Abstract LB-182: Optimized antibody modification and radiolabeling conditions for zirconium-89 immuno-PET

Abstract

Abstract OBJECTIVES: The PET imaging isotope 89Zr (t1/2 = ∼3.3 d) is the most commonly used isotope with antibodies (Abs) (immuno-PET). The chelator desferrioxamine (DFO) has been conjugated to many Abs, radiolabeled with 89Zr, and imaged in many animals and humans. It is generally known that conjugating too many chelators onto Abs will decrease binding affinity and immunoreactivity. We have conjugated DFO to the HER2/neu targeting Abs trastuzumab in varying quantities, and then radiolabeled with 89Zr under a comprehensive set of conditions and evaluated their effects on radiochemical yields (RCY), specific activity (SA), immunoreactivity, and ultimately in vivo tumor targeting. We believe a comprehensive study on these variables has not been sufficiently performed and would be of substantial value to anyone attaching isotopes to antibodies. METHODS: The chelator p-SCN-Bn-DFO was conjugated to trastuzumab (5, 10, 20, 40, 60, 100, 150, 200 equivalents), purified, and the number of attached chelators determined by MALDI-TOF mass spectrometry. These different modifications of DFO-trastuzumab were then radiolabeled and RCY were monitored from 15-60 min by iTLC (25 °C and 37 °C, n = 3, 50 mM EDTA mobile phase). The immunoreactivity of these different conjugates was determined by a cell binding assay with SKOV3 cells, and binding affinity by surface plasmon resonance. This was followed by 89Zr radiolabeling with varying quantities of Abs (5, 25, 50 μg), different buffers (PBS, chelexed PBS, TRIS/HCl, HEPES), different volumes (100, 300, 1000 μL), and activities (0.2, 1.0 mCi). To test radiolysis effects, 89Zr-DFO-trastuzumab (∼200 μCi) was radiolabeled and then placed in 100 μL of different buffers (PBS, TRIS/HCl, saline, HEPES, PBS + ascorbic acid, and PBS + gentisic acid) for 48 h before determining immunoreactivity. Representative samples (5, 10, 20, and 200 equiv. DFO conjugations) were radiolabeled and injected in mice bearing subcutaneous SKOV3 xenografts (n = 4 per group), and imaged by PET at 24, 72, and 120 h p.i. with biodistributions done at 120 h. RESULTS: Regardless of the equiv. chelator used during conjugations, a maximum average number of chelates/Abs of only ∼10-12 could be attached. To achieve maximum SA, a small quantity of Abs was used (5 μg), achieving RCY of ∼50-90% and SA of ∼15-35 mCi/mg (depending on chelates/Abs). The optimal conjugation level and radiolabeling conditions were 1-3 DFO per Abs (5-10 equiv. conjugation), chelexed PBS as buffer, minimum buffer volume, and no radioprotectant (gentisic acid or ascorbic acid). The in vivo effects of these chelate conjugations were dramatic, revealing a trend where higher amounts of attached DFO correlated to lower tumor uptake and higher liver and spleen uptake. The 89Zr-DFO-Trastuzumab that was prepared by reacting 5 equiv. of chelator (∼1.4 ± 0.5 chelates/Abs) had tumor and liver uptake at 120 h p.i. of 38.7 ± 3.8 and 6.3 ± 4.1%ID/g, respectively, where the 200 equiv. conjugation (∼10.9 ± 0.7 chelates/Abs) had values of 16.2 ± 3.2 and 27.5 ± 4.1%ID/g, respectively (p < 0.05). CONCLUSIONS: Increasing the degree of DFO attachment to trastuzumab resulted in a concomitant increase in maximum SA values (up to ∼35 mCi/mg), and a decrease in immunoreactivity/binding affinity, which translated to inferior in vivo performance. Optimized radiolabeling conditions provided very high SA of 17.5 ± 2.2 mCi/mg with only ∼1.4 ± 0.5 chelates/Abs. Citation Format: Eric W. Price, Jonathan M. Glaser, Kimberly J. Edwards, Jason S. Lewis. Optimized antibody modification and radiolabeling conditions for zirconium-89 immuno-PET. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr LB-182.