The diagnostic and prognostic value of immunophenotyping in acute leukemia

The immunophenotype of acute myeloid leukemia: is there a relationship with prognosis?

Immunophenotyping of acute myeloid leukemia has controversial implications with regards to prognosis. Many associations have been described between individual antigen expression on myeloid blasts and prognosis, however few are consistent. Markers with a consistent prognostic association that have been demonstrated in more than one study have been CD13, CD14, and CD15. The association of the expression of CD11b with poor prognosis appears definite, as does CD7 expression. When compared with the expression of a single antigen, a pattern of antigen expression is likely to have a more significant relationship to prognosis. This is exemplified by the panmyeloid phenotype (expression of 5 myeloid antigens) which appears to be associated with a good prognosis and may differentiate a subgroup within an otherwise intermediate prognosis group of patients. Further analysis with the inclusion of novel antibodies and the combination of multiple antibodies to create further subgroups such as the panmyeloid phenotype will continue to enhance knowledge in this area.

Analysis of the immunophenotypes of de novo acute lymphoblastic leukaemia (ALL) in the Sultanate of Oman

Acute lymphoblastic leukaemia (ALL) immmunophenotypes were analysed by flow cytometry in 65 Omani patients (46 children, 19 adults). Common ("CALLA-positive") ALL was the most frequently encountered (70%) B-cell lineage immunophenotype. Among T-cell lineage ALL patients, mature T-cell ALL was the least frequent (7%). Expression of certain surface markers including CD20 and CD6 appears to have an effect on some clinical and haematological features but FAB morphology was not useful in predicting cell lineage immunophenotypes. Other interesting findings, currently of uncertain significance, include the sizeable proportions of pre-B-ALL (group V) and cortical thymocytic ALL (stage III). Results of this study should help strengthen the emerging leukaemia database of the recently established Oman National Cancer Register and thereby contribute to a successful global attack against the haematological malignancies.

Immunophenotypic analysis of acute lymphocytic leukemia

Acute lymphoblastic leukemia (ALL) is one of the most common hematologic malignancies. Flow cytometry is an integral part of ALL diagnosis and also provides significant patient prognostic information. This article is a practical review of the basic principles of the flow cytometric evaluation of acute leukemias, the interpretation of flow cytometric data, and the management of practical problems such as aberrant antigen, hematogones, bone marrow regeneration, and minimal residual disease.

Flow cytometry in the diagnosis of acute leukemia

Acute leukemias are a heterogeneous group of malignancies with varying clinical, morphologic, immunologic, and molecular characteristics. Many distinct types are known to carry predictable prognoses and warrant specific therapy. Distinction between lymphoid and myeloid leukemias, most often made by flow cytometry, is crucially important. Several advances in flow cytometry, including availability of new monoclonal antibodies, improved gating strategies, and multiparameter analytic techniques, have all dramatically improved the utility of flow cytometry in the diagnosis and classification of leukemia. Acute leukemias reflect the pattern of antigen acquisition seen in normal hematopoietic differentiation, yet invariably demonstrate distinct aberrant immunophenotypic features. Detailed understanding of these phenotypic patterns of differentiation, particularly in myeloid leukemia, allows for more precise classification of leukemia than does morphology alone. However, morphologic and differentiation-based classifications of leukemia are limited in their prognostic value; cytogenetics and molecular genetics appear to be most important for identifying entities with distinct prognoses and clinical behavior. Increasingly, many of these genetically distinct subgroups of leukemia have been found to be closely associated with distinct immunophenotypes. For example, translocations such as t(8;21), t(15;17), and inv(16) in acute myeloid leukemia (AML), and t(1;19) and t(12;21) in acute lymphoblastic leukemia (ALL) have distinctive immunophenotypic profiles. Thus, in addition to classification into differentiation-based subtypes, detailed flow cytometric studies can define complex antigenic profiles that are associated with specific molecular defects and well-defined biology. In summary, multiparameter flow cytometry is an invaluable tool in the diagnosis, classification, and monitoring of patients with acute leukemia. Semin Hematol 38:124-138.

Fifteen cases of t(1;19)(q23;p13.3) identified in an Australian series of 122 children and 80 adults with acute lymphoblastic leukemia

The t(1;19)(q23;p13) has been reported in up to 6% of cytogenetically abnormal cases of acute lymphoblastic leukaemia (ALL), associated with a pre-B-ALL phenotype. In the 5-year period 1995-1999, we detected t(1;19) in 13 children and 2 adults with newly diagnosed ALL. This represented 10% of pediatric and 2.5% of adult diagnostic ALL samples successfully cultured in one center during this time. There were 9 males and 6 females. The mean age at diagnosis for the 13 children was 6.5 years (range 1.5 to 14 years) and the 2 adults were aged 42 and 45 years. The unbalanced t(1;19) occurred in 7 of 13 children (54%), contrary to the reported excess of unbalanced translocations at 75%; both adults had the unbalanced translocation. At diagnosis, the t(1;19) was the sole abnormality in 4 patients (26%), and in the remainder (74%) was part of a complex karyotype, which included i(7q) (2 patients), hyperdiploidy (2 patients) and del(6q) (2 patients). Correlation of karyotype with white cell, blast and platelet counts, cell surface markers, initial response to chemotherapy and short-term outcome showed no difference between the balanced and unbalanced forms of the translocation in children or whether t(1;19) was present as the sole abnormality or part of a complex karyotype.

