The Lifespan of an MLL-Rearranged Therapy-Related Paediatric AML

LETTER TO THE EDITOR From initiation to eradication: the lifespan of an MLL-rearranged therapy-related paediatric AML

Bone Marrow Transplantation (2015) 50, 1382–1384; doi:10.1038/ amplified by RT-PCR. DCP1A is predominantly cytoplasmic as part bmt.2015.155; published online 6 July 2015 of a protein complex involved in degradation of mRNAs.8 Upon stimulation by TGFB1 and interaction with SMAD4, it becomes nuclear contributing to the transactivation of TGF-α target genes.9 Therapy-related AML (tAML) is one of the most prevalent DCP1A is a novel MLL fusion partner in AML. The in vitro secondary malignant neoplasms in paediatric cancer survivors. transforming capacity of MLL-DCP1A was weaker than the more Estimates of incidence and association with causative chemo- common MLL-ENL fusion but it could nevertheless be clearly therapeutic agents are mainly based on epidemiological data. verified by colony formation in retroviral transduction-replating 10 Individual courses revealing the exact timing of the first clone assays (Figure 1d). appearance and its development under treatment are limited The cumulative sampling of bone marrow during preceding owing to the overall low incidence, which does not justify neuroblastoma treatment offered the unique possibility to repeated bone marrow sampling in order to detect the occurrence quantitatively backtrack the origin of the tAML by quantitative of potentially leukemic cells. real-time PCR using the genomic MLL-DCP1A breakpoint as a Here we report a unique case of tAML arising after neuro- molecular marker. A low-grade presence of the leukemic clone − 4 blastoma treatment that was tracked by the presence of a (10 in mononuclear cells) was already detectable in bone MLL-DCP1A translocation. Kinetics of the malignant clone marrow at day 67 after the first two N8 cycles consisting implied disease initiation by the first etoposide doses, allowed a of topotecan 1 mg/m2 per day (day 1–7), cyclophosphamide quantitative follow-up during chemotherapy and suggested clone 100 mg/m2 per day (day 1–7) and etoposide 100 mg/m2 per day eradication by a GvL reaction. (day 8–10). Topotecan is a topoisomerase I inhibitor but an A 3-year 9-months-old girl presented with N-Myc-, del1p36+, increased carcinogenic risk for malignancies with chromosomal MIBG+ disseminated neuroblastoma and received treatment on aberrations after topotecan treatment has not yet been observed. the NB2004 trial protocol (NCT00410631). Chemotherapy was Secondary haematological malignancies induced by alkylating initiated with two N8 cycles (topotecan, cyclophosphamide and agents, such as cyclophosphamide, typically occur with a latency etoposide) as the experimental arm of a trial assessing topotecan of 5–7 years and show cytogenetic abnormalities involving for treatment of high-risk neuroblastoma1 followed by three chromosomes 5 and 7. Topoisomerase-II inhibitors like etoposides N5 (cisplatin, etoposide and vindesine), two N6 (vincristine, and anthracyclines induce secondary leukaemia with a short dacarbacin, ifosfamide and doxorubicine) and a third N8 cycle, latency period of 2–3 years that frequently carry translocations thoracotomic tumour resection, 131-I-MIBG treatment and involving 11q and 21q suggesting a causative involvement myeloablative megatherapy (melphalan, etoposide and carbo- in the current case. Breakpoint characteristics, microhomology platin) and subsequent autologous stem cell transplantation and duplication of the GATCA nucleotides in the reciprocal (ASCT). Retinoid acid (RA)-maintenance therapy was started DCP1A-MLL fusion document a NHEJ repair signature, consistent 63 days thereafter.2 Restaging for persistent neuroblastoma was with a topoisomerase II-related rearrangement. The MLL-DCP1A negative, but progressive neutropenia developed despite RA-dose breakpoint in this particular case occurred in position 1181 of the adjustment (Figure 1a). MLL breakpoint cluster region, 5 Kb distant from the localization of A diagnostic bone marrow aspiration 216 days after ASCT a topoisomerase II recognition site, a hot-spot of tAML.11 revealed tAML FAB M4 with the expression of CD33, CD38, CD14, CD34+ cell selection from mobilized peripheral stem cells was CD15, CD45, CXCR4 and cytoplasmic MPO. Treatment protocol applied as purging strategy for residual neuroblastoma cells AML-BFM 2004 was initiated with prolonged bone marrow aplasia according to the treatment protocol. Unfortunately, MLL-DCP1 A observed after the second L-DNR/FLAG.3 Unrelated bone marrow myeloid cells were also enriched by this procedure (Figure 1a). transplantation was initiated by conditioning with thiotepa, Despite intensive chemotherapy treatment of neuroblastoma the fludarabine, treosulfan and antithymocyte globulin. GvHD pro- malignant clone expanded and it was only partially reduced after phylaxis with cyclosporin A (CSA) started at day-1. In addition, MIBG and high-dose chemotherapy preceding ASCT. In contrast, 10 mg/m2 MTX were given on days +1, +3 and +6. Grade II skin cells experimentally transformed by MLL-DCP1A alone responded GvHD and grade I gut GvHD developed around day 32 after to all-trans RA treatment with proliferation arrest and terminal allogeneic transplantation, but responded within days to short- differentiation (Figures 1e and f) suggesting the accumulation of term systemic steroids. Few other individuals with MLL-rearranged additional mutations beyond the MLL translocation. leukaemia after neuroblastoma treatment have been reported in During routine immune suppression by CSA after allogeneic − − detail; outcome was predominantly unfavourable in these transplantation, MRD was detectable in the range of 10 3 to 10 4 cases.4–6 where normal chimaerism analysis by STR-repeat PCR could not The occurrence of AML within o2 years after treatment detect this developing molecular relapse because of the with etoposide was highly suggestive of the presence of methodological-sensitivity limit. Facing an imminent molecular MLL-rearrangements. Cytogenetic analysis (46,XX,t(3;11)(p21;q23)) relapse it was decided to stop CSA ahead of schedule to allow a confirmed an 11q23 aberration with translocation to 3p21 and more intense GvL effect. Indeed, tapering and eventually interphase split-FISH demonstrated involvement of MLL discontinuation of CSA resulted in gradual MRD reduction leading (Figure 1b). Genomic LDI-PCR7 identified the fusion in intron 6 to a permanent molecular remission for 5 years (Figure 1a). of DCP1 A (DCP1 decapping enzyme homolog A; Figure 1c). In This example underscores the benefit of highly sensitive frame MLL-DCP1A and DCP1A-MLL fusion transcripts were quantitative PCR techniques using the unique fusion gene as Letter to the Editor 1383

