A Phase II Study of Midostaurin and 5-Azacitidine for Elderly Patients with Acute Myeloid Leukemia

Introduction: Acute myeloid leukemia (AML) affects predominantly elderly patients (pts); however, their responses to standard chemotherapy (chemo) are inferior to those of younger adults. Fms-like tyrosine kinase 3 (FLT3) is a class III receptor tyrosine kinase that regulates growth and proliferation of myeloid cells, is expressed in 80% of AML samples, and contains an activating mutation in approximately 30% of AML pts. Midostaurin (mido) is a multikinase inhibitor with activity towards FLT3, and was recently approved in combination with standard induction chemotherapy for treatment of FLT3-mutant (FLT3-mut) AML. The phase I portion of this trial investigated the safety and tolerability of mido in combination with azacitidine (aza), a standard treatment (tx) for AML in elderly pts unfit for standard induction chemotherapy (Clin Lymphoma Myeloma Leuk 2015, 15:428-432). The maximum tolerated dose (MTD) was 75 mg orally (PO) twice daily (BID) in combination with aza. Here we report further safety and efficacy results of a preplanned expansion cohort.

Methods: Stage 1 of this Phase II study enrolled 17 previously untreated pts with AML (including 9 previously untreated pts from Phase I treated at the MTD). Stage 2 enrolled 9 additional pts for a total of 26 evaluable pts. All patients were unfit for standard induction chemotherapy (chemo). Pts received mido 75 mg PO BID on days 8-21 and aza 75 mg/m² intravenously (IV) on days 1-7 of 28-day cycles. Response was assessed after cycles 2 and 4. Additional cycles were offered to pts who demonstrated clinical benefit. Dose adjustments for QT_c prolongation, hematologic and non-hematologic toxicities were allowed. Concomitant administration of strong CYP3A4 inhibitors and inducers was discouraged; fluconazole at a dose of 200 mg daily was allowed. Response was determined per 2003 IWG criteria. Adverse events were graded according to version 4 of the Common Terminology Criteria for Adverse Events. FLT3 activity was measured via the plasma inhibitory assay (Blood 2006, 108:3477-3483).

Results: The median age was 74 (range = 59-85), 57.7% were female, 53.8% were ECOG 1 performance status (0-2). 53.8% of pts had unfavorable risk disease, with complex cytogenetics as the most common karyotype (42.3%). 7 pts had secondary AML. No pts had documented FLT3-ITD mutations. 25 pts received at least 1 day of mido with aza as 1 pt had rapid rise in leukocytes after enrollment and was felt inappropriate to start mido therapy. The treatment emergent adverse events (TEAEs) reported by > 20% of pts were nausea (58%), constipation (58%), diarrhea (54%), fatigue (50%), edema (46%), dyspnea (35%). Grade III/IV infection (infxn), including bacterial, fungal, and viral, was common with 9 (35%) of pts reporting at least 1 instance of documented infxn and an additional 4 (15.4%) pts with uncomplicated febrile neutropenia. Additional Grade III and IV TEAEs reported in > 10% of pts were hypotension (15%), syncope (15%). 10 pts were removed early before two cycles; 8 for infxn, 1 for syncopal events, and 1 for consent withdrawal. There were 3 on study deaths, all due to infxn. Mean number of days on study was 115 (median = 58, range 0-576). Of 16 pts evaluable for response, there were 4 complete remissions, 1 complete remission with incomplete recovery of neutrophils, and one partial response with hematologic improvement in platelets. Median OS was 262 days (95% CI, 203-472). For pts with FLT3 activity results (n = 7), activity decreased by over 70% between the start and mid-point of each cycle (range, 45-89%).

Conclusions: This is the first study investigating the efficacy and toxicity of combination aza and mido in previously untreated, elderly AML pts without a FLT3 mutation. Duration of tx and overall response rate were similar to those reported in a Phase II trial of mido and aza which included FLT3-mut AML and previously treated pts, and utilized a lower dose of mido (50 mg PO BID) (Am J Hematol 2015, 90:276-81). However, the degree of FLT3 inhibition and incidence of characteristic side effects of mido (e.g ., nausea) was higher in our study, likely explained by the higher dose of mido administered. Because no FLT3-mut pts were enrolled, the effect of FLT3 mutation status on response cannot be determined for this dosing schedule. The combination of standard dose aza with mido is feasible and warrants further investigation in AML pts unfit for intensive therapy.

Figure

View largeDownload slide

Figure

View largeDownload slide

Close modal