Aggressive Langerhans cell histiocytosis ... - Wiley Online Library

16 downloads 370187 Views 317KB Size Report
Sep 14, 2007 - from the Langerhans cell tumor late in its course with an antibody raised .... Adobe photoshop elements (v.2.0) software (Adobe Systems, San.
Aggressive Langerhans cell histiocytosis following T-ALL: Clonally related neoplasms with persistent expression of constitutively active NOTCH1 Scott J. Rodig,1 Ethan G. Payne,1 Barbara A. Degar,2,3 Barrett Rollins,3 Andrew L. Feldman,4 Elaine S. Jaffe,5 Arlene Androkites,2 Lewis B. Silverman,2 Janina A. Longtine,1 Jeffery L. Kutok,1 Mark D. Fleming,6* and Jon C. Aster1* Langerhans cell histiocytosis (LCH) and related entities are neoplasms of unknown pathogenesis. Here, we describe studies assessing the role of NOTCH1 mutations in LCH, which were based on a case of fatal Langerhans cell tumor after T-cell acute lymphoblastic leukemia (T-ALL). Although the two types of neoplasm in this patient were temporally and pathologically distinct, molecular analyses showed that they harbored the same T-cell receptor gene rearrangements and two activating NOTCH1 mutations involving exons 27 and 34. The exon 27 mutation altered a conserved cysteine residue in the N-terminal portion of the NOTCH1 heterodimerization domain, while the mutation in exon 34 introduced a premature stop codon that results in the deletion of C-terminal negative regulatory PEST domain. Analysis of cDNA prepared from the aggressive Langerhans cell tumor showed that the NOTCH1 mutations were aligned in cis, a configuration that caused synergistic increases in NOTCH1 signal strength in reporter gene assays. Immunohistochemistry confirmed that the Langerhans cell tumor also expressed NOTCH1 protein. Although these data suggested that NOTCH1 mutations might contribute to the pathogenesis of typical sporadic LCH and related neoplasms occurring in the absence of T-ALL, an analysis of 24 cases of LCH and Rosai–Dorfman Disease occurring in patients without an antecedent history of T-ALL revealed no mutations. Thus, activating NOTCH1 mutations appear to be unique to aggressive Langerhans cell tumors occurring after T-ALL. Persistent expression of NOTCH1 in such tumors suggests that Notch pathway inhibitors could have a role in C 2007 Wiley-Liss, Inc. the treatment of these unusual neoplasms. Am. J. Hematol. 83:116–121, 2008. V

Introduction Langerhans cell histiocytosis (LCH) is a rare, clinically heterogenous neoplasm of immature dendritic cells that is most common in children [1–4]. In its most aggressive form, LCH and its histologic variant Langerhans cell sarcoma (LCS), affect multiple organ systems and is often fatal despite chemotherapy. The etiology of LCH and LCS are unknown; most cases are cytogenetically normal, and a specific molecular defect has yet to be linked to the disease. LCH/LCS has been observed in association with nonHodgkin and Hodgkin lymphomas, and more recently, with T-cell acute lymphoblastic leukemia/lymphoma (T-ALL) [5,6]. Remarkably, coincident LCH/LCS and T-ALL show a strong tendency to share the same T cell receptor (TCR) gene rearrangements, indicating that the two tumors are clonally related [6]. Progenitors with lymphoid potential can give rise to dendritic cells [7], but the molecular mechanisms that regulate dendritic cell development from multipotent progenitors remains incompletely understood. Similarly, the basis for the epigenetic reprogramming that converts T-ALL to LCH/LCS is also unknown. Notch signaling regulates normal pre-T cell development [8], and gain-of-function mutations in NOTCH1 appear to be the most common acquired genetic lesion found in human T-ALL [9]. NOTCH1 encodes a transmembrane receptor that participates in a novel signaling pathway in which a series of regulated ligand-induced proteolytic cleavages allows the intracellular domain of NOTCH1 to translocate to the nucleus (for recent review, see [10]). Here, Notch combines with other factors to form a short-lived transcriptional activation complex that turns on the expression of downstream target genes. The outcome of Notch

signaling is varied with respect to context and dose, but within multipotent progenitors it often results in the adoption of one cell fate at the expense of a second. For example, NOTCH1 signals in multipotent lymphoid progenitors induce T cell development and block early stages of B cell development [8]. The effect of Notch signaling on dendritic cell development is less certain. Within the thymus, strong Notch signals block dendritic cell differentiation [11,12], but in other contexts it appears that Notch can have a positive effect on dendritic cell development [13,14]. Here we describe studies that were inspired by a case of fatal LCH/LCS that occurred in a patient who was in remission following treatment of T-ALL. Molecular analysis of