CD56 expression in acute promyelocytic leukemia: a possible indicator of poor treatment outcome?

PURPOSE

Blast expression of CD56 is frequent in patients with t(8;21)(q22;q22) acute myeloid leukemia and is associated with an inferior outcome. The expression of CD56 has rarely been reported in acute promyelocytic leukemia (APL) and has not been clinically characterized. Therefore, we examined the prognostic significance of CD56 expression in APL.

PATIENTS AND METHODS

We identified all reported cases of CD56+ APL in the medical literature and collected clinical, biologic, and therapeutic details.

RESULTS

Data were obtained for 12 patients with CD56+ APL (> 20% blast expression of CD56), including four cases from a single institution with a total of 42 APL patients. All of the CD56+ APL patients had documented cytogenetic presence of t(15;17), and of the nine reported isotypes, eight (89%) were S-isoform. Only six CD56+ patients (50%) attained complete remission (CR); the other six individuals died within 35 days of presentation. Of the six patients who attained a CR, three (50%) relapsed at 111, 121, and 155 weeks, whereas three remained in continuous CR at 19, 90, and 109 weeks. Comparison of the control CD56- to CD56+ APL patients demonstrated that the latter group had a significantly lower fibrinogen level (P = .007), and among patients for whom data were available, there was a higher frequency of the S-isoform (P = .006). Additionally, the CR rate (50% v 84%, P = .025) and overall median survival (5 v 232 weeks; P = .019) were significantly inferior for CD56+ APL patients.

CONCLUSION

CD56+ acute promyelocytic leukemia is infrequent, seems to occur more frequently with the S-isoform subtype, and may be associated with a lower CR rate and inferior overall survival.

Prognostic Significance of Fluorescence Intensity of Surface Marker Expression in Childhood B-Precursor Acute Lymphoblastic Leukemia. A Pediatric Oncology Group Study

This report describes the prognostic significance of the intensity of surface membrane antigen expression in a series of 1,231 children older than 1 year with newly diagnosed B-precursor acute lymphoblastic leukemia (ALL) treated on Pediatric Oncology Group (POG) treatment protocols. All patients had dual-color flow cytometric immunophenotyping performed at a central reference laboratory with a standard panel of monoclonal antibodies. The flow cytometers used in the study were calibrated with a standard fluorescence microparticle that permitted conversion of relative fluorescence channels to standard units of mean equivalents of soluble fluorochrome (MESF). In univariate analysis, fluorescence intensity of CD45 and CD20 was significantly associated with event-free survival (EFS), whereas other markers showed no significant correlation with outcome. Patients whose blasts were greater than the 75th percentile of intensity for CD45 (corresponding to 18,000 MESF units with CD45-FITC, or about 8% of the intensity of normal lymphocytes) fared significantly worse than those with lower-density CD45, and those whose blasts were greater than the 25th percentile of intensity for CD20 (corresponding to 17,900 MESF units with CD20-PE) had a poorer EFS. The intensity of both CD45 and CD20 was independently correlated with outcome. There was no significant correlation between intensity of expression of either antigen and traditional clinical risk factors, ploidy, or t(9; 22) or t(1; 19). All patients with t(4; 11) had CD45 intensity greater than the 75th percentile, but CD45 intensity retained its prognostic significance after adjusting for t(4; 11). In multivariate analysis, both CD45 intensity greater than the 75th percentile and CD20 intensity greater than the 25th percentile were significantly correlated with poor outcome independently of previously reported poor prognostic factors including National Cancer Institute (NCI) risk group, ploidy, trisomies of 4 and 10, and adverse translocations including t(1; 19), t(9; 22), and t(4; 11). We conclude that in childhood B-precursor ALL, the intensity of expression of CD20 and CD45 provides prognostic information not available from simple consideration of antigen expression as positive or negative, and adds to that obtained from traditional clinical and biologic risk factors.

Aminopeptidase-N (CD13; gp 150): contrasting patterns of enzymatic activity in blood from patients with myeloid or lymphoid leukemia

We have investigated the compartmentalization of aminopeptidase-N-like activity in various blood fractions obtained from patients with acute lymphoid (ALL) or myeloid (AML) leukemia. The primary difference appears not to be the absolute level of overall activity, but rather the relative proportions of the different forms of activity detected. Thus, despite similar levels of total aminopeptidase-N-like activity detected in cells from different leukemic groups, true aminopeptidase-N/CD13 activity was only detected in cells derived from AML patients. Even in these patients, however, most of the detected aminopeptidase-N-like activity ( > 80%) could not be attributed to aminopeptidase-N/CD13. In marked contrast, plasma from leukemic patients also contained substantial total aminopeptidase-N-like activity, of which (irrespective of leukemic group) most could be attributed to aminopeptidase-N/CD13. Whilst slightly higher levels of total activity were obtained in plasma from AML patients compared to ALL patients, there was no difference in the relative proportion attributable to aminopeptidase-N/CD13 (approximately 80% of total aminopeptidase-N-like activity). Evaluation of total aminopeptidase-N-like activity present in whole blood gave differential patterns, and whilst only a proportion (20-40% of total aminopeptidase-N-like activity) could be attributed to true aminopeptidase-N/CD13, blood from patients with CD13+ AML showed the greatest activity so attributable. In total, our results outline the complexities of peptidase activities present within blood of leukemic individuals, and may, in part, explain the variability of previous studies attempting to associate prognostic features with phenotypic expression of CD13.