Efficacy of Vemurafenib in Hairy-Cell Leukemia

Abstract

To the Editor: The BRAF V600E mutation is present in nearly all cases of hairy-cell leukemia. 1 This finding has led to the introduction of BRAF inhibitors for the treatment of chemotherapy-re-sistant hairy-cell leukemia, and patients have had a good response to the oral inhibitor vemu-rafenib. 2-4 Constitutive phosphorylation of both extracellular signal-regulated kinase (ERK) and mitogen-activated protein–ERK kinase (MEK) has been considered to be a direct consequence of BRAF activation, with BRAF inhibition result-ing in cell death through suppression of this pathway in hairy-cell leukemia. However, data to support this theory are limited, since most pa-tients present with pancytopenia. We evaluated a patient with purine analogue– refractory hairy-cell leukemia who had biallelic BRAF V600E mutations and a high leukemic bur-den during treatment with vemurafenib. Because of the high numbers of circulating hairy-cell leukemia cells, it was possible to study the ef-fects of vemurafenib directly in vivo. Vemurafenib induced complete clinical remission with reduc-tion of the viability of CD103+ hairy-cell leuke-mia cells during therapy (Fig. 1A). A pull-down and kinase assay showed inhibition of BRAF in leukemic cells in vivo (Fig. 1B). However, BRAF inhibition was not associated with any major changes in phosphorylation of either MEK or ERK in vivo, as shown by means of both immu-noblot and flow cytometry (Fig. 1C), despite prolonged exposure to vemurafenib. Our experiments showed an unanticipated un-coupling between the decrease in BRAF activity (together with increased cell death) and MEK– ERK inhibition in vivo; this was not dependent on the duration of exposure to vemurafenib. We cannot rule out the possibility that BRAF inhibi-tion in vivo eventually resulted in suppression of ERK activation in some anatomical compart-ment other than the blood before leukemic cell death, but this possibility appears to be unlikely. First, our in vivo data clearly showed inhibition of BRAF without any change in phosphorylated ERK levels in leukemic cells, as shown by means of both immunoblot and flow cytometry, while cells were dying (as shown by the increasing level of propidium iodide staining). Second, the lack of effect of BRAF inhibitors on phosphory-lated MEK and ERK was previously reproduced during prolonged incubation in vitro of hairy-cell leukemia cells obtained from the same pa-tient, whereas the MEK 1 and MEK 2 inhibitor PD325901 successfully blocked ERK phosphory-lation and induced significant cell death. 5 An alternative signaling pathway, as yet un-characterized, may therefore be affected by ve-murafenib, either directly or through BRAF inhi-bition, and it may have a strong impact in hairy-cell leukemia cell survival in vivo. These data have implications for the design of possible combination therapies.