MODELING RESISTANCE TO TYROSINE KINASE INHIBITORS IN TEL/ABL+ ACUTE LYMPHOBLASTIC LEUKEMIA

ABSSUB-5112

Background: TEL/ABL+ acute lymphoblastic leukemia (ALL) represents a rare leukemia subtype with unfavorable prognosis. The activity of fusion kinase can be inhibited by tyrosine kinase inhibitors (TKI), which might be beneficial in the therapy of this ALL.

Aims: We aimed to study potential mechanism of TKI resistance in TEL/ABL+ ALL.

Results: Via long-term exposure of parental sensitive TEL/ABL+ cell line to slowly increasing doses of imatinib mesylate (IM) we have established panTKI-resistant daughter cell line. To control for the effect of long-term culturing, daughter sensitive cell line was established in parallel to resistant one and always used together with parental sensitive cell line in analyses comparing cells with resistant and sensitive phenotype.

Using deep sequencing, FISH, qPCR and Western blot we excluded the most common mechanisms of IM resistance described in BCR/ABL+ ALL (kinase domain mutations, genomic amplification or enhanced fusion gene expression) as the cause of resistance in our model.

To study genomic background of resistance, we performed  genomic profiling and whole exome sequencing. First round of profiling on high density SNP arrays did not reveal any changes acquired by resistant cells compared to sensitive cells. Second round of profiling on SNP array with different coverage revealed that the resistant cells acquired a 60 kb intragenic deletion in KDM6A gene encoding lysine-specific histone demethylase. This deletion is predicted to result in either the expression of aberrant KDM6A or in loss of function and we are currently studying its impact on transcript and protein level.

Whole exome sequencing identified 25 non-synonymous nucleotide variants (SNVs) within gene coding regions that were gained by resistant cells and not detected in sensitive cells. Twelve SNVs affect genes expressed in the studied cell line and annotated to various biological processes including intracellular signaling (e.g. G protein GNB1 or JAK/STAT signaling component STAM2). These genes have not been associated with TKI resistance before and their potential contribution to resistance needs to be further clarified.

To study gene expression changes associated with resistance we have performed gene expression profiling on microarrays and identified a set of genes differentially expressed between resistant and sensitive cells. The expression data for the top up/down-regulated genes were confirmed by qPCR on a set of multiple independently harvested and processed samples. The expression of several candidates was further assessed with regard to the presence of IM in growth media and 2 different expression patterns were observed. First group of genes, represented by SORBS2, was differentially expressed in resistant cells compared to sensitive cells irrespective of IM presence. In contrary, the change in expression of second subgroup of genes (involving BCL6 and SOCS2) was induced in resistant but not sensitive cells by high but still completely tolerated IM levels. Of interest, upregulation of BCL6 was recently described as a cause of IM resistance in BCR/ABL+ cells and its inhibition was shown to restore IM response (Duy, Nature, 2011).

Summary/Conclusion: To conclude, we have established a new model of TKI-resistance that is not driven by the most common resistance mechanisms described so far. Using genomic and expression profiling we have identified several genetic alterations and gene expression changes associated with resistant phenotype that may contribute to resistance mechanism and will be further studied.

Support: IGA MZ NT/13170-4; UNCE 204012; RVO-FNM64203; RVO-VFN64165;GAUK 694414

Keywords: ALL, Imatinib resistance

http://www.ehaweb.org/