A Phase 1b Study of Vadastuximab Talirine As Maintenance and in Combination with Standard Consolidation for Patients with Acute Myeloid Leukemia (AML)

Background

Post-remission therapies for patients with AML such as high-dose cytarabine (HiDAC) and allogeneic stem cell transplant (alloSCT) have led to improved outcomes for younger patients, but disease recurrence remains prevalent with ~40% 5-year OS. CD33 is a cell surface receptor expressed in ~90% of AML, representing a promising target for therapy. Vadastuximab talirine (33A) is a CD33-directed antibody conjugated to 2 molecules of a pyrrolobenzodiazepine (PBD) dimer.

Methods

This phase 1b dose-escalation study (NCT02326584) evaluates the safety and anti-leukemic activity of 33A in combination with consolidation therapy (HiDAC) or as a single agent for maintenance therapy. AML patients (ECOG status 0-1) must be in 1st remission (CR or CRi) after standard induction therapy and be able to receive HiDAC (consolidation cohort) or be in 1st remission and have completed planned post-remission therapies, either chemotherapy and/or alloSCT (maintenance cohort). For maintenance post-alloSCT, patients were between Day 60 and 100 post-transplant without significant GVHD. Prior to HiDAC administration (3 gm/m2 q12h Day 1, 3, 5), 33A is given on Day 1 for up to 4 cycles (28-day cycle). For maintenance therapy, 33A is given as a single agent on Day 1 for up to 8 cycles (6-wk cycle).

Results

Consolidation cohort: 21 patients (57% male) with a median age of 52 years (range, 21-64) were treated with 5, 10, or 20 mcg/kg of 33A with HiDAC. Patients received a median of 2 cycles (range, 1-4). As anticipated, all patients experienced Grade 4 myelosuppression. At 20 mcg/kg, 1 DLT (lack of recovery of platelets [25K] and/or ANC [500] by Day 42) occurred in Cycle 1. At 10 mcg/kg, no DLTs were observed but delay of subsequent cycles of treatment occurred in 4 of 10 patients, primarily due to thrombocytopenia. No DLTs were observed in the 8 patients treated at 5 mcg/kg and 1 non-hematologic-related dose delay was reported (otitis externa). Non-hematologic treatment-emergent adverse events (AE) in ≥25% of patients regardless of relationship included nausea (38%) and fatigue (33%). No infusion-related reactions (IRRs) or events of veno-occlusive disease were reported. The 30- and 60-day mortality rates were 0%. Of the 19 efficacy evaluable patients, 15 (79%) have maintained remission, 18 patients are alive and 3 patients (14%) remain on treatment. Reasons for treatment discontinuation were completion of planned consolidation therapy (38%), AE (thrombocytopenia, 14%), leukemic relapse (5%), and other non-AE (29%). Nine patients (43%) went on to receive an alloSCT.

Maintenance cohort: 22 patients (41% male) with a median age of 45.5 years (range, 23-71) have been treated with 5 mcg/kg of 33A. Patients were a median of 6.2 months from diagnosis (range, 3.4-21.5); 12 patients completed chemotherapy-based treatment alone and 10 patients completed standard chemotherapy with an alloSCT in 1st remission. Patients received a median of 3 cycles (range, 1-6); no DLTs were reported. AEs reported in ≥15% of patients were fatigue (41%), neutropenia (41% [36% ≥G3]), nausea (36%), thrombocytopenia (36% [27% ≥G3]), diarrhea, dyspnea, headache, and vomiting (18% each); no IRRs were observed. Of the 20 efficacy evaluable patients, 15 (75%) have maintained remission. Reasons for treatment discontinuation were AEs (41%, primarily myelosuppression), leukemic relapse (14%), completion of planned therapy (9%), and other non-AE reasons (19%); 4 patients (18%) remain on treatment. Median OS is not yet reached and 19 patients are alive.

Pharmacokinetic data in patients receiving post-remission therapy with 33A demonstrate that exposure appears to be greater than in patients with active disease, possibly due to a decrease in target-mediated disposition.

Conclusions

33A can be safely administered in combination with HiDAC and as monotherapy in the post-remission setting. In combination with HiDAC, non-hematologic toxicities of 33A were consistent with effects reported with HiDAC alone. As a single agent, 33A administered as maintenance post-chemotherapy and/or alloSCT results in predictable on-target myelosuppression, with mild non-hematologic adverse effects.