SEARCH:   Advanced

Prepublished online as a Blood First Edition Paper on April 3, 2003; DOI 10.1182/blood-2002-09-2913. This Article Abstract Full Text (PDF) All Versions of this Article: 2002-09-2913v1 102/3/1000    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Rights and Permissions Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Asnafi, V. Articles by Macintyre, E. Search for Related Content PubMed PubMed Citation Articles by Asnafi, V. Articles by Macintyre, E. Related Collections Immunobiology Neoplasia Oncogenes and Tumor Suppressors Clinical Trials and Observations Social Bookmarking                   What's this?  Previous Article  |  Table of Contents  |  Next Article  Blood, 1 August 2003, Vol. 102, No. 3, pp. 1000-1006 NEOPLASIA CALM-AF10 is a common fusion transcript in T-ALL and is specific to the TCR lineage Vahid Asnafi, Isabelle Radford-Weiss, Nicole Dastugue, Chantal Bayle, Daniel Leboeuf, Christiane Charrin, Richard Garand, Marina Lafage-Pochitaloff, Eric Delabesse, Agnès Buzyn, Xavier Troussard, and Elizabeth Macintyre From the Laboratoire d'Hematologie, Service d'Histologie, Embryologie, et de Cytogenetic, Service d'Hematologie Clinique, Centre Hospitalier Universitaire (CHU) Necker Enfants Malades and University Paris V; Laboratoire d'Hematologie et Genetique des Hemopathies, CHU Purpan, Toulouse; Laboratoire d'Hematologie, Institut Gustave Roussy, Villejuif; Laboratoire d'Hematologie, CHU Edouard Herriot, Lyon; Institut de Biologie des Hopitaux, CHU Nantes; Departement de Biologie, Unite de Genetique Hematologique, Institut Paoli Calmette, Marseille; and Laboratoire d'Hematologie, CHU Caen, France    Abstract Top Abstract Introduction Patients, materials, and methods Results Discussion References   The t(10;11)(p13-14;q14-21) associated with CALM-AF10 is considered to be rare and associated with a variety of acute lymphoid and myeloid leukemias. Twelve (9%) of 131 unselected T-cell acute lymphoid leukemias (T-ALLs) expressed CALM-AF10 by reverse transcription–polymerase chain reaction or fluorescence in situ hybridization (or both), including 8% of children and 10% of adults, of whom only half demonstrated a t(10;11) by classical cytogenetics. CALM-AF10 was not found in T-cell–receptor (TCR) lineage T-ALLs, as defined by expression of TCR, cytoplasmic TCR, or TCRVDJ rearrangement in immature cytoplasmic TCR- cases, compared with 19% of TCR T-ALLs and 33% of immature / T-ALLs. The latter differed from their CALM-AF10- immature counterparts by a CD5+/CD2-phenotype, as found in TCR but not TCR T-ALLs and in their TCR and TCR configurations, altogether suggesting that CALM-AF10+ immature /T-ALLs are TCR precursors and that, within T-ALL, CALM-AF10 is specific for this lineage. Nine of 12 immature CALM-AF10 T-ALLs demonstrated 3' fusion transcripts, whereas 6 of 7 TCR T-ALLs demonstrated 5' fusion transcripts. The latter retain the AF10 extended LAP/PHD domain necessary for homo-oligomerization. All 8 patients with CALM-AF10+TCR T-ALLs are alive, compared with only 3 of 12 with immature CALM-AF10+ T-ALLs. Six CALM-AF10+ non-T acute leukemias all expressed CD7 and demonstrated T-restricted TCR rearrangements, suggesting that they may also be related to the TCR lineage. CALM-AF10 is therefore the most common fusion protein in T-ALL. It requires molecular and immunophenotypic characterization for appropriate prognostic evaluation and should be included in diagnostic screening panels of T-ALL and immature acute leukemias. Analysis of immature CALM-AF10+ leukemias will also facilitate analysis of the early stages of development of the TCR lineage.    Introduction Top Abstract Introduction Patients, materials, and methods Results Discussion References   T-lineage acute lymphoid leukemias (ALLs) are characterized by a relatively high proportion of cases with a normal (30%) or failed (20%) karyotype. The most commonly recognized abnormality corresponds to those involving the T-cell–receptor / (TCR/) locus on chromosome (chr.) 14q11.1 Oncogenic fusion proteins are described in fewer than 10% of cases, with the most common corresponding to mixed-lineage leukemia (MLL)/chr.11q23 fusions.2 These must be distinguished from other chr.11q abnormalities such as the t(10;11)(p13-14;q14-21), which leads to expression of the CALM-AF10 fusion transcript (FT).3 Distinction of this abnormality from t(10;11)(p13-14;q23) involving MLL-AF10 can be difficult and fluorescence in situ hybridization (FISH) or reverse transcription–polymerase chain reaction (RT-PCR) analysis is often