¹Clinica Oncoematologica Pediatrica, Dipartimento di Pediatra, Università di Padova, Italy

²Servicio de Citometria, Universita de Salamanca, Spain

³Centro Ricerche 'M. Tettamanti', Clinica Pediatrica, Università di Milano, Monza, Italy

⁴Dipartimento di Scienze Pediatriche, Università di Torino, Italy

Correspondence to: L De Zen, Clinica di Oncoematologia Pediatrica, Dipartimento di Pediatria, Università di Padova, via Giustiniani 3, 35128 Padova, Italy; Fax: 0039-049-8211465

The t(12;21)(p13;q22) fusion gene is the most frequent genetic lesion described in precursor B cell acute lymphoblastic leukemia (ALL) of childhood occurring in a quarter of cases. This gene rearrangement is associated with a good outcome presenting a high response rate to chemotherapy. In spite of its potential clinical relevance, the t(12;21) translocation usually goes undetected with conventional cytogenetic procedures. In the present study we utilized an objective flow cytometric approach (multiparametric quantitative analysis) for the phenotypic characterization of this type of ALL. We studied a total of 74 precursor B-ALL children, including 21 t(12;21)⁺ and 53 t(12;21)^- cases. Our results show that the t(12;21)(p13;q22)⁺ ALLs display a higher intensity of CD10 (P = 0.0016) and HLADR (P = 0.005) expression together with lower levels of the CD20 (P = 0.01), CD45 (P = 0.01), CD135 (P = 0.003) and CD34 (P = 0.03) antigens as compared to the t(12;21)^- cases. Moreover, as regards CD34 expression, we observed a more heterogeneous antigen expression within individual patients with higher coefficients of variation (median of 202 vs 88, P = 0.0001). A multivariate analysis disclosed that with the immunophenotypic approach used identification of t(12;21)⁺ cases can be achieved with a sensitivity of 86% and a specificity of 100%. We conclude that childhood precursor B-ALL carrying the t(12;21) translocation display characteristic phenotypic features which could provide a rapid, simple, sensitive and specific screening method to select for those cases that should undergo confirmatory molecular analysis. Leukemia (2000) 14, 1225-1231.

immunophenotype; genotype; cytometry

Introduction

The t(12;21)(p13;q22) fusion gene is the most frequent genetic lesion described in leukemic B cells from children with acute lymphoblastic leukemia (ALL) occurring in around a quarter of cases.^1,2,3,4 The molecular consequence of the t(12;21)(p13;q22) translocation is the fusion of two known genes named ETV6, mapped to 12p13, and AML1, located at 21q22. This translocation results in the transcription of a chimeric protein which consists of the helix-loop-helix domain of ETV6 and the entire AML1 gene, including both its DNA binding and transactivation domains.^5,6 The ETV6/AML1 fusion transcript is likely to be important for leukemogenesis.⁷ Recent reports^1,3,4,8 have shown that this gene rearrangement is associated with a clinically defined subgroup of precursor-B-ALL affecting children with ages from 1 to 12 years who show non-hyperdiploid leukemic lymphoblasts and do not display hyperleukocytosis. Moreover, the presence of the ETV6/AML1 transcript has been found to be associated with a good prognosis since patients with t(12;21) translocation, as compared with other patients, display a higher response rate to chemotherapy.⁴

In spite of its potential clinical relevance, the t(12;21) translocation involving the ETV6/AML1 genes usually goes undetected with conventional cytogenetic procedures. Thus, either fluorescence in situ hybridization or more frequently molecular biology techniques based on the polymerase chain reaction (PCR)⁹ have to be applied for its unequivocal detection, with the handicap that from each four childhood B-ALL cases analyzed, three would be negative. Accordingly, any technique that could be routinely used to screen childhood precursor B-ALL for the t(12;21) translocation would be welcome.

In past years immunophenotyping has proven to be useful for the diagnosis and subclassification of precursor-B acute lymphoblastic leukemia (B-ALL).^{10,11,12,13,14,15} Accordingly, assessment of the B cell origin of leukemic blasts and their assignment to a particular phenotypic subtype have been based on the immunologic similarities of leukemic cells and normal bone marrow B cell precursors.^16,17,18 However, in recent years several reports have shown that in most precursor B-ALL cases neoplastic cells display aberrant phenotypes.^{19,20,21,22,23} Some of these leukemia-associated phenotypes have been associated with specific genetic abnormalities.^{24,25,26,27,28,29} In this regard, large multicentric studies have been conducted in order to explore the potential existence of an association between specific cytogenetic abnormalities such as the t(1;19)(q23;p13)^30,31,32 and t(4;11)(q21;p23)^33,34 translocations and peculiar phenotypes of leukemic B cells.

