Background: Proteins such as cytokines and growth factors are important in cellular communication. They are also known to mediate messages between cancer cells and the immune system. CML is known to be an immunogenic cancer and function of the immune system has been linked to good treatment responses and successful treatment discontinuation. To understand the crosstalk between leukemic cells and the microenvironment we analyzed the plasma cytokine profile of CML patients with multiplex proteomic array at diagnosis and during treatment.

Methods: Peripheral blood (PB) samples were collected from 33 CML patients at diagnosis and during treatment (3 and 12 months after treatment initiation). Patients received either bosutinib, imatinib or dasatinib as first-line therapy. The dasatinib treated patients received low-dose IFN-a after 3 months of dasatinib monotherapy. In addition, PB of 17 healthy volunteers was collected. The plasma cytokine profile was studied with multiplexed cytokine and growth factor panel that consists of 92 proteins (Proseek Multiplex Inflammation I96×96, Olink). Wilcox test was used to determine significant differences between patient and healthy volunteers, and p<0.05 was used as a cut-off value. Ingenuity pathway analysis (IPA) was performed to characterize aberrant disease associated pathways in CML.

Results: The analysis of the cytokine panel showed that 43 proteins were upregulated in CML patients at diagnosis compared to the healthy. The IPA demonstrated that these cytokines and growth factors were related to the granulocyte adhesion and diapedesis and Th1 pathway. Top upstream regulators of these cytokines were NFKB1, RELA, STAT3, NFKBIA and RELB. The most pronounced differences were observed in the following proteins: Interleukin 10 receptor, beta subunit (IL-10RB), tumor necrosis factor receptor superfamily member 9 (TNFRSF9), CD5, Chemokine (C-C motif) ligand 23 (CCL23), Caspase-8 (CASP-8), CD244, adenosine deaminase (ADA), tumor growth factor a (TGFA), hepatocyte growth factor (HGF), tumor necrosis factor superfamily member 14 (TNFSF14), S100 calcium-binding protein A12 (EN-RAGE), urokinase-type plasminogen activator (uPA), vascular endothelial growth factor A (VEGFA) and oncostatin M (OSM). The levels of most proteins normalized to the levels of the healthy controls already after 3 months of treatment initiation regardless of the tyrosine kinase inhibitor (TKI) used suggesting that these proteins were related to the leukemic growth. Only the levels of IL-10RB and TNFRSF9 remained high compared to the healthy even during treatment.

Interestingly, some of the protein levels at diagnosis correlated with BCR-ABL1 values (IS) at 3 months. Proteins such as Flt3L (r=-0.45, p=0.01), MCP-4 (r=-0.42, p=0.02), AXIN1 (r=-0.39, p=0.03), CXCL10 (r=-0.38, p=0.04) and CCL19 (r=-0.44, p=0.01) had a negative correlation and IL-4 (r=0.40, p=0.03) had a positive correlation, but only the Flt3L (r=-0.44, p=0.01) correlated also with the BCR-ABL1 values at 12 months. Patients who were treated with dasatinib therapy had increased levels of Flt3L at 3 months (p=0.008) and the level further increased at 12 months after the addition of IFN-a compared to the patients who received only imatinib or bosutinib (p=0.01).

Conclusions: Multiplex plasma proteomic analysis provides tools for biomarker discovery and treatment monitoring. During first 3 months of TKI therapy most of the plasma cytokines and growth factors which were upregulated at the time of diagnosis returned to the levels of healthy controls. Flt3L levels at the diagnosis correlated negatively with the treatment response during the TKI therapy. As Flt3L is known to increase stem cell proliferation, low levels of Flt3L may lead to more quiescent stem cells and possible therapy resistance. During dasatinib and IFN-a treatment Flt3L levels increased above the levels noted in healthy controls.


Mustjoki: Ariad: Research Funding; Pfizer: Honoraria, Research Funding; Novartis: Honoraria, Research Funding; BMS: Honoraria, Research Funding; Celgene: Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.