A Phase 1 and Correlative Biological Study of CSL360 (anti-CD123 mAb) in AML

Abstract

CD123 (IL-3Rα) is a phenotypic marker of putative leukemic stem cells (LSC) in AML (Jordan, Leukemia 2000; 14:1777). We and others have found that CD34+38− cells from AML patients (pts) express high levels of CD123, in contrast to absence of expression on CD34+38− cells in normal individuals. Binding of CD123 by monoclonal antibody (mAb) 7G3 inhibits IL-3-dependent signalling and proliferation in vitro. In a NOD-SCID xenograft model 7G3 inhibits human AML engraftment, but not normal human hematopoiesis (Lock ASH 2007; Abs161). CSL360, a recombinant chimeric IgG1 mAb derived from 7G3, binds the same epitope. CSL360 concentrations ≥ 0.1μg/mL in vitro inhibited 90% AML cell growth in the presence of supraphysiological IL-3 levels. Preclinical toxicology studies with doses up to 100 mg/kg weekly × 4 in cynomolgus monkeys showed no CSL360-related effects in clinical signs, hematology, chemistry, urinalysis, gross pathology or histopathology. A Phase 1 study of safety, pharmacokinetics (PK) and bioactivity of CSL360 in relapsed, refractory or high risk AML began in March 2007. Pts receive 12 weekly iv infusions if not withdrawn early due to treatment-related toxicity or disease progression. Additional treatments may be given to pts who achieve a response. Bone marrow aspirates/trephine samples are obtained at screening, after dose 3 and before doses 5 and 11. More than 180 infusions have been administered to 26 AML pts (21 M, 5 F; 17 de novo, 8 MDS-related, 1 treatment-related AML) in 5 dose level cohorts: 0.1, 0.3, 1.0, 3.0 and 10 mg/kg. There is no intra-patient dose escalation. PK parameters over the dose range, estimated in 19 pts over 7 days after doses 1 and 4, were linear with dose-proportional increases in the AUC and Cmax; dose 1 Cmax ranged from 0.62 – 287.33 μg/mL and dose 4 Cmax from 1.02 – 178.22 μg/mL. CSL360 mean plasma half-life (dose 1, 83 ± 33 h; dose 4, 117 ± 59 h) appears to be independent of dose and treatment number. Dose 1 systemic clearance (0.21 ± 0.16 L/h) and volume of distribution (0.39 ± 0.22 L/kg) were relatively low, consistent with this size IgG. In all pre-treatment samples anti-CSL360 antibody titers were negative, determined by enzyme immunoassay. Anti-CSL360 antibodies were detected post-treatment in 8/12 pts; these antibodies have not been fully quantified or characterised. CSL360 has been well tolerated; a MTD has not been defined. Seven pts received all 12 doses, 13 pts were withdrawn due to progressive disease or investigator's decision, 3 pts were withdrawn in association with infections, 2 pts withdrew consent, and 1 pt is ongoing. Three serious adverse events have been considered possibly related or related to CSL360: 1 invasive fungal infection (Gr 5), and 2 infusion reactions (Gr 2; hospitalised). Other adverse events are consistent with expectations for the disease population. Of 8 pts in the 3 mg/kg and 10 mg/kg cohorts who are evaluable for response after ≥ 4 doses, 1 complete response (CR) has been observed. A 22 yr old male, de novo FAB M1 cytogenetically normal AML, who had relapsed post-2 allogeneic SCT, achieved a morphologic leukemia free state after 3 doses (3.0 mg/kg) and CR after 12 doses, sustained for > 9 weeks. The pt received 17 doses before withdrawal to treat co-morbidities. Flow cytometry studies with anti-CD123 antibodies demonstrated dose-dependent CSL360 coating of both AML blasts and LSC. Saturation of target antigen on marrow and blood cells was observed 1 day after dosing at 0.3mg/kg, associated with decreased expression of CD123 detected by an antibody to a different epitope. At higher dose levels saturation of CD123 was maintained 7 days post dosing, associated with ongoing reduction in surface CD123 expression. In a representative sample, plasma from a pt treated at 10 mg/kg specifically inhibited IL-3-induced proliferation of AML blasts ex vivo, indicating sufficient circulating concentration of CSL360 to inhibit IL-3 mediated effects in vivo. Effects of CSL360 on proliferation and apoptosis of AML cells in treated patients are being investigated. These preliminary results show anti-CD123 mAb therapy with CSL360 is safe and tolerable; biological effects have been demonstrated; a sustained CR was achieved in 1 advanced, refractory AML pt. The study is continuing, with 20 evaluable patients to be accrued and treated at 10 mg/kg weekly; at this dose level the PK and correlative assays predict that complete blockade of IL-3 signalling through CD123 can be achieved in vivo.