Ponatinib in patients with refractory acute myeloid leukaemia: findings from a phase 1 study

Activating mutations in the FMS-like tyrosine kinase-3 (FLT3), a tyrosine kinase receptor important in haematopoiesis, are among the most common molecular aberrations in acute myeloid leukaemia (AML), occurring in 30% of adult patients (Levis & Small 2003). Common FLT3-activating mutations include FLT3 internal tandem duplications (FLT3-ITDs), detected in about 23% of AML patients, and point mutations within the tyrosine kinase domain, found in about 8% (Levis & Small 2003). These mutations result in a constitutively active FLT3 receptor, leading to growth factor–independent proliferation and survival of leukaemic cells and conferring poor prognosis (Levis & Small 2003). Clinical studies of single-agent first-generation FLT3 inhibitors have demonstrated clinical activity, with responses that are typically short-lived and mostly partial or complete responses with incomplete haematopoietic recovery. This may be due to suboptimal potency and/or pharmacokinetics, leading to insufficient or transient target inhibition, or concomitant c-kit inhibition (Knapper 2011). Recently, high potency second-generation FLT3 inhibitors (eg, quizartinib) have shown substantial efficacy as monotherapy, suggesting a potency threshold for clinical benefit (Knapper 2011). The validation of FLT3-ITD as a therapeutic target has rekindled interest in developing and testing new potent FLT3 inhibitors in AML patients with FLT3-ITD mutations (Smith et al, 2012). Ponatinib is a novel, orally administered tyrosine kinase inhibitor (TKI) and a potent pan–BCR-ABL1 inhibitor (O’Hare et al, 2009). Based on results in patients with chronic myeloid leukaemia (CML) and Philadelphia chromosome–positive acute lymphoblastic leukaemia (Ph+ ALL) in phase 1 and phase 2 clinical trials (Cortes et al 2012a, Cortes et al (2012b), ponatinib (45 mg once daily) has been approved in the United States for the treatment of patients with CML and Ph+ ALL that is resistant or intolerant to prior TKI therapy. Preclinical studies revealed that ponatinib also potently inhibits FLT3, leading to apoptosis of leukaemic cell lines carrying the FLT3-ITD mutation and tumour regression in xenograft models, suggesting the potential for activity in patients with AML (Gozgit et al, 2011). Additionally, ponatinib appears to retain activity against the clinically-relevant quizartinib-resistant mutant FLT3-ITD F691L (Smith et al, 2013). Here we report the first clinical experience with ponatinib in 12 AML patients included in the phase 1 study. Methods are described in the on-line supporting information. The median age of these patients was 49 (30-72) years. The median time from diagnosis to treatment was 1 year. Patients received a median of 3 (1-7) prior therapies; 58% had received 3 or more prior therapies (Table I and Table S1). Mutational analysis in a central laboratory confirmed the presence of FLT3-ITD in 7 patients (58%). Three additional patients did not have an adequate DNA sample at study entry; however, they had a history of FLT3-ITD mutation—as reported by the investigator—and they are included in the FLT3-ITD mutation– positive group for these analyses. Three patients (all FLT3-ITD mutation positive) were previously treated with one or more FLT3 inhibitors (sorafenib, quizartinib, and/or IMC-EB10); one patient progressed on IMC-EB10 and had a partial response to sorafenib, one patient had a complete response to sorafenib and a partial response to quizartinib, and one patient had a partial response to quizartinib. Seven patients (70%) with FLT3-ITD mutation were FLT3 inhibitor– naive (Table I). The median treatment duration was 52 (10-173) days. At the time of analysis, all patients had discontinued ponatinib: 5 (42%) due to death (all unrelated to ponatinib), 3 (25%) due to adverse events (AEs: unrelated central nervous system [CNS] haemorrhage, possibly related acute pancreatitis, unrelated graft vs host disease), 2 (17%) due to progressive disease (PD), and 2 (17%) due to investigator decision (Table I). Table 1 Selected baseline characteristics, treatment duration, response, and reasons for discontinuation by individual patients with AML Nine patients experienced at least one treatment-related AE. The most common treatment-related AEs occurring in 2 or more patients were pancreatitis (n=3) and petechiae (n=2). Three patients experienced a treatment-related serious AE (SAE) of pancreatitis (all grade 2), which was a dose-limiting toxicity in this trial (Cortes et al, 2012a). Pancreatitis resolved in 2 patients after dose interruption, lasting 3 days in one patient and 8 days in the other. These 2 patients continued therapy at a reduced dose (30 mg) and were subsequently re-escalated to 45 mg without recurrence. The third patient discontinued therapy per investigator decision. Additional details regarding treatment-emergent AEs and SAEs can be found in Table S2. Seven patients died during the study for reasons not related to ponatinib: disease progression (n=3), multiorgan failure (n=2), pneumonia and sepsis (n=1), and CNS haemorrhage (n=1) (Table I). Ponatinib had an acceptable safety profile in this small group of patients with refractory AML, similar to that observed in patients with CML and Ph+ ALL. Few treatment-related AEs were reported; the most common was pancreatitis, which was manageable, and re-challenge with ponatinib was possible in most cases. The geometric mean maximal concentration and area under the curve of single-dose ponatinib at day 1, cycle 1 in AML patients were 97 nM and 1441 nM*h, respectively, similar to findings across all 31 patients receiving 45 mg ponatinib (98.8 nM and 1360.1 nM*h). The overall response rate (RR, partial remission or better) was 3/12 (25%): 2 patients achieved complete remission with incomplete blood count recovery and one patient experienced partial remission (Table I, Fig 1). These 3 responders carried FLT3-ITD mutations and were all FLT3 inhibitor–naive; the duration of ponatinib treatment in these patients was 3 to 6 months. Among 10 patients with FLT3-ITD mutations, RR was 3/10 (30%). Among 7 patients with FLT3-ITD mutations who were FLT3 inhibitor–naive, RR was 3/7 (43%). Three patients (2 FLT3-ITD negative) had stable disease, as they did not meet criteria for complete/partial remission or PD; however, peripheral blood blasts in 2 of these patients decreased considerably (~60-90%) during the first treatment cycle. The RR reported with quizartinib in phase 1 testing was 30% (Cortes et al, 2009) and 10% with sorafenib (Borthakur et al, 2011). Although the sample size reported here is small, these results suggest that ponatinib has clinical activity in AML patients with FLT3-ITD, requiring confirmation in a larger cohort of patients and with additional focus on optimization of response (eg, combination therapy) and response durability. Figure 1 Course of the disease in 3 responders during ponatinib treatment