a N8 N8 N5 N6 N5 N6 Surgery N8 N5 MIPG-J131 DNX/FLAG DNX/FLAG MUD-SCT 100 GCSF PDN 4000 Retinoic A. CSA MLL-DCP1A+/MNC 10-1 PMN/µl 10-2 3000

10-3

10-4 CD34 pos. 2000 Unpurified 10-5 1000 n.q. Flow through

neg. PBSC-MACS 0 0 365 d 730 d 7 y Time after initial diagnosis

b c

11 Der(11) 123 5’MLL

Der(3)

91011 3’MLL

dfe MLL-DCP1 MLL-ENL neo 96h treatment Gr-1 timecourse 0d 100.0

90.0 Gr-1 80.0 c-kit 100 0d = 70.0 2d MLL-ENL 4d

= (%) Proliferation 60.0 7d MLL-DCP1 50.0 0.01 0.1 (µM ATRA) 1 Gr-1 Relative colony # (%) Relative colony 10 0 Figure 1. Clinical course, genetic and functional characterization of the MLL-DCP1A rearranged tAML. (a) Quantification of MLL-DCP1A-positive cells during neuroblastoma and subsequent AML treatment (black dots, left y-axis). Therapeutic elements are given in the top line. Grey dots show the neutrophil count during RA-maintenance therapy (right y-axis). Reduced doses of RA and CSA are indicated by the lighter colouring. The effect of CD34 selection on the purity of autologous stem cell product is indicated by symbols at day 140. (b) Karyogram demonstrating the presence of a translocation t(3;11). Involvement of the MLL gene is confirmed by a split signal of the green 5ʹ MLL and red 3ʹ MLL probe in metaphase FISH. (c) Alignment of the genomic MLL-DCP1A (der11) and reciprocal DCP1A-MLL (der3) fusion sequence; microhomologies are underlined. (d) MLL-DCP1A transforms primary cells in a colony-formation assay. Primary murine hematopoietic cells were retrovirally transduced with MLL-DCP1A, MLL-ENL or empty vector (neo) as control. Cells were plated for three rounds in methylcellulose with non- transformed cells exhausting their proliferative capacity, thus ablating colony formation. The left panel shows stained colonies arising in the third round of replating. A numerical evaluation is depicted in the bar chart. (e) Response of MLL-DCP1A transformed cells towards all-trans RA. Cells transformed by retroviral transduction of MLL-DCP1A were cultivated in the presence of increasing amounts of ATRA. Proliferation was determined by standard MTT assay in triplicates. Average and s.d. of relative proliferation rates are shown. (f) MLL-DCP1A transformed cells differentiate in the presence of ATRA. MLL-DCP1A transduced cells were cultivated in the presence of 1 μM ATRA and the differentiation- specific surface marker Gr-1 as surrogate measurement for cellular maturation was determined at the indicated time points. The inset shows a double c-kit/Gr-1 staining indicating that MLL-DCP1A cells are c-kit negative. molecular marker for MRD detection.12 Alternatively, routine precise monitoring of therapy effects after application of novel methods are limited by their sensitivity (morphology and FISH) agents and immune interventions. or their ability to discriminate definitively between blasts and benign progenitor cells in regenerating bone marrow (flow cytometry and gene-expression profiling). With the development of patient-specific MRD assays, clinical management and CONFLICT OF INTEREST biological understanding of high-risk leukaemia may benefit from The authors declare no conflict of interest.