In spite of its relative frequency the ETV6/AML1-positive leukemias have not yet undergone an extensive immunophenotypic characterization. In fact, previously reported series comprise flow cytometric studies in which only a restricted number of surface antigens was analyzed for its presence or absence on the leukemic cells. These reports have found that ETV6/AML1-positive cases are CD10 positive and frequently myeloid antigen positive,^8,35 none of these characteristics being specific for this group of B-ALL cases. The exception to this is a recent study published by Borowitz et al³⁶ in which a CD9^-/CD20^-/CD45⁺/CD13⁺/CD34⁺ pattern was shown to be more frequent in t(12;21)⁺ cases although presenting a relatively low sensitivity to be routinely applied for the screening of t(12;21)⁺ precursor B-ALL cases. In the present study we have utilized an objective flow cytometric approach for the phenotypic characterization of B-ALL cases, which consists of a multiparametric quantitative analysis of antigen expression on leukemic B cells specifically identified on the basis of a large panel of predefined triple-staining combinations of monoclonal antibodies. This approach has been demonstrated to be useful for screening among precursor B-ALL patients of those cases with t(12;21) translocation, since it displays both a high sensitivity (86%) and specificity (100%).

Materials and methods

Samples

A total of 74 childhood precursor-B acute lymphoblastic leukemias diagnosed according to previously established criteria³⁷ included in AIEOP ALL 91 protocols were collected for the study. The only selection criteria were the availability of frozen cells at diagnosis and that all cases were studied for the t(12;21) translocation. Among the patients analyzed 38 were males and 36 females with a median age of 5.3 years (range 0-15 years).

Immunophenotypic studies

All studies were performed using a direct immunofluorescence technique on frozen mononuclear cells obtained at diagnosis from the bone marrow samples separated by Ficoll-Hypaque technique. A number of 0.5 ´ 10⁶ cells in a final volume of 100 l was used per test. Monoclonal antibodies (MoAbs) directly conjugated with fluorescein isothiocyanate (FITC), phycoerythrin (PE) and the phycoerythrin-cyanine 5 (PE-Cy5) fluorochrome tandem conjugate were combined in each tube. The fluorochrome and source of each MoAb used in the present study were as follows: CD20-PE, CD15-FITC, CD34-PE, CD38-PE and HLADR-PE were purchased from Becton Dickinson (San José, CA, USA); CD33-PE, CD13-PE, CD10-FITC, CD19-FITC and CD2-FITC were from Coulter (Miami, FL, USA), CD19-PE-Cy5, CD45-PE-CY5 and CD65-FITC were obtained from Caltag Laboratories (San Francisco, CA, USA); CD135-PE was from Immunotech (Marseille, France), CD24-FITC from Ortho Diagnostic Systems (Raritan, NJ, USA) and CD14-PE from DAKO (Glostrup, Denmark). The three color combinations used are illustrated in Table 1.

One triple combination with isotype controls and the unstained cells were used as negative controls. After adding the MoAbs, cells were vortexed and incubated for 15 min at room temperature in the dark and thereafter they were washed twice in phosphate-buffered saline (PBS). Then they were resuspended in 1 ml of PBS and analyzed in an Epics XL (Coulter) flow cytometer. The instrument was calibrated before data acquisition using a CD4FITC/CD8PE/CD3PE-Cy5 positive control. A minimum of 10000 events was collected for each combination of MoAbs.

For the analysis process an immunological gate including all blast cells was established in all tube combinations based on the expression of the CD19 B cell-associated marker. For that purpose a display of SS-LOG (side scatter with logarithmic amplification) vs CD19 was used to construct the so-called immunological gate; additionally, a combination of the light scatter parameters (forward Scatter-FS vs SS-LOG) was used in order to eliminate dead cells and debris. By using this gating strategy more than 95% of the cells included were blasts. The ability of the SS-LOG/CD19 gate in detecting more than 95% of blast cells was seen by results from CD19-CD34-CD45 combination; in fact never in the examined samples did more than 3% of CD19-positive cells show a high expression of CD45 in the absence of CD34. All the other markers were analyzed in biparametric histograms gated on CD19⁺ blast cells. For that purpose data were stored in list mode files using the System II software program (Coulter). For each MoAb explored in this study the mean fluorescence intensity (MFI) expressed as mean fluorescence channel (arbitrary units scaled from 0 to 1024), the standard deviation (s.d.), the coefficient of variation (CV) and the number of different subpopulations found on the basis of multimodal distribution, was reported. As to obtaining an absolute value with intra-inter laboratory reproducibility issues, the MFI values were converted into molecules equivalent to soluble fluorochrome (MESF) by using a DAKO bead set. Briefly, this kit contains a set of calibrated standards, with five populations of microbeads displaying increasing and predetermined amounts of fluorescence (expressed in terms of number of MESF) and one reference blank population. After analyzing the fluorescence beads with the same instrument setting as those used for the analysis of the patient samples, we built up a calibration plot to convert MFI into MESF values. The linear regression equation, correlating the channel number with the specific MESF value, was calculated using the TallyCall software (DAKO). Instrument calibration for the measurement of fluorescence intensity in MESF units was performed on a daily basis.

The median of MESF values obtained from isotype controls and unstained cells was calculated. An antigen was considered positive when its MESF value resulted higher than the median +2 s.d.

