Poster Session I: Acute lymphoblastic leukemia - Biology & translational research
De novo or acquired resistance to glucocorticoids, the backbone regimens for remission-induction in pediatric B-precursor acute lymphoblastic leukemia (B-ALL), poses a significant therapeutic hurdle and is a strong predictor of inferior survival. With the mechanisms being insufficiently understood, strategies for restoring glucocorticoid sensitivity have been limited. We previously reported that CD9 negativity was associated with poorer prednisone response in a cohort of pediatric B-ALL patients (Leung et al, EHA, 2017), but the functional linkage remains undefined.
(1) To establish the glucocorticoid sensitivity pattern of B-ALL cell lines and its relationship with CD9 expression; (2) to investigate the definitive role of CD9 in regulation of glucocorticoid response using gain- and loss-of-function approaches; and (3) to determine if CD9 could affect the expression of glucocorticoid receptor.
An in vitro drug sensitivity profiling, comprising 7 standard chemotherapeutic agents (6-mercaptopurine, cytarabine, daunorubicin, vincristine, methotrexate, prednisolone and dexamethasone), was performed on 4 CD9+ (697, BV-173, RS4;11, SUP-B15) and 3 CD9− (REH, SEM, KOPN-8) B-ALL cell lines. Responses to prednisolone and dexamethasone were re-assessed after lentiviral transduction of CD9 overexpression or knockout vectors into B-ALL cell lines. The expression of glucocorticoid receptor NR3C1 in B-ALL cell lines was measured by qPCR and Western blot.
Unsupervised clustering analysis of drug profiling data revealed a unique pattern of glucocorticoid response, which was manifested by the preferential resistance of CD9− cells to prednisolone and dexamethasone, but not to 6-mercaptopurine, cytarabine, daunorubicin, vincristine and methotrexate. Overexpression of CD9 in the CD9− SEM cell line significantly enhanced its sensitivity to prednisolone (IC50: 6,231 vs. 55,346 nM; P = 0.017) and dexamethasone (IC50: 351.5 vs. 990.6 nM; P = 0.005), when compared to cells transduced with control vector. In contrast, CD9 knockout in the CD9+ 697 cell line significantly decreased its sensitivity to prednisolone (IC50: 321.1 vs. 242.7 nM; P = 0.018) and dexamethasone (IC50: 16.2 vs. 13.8 nM; P = 0.032). Such difference was not observed for other chemotherapeutic agents, indicating the specific influence of CD9 on glucocorticoid response. The glucocorticoid receptor NR3C1 was ubiquitously expressed in CD9+ and CD9− cell lines, and was not significantly altered upon CD9 overexpression/knockout.
Our results collectively indicate that CD9 negativity was definitively linked to glucocorticoid resistance, which could be partially reversed by CD9 reactivation through a NR3C1-independent mechanism. Comprehensive understanding of the interaction between CD9 and glucocorticoid susceptibility could lead to improved therapeutic strategies for resistant pediatric B-ALL. The findings could also be exploited for other related malignancies (e.g. AML), of which glucocorticoids remain largely inefficacious.