular or immature phenotypes, and molecular and cytogenetical abnormalities. In order to focus the significance of immunophenotypic analysis, we have reviewed the immunophenotypic studies of 145 infants under 18 months of age treated at the AIEOP centers from 1984 to 1992. Children have been divided in three age groups of six months each; WBC count at diagnosis has been evaluated both as mean values and within different categories (< 10.10(9)/L, > 100.10(9)/L). These have been studied in correlation with immunophenotype and with the expression of single, specific markers. A significant correlation has been found between young age, high WBC count and immature phenotypes. Common ALL was more frequent in older children and showed lower WBC counts. Moreover, event-free survival was significantly better in older children with WBC count < 100.00/mm3, with CD10+, MyAg- ALL. Therefore, we suggest that immunophenotypic analysis is still an important prognostic factor and can be usefully used, together with simple clinical data, to plan therapy for ALL in infants.

Peroxidase expression in acute unclassified leukemias: ultrastructural studies in combination with immunophenotyping

The lineage affinity of 61 cases of acute unclassified leukemias (AUL) was reevaluated by ultrastructural analysis of peroxidase expression (POEM) in combination with immunophenotyping. In 24 cases a significant proportion of the blasts displayed ultrastructural myeloperoxidase (UMPO) and in another 3 cases platelet peroxidase (UPPO) suggesting the allocation of these leukemias to the myeloid or megakaryoblastic lineage. No significant correlation between myeloid surface marker expression and POEM positivity could be detected, while the presence of the CD 19 or CD 24 antigen significantly correlated with POEM negativity. The detection of lymphoid markers on POEM+ blasts indicates, that these leukemias might be derived from bipotential progenitors, which have retained their lymphoid antigens during myeloid differentiation. In one case a bilineage leukemia with a POEM+/CD 19- and a POEM-/CD 19+ population could be identified by immunoelectron microscopical studies (IEM). The remaining 34 AUL cases were POEM negative. Combined data suggest that these cases probably derive from early lymphoid progenitors. Taken together, AUL are heterogenous in presentation and in their cellular origin including a major portion of cases with an unequivocal myeloid differentiation detectable by ultrastructural analysis of myeloperoxidase expression.

Risk-directed therapy for childhood acute lymphoblastic leukemia. Results of the Associazione Italiana Ematologia Oncologia Pediatrica '82 studies

BACKGROUND

In 1982, the Associazione Italiana Ematologia Oncologia Pediatrica (AIEOP) started its third-generation study, aiming to improve previous results obtained by AIEOP '79 study and to deliver a standardized treatment to most Italian children with acute lymphoblastic leukemia (ALL).

METHODS

We treated 902 children (older than 1 year and younger than 15 years of age) with newly diagnosed ALL in multicenter studies of risk-directed therapy (111 low risk [LR] from Study 8201; 570 average risk [AR] from Study 8202; and 117 and 104 high risk [HR] from Studies 8303 and 8503, respectively). Induction therapy was composed of vincristine, prednisone, and asparaginase for LR or AR patients and these agents plus daunorubicin, (Study 8503) or vincristine, prednisone, cytarabine, and intermediate-dose methotrexate (Study 8303) for HR patients. Central nervous system (CNS) preventive therapy consisted of intrathecal methotrexate only (LR), intrathecal methotrexate plus 18 Gy cranial irradiation (AR and HR Study 8503), or high-dose (HD) cytarabine (HR Study 8303). Reinduction therapy was vincristine/prednisone/daunorubicin for AR patients with cyclophosphamide added for HR patients in Study 8303 and HD asparaginase in Study 8503. LR patients did not receive intensification therapy. Continuation therapy comprised 6-mercaptopurine plus methotrexate and monthly pulses with vincristine plus prednisone for all patients, except for HR patients in Study 8303 who also received teniposide plus cytarabine. Weekly HD asparaginase was also given in Study 8503. Duration of treatment was 24 months for Studies 8201 and 8202, 15 months for Study 8303, and 22 months for Study 8503. The overall complete remission (CR) rate was 94.7% (97.3% for LR, 94.9% for AR, and 93.2% for HR).

