Appendix:
Pharmacokinetics
ELISA assay format utilized anti-DM4 coated tubes to capture the huMy9-6-DM4 immunoconjugate and [1125]-labeled CD33 antigen for immunoconjugate detection. To reduce the cross-reactivity with CD33 antigens, standard quality controls and study samples were incubated with huMy9-6 antibody (10µg/mL) for 1h at room temperature before pipetting. The limit of quantization was 40ng/mL.
Competitive ELISA method is based on the competition for CD33 coated tubes of biotin-HuMy9-6 with the antigen present in the test sample. The limit of quantification was 5µg/mL.
Pharmacodynamics
CD33 density and occupancy were performed using two different mouse monoclonal antibodies (MAbs) of the same IgG1 isotype that recognize different epitopes of the CD33 antigen. MAb1 is competitive with huMy9-6-DM4 immunoconjugate and thus can not bind in the presence of the immunoconjugate on the cell surface, indicating the number of free, unoccupied CD33 target antigenic sites. MAb2 recognizes a different domain, binds independently of the presence or absence of bound immunoconjugate to the same antigen, thus indicating the number of total, occupied and unoccupied CD33 antigens, still present at cell surface. Samples stained on-site, stabilized by fixation, sent to a core-lab and analysed using a COULTER® EPICS® XL-MCLTM Flow Cytometer from Beckman Coulter (ROISSY, CDG, F) powered with the XL SYSTEM IITM SOFTWARE (Beckman Coulter) using 2- or 3-colour fluorescence analysis, for peripheral blood and bone marrow samples respectively. Differential analysis of leukocyte subsets and measurement of fluorescence intensity values was operated using an external flow cytometry software (FCS-Express®, Denovo Software, Los Angeles, USA). Mean fluorescence intensity (MFI) elicited by the MAb bound to the CD33 was compared to a calibration curve prepared using CellQuant Calibrator (Biocytex, France) to determine the Antibody Binding Capacity (ABC). Correction with the apparent ABC measured using an isotype-matched irrelevant mouse IgG1 provided the specific ABC (sABC) i.e. the number of MAb specifically bound per cell.
The quantization of DM4 on the blast surface was performed using an anti-DM4 specific MAb and the same QFCM approach as for CD33 MAbs. In order to describe the evolution of huMy9-6-DM4 conjugation ratio in vivo, irrespective of baseline and/or actual CD33 density, results were expressed as DM4/CD33 ratio by dividing the number of cell-bound anti-DM4 MAb molecules (sDM4) by the number of total CD33 molecules at each time-point. Post-treatment values were compared with baseline values obtained via an in vitro drug spiking assay under saturating concentration of AVE9633 (2µg/mL). The P-gp expression levels and functionality were measured using the anti P-gp170 specific MAb UIC2 in the absence and presence of vinblastin, a P-gp inducing substrate.16 The proportion of P-gp molecules that could be activated after in vitro incubation with vinblastin (Proportion of Activatable P-gp, PAP) was used as a measure of biological function [16].
1