required. The t(10;11)(p13-14;q14-21) corresponds to a rare translocation initially described in immature T-ALL by the Groupe Français de Cytogénétique Hématologique (GFCH).4-6 It has also been described in lymphomas and acute leukemias (ALs) of several lineages, including myeloid, megakaryocytic, eosinophil, and undifferentiated leukemias (AULs), and as such has been considered to be nonlineage specific.3,7-12 Several different CALM-AF10 FTs have been described, without any evident correlation with leukemia subtype.9,10,13 The incidence of CALM-AF10 in T-ALL is unknown. T-ALLs correspond to a heterogeneous group of ALs arrested at various stages of thymic development.14,15 We have recently used immunophenotypic, genotypic characterization of TCR status and RAG-1/pT real-time quantitative-PCR (RQ-PCR) to demonstrate that T-ALLs reproduce normal stages of thymic development.16 Among surface (s) TCR- T-ALLs, cytoplasmic (c) TCR expression identifies immature (IM) TCR lineage-restricted cases (pre-) and TCR rearrangement status allows separation of their immediate precursors (IM) from IM/, which are likely to include TCR lineage precursors (Figure 1). Thymic precursors undergo TCR rearrangement in a reproducible order.17,18 The TCR locus is the first to rearrange, starting with D2-D3, followed by D2-J1 or V2-D3 and, subsequently, complete V(D)J rearrangement or deletion during TCR rearrangement. V2-D3 and D2-D3 rearrangements can occur in the absence of T-lymphoid transcription factors, whereas J1 rearrangements require T cell–specific factors.19 They are restricted to T-ALL, when they occur in both TCR and TCR lineages. Detection of TCR lineage precursors is facilitated by cTCR, CD1a expression, and CD4/8 double positive (DP) on sCD3- cells, all of which identify cells undergoing selection. In contrast, detection of normal TCR precursors has proved difficult, due to the absence of specific cell surface markers.20 Identification of such markers would facilitate analysis of TCR lineage development. In the present study, we demonstrate that CALM-AF10 occurs in approximately 10% of pediatric and adult T-ALLs and is restricted to T-ALLs expressing TCR and to their precursors. Molecular detection of this FT will not only allow more appropriate management of this subcategory of T-ALL but will provide TCR precursors for analysis of the development of the TCR lineage. View larger version (16K): [in this window] [in a new window]   Figure 1.. Categories of T-ALL. T-ALLs were divided into 3 major immunophenotypic categories: (1) lineage cases (n = 66), which include cases expressing TCR (n = 24) and pre- T-ALLs, which are TCR- but express cTCR+ (n = 42); (2) T-ALLs expressing TCR (n = 32); and (3) immature (IM) T-ALLs, which are TCR- and cTCR - (n = 46). IM T-ALLs were classified on the basis of their TCR, TCR, and TCR gene configurations. IM0 cases demonstrate a germline configuration at all 3 loci and include the most immature, nonlineage-restricted cases. Progressive T-lymphoid restriction is associated with TCR rearrangement (IM), followed by TCR and incomplete TCR DJ in IM, and finally by complete TCR V(D)J rearrangement in IM, the immediate precursors of lineage T-ALLs.16 The incidence of CALM-AF10 in 131 unselected T-ALLs in each subgroup is indicated. sTCR indicates surface TCR; cTCR, cytoplasmic TCR; GL, germline.      Patients, materials, and methods Top Abstract Introduction Patients, materials, and methods Results Discussion References   Patients and diagnostic analysis Diagnostic peripheral blood or bone marrow samples from 144 T-ALLs, defined by expression of c/sCD3 and sCD7, were analyzed after the patients provided informed consent according to the Declaration of Helsinki. These included 131 unselected consecutive samples corresponding to all patients from a given center with available material and 13 selected cases with TCR (n = 5), IM, or t(10;11) T-ALL (n = 8). Patients came from 17 clinical centers. The majority of samples had more than 80% blasts. The majority of adult samples (n = 89) were taken prospectively within the LALA94 multicenter trial (coordinator Denis Fière) and the majority of children (n = 55) within the FRALLE93 or FRALLE2000 trials (coordinator André Baruchel). Approval was obtained from the LALA and FRALLE review boards (CCPPRB) for these studies. T-ALLs were considered adult if the individual was older than 15 years. Details of patient classification, DNA and RNA extraction, immunophenotype, and TCR analysis are as described previously.16 Non-T ALs were