1 Department of Pathology, Brigham and Women’s Hospital, Boston, Massachusetts; 2Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts; 3Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts; 4Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota; 5Laboratory of Pathology, National Cancer Institute, Bethesda, Maryland; 6Department of Pathology, Children’s Hospital, Boston, Massachusetts

Contract grant sponsor: JCA; Contract grant number: CA082308. Contract grant sponsors: BAD and MDF; Contract grant number: AI050225. *Correspondence to: Jon C. Aster, Department of Pathology, Brigham and Women’s Hospital, Boston, MA. E-mail: [email protected] or Mark D. Fleming, Department of Pathology, Children’s Hospital, Boston, MA. E-mail: mark.fl[email protected]. Received for publication 12 June 2007; Revised 28 June 2007; Accepted 30 June 2007 Am. J. Hematol. 83:116–121, 2008. Published online 14 September 2007 in Wiley InterScience (www.interscience. wiley.com). DOI: 10.1002/ajh.21044

C 2007 Wiley-Liss, Inc. V

American Journal of Hematology

116

http://www3.interscience.wiley.com/cgi-bin/jhome/35105

Figure 1. Morphology and immunophenotype of the patient’s T-ALL and aggressive LCH/LCS. (A) Lymphoblasts (Wright– Giemsa stain, 1,0003), seen in a bone marrow aspirate smear at the time of the patient’s initial leukemia presentation. (B) Langerhans cell tumor (H&E stain, 4003; inset, 1,0003), seen in a skin biopsy taken 18 months following remission of T-ALL. This biopsy was stained with (C) anti-S100, (D) anti-langerin, (E) anti-CD1a, and (F) anti-TdT (4003; (inset F)-control section showing positive nuclear staining of thymocytes in normal thymus, 1,0003), and examined by electron microscopy (G), which revealed Birbeck granules (arrows). (H) Extensive skin lesions seen late in the clinical course. (I) Cellular atypia consistent with LCS, seen in a biopsy from late in the clinical course (H&E stain, 1,0003).

TCR rearrangements revealed that the aggressive LCH/LCS was clonally related to the prior T-ALL, and that both neoplasms harbored identical, synergistic activating mutations in NOTCH1 [9]. Although it seemed possible a priori that the lineage conversion seen in this tumor might be caused by changes in NOTCH1 expression (given its regulatory role in T cell development, and evidence suggesting that Notch signals inhibit intrathymic dendritic cell development), the Langerhans cell tumor expressed mutated NOTCH1 mRNA and stained for NOTCH1 protein. On the basis of these findings, we investigated the alternative possibility that NOTCH1 mutations might be associated with sporadic forms of LCH. However, analyses of the mutational ‘‘hotspots’’ in an additional 24 tumors of Langerhans cell or histiocytic origin failed to uncover additional NOTCH1 mutations. Thus, aggressive Langerhans cell neoplasms following T-ALL appear to be a uniquely associated with expression of constitutively active forms of NOTCH1 in a subset of cases. Results The index case was a patient who presented at the age of 3 years 7 months with a white blood cell count of