© 2015 Macmillan Publishers Limited Bone Marrow Transplantation (2015) 1382 – 1384 Letter to the Editor 1384 1,6 2,6 1 3 4 M Krumbholz , J Bradtke , D Stachel , O Peters , B Hero , 4 Robinson BW, Cheung NK, Kolaris CP, Jhanwar SC, Choi JK, Osheroff N et al. W Holter1,7, R Slany5 and M Metzler1 Prospective tracing of MLL-FRYL clone with low MEIS1 expression from 1Department of Paediatric Oncology, University Hospital Erlangen, emergence during neuroblastoma treatment to diagnosis of myelodysplastic Erlangen, Germany; syndrome. Blood 2008; 111:3802–3812. 2Oncogenetic Laboratory, Molecular Pathology, Institute of 5 Pegram LD, Megonigal MD, Lange BJ, Nowell PC, Rowley JD, Rappaport EF et al. Pathology, University Hospital Giessen and Marburg, t(3;11) translocation in treatment-related acute myeloid leukemia fuses MLL with the GMPS (guanosine 5' monophosphate synthetase) gene. Blood 2000; 96: Giessen, Germany; – 3 4360 4362. Department of Paediatics, St. Hedwig Hospital, 6 Megonigal MD, Cheung NK, Rappaport EF, Nowell PC, Wilson RB, Jones DH et al. Regensburg, Germany; Detection of leukemia-associated MLL-GAS7 translocation early during 4 Department of Paediatric Oncology, University of Cologne, chemotherapy with DNA topoisomerase II inhibitors. Proc Natl Acad Sci USA 2000; Cologne, Germany and 97:2814–2819. 5Department of Genetics, University of Erlangen, Erlangen, Germany 7 Meyer C, Schneider B, Reichel M, Angermueller S, Strehl S, Schnittger S et al. E-mail: [email protected] Diagnostic tool for the identification of MLL rearrangements including unknown 6These authors contributed equally to this work. partner genes. Proc Natl Acad Sci USA 2005; 102: 449–454. fi 7Current address: Department of Paediatrics, St Anna Children's 8 Lykke-Andersen J. Identi cation of a human decapping complex associated with hUpf proteins in nonsense-mediated decay. Mol Cell Biol 2002; 22: Hospital, Medical University, Kinderspitalgasse 6, 8114–8121. Vienna 1090, Austria 9 Bai RY, Koester C, Ouyang T, Hahn SA, Hammerschmidt M, Peschel C et al. SMIF, a Smad4-interacting protein that functions as a co-activator in TGFbeta signalling. Nat Cell Biol 2002; 4:181–190. REFERENCES 10 Strehl S, Borkhardt A, Slany R, Fuchs UE, Konig M, Haas OA. The human LASP1 1 Simon T, Langler A, Harnischmacher U, Fruhwald MC, Jorch N, Claviez A et al. gene is fused to MLL in an acute myeloid leukemia with t(11;17)(q23;q21). Topotecan, cyclophosphamide, and etoposide (TCE) in the treatment of high-risk Oncogene 2003; 22:157–160. neuroblastoma. Results of a phase-II trial. J Cancer Res Clin Oncol 2007; 133: 11 Whitmarsh RJ, Saginario C, Zhuo Y, Hilgenfeld E, Rappaport EF, Megonigal MD 653–661. et al. Reciprocal DNA topoisomerase II cleavage events at 5'-TATTA-3' sequences 2 Simon T, Berthold F, Borkhardt A, Kremens B, De Carolis B, Hero B. Treatment and in MLL and AF-9 create homologous single-stranded overhangs that anneal to outcomes of patients with relapsed, high-risk neuroblastoma: results of form der(11) and der(9) genomic breakpoint junctions in treatment-related AML German trials. Pediatr Blood Cancer 2011; 56:578–583. without further processing. Oncogene 2003; 22: 8448–8459. 3 Kaspers GJ, Zimmermann M, Reinhardt D, Gibson BE, Tamminga RY, Aleinikova O 12 Krumbholz M, Jung R, Bradtke J, Reinhardt D, Stachel D, Metzler M. et al. Improved outcome in pediatric relapsed acute myeloid leukemia: results Response monitoring of infant acute myeloid leukemia treatment by quantifica- of a randomized trial on liposomal daunorubicin by the International BFM tion of the tumor specific MLL-FNBP1 fusion gene. Leuk Lymphoma 2015; 56: Study Group. J Clin Oncol 2013; 31:599–607. 793–796.