RESULTS

Overall 7-year event-free survival (EFS) was 53.6% (standard error [SE], 1.8). EFS was 60.8% in LR (SE, 4.7), 60.6% in AR at 7 years (SE, 4.7), and 18.5% in Study 8303 (HR) at 5 years (SE, 3.8). Because of the poor result in HR patients, a successor study (8503) was developed that yielded a 5-year EFS of 46.1% (SE, 5.1). Site-specific relapse rates were 18.5% (LR), 13.4% (AR), 35.1% (HR on 8303), and 18.3% (HR in Study 8503) for bone marrow and 9.2%, 7.9%, 17.5%, and 19.3%, respectively, for the CNS (isolated). Isolated testicular relapse was observed in 3.9% of male patients.

CONCLUSIONS

This risk-directed therapy cured at least 50% of patients with ALL with relatively nonintensive therapy. The 80% overall survival rate for LR and AR patients at 7 years suggested that the total cure rate may be higher than 50% because of the significant salvage rate for patients who received antimetabolite-based therapy initially.

The immunophenotype in infant acute lymphoblastic leukaemia: correlation with clinical outcome. An Italian multicentre study (AIEOP)

A detailed analysis of immunophenotype of 112 infants aged less than 18 months with acute lymphoblastic leukaemia (ALL) was performed. Patients were divided into three groups on the basis of age at presentation (under 6 months: group 1: 6-12 months: group 2; 13-18 months: group 3). There were three cases of T-ALL (2.6%). The proportion of other subtypes was: common ALL in 59 patients (52.68%), pre-B ALL in 15 patients (13.3%), pre-pre-B ALL in 27 (24.1%) and acute undifferentiated leukaemia (AUL) in eight patients (7.14%). In non-T ALL, positivity to CD10 (corresponding to C-ALL and pre-B ALL) was distributed in the three age groups as follows: 38.88% (group I) 65.38% (group II) and 86.36% (group III). Conversely, immature phenotypes (pre-pre-B and AUL) were found more often in the younger patients of groups I and II, as well as anomalous phenotypes, such as the presence of myeloid antigens (MyAg) and of CD7. Prognostic significance was evaluated as event-free survival (EFS) by statistical analysis. A better outcome in CD10-positive ALL than in CD10-negative ones (48% v. 25% of long-term survivors) was demonstrated in all infants. Similarly, EFS was significantly better in MyAg-negative than in MyAg-positive cases. These results were confirmed also when adjusting for white blood cell count. This allowed the identification of CD10-negative, MyAg-positive ALL, which were relatively more frequent in infants and had a poorer clinical outcome with the current therapies. This study stresses the prognostic relevance of the immunological study in infant leukaemias and its utility in choosing different therapeutic modalities for poor risk phenotypes.

Electron microscopy: a contribution to further classification of acute unclassifiable childhood leukemia

The ultrastructural, light microscopical and immunological features of twelve cases of acute childhood leukemia are described. Nine cases were unclassifiable by light microscopy, morphology and cytochemistry, and three were difficult to classify because of a low percentage of Sudan-Black B positive blasts. By means of electron microscopy (including peroxidase cytochemistry), two main groups were seen: 1. Acute myeloid leukemia, in which could be distinguished a) a more differentiated myeloid leukemia, b) a leukemia with megakaryoblastic involvement and c) a minimally differentiated acute myeloid leukemia with granules present and 2. lymphoblastic leukemia. One case could not be classified. The first group included two possible cases of a hybrid leukemia with CD19 or CD10 positivity as well as ultrastructural peroxidase activity. We conclude that electron microscopy aids to further classification of minimally differentiated and hybrid acute leukemias.