Anti-CD45 Pretargeted Radioimmunotherapy in a Nonhuman Primate Model.

The efficacy of radiolabeled antibody (RAb) therapy has been validated by high response rates in hematologic malignancy, yet patients treated with conventional non-myeloablative doses of radioimmunotherapy (RIT) usually relapse. Low tumor-to-normal organ ratios of absorbed radioactivity are the result of ongoing non-specific radiation exposure through the circulation; this accumulates over time and is dose limiting. Our group has documented the efficacy of pretargeted RIT in murine models, using an anti-human (h)CD45 antibody (BC8). We have engineered, expressed, and purified a recombinant tetravalent single chain antibody (Ab)-streptavidin fusion protein [(scFv)₄ -SA] directed against (h)CD45. The tetrameric fusion protein (FP) retains the full antigen binding capacity of intact Ab. Because the tumor-reactive FP is not bound to a therapeutic radionuclide, it can localize to tumor sites without subjecting the rest of the body to non-specific irradiation. After maximal accumulation of BC8-FP in the tumor, a small molecular weight radioactive moiety with high affinity for the tumor-reactive FP is administered (radio-DOTA-biotin). This second reagent penetrates tumors rapidly where it binds to the pretargeted FP. Unbound radio-DOTA-biotin molecules are small enough to be rapidly cleared from the blood and excreted in the urine within minutes. In a series of studies involving fourteen fascicularis Macaques, we have shown that this two-step pretargeting approach using BC8-FP is feasible, safe and effective. We have identified that the blood clearance kinetics of radiolabeled BC8-FP are comparable to conventionally radiolabeled intact Abs (t_1/2= 40.6 and 50.2 hrs respectively), and have determined that the optimal time-point for radio-DOTA-biotin administration is 48 hrs after FP infusion. We found BC8-FP uptake in target tissue (lymph node and spleen) to be very specific when compared with a control anti-TAG72 (scFv)₄-SA (CC49-FP). Measured on a gamma counter at 96 hrs post infusion, the percent injected dose per gram (%ID/g) was 0.28±0.01 for BC8-FP and 0.03±0.01 for CC49-FP in lymph nodes (LN); and 0.33±0.02 and 0.07±0.002, respectively, in spleen. Serial LN biopsies and terminal spleen tissue harvests in animals co-injected with equimolar concentrations of intact BC8 antibody and BC8-FP, revealed equivalent initial uptake in target tissue, but superior retention of BC8-FP over time. After 96 hrs, the %ID/g of ¹¹¹In-BC8-FP in lymph node was 0.93±0.02, compared to 0.046±0.013 for BC8 Ab. In spleen the %ID/g was 0.158±0.0 for BC8-FP versus 0.057±0.0 for BC8 Ab. With two-step pretargeting, LN-to-blood and spleen-to-blood ratios 48 hours after ¹¹¹In-DOTA-biotin were 10.3:1 and 13.9:1, respectively, while for conventional BC8 Ab they were 2.6:1 and 3.0:1 respectively. These studies demonstrate that multi-step pretargeting can improve tumor-to-normal organ ratios of radionuclide delivery in a nonhuman primate model and offer the promise of more durable responses in patients with hematologic malignancy. Further investigation of radiolabeled BC8 as a rational therapeutic agent for both myeloid leukemia and non-Hodgkin’s lymphoma is warranted. Additional studies are planned involving introduction of a dendrimeric N-acetyl galactosamine “clearing agent” to further improve the therapeutic index. Future patient clinical trials are anticipated.