American Journal of Hematology DOI 10.1002/ajh

270,000 cells/ll and an anterior mediastinal mass. A bone marrow aspirate smear obtained at presentation showed morphologically typical lymphoblasts (Fig. 1A). These were confirmed to be pre-T cells by flow cytometry, which demonstrated a population of blasts expressing CD2, CD3, CD5, CD7, CD8 (subset), CD34, CD45, HLA-DR, TdT, and CD1a. Cytogenetic studies were not obtained. The patient received induction chemotherapy per a cooperative group protocol based on the Berlin–Frankfurt–Munster (BFM) backbone and achieved a complete remission [15]. After 4 months, therapy was switched to the DFCI Consortium protocol [16]. Eighteen months into the course, a tender ulcer appeared on the posterior trunk. A skin biopsy revealed a dermal and epidermotropic infiltrate comprised of Langerhans cells (Fig. 1B), which expressed S-100, langerin, and CD1a in >95% of tumor cells (Fig. 1C–E, respectively). The tumor cells were negative for TdT (Fig. 1F) and T cell antigens (not shown). Although the Langerhans cells showed the bland monomorphic morphologic features typical of LCH, scattered mitotic figures were seen (Fig. 1B inset; up to 2 mitoses per hpf). By electron microscopy, the tumor cells contained numerous Birbeck granules

117

(Fig. 1G). Cytogenetic studies revealed a normal female karyotype. Despite multiple therapeutic interventions, over the next 12 months the skin lesions increased in size and number (Fig. 1H), leading to recurrent skin infections. Histologic examination of tissue obtained during surgical debridement of infected lesions showed sheet-like growth of the neoplastic cells and increased cellular atypia as compared to the earlier biopsy (Fig. 1I); frequent mitotic figures (up to 5 per hpf); and loss of expression of Langerhans cell antigens, as only approximately 20% of neoplastic cells retained reactivity for langerin, S100, CD1a, and lysozyme (data not shown). These morphologic and immunophenotypic changes were consistent with progression to LCS. The patient experienced several episodes of bacterial sepsis and cutaneous aspergillus infection related to severe immune suppression and impaired skin integrity, and she died of presumed sepsis at the age of 6 years 5 months. An autopsy was not performed. Molecular analysis demonstrated that the T-ALL and Langerhans cell tumors were clonally related. DNA from both tumor types yielded the same bi-allelic clonal TCR rearrangement when amplified with TCRg chain V10 region primers (Fig. 2A), whereas a remission peripheral blood yielded a polyclonal set of PCR products (Fig. 2A, bottom panel). The presence of the same clonal TCR rearrangement in the Langerhans cell tumor early and late in its course was further confirmed by Southern blot analysis with a TCRg J region probe (Fig. 2B), which showed identical TCRg rearrangements in two temporally distinct biopsy samples. The NOTCH1 gain-of-function mutations that are seen in T-ALL preferentially involve the extracellular heterodimerization domain (encoded by exons 26 and 27) and the intracellular negative regulatory PEST domain (encoded by exon 34) [9]. Sequencing of genomic DNA isolated from the T-ALL and the LCH revealed identical heterozygous NOTCH1 mutations in exon 27 and exon 34 in both tumor types (Fig. 2C). The mutation in exon 27 results in a novel cysteine to arginine substitution at residue 1693 (C1693R). The mutation in exon 34 (C7321T) creates a stop codon at amino acid residue 2441 (Q2441X) and leads to the deletion of the C-terminal 115 amino acids, including the PEST domain. In T-ALL, compound mutations are typically aligned in cis in one of the two NOTCH1 alleles, a configuration that produces synergistic increases in NOTCH1 signaling [9]. Analysis of cDNA prepared from the Langerhans cell tumor late in its course revealed that the mutations were present in 50% of NOTCH1 transcripts and aligned in cis (data not shown). Furthermore, immunostaining of a biopsy obtained from the Langerhans cell tumor late in its course with an antibody raised against the intracellular domain of NOTCH1 showed a nuclear and perinuclear pattern of reactivity (Fig. 2E), which is similar to the pattern seen in a subset of TALLs (SJR, JCA, unpublished data). Finally, in a functional assay, expression of the C1693R and Q2441X mutations in cis activated NOTCH1 signaling synergistically (Fig. 2F), providing evidence that these mutations are pathogenically significant. On the basis of the findings above, we also sequenced NOTCH1 alleles in one additional case of coincident LCH/ LCS and T-ALL, which was shown previously to be clonally related [6]. However, NOTCH1 mutations were absent from this tumor (Table I), rendering it noninformative with respect to the possible involvement of NOTCH1. The persistent expression of activated NOTCH1 in this case of aggressive LCH/LCS raised the possibility that such abnormalities might also be found in sporadic forms