all included following identification of a t(10;11) by classical karyotype analysis, in collaboration with the GFCH. They were treated on a variety of clinical protocols in 5 centers. RT-PCR detection and characterization of CALM-AF10 For diagnostic screening of CALM-AF10, 0.2 µg cDNA equivalent, was amplified in 2 separate RT-PCR assays, using CALM S1770 (GCAATCTTGG CATCGGAAAT) and either AF10 AS559 (CGATCATGCG GAACAGACTG) or AF10 AS1002 primers (GCGCTTCAAT GATCCAGATAT AGAG; Figure 1). RT-PCRs were performed in 50 µL with MgCl2 2.5 mM; deoxyribonucleoside triphosphate (dNTP) 0.2 mM; dimethyl sulfoxide (DMSO) 25%; primers 0.4 µM each; Taq Gold (Applied Biosystems, Warrington, United Kingdom) 2 U; gold buffer II 1 time for 35 cycles at 94°C for 30 minutes, 60°C for 1 minute, and 72°C for 1 minute with an initial cycle of 94°C for 8 minutes and a final elongation of 10 minutes at 72°C. PCR products were separated on 2% agarose and visualized with ethidium bromide. Positive cases were further characterized by cDNA amplification and a nested RT-PCR: CALM S2007 (TCCTGTAATG ACGCAACCAA CC) and AF10 AS781 (CCCGTTTGCT CTTTTTCAGC TT) for 5' FT or AF10 AS1002 for 3' FT, and direct sequencing using BigDye terminator cycle kits (Applied Biosystems) on an ABI 3700. Karyotype and FISH analysis Conventional cytogenetic analysis was performed on 24-hour unstimulated bone marrow cultures with or without synchronization according to standard procedures. Karyotypes were described according to Mitelman.21 At least 20 mitoses were analyzed before considering a case normal. For FISH, 2 adjacent AF10 yeast artificial chromosomes (YACs), 807B3 and 815C7, from the Centre d'Etudes du Polymorphisme Humaine (CEPH) library were labeled with fluorescein isothiocyanate (FITC). For CALM, 2 tetramethyl rhodamine-labeled CEPH YACs, proximal 785C1 and distal 914D9, have been previously identified to detect CALM rearrangement as a split signal.3 Dual-color analysis was performed as previously described,22 and 100 metaphases or interphase nuclei were analyzed.    Results Top Abstract Introduction Patients, materials, and methods Results Discussion References   CALM-AF10 is a common FT in T-ALL and is frequently undetected by classical cytogenetics CALM-AF10 RT-PCR analysis of 131 consecutive adult and pediatric T-ALLs identified 12 positive cases (9%), including 4 (8%) of 49 children under 15 years of age, and 8 (10%) of 82 adults aged 15 years or older (Table 1). Seven (58%) of these were not detected by classical cytogenetics. In collaboration with the GFCH, a further 13 selected cases with TCR or t(10;11) T-ALL were analyzed. Eight were CALM-AF10+, including 3 of 5 TCR and 5 of 8 IM or t(10;11) T-ALLs. One case (UPN4562) with a normal karyotype, which could not be tested by RT-PCR but showed a CALM-AF10 fusion with duplication of the derivative 11 in interphase nuclei by FISH, was detected as a split CALM signal, which colocalized with AF10 (data not shown). Of the 20 CALM-AF10 T-ALLs, 12 (60%) demonstrated a t(10;11) by conventional analysis, 3 had a normal karyotype, 3 were classified as failures, and 2 demonstrated other abnormalities. FISH analysis confirmed the CALM-AF10 in all 10 cases tested. Optimal CALM-AF10 detection therefore requires molecular screening by RT-PCR or FISH. View this table: [in this window] [in a new window]   Table 1.. Characteristics of CALM-AF10+ T-ALLs   Characteristics of CALM-AF10 T-ALLs CALM-AF10 was slightly more frequent in male patients (M/F ratio 2.8 compared with 2.3 for CALM-AF10- cases). The incidence of CALM-AF10 did not vary with age; positive cases ranged from 3 to 43 years (mean, 20.7 years). This was slightly lower than the mean age for CALM-AF10- cases (22.3 years 3 months to 78 years). As shown in Table 2, the majority of pediatric CALM-AF10 T-ALLs were TCR, whereas the majority of adult cases were IM. View this table: [in this window] [in a new window]   Table 2.. Immunophenotype and clonally rearranged TCR alleles in CALM-AF10 T-ALLs compared to negative cases with a similar stage of maturation arrest   Immunophenotype and TCR genotype T-ALLs were divided into 3 major immunophenotypic categories. None of the 78 lineage T-ALLs and their immediate IM precursors expressed CALM-AF10. In contrast, 8 (25%) of 32 of TCR and 12 (41%) of 29 of IM/ T-ALLs were CALM-AF10+, including 5 (19%) of 27 unselected TCR and 7 (33%) of 21 unselected IM/ T-ALLs (Figure 1). All 5 IM0 were CALM-AF10-. We therefore undertook to determine whether IM/ cases could represent TCR precursor T-ALLs. No