Update of a Phase I Study of Sorafenib in Patients with Refractory/Relapsed Acute Myeloid Leukemia or High-Risk Myelodysplastic Syndrome.

The critical importance of the Ras, VEGF, and FLT3 pathways in the pathogenesis of myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) has been well established. FLT3 abnormalities, internal tandem duplication (ITD) and point mutations, occur in about 30% of pts with AML and the FLT3-ITD mutation independently confers poor prognosis. Sorafenib is an oral multikinase inhibitor targeting the above pathways and is highly potent against FLT3-ITD mutants (IC₅₀ 1–3 nM) (ASH abstract, 2006). We are conducting a phase I trial to evaluate the safety and efficacy of two different schedules of sorafenib. To date, 21 patients (pts) with refractory/relapsed AML (n=20) and high risk MDS (n=1) have been enrolled. Pts were randomized to sorafenib for 5 days per week for 21 days (arm A; n=11) or for 14 days every 21 days (arm B; n=10). In both arms the starting dose level (DL) is 200 mg twice daily. Successive dose levels are 600, 800, and 1200 mg daily in a standard 3+3 design. Peripheral blood (PB) and bone marrow (BM) samples were obtained for evaluation of FLT3 status and phosphorylated and total FLT3 and ERK expression. Median age is 62 years (range, 33–82), number of prior therapies 2 (range, 1–5), time from diagnosis to sorafenib treatment 9 months (range, 2–46), and median duration on study was 1.2 months (range, 0.1–3.4). Twenty pts are evaluable. 9/20 (45%) pts received ≤ 1 cycle of sorafenib because of disease progression (n=6), self-discontinuation (n=2), or no benefit (n=1), of whom 5 (56%) were FLT3-ITD negative, 3 (33%) were FLT3-ITD positive, and 1 (11%) was not tested. In contrast, 11/20 (55%) pts received > 1 cycle of sorafenib, of whom 8 (73%) were FLT3-ITD positive and 3 (27%) were FLT3-ITD negative; reasons for discontinuation were disease progression (n=5), self-discontinuation (n=2), stem cell transplant (n=2), or no benefit (n=2). Sorafenib has been well tolerated with 1 pt achieving a DLT of grade 3 hyperbilirubinemia at the 800 mg daily dose in arm B, but the MTD has not been reached; this cohort has been expanded. The only other grade 3 toxicity has been pleural effusion at the 600 mg daily dose in arm A, not considered a DLT because it occurred during cycle 2. A ≥ 50% reduction in PB or BM blasts was obtained in 11/20 (55%) pts. 9/11 (82%) pts harbored the FLT3-ITD mutation and had a median duration of response of 42 days (range, 15–87). In these 9 pts, the median PB absolute blast count at baseline and after maximal response to sorafenib was 10.3 (range, 0.2–18.7) and 0 (range, 0–1)(p=0.008). Median BM blast percentage at baseline and after maximal response to sorafenib was 72% (range, 14–96) and 42% (range, 12–58) (p=0.002), with 1 pt achieving a morphologic complete remission in the BM. Serial determinations of phosphorylation status following sorafenib (at 0, 2, 24,120 hours) in pts with the FLT3-ITD mutation demonstrated inhibition of phospho-FLT3 in 3/3 and phospho-ERK in 5/5 pts. In conclusion, sorafenib administration is safe in AML and appears to preferentially target the FLT3-ITD mutation. This study continues to accrue pts to define the MTD and it will be followed by combination studies of standard chemotherapy with sorafenib, with an emphasis on targeting pts with AML expressing the FLT3-ITD mutation.