118

of LCH. We thus obtained eighteen biopsies of isolated LCH, two biopsies of Rosai–Dorfman disease, and six biopsies of disseminated LCH (also known as Letterer–Siwe disease) from four patients, all without antecedent histories of T-ALL. These cases were derived from pediatric patients and included a wide range of clinical presentations (Table I). Exons 26, 27, and 34 of NOTCH1 were PCR amplified and sequenced. To increase the sensitivity of the analysis in the six biopsies of aggressive LCH, we cloned the products and sequenced sixteen individual clones derived from each PCR product. No NOTCH1 mutations were detected in any of these twenty-six samples, highlighting the unique nature of the index case. Discussion Tumors of Langerhans cell origin are rare and include LCH, which most commonly presents as an isolated mass in the bone of children and young adults. In this most common form, the disease is generally self-limited and has an excellent clinical outcome. However, more aggressive forms of LCH also exist, including multifocal and unisystem disease (Hand–Schuller–Christian disease); multifocal, multisystem disease (Letterer–Siwe disease); and poorly differentiated LCS. There are few effective treatments for aggressive tumors of Langerhans cell origin, and the prognosis is generally poor. Furthermore, the biologic defects that underlie these tumors remain unknown. Unusual cases sometimes lead to insights pertaining to disease pathogenesis that are of broad relevance. An example is the involvement of NOTCH1 in a rare (7;9) translocation in human T-ALL [17], which presaged the discovered of frequent activating point mutations in NOTCH1 in this aggressive human cancer. In a similar vein, we viewed the opportunity to study a case of aggressive LCH/ LCS occurring after T-ALL as a chance to gain new insight into mechanisms that might explain the changes in gene expression that underlie lineage switches in clonally related neoplasms, and possibly the molecular pathogenesis of sporadic forms of LCH. We first hypothesized that the conversion of T-ALL to a tumor of Langerhans cells might be mediated by altered expression and activity of NOTCH1, a member of the NOTCH receptor family that directly influences programs of differentiation through effects on gene expression. Surprisingly, however, we found aggressive LCH/LCS following T-ALL to be associated with persistent expression of NOTCH1 bearing dual activating mutations. One of the mutations discovered in our index case, C1693R, has not been described in T-ALL or other cancers previously. However, C1693 is highly conserved in NOTCH receptors (Fig. 2D), and mutations involving this site have been shown to cause gain of function developmental phenotypes in Drosophila [18]. Thus, to the extent that we are able to analyze our reference case, we find no evidence to suggest that the lineage switch was mediated through down-regulation of NOTCH1. It remains possible that the aggressive LCH/LCS subclone acquired alterations that either (i) interfered with the function of the NOTCH1 transcriptional activation complex, thus inhibiting the expression of Notch target genes; or (ii) which directly influenced the activity of transcription factors implicated in Langerhans cell differentiation, such as Spi-B [12]. Unfortunately, RNA was not available (and DNA was limited) from the T-ALL in our index case, precluding more detailed genomic comparison of the two tumors. Alternatively, the persistent expression of activated NOTCH1 in our index case raised the question of whether NOTCH1 mutations have a role in the etiology of sporadic forms of LCH that are unrelated to T-ALL. However,