significant immunophenotypic differences were observed between CALM-AF10+ and CALM-AF10- TCR T-ALLs. The immunophenotypic features that distinguished TCR-expressing T-ALLs from their lineage counterparts included less frequent expression of CD2, CD1a, and CD4/8 DPs (P < .01) but more frequent expression of CD117, CD13, or CD33 (P .01). As expected, IM T-ALLs were more frequently CD34+, CD117+, CD13/33+, CD1a-, and CD4/8 double negative (DN) than cases expressing TCR. Within the IM/ category, the most striking immunophenotypic differences involved CD2 and CD56. IM/ CALM-AF10+ T-ALLs were virtually all negative for CD2 and CD56, whereas CD2 was expressed by 71% (P < .01) and CD56 by 43% (P = .02) of IM/ CALM-AF10- cases. The only T-lymphoid antigens to be reproducibly expressed by CALM-AF10+ IM/ T-ALLs were CD5, TdT, and, by definition, cCD3 and CD7. The absence of T-lymphoid antigen expression was not, however, due to relative immaturity, because expression of CD34, CD117, and CD13/33 was similar in IM/ CALM-AF10+ and CALM-AF10- cases. A CD5+CD2- phenotype was particularly common in TCR T-ALLs, because it was found in 53% (17 of 32) of TCR compared with only 7% of TCR lineage cases (P < .01). This phenotype was also much more frequent in IM/ CALM-AF10+ compared with CALM-AF10- cases (P = .02), suggesting that it may identify TCR precursors. The type of TCR rearrangements differed in CALM-AF10+ and CALM-AF10- IM/ T-ALLs. Rearranged TCR alleles in CALM-AF10+ cases were predominantly D2-J1 and 76% involved a J segment, indicative of T-lymphoid restriction. The majority (61%) of rearranged alleles in CALM-AF10- cases were non–T restricted. TCR rearrangements in CALM-AF10+ cases involved the functional VfI/V9 segments in the majority, in contrast to CALM-AF10- cases, when half involved the V10 and V11 pseudogenes. CALM-AF10+ IM/ T-ALLs therefore show no evidence of TCR lineage restriction, but are more likely to have undergone J and end-stage TCR rearrangement than CALM-AF10- cases at a similar stage of maturation arrest. These TCR profiles reinforce the immunophenotypic evidence of a TCR lineage origin for immature CALM-AF10+ T-ALLs. AF10 content correlates with different stages of maturation arrest CALM-AF10 FTs demonstrate extensive diversity and alternative splicing (Table 1; Figure 2). Detailed analysis of CALM-AF10+ cases with specific primers, followed by direct sequencing, confirmed the specificity of the different RT-PCR products. All 19 sequenced were in-frame. They included all previously described FTs. AF10 sequences starting at nucleotides 424 or 589 were considered as 5' FTs. These retained the extended LAP/PHD, which was lost in 3' FTs at bp 883/979. Nine CALM-AF10+ demonstrated 5' FTs and ten 3' FTs, but the distribution differed strikingly in immature and mature T-ALLs; 3' FTs predominated in IM/ T-ALLs (9 of 12), compared with only 1 of 7 TCR cases (P = .01). These data suggest that the presence of 5' FT sequences, including the extended LAP/PHD, are necessary for maturation toward the TCR lineage, whereas their absence leads to a more immature stage of maturation arrest. View larger version (54K): [in this window] [in a new window]   Figure 2.. CALM-AF10. (Top) CALM-AF10 fusion gene. Wild-type AF10 functional domains are indicated and the cysteines of the extended LAP/PHD motif are detailed. CALM and AF10 exon nomenclatures are reference sequences NM007166 and NM004641, respectively. Nucleotide nomenclature is identical to that used previously.10 Identified FT points within CALM and AF10 are shown as vertical arrows with solid lines, alternative splice sites as vertical arrows with dotted lines. Position of RT-PCR (black) and sequencing (gray) primers are indicated. (Bottom) RT-PCR products using AF10 AS1002 (upper) or AF10 AS559 (lower) primers in CALM-AF10+ patients with IM or TCR T-ALLs. Unique patient numbers (UPNs) are indicated above each lane. MWM indicates molecular weight markers.   