A Phase 1 Trial of Imatinib Mesylate with Daunorubicin and Cytarabine for Patients with C-Kit Positive Relapsed AML.

C-kit is a tyrosine kinase receptor expressed on more than 10% of blasts in 95% of relapsed AMLs. Signaling pathways, such as STAT3 and STAT5, may be activated downstream of c-kit. Imatinib mesylate (IM) is a potent c-kit inhibitor and has activity in relapsed/ refractory AML. We conducted a Phase 1 trial of IM in combination with cytarabine (A) and daunorubicin (D) in pts with c-kit+ relapsed AML.

Methods: All pts were treated at the Cleveland Clinic from 2003–2008. Cytogenetic (CG) risk was defined by CALGB criteria. Eligibility criteria included: age ³18 yrs, AML in first relapse (excluding APL), relapse more than 6 mos from induction therapy, ECOG performance status 0–2, ³20% blasts c-kit+ (CD117+ by flow cytometry). Phosphorylated (P) (activated) STAT3 (nuclear and cytoplasmic) was assessed by immunhistochemistry (IHC) on pre-tx samples. IHC studies for P-STAT5 are ongoing. Cases were defined as positive if any staining was present in blasts. All pts received A 100 mg/m²/d by continuous infusion for 7 days and D 45 mg/m² IV for 3 days (7+3). IM dose was escalated using a standard 3 + 3 design. Six additional pts were then treated at the MTD. Planned dose levels were: −1 (300 mg), 1 (400 mg), 2 (600 mg), 3 (800 mg). IM was administered with the first dose of chemotherapy and continued daily until disease progression, intolerance, removal from study for another treatment (including alloBMT), or dose-limiting toxicity (DLT). A Day 14 BM was performed; pts with persistent leukemia but ≥ 50% reduction in BM blasts continued IM and received 5 days of A (100 mg/m²/d) and 2 days of D (45 mg/m²/d).

Results: Twenty-one pts have been enrolled: median age was 47 yrs (range 24–75) and 48% were male. The median time from CR to relapse was 439 d (range 216–1100). CG risk: poor 14%, intermediate 71%, and good 14%. The median percentage of c-kit+ blasts was 89% in the bone marrow (range 52–98). In patients expressing nuclear P-STAT3, 2–6% of the blasts stained positive. Eleven pts had no nuclear P-STAT3. No pts had cytoplasmic P-STAT3. At 400 mg IM, 2 of 6 pts had a DLT; both were ³ Grade 3 hyperbilirubinemia. Pharmacodynamic/ pharmacokinetic studies to better understand the etiology of the hyperbilirubinemia are ongoing. Subsequent pts were treated with 300 mg IM. One pt died during induction and was not evaluable for response. Eleven of the 19 evaluable pts (58%) achieved CR or CRp . Pts received IM for a median of 23 d (range 9–574+). Three discontinued IM to proceed to alloBMT. Two pts continue on IM and remain in a remission at 52+ and 574+ days. C-kit expression did not correlate with CR. However, patients with no nuclear P-STAT3 expression had a higher CR rate (82%) than patients with P-STAT3 nuclear expression (33%) (p=0.18).

Conclusions: The MTD of IM in combination with 7+3 was 300 mg. Since the hyperbilirubinemia appeared to be reversible with discontinuation of IM, it may be possible to escalate IM doses further and re-challenge. The CR rate is encouraging, and merits evaluation in a Phase 2 study. In particular, pts with no nuclear P-STAT3 had a CR rate of 82%; whereas, pts with nuclear P-STAT3 expression had a CR rate of 33%. STAT3 may be activated independently of c-kit; therefore, directly targeting STAT3 in this latter group of patients may potentially improve prognosis. However, these are preliminary data, and will need to be evaluated in a larger study.