RT-PCR assay

All samples were analyzed for the presence of TEL/AML1 fusion gene products. The sequence of amplification primers was made by the set B12-(5'-CGTGGATTTTTCAATACATGTCTCA-3') of TEL and AM3 (5'-AACTGCCTTCGT CTCTATCTTTGTCCTTGG-3') for AML1, with minor modifications according to the published TEL sequence.³⁸ After amplification, 10 l of PCR products were run on a 2.5% agarose gel stained with ethidium bromide and visualized under an ultraviolet (UV) lamp. Representative PCR products were cloned into the plasmid vector pMOS (Amersham, Buckinghamshire, UK) and sequenced by the dideoxynucleotide chain termination method modified for use with double-stranded DNA templates.⁸

Statistical methods

For each MESF and CV value the mean, median and standard deviation (s.d.) were calculated. Univariate analysis was performed using the Student's t-test and Kolmogorov-Smirnov tests to analyze the different distribution of quantitative variables between t(12;21)-positive and -negative cases. An expert system was undertaken in two ways: one choosing the best cut-off values according to the sensitivity and specificity tests (test of Yonden) and the other with reference to the quartile of the whole population.

Results

From the 74 precursor-B ALL cases analyzed in the present study, 21 (28%) were positive for the ETV6/AML1 fusion gene product detected by RT-PCR whereas the remaining 53 patients were negative. In this group 16/53 (30%) resulted as hyperdiploid, whereas 37/53 (70%) were diploid.

Background fluorescence levels were calculated for each group of fluorochrome conjugates. Based on these findings, a case was considered positive for a FITC, PE and PE-Cy5 conjugated MoAb when the MESF value resulted higher than 900, 400 and 400, respectively (median +2 s.d. of the controls).

Based on these criteria, the resulting distribution of antigen expression in t(12;21)-positive and -negative cases was as described in Table 2. The presence of reactivity for CD34, CD20, CD13, CD15 and CD135 was significantly different in the two groups of precursor B-ALL patients. In fact, a significantly higher proportion of CD34⁺ and CD13⁺ cases was found in t(12;21)⁺ leukemias while the percentage of CD20⁺, CD15⁺ and CD135⁺ patients was lower among these cases. No significant differences were detected as regards the incidence of positive cases for the CD45, CD10 and HLA-DR antigens.

Table 3 shows the results obtained with regards to the relative intensity of antigen expression in precursor B-ALL cases according to the presence or absence of t(12;21) translocation. It can be seen that for markers which were positive in both groups of patients, different amounts of antigen were present. Differences could be observed both in intensity and homogeneity/heterogeneity of the antigen expression distribution as assessed by the median MESF and CV values, respectively (Table 4).

Univariate analysis comparing ETV6/AML1-positive and -negative ALL cases showed that the median fluorescence intensity obtained for CD10, CD135, CD20, CD34, CD45 and HLA-DR antigens was significantly different in both groups of patients, thus disclosing a difference for CD10 and CD45 not previously detected with positive and negative criteria. ETV6/AML1-positive cases (Figure 1) were characterized by a high median intensity of expression of CD10 and HLA-DR and low intensity of CD20, CD45, CD135, CD34.

Regarding the myeloid antigens, CD15 showed a lower median intensity in ETV6/AML1⁺ cases while CD13 expression was slightly more intense in these patients although for both antigens differences did not reach high statistical significance (P = 0.07).

No differences were detected in the CD19, CD24, CD33 and CD38 fluorescence intensity between the two groups of patients.

Upon analyzing the distribution pattern of each of the antigens explored as assessed by the mean CV of the fluorescence intensity, we found that ETV6/AML1⁺ leukemias were characterized by a higher heterogeneity in the levels of CD34 expression (median CV of 201% vs 88%, P = 0.0001). In fact, 20 out of the 21 ETV6/AML1⁺ cases (95%) showed a bimodal distribution for CD34 with two clearly distinct subpopulations of leukemic cells displaying different levels of this marker on their surface (Figure 1a). Also CD13 and CD33 showed a higher heterogeneity of expression among the ETV6/AML1⁺ cases but the differences with ETV6/AML1^- patients did not reach high statistical significance.

In order to analyze the combination of the phenotypic variables displaying the highest sensitivity and specificity for the identification of those B-ALL cases displaying the t(12;21) translocation, we performed a multivariate analysis. All individual phenotypic characteristics which showed a significantly different distribution on t(12;21)⁺ cases, as compared to t(12;21)^- leukemias, were included. For each of these markers the best individual sensitivity and specificity cut-off value was calculated (Table 5).

As shown in Table 6, in 18 out of the 21 (86%) t(12;21)⁺ B-ALLs leukemic cells displayed all the chosen criteria for t(12;21), while none of the t(12;21)^- cases simultaneously showed these six phenotypic features. Therefore, the sensitivity of these parameters to predict for the existence of t(12;21) translocation in childhood precursor B-ALL cases, was 86% with a specificity of 100% (Yune test 0.857). Two (9.5%) t(12;21)-positive cases responded to five criteria while only one (5%) responded to four conditions: one case showed a lower HLA-DR (64000 MESF) expression, another was characterized by a relatively higher reactivity for CD20 (800 MESF) and the remaining one showed a unimodal CD34 distribution (CV of 60%) with a low HLADR expression (33000 MESF).

The analysis performed considering only positivity against negativity for the CD34, CD20, CD15, CD13 and CD135 antigens, demonstrated only 70% sensitivity and 90% specificity.