Prognosis of CALM-AF10 depends on the stage of maturation arrest All 3 adults with TCR CALM-AF10+ are in complete remission at 46 to 64 months from diagnosis, whereas 8 of 9 adults with IM T-ALLs have died, with the only survivor being one of the 3 having undergone allogeneic transplantation. Similarly, all 5 children with CALM-AF10+ TCR ALL are in remission 8 to 115 months after induction, whereas 2 of the 3 pediatric patients with IM T-ALLs have relapsed and the third has undergone allogeneic transplantation (Table 1). Although the numbers are limited, CALM-AF10 in IM T-ALL identifies a poor prognostic subgroup, whereas there is no evidence that it is pejorative in TCR T-ALL. All non-T lineage CALM-AF10+ ALs demonstrate TCR rearrangement Six cases of cCD3- AL with CALM-AF10 were analyzed (2 acute myeloid leukemia [AML], 1 B-lineage ALL, 3 AUL; Table 3). Half demonstrated mediastinal enlargement. Surprisingly, 2 had 5' FTs, which included the entire ext-LAP/PHD. All had undergone TCR rearrangement, which was predominantly bialleleic. Interestingly, 1 of the 2 cases with 5' FT was the only one to have undergone TCR V(D)J rearrangement (V1-J1) and the only one with bialleleic end-stage TCR rearrangement. Three of the 4 cases with 3' FT had undergone D2-J1, similar to the profiles seen in IM T-ALLs. All but one had undergone TCR rearrangement. TCR profiles were in general immature, because half involved V11. Immunophenotypes were variable, but were universally CD7+, CD34+, or CD117+ but CD2- and cCD3-. CALM-AF10 was originally described in the U937 myelomonocytic cell line, derived from a patient with diffuse histiocytic lymphoma.7 We confirmed the 3' CALM-AF10 in the clone used here. Interestingly, although TCR was not rearranged, this line also demonstrated early, nonrestricted, TCR V2-D3 rearrangement. View this table: [in this window] [in a new window]   Table 3.. Characteristics of CALM-AF10+ non-T ALs      Discussion Top Abstract Introduction Patients, materials, and methods Results Discussion References   In this manuscript, we demonstrate that CALM-AF10 FTs occur in approximately 10% of T-ALLs and are restricted to TCR and IM/ cases, whose immunophenotypic and TCR configurations are highly suggestive of TCR precursors. Correlation of the type of CALM-AF10 FT with leukemic subtype suggests that AF10 content differs with the stage of maturation arrest. Mature CALM-AF10 cases expressed CD5, TCR, and CD4 or CD8 (or both) and, as such, were easily recognizable as T-ALL. In contrast, immature cases expressed few T-lineage markers other than CD5 in two thirds, TdT, cCD3, and CD7 and were often positive for CD13, CD33, or CD34. They are therefore easily mistaken for AML M0 if cCD3 is not assessed. Several arguments demonstrate, however, that they are indeed T-ALLs. All were myeloperoxidase negative, most by both cytochemistry and immunophenotype. All expressed cCD3 and had undergone at least TCR and mostly TCR rearrangement. The identification of J rearrangements in the majority demonstrates that they are T-lineage restricted.19 The IM CALM-AF10+ T-ALLs resembled TCR T-ALLs, particularly with regard to their CD2 negativity and TCR and TCR profiles, which were indicative of differentiation toward functional TCR cells, with no evidence of TCR lineage differentiation, because none had undergone TCR V(D)J or TCR deletion (IM). CD2 expression was constant in TCR lineage cases but present in less than half of TCR T-ALLs and virtually no IM CALM-AF10+ T-ALL, compared with over 70% of CALM-AF10- IM/ cases. This was not due to relative immaturity, at least based on expression of CD34, CD117, and CD13/33. CD5 and CD2 are expressed at an early stage of thymic development and there is no evidence of a CD5+CD2- population during T lymphopoiesis.20,23 CD2 negativity and, more specifically, a CD5+CD2- phenotype is therefore relatively specific to TCR lineage leukemia, including their precursors. CD2 negativity is well recognized in TCR T-ALL24,25 and has also been described in a subset of human epithelial cells and the majority of ruminant normal TCR cells.26,27 CD2 is the ligand for the leukocyte function–associated antigen 3 (LFA-3) adhesion molecule expressed by antigen-presenting cells.28 It acts as a costimulatory molecule in major histocompatibility complex (MHC)–mediated activation and has also been suggested to play a role in thymocyte development.29,30 Although the functional significance of CD2 on TCR cells is unknown, it may modulate MHC class I–like activation on epithelial TCR cells by engagement of molecules such as MICA/B.26 CD56 was also uniformly negative in CALM-AF10+ IM T-ALLs, but was expressed by half the CALM-AF10- IM/ cases. This suggests either that CD56 is expressed preferentially by immature T-ALLs that are committed to lineages other than TCR, or that its expression is down-regulated earlier in the presence of CALM-AF10. Taken together, the immunophenotypic and TCR genotypic features of CALM-AF10+ IM T-ALLs demonstrate that these leukemias correspond to TCR precursors. Identification of physiologic TCR precursors has proven difficult, largely due to