Oncogenic addiction to the Fms-like tyrosine kinase 3 (FLT3) is a hallmark of acute myeloid leukemia (AML) that harbors the FLT3–internal tandem duplication (FLT3-ITD) mutation. While FLT3 inhibitors like sorafenib show initial therapeutic efficacy, resistance rapidly develops through mechanisms that are incompletely understood. Here, we used RNA-Seq–based analysis of patient leukemic cells and found that upregulation of the Tec family kinase BMX occurs during sorafenib resistance. This upregulation was recapitulated in an in vivo murine FLT3-ITD–positive (FLT3-ITD+) model of sorafenib resistance. Mechanistically, the antiangiogenic effects of sorafenib led to increased bone marrow hypoxia, which contributed to HIF-dependent BMX upregulation. In in vitro experiments, hypoxia-dependent BMX upregulation was observed in both AML and non-AML cell lines. Functional studies in human FLT3-ITD+ cell lines showed that BMX is part of a compensatory signaling mechanism that promotes AML cell survival during FLT3 inhibition. Taken together, our results demonstrate that hypoxia-dependent upregulation of BMX contributes to therapeutic resistance through a compensatory prosurvival signaling mechanism. These results also reveal the role of off-target drug effects on tumor microenvironment and development of acquired drug resistance. We propose that the bone marrow niche can be altered by anticancer therapeutics, resulting in drug resistance through cell-nonautonomous microenvironment-dependent effects.
Jolieke G. van Oosterwijk, Daelynn R. Buelow, Christina D. Drenberg, Aksana Vasilyeva, Lie Li, Lei Shi, Yong-Dong Wang, David Finkelstein, Sheila A. Shurtleff, Laura J. Janke, Stanley Pounds, Jeffrey E. Rubnitz, Hiroto Inaba, Navjotsingh Pabla, Sharyn D. Baker
Usage data is cumulative from April 2024 through April 2025.
Usage | JCI | PMC |
---|---|---|
Text version | 620 | 124 |
130 | 44 | |
Figure | 333 | 15 |
Supplemental data | 275 | 17 |
Citation downloads | 105 | 0 |
Totals | 1,463 | 200 |
Total Views | 1,663 |
Usage information is collected from two different sources: this site (JCI) and Pubmed Central (PMC). JCI information (compiled daily) shows human readership based on methods we employ to screen out robotic usage. PMC information (aggregated monthly) is also similarly screened of robotic usage.
Various methods are used to distinguish robotic usage. For example, Google automatically scans articles to add to its search index and identifies itself as robotic; other services might not clearly identify themselves as robotic, or they are new or unknown as robotic. Because this activity can be misinterpreted as human readership, data may be re-processed periodically to reflect an improved understanding of robotic activity. Because of these factors, readers should consider usage information illustrative but subject to change.