Abstract 566: Preclinical and translational pharmacokinetic (PK) modeling of the self-assembling and disassembling (SADA) bispecific fusion protein CD38-SADA for first-in-human (FIH) pretargeted radioimmunotherapy (PRIT)

Abstract Background: PRIT is designed to deliver tumor-targeted radiation while limiting exposure to healthy tissues. CD38-SADA binds CD38 and select radioisotopes chelated to tetraxetan (DOTA) and contains a p53-derived domain mediating the self-assembly of tetramers from monomers in a reversible, concentration-dependent manner. Preclinical studies have demonstrated dose-dependent anti-tumor efficacy of two-step CD38-SADA PRIT with 177Lu-DOTA administered 48h after an initial CD38-SADA infusion. Here, we report preclinical PK models and human simulations that informed the FIH study design and dosing regimen. Methods: Total plasma concentrations of CD38-SADA (not cross-reactive with mouse CD38) were measured in 2 repeat dose studies following administration on D1, 8, and 15 in transgenic mice expressing human CD38 (huCD38 tg) and in a single dose study in huCD38 tg and wild-type (WT) mice. A PK model with Michaelis-Menten approximation of target-mediated drug disposition (TMDD) was fitted to the combined data. The PK model also incorporated in vitro mass photometry estimates of the CD38-SADA tetramer: monomer equilibrium. The model was scaled for human PK simulations using body weight-based allometry and theoretical scaling coefficients. For TMDD scaling, 2 scenarios were simulated: 1) Same CD38-mediated elimination in human as estimated in huCD38 tg mice. 2) No CD38-mediated elimination, i.e., the most conservative case resulting in the highest predicted CD38-SADA exposure. Results: The mouse PK model adequately described the observed CD38-SADA plasma concentration-time profiles. Cmax was dose-proportional, with faster elimination observed in huCD38 tg mice vs WT mice at low doses, consistent with TMDD through binding to CD38 receptors. CD38-SADA monomers and tetramers demonstrated parallel linear and CD38-mediated nonlinear elimination from the central compartment. Linear clearance of CD38-SADA monomers was approximately 20 times faster than CD38-SADA tetramers (CLmon 1.21 vs CLtet 0.058 mL/h). At the planned FIH starting dose of 3 mg/kg, the predicted exposure provided a sufficient toxicological margin with/without TMDD scaling: Cmax = 57.68 vs 58.71 μg/mL; AUC0-inf = 1070 vs 3020 μg∙h/mL, respectively. The predicted tetrameric prevalence was ≥90% at the Cmax concentrations and >50% at total plasma CD38-SADA concentrations >1 μg/mL, which was predicted to last for 40 to 136 hours depending on the TMDD scaling scenario. Conclusions: A 2-compartmental PK model with nonlinear CD38-mediated elimination of CD38-SADA described the observed total CD38-SADA plasma concentration-time in huCD38 tg mice. PK simulations supported the initial FIH dosing for trial NCT05994157, CD38-SADA/177Lu-DOTA PRIT in patients with relapsed/refractory non-Hodgkin lymphoma. Citation Format: Brian H. Santich, Charlotte A. Russell, LinLin Wang, Maria Düring, Lionel Renaud, Matthias Machacek, Norman LaFrance. Preclinical and translational pharmacokinetic (PK) modeling of the self-assembling and disassembling (SADA) bispecific fusion protein CD38-SADA for first-in-human (FIH) pretargeted radioimmunotherapy (PRIT) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 566.