Comparison of 3,2-HOPO and DFO-based zirconium-89 antibody complex targeting mesothelin

Background: Mesothelin (MSLN) targeted thorium-227 (MSLN-TTC; [²²⁷Th]Th-3,2-HOPO-MSLN-mAb) has demonstrated in vivo efficacy in MSLN positive tumors^1,2. This MSLN antibody (MSLN-mAb, BAY 861903) based TTC is currently being evaluated in Phase I clinical trial (NCT03507452). Due to the low gamma emission with low abundance of measurable photons in the decay chain of thorium-227, tumor imaging using thorium-227 based conjugate is technically challenging^3,4. Besides, forming a complex with thorium-227, the 3-hydroxypyridin-2-one (3,2-HOPO) chelator can also form complex with zirconium-89.^,6. This approach thus would allow for radiolabeling the 3,2-HOPO-MSLN-mAb either with therapeutic or imaging radionuclide. Thus, the 3,2-HOPOH-MSLN-mAb was labeled with zirconium-89 to produce [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb. In parallel, the MSLN-mAb was conjugated to deferoxamine (DFO; [⁸⁹Zr]Zr-DFO-MSLN-mAb), one of the widely used chelator of zirconium-89. Both the zirconium-89 conjugates were compared in vitro and in vivo using MSLN positive tumor xenograft mouse models.

Methods: 3,2-HOPO-MSLN-mAb (BAY 2287409) and DFO-MSLN-mAb conjugates were labeled with zirconium-89 to yield [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb and [⁸⁹Zr]Zr-DFO-MSLN-mAb respectively. Radiochemical purity (RCP) was determined by size exclusion HPLC. The zirconium-89 conjugates were evaluated in vitro (binding affinity) and in vivo for biodistribution and PET imaging of HT29-MSLN and patient-derived (PDXs, NCI-Meso21 and NCI-Meso16) tumor xenografts. After injecting (i.v) tumor bearing mice (Athymic, nu/nu, female) with the radioactive conjugates, biodistribution and imaging was performed on days 1, 3, and 6 for HT29-MSLN xenografts and on day 3 for PDXs. Tissue associated radioactivity was determined by gamma counter and used to calculate % injected dose/g (%ID/g), tissue:blood (T:B), and tissue:muscle (T:M) ratios.

Results: The RCP of [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb and [⁸⁹Zr]Zr-DFO-MSLN-mAb was 52-76% (n=20) and 90-92% (n=8) respectively. [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb and [⁸⁹Zr]Zr-DFO-MSLN-mAb exhibited a low nanomolar binding affinity (K_d=0.16-2.3 nM) for MSLN. Pharmacokinetics over the time-course was similar for both the zirconium-89 conjugates except for blood, tumor, and femur. [⁸⁹Zr]Zr-DFO-MSLN-mAb showed higher HT29-MSLN tumor uptake (28-33 %ID/g) at all time-points compared to [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb (7-11 %ID/g). Similarly, on day 3, PDX tumor accumulation of [⁸⁹Zr]Zr-DFO-MSLN-mAb (15.88 -19.49%ID/g) was higher than [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb (7.95-13.07%ID/g). T:B and T:M ratios were also lower for [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb than the zirconium-89 DFO conjugate. However, femur uptake of [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb (6.74%ID/g) was ~2-fold higher compared to [⁸⁹Zr]Zr-DFO-MSLN-mAb (3.57%ID/g) at day 1 and then increased to ~3-4-fold over 6 days. At all times, PET imaging results paralleled the biodistribution pattern of both the zirconium-89 conjugates.

Conclusions: In vitro, both conjugates exhibited a high binding affinity for MSLN. In vivo, [⁸⁹Zr]Zr-DFO-MSLN-mAb showed higher tumor uptake and lower femur uptake than [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb. As [⁸⁹Zr]Zr-3,2-HOPO-MSLN-mAb uses the same chelator as [²²⁷Th]Th-3,2-HOPO-MSLN-mAb), the same 3,2-HOPO-MSLN-mAb conjugate could be better at studying organ distribution and lesion uptake of the MSLN-TTC, with the caveat that detection of MSLN positive tumors in the lower extremity might be more difficult if high femur uptake is also seen in humans.