Discussion

For a long time neoplastic cells from acute leukemias were believed to reflect the phenotypes of the earlier stages of normal hemopoietic cells blocked in their differentiation.³⁹ Thus in the past two decades immunophenotypic studies have been used for the subclassification of precursor B-ALL according to the expression of B cell maturation antigens.^{10,11,12,13,14,15} However, more recent studies have shown that by far the majority, if not all, precursor B-ALL cells display immunophenotypic features which are usually not detected in the normal bone marrow B cell precursors - the so-called phenotypic aberrations.^{19,20,21,22,23} Accordingly, cross-lineage antigen expression and aberrant patterns of B cell maturation are usually detected in these patients. Based on these findings, it has been suggested that these leukemia-associated phenotypes could reflect, at least to a certain extent, the genetic aberrations present in the leukemic cells; accordingly for those patients displaying an identical genetic lesion, a similar phenotypic behavior could be expected. Indeed, the association between specific phenotypic subgroups and specific chromosomal abnormalities has been reported in the past.^{24,25,26,27,28} In spite of this, few studies have been aimed at analyzing the utility of the immunophenotypic study for the screening of specific genetic aberrations,^{30,31,32,33,34} even if it is routinely performed in diagnosis of acute leukemias so resulting in a very good screening tool.⁴⁰

In contrast to other chromosome translocations such as the t(15;17) aberration identified in AML patients,⁴¹ no clear correlation that could be used as a highly sensitive and specific parameter for the screening of t(12;21)⁺ cases, has been found between the t(12;21) translocation, present in precursor B-ALL patients, and other disease characteristics. Furthermore, this chromosome aberration is usually not detected with conventional cytogenetics and either FISH or PCR techniques have to be used to identify t(12;21)⁺ cases.^8,36 Nevertheless, these techniques are either not available or not routinely performed in the majority of centers. These findings together with the reported association between t(12;21) and a better prognosis in children with precursor B-ALL,^1,3,4,8 point to the need for a rapid and simple technique that could be used in the screening of precursor B-ALL patients carrying the t(12;21) translocation, in which appropriate molecular studies should be undertaken.

In the present study we have analyzed the immunophenotype of the leukemic cells from a group of 74 children with precursor B-ALL selected only for PCR ETV6/AML1 translocation. Twenty-one (28%) were shown to carry the t(12;21) translocation. This percentage corresponds to the normal B-ALL distribution.^1,2,3,4 In t(12;21)-negative leukemias an excess of CD10-negative cases resulted as reported in Table 2. This observation depends on casual selection of frozen cells in tested patients. In fact the percentage of CD10^- cases is higher than expected in normal childhood B-ALL distribution (10% vs 4%).³⁷ For this reason we also performed statistical analysis on each subgroup, so obtaining the same values of sensitivity and specificity (data not shown).

Our aim was to investigate whether the t(12;21)⁺ cases showed a specific immunophenotyping which could be applied for the screening of this genetic abnormality. For that purpose we decided to use a multiparametric objective approach for the immunophenotypic analysis in which, prior to their phenotypic characterization, B-lineage leukemic cells were specifically identified on the basis of their reactivity for the CD19 B cell-associated antigen. In order to increase the ability to discriminate leukemic from normal B cell precursors and avoid misinterpretation of positive and negative antigens,^42,43,44,45 the immunological characterization of the leukemic cells was performed using objective criteria based on the evaluation not only of the presence/absence of reactivity for a certain antigen, by which low sensitivity and specificity were previously obtained as recently underlined by Hrusak et al,⁴⁶ but also on the quantitative levels of antigen expression and its distribution on the whole leukemic cell population.

In order to provide reproducible results at various times in the same laboratory and in different laboratories, the absolute levels of antigen expression were translated from arbitrary units (fluorescence channels) into molecules equivalent to soluble fluorochrome (MESF), specific information being provided on each fluorochrome-conjugated reagent used. The coefficient of variation found for the expression of each antigen was used as a measurement of the degree of variability on the levels of antigen expression in the whole population of leukemic cells from each individual patient. The possible bias connected with frozen cells and peculiar batches of reagents could be excluded; in fact, Ginaldi et al⁴⁷ denied differences in antigen intensity between frozen and fresh blast cells; moreover, no significant variability in fluorescence intensity was found in different antibody batches (data not shown).

Our results showed that significant differences exist between the immunophenotypic characteristics of t(12;21)⁺ and t(12;21)^- precursor B-ALL patients, the former being associated with higher levels of expression of CD10, HLADR and lower reactivity of CD20, CD135, CD45 as well as a higher heterogeneity of CD34 expression among the leukemic cell population of individual patients. While the positivity of CD13 antigen was statistically relevant in this cohort, in contrast its expression resulted as higher in t(12;21) cases but not statistically significant with TEL/AML1-negative leukemias. This apparent discrepancy might depend either on the different modality of positive evaluation or on the low number of patients tested. These results confirm the lack of prognostic impact of myeloid markers in pediatric ALLs.⁴⁸ Moreover, we confirm previous findings from other groups who have found an association between the t(12;21)⁺ and common-ALL phenotype⁸ as well as a higher expression of CD10 and a lower expression of CD45 and CD20 antigens.⁴⁹ On the other hand, CD20-negative, low CD45 and high CD10 expression have been correlated with those precursor B cells ALLs such as present a good outcome.^50,51,52 Our results provide evidence for the association of these clinical and immunophenotypic features with the t(12;21) translocation in childhood precursor B-ALL. The individual relevance of CD9 and KOR-SA3544 in identifying the TEL-AML1 translocation³⁶ has not been evaluated because these two antigens are not included in the antibody combinations routinely utilized in the AIEOP immunophenotyping analysis at diagnosis.