the absence of specific surface markers other than the TCR. Our data suggest that TCR precursors will be found within the cCD3+/CD7+/CD5+/CD2- compartment, which is likely to demonstrate TCR D2-J1 or TCR V9/VfI rearrangements. TCR lineage specification may also precede cCD3 expression, because 5 non–T-ALLs with CALM-AF10 showed TCR D2-J1 rearrangement. TCR rearrangements occur in 3% to 14% of AML cases, especially those that express TdT or lymphoid surface antigens or both.31 TCR D2-J1 has also been identified in very immature CD7+ T-ALLs, which are predominantly CD2-, CD34+, CD38+.32 Only half expressed cCD3 or CD5, and as such would be classified as AML. Detection of cCD3 has, however, only recently been widely used in classification of ALs and its reproducible interpretation in multicenter studies is difficult, because of variable permeabilization techniques.33 CALM-AF10+ non-T ALs may correspond to the earliest TCR precursor, or at least to a precursor with TCR potential. They are likely to represent expansions from a thymic rather than a bone marrow precursor, based on their frequent association with mediastinal enlargement, CD7 expression, and TCR rearrangements. It is likely that this corresponds to the recognized multipotent thymic CD34+, CD33+, CD7+/+, CD45RA+, CD2/5/1a- precursor.20,23,34 These ALs, however, demonstrate considerable differentiation, as evidenced by myeloperoxidase expression in 2 and sCD22 in one. It is possible that other unidentified abnormalities prevent T-lymphoid differentiation in a thymic precursor that has undergone TCR rearrangement and that the B-lymphoid or myeloid differentiation reflects default differentiation, as previously described for murine thymocyte differentiation toward the myeloid lineage in response to interleukin 2 (IL-2) in cells with TCR rearrangement.35 This possibility is in keeping with the recently recognized plasticity of hematopoietic progenitors and suggests that such plasticity be retained by early thymocytes. CALM-AF10+ non-T AL could also correspond to expansions of an extrathymic precursor. The only recognized extrathymic precursors that undergo TCR and TCR rearrangement are those of the TCR lineage, a significant proportion of which develop in the intestine.36,37 Determination of the physiologic relevance of the model of TCR differentiation proposed here requires in vitro cell culture from selected human progenitor populations. Our data facilitate selection of the starting population. Identification of precursors will also allow further analysis of the genetic mechanisms leading to their leukemic transformation and determination of whether these differ from those involved in lineage T-ALLs. The majority of cases with 5' FTs containing the LAP/PHD AF10 domain are TCR ALLs, whereas those having lost it are arrested at an earlier stage. This protein/protein interaction motif mediates homo-oligomerization and is conserved in several proteins, including MLL.38-40 In AML with MLL-AF10, the AF10 ext-LAP/PHD is always deleted, as are the MLL PHD fingers.41 Both retain the COOH terminal leucine zipper (LZ) domain necessary for malignant transformation.42 The presence of the AF10 LZ in the absence of the PHD fingers therefore leads to an immature stage of maturation arrest in both MLL-AF10 and CALM-AF10, whereas the presence of both motifs allows differentiation to a mature TCR phenotype. Although deletion and mutational analyses of these different FTs, as well as intracellular localization analysis, will be necessary for definitive proof, it is likely that the AF10 protein plays a significant role in the early stages of hematopoietic differentiation, particularly with regard to TCR lineage commitment. This is similar to the BCR-ABL leukemias, when BCR content determines, or is at least associated with, distinct stages of maturation arrest.43-45 The CALM protein plays a role in clathrin-mediated endocytosis and the recruitment of proteins into coated pits.46 CALM may simply provide the promoter, which would imply that it is expressed in early, nonrestricted lymphoid precursors. Because, however, virtually all the CALM protein, including the clathrin-binding domain, is maintained in the fusion protein, it may also lead to delocalization of AF10. CALM is normally localized to the cell membrane and to the endocytotic vesicles, whereas AF10 is normally found in the cytoplasm and the nucleus. Murine AF10 is expressed predominantly in the testis and kidney, but also at lower levels in the thymus and brain. It may therefore play a