Background: Mutations in the BCR-ABL1 kinase domain are a well-documented mechanism of resistance to tyrosine kinase inhibitors (TKIs), but less is known about primary resistance independent of BCR-ABL1 kinase activity. We reported a gene expression classifier of TKI-naïve CD34+ cells from chronic phase chronic myeloid leukemia (CP-CML) patients that predicts cytogenetic response to imatinib (McWeeney et al. Blood 2010). The expression signature associated with primary cytogenetic failure showed overlap with previously reported signatures of blast phase CML (BP-CML), suggesting that primary TKI resistance and advanced disease are biologically similar.

Results: To identify critical genes involved in primary TKI resistance, we performed principal component analysis on the expression signature and identified the hematopoietic cell cycle regulator, MS4A3, as a key factor within this classifier. Importantly, low MS4A3 expression not only correlated with primary TKI resistance, but also with shorter overall survival (p<0.01), prompting us to study the role of MS4A3 in more detail. Since the signatures for primary cytogenetic failure and BP-CML overlapped, we next assessed whether MS4A3 expression was downregulated in BP-CML versus CP-CML. qRT-PCR confirmed that MS4A3 mRNA levels are markedly reduced (by >92%) in CD34+ cells from BP-CML patients (n=17; p<0.01) versus CP-CML patients (n=23) and normal controls (n=3). There was no significant difference between normal and chronic phase CD34+ cells, suggesting that MS4A3 is specifically downregulated upon CML blastic transformation. Microarray data from Oehler et al. (unpublished observations) and Zheng et al. (Leukemia 2006) also showed that MS4A3 mRNA levels are reduced in CD34+ cells from BP-CML patients (n=16; p<0.001) compared to CP-CML patients (n=19). Immunoblot analyses confirmed that MS4A3 protein was detectable in CD34+ cells from newly diagnosed CP-CML patients, but not in samples from patients with primary resistance who failed multiple TKIs but lack clinically relevant BCR-ABL1 mutations. Finally, in CP-CML samples from newly diagnosed patients, MS4A3 mRNA levels are 12-fold reduced in primitive CD34+38- stem cells compared to more committed CD34+38+ progenitor cells (n=3; p<0.05), which are innately resistant to BCR-ABL1 inhibition. Consistent with BCR-ABL1 kinase independence, MS4A3 mRNA and protein levels were unaffected by ex vivo TKI treatment.

MS4A3 expression is also low in BP-CML cell lines, including K562, KYO-1, BV-173, KCL-22, and KU-812, with the notable exception of LAMA-84 cells. Thus, to understand the functional role of MS4A3 for TKI resistance, we introduced a doxycycline-inducible shRNA targeting MS4A3 (shMS4A3) into LAMA-84 cells. qRT-PCR confirmed 50-90% MS4A3 knockdown in the presence of doxycycline (0.1 µg/mL). Consistent with its role as a tumor suppressor, MTS assays revealed that MS4A3 knockdown increased the imatinib IC50 (n=3; p<0.05) and abolished the effects of imatinib in colony formation assays (n=2; p<0.05). We next assessed the effects of shMS4A3 in CD34+ cells from newly diagnosed CP-CML patients. qRT-PCR confirmed ~50% MS4A3 knockdown in primary cells. Despite the incomplete knockdown, shMS4A3 enhanced colony formation in the presence of imatinib in a dose-dependent manner (n=4; p<0.00001), and abolished imatinib-induced apoptosis (n=3; p<0.001). We also assessed the effects of ectopic MS4A3 expression in CD34+ cells from advanced phase CML (AP-CML) patients. qRT-PCR confirmed >2-fold upregulation of MS4A3. As expected, ectopic MS4A3 reduced colony formation by 55% in AP-CML (n=2; p<0.01), and enhanced sensitivity to imatinib-induced apoptosis by 35% (n=2; p<0.01). Neither shMS4A3 nor ectopic MS4A3 had any effect on survival of normal cord blood CD34+ cells (n=2).

Conclusion: Our results suggest that MS4A3 is a tumor suppressor protein in CML that governs TKI responsiveness and is regulated in a BCR-ABL1 kinase-independent manner. MS4A3 loss confers TKI resistance to CP-CML patients destined to exhibit primary cytogenetic failure, and in BP-CML patients with refractory resistance. MS4A3 may also contribute to the innate resistance of primitive CML stem cells. Studies to identify the mechanism of MS4A3 downregulation in TKI resistance and how its loss biochemically impairs TKI response is currently underway and will be reported.


Agarwal:CTI BioPharma: Research Funding. Deininger:BMS: Other: Consulting & Advisory Role, Research Funding; Novartis: Other: Consulting or Advisory Role, Research Funding; Celgene: Research Funding; Genzyme: Research Funding; Gilead: Research Funding; ARIAD Pharmaceutical Inc.: Other: Consulting or Advisory Role; Incyte: Other: Consulting or Advisory Role; Pfizer: Other: Consulting or Advisory Role.

Author notes


Asterisk with author names denotes non-ASH members.