P653: CML LONG-TERM LEUKEMIC STEM CELL PERSISTENCE IS INFLUENCED BY MYELOFIBROSIS AT DIAGNOSIS

HemaSphere, vol. 7, pp. e8231722

Abstract

Topic: 7. Chronic myeloid leukemia - Biology & Translational Research Background: Chronic myeloid leukemia (CML) is caused by the translocation t(9;22) acquired in the hematopoietic stem cell (HSC) compartment, leading to constitutively activated BCR-ABL1 tyrosine kinase. The oncoprotein induces expansion of the leukemic stem cell (LSC) pool but also drives differentiation. The malignant cells impact on the niche through e.g. inflammatory factors, sometimes resulting in myelofibrosis (MF), which is an adverse prognostic marker for tyrosine kinase inhibitor (TKI) response. Aims: Since LSC persistence is likewise supported by inflammatory processes, we studied the outcome of MF at diagnosis on treatment-free remission (TFR), in a small patient cohort. Moreover, we assessed the differentiation potential of CML-mesenchymal stromal cells (MSCs) vs healthy donor (HD) and the effect of fibrotic remodeling on their expression profile, aiming to identify specific factors that could be responsible for TFR failure. Methods: Clinical parameters of CML patients were correlated with MF or IGFBP2 (Insulin-Like Growth Factor Binding Protein 2) serum levels, measured via ELISA in peripheral blood (pB) serum. HD and CML MSCs were subjected to trilineage differentiation assays and RNAseq. Co-culture experiments were performed using HS-5 stromal and K562 CML cells in a transwell and submitted to qRT-PCR. Results: In our cohort, MF was evident in 42% (25 out of 60) of patients at diagnosis. Interestingly, our TKI-discontinuation cohort revealed that 4 out of 4 patients who failed TFR, displayed MF at diagnosis whereas MF was evident in only 2 out of 8 in patients after 1 year of successful TFR. In addition, CML patients with MF grade 2 had 3-times higher pB blast cell counts compared to CML patients with no MF (p<0.01). Primary HD vs CML MSC differentiation potential showed increased osteogenic (fold change (FC) 23.6 vs. FC 74.0) and adipogenic (FC 1.5 vs. 3.7) differentiation in CML samples, as suggested previously. This was confirmed in the RNAseq analysis, which revealed an up-regulation of osteogenic pathways, that had also been assigned to LSC support, and enhanced expression of extracellular matrix binding proteins. Moreover, we observed an increase of IGFBP2 expression in CML samples, also by qRT-PCR (p<0.05). This was confirmed in the serum of HD and CML patients at diagnosis (HD: 78 ± 58 ng/ml, CML: 234 ± 104 ng/ml serum IGFBP2, p<0.001). Higher IGFBP2 serum levels in CML patients correlate with a higher pB blast count (0% pB blasts: 144 ± 82 ng/ml, 1-12% pB blasts: 256 ± 117 ng/ml serum IGFBP2, p<0.01) and with non-optimal response (nOR) after 12 months (M) of treatment (OR at 12M: 172.6 ± 62 ng/ml, nOR at 12M: 337 ± 45 ng/ml IGFBP2 at diagnosis or 3M, p<0.01). In HS-5 and K562 coculture experiments, HS-5 stromal cells expressed significantly more IGFBP2, compared to the HS-5 monoculture (p<0.0001), suggesting that the hematopoietic cells might redirect the niche to up-regulate IGFBP2. In a publicly available microarray dataset (GEO number GSE47927), IGFBP2 was also highly expressed in CML-HSCs, which further increased during disease progression, suggesting that in vivo, elevated IGFBP2 levels are provided by the niche but also by the LSCs themselves. Summary/Conclusion: Using this small TKI-discontinuation patient cohort, our data suggests that MF upon diagnosis creates a niche that allows for long-term LSC persistence, thus inducing TFR failure. One of the factors possibly influencing this niche is IGFBP2, whose presence in CML patients seems to correlate with TKI response and potentially CML disease progression. Keywords: Chronic myeloid leukemia, Stem cell niche, Myelofibrosis

Authors 12

1. Henrike Jacobi
   Affiliation as printed

   Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Aachen, Germany

2. Margherita Vieri
   Affiliation as printed

   Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Aachen, Germany

3. Carolina Yaeko Namasu
   Affiliation as printed

   Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Aachen, Germany

4. Laura Flüter
   Affiliation as printed

   Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Aachen, Germany

5. Ivan G. Costa Aachen Institute for Computational Genomics

   RWTH Aachen University

   Affiliation as printed

   Institute for Computational Genomics, Aachen, Germany

6. Tiago Maié Aachen Institute for Computational Genomics

   RWTH Aachen University

   Affiliation as printed

   Institute for Computational Genomics, Aachen, Germany

7. Nicolas Chatain
   Affiliation as printed

   Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Aachen, Germany

8. Fabian Beier
   Affiliation as printed

   Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Aachen, Germany

9. Michael Huber Aachen Institute of Biochemistry and Molecular Immunology

   RWTH Aachen University

   Affiliation as printed

   Institute of Biochemistry and Molecular Immunology, Aachen, Germany

10. Martina Margrit Crysandt
    Affiliation as printed

    Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Aachen, Germany

11. Tim H. Brümmendorf
    Affiliation as printed

    Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Aachen, Germany

12. Mirle Schemionek
    Affiliation as printed

    Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Aachen, Germany