role in normal T lymphopoiesis.39,47 Molecular screening for CALM-AF10 is justified because more than 50% of unselected cases were not detected by classical karyotype analysis. This was due predominantly to the fact that the mitoses analyzed did not correspond to the leukemic clone. Undetected CALM-AF10 was rare in cases with clonal abnormalities. Routine molecular screening by RT-PCR or FISH for this abnormality is therefore advisable. RT-PCR has the advantage of allowing distinction of the different forms of fusion transcript, which is important for prognostic assessment. RT-PCR will also allow baseline analysis for subsequent quantitative PCR follow-up. Real-time quantification of CALM-AF10 is likely to be complicated by the extensive variability of the FTs, which can also complicate diagnostic screening, if high-stringency conditions are not used. Alternatively, FISH allows rapid distinction between CALM-AF10 and MLL-AF10. The CALM-AF10 FT is situated on the derivative 10 chromosome, reflecting the 5' telomere, 3' centromere orientation of CALM and AF10. AF10-CALM FTs are in-frame4,48 but do not contain oncogenic motifs.13 A hybrid approach, with karyotype and FISH screening, followed by RT-PCR characterization of positive or doubtful cases, may be optimal. Screening of AML with normal or failed cytogenetics is also justifiable because CALM-AF10 is rarely missed in patients with an abnormal karyotype, which is more frequent in AML than T-ALL. CALM-AF10 IM T-ALLs clearly have a poor prognosis, both with regard to initial response to induction chemotherapy and overall survival. Because the CALM-AF10+ cCD3- patients with AL were treated on a variety of different protocols, we did not attempt to assess their clinical outcome, particularly because 4 of 6 underwent allografting, but CALM-AF10 is recognized to have a poor prognosis in IM AL.3 Given similarities between the functional domains in the MLL-AF10 and 3' CALM-AF10 fusions and in their stage of maturation arrest, it may be appropriate to regroup both of these minor subcategories into a single, high-risk immature AL protocol. In contrast, TCR, CALM-AF10 T-ALLs demonstrate comparable survival to TCR lineage cases on, at least adult, T-ALL protocols (data not shown). In practical terms, if the characteristics of CALM-AF10 AL demonstrated here are confirmed, it will be reasonable to intensify CALM-AF10+ AML and sCD3- T-ALLs, but not sCD3+ cases, which probably all belong to the TCR lineage. This is more compatible with initial, rapid assignment to induction chemotherapy in an appropriate protocol than stratification by type of FT, at least with current diagnostic practice. In conclusion, we show that CALM-AF10 is a common FT in T-ALL, when it is restricted to cases of the TCR lineage. Because the prognostic significance appears to depend on stage of maturation arrest, appropriate assessment of these cases requires prospective immunophenotypic and molecular subtype analysis within the context of T-ALL and AML protocols. It will obviously be interesting to determine the transcriptional profiles of these cases and to compare them with the signatures of currently recognized T-ALL and AML categories.    Acknowledgements   We thank Dorothée Menage for secretarial assistance, Florence Achenbaum-Cymbalista (Hôtel Dieu, Paris), Jean-Michel Boiron (Bordeaux), Judith Landman-Parker, (Trousseau, Paris), Anne Hagemeijer and Peter Vandenberg (Leuven), Bernard Lenormand (Rouen), Pascale Cornillet-Lefevre (Reims), and all the biologists and clinicians of the LALA adult and FRALLE pediatric French ALL groups for providing T-ALL samples and results.    Footnotes   Submitted September 24, 2002; accepted March 22, 2003. Prepublished online as Blood First Edition Paper, April 3, 2003; DOI 10.1182/blood-2002-09-2913. Supported by the fondation contre la leucémie de la Fondation de France, l'Association de la Recherche sur le Cancer (ARC), the Direction de Recherche Clinique de l'Assistance Publique-Hôpitaux de Paris (PHRC 97-106), and the Biomed-2 BMH4-CT98-3936 concerted action. The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked "advertisement" in accordance with 18 U.S.C. section 1734. Reprints: Elizabeth Macintyre, Laboratoire d'Hématologie, Tour Pasteur, Hôpital Necker, 149-161, rue de Sèvres, 75743 Paris Cedex 15, France; e-mail: elizabeth.macintyre{at}nck.ap-hop-paris.fr.    