Bone Marrow (BM) Microenviroment Factors As Early Markers of Response in Patients with Newly Diagnosed Chronic Phase Chronic Myelogenous Leukemia (CML-CP) Treated with Nilotinib

Introduction

Tyrosine Kinase Inhibitors (TKI) have completely changed the scenario of CML and dramatically improved the outcomes. Thus, early identification of patients expecting poor outcome is crucial to offer alternative TKI regimens or in some selected cases stem cell transplantation before disease progression may occur. The Evaluating Nilotinib Efficacy and Safety in Trial as First-Line Treatment (ENEST1st) is a phase 3b is an open-label study of nilotinib 300 mg twice daily (BID) in adults with newly diagnosed BCR-ABL positive CP-CML. Aim of the ENEST1st sub-study N10 was to investigate BM microenvironment markers that regulate leukemic stem cells in the bone marrow (BM) niche of Nilotinib-treated patients.

Methods

The study enrolled patients in 21 Italian ENEST1st participating centers. Response was based on ELN recommendations (Baccarani M, et al. Blood 2013 122:872-884). In an interim analysis, molecular and cytogenetic response by 24 months was assessed. Mononuclear cells were collected from BM and PB samples at the screening visit (V0) and after 3 months of treatment (V4). RT-qPCR for the expression of 10 genes (ARF, KIT, CXCR4, FLT3, LIF, NANOg, PML, PRAME, SET and TIE), involved in the stemness and hematopoietic stem cells survival signaling regulation was conducted. RT-qPCR data were normalized by the expression of GUS mRNA (normalized copy number, NCN). Plasma samples were collected at different time points from both BM or PB samples. Concentrations of 20 different analytes, including IL-1a, IL-3, M-CSF, SCF, SDF1-a, TRAIL, HGF, PDGF-bb, IL1b, IL-6, IL-7, IL-8, IL-10, IL-12, IL-15, G-CSF, GM-CSF, MIP-1a, TNF-a, and VEGF, were simultaneously evaluated using commercially available multiplex bead-based sandwich immunoassay kits.

Results

33 out of 37 patients enrolled were available for an interim molecular analysis at 24 months: an optimal response was achieved in 25 patients, a warning response in 5 patients and a failure response in 3 patients. We observed a significant correlation between the expression of two genes involved in the regulation of stem cell pluripotency (NANOg) or cytokine signaling (SET) and patient outcome. Indeed, NANOg and SET mRNA were significantly down-regulated in PB samples at diagnosis of patients with optimal response compared to patients with warning/failure response (NANOg mRNA: 0.3±0.25 NCN vs 0.6±0.7 NCN, respectively; p=0.05; SET mRNA: 0.2±0.3 NCN vs 2.3±4.2 NCN, respectively; p=0.03). We also investigated the plasma level of several factors involved in the hematopoietic stem cells (HSCs). Some of these markers showed a significant correlation with patient's outcome when evaluated at diagnosis in either PB or BM samples. Indeed, high level of IL12 (in the BM samples), or HGF, mCSF and SCF (in the PB samples) were associated to a worst prognosis markers, since significantly correlating with no MMR@12months (IL12, p=0.03), intermediate/high Socal score (mCSF, p=0.03; SCF, p=0.03), no reduction of MMR below to 1 at 3 month (SCF, p=0.04) or warning/failure response to Nilotinib treatment (HGF, p=0.03; SCF, p=0.04). Indeed, we find a lower levels of PDGFb, SDF1, TNFa, TRAIL (in the BM samples), and HGF, SDF1, TRAIL (in the PB samples) in those patients with intermediate/high Hasford or Sokal score (PDGFb, p=0.0007; SDF1, p=0.02), warning/failure response to Nilotinib treatment (HGF, p=0.03) or lacking of MMR4.0 (SDF1, p=0.01; TNFa, p=0.02; TRAIL, p=0.05).

Conclusion/Summary

Taken together, our results suggest that the expression analysis of genes involved in cell pluripotency (NANOg) and/or cell signaling (SET) at baseline, may indicate early achievement of deep molecular response in shown CML-CP patients treated with nilotinib. In addition, in patients with optimal response to Nilotinib, high concentration of SDF-1, TRAIL (inversely correlated with BCR-ABL, and associated to an higher susceptibility to apoptosis in the leukemic blasts) were observed as well as BM TNF (cell-extrinsic and potent endogenous suppressor of HSC activity). A lower concentration of several factors associated to hematopoietic progenitor cell growth and survival (including HGF, SCF and IL12) were observed compared to patients failing to achieve response to Nilotinib. These data strongly suggest that stromal microenvironment supports the viability of BCR-ABL cells in BM niches through direct feeding, or environment releasing of survival factors.