Multivariate analysis was performed in order to explore the value of those immunophenotypic features to distinguish between t(12;21)-positive and both hyperdiploid and diploid precursor B-ALL cases negative for the t(12;21) translocation. Our results show that a level of CD10 and HLADR expression higher than 10000 and 80000 MESF, respectively, a CD20, CD45 and CD135 reactivity lower than 500, 400 and 5000 MESF, respectively, and bimodal CD34 was the best combination of parameters to discriminate between both groups of patients. In this sense it should be noted that most (86%) of the precursor B-ALL cases which were t(12;21)⁺, simultaneously displayed these six phenotypic criteria, while none of the t(12;21)^- cases showed this phenotype. Accordingly, we show that the presence of these six phenotypic characteristics in childhood precursor B-ALL blast cells is specific (100%) for the t(12;21) translocation, although there is a small group of these patients (three, 14%) in which one or two of the six criteria listed above is lacking; in 2/3 of them additional cytogenetic abnormalities were found. This association between t(12;21) and abnormal karyotypes has been recently reported by Raynaud et al.⁵³

In the present study we conclude that this multiparametric quantitative immunophenotyping approach seems to be able to identify homogeneous groups of ALLs also corresponding to genotypically determined types such as t(12;21) and also t(4;11) and hyperdiploid cases (De Zen manuscript in preparation). Our results enhanced the two properties reported in the Borowitz study: that an accurate phenotypic prediction of cytogenetic and molecular genetic abnormalities could not be obtained by either positive or negative antigen definition; only a more accurate description of the antigen expression pattern was able to demonstrate significant predictability.³⁶ Finally, we demonstrated that the childhood precursor B-ALL cases carrying the t(12;21) translocation display characteristic phenotypic features easily identifiable by routinely used immunophenotyping with multiparametric quantitative approach. This method provides an inexpensive, rapid, simple, sensitive and specific technique that could be used to screen for those patients in which confirmatory molecular studies have to be performed for the identification of t(12;21)⁺ childhood precursor B-ALL cases.

Acknowledgements

We thank all the investigators from AIEOP Institutions for their collaboration; Maurizio Aricò, MD for helpful discussion and Mr Denis Swift for the editorial scrutiny. This work was supported by AIRC (Associazione Italiana per la Ricerca sul Cancro), MURST ex 40% and 60%; Fondazione Città della Speranza.

1 McLean TW, Ringold S, Neuberg D, Stegmaier K, Tantravahi R, Ritz J, Koeffler HP, Takeuchi S, Janssen JWG, Seriu T, Bartram CR, Sallan SE, Gilliland DG, Golub TR. TEL/AML1 dimerizes and is associated with a favourable outcome in childhood acute lymphoblastic leukemia. Blood 1996; 88: 4252-4258, MEDLINE

2 Romana SP, Poirel H, Leconiat M, Flexor MA, Mauchauffé M, Jonveaux P, Macintyre EA, Berger R, Bernard OA. High frequency of t(12;21) in childhood B-lineage acute lymphoblastic leukemia. Blood 1995; 86: 4263-4269, MEDLINE

3 Shurtleff SA, Buijs A, Behm FG, Rubnitz JE, Raimondi SC, Hancock ML, Chan GC-F, Pui C-H, Grosveld G, Downing JR. TEL/AML1 fusion resulting from a cryptic t(12;21) is the most common genetic lesion in pediatric ALL and defines a subgroup of patients with an excellent prognosis. Leukemia 1995; 9: 1985-1989, MEDLINE

4 Rubnitz JE, Pui CH, Downing JR. The role of the TEL fusion gene in pediatric leukemias. Leukemia 1999; 13: 6-13, MEDLINE

5 Golub TR, Barker GF, Bohlander SK, Hiebert SW, Ward DC, Bray-Ward P, Morgan E, Raimondi SC, Rowley JD, Gilliland DG. Fusion of the TEL gene on 12p13 to the AML1 gene on 21q22 in acute lymphoblastic leukemia. Proc Natl Acad Sci USA 1995; 92: 4917-4921, MEDLINE

6 Romana SP, Mauchauffé M, Le Coniat M, Chumakov I, Le Paslier D, Berger R, Bernard OA. The t(12;21) of acute lymphoblastic leukemia results in a TEL-AML1 gene fusion. Blood 1995; 85: 3662-3670, MEDLINE

7 Raynaud S, Cavé H, Baens M, Bastard C, Cacheux V, Grosgeorge J, Guidal-Giroux C, Guo C, Vilmer E, Marynen P, Grandchamp B. The 12;21 translocation involving TEL and deletion of the other TEL allele: two frequently associated alterations found in childhood acute lymphoblastic leukemia. Blood 1996; 87: 2891-2899, MEDLINE

