Kategorie: Myelodysplastický syndrom
Téma: Myelodysplastic syndromes - Biology
Číslo abstraktu: S675
Autoři: Hetty H. Helsmoortel; Tim Lammens; Farzaneh Ghazavi; Aurélie Caye; Dr. Helene Cavé; prof. MD Christian M. Flotho; Prof. MD Charlotte M. Niemeyer, PhD; prof. MUDr. Jan Starý, DrSc.; Silvia Bresolin, PhD; Riccardo Masetti; MD Anne Uyttebroeck, PhD; MD Marry M. van den Heuvel-Eibrink, PhD; Yves Benoit; Prof. Frank Speleman; Pieter Van Vlierberghe; MD Barbara De Moerloose
Background: Juvenile myelomonocytic leukemia (JMML) is an aggressive clonal myeloid neoplasm of early childhood associated with mutations in Ras pathway genes (PTPN11, KRAS, NRAS, CBL and NF1). Elevated fetal hemoglobin (HbF) levels and monosomy 7 are frequently observed. Stem cell transplantation is the only available curative treatment option but only provides an event-free survival of about 50%.
Aims: Gain insight in the molecular networks involved in JMML pathogenesis based on mRNA, microRNA and long non-coding RNA transcriptome analysis of JMML samples.
Methods: Expression of 27958 mRNA probes and 23042 lncRNA probes was assessed in diagnostic bone marrow or peripheral blood mononuclear cells of 63 JMML patients and 5 healthy donors, using a custom designed Agilent array. In addition, cDNA of 768 microRNAs was pre-amplified and quantified using miRNA specific Taqman probes.
Results: Unsupervised clustering of an initial cohort of 14 patients generated two subgroups with let-7e and RNA-binding protein LIN28B amongst the most significantly differentially expressed genes. In the final cohort, relative higher LIN28B expression was observed in 35 of 63 cases (55.6%) and was defined as the average of the healthy donors plus three standard deviations. Univariable Cox regression showed that logarithmic LIN28Bexpression as a dichotomous variable can predict overall survival (p=0.035, exp(B) = 4.227, CI(95%) = 1.108 – 16.125). Patients with higher LIN28B mRNA levels experience a significant worse overall survival (Kaplan-Meier plot, p=0.022). HbF and platelet count were also significant prognostic factors, as described previously (p=0.023 and 0.027 respectively). There was no association between LIN28B expression and Ras pathway mutation status.
We observed the strongest miRNA anti-correlation between LIN28B and five let-7 family members (d, b, g, e and a), and the second highest positive mRNA correlation between LIN28B and HMGA2. Recently, it was shown that the LIN28B – let-7 – HMGA2 axis determines higher self-renewal of fetal hematopoietic stem cells (Copley, 2013). This indicates that LIN28B confers augmented self-renewal to leukemic hematopoietic stem cells in JMML and – since this is an early childhood disease – this is potentially already initiated during embryogenesis.
JMML patients frequently show elevated HbF levels at diagnosis. A positive correlation was found betweenLIN28B expression and HbF levels (rs=0.64, p<0.001, N = 41). Interestingly, our gene expression profiling data showed that both probes corresponding to HBG1 (encoding the human gamma globin chain) and HBBP1(encoding a lncRNA-affiliated hemoglobin beta pseudogene) were strongly correlated with LIN28B expression in our patient series. This emphasizes the central role for LIN28B in the fetal (leukemic) hematopoietic stem cell system.
Strikingly, patients with monosomy 7 (n=7/56) never displayed increased LIN28B expression (Chi-square p = 0.0017), suggesting the presence of a LIN28B activating transcription factor on chromosome 7. We identifiedMNX1 (HLXB9) as a possible activator of LIN28B based on a very strong correlation and siRNA knockdown.
Summary/Conclusion: LIN28B is overexpressed in more than half of JMML patients and significantly predicts poor outcome. Strong correlation with HbF levels, HBG1 and HMGA2 gene expression suggests thatLIN28B regulates a leukemic hematopoietic stem cell network. Moreover, patients with monosomy 7 seem to be protected against LIN28B overexpression due to the absence of the putative activator MNX1. All together, LIN28Bhas a link with most characteristics of the disease and plays a central role in regulatory networks in JMML.
Keywords: Gene expression profile, JMML, Leukemic stem cell, Prognostic factor
Datum přednesení příspěvku: 14. 6. 2014