We have shown that the proteasome inhibitor bortezomib is able to disrupt the physical interaction between the transcription factor Sp1 and NF-kB(p65), inhibiting Sp1 DNA-binding activity in myeloid blasts. (Liu, Blood 2008) We have also observed that both KIT and FLT3 promoters harbor Sp1 binding sites and that Sp1 induces transcription of these genes. We hypothesize that the Sp1/NF-kB pathway is a modulator of tyrosine kinase (TK) activity, and that administration of bortezomib with the tyrosine kinase inhibitor midostaurin will down regulate TKs at the transcription level and inhibit kinase activity in AML. The expectation is that two fold molecular targeting in combination with chemotherapy will result in a greater antileukemic effect.


Adults with relapsed or refractory AML by WHO criteria and preserved organ function with ECOG ≤ 2 were eligible. Patients enrolled at dose levels (DL) 1 and 2 received midostaurin 50mg po BID d 1-14, bortezomib was dose escalated – DL 1: 1.0mg/m2 IV d 1,4,8,11; DL 2: 1.3mg/m2 IV d 1,4,8,11. Cycles were repeated every 28 d or until achievement of morphologic complete remission (CR) or CR with incomplete count recovery (CRi), for up to 3 cycles. Patients enrolled at DL 3 received mitoxantrone 4mg/m2, etoposide 40mg/m2, cytarabine 1g/m2 d1-6 (MEC) followed by midostaurin 50mg bid d 8-21 and bortezomib 1.3mg/m2 IV d 8,11,15,18. Due to the development of grade 3 peripheral neuropathy (PN) that was a dose limiting toxicity (DLT) in two patients treated at DL 3, bortezomib administration was changed from IV to subcutaneous (SQ) and the dose and schedule was changed to 1.0mg/m2 on d 8 and 15 on (DL 3A) - doses of midostaurin and MEC remained constant. Patients enrolled at DL 3 or 3A received one cycle of treatment, but were able to continue with maintenance bortezomib and midostaurin only for up to three cycles if they did not proceed to allogeneic stem cell transplantation (SCT). Responses were graded by International Working Group criteria for AML (Cheson, JCO 2013).


11 patients enrolled to DL 1 and 2; 7 had relapsed disease (5 with CR1< 1 year) and 4 were primary refractory. The median age was 67 (range, 64-76) and 4 patients were FLT3-ITD or FLT3-TKD positive. According to European LeukemiaNet (ELN) genetic risk groups, 6, 3, and 2 patients had intermediate I, intermediate II and adverse risk disease, respectively. 2 patients received more than one cycle of chemotherapy (range, 1-2), all developed progressive disease and were removed from study. There were no DLTs at DL 1 or 2. 13 patients enrolled to DL 3 and 3A, 8 had relapsed disease (3 with CR1< 1 year) and 5 were primary refractory. The median age was 53 (range, 19-70) and 4 patients were FLT3-ITD positive. According to ELN genetic risk, 2, 7, 1 and 3 patients had favorable, intermediate I, intermediate II, and adverse risk, respectively. Of the first 3 patients treated at DL 3, one developed grade 3 PN which was a DLT. 3 additional patients were enrolled and 1 patient with a history of anthracycline exposure had a grade 3 decrease in EF. An additional 3 patients were treated, 1 developed grade 3 diarrhea and 1 developed grade 3 PN, both DLT. Given that 8/9 (88%) patients treated at DL 3 achieved a CR or CRi, we chose to further explore the regimen, however, given the concerns for PN, the dose of bortezomib only was modified (DL 3A). 4 additional patients were treated at DL 3A, 2 achieved CR, 1 achieved CRi. No patients have developed PN and there were no DLTs at this dose level. Death within 8 weeks occurred in 8 patients (34.7 %) due to sepsis (1), pneumonia (1), enterococcus bacteremia (1), or progressive disease (5). Overall the objective response rate was 47.8% (11/23), however the CR/CRi response in those who received bortezomib, midostaurin and chemotherapy was 84.6% (11/13). The CR/CRi rate in those with a FLT3 mutation was 75% (3/4). 7 patients have gone on to receive an allogeneic SCT.


The combination of bortezomib, midostaurin and intensive chemotherapy was active in this group of patients with relapsed/refractory AML. DLT of PN was observed, however with a decrease in dose and frequency of bortezomib this has not recurred. Correlative studies to evaluate bortezomib and midostaurin in modulating tyrosine kinase activity are ongoing.

Off label use

Bortezomib in AML


Off Label Use: Bortezomib in AML.

Author notes


Asterisk with author names denotes non-ASH members.