8 Borkhardt A, Cazzaniga G, Viehmann S, Valsecchi MG, Ludwig WD, Burci L, Mangioni S, Schrappe M, Riehm H, Lampert F, Basso G, Masera G, Harbott J, Biondi A. Incidence and clinical relevance of TEL/AML1 fusion genes in children with acute lymphoblastic leukemia enrolled in the German and Italian multicenter therapy trials. Blood 1997; 90: 571-577, MEDLINE

9 Cayuela JM, Baruchel A, Orange C, Madani A, Auclerc MF, Daniel MT, Schaison G, Sigaux F. TEL/AML1 fusion RNA as a new target to detect minimal residual disease in pediatric B-cell precursor acute lymphoblastic leukemia. Blood 1996; 88: 302-308, MEDLINE

10 Bene MC, Castoldi G, Knapp W, Ludwig WD, Matutes E, Orfao A, van't Veer MB. Proposals for the immunological classification of acute leukemias. Leukemia 1995; 9: 1783-1786, MEDLINE

11 Jennings CD, Foon KA. Recent advances in flow cytometry: application to the diagnosis of hematologic malignancy. Blood 1997; 90: 2863-2892, MEDLINE

12 De Rossi G, Grossi C, Foa R, Tabilio A, Vegna L, Lo Coco F, Annino L, Camera A, Cascavilla N, Ciolli S, Del Poeta G, Liso V, Mandelli F. Immunophenotype of acute lymphoblastic leukemia cells: the experience of the Italian Cooperative Group (Gimena). Leuk Lymphoma 1993; 9: 221-228, MEDLINE

13 Ludwig WD, Reiter A, Loffler H, Gokbuget, Hoelzer D, Riehm H, Thiel E. Immunophenotypic features of childhood and adult acute lymphoblastic leukemia (ALL): experience of the German Multicentre Trials ALL-BFM and GMALL. Leuk Lymphoma 1994; 13: (Suppl. 1) 71-76, MEDLINE

14 Foon KA, Tood RF III. Immunologic classification of leukemia and lymphoma. Blood 1986; 68: 1-31, MEDLINE

15 Paietta E. Proposal for the immunological classification of acute leukemias. Leukemia 1995; 9: 2147-2148, MEDLINE

16 Janossy G, Bollum FJ, Bradstock KF, Ashley J. Cellular phenotypes of normal and leukemic hemopoietic cells determined by analysis with selected antibody combinations. Blood 1980; 56: 430-441, MEDLINE

17 Ryan DH, Chapple C, Kossover SA, Sandberg AA, Cohen HJ. Phenotypic similarities and differences between cALLA-positive acute lymphoblastic leukemia cells and normal marrow cALLA-positive B cell precursors. Blood 1987; 70: 814-821, MEDLINE

18 Lamkin T, Brooks J, Annett G, Roberts W, Weinberg K. Immunophenotypic differences between putative hematopoietic stem cells and childhood B-cell precursor acute lymphoblastic leukemia cells. Leukemia 1994; 8: 1871-1878, MEDLINE

19 Hurwitz CA, Loken MR, Graham ML, Karp JE, Borowitz MJ, Pullen DJ, Civin CI. Asynchronous antigen expression in B lineage acute lymphoblastic leukemia. Blood 1988; 72: 299-307, MEDLINE

20 Ross CW, Stoolman LM, Schnitzer B, Schlegelmilch JA, Hanson CA. Immunophenotypic aberrancy in adult acute lymphoblastic leukemia. Am J Clin Pathol 1990; 94: 590-599, MEDLINE

21 Drexler HG, Thiel E, Ludwig WD. Review of the incidence and clinical relevance of myeloid antigen-positive acute lymphoblastic leukemia. Leukemia 1991; 5: 637-645, MEDLINE

22 Greaves MF, Chan LC, Furley AJW, Watt SM, Molgaard HV. Lineage promiscuity in hematopoietic differentiation and leukemia. Blood 1986; 67: 1-11, MEDLINE

23 Ciudad J, San Miguel JF, Lopez Berges MC, Vidriales B, Valverde B, Ocqueteau M, Mateos G, Cabellero MD, Hernendez J, Moro MJ, Mateos MV, Orfao A. Prognostic value of immunophenotypic detection of minimal residual disease in acute lymphoblastic leukemia. J Clin Oncol 1998; 16: 3774-3781, MEDLINE

24 Pui CH, Williams DL, Roberson PK, Raimondi SC, Behm FG, Lewis SH, Rivera GK, Kalwinsky DK, Abromowitch M, Crist WM, Murphy SB. Correlation of karyotype and immunophenotype in childhood acute lymphoblastic leukemia. J Clin Oncol 1988; 6: 56-61, MEDLINE

25 Van Denderen J, Van der Plas D, Meeuwsen T, Zegers N, Boersma W, Grosveld G, Van Ewijk W. Immunologic characterization of the tumor-specific bcr-abl junction in Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood 1990; 76: 136-141, MEDLINE

26 Tien HF, Wang CH, Lee FY, Liu MC, Chuang SM, Chen YC, Shen MC, Lin DT, Lin KH, Chuu WM. Cytogenetic study of acute lymphoblastic leukemia and its correlation with immunophenotype and genotype. Cancer Genet Cytogenet 1992; 59: 191-198, MEDLINE

