Phase 1 Study of UCN-01 in Combination with Perifosine in Patients with Relapsed and Refractory Acute Leukemias and High Risk Myelodysplastic Syndromes (MDS)

The phosphatidylinositol-3-kinase (PI3K)/Akt signaling pathway controls key cellular processes and is frequently activated in both acute and chronic myeloid leukemia (AML, CML) as well as high-risk MDS, thus representing an attractive therapeutic target. UCN-01 inhibits a number of serine-threonine kinases, including phosphatidylinositide-dependent kinase 1, which phosphorylates and activates Akt. Perifosine, an alkylphospholipid, targets the pleckstrin homology domain of Akt, thereby inhibiting its plasma membrane translocation and activation. Synergistic inhibition of Akt by UCN-01 and perifosine has been demonstrated in prostate and lung cancer cell lines (Clin Cancer Res 10: 5242, 2004).

We conducted a Phase I study to determine the maximum tolerated dose (MTD) and recommended Phase II dose of UCN-01 and perifosine, using a standard 3+3 dose escalation design. Patients (pts) >18 years with poor-risk, or relapsed/ refractory AML or acute lymphoblastic leukemia (ALL), CML in accelerated/blastic phase or MDS failing standard therapy, and peripheral blast count <30×10⁹/l were eligible. Perifosine was administered on day (d)1 at 150 mg orally every 6 hours, then 100 mg daily from d2 continuously. UCN-01 was given as a 3-hour infusion on d4 at one of 3 dose levels (DL) (DL1 40/DL2 65/DL3 90 mg/m²). 28-day cycles were repeated depending on response/tolerability, with UCN-01 given at 50% dose. Antiemetic prophylaxis (perifosine) and hydration (UCN-01) were required.

Thirteen pts with AML (11), ALL (1), and MDS (1) were treated (DL1, n=6; DL2, n=4; DL3, n=3). Median age was 69 years (range, 20–85), 77% pts were male, and median number of prior treatments was 3 (range, 1–6). Nine of 11 AML patients had secondary (7)/therapy-related (2) disease, 4 had adverse and 5 intermediate karyotypes, and all were either primary refractory (4) or had first remission duration < 12 months (7). Pts received median 1 (range, 1–3) cycle of therapy. Five pts did not complete 1^st cycle due to disease progression (n=3; DL1), cholecystitis/sepsis (n=1; DL3), or drug-related toxicity (n=1; DL3), and 4 received hydroxyurea temporarily to control increasing blast counts. The most frequent drug-related toxicities (NCI CTC v3) were grade (gr) 1/2 nausea and diarrhea (62% each), vomiting (54%), fatigue and hyperglycemia (31% each) anorexia, hypocalcemia and constipation (23% each), increased AST/ALT (15%), hypokalemia and hypotension (8% each). Few ≥gr 3 toxicities were seen, all at DL3: fatigue (gr 3), hyperglycemia (gr 4), hypokalemia (gr 3), pericarditis/tamponade/hypotension (gr 4) and respiratory distress (gr 4). MTD was therefore defined as 65 mg/m² based on 2 pts with dose limiting toxicities at DL3. One AML pt (DL1) had disease stabilization for 3 cycles and one MDS pt (DL2) had improvement in neutrophil count. Given lack of sufficient evidence of clinical efficacy, the DL2 cohort was not expanded to 6 pts as planned. However, no gr 3/4 drug-related toxicities were observed in 10 pts treated at DL1/2.

Western blot analysis did not consistently demonstrate appreciable direct Akt inhibition in ficolled blood/marrow mononuclear cells from 5 pts (DL1/2) on d4 (perifosine alone) or d21-28, however, a 38–74% reduction in phosphorylated ribosomal protein S6 in leukemic blasts from 3 pts on d4 was noted by intracellular flow cytometry, consistent with inhibition of signaling through the Akt downstream target p70S6 kinase.

The average steady state concentration (C_ss) of perifosine was 4.75±2.53 μg/mL, consistent with published data (Eur J Cancer 38: 1615, 2002), with 1 pt having an unusually high C_ss of 9.25 μg/mL. The end-of-infusion concentration for UCN-01 was also consistent with published data (Clin Cancer Res 13: 2667, 2007), with average C_max 13.2±0.424 μg/mL and 20.7±3.08 μg/mL at doses of 40 and 65 mg/m², respectively.

UCN-01 (40 or 65 mg/m²)/perifosine can be safely administered, but this regimen lacked clinical efficacy. This approach may have failed because of insufficient Akt inhibition in vivo and/or need to inhibit more than one signaling pathway to produce clinical response. Future studies with more potent Akt inhibitors should examine the validity of this approach in conjunction with pharmacodynamic monitoring.

Carroll:Glaxo Smith Kline, Inc.: Research Funding; Sanofi Aventis Corporation: Research Funding; TetraLogic Pharmaceuticals: Research Funding; Agios Pharmaceuticals: Research Funding. Gojo:Keryx Inc.: Research Funding. Off Label Use: Neither UCN-01 nor perifosine have an approved indication.