Venetoclax Resistant Philadelphia Chromosome Positive Acute Lymphoblastic Leukemia Cells Are Also Resistant to ABL Inhibitors, Which Can be Overcome By Targeting the PI3K/AKT/ mTOR Pathway

Introduction: Venetoclax is a BH3-mimetic shown to specifically inhibit BCL-2 and has activity in hematologic malignancies addicted to BCL-2, including Ph+ ALL. However, when used as a single agent resistance develops through upregulation as well as post-translational modifications of MCL-1 from a variety of different pathways. It remains unknown what effects if any BCL-2 resistance will have on ABL kinase inhibition. In these studies, we investigated whether acquired resistance to venetoclax affected kinase sensitivity in a pre clinical model of Ph+ ALL.

Methods: A venetoclax resistant cell line (SUPVR2) was generated by progressively exposing the Ph+ ALL cell line SUPB15 to increasing levels of venetoclax, and then characterized to the parent cell line. Expression of both pro and anti apoptotic members of the BCL-2 family of proteins was assessed via immunoblot. Sensitivity of resistant cells was then compared to parental cells across a library of small molecule kinase inhibitors. Briefly, cells were incubated with a panel of 130 small molecule inhibitors at varying concentrations for 72 hours, and then assessed using the colorimetric MTS assay as an indirect measure of viability. Drug effectiveness was evaluated based on the IC50 of each drug on the panel against each cell line. SUPB15 and SUPVR2 cells were also exposed to increasing concentrations of dasatinib or venetoclax after treatment with 40 µM of siRNA targeting ABL-1, MCL-1 or a non-specific control (NS). Drug combinations were assessed by co-culturing resistant cells with varying concentrations of each drug in combination or as single agent, and then synergy calculations were perfromed using Calculsyn software.

Results: The IC50 of the SUPVR2 cells to venetoclax was >100µM, compared to 10nM for SUPB15. Levels of MCL-1 were higher and BCL-2 were lower in SUPVR2 cells as compared to SUPB15, but no differences were seen in BCL-xL or the anti-apoptotic molecules BIM, BAX/BAK, PUMA or NOXA. When assessed with the full drug panel, SUPVR2 were found to have reduced sensitivity compared to SUPB15 to the ABL TKIs nilotinib (IC50 257.7nM vs 10.7nM ) and dasatinib (306.4nM vs 4.6nM) but not ponatinib (1.5nM vs 1nM). In contrast, SUPVR2 cells showed enhanced sensitivity to molecules that targeted the PI3K/mTOR pathway (rapamycin (0.94nM vs15.6nM), PP242 (39.4nM vs 72.5nM), INK-128 (40.6nM vs 91.0nM), BEZ235 (29nM vs 531.1nM) and PI-103 (71.3nM vs 123.3nM). Although SUPVR2 did still harbor the BCR-ABL transcript, siRNA inhibition of ABL had no impact on viability compared to SUPB15, suggesting that the SUPVR2 are no longer ABL-dependent. siRNA inhibition of MCL-1 expression restored sensitivity of the resistant cells to venetoclax, but only partially restored sensitivity to dasatinib. Co-incubation of cells with dasatinib and either rapamycin, PP-242 or PI-103 led to restoration of dasatinib sensitivity in a synergistic manner.

Conclusions: Resistance to venetoclax in SUPB15 leads to concurrent resistance to ABL kinase inhibitors, which appears to occur via activation of an ABL-independent pathway. Although it is not clear whether this resistance will develop if both an ABL kinase inhibitor and a BCL-2 inhibitor are administered concurrently, our results indicate that inhibitors of PI3K/mTOR may prevent the emergence of venetoclax and ABL resistance.