27 Ludwig WD, Bartram CR, Harbott J, Koller U, Haas O, Hansen-Hagge T, Heil G, Seibt-Jung H, Teichmann J, Ritter J, Knapp W, Gadner H, Thiel E, Riehm H. Phenotypic and genotypic heterogeneity in infant acute leukemia. I. Acute lymphoblastic leukemia. Leukemia 1989; 3: 431-439, MEDLINE

28 Ludwig WD, Bartram CR, Thiel E, Teichmann JV, Harbott J, Reiter A, Riehm H. Childhood acute lymphoblastic leukemia with co-expression of myeloid antigens (My+ALL): incidence, genotype, and clinical significance. Blood 1989; 74: 197a (Abstr.),

29 Kita K, Nakase K, Miwa H, Masuya M, Nichii K, Morita N, Takalura N, Otsuij A, Shirakawa S, Ueda T, Nasu K, Kyo T, Dohy H, Kamada N. Phenotypical characteristics of acute myelocytic leukemia associated with the t(8;21)(q22;p22) chromosomal abnormality. Frequent expression of immature B-cell antigen CD19 together with stem cell antigen CD34. Blood 1992; 80: 470-477, MEDLINE

30 Devaraj PE, Foroni L, Janossy G, Hoffbrand AV, Secker-Walker LM. Expression of the E2A-PBX1 fusion transcripts in t(1;19)(q23;p13) and der(19)t(1;19) at diagnosis and in remission of acute lymphoblastic leukemias with different B lineage immunophenotype. Leukemia 1995; 9: 821-825, MEDLINE

31 Pui CH, Raimondi SC, Hancock ML, Rivera GK, Ribeiro RC, Mahmoud HH, Sandlund JT, Crist WM, Behm FG. Immunologic, cytogenetic and clinical characterization of childhood acute lymphoblastic leukemia with the t(1;19)(q23;p13) or its derivative. J Clin Oncol 1994; 12: 2601-2606, MEDLINE

32 Borowitz MJ, Hunger SP, Carroll AJ, Shuster JJ, Pullen DJ, Steuber CP, Cleary ML. Predictability of the t(1;19)(q23;p13) from surface antigen phenotype: implications for screening cases of childhood acute lymphoblastic leukemia for molecular analysis: a Pediatric Oncology Group study. Blood 1993; 82: 1086-1091, MEDLINE

33 Pui CH, Frankel LS, Carroll AJ, Raimondi SC, Shuster JJ, Head DR, Crist WM, Land VJ, Pullen DJ, Steuber P, Behm FG, Borowitz MJ. Clinical characteristics and treatment outcome of childhood acute lymphoblastic leukemia with the t(4;11)(q21;q23): a collaborative study of 40 cases. Blood 1991; 77: 440-447, MEDLINE

34 Pui CH. Acute leukemias with the t(4;11)(q21;q23). Leuk Lymphoma 1992; 7: 173-179, MEDLINE

35 Baruchel A, Cayuela JM, Ballerini P, Landman-Parker J, Cezard V, Firat H, Haddad E, Auclerc MF, Valensi F, Cayre YE, Macintyre EA, Sigaux F. The majority of myeloid-antigen-positive (My⁺) childhood B-cell precursor acute lymphoblastic leukaemias express TEL/AML1 fusion transcripts. Br J Haematol 1997; 99: 101-106, MEDLINE

36 Borowitz MJ, Rubnitz J, Nash M, Pullen DJ, Camitta B. Surface antigen phenotype can predict TEL-AML1 rearrangement in childhood B-precursor ALL: a Pediatric Oncology Group study. Leukemia 1998; 12: 1764-1770, MEDLINE

37 Conter V, Aricò M, Valsecchi MG, Rizzari C, Testi A, Miniero R, Di Tullio MT, Lo Nigro L, Pession A, Rondelli R, Messina C, Santoro N, Mori PG, De Rossi G, Tamaro P, Silvestri D, Biondi A, Basso G, Masera G. Intensive BFM chemotherapy for childhood ALL: interim analysis of the AIEOP-ALL 91 study. Haematologica 1998; 83: 791-799, MEDLINE

38 Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of the PDGF receptor to a novel ets-like gene, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation. Cell 1994; 77: 307-316, MEDLINE

39 Janossy G, Bollum FJ, Bradstock KF, McMichael A, Rapson N, Greaves MF. Terminal transferase-positive human bone marrow cells exhibit the antigenic phenotype of common acute lymphoblastic leukemia. J Immunol 1979; 123: 1525-1529, MEDLINE

40 Borowitz MJ, Rubnitz J, Nash M, Pullen DJ, Camitta B. Replay: immunophenotypic prediction of TEL/AML1 rearrangement in childhood ALL (letter). Leukemia 1999; 13: 983,

41 Orfao A, Chillon MC, Bortoluci AM, Lopez Berges MC, Garcia-Sanz R, Gonzalez M, Tabernero MD, Garcia Marcos MA, Rasillo AI, Hernendez-Rivas J, San Miguel JF. The flow cytometric pattern of CD34, CD15 and CD13 expression in acute myeloblastic leukemia is highly characteristic of the presence of PML-RARalfa gene rearrangements. Haematologica 1999; 84: 405-412, MEDLINE

