Background : Acute myeloid leukemia (AML) is a heterogeneous disease with a variety of chromosomal and molecular aberrations, rendering AML a challenging disease to cure with targeted agents. Despite the genetic variability, AML uniformly responds to radiation therapy. We have previously shown therapeutic efficacy targeting CD45+ AML using the beta-emitter 90Y and a streptavidin-biotin (SA) pretargeted radioimmunotherapy (PRIT) system. Theoretical concerns, such as endogenous biotin in humans, may limit translation of this approach to the clinic. To circumvent such hurdles, we developed bispecific antibody constructs targeting both CD45 and 90Y-labeled-DOTA-biotin, and evaluated this approach in murine models of AML.

Methods : We constructed a pair of plasmids using the pfuse system that code for heavy and light chains for antibody bispecific to murine CD45 (30F11) and Y-DOTA (C825). Plasmid pairs were co-transfected into HEK 293T cells for production and purificaiton of the bispecific antibody. Binding to CD45 and 90Y-labeled-DOTA-biotin was confirmed, after which biodistribution studies with 30F11-Fc-C825 were performed in a disseminated murine leukemia model. B6SJLF1 mice (n=5 per group) were injected intravenously with 105 SJL murine leukemia cells, followed 2 days later with 0.67 nmol 30F11-Fc-C825, then 5 μg DOTA-Y-dextran clearing agent 23 hr later to clear any unbound bispecific construct from circulation. Mice were then injected with 50 µCi 90Y-DOTA-biotin IP 24 hr after the bispecific antibody infusion. Tissues were harvested at 6 and 24 hr after radiolabeled DOTA-biotin, and then analyzed for radioactivity to calculate the percent injected dose per gram of tissue (% ID/g).

Results : Target organs (spleen and bone marrow) from leukemia bearing mice treated with 30F11-Fc-C825 had 9.0 ± 1.5 and 8.1 ± 1.2% ID/g, respectively, 24 hr post injection. Non-target organs such as kidneys and lungs each had <0.5% ID/g at this time point. Targeting translated into a therapeutic benefit, as mice bearing SJL leukemia treated with 30F11-Fc-C825 and 800 µCi of 90Y-DOTA-biotin had median overall survival of 43 days, compared to leukemic control mice (30 days, p<0.0001).

To investigate this approach in a clinically relevant human leukemia model, an anti-human CD45 (BC8) scFv fusion protein was similarly engineered onto a high-affinity anti-Y-DOTA (C825) scFv fusion protein. This bispecific construct (BC8-Fc-C825) was expressed and purified from CHO cells. Binding to human CD45 and 90Y-labeled-DOTA-biotin was confirmed by flow cytometry. In biodistribution studies, female athymic mice (n=5 per group) were injected subcutaneously with 107 human AML cells (HEL). Mice were then injected intravenously with 1.4 nmol BC8-Fc-C825, BC8-SA, or non-targeting CC49-Fc-C825 as the first step before receiving clearing agent. Radiolabeled 90Y-DOTA-biotin (50 µCi) was injected IP 24 hr after bispecific delivery. HEL tumors from mice treated with BC8-Fc-C825 showed peak uptake at 24 hr post injection with 7.75 ± 0.02% ID/g, with minimal uptake in non-target organs [e.g., 0.5 ± 0.3% ID/g in kidneys]. This compared favorably with HEL tumors from mice treated with BC8-SA as the first step, which had peak uptake of 6.7 ± 0.01% ID/g 24 hr post-injection. Therapy studies took athymic mice (n=10 per group) that had been given subcutaneous HEL xenografts, before receiving BC8-Fc-C825, BC8-SA, or non-binding control CC49-Fc-C825. Treated mice were given clearing agent, followed by 800 to 1500 µCi of 90Y-labeled DOTA-biotin. Mice bearing HEL tumors treated with BC8-Fc-C825 followed by 1400 µCi of 90Y-DOTA-biotin, resulted in 6 of 10 mice surviving 170 days post-injection, whereas untreated controls and HEL-bearing mice treated with non-targeting negative control CC49-Fc-C825 and 1400 µCi of 90Y-DOTA-biotin required euthanasia due to tumor size by day 26 and 32, respectively.

Conclusion : Our bispecific constructs both showed selectivity for CD45+ targets and 90Y-DOTA. In two leukemia models, targeting of 90Y-DOTA to target hematopoietic tissues translated into survival benefits. Given the therapeutic efficacy, bispecific constructs with the capacity to pretarget CD45 and capture 90Y-DOTA should be added to the armamentarium of PRIT approaches.


Orozco: Actinium Pharmaceuticals: Other: Research Funding to Institution for sponsored Clinical Trials. Till: Mustang Bio: Patents & Royalties, Research Funding; Genentech: Research Funding. Gopal: Seattle Genetics: Consultancy, Research Funding. Orcutt: inviCRO: Employment, Equity Ownership. Pagel: Actinium Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees. Press: Roche: Honoraria, Research Funding; BMS: Honoraria; Bayer: Consultancy.

Author notes


Asterisk with author names denotes non-ASH members.