Arsenic and old FLT3

FLT3-internal tandem duplication (FLT3-ITD) mutations are common in acute promyelocytic leukemia (APL) and render the disease more difficult to manage or cure. In this issue of Blood, Esnault et al begin to unravel how the mutation-activated FLT3 receptor impedes the effects of all-trans retinoic acid (ATRA) and how arsenic counters this.¹ 

FLT3 activating mutations have thus far defied formal classification as an acute myeloid leukemia (AML) subtype. “FLT3-mutant AML” is not a World Health Organization–designated category of the disease, and for good reason. These mutations seem to delight in showing up in different subtypes of AML and making a bad situation worse. Take APL for example. FLT3-ITD mutations are found in roughly one-third of all patients with APL, and in the pre-arsenic era, they rendered APL more problematic to manage and adversely affected the survival of patients with an otherwise curable disease.^2,3 

Esnault et al crossed an FLT3-ITD knock-in transgenic mouse line with an established promyelocytic leukemia/retinoic acid receptor alpha (PML/RARA) mouse line to generate the double-mutant murine APL. There was no obvious difference in phenotype between the mice with PML/RARA and those with PML/RARA/FLT3-ITD, but the double-mutant mice showed a blunted response to ATRA, both in differentiation and in survival. The reorganization of nuclear bodies that would normally occur in response to ATRA did not occur in the double-mutant mice nor was PML/RARA degraded. The use of arsenic, however, induced differentiation, RARΑ degradation, and p53 induction in both models.

This demonstration that arsenic abrogates the FLT3-ITD–conferred resistance to ATRA makes sense in light of the clinical data. Shortly after they were discovered, FLT3-ITD mutations were noted to occur frequently in APL and were typically associated with a high white blood cell count at presentation, defining such patients as high risk by conventional criteria.²  Treatment of FLT3-ITD–mutated APL with ATRA combined with chemotherapy was often successful but still resulted in worse overall survival compared with APL patients lacking such mutations.³  The introduction of arsenic into the management of APL has led to remarkable survival rates,⁴  but high-risk patients continue to be somewhat problematic. Interestingly, the findings from one clinical study are particularly salient to this topic.⁵  Iland et al reported that high-risk APL patients treated with a regimen using both ATRA and arsenic fared slightly worse than low- or intermediate-risk patients but not if they had an FLT3 mutation. In other words, arsenic seemed to abrogate the prognostic effects of FLT3 mutations specifically. The findings of Esnault et al, therefore, provide important scientific support for the clinical results reported by Iland et al⁵  and suggest that APL patients with FLT3 mutations will benefit from treatment with induction regimens incorporating both ATRA and arsenic, and they probably do not need an FLT3 inhibitor. For practical purposes, this essentially means that all APL patients should have arsenic as a component of induction, not just those with low- and intermediate-risk disease.

There is, perhaps, a larger story in all of this. Work over the past several years has yielded a model of AML in which the disease develops from the sequential acquisition of driver mutations.^6,7  In general, any individual driver mutation is insufficient to cause the disease. Likewise, targeting any individual driver mutation with single-agent therapy (such as with ATRA alone, an FLT3 inhibitor alone, or an IDH inhibitor alone) is insufficient to cure the disease. APL is the first subtype of AML that can be consistently cured (provided the patient gets to the hospital in time). However, to accomplish that, it seems that the mutations driving each particular case have to be dealt with in some way. For the other AML subtypes driven by multiple different mutations, we can expect the same.