American Journal of Hematology DOI 10.1002/ajh

Figure 2. Molecular and functional analysis of antigen receptor rearrangements and NOTCH1 mutations. (A) Analysis of TCRc rearrangements by PCR. DNAs isolated from the patient’s T-ALL (top panel), early aggressive Langerhans cell tumor (middle panel), and peripheral blood during remission of the T-ALL (bottom panel) were amplified with primers specific for the Vc10 region of the TCR-c chain gene and separated by capillary gel electrophoresis. The T-ALL and LCH DNAs yielded two identical PCR products, 143 and 157 base pairs in size, whereas the peripheral blood DNA gave a polyclonal set of PCR products. There were no clonal rearrangements using primers to V regions 1–8, 9, or 11. (B) Analysis of TCRc rearrangements by Southern blot. DNAs isolated from unrelated germline control tissue (left lane), early (middle lane), and late (right lane) Langerhans cell tumors (LCH) were digested with BglII and probed with a fragment from the TCR-c chain gene J region. (C) DNA sequences of exon 27 (left column) and exon 34 (right column) of NOTCH1 derived from the patient’s T-ALL and the Langerhans cell tumor (LCH) late in its course, showing normal (top row) and mutated alleles (middle and bottom rows). In both tumors, approximately 50% of sequences from individual clones were mutated. Superscript numbers correspond to the cDNA sequence. (D) Conservation of mutated NOTCH1 residues (in red) based on sequence alignments. (E) Detection of NOTCH1 protein in the patient’s early Langerhans cell tumor (skin biopsy in Fig. 1B) by immunostaining with anti-NOTCH1. Nuclear and perinuclear dot-like staining is observed. (F) Functional assay of NOTCH1 mutations. U2OS cells were transfected in triplicate with 10 ng of the pcDNA3NOTCH1 plasmids encoding normal NOTCH1 or forms of NOTCH1 bearing the indicated mutations, a NOTCH-responsive luciferase reporter plasmid (250 ng), and an internal Renilla luciferase control plasmid (5 ng). At 44–48 h post-transfection, whole cell lysates were prepared and analyzed using a dual luciferase assay. Fold stimulations are expressed relative to activities observed with an empty pcDNA3 plasmid. Error bars correspond to standard deviations.

American Journal of Hematology DOI 10.1002/ajh

119

TABLE I. Summary of NOTCH1 Analysis in Langerhans Cell Tumors

Histology, immunohistochemistry, and imaging

Case

Age

Sex

Biopsy site

Diagnosis

NOTCH1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

19 m 7y 4y 3y 2y 8y 2y 12 y 2y 3y 5y 6y 2y 19 y 4y 5y 12 y 2y 6y 1y 3y 3y 12 m 3y 8y

M F M M F F M M F M M M M F M M M M F F F M F F M

Bone marrow Posterior pubic ramus Skull Orbit Skull Skull Ear canal Skull Scapula Mastoid bone Skull Orbit Lymph node Lung Left ischium Periorbital soft tissue Skull Temporal bone Lymph node Lymph node Skin Skin Skin Skin Liver, lung

LCH LCH LCH LCH LCH LCH LCH LCH LCH LCH LCH LCH LCH LCH LCH LCH LCH LCH RDD LCH/RDD dLCH dLCH dLCH dLCH LCH/LCS*

Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal**

m 5 months; y 5 years; M 5 male; F 5 female; LCH 5 Langerhans cell histiocytosis; RDD 5 Rosai–Dorfman disease; RDD/LCH 5 tumor with features of Rosai–Dorfman disease and Langerhans cell histiocytosis; dLCH 5 disseminated LCH (Letterer–Siwe disease); Normal 5 unmutated sequence, LCH/LCS* 5 aggressive Langerhans cell tumor following T-ALL (ref. 6), ** 5 normal NOTCH1 in presenting T-ALL and subsequent sarcoma.

sequence analyses of the mutational ‘‘hotspots’’ of NOTCH1 in an additional 26 tumors of Langerhans cell and histiocytic origin from 24 patients failed to identify additional cases with mutated NOTCH1 alleles. Although only a subset of T-ALLs harbors NOTCH1 mutations, the lack of NOTCH1 mutations in such a large cohort of cases suggests that activating mutations in NOTCH1 are an uncommon event in isolated LCH, and that aggressive LCH/LCS following T-ALL has distinctive genetic features. Given the poor response of aggressive forms of LCH to therapy, as seen in our index case, it is of interest to consider if Langerhans cell tumors with NOTCH1 mutations continue to depend on Notch signals for growth, as is the case in TALLs with NOTCH1 mutations [19,20]. Such a relationship would provide a rationale for the use of Notch pathway inhibitors, which are currently under study in T-ALL, in patients with Langerhans cell tumors following T-ALL.