References Top Abstract Introduction Patients, materials, and methods Results Discussion References   Schneider NR, Carroll AJ, Shuster JJ, et al. New recurring cytogenetic abnormalities and association of blast cell karyotypes with prognosis in childhood T-cell acute lymphoblastic leukemia: a pediatric oncology group report of 343 cases. Blood. 2000;96: 2543-2549.[Abstract/Free Full Text] Hayette S, Tigaud I, Maguer-Satta V, et al. Recurrent involvement of the MLL gene in adult Tlineage acute lymphoblastic leukemia. Blood. 2002;99: 4647-4649.[Free Full Text] Dreyling MH, Schrader K, Fonatsch C, et al. MLL and CALM are fused to AF10 in morphologically distinct subsets of acute leukemia with translocation t(10;11): both rearrangements are associated with a poor prognosis. Blood. 1998;91: 4662-4667.[Abstract/Free Full Text] Dreyling MH, Martinez-Climent JA, Zheng M, Mao J, Rowley JD, Bohlander SK. The t(10; 11)(p13;q14) in the U937 cell line results in the fusion of the AF10 gene and CALM, encoding a new member of the AP-3 clathrin assembly protein family. Proc Natl Acad Sci U S A. 1996;93: 4804-4809.[Abstract/Free Full Text] Groupe Francais de Cytogenetique Hematologique (GFCH). t(10;11)(p13–14;q14–21): a new recurrent translocation in T-cell acute lymphoblastic leukemias. Genes Chromosomes Cancer. 1991;3: 411-415.[Medline] [Order article via Infotrieve] Groupe Francais de Cytogenetique Hematologique. Collaborative study of karyotypes in childhood acute lymphoblastic leukemias. Leukemia. 1993;7: 10-19.[Medline] [Order article via Infotrieve] Kobayashi H, Thirman MJ, Rowley JD. U937 cell line has a t(10;11)(p13–14;q14–21) rather than a deletion of 11q. Genes Chromosomes Cancer. 1995;13: 217-218.[Medline] [Order article via Infotrieve] Kumon K, Kobayashi H, Maseki N, et al. Mixedlineage leukemia with t(10;11)(p13;q21): an analysis of AF10-CALM and CALM-AF10 fusion mRNAs and clinical features. Genes Chromosomes Cancer. 1999;25: 33-39.[CrossRef][Medline] [Order article via Infotrieve] Carlson KM, Vignon C, Bohlander S, Martinez-Climent JA, Le Beau MM, Rowley JD. Identification and molecular characterization of CALM/AF10 fusion products in T cell acute lymphoblastic leukemia and acute myeloid leukemia. Leukemia. 2000;14: 100-104.[CrossRef][Medline] [Order article via Infotrieve] Bohlander SK, Muschinsky V, Schrader K, et al. Molecular analysis of the CALM/AF10 fusion: identical rearrangements in acute myeloid leukemia, acute lymphoblastic leukemia and malignant lymphoma patients. Leukemia. 2000;14: 93-99.[CrossRef][Medline] [Order article via Infotrieve] Jones LK, Chaplin T, Shankar A, et al. Identification and molecular characterisation of a CALM-AF10 fusion in acute megakaryoblastic leukaemia. Leukemia. 2001;15: 910-914.[CrossRef][Medline] [Order article via Infotrieve] Salmon-Nguyen F, Busson M, Daniel M, Leblanc T, Bernard OA, Berger R. CALM-AF10 fusion gene in leukemias: simple and inversion-associated translocation (10;11). Cancer Genet Cytogenet. 2000;122: 137-140.[CrossRef][Medline] [Order article via Infotrieve] Narita M, Shimizu K, Hayashi Y, et al. Consistent detection of CALM-AF10 chimaeric transcripts in haematological malignancies with t(10;11)(p13; q14) and identification of novel transcripts. Br J Haematol. 1999;105: 928-937.[CrossRef][Medline] [Order article via Infotrieve] Breit TM, Wolvers-Tettero IL, van Dongen JJ. Phenotypic and genotypic characteristics of human early T-cell differentiation: the T-cell acute lymphoblastic leukaemia model. Res Immunol. 1994;145: 139-143; discussion 155-138.[CrossRef][Medline] [Order article via Infotrieve] Gomez E, San Miguel JF, Gonzalez M, et al. Heterogeneity of T cell lymphoblastic leukaemias. J Clin Pathol. 1991;44: 628-631.[Abstract/Free Full Text] Asnafi VBK, Boulanger E, Comba B, et al. Analysis of TCR, pT{alpha} and RAG-1 in T acute lymphoblastic leukemias improve understanding of early human T lymphoid lineage commitment. Blood. 2002;101: 2693-2703. Blom B, Res P, Noteboom E, Weijer K, Spits H. Prethymic CD34+ progenitors capable of developing into T cells are not committed to the T cell lineage. J Immunol. 1997;158: 3571-3577.[Abstract] Blom B, Verschuren MC, Heemskerk MH, et al. TCR gene rearrangements and expression of the pre-T cell receptor complex during human T-cell differentiation. Blood. 1999;93: 3033-3043.[Abstract/Free Full Text] Lauzurica P, Krangel MS. Enhancer-dependent and -independent steps in the rearrangement of a human T cell receptor delta transgene. J Exp Med. 1994;179: 43-55.[Abstract/Free Full Text] Spits H, Blom B, Jaleco AC, et al. Early stages in the development of human T, natural killer and thymic dendritic cells. Immunol Rev. 1998;165: 75-86.[CrossRef][Medline]