42 Faraht N, Lens D, Zomas A, Morilla R, Matutes E, Catovsky D. Quantitative flow cytometry can distinguish between normal and leukemic B-cell precursors. Br J Haematol 1995; 10: 640-646,

43 Dworzak MN, Fritsch G, Fleischer C, Printz D, Froschl G, Buchinger P, Mann G, Gadner H. Comparative phenotype mapping of normal vs. malignant pediatric B-lymphopoiesis unveils leukemia-associated aberrations. Exp Hematol 1998; 26: 305-313, MEDLINE

44 Lavabre-Bertrand T, Duperray C, Brunet C, Poncelet P, Exbrayat C, Bourquard P, Lavabre-Bertrand C, Brochier J, Navarro M, Janossy G. Quantification of CD24 and CD45 antigens in parallel allows a precise determination of B-cell maturation stages: relevance for the study of B-cell neoplasias. Leukemia 1994; 8: 402-408, MEDLINE

45 Weir EG, Cowan K, LeBeau P, Borowitz MJ. A limited panel can distinguish B-precursor acute lymphoblastic leukemia from normal B precursors with four color flow cytometry: implication for residual disease detection. Leukemia 1999; 13: 558-567, MEDLINE

46 Hrusak O, Trka J, Zuna J, Bartunkova J, Stary J. Are we ready to curtail testing for TEL/AML1 fusion (letter)? Leukemia 1999; 13: 981-982, MEDLINE

47 Ginaldi L, Matutes E, Farahat N, De Martinis M; Morilla R, Catovsky D. Differential expression of CD3 and CD7 in T-cell malignancies: a quantitative study by flow cytometry. Br J Haematol 1996; 93: 921-927, MEDLINE

48 Putti MC, Rondelli R, Cocito MG, Aricò M, Sainati L, Conter V, Guglielmi C, Cantù-Rajnoldi A, Consolini R, Pession A, Zanesco L, Masera G, Biondi A, Basso G. Expression of myeloid markers lack prognostic impact in children treated for acute lymphoblastic leukemia: Italian experience in AIEOP-ALL 88-91 studies. Blood 1998; 92: 795-801, MEDLINE

49 Borowitz MJ, Shuster J, Carroll AJ, Nash M, Look AT, Camitta B, Mahoney D, Lauer SJ, Pullen DJ. Prognostic significance of fluorescence intensity of surface marker expression in childhood B-precursor acute lymphoblastic leukemia. A Pediatric Oncology Group Study. Blood 1997; 89: 3960-3966, MEDLINE

50 Lavabre-Bertrand T, Janossy G, Ivory K, Peters R, Secker-Walker L, Porwit-MacDonald A. Leukemia-associated changes identified by quantitative flow cytometry: I. CD10 expression. Cytometry 1994; 18: 209-217, MEDLINE

51 Behm FG, Raimondi SC, Schell MJ, Look AT, Rivera GK, Pui C-H. Lack of CD45 antigen on blast cells in childhood acute lymphoblastic leukemia is associated with chromosomal hyperdiploidy and other favorable prognostic features. Blood 1992; 79: 1011-1016, MEDLINE

52 Borowitz MJ, Guenter KL, Shults KE, Stelzer GT. Immunophenotyping of acute lymphoblastic leukemia by flow cytometric analysis: use of CD45 as right-angle light scatter to gate on leukemic blasts in three-color analysis. Am J Clin Pathol 1993; 100: 534-540, MEDLINE

53 Raynaud SD, Dastugue N, Zoccola D, Shurtleff SA, Mathew S and Raimondi SC. Cytogenetic abnormalities associated with the t(12;21): a collaborative study of 169 children with t(12;21)-positive acute lymphoblastic leukemia. Leukemia 1999; 13: 1325-1330, MEDLINE

Figure 1  Characteristic pattern of t(12;21)-positive ALL in CD19 gated cells. (a) Bimodal distribution (high CV) of CD34; (b) high intensity of HLA-DR expression; (c) high expression of CD10 and low intensity of CD20; (d) low/absent CD135 and CD45 expression.

Table 1  Triple-staining combinations of monoclonal antibodies used in the present study. The first is used for blast identification, the others for blast characterization

Table 2  Distribution of cases showing positive expression of antigens calculated according to positive and negative criteria and chi-square test for comparison between t(12;21)⁺ and t(12;21)^- precursor B-ALLs

Table 3  Univariate analysis for MESF values in t(12;21)-positive and -negative cases

Table 4  Univariate analysis for coefficient of variation (CV) values in t(12;21)-positive and -negative cases. Only the MoAbs with statistical significance are reported

Table 5  Sensitivity and specificity of the individual phenotypic characteristics included in the multivariate analysis for the detection of t(12;21)⁺ leukemias

Table 6  Distribution of precursor B-ALL cases according to the number of satisfied immunophenotypic criteria: all the six characteristics agree with t(12;21)⁺ leukemias (18/21) while none of the t(12:21)^- cases display them simultaneously