Bone marrow aspirates, and skin and soft tissue biopsies were processed and stained according to standard procedures with antibodies against S-100 (DAKO, Glostrup, Denmark), langerin (Novocastra, Newcastle upon Tyne, UK), CD1a (Beckman-Coulter, Fullerton, CA), or TdT (Supertechs, Bethesda, MD) according to the manufacturers’ directions. Staining for NOTCH1 was performed using an antiserum raised against the transcriptional activation domain as described [21]. Electron microscopy was performed in the Department of Pathology at Children’s Hospital, Boston using standard methodologies. Blood smears were photographed on a Nikon Eclipse E600 microscope with a 1003/0.30 oil immersion lens and an RT Slider SPOT 2.3.1 camera (Diagnostic Instruments, Sterling Heights, MI) using SPOT Advanced software (v. 3.5.9). Biopsy specimens were photographed on an Olympus BX41 microscope (Olympus America, Melville, NY) with Olympus UplanFl 1003/1.30 oil immersion lens and an Olympus Q-color 5 camera and Adobe photoshop elements (v.2.0) software (Adobe Systems, San Jose, CA).

Molecular analyses DNA was isolated from bone marrow aspirate material (T-ALL), skin (LCH), and remission peripheral blood mononuclear cells of the index case using either the Gentra PureGene DNA Isolation kit (Gentra systems, Minneapolis, MN) or the Qiagen QiAmp DNA mini kit (QiagenUSA, Valencia, CA) per the manufacturers’ directions. DNA from the 26 additional biopsies containing tumors of Langerhans cell or histiocytic origin was isolated from formalin-fixed, paraffin-embedded tissue using the Qiagen QiAmp DNA mini kit. Southern blot analysis was performed by hybridizing BglII-digested genomic DNA isolated from a fresh frozen skin biopsy with 32P-labeled probe derived from the junctional region of the TCRg chain gene. TCRg rearrangement studies were also performed on samples of genomic DNA obtained from the bone marrow, skin, and peripheral blood of the index case using the TCRg Gene Clonality Assay (BIOMED-2) from InVivoScribe Technologies (San Diego, CA). Amplified PCR products were fractionated by capillary gel electrophoresis on an ABI 3100 Avant Genetic Analyzer. PCR amplification of the NOTCH1 exons 26, 27, and 34 from genomic DNA, and long-range RT-PCR amplification of NOTCH1 cDNA were performed as described [9]. PCR products amplified from DNA obtained from the 6 biopsies of aggressive LCH were TA-cloned into the vector pCR2.1, and 16 independent clones from each amplification product were sequenced. For the remaining 20 cases, which were either typical LCH or histiocytic lesions, PCR products were sequenced directly. Sequences were analyzed for the presence of mutations using Mutation Surveyor v2.51 software (SoftGenetics LLC, State College, PA).

NOTCH reporter gene assays Reporter assays were performed by co-transfecting U2OS cells with full-length NOTCH1 cDNAs cloned into the plasmid pcDNA3, a NOTCH-responsive luciferase reporter plasmid, and a Renilla luciferase internal control plasmid, as described [9]. Site-directed mutagenesis was performed using the QuikChange Kit (Stratagene).

Acknowledgments The authors thank the patient and her family for participating in this study. Howard Mulhern and Dimity Hall provided expert electron microscopy and molecular diagnostics technical assistance, respectively. References

Materials and Methods Case material Additional tumors of Langerhans cell or histiocytic origin were derived from the formalin-fixed paraffin-embedded tissue files of Children’s Hospital, Boston. Included were eighteen biopsies of self-limited or isolated LCH, two biopsies of Rosai–Dorfman disease, and six biopsies of aggressive, disseminated LCH (Letterer–Siwe disease). These latter six samples were obtained from four patients, two of which had two biopsies each. One case of aggressive LCH/LCS, which followed an antecedent T-ALL and was reported previously [6], was also analyzed.

120

1. Jaffe ES, Stein H, Vardiman JW. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. World Health Organization Classification of Tumours. Lyon: IARC Press; 2001. 2. Nicholson HS, Egeler RM, Nesbit ME. The epidemiology of Langerhans cell histiocytosis. Hematol Oncol Clin North Am 1998;12:379–384. 3. Lieberman PH, Jones CR, Steinman RM, et al. Langerhans cell (eosinophilic) granulomatosis. A clinicopathologic study encompassing 50 years. Am J Surg Pathol 1996;20:519–552. 4. Willman CL, Busque L, Griffith BB, et al. Langerhans’-cell histiocytosis (histiocytosis X)—A clonal proliferative disease. N Engl J Med 1994;331:154–160. 5. Egeler RM, Neglia JP, Arico M, et al. The relation of Langerhans cell histiocytosis to acute leukemia, lymphomas, and other solid tumors. The LCH-Malignancy Study Group of the Histiocyte Society. Hematol Oncol Clin North Am 1998;12:369–378.

American Journal of Hematology DOI 10.1002/ajh

6. Feldman AL, Berthold F, Arceci RJ, et al. Clonal relationship between precursor T-lymphoblastic leukaemia/lymphoma and Langerhans-cell histiocytosis. Lancet Oncol 2005;6:435–437. 7. Anjuere F, del Hoyo GM, Martin P, Ardavin C. Langerhans cells develop from a lymphoid-committed precursor. Blood 2000;96:1633–1637. 8. Maillard I, Fang T, Pear WS. Regulation of lymphoid development, differentiation, and function by the Notch pathway. Annu Rev Immunol 2005;23:945–974. 9. Weng AP, Ferrando AA, Lee W, et al. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science 2004;306:269–271. 10. Bray SJ. Notch signalling: A simple pathway becomes complex. Nat Rev Mol Cell Biol 2006;7:678–689. 11. De Smedt M, Hoebeke I, Reynvoet K, et al. Different thresholds of Notch signaling bias human precursor cells toward B-, NK-, monocytic/dendritic-, or T-cell lineage in thymus microenvironment. Blood 2005;106:3498–3506. 12. Dontje W, Schotte R, Cupedo T, et al. Delta-like1-induced Notch1 signaling regulates the human plasmacytoid dendritic cell versus T-cell lineage decision through control of GATA-3 and Spi-B. Blood 2006;107:2446–2452. 13. Cheng P, Nefedova Y, Miele L, et al. Notch signaling is necessary but not sufficient for differentiation of dendritic cells. Blood 2003;102:3980–3988. 14. Hoshino N, Katayama N, Shibasaki T, et al. A novel role for Notch ligand Delta-1 as a regulator of human Langerhans cell development from blood monocytes. J Leukoc Biol 2005;78:921–929.

American Journal of Hematology DOI 10.1002/ajh

15. Reiter A, Schrappe M, Ludwig WD, et al. Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients. Results and conclusions of the multicenter trial ALL-BFM 86. Blood 1994;84:3122–3133. 16. Silverman LB, Gelber RD, Dalton VK, et al. Improved outcome for children with acute lymphoblastic leukemia: Results of Dana–Farber Consortium Protocol 91-01. Blood 2001;97:1211–1218. 17. Ellisen LW, Bird J, West DC, et al. TAN-1, the human homolog of the Drosophila notch gene, is broken by chromosomal translocations in T lymphoblastic neoplasms. Cell 1991;66:649–661. 18. Lieber T, Kidd S, Alcamo E, et al. Antineurogenic phenotypes induced by truncated Notch proteins indicate a role in signal transduction and may point to a novel function for Notch in nuclei. Genes Dev 1993;7:1949– 1965. 19. Weng AP, Millholland JM, Yashiro-Ohtani Y, et al. c-Myc is an important direct target of Notch1 in T-cell acute lymphoblastic leukemia/lymphoma. Genes Dev 2006;20:2096–2109. 20. Palomero T, Lim WK, Odom DT, et al. NOTCH1 directly regulates c-MYC and activates a feed-forward-loop transcripitonal network promoting leukemic cell growth. Proc Natl Acad Sci USA 2006;103:18261–18266. 21. Pear WS, Aster JC, Scott ML, et al. Exclusive development of T cell neoplasms in mice transplanted with bone marrow expressing activated Notch alleles. J Exp Med 1996;183:2283–2291.

121