Gefitinib adopts orphan kinases in leukemia

Comment on Stegmaier et al, page 2841

Gefitinib, an orally active small-molecule drug that is used for lung cancer therapy based on its EGFR kinase inhibitory properties, is shown to promote differentiation of leukemic cell lines and primary blasts from AML patients in an EGFR-independent fashion.

Gefitinib (Iressa, Astra Zeneca, London, United Kingdom) is a selective, small-molecule inhibitor of adenosine triphosphate (ATP) binding by epidermal growth factor receptor (EGFR). The drug is currently approved for treatment of chemotherapy refractory, advanced non–small cell lung cancer, although results from phase 3 randomized studies have not confirmed gefitinib-associated survival benefit. (Note: it is currently subject to new Federal Drug Administration [FDA] regulations that restrict its access.¹ ) However, tumors carrying sensitizing EGFR mutations within exons 18 to 21 (tyrosine kinase domain) might be particularly vulnerable to the drug, and such mutations might be used as molecular predictors of response. Regardless, in this issue of Blood, Stegmaier and colleagues demonstrate an apparently EGFR-independent differentiation effect of gefitinib on both acute myeloid leukemia (AML) cell lines and primary patient-derived AML blasts in vitro.

The authors applied a high-throughput strategy that included use of gene expression analysis, mass spectrometry, and functional tests to screen for candidate molecules that are capable of inducing myeloid differentiation.²  Such mechanistically insightful strategy predefined the treatment goal as a specific gene expression signature reflecting a desired whole-cell phenotype (ie, myeloid differentiation), and it successfully leveraged the high information content inherent in the gene expression analysis independent of a specific molecular target. Subsequently, the observations for one of the lead compounds, dianilinophthalimide (DAPH1), an EGFR kinase inhibitor, was extended through the study of gefitinib. The latter is a more attractive study drug than DAPH1 since it is already FDA approved. By morphologic criteria, 10 μM gefitinib induced neutrophilic differentiation of HL-60 and Kasumi-1 cells, as well as monocytic differentiation of U937 cells, suggesting activation of a broad differentiation program that may be cell-context dependent. Gefitinib also induced nitro-blue tetrazolium (NBT)–reducing activity in both HL-60 and U937 cells, and drug treatment induced global changes in gene expression at 24 hours that were consistent with neutrophilic differentiation.

Interestingly, gefitinib's effects in the aforementioned assays did not appear to be dependent on either its anti-EGFR or anti-ERBB2 (HER-2/neu) activity. The authors failed to detect EGFR and ERBB2 proteins in HL-60 and Kasumi-1 cells, and treatment with the monoclonal anti-EGFR and anti-ERBB2 antibodies, cetuximab and trastuzumab, respectively, failed to induce myeloid differentiation of cells. This is consistent, as recently suggested,³  with inhibition of 1 or more cellular kinases that may represent hitherto uncharacterized gefitinib targets. Gefitinib also decreased the viability of blasts derived from 6 of 8 AML patients across several French-American-British (FAB) subtypes at a 50% effective concentration (EC50) of less than 5 μM (clinically achievable range)—for several patients, morphologic evidence of differentiation was evident, and in some, global expression profiling confirmed activation of the myeloid differentiation program. Nevertheless, it remains to be seen whether the differentiation model using HL-60 cells (an acute promyelocytic leukemia [APL] cell line) has set a clinically useful threshold for positive hits in the primary chemical screen. Further preclinical testing using other leukemia cell lines as well as primary blasts from additional AML patients may help identify predictors of response and help assess the robustness of the gefitinib induced differentiation response prior to its translation in the clinic. ▪