Background CD33 is expressed on myeloblasts in the majority of patients with AML. Vadastuximab talirine (33A) is a potent, highly stable antibody-drug conjugate composed of a humanized CD33-directed antibody conjugated at engineered cysteine residues to 2 molecules of pyrrolobenzodiazepine (PBD) via protease cleavable linkers. Upon binding to CD33, 33A is internalized and PBD is released leading to DNA cross-linking and cell death. RO on targeted cells following treatment administration is a pharmacodynamic measure of myeloblast exposure to 33A.

Methods A phase 1 study (NCT01902329) assessed the safety and antileukemic activity of 33A as monotherapy. Best response was evaluated as CRc (CR + CR with incomplete count recovery), morphologic leukemia-free state, or resistant disease. Peripheral blood and bone marrow (BM) aspirates were collected for pharmacokinetic (PK) and biomarker analysis, including assessment of minimal residual disease negative (MRDneg) remission by 10-color flow cytometry. Calculated 33A PK parameters included the observed maximum concentration (Cmax) and the area under the concentration-time curve to the last quantifiable value (AUClast). Myeloblast CD33 levels were measured by multi-color flow cytometry using 2fluorophore conjugated anti-CD33 reagents (1 competitive with 33A). To assess RO, the relative fluorescence values for each reagent were compared in samples drawn before and 2 hours after IV 33A administration. RO was obtained from 71 evaluable patients; 57 (80%) had relapsed/refractory disease (median age 72); 14 patients were treatment naïve (median age 76).

Results Cmax and AUClast increased with increasing dose. Dose-normalized (DN) AUClast declined with increasing baseline BM cellularity and myeloblast percentage. Greater RO was associated with greater Cmax and AUClast. In a univariate analysis, variables associated with odds of CRc included treatment naivety, DN Cmax, and AUClast. Variables associated with odds of MRDneg included RO, DN Cmax, Cmax and AUClast. RO ranged from <10% to >80%; across RO tertiles (32%, 55%), the odds of MRDneg remission increased from 4% (1/23) to 25% (6/24) to 50% (12/24). After accounting for dose, baseline BM myeloblasts, treatment naivety, and NPM1 mutation, the odds of CRc associated with DN Cmax and cytogenetic risk by MRC, while the odds of MRDneg associated with DN Cmax, Cmax, RO, and DN AUClast.

Conclusions Dose normalized serum exposure to 33A declined with increased disease burden, as assessed by BM cellularity and myeloblasts. Exposure-related RO on peripheral blood myeloblasts confirmed CD33 targeting by 33A in patients. Together, these results support target-mediated disposition of 33A. An association of the odds of achieving MRDneg with RO, Cmax, and AUClast was observed with 33A monotherapy treatment. High RO was not required for patients to achieve CRc or MRDneg remission. Further efforts are underway to better understand the pharmacologic and molecular drivers of the deep, MRDneg remissions mediated by 33A.

Citation Format: Charles Biddle-Snead, Jenna L. Voellinger, Phoenix Ho. Assessment of myeloblast CD33 receptor occupancy (RO) by vadastuximab talirine in patients with acute myeloid leukemia (AML) receiving monotherapy treatment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr CT120. doi:10.1158